Hydraulic Handbook PDF

THE HYDRAULIC
HANDBOOK
9th Edition
THE HYDRAULIC
HANDBOOK
9th Edition
By
T. Hunt and N. Vaughan
ISBN 1 856172503
Copyright 1996 ELSEVIER SCIENCE LTD
All rights reserved

This book is sold subject to the condition tlult it slulll not by way of trade or otherwise be
resold, lent, hired out, stored in a retrieval system, reproduced or translated into a machine language,
or otherwise circulated in any form of binding or cover other tluln tlult in which it is published,
without the Publisher's prior consent and without a similar condition including this
condition being imposed on the subsequent purclulser,
Other engineering publications by Elsevier Advanced Technology include:

Condition Monitor Newsletter
Handbook of Maintenance Management
Filtration and Separation Magazine
Membrane Technology Newsletter
Handbook of Filter Media
Seals and Sealing Handbook
Handbook of Hose, Pipes, Couplings and Fittings
Handbook of Power Cylinders, Valves and Controls
Pneumatic Handbook
Pumping Manual
Pump User's Handbook
Submersible Pumps and their Applications
Centrifugal Pumps
Handbook of Valves, Piping and Pipelines
Handbook of Fluid Flowmetering
Handbook of Noise and Vibration Control
Handbook of Meclulnical Power Drives
Industrial Fasteners Handbook
Handbook of Condition Monitoring
Published by
Elsevier Advanced Technology
The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, UK
Tel +44 (0) 1865-843842
Fax +44 (0) 1865-843971
Printed in Great Britain by BPC Wheatons Ltd, Exeter

Preface
The harnessing of hydraulic power has come a long way since its inception at the Flood!
It has also advanced dramatically since the first edition of this handbook in the 1940s. But
it is the more recent changes since the 8th edition in 1983 which have caused this new
edition to become necessary.
Much greater dependence on microprocessor control and the requirement for smaller
quieter units of higher power have caused the major differences, particularly in valves.
Conversely the growth in the capacity of hydraulic motors has dwarfed those of only a
decade ago. But there is also the improved reliability through design and use of hydraulic
systems; unitary or combined construction has increased not only with multi-pumps, but
also with the integration of electronics with mechanical components; there is the greater
awareness of the value of water hydraulics, although the major growth here is yet to come.
Some chapters, as always with a new edition, follow the lines of the earlier edition, but
the major of the Hydraulic Handbook has been completely rewritten to enable users to
enter the 21st century with confidence.
Acknowledgements
(excluding those companies acknowledged with illustrations and tables)
Mr Chris Rice (British Aerospace pic, Filton) Mr Denton French (Mannesman Rexroth Ltd)
Mr Brian Holmes (Hiigglunds Drives Ltd) Mr Kevin Parslow (SAI (GB) Ltd)
Mr Andrew Pries (Enerpac Ltd) Mr Anthony Boyd (Parker Hannifin Ltd)
Mr Karen Shillito (Hauhinco Water Mr Matt Maher (Poclain Hydraulics Ltd)
Hydraulics Ltd) Mr Dick Scutt (David Brown Hydraulics Ltd)
Mr Glen Quickfall (Power Team) Mr Graham Miller (Fawcett Christie Ltd)
Mr Paul Newnham (Denison Hydraulics Mr Keith Mercer (Sauer Sundstrand Ltd)
UK Ltd) Mr Tony Wynn (Ace Controls International
Mr Chris Boll (Robert Bosch Ltd) Ltd)
Mr Mike Day (The Pall Group) Mr Joseph O'Connor (Shell Oils)
Mr Tony Clements (Linde Hydraulics Ltd) Mr David Woolley (Serck Access (UK) Ltd)
Mr Ray Roderick (Flu pac Ltd) Mr John Bentley (Oilgear Towler Ltd)
Mr Warran Lamb (Hose-Fit Ltd) Susie Bergin (V A Technology Ltd)
Mr Brian Withington (Occo Coolers Ltd)
vii
Contents
SECTION ONE
1.1 SI Units for Fluid Power ................................................................................. 1
1.2 Notation and Symbols ..................................................................................... 5
1.3 The Hydraulic Advantages .............................................................................. 9
1.4 Properties of Fluids ......................................................................................... 15
1.5 Basic Hydraulic Theory ................................................................................... 27
1.6 The Hydrostatic Principle and Basic Circuits ................................................. 41
SECTION TWO
Part 2a
2a.l Hydraulic Pumps ............................................................................................. 49
2a.2 Pump Drivers and Couplings .......................................................................... 77
2a.3 Hydraulic Motors ............................................................................................ 85
2a.4 Hydraulic Hand Pumps ................................................................................... 93
2a.5 Hydraulic Cylinders ........................................................................................ 95
2a.6 Rotary Actuators .............................................................................................. 125
Part 2b
2b.l Hydraulic Valve Classification ........................................................................ 133
2b.2 Valve Construction & Connection .................................................................. 139
2b.3 Directional & Flow Control Valves ................................................................ 157
2b.4 Pressure Control Valves ................................................................................... 175
2b.5 Proportional Valves ......................................................................................... 187
2b.6 Servovalves ..................................................................................................... 201
Part 2c
2c.l Accumulators .................................................................................................. 213
2c.2 Shock Absorbers .............................................................................................. 235
2c.3 Coolers ............................................................................................................ 239
2c.4; Heaters ............................................................................................................ 249
2c.5 Reservoirs and Tanks ....................................................................................... 251
2c.6 Tubes & Pipework Calculations ...................................................................... 259
2c.7 Pipework Couplings & Fittings ....................................................................... 275
2c.8 Hydraulic Hose ................................................................................................ 285
2c.9 Hose Couplings and Fittings ........................................................................... 297
2c.l0 Seals ................................................................................................................ 305
2c.ll Protectives Covers ........................................................................................... 321
SECTION THREE
3.1 Hydraulic Fluids .................................................................................................. 325
3.2 Filters and Filtration ............................................................................................. 341
3.3 Contamination Control ......................................................................................... 357
3.4 Cleanliness Coding Systems ................................................................................ 367
viii
SECTION FOUR
4.1 Linear Circuits ..................................................................................................... 373
4.2 Actuator Perfonnance .......................................................................................... 385
4.3 Hydrostatic Transmissions ................................................................................... 399
4.4 Design Considerations ......................................................................................... 419
4.5 Computer Aids and Simulation ............................................................................ 423
SECTION FIVE
5.1 Servosystems ....................................................................................................... 431
5.2 Electro-Modulated Hydraulics ............................................................................ 447
5.3 Pneumatic Logic Controls .................................................................................... 455
5.4 Computers for Control ......................................................................................... 459
SECTION SIX
6.1 Faults and Leakage .............................................................................................. 465
6.2 Instrumentation .................................................................................................... 471
6.3 Testing Systems and Components ........................................................................ 481
6.4 Monitoring ........................................................................................................... 489
6.5 Maintenance Programmes .................................................................................... 501
6.6 Vibration and Noise ............................................................................................. 511
6.7 Legislation and Safety .......................................................................................... 521
SECTION SEVEN
7.1 Miniature Hydraulics ........................................................................................... 525
7.2 High Temperature Hydraulics .............................................................................. 531
7.3 Ultra-High Pressure Hydraulics ........................................................................... 537
7.4 Water Hydraulics .................................................................................................. 547
7.5 Hydro-Pneumatics ............................................................................................... 553
7.6 Standards .............................................................................................................. 559
7.7 Training ................................................................................................................ 575
SECTION EIGHT
8.1 Mechanical Handling ........................................................................................... 577
8.2 Industrial Robots .................................................................................................. 591
8.3 Machine Tools and Automation ........................................................................... 597
8.4 Workshop Tools ................................................................................................... 605
8.5 Injection Moulding Machines ............................................................................. 611
8.6 Hydraulic Presses ................................................................................................ 617
8.7 Hydraulic Ring Mains and Power Packs ............................................................ 637
8.8 Mobile and Agricultural Hydraulics ................................................................... 643
8.9 Land-based Transport .......................................................................................... 655
8.10 Hydraulics in Aerospace .................................................................................... 669
8.11 Marine Hydraulics .............................................................................................. 679
8.12 Offshore and Subsea .......................................................................................... 689
8.13 Mining and Intrinsic Safety ................................................................................ 695
8.14 Hydraulics in Medicine ..................................................................................... 707
ADVERTISERS BUYERS GUIDE ........................................................................ 713
EDITORIAL INDEX ............................................................................................... 729
Hiigglunds drive systems provide your machines with a maximum of
flexibility thanks to the stepless speed control. No gearboxes or clutches
means no trouble and longer lifespan of the drive as well as the machine.
This together with the fact that the hydraulic motor is fitted directly to the
machine shaft makes it possible to save space. The power unit can be lo-
cated just about anywhere you want it. A drive system from Hiigglunds
Drives is a compact, simple concept, easy to install with very low
maintenance.
In many applications it is essential that the machine can start irrespective
of load conditions. With our drive systems you will get very high start up
torque through to maximum speed. The drive system can deal with
frequently occurring starts/stops or reversal without loss of production
capacity. The extremely low inertia eliminates machine stresses and strains
and the excellent low speed performance is controllable in four quadrants.
A low-speed drive from Hiigglunds Drives, fitted directly on the machine
shaft, provides integral overload protection. As there are no gearboxes or
clutches the system is insensitive to shockloads. The hydraulic drive
system controls the power within very close tolerances (also when stalled)
which protects the whole machine against damage. The truly shockproof
design makes the 'Whole system tough and reliable.
When you get a drive system from Hiigglunds Drives you get a complete
system with power units, controls and installation. You can transform your
machine from the slave of an operator into an intelligent high production
unit.
All in all a drive system from Hiigglunds Drives is a reliable drive that
will live up to your expectations.
Do away with problem gearboxes.
Send for a free brochure today and find out more about Hiigglunds Drives
and what we can do for vou.
SECTION 1
The International Association for Hydraulic Research (IAHR), founded in 1935, is the world's foremost
independent organisation of engineers and scientists engaged in hydraulic engineering. IAHR promotes the
exchange of knowledge through congresses, symposia, working groups and publications in the fields of river and
coastal hydraulics, water resources, environment, disaster prevention risk analysis, energy and industrial
processes.
The structure of IAHR is composed of Technical and Regional Divisions. The Technical Sections cater to the
applied and scientific needs of our members. The Sections regularly organise conferences and symposia in their
own specific fields, thus providing a basis for worldwide cooperation, or the development of monographs on
specific topics.
Benefits of membership
Individual members receive: the Journal of Hydraulic Research and IAHR Bulletin (6 times/year); discount on
the proceedings of the biennial congresses and other IAHR publications; Register of Members; reduced
registration fees at biennial congresses and most co-sponsored congresses.
Corporation members receive: the IAHR Bulletin (6 times/year); Register of Members. This category is open to
engineers and scientists whose employer is a corporate member of IAHR.
Corporate members (universities, research institutes, engineering firms, consultants) receive: two copies of the
Journal of Hydraulic Research and IAHR Bulletin (6 times/year); Proceedings of the biennial congresses; two
copies of the Register of Members; other benefits as shown under "individual"
Membership fees depend on category of membership and country of residence
For more information contact:

IABR Secretariat P.O. Box 177 2600 MH Delft, the Netherlands,
tel: +31-15-2858585, fax: +31-15-2858582, e-mail: iahr@ ldelft.nl, internet: http:// .iahr.nl
SI UNITS FOR FLUID POWER
SI UNITS were agreed internationally in the late 1960's. At the time, they were an
extension and refinement of the previous metric system. Because the Systeme Interna-
tional d'Unites (SI) was designed to be a coherent system of units - where the product or
quotient of any two unit quantities is the unit of the resultant quantity - there is an agreed
restricted number of seven Base Units in accord with ISO 1000: 1992 (Ed. 3) - see also ISO
31:1992. These are;
metre - m the unit length
kilogram kg the unit of mass
second - s the unit of time
ampere - A the unit of electric current
degree kelvin K the unit of thermodynamic temperature
mole - mol the amount of substance
candela cd the unit of luminous intensity
2 THE HYDRAULIC HANDBOOK
There are also two Supplementary Units relating to angle

radian - rad the plain angle
steradian sr the solid angle
A number of Derived Units may also be used. They are derived from the Base Units, or
are especially relevant in certain applications. These frequently have special names.
Great care must be taken when considering weight and mass and force. The unit of mass
is the kilogram (kg), and the unit of force is the newton (N). Thus, because
Force = Mass x Acceleration,
the units ofthe newton are kg m S-2.Weight, however, is also a force and hence should
be rated in terms of newton. Although weight is commonly called kgf, this is not an SI unit
and should be discouraged.
In addition there are a few Non-SI Units which are permissible because of their practical
importance, or for use in specialised fields. These are:
bar bar a unit of pressure = 105 Pa == 1 atmosphere
litre L or I a unit of volume = 10-3 m3
tonne t a unit of mass [metric ton] = 103 kg = I Mg
poise P a unit of dynamic viscosity = 10-1 kg m-I sol
stokes St a unit of kinematic viscosity = 10-4m2 S-I
(Note: 1 cP = 1 mPa g 1 cSt = 1 mm2 g-l )
and the commonly used units of time
minute min = 60 s
hour h = 60 min
day d = 24 h.
In order to keep the numerical values to reasonable lengths, the use of prefixes is
recommended. However, compound prefixes are not to be used (i.e. both numerator and
denominator) and the prefix should normally only be used in the numerator. Only one
prefix is recommended in forming a multiple of a compound SI unit. (Prefixes must not
be used for the non-SI units of time.) Thus the following
are acceptable: but not acceptable:

L / s , or L S-1 mL / ms , or mL ms-l
mL / min , or mL min-l mL / mmin , or mL mmin-l.
It will be apparent to the reader that great care must be taken when using 'm' which can
be both 'milli' and 'metre'. It is normal to leave a space after the 'm' if it is being used as
a 'metre', but not if it is purely a prefix.
The SI prefixes for the fractions and multiples are shown in Table 1.1.2:
Note: the non multiples of 103, i.e. the first two rows, are not recommended, but may
SI UNITS FOR FLUID POWER
be used. Examples in fluid power are those of mineral oil viscosity, where 'cP' and 'cSt'
are more common than 'P' or 'St'.
Whilst it is expected that SI units are used in all new work, there will be examples of
other units, such as the Imperial units, which are still in use in older systems. In order to
help in the understanding of these, Imperial units are frequently quoted (in brackets) after
the SI quantity.
A table of Conversion Factors is given in Table 1.1.3.
5
NOTATION AND SYMBOLS
. IT IS IMPORT ANT to be able to recognise the notation and figures as they appear in the
text. There are standards for both in the International Standards Organisation (ISO) and
these will be adhered to throughout. In order to clarify the situation, and indicate other
notation used in the book which is not in the ISO lists, a full review is given on the next
two pages. These are separated into English and Greek Characters for convenience.
The full range of graphical symbols, and many examples of complex arrangements of
components in fluid power circuit diagrams, are given in the ISO 1219-1:1991 standard.
The small selection in this chapter is a brief summary of the basic symbols only, and no
attempt has been made to show the various combinations possible.
9
THE HYDRAULIC ADVANTAGES
POWER MAYbe transmitted in several ways. For instance, the following are all
possibilities:
Electrics - by means of electrons at a potential

Pneumatics - by means of a gas under a pressure
Mechanics - by means of a mechanical structure under a load
Hydraulics - by means of a liquid under a pressure.
There are, of course, other power transmission media - the most important of all, which
affects every one of us, is radiation from the sun. Radiation may be used as a transmission
medium on the earth as well.
None, however, has a position oftotal superiority.
Each has to be assessed not only in its general applicability to a particular situation, but
also to its particular suitability. One very common application, which has been tried in all
four types of transmission is that of the bicycle or tricycle. A pneumatic bicycle was on
show at the 1951 Festival of Britain exhibition; Sinclair introduced the electric cycle,
Danfoss the hydraulic bicycle, but so far the mechanical transmission has proved to be not
only the first but also the best because of its simplicity and efficiency.
There are variations in each transmission. For instance, a gear train is not going to
achieve the same result as sprockets and chain, although both are using mechanics for the
transmission. Similarly, an open channel for hydraulics is not going to achieve the same
result as an enclosed pipe.
In fact, each of the 'media' has advantages and disadvantages, and, as improvements
in design and manufacture take place, the relative advantage may vary. Some of the
general conclusions which may be made at the current time are outlined in Table 1.3.1.
However, there are no clear cut distinctions, as mentioned above, and there will be
occasions when the design, or materials available, may completely reverse the situation.
It is possible, also, that the final decision may be based on cost alone.
Factors to be considered
Apart from price there are several factors which need to be considered before a choice of
transmission is made. Table 1.3.1 highlights a few ideas. Consider the following:
What starting torque is required?
How much space is available?
Are the conditions constant or variable?
What is the typical duty cycle?
What power level is required?
What rotary speed is required?
Is the environment potentially hazardous?
How safe has the system to be?
1lIE HYDRAULIC ADVANTAGES II
Is the environment 'unfriendly'?

What finesse of control is required?
What total efficiency is expected?
What total reliability is required, including control?
How are failure conditions to be monitored?
What distance has the transmission to cover?
Although each category mentioned above would have a variety of answers depending
upon the exact requirement, most of them are generally considered favourable with
hydraulic systems.
Figure 1.3.1 is a suggestion of the most commonly acceptable regions of applicability.

It shows where hydraulics may be more suitable, even in rotary applications where
electrical motion would normaIly be superior. The hydraulic region is emphasised on the
high power end, but shows little application on the very small torque required for miniature
systems and controls.
Typical hydraulic applications

The very wide range of possible components enables the user of hydraulics to exercise
considerable flexibility in arriving at a solution. As mentioned above and illustrated in
figure 1.3.1, hydraulic applications are most valuable where high force and power
concentration and capacity is required in limited space. Such compact designs incorporate
relatively small masses which reduce dynamic forces and, hence, encourage reliability and
accuracy.
Applications are many in both the static and the mobile fields. Examples of static
applications are plastics processing machines, presses, cutting machine tools and forging
machinery. The mobile field covers a considerable range from mining to construction,
from agriculture to marine, from road to rail; or the even greater extremes of aerospace to
underwater. Robotics vies with electrical and pneumatic, but there are many occasions
when hydraulic systems are used.
The high power required at slow speed for such applications as bridge moving, or the
rapid acceleration ofIarge masses from stationary as with military functions, are both ideal
features for hydraulic power. Quite frequently the cost of hydraulic designs in comparison
with electro-mechanical designs may be some 30% lower, even for medium output, and
even less for the greater outputs.
Electro-hydraulic systems have advanced considerably due to the development of
electronic digital microprocessor-control, improving the energy utilisation and precision
of the system. Hydraulic fluids, too, now cover a much wider range enabling higher
pressures and temperatures to be achieved.
Some applications involve a combination of hydraulics with one ofthe other media. For
instance, electro-hydraulics are common with injection moulding machines, whilst
mechanical-hydraulics are seen in many mobile machines such as automobiles, earth-
movers and mobile cranes.
Examples of hydraulic operations

The examples given illustrate typical situations where an hydraulic transmission scores
over electrical, mechanical and pneumatic transmissions.
Example of an hydraulic drive in a process plant
The requirement was to drive a number of agitators on the principal reactors on a
polystyrene production line. The output of close to 50 000 tonnes per annum was not only
large but needed the maximum reliability to avoid unscheduled and costly shutdowns.
Hagglunds hydraulic motors were coupled directly to the process equipment avoiding the
requirement of gear boxes associated with electromechanical drives. When finally
installed near Manchester, the power of the drive system exceeded 2 MW - the largest
direct drive installation in the UK (and possibly Europe). The key feature of the Hagglunds
hydraulic drives is their ability to deliver very high torque at low speed and throughout the
speed range.
Start-up is particularly critical. At high viscosities, an electromechanical drive will
apply so much effort to start the mix that there is a very real risk of damage being suffered
by the gearbox or of mixer blade shear. Another advantage of the hydraulic drive is that
at high torque it does not trip, but holds the torque indefinitely, The pump senses this and
will 'back-off' so as not to introduce unwanted heat into the drive. When an electric drive
trips, the process stops completely. The pressure in the drive is also an indication of what
is happening within the reactor or the process pump, and alarm signals are generated if the
pressure is abnormal. [Process Industry Journal May 1994J
THE HYDRAULIC ADVANTAGES 13
Example of a hydraulically operated aircraft gangway

For many years it has been the tradition to use electro-mechanically operated telescopic
carriers to manipulate aircraft mobile gangways. However, Mannesmann Rexroth were
able to show that using hydraulics was more than a convincing alternative - it had positive
advantages. One of the main issues was the safety aspect both to passengers and aircraft;
another was an ability to manoeuvre the gangway by hand should a power failure occur;
but perhaps the most important for a busy airport (Munich II) was the speed of operation
if the maximum passenger flow was to be achieved.
Both the transport aspects (including steering) and the raising and lowering of the wheel
driven tubular passenger structure were hydraulically operated, transmission being
provided by a multiple pump comprising: two variable displacement (swashplate) axial
piston pumps, each independently controlled so that they can independently drive a wheel
motor; one gear pump for lifting; a combined drive motor and power pack mounted on the
cross-beam ofthe vertical supports; and, located in the same place, the oil reservoir with
filters and monitor. A number of ingenious designs are incorporated in the total system;
should the power fail, a hand pump may be used with series of stopcocks and check val ves
enabling the gangway to be lowered in a controlled manner, and the brakes to be released
for the unit to be towed. [Drives and Controls March 1994]
Example of a drive for a travelling saw

The conventional method for driving a travelling saw would usually be by electric motor.
In the particular application - on a continuous non-ferrous casting plant - this would have
been possible but would have required closed loop control due to the complication of two
types of saw blades which were to be used, i.e. HSS and HM tips. The drive power of lOa
kW would necessitate a redesign of the existing plant due to the moved mass and the
dimensions of the drive.
More acceptable solutions are provided by hydraulics. For instance, it would be
possible to use a hydrostatic, closed loop drive with primary adjustment of the pump - the
speed being adjusted by changing the displacement of the pump. However, the operating
pressure between pump and motor changes as a function of the torque at the saw blade
(which can occur due to eccentricities) and therefore in the hydraulic motor. This idea can
lead to considerable variations in the speed of the sawblade. It would also be possible that
there would be a lower loop gain and hence a tendency to oscillate.
Another solution, which was used, was that of using a hydraulic drive controlled by a
secondary pump unit with a digital speed sensor. This works as a speed controlled drive
with constant operating pressure and the torque fluctuations are compensated for by
analogue adjustment of the motor's swept volume. The stiff drive concept keeps speed
fluctuations to a minimum. The size of this unit was conveniently fitted with no major
changes to the plant, and the entire range of cutting speeds was steplessly controlled while
maintaining a highly constant angular velocity. A VT12000 control unit was used for the
control and monitoring electronics, with command values being provided by a PLC
[Rexroth Information Quarterly 2/1994].
15
PROPERTIES OF FLUIDS
AN HYDRAULIC fluid has several functions to perform. For instance it must

transmit power efficiently
lubricate the moving parts
provide a means of cooling
carry away contamination, and
(with water) provide microbial degradation
The effectiveness of each fluid to perform these functions is determined by both the design
of the system and the basic fluid properties. There are well over 20 different characteristics
of fluids and their importance varies with the application. Initially the basic fluid
properties are examined (including a discussion of the Newtonian aspect), followed by a
range of other features of less general interest, but possibly of great specific importance.
The basic fluid properties are
viscosity
density (or specific gravity if compared with water),
specific heat
compressibility
Other features of interest to the fluid power user are
vapour pressure
surface tension
thermal expansion
thermal conductivity
aniline point
pour point temperature
cloud point temperature
flash point temperature
fire point temperature
spontaneous ignition (or auto ignition) temperature
air release value

water separability
neutralisation value
resistivity and conductivity
as well as such items as good lubricity, filterability, foam resistance, demulsification time,
and wear and corrosion resistance, and, of course with the water fluids, inclusion of a wide
spectrum biocide. Compatibility is another essential item as regards the metals and seals
used in any system. Most of the features may be adjusted by the use of additives; these will
vary from one oil manufacturer to another, who, generally will keep their precise content
a closely guarded secret.
Newtonian or non-Newtonian?
A Newtonian fluid, as with most mineral oils, is one which maintains its viscosity no
matter what the shear rate may be. A non-Newtonian fluid, like an emulsion, changes its
viscosity with the rate of shear; thus where the shear rates are high, e.g. between the tips
of vanes and the ring in a vane pump, the viscosity of an emulsion falls significantly. For
this reason it is necessary to use an emulsion of a slightly higher viscosity than a mineral
oil for the same application. This is advantageous in that leakage is reduced.
Viscosity
The viscosity of a fluid is a measure of its internal resistance to flow. Dynamic (Absolute)
viscosity (11)is defined in terms of the force between two parallel laminae or layers of fluid
at a certain slip velocity between them. The unit of dynamic viscosity is newton second
per metre squared (N s m-2 or Pa s). The non-SI, but acceptable, unit the poise is 0.1 N s
m-2 Because the dynamic viscosity of real fluids determined in poise is invariably a
fractional quantity, the more usual unit employed for expressing dynamic viscosity is the
centipoise (cP), or one hundredth of a poise. The significance of dynamic viscosity is that
it is effectively afriction coefficient.
For engineering calculations it is usually more convenient to employ kinematic
viscosity (v) rather than dynamic viscosity, this being determined as the absolute dynamic
viscosity divided by the mass density of the fluid. The SI unit is thus metre squared per
second (m2 S-l). The derived non-SI unit is the stoke (St) but for the same reason as above
the practical unit is invariably taken as a centistoke or one hundredth of a stoke (cSt).
Kinematic viscosity is used for the calculation of flow characteristics, and thus dynamic
pressure. (Mineral oils have a considerable range of viscosity, from perhaps 5 cSt, to over
10 000 cSt at sub zero Celsius temperatures.)
Other practical viscosity values

For consistency (the whole reason for having SI units) all fluid viscosity values should be
quoted in centistokes. If, however, fluid viscosity is quoted in some other arbitrary scale,
such as Redwood No 1 seconds, Saybolt Universal Seconds (SUS), or Engler degrees, it
should be understood that there are no exact conversions between these scales, nor for
conversion of arbitrary viscosities in seconds or degrees to kinematic viscosities in
centistokes, fe/sec, or in2/sec units. An approximate conversion can be made by reference
to conversion scales or conversion tables, but again care should be exercised because such
conversions apply only at the same temperature as the original measurement.
In the case of a non-Newtonian fluid the instantaneous viscosity ofthe fluid is dependent
00 the shear stress in that fluid at that particular moment. If necessary, a specific viscosity
figure can be obtained with a viscometer which ensures a uniform shear rate throughout
measurement. Such a figure will, however, have limited practical value, unless the shear
stability characteristics of the fluid are also known.
The variation of viscosity with temperature is one of the most significant parameters
with hydraulic fluids, affecting both the performance and selection of a fluid. This can be
fully expressed by plotting a characteristic curve for the fluid on a chart with a 10g.1og
kinematic viscosity scale versus log absolute temperature (Figure 1.4.1). Such scales
cause the characteristic curve for most fluids to be linear. Given a number of spot readings
for viscosity and temperature, a close approximation to the viscosity-temperature charac-
teristics of that fluid at intermediate temperatures can be obtained by joining these points
with a straight line, and for temperature outside the range covered by the spot values, by
extending this line in either direction.
ISO viscosity classification

Viscosity classification for industrial liquid lubricants defined by BS 4231, and later ISO
3448, establishes 18 viscosity grades in the range 2 -1500 cSt covering approximately (in
the case of mineral oils) the range from kerosene to cylinder oils and thus also embracing
the normal range of mineral-oil-based hydraulic fluids. Classification is based on the
principle that the mid-point kinematic viscosity of each grade should be approximately
50% higher than that of the preceding one. Using this numbering system, oil viscosities
are quoted as ISO viscosity grade (number), or ISO VG (number)-see Table 1.4.1. Figure
1.4.2 shows a selection of kinematic viscosities at various temperatures with different
Viscosity Indices.
PROPERTIESOF FLUIDS 19
Viscosity Index
Viscosity Index (VI) is a single number representation of the viscosity temperature
characteristics of a fluid. The higher the viscosity index the smaller the change in viscosity
with temperature, and vice versa, although this is only a general guide as to actual change.
Most mineral oils will have VI values around 100, but special oils are made with an
Extended Viscosity Index in the high lOO's to over 200. Very low VI's can also occur-
to possibly 15. Examples are indicated on Figure 1.4.1.
Density
The density of a fluid is defined as the mass per unit volume, namely,
A typical value for a mineral oil would be 870 kg m-3, whilst a synthetic fluid could be
1200 kg m-3
Specific gravity is a dimensionless quantity and is the ratio of the density of a fluid to
the density of water. For the mineral oil again, the typical value would be 0.87. In the case
of equations for engineering calculations it is often desirable to state density as a factor,
in terms of specific gravity, thus avoiding any possible confusion between the true
numerical values of density which are to be employed in the formula.
The significance of specific gravity as a hydraulic fluid parameter is that it gives an
indication of the weight of the fluid in the system, or more directly a comparison of fluid
weights for a given system where different fluids may be considered. Also the higher the
specific gravity of the fluid, the more difficult it is to lift the fluid in the suction part of the
system; the design ofthe suction side may therefore need particular care in order to avoid
the possibility of cavitation and erratic pump operation.
Specific Heat
The specific heat of a fluid is a measure of the amount of heat absorbed by a mass of fluid
when raised 1C. Its units are thus kj kg-1C-I The specific heat is not constant but varies
with temperature. For practical calculation a constant value is often assumed based on a
nominal temperature range, although most oil companies will quote the specific heat at
definite temperatures, such as 20e.
The actual specific heat of a hydraulic oil may be appreciably modified by the presence
of additives, and also of contaminants in the fluid.
Compressibility of Fluids
Unlike air, hydraulic fluids are considered as 'incompressible' . However this is not strictly
true although the compressibility is much less than air. Compressibility is the reciprocal
of the fluid bulk modulus (B) - bulk modulus being the ratio of the pressure change to the
volumetric strain in the fluid (like Young's Modulus for solids). However, the bulk
modulus is not constant for liquids, tending to increase with temperature and decrease non-
linearly with pressure - see Figure 1.4.3.
The instantaneous value of the bulk modulus at any pressure is called the tangent bulk
modulus (BTAN), and the mean value of the bulk modulus from atmospheric pressure to any
pressure P is called the secant bulk modulus (BSEc)
where Vo = volume at atmospheric pressure
The values of the tangent and secant moduli will tend to coincide at lower pressures (i.e.
as P approaches atmospheric pressure), and for pressures up to about 7 MPa (70 bar) the
PROPERTIES OF FLUIDS 21
difference can usually be ignored. For general engineering calculations a 'typical' bulk
modulus value may be quoted and used for pressures up to 70 MPa (700 bar). Logically
this should be the secant modulus. For working at specific high pressures, however, the
tangent modulus should be used, if known (Table 1.4.3). (See also Table 1.4.4).
Typical working figures are a reduction in initial volume of 0.5% per 7 MPa (0.00735%
per atmosphere) for hydraulic oils and 0.4% per 7 MPa (0.0059% per atmosphere) for
water. Such figures are reasonably valid for pressures up to 70 MPa (700 bar) and over
a temperature range of 10 - lOOoe .
The compressibility of a normal hydraulic fluid at 200e and 7 MPa (70 bar) can also be
estimated quite accurately from its kinematic viscosity at 22e , using the empirical
formula
'compressibility' = 0.0129 - 0.000737 log v (mm3 N-1 )
where v = viscosity (cSt).
Alternatively, 'compressibility' may be expressed in terms of relative density, namely.
As a direct result of compressibility, the density of any real fluid will increase with
pressure. In very high pressure systems this may be more significant than the volumetric
change. It should be noted that the presence of free air in the oil, and any flexibility in pipe
wall, will change the apparent compressibility.
Vapour Pressure
The vapour pressure of a fluid is the pressure exerted by the saturated vapour in contact
with the surface of the fluid at a specified temperature. The higher the vapour pressure the
more volatile the fluid, and/or the nearer it is to boiling point. Fluids with a high vapour
pressure, either due to their volatile nature or because of a high operating temperature, are
therefore prone to 'flash' into vapour under suction conditions, thus setting a specific limit
to the net positive suction head a pump can accommodate without cavitating.
Additionally, as the boiling point of a fluid approaches, the more volatile fractions will
come off first, progressively changing the nature of the fluid. This is seldom significant
in oil fluids at normal working temperatures, but with water fluids progressive loss of
water may be experienced at quite moderate working temperatures.
Surface Tension
Surface tension may be significant, affecting:
(i) foaming characteristics at the free liquid surface or interface between two non-
miscible fluids;
(ii) the ability of the fluid to 'wet' a metal surface;
(iii) inherent leakage past seals and at joints, etc.
Characteristics (i) and (ii) can be adequately controlled by additives if necessary. Item
(iii) is not normally significant in the case of oil fluids which have adequate surface
tension to make sealing relatively easy. Fluids which have a surface tension of less than
0.03 N mm-2, however, are troublesome to seal, needing particular attention to joints, and
one must often accept that some leakage will be inevitable with practical designs of seals.
Thermal Expansion
The coefficient of volumetric expansion of an oil remains practically constant over the
usual range of working temperatures encountered in hydraulic systems; 0.0006 per C is
a typical value. Differences in value are largely related to the specific gravity of the fluid.
The coefficient of volumetric expansion, however, decreases rapidly with increasing
pressure, e.g. of the order of 0.000025 per C per 7 MPa (70 bar). The relative volume of
a pressurised fluid is thus less than that predicted on the basis of volume correction
coefficients alone.
Thermal Conductivity
The thermal conductivity is a measure of the ability of a fluid to dissipate or transfer heat.
In a practical system heat dissipation may be hindered by the formation of boundary layer
films and thus fluids which do not 'wet' the internal surfaces tend to have lower thermal
conductivities. The lower the thermal conductivity of the fluid the higher its working
temperature will tend to be, under similar operating conditions. Thus mineral oils, having
a generally low thermal conductivity, will tend to run at higher working temperatures than
water-based or water-glycol fluids used in a similar system. The higher the electrical
conductivity of a fluid, the higher the thermal conductivity is likely to be.
Thermal conductivity may be correctly quoted in either
Aniline Point
The aniline point of a mineral oil is the lowest temperature at which the oil is completely
miscible with an equal volume of freshly distilled aniline. It is a general indication of the
aromatic content of the oil (paraffinic oils having a high aniline point and aromatic oils a
low aniline point), and because of this is sometimes used as a form of compatibility index.
The aniline temperature is likely to be around 100C,
Pour Point
This is the temperature at which the thickening action of separation of the waxy
constituents is so marked that the fluid ceases to flow. On the viscosity curve this would
be marked by an abrupt rise in viscosity. The pour point (temperature) is determined with
regard to specific flow conditions. For heavy oils a typical figure could be as high as 4C,
whilst for fluids required in Arctic conditions it may be as low as minus 50'C, or even
minus 70C with a silicone fluid.
Cloud Point
The cloud point of a mineral oil is that temperature at which waxes or other solids normally
present in solution tend to crystallise out, or come out of solution. This can lead to clogging
or partial choking of the system.
Flash Point
The flash point temperature is the temperature above which the fluid will ignite in the
presence of an open flame. The flash point broadly defines the relative fire hazard of a
fluid. The flash point may also serve as an indication of the type of an oil or blend, since
the more volatile the oil the lower the flash point and vice versa. The flash point may be
determined by various standard 'open' or 'closed' cup tests. These tests also indicate the
tendency of the fluid to ignite when brought into contact with a hot surface. They are
particularly relevant where spills may leak on to hot exhaust pipes, etc. The flash point
temperature in general will be between 1DOoC and 300C,
Fire Point Temperature
The fire point temperature is the temperature above which the fluid, once ignited, will
continue to burn on its own. This is likely to be some 10 - 30C above the flash point
temperature, or maybe, 50C for phosphate esters.
Spontaneous Ignition Temperature - SIT

Auto Ignition Temperature - AIT
This is the temperature at which a fluid will ignite spontaneously without the presence of
any other flame. It is a measure of the tendency of the fluid to withstand overheating and
contact with hot surfaces. This is considered to be one of the most relevant tests for
hydraulic fluids.
Conventional hydraulic oils vaporise at relatively low temperatures and ignite with a
sharp report. The water containing types of water-in-oil emulsions and water-glycol
mixtures form a steam blanket which must be dispersed so that extremely high tempera-
tures are needed for spontaneous ignition.
Phosphate ester fluids, on the other hand, with their innate resistance to fire, cannot
support combustion but will ignite spontaneously at very high temperatures, of the order
of 600C, silicones, perhaps, 500e. The behaviour of the various fluids in this test differs;
interpretation of the results is difficult and not amenable to tabulation.
Air Release Value

Mineral oils usually contain between 7 - 10% by volume of air in solution. It is, however,
the free air (in the form of tiny bubbles) which needs to be released from the oil into the
atmosphere. Should the air be retained, the compressibility increases and oxidation of the
oil would be encouraged. Air release is measured by determining the number of minutes
taken to release 0.2% air from the liquid held at 50e.
Water Separability
Water also needs to be released from the oil, by separation. This allows it to be drained off.
Water in the oil has similar effects to air, and may also encourage greater corrosion. On
the other hand, for water emulsions, it is important that the water remains in solution and
the fluid is stable.
Neutralisation Value
It is important to know the acidity or alkalinity of an hydraulic fluid both with reference
to the life of the fluid and the effect on the system components. If acids are likely to be
formed in the system, then a high initial alkalinity is advisable; conversely, if salts are
likely to produce a rising alkalinity, then some initial acidity level would help. However,
too great a level in either direction could cause corrosion in the system.
The Neutralisation Value or acidity value (the Total Acid Number- TAN) is obtained by
the mg measure of how much potassium hydroxide alkali has to be added to neutralise I
g of the fluid. The alkalinity value (the Total Base Number- TBN) is obtained by the mg
measure of how much potassium hydroxide alkali just neutralises the amount of acid
necessary to neutralise 1 g of the fluid.
Resistivity and Conductivity
It is generally considered that spool valve corrosion/erosion can be badly affected by solid
particulate and chlorine content in the fluid. In addition to reducing these levels, it is also
considered that a high resistivity improves the situation. Levels of the order of 50 MW m
are likely to eliminate such electro-kinetic effects. However, in practice it may be difficult
to maintain such a high level due to the increased moisture content. It should be pointed
out that resistivity (r) is the inverse of conductivity (s). The units are illustrated in the
following equations, in reference to a volume of fluid
Corrosion resistance, as well as fire resistance, are also important 'resistivity' properties
for liquids in certain situations. This particularly applies to mining.
Tests
The Institute of Petroleum has a number of standard tests which are used to provide
consistent results for each of the properties mentioned above. These are listed in the
chapter on Standards.
27
BASIC HYDRAULIC THEORY
HYDRAULIC SYSTEMS are a combination of hydrostatics and hydrodynamics. Hydro-

statics relates to stationary fluids where power is transmitted by pressure energy.
Hydrodynamics relates to moving fluids where forces are generated by motion. Thus in
practice a mixture of the two may be present; for example, actuators work on a hydrostatic
principle, whilst flow through the pipelines to the actuators conform to hydrodynamic
laws.
Hydrostatics - stationary fluids

In a purely gravitational hydrostatic system where all pressures are relative to atmospheric,
a static pressure is produced by virtue of the mass of fluid supported by a base area;
the pressure (P) being dependent only on the height of the column of liquid (h) and its
density, (p)
P=hxpxg
where g = acceleration due to gravity,
In the case of fluid in a container (e.g. a tank or reservoir), hydrostatic pressure exerts
a force F on the base area:
F=AxP
where A = base area.
The pressure at any point within the fluid is determined by its depth below the surface
of the liquid, or head (h).
The pressure at any point in a static fluid is the same in every direction. The force exerted
on any surface wetted by the fluid is thus equal to the product of the fluid pressure (at the
appropriate depth) and the surface area.
A closed system, or confined fluid, is described by Pascal's Hypothesis which states that
if a force is applied to any part of an enclosed fluid system, the extra pressure resulting is
the same throughout the whole of the enclosed fluid system; it also acts perpendicularly
to all surfaces in contact with the fluid. Although, in theory, the weight of fluid adds to the
28
pressure of the lower parts, for almost all practical hydrostatic applications this is
considered so negligible as to be completely ignored.
Since forces exerted on surfaces by hydrostatic pressure are proportional to areas, force
multiplication is possible by using, for instance, pistons of different area (see Figure 1.5.1).
Here F 1applied to piston Al develops a pressure ofF 1/Al which is transmitted throughout
the fluid. Thus the resulting pressure on piston area A2 is (FI/AI ), or:
with a force multiplication of A2 / Ai .

This is a valid hydrostatic system since the displacement of fluid is small enough for
velocity loss to be neglected. In a practical system, modifying factors, such as the weights
of the respective pistons (M1g andM2g) and the friction of the piston seals (fl and f2), may
be included - see Figure 1.5.2. The true output would then be calculated as:
BASIC HYDRAULIC THEORY 29
Respective piston travels are not modified in any way (provided no fluid leaks occur)
and are:
These equations define the performance in a basic hydrostatic system of force

multiplication over pressure ranges where the compressibility of the fluid is negligible.
Not all such systems are necessarily force multipliers, however. They are equally suitable
for force transfer with high efficiency. Thus the typical automobile brake system may, in
fact, employ slave cylinders of smaller diameter than the master cylinder. All the force
multiplication requires is applied at the mechanical output, where a maximum pedal
pressure of the order of450 N with amastercylinderof25 mmdiametercan yield a system
pressure of the order of 4.2 - 5.6 MPa (42-56 bar), with virtually 100% force transfer for
the slave cylinders.
Pumps in hydrostatic systems

Where a hydraulic pump acts as the transmitter in a force transfer system, the piston is
'sized' on the basis ofthe output force required and the fluid pressure available from the
pump, i. e.
where =
S stroke
t = time to complete stroke
Pump output power required then follows as:
power out = F S / t .
It is assumed here that fluid velocity components are negligible, i.e. the system is truly
hydrostatic. (The theory is true for low inertia systems; the acceleration time for other
systems is discussed later.) In practical systems it may be necessary to take into account
velocity components and back-pressure effects; also seal friction and, in the case of large
vertical pistons, the piston weight.
Hydrodynamics - moving fluids

Fluid under pressure will always tend to fill a pipe and for all practical purposes in
engineering hydraulics, pipes will always be filled by pressurised flow. The velocity of
flow through pipes of changing or variable cross section will then automatically adjust to
maintain continuity of flow. Thus at a change of section (Figure 1.5.3):
where p = fluid density.
Equally the fluid will also be subject to some static pressure generating the flow (unless
it is flowing down an inclined pipe under gravity); and possibly potential pressure (due to
a change in weight along the length of flow).
According to Bernoulli's principle, the total energy of a fluid is constant. It follows that
the sum of the static pressure, dynamic pressure and potential pressure (neglecting
frictional losses) is constant for any flow line cross section.
In most practical systems potential pressure can be neglected, assuming simply that
static pressure falls with increasing velocity pressure (i.e. increasing flow velocity) and
vice versa.
Laminar or turbulent flow

Whether a fluid is turbulent or laminaris determined by the Reynold's Number Re. This
dimensionless quantity is given by
where v = mean fluid velocity (m S-I)

d = pipe hydraulic diameter (m)
v = fluid kinematic viscosity (m2 S-I)
The hydraulic diameter can be taken as the pipe diameter in most cases, but where the
bore is non-circular an estimate can be made using
The exact value of Reynold's number at which a change occurs from laminar to
turbulent will ultimately depend on the precise pipe characteristics. In order to avoid this
unknown transition region, and be confident in the flow characteristics, it is normal to state
that
Reynolds number below 1500 - Laminar flow
Reynolds number above 3200 - Turbulent flow.
The pressure drop in a pipe, because of frictional losses, may be expressed as
layers of fluid. The pressure drop is thus dependent on both the Reynold's number and the
smoothness of the pipe bore. Although empirical approximations have been devised, it is
usual easier to read the friction factor value, 'ft' off a chart such as a Moody diagram - see
Figure 1.5.4. The Moody diagram includes an expression for relative roughness (kid),
typical values of 'k' are given in Table 1.5.2 for pipes and tubes measured in mm. (The
table gives a number of non-hydraulic pipes for completeness.)
generally applicable to gases than liquids. Thermal terms are generally ignored when
dealing with liquids, when an additional term is introduced to take into account the
conversion of some mechanical energy into thermal energy due to viscous friction. A
general expression of the energy equation can be reformed to become:
In a practical system the head loss is always positive and in the direction of flow. In a
hydrostatic system work is added to the system (e.g. by the pump). In the case of a
hydrokinetic system work is both added (by the pump) and removed (by the hydrokinetic
machine).
Efficiency
The efficiency of a hydraulic system is not easy to define, and may be even harder to
determine. The mechanical aspects of the system (which involve hydraulics in some form)
may be tested, and a combined efficiency calculated. The efficiency will vary consider-
ably depending on the conditions - viscosity, etc. - and hence it may only be meaningful
at the design stage and operation. This would be the best achievable efficiency. At any
other condition, the efficiency is will be less.
Three efficiencies are normally considered-
Mechanical efficiency
Volumetric efficiency and
Overall efficiency
The overall efficiency of a system is the comparison of the input power required to drive
the system, to the output power generated. (Output power being measured only in those
aspects which are considered useful- work done, rather than heat generated, unless the
purpose of the system is to provide heating.)
Mechanical efficiency of a power component is calculated by comparing the input
power required to the particular component, to the output power coming from that
component. (Again this relates to the useful aspects of the mechanism.) Thus, for instance,
for a pump, an input torque would be considered, and compared with the theoretical torque
required to produce a certain pressure at a certain speed, thus
being compared with the actual volume being displaced. Because the losses are usually
due to slip and leakage, the efficiency tends to drop with pressure; at optimum design
efficiencies between 95% and 98% would be expected. Pump and motor efficiencies are
the inverse of each other, thus
In the case of actuators, the efficiency can be examined by combining the actuator
effects with the pump providing the supply. Thus the force / velocity emanating from the
actuator is compared with the mechanical input to the pump. The actuator has an
intermittent operation, and so a particular part, or the total part, of the cycle must be chosen.
These efficiencies will be discussed more fully in the chapters concerned with the
specific components.
Flow through changing cross section and orifices

How rate is modified when a fluid passes through a change in cross section. Although this
may be a regrettable energy loss in many cases, it can also be utili sed in flow control.
Control valves employ many different orifice configurations, each design having a special
effect on the flow.
Losses in flow energy through changing cross sections are largely due to separation,
although frictional losses will also be present. Such losses, as a head (hL), are most
conveniently expressed in terms of the velocity head (v2/2g) at the downstream end, where
v is the downstream velocity. Thus
This is valid for all general cases except that for expansion of cross section. The
effective velocity value is the difference between upstream and downstream flow
velocities. In most orifices employed in fluid power systems, the upstream area is much
greater than the orifice area, and hence approach velocity effects are negligible.
Typical values of Kdor entries and exits are given in Fig 1.5.5. In the case of a sudden
contraction (Figure 1.5.6) the loss coefficient is a function of the diameter ratio d2/d], as
shown. This can also be derived mathematically as:
41
THE HYDROSTATIC PRINCIPLE

AND BASIC CIRCUITS
The previous sections have outlined the basic principles appropriate to the operation of
hydraulic systems. However, the application of this theoretical background to the most
common pump supplied systems is worth further consideration and explanation.
A linear circuit
To illustrate these principles a simplified circuit is shown in Figure 1.6.1 using standard
symbols (see Notation and Symbols for a summary of the main symbols used). This circuit
shows a pump taking fluid from a tank and supplying it to a single acting cylinder via a
directional control valve. There is also a pressure relief valve connected between the pump
outlet and the tank. The symbols do not give any detail of the construction of the
components, they only describe their function within the system. The details of the
component construction which could provide these functions and their principal alterna-
tives are described in Section 2. However, at this stage their function can be understood
and hence the operation of the circuit can be deduced.
The symbol indicates the type of pump used almost exclusively in hydraulic systems,
a positive displacement pump. The term positive displacement means that it moves a fixed
volume of fluid for each rotation of its input shaft irrespective of its internal construction.
If it is rotating at constant speed then the pump will give a constant flow (strictly a
volumetric flow rate) output. The symbol Qp is used to indicate the flow from the pump
which is given units of say L/min. This flow can be related to the size of the pump and its
speed by:
where Np = the pump speed

Dp = the pump volumetric capacity
The directional control valve symbol in Figure 1.6.1 shows a 3 position, 3 way valve. Each
box within the symbol denotes an alternative position for the valve and the external lines
show the port connections giving the number of "ways" flow can be directed. As shown
the valve is in its centre position, the box indicates that the pump flow is being directed
back to the tank, and the cylinder connection is blocked. If the top box of the valve is
selected then the reader must imagine this box occupying the position between the external
lines and then the valve will direct the pump flow to the cylinder.
The flow into the cylinder will cause the piston to move the load mass M upwards. If
the lower box is selected the piston can move downwards and return to a lower position
as flow returns to tank. However, this setting of the valve position does not stop the pump,
and it will still produce a flow which must be provided with a suitable flow path to prevent
damage. In the circuit of Figure 1.6.1, the flow can return to tank via the relief valve. This
is a valve which will open automatically when the pressure on the upstream side reaches
a pre-set level PMAX. A more detailed consideration of this circuit in terms of the up and
down movements of the piston is given below.
Piston upstroke
There is a simple relationship between the upward movement of the piston and the flow
supplied by the pump. After a transient period of acceleration the piston will move
upwards with a steady velocity Vel, which is related to the pump flow by the piston area
A.
The larger the pump flow then the more rapid will be the piston velocity. The larger the
piston area for a given pump flow then the lower will be the piston velocity. This
relationship is true provided that all the pump flow passes to the cylinder and there is no
additional or alternative flow path e.g. through the relief valve. Fluid will always find the
easiest flow path and in this context easiest refers to the path of least resistance or lowest
pressure. To establish that this is true requires some consideration of the pressures in the
circuit. This circuit operates on the same principles of hydrostatic pressures and force
balance that were described in Section 1.5 above, although this may not be immediately
apparent. If the piston is supporting a mass M, having a weight of Mg, then this must be
balanced by an equal and opposite force on the piston produced by pressure in the fluid.
'The subscript E indicates the "extend" stroke ofthe piston, and is used to be consistent with later sections and
a general system of notation.
THE HYDROSTATIC PRINCIPLE AND BASIC CIRCUITS 43
Mg = piA
where PL = the fluid pressure balancing the load
A = the piston cross-sectional area
This is true in the steady state Le. when the mass is not accelerating, possibly the
majority of time even if the mass is moving. An additional component of force will be
required when the mass is accelerating upwards and similarly a decrement when it is
decelerating to rest. If these cases and friction are neglected at this stage then the pressure
is determined wholly by the load supported on the piston i.e. the mass M.
To emphasise this point the magnitude of pressure in the system is established by the
cylinder and its load, NOT by the pump. The pump provides the input effort, in terms of
a flow to cause movement in the system but does NOT set the pressure level required for
movement.
If the load is increased such that the pressure required is higher than the setting of the
relief valve then the flow path through the relief valve will become the lower resistance
and the pump flow will take this route. This will act to limit the maximum pressure in the
system to protect against damage to components or seals. This condition will also define
the maximum force available from the piston, often called the stall thrust. In fact, in the
simple circuit shown there would not be a force balance across the piston in such
circumstances, and flow would come out of the cylinder as discussed below, and the piston
and load would fall in this condition. There are simple additions which could be made to
the circuit which would prevent this in practice.
Piston downstroke
A closer examination of the conditions when the valve is selected to lower the piston will
show a similar effect, however there are now two separate parts of the circuit- the return
flow side and the pump side, and these are considered in turn.
The piston pressure must be the same value of PL, if the mass is the same and
acceleration and friction effects are still neglected. This pressure is now the dri ving effort
causing flow in this part of the circuit. The size of the flow will be determined by the flow
restrictions between the cylinder outlet and the tank. As shown in Figure 1.6. I, the only
restrictions will be from the pipe work connecting the components and the directional
control valve. These will usually be sized to give a low pressure drop and not restrict the
outflow, so the piston can descend very rapidly. It would be normal to include extra valves
in the cylinder connection to provide a restriction as shown in Figure 1.6.2.
There are two extra valve components in Figure 1.6.2 - a non-return valve, usually called
a check valve; and a variable orifice restrictor valve sometimes called a needle valve (the
oblique arrow symbol indicates that the orifice area is adjustable).
The check valve in Figure 1.6.2 allows flow to pass freely in the direction feeding the
cylinder but does not allow any flow in the reverse direction. The restrictor valve offers
a similar flow restriction in both directions and its characteristic is usually described by
the orifice equation introduced in Section 1.5 above. When flow is being supplied to the
cylinder it will pass preferentially through the check valve and there will be little effect
on the system performance. However, when the piston is descending the flow must pass
through the restrictor and this will be adjusted to be the most significant restriction to the
return flow, in comparison with the pipe work or the directional control valve. This
component thus provides the dominant pressure restriction, and the return flow rate to the
tank will depend on this and the load pressure. Thus the return flow QR can be related with
an equation for orifice flow:
On the pump side, as has been stated above, the pump output flow must be provided with
an acceptable route out of this part of the circuit. If this is not provided then the fluid will
find an unacceptable route by making a hole in the system or else the pump and its driver
will stall. Thus in this circuit the flow can only pass through the relief valve, and this will
set the pressure on the pump side of the directional control valve. When the setting on the
relief valve is reached, the valve will open and allow the flow to pass back to tank. The
relief valve is constructed so that it only opens sufficiently to let this flow pass via a
restriction and hence it maintains the pressure in this part of the circuit at the maximum
level.
Note again that it is not the pump which has set the system pressure. In both the above
cases it is the component which has been supplied with the pump flow which sets the
pressure. The common feature is the resistance experienced by the flow from the pump,
either from the cylinder and load or from the relief valve, and in all cases it is this which
establishes the working pressure level.
A numerical example
A simple numerical example will help in the use of typical units. A mass of 1700 kg is
supported by a cylinder with a piston diameter of 40 mm. The cylinder is supplied from
'The subscript R is used to denote the "retract" stroke of the piston.

THE HYDROSTATIC PRINCIPLE AND BASIC CIRCUITS 47
also be the power transmitted to the load. Losses of pressure in pipe work and valves
represent sources of power loss. It is again important to realise that these pressure losses
will mean that the pressure at the pump must be higher than the load pressure by this
amount i.e. the load pressure must still balance the load force or torque. Additional losses
will occur due to frictional forces in the actuator, or motor.
Pumps are not the only sources of power in hydraulic circuits. External forces on load
actuators may act to return energy to the hydraulic system. If this energy can be harnessed
rather than dissipated then the overall energy efficiency of the system can be improved.
There are a number of techniques where energy recovery through regeneration or
secondary systems can be used to achieve this and these are discussed in later sections.
Further consideration to the effects of these losses, the effects of friction and mass
acceleration forces are given in Section 4, together with discussion of the transient effects
in circuits.
SECTION 2
a certain fixed volume of fluid (the ideal volume) is transferred to the hydraulic system;
the actual volume depending on the volumetric efficiency at the particular fluid/pressure
characteristics. Piston pumps, gear pumps and vane pumps are the most common types of
positive displacement pump; screw pumps are occasionally used.
Although gear pumps may be the simplest form of pump design, piston pumps compete
with them as regards main power units. Piston pumps include in-line, vee and other multi-
cylinder configurations; radial piston pumps with cylinders disposed radially around a
cam crankshaft; rotary piston pumps where the cylinders are rotated around a fixed shaft;
and axial piston pumps with cylinder reciprocating motion being caused by a swash-plate.
Generally pumps are classed as either 'fixed displacement' or 'variable displacement' .
In the case of the variable type, the output per revolution is varied by a control mechanism,
which in more recent versions has been an integral electric hydraulic feature capable of
controlling pressure, flow or power.
In deciding the appropriateness of a pump there are a number of features which need
to be considered. Obviously the pump output and maximum pressure ability are the prime
consideration, but also regard must be taken to the need, or otherwise, of variable output,
whether it will be in open or closed circuit, the dimensional size and weight, and the cost.
It may also be a requirement to operate in the dual capacity of pump or motor. Noise has
become a critical factor, and major design features are included in some gear and vane
pumps to reduce the levels. Fluid compatibility is also essential. A positive head at the
pump inlet may also be required.
A summary of hydraulic pump types and typical maximum operating parameters are
given in Table 2A.I.I.
Piston pumps
Piston pumps offer high volumetric efficiencies together with virtually no limit on
capacity, and thus cover a wide range of delivery requirements. Because of the greater
complexity of construction, however, they are seldom competitive in smaller sizes with
gear or vane pumps unless high system pressures are required. In this respect they are
superior to all other types of pump, although the pressure rating of a piston pump is
governed by the types of valve which can be employed with the design. Configurations
relying on porting, sliding or rotary valves are limited in the maximum pressures they can
develop; in general, the higher pressures can only be achieved with seated valves.
In-line piston pumps

Multi-cylinder pumps of in-line configuration are generally robust, large units designed
for long life applications requiring high pressures and large deliveries. They are consid-
erably larger than most other designs and therefore, are invariably of horizontal configu-
ration. Apart from the fact that they are capable of developing the highest pressures of any
type of hydraulic pump, volumetric efficiencies in excess of 97% can be realised, an
important consideration when large volumes of fluid are being handled. Speeds are
generally restricted to 1()()- 600 rev/min, although so called high-speed pumps of this type
are rated for continuous running up to 1500 rev/min. Smaller in-line pumps are either of
horizontal or vertical configuration, and may be designed for higher operating speeds.
HYDRAULIC PUMPS 51
Individual designs may differ in detail, notably in the method of reciprocating the
cylinders. These may be directly driven by a crankshaft and connecting rods, or displaced
by cams or push rods and returned by springs. Pistons are plain, working with fine
clearances in the bores, with suitable packing seals. Valves are invariably of the
contaminant insensitive seated type (e.g. poppet valves) and normally operated automati-
cally by pressure difference; both conical seats and flat faces are used.
In-line pumps lend themselves to adaptation for handling fluids with low lubricity. In
this case a separate pressure lubrication system may be incorporated to supply the
requirements of the pump unit, particularly the bearings.
Figure 2A.I.I shows the cross section of one example.
Radial piston pumps

The radial piston pump is a more compact type capable of a similar performance to an in-
line piston pump. It also lends itself to variable-deli very performance. Pumps of this class
embrace both radial (fixed cylinder, rotating cam plate) and rotary designs (fixed cam
plate, rotating cylinder block); these are described below.
As-with the in-line pump, the fine clearances and long leakage path provided by the
pistons make high pressures readily obtainable, although this is limited in designs where
port-type rather than seating valves have to be employed, because of the rnovements
involved.
Where the entire cylinder block rotates about a stationary cam, porting has to be
employed, limiting the maximum pressure which can be developed to about 30 MPa (300
bar). Speed is also limited by the mass of the rotating cylinder block.
A typical configuration is shown in Figure 2A.I.2, where the rotating cylinder block is
mounted on a stationary eccentrically positioned pintle carrying diametrically-opposed
inlet and outlet ports, the whole cy linder block being surrounded by the casing of the pump
or a tracking ring.
Variable-delivery performance is readily achieved by making the eccentricity of the
tracking ring variable and thus the design is particularly versatile in this respect - see
Figure 2A.I.3.
The rotary type radial piston has found its main uses as an aircraft hydraulic pump and
for marine applications, but is less favoured than axial piston types for general industrial
hydraulics. It is not competitive with the in-line piston pump for high pressure services.
The alternative configuration, or true radial, employs a fixed cylinder block with the
cylinders radially disposed and a rotating cam or eccentric driving the pistons. Piston
return is accomplished either by suction or spring loading. In the case of higher speed
pumps, positive-return drives may be provided to ensure that the pistons return satisfac-
torily during the suction stroke. Higher speeds, and very much higher pressures, can be
realised than with rotary piston pumps, and the type can also have a variable delivery by
providing axial movement of the cam. It also lends itself admirably to tapping off separate
deliveries from individual cylinders or combinations of cylinders, to obtain multiple
outlets at different pressures, if required.
Axial piston pumps

In the axial piston pump the cylinders are disposed axially (parallel to the crankshaft) and
reciprocated by an inclined swash plate. If the cylinder block is fixed and the swash plate
oscillates the type is known as a wobble-plate pump, (Figure 2A.I.4). If the swash plate
angle is fixed, then the pump is a fixed-capacity type. It is, however, a relatively simple
matter to arrange for the swash-plate angle to be adjustable, thus giving the pump variable
delivery characteristics.
The rotating swash-plate pump is more usually produced as a fixed displacement type
and also differs from the rotating-block type in that the direction of flow is independent
of the direction of rotation. There are numerous variations in design detail, one of the main
problems being to minimise wear produced by the sliding contact of the pistons. Instead
of a plain swash plate the pistons often butt against a separate reaction plate or wobble plate
carried on bearings in the swash plate. The swash plate may then be constrained to remain
stationary, or in some designs may be driven through a gear at a relatively slow speed to
produce even wear on its surface.
The rotating block configuration (Figure 2A.l.5) has the advantage that it is easy to
arrange for the fixed swash-plate to be tiltable, to provide infinitely variable flow
characteristics, and it is thus a favoured type for specialised applications such as
hydrostatic drives. On the other hand, porting has to be employed, which limits the
maximum pressure capacity of the pump, and the design can become complicated when
positive drive methods are used to ensure that the pistons always remain in contact with
the swash plate. Often further complications are added by devices, such as floating port
plates, incorporated to reduce internal leakage at higher pressures. Slipper pads, at the
swash plate end of the pistons, are either lubricated through small central holes or are
designed to flex slightly under elastohydrodynamic forces with oil being forced under
from the sides of the pad.
It is interesting to note that the basic configuration of axial piston pumps has virtually
remained unchanged since the initial patent by Denison in 1937 (and filed in the USA in
1945) - see Figure 2A.1.6.
HYDRAULIC PUMPS 55
In order to increase the performance of axial piston pumps, there is a tendency to

increase the piston size. To maintain the same response the piston mass must be reduced
by improved materials or thinner (or hollow) sections. Typically in this design, compen-
sator or de-swash response times are around 300 ms, but can be down to 50 ms or less on
more advanced models.
Bent-axis piston pumps

The bent-axis or tilted-body pump employs a driving disc mounted at right angles to the
shaft, and the driven member and pump body rotate together, see Figure 2A.I.7. Angular
displacement of the driven member and body will then drive the pistons with a recipro-
cating motion, the stroke being proportional to the angular displacement. Piston drive is
normally through ball ended connecting rods providing a simple form of universal joint
coupling which also serves to rotate the cylinder block drive from the disc end. The block
itself must, of course, be suitably supported so that it is free to rotate.
FIGURE 2A.1.7 - Bent-axis piston pump [Denison Hydraulics].
As with all rotating-block pumps, seating type valves cannot be employed, so pressure
rating is somewhat restricted. The configuration does, however, readily provide for
infinitely variable flow, simply by altering the angle of the driven member relative to the
block. It is also suitable for running at relatively high speeds, and is superior to the rotating
block axial piston pump in this respect. Port connections do, however, call for a more
complicated design of internal porting.
Characteristics of piston pumps are summarised in Table 2A.I.2.
HYDRAULIC PUMPS 57
Vane pumps
Vane pumps are particularly suited to medium-pressure, medium-speed duties and hence
have the advantage over gear pumps that the rotor can be hydraulically balanced, thus
minimising bearing loads. The higher speeds (up to 2500 rev/min) have enabled vane
pumps to vie with piston pumps in some applications - previously their main application
was solely for low- and medium- pressure systems requiring a compact low-cost pump
(e.g. machine-tool hydraulic systems), their versatility being an attractive feature.
Currently vane pumps are capable of developing pressures up to 32 MPa (320 bar), with
maximum deliveries up to 410 - 455 L/min. In the case of variable delivery pumps,
maximum pressure rating is usually 7 MPa (70 bar). Minimum speed for vane pumps is
of the order of 200 - 450 rev/min.
The simple single-cell vane pump is unbalanced hydraulically and tends to suffer from
low volumetric efficiency because of the difficulty in controlling internal leakage. The
geometry does, however, readily lend itself to providing variable delivery characteristics,
simply by adjustment of the relative positions of the rotor spindle and outerring modifying
the capacity of the 'pockets' swept during revolution - Figure 2A.I.8. The casing can also
readily be fitted with a replacement liner to take wear.
A degree of balance can be achieved by adopting a two-cell or three-cell configuration
with a single rotor although this complicates the design and can compromise the pumping
performance. The usual form of balanced design incorporates two inlet ports and two
outlet ports on opposite sides of the motor, with an elliptical shaped cam ring. Both sides
of the rotor then have equal pressure excited in opposite directions, cancelling out side-
thrust on the rotor shaft and its bearings which now have only to carry the external load.
The most popular form of vane pump is the fixed delivery, balanced type, which apart
from having a higher volumetric efficiency than single vane pumps, can generate higher
pressures without slippage - currently, up to 32 MPa (320 bar).
Fixed delivery vane pumps

Fixed delivery vane pumps have had considerable success in industrial and mobile
machines. The heart of the pump is the rotating group, which basically consists of a rotor
which carries sliding vanes that sweep around the elliptically shaped inside wall of a cam
ring. During rotation, the varying volume of oil trapped in between the vanes, causes a
pumping action. With this type of pump different cam ring sizes denote different pump
capacities or flows. One prime area of concern to the pump designer, is the vane tip wear,
mainly due to the continuous sliding contact between the vane and the cam ring. One
patented design uses double lip configurations for each vane, so that one tip operates in
a pressure mode and the other in a suction mode.
In conventional vane pumps, the vanes are spring loaded in the rotor to maintain a
constant force against the cam ring irrespective of system pressure. In other designs, the
rear of each vane is hydraulically loaded by a hold-out-pin, where the hold-out pressure
is directly proportional to pump discharge pressure; this ultimately reduces wear, and
helps the pump run at lower speeds and pressures. See Figure 2A.1.9. (and also Figure
2A.1.15).
Variable delivery vane pumps

Variable-delivery vane pumps are normally unbalanced designs. The displacement of the
pump is changed by movement of the cam ring relati ve to the rotor, effectively modifying
the 'pocket' volumes. Ring movement in such cases is usually spring controlled with a pre-
determined spring pre-compression opposing the horizontal force developed on the ring.
Vertical movement of the ring is constrained by a thrust block. The spring rate then
determines the change in pressure necessary to move the ring from full flow rate to zero.
Deli very remains directly proportional to the ring setting or effecti ve 'stroke' , between the
range of full eccentricity and concentricity, unless the drop-off 'knee' is reached, when the
ring will move automatically to reduce the stroke to compensate for increasing pressure.
The pump is thus pressure-compensated and does not need a relief valve, as an over-
pressurisation will simply reduce the deli very to zero by moving the ring to the concentric
HYDRAULIC PUMPS 59
position. The main requirement is to establish a suitable delivery/eccentricity gradient so

that the pump cannot become unstable and attempt to oscillate over its normal working
range.
Gear pumps
Gear pumps come in two totally different designs - external and internal- although both
incorporate the conventional looking gear wheel or annulus gear. The form of the gear
teeth may vary slightly from the usual involute shape, and be positively chunky in style
in some cases, but the design is such as to provide a good overall efficiency with the
minimum of internal leakage. Whilst basic external spur gear pumps tend to be noisier,
additional features in either type can significantly reduce noise levels.
External-gear pump
The external gear pump is capable of developing higher fluid pressures than a vane pump
and can also be run at higher speeds. Its original limitation was high internal leakage which
has been overcome by the introduction of pressure balancing methods. Modern gear
pumps are rated for up to 30 MPa (300 bar) at 3000 - 6000 rev/min. Again, in special
designs, speeds of up to 50 000 rev/min may be achieved. Average speed for gear pumps
tends to be within the range 400 - 2000 rev/min.
The basic elements of an unbalanced and a balanced external gear pump are shown in
Figure 2A.I.I O. Two inter-meshing gears of the same diameter and form are mounted on
separate spindles and housed in a close fitting casing. Inlet and outlet ports are formed
directly in the sides of the casing, in line with the point of meshing. One gear shaft is driven
whilst the other idles. Both shafts are carried in low-friction rolling plain bearings.
During rotation, as each pair of teeth inter-mesh on the inlet side, the volume on that side
is reduced by the volume of two tooth spaces, providing a suction effect. Oil flowing into
the suction space is then trapped on each side by a tooth crest approaching the bore of the
housing and carried round to the delivery side by the 'pockets' between adjacent pairs of
teeth. On the delivery side, the oil is displaced from the delivery port under pressure.
For maximum volumetric efficiency there should be no leakage between the teeth, and
no leakage across the end faces of the gears. Even if such internal leakages are reduced to
a practical minimum, there may be a further penalty to pay in the high bearing loads and
unbalancing loads introduced by fluid trapped in the teeth pockets, which can result in
increased friction and lowered mechanical efficiency. Thus a compromise may have to be
reached in order to achieve maximum overall efficiency, although usually attention to
increasing volumetric efficiency results in higher gains than any corresponding reduction
in mechanical efficiency.
As a consequence a premium is placed on detail design, precision workmanship and
rigidity, in order to produce a high-performance gear pump with a high overall efficiency.
Although the mechanical efficiency is comparatively low, the overall efficiencies can
exceed 90%. Development of the high efficiency gear pump took place primarily for
aircraft hydraulic systems, where the design problems were further aggravated by the high
system pressures required, e.g. 21 MPa (210 bar). Lower cost gear pumps of up to 25 MPa
(250 bar) and higher rating efficiency, are now available for industrial hydraulics, whereas
previously industrial hydraulic gear pumps were of simpler design and lower efficiencies
and used mainly for delivery pressures below 3.5 MPa (35 bar ).
The basic design of this pump may be balanced by tooth venting although on low
pressure, relief may be provided by allowing the gears to run with a small backlash. On
high-pressure pumps, grooves may be cut in the side plates to provide communication
between the inter-tooth spaces (as shown in the second illustration of Figure 2A.l.1 0).
The external gear pump is available at very low flow output per revolution, and hence
is suitable for miniature hydraulics at high pressures.
The tooth form may not be precisely involute; an improved displacement can be
obtained by using an extended addendum involute form. Different tooth forms will also
provide different degrees of sealing. Whilst the stub gear remains about the best for
sealing, its displacement per revolution may be less than that which can be achieved with
other tooth forms. Single- and double-helical forms are also used, but mainly where a
quieter running pump is required. Indeed the helical form normally achieves a consider-
able reduction in noise level, often without loss in efficiency and with very small cost
increase. Reducing the number of teeth has the advantage of increasing the displacement
per revolution, regardless of gear form.
The usual method of producing a range of gear pumps of different capacities is to adopt
one or more gear diameter sizes and then manufacture each diameter size in several
different face widths, i.e. extend the capacity by increasing the face width of the gears in
lengthened casings. This enables the same detail design features to be preserved through-
out the series at minimum cost, although, in the case of high-pressure pumps, altering the
length of the gears will modify the casing stress and bearing loads. Decreasing the gear
HYDRAULIC PUMPS 61
length will decrease the casing stress but increase bearing loads, and vice versa.
Geometrically, the optimum form for a high-pressure pump is given by a gear width equal
to the pitch circle diameter of the gear.
Methods of detail design, aimed at reducing internal leakage across the end faces of the
gears are based on pressure-loading techniques. One such method is to fit side plates
between the gear faces and end covers, with high-pressure oil fed to specific areas on the
outside of these plates. The plates are thus pressurised to bring them into contact with the
gear ends and prevent end leakage, with a force proportional to internal pressure of the
pump. Numerous variations on this theme have been developed by individual manufac-
turers and although this involves additional components, the overall cost is usually less
than the cost of the more precise manufacture necessary to achieve similar end-leakage
control with fixed side clearances.
Most gear pumps are of the two- or three-piece design. Whilst the three-piece
arrangement (body, flange and cover) allows the internal forces on the body and bearings
to be constrained and balanced, the two-piece (with the gears and one set of bearings)
requires a secure means of alignment. Doweling is usually acceptable for the lower
pressures but if the full efficiency is to be maintained over the whole pressure range a
design like that shown in Figure 2A.I.II, with an oval recess which becomes tighter with
pressure, may actually improve the efficiency with the higher pressures.
FIGURE 2A.l.11- The Powermaster patented interlocking body [Ultra Hydraulics].
Gear pumps should not normally be run at speeds above the manufacturer's rated
maximum (although little harm is likely to result from such running for short periods under
no-load conditions); nor should they be run continuously at high pressures and very low
speeds. Otherwise they can be regarded as a particularly versatile type of pump which can
be run at any speed. Variable delivery operation would not normally be obtained by
varying the speed, however. It would be better to use two or more gear pumps, with off-
loading of individual pumps as necessary. Multiple units can then be mounted in tandem,
with a common drive.
Gear noise may be reduced by the use of helical gearing (as in Figure 2A.1.l2) or by
has some specific advantages over the external gear pump, notably the lower localised
fluid pressures generated and the lower shearing forces on the fluid. The long arc of
meshing means that there are much lower operating noise levels than a comparable
external gear pump, and the unit readily lends itself to multi-staging. Constant running
pressures up to 25 MPa (250 bar) are possible.
In the simplest form of the internal-gear pump, the eccentrically mounted inner gear
drives the outer gear, and some means of blocking the high-pressure side from the low-
pressure side is needed in order to prevent back-flow. The simplest way of achieving this
is by the introduction of a crescent shaped 'filler', filling the clearance space on the non-
meshing side of the internal gear. Precision manufacture is called for to give the close
clearances necessary to minimise internal leakage, but an internal gear pump of this type
is notable for the large displacement which it can offer relative to its overall size.
The 'Gerotor' pump is another form of internal-gear pump which is now produced in
designs capable of developing up to 14 MPa (140 bar) pressure per stage. It is low cost and
usually is included to supplement the main flow, e.g. to supercharge an axial piston pump.
In the main, however, all internal gear pumps are low-pressure types.
Another design which omits the usual filler, is the Superlip-System compensated
internal gear pump, as shown in Figure 2A.l.15. This combines the advantages of an
internal gear pump with those of a vane pump - being small, economically priced and
having a high overall efficiency (over 90%, even at low pressures). The crown of each of
the internal gear teeth has inserted within it a small sealing element which accurately seals
the 1800 displacement pressure region preventing pressure surges and high noise levels.
Radially arranged bore holes in the annulus gear carry the hydraulic oil to and from the
working region. The sealing process occurs at a sliding speed of almost zero, and hence
results in long life.
Gear and vane pumps may also be put in tandem in two's or three's - dual and triple
pumps - such as the triple vane pump in Figure 2A.I.17 A. The compact nature of these
make them suitable for mobile machinery where rapid movement is required, such as with
fork lift trucks.
One advantage of a dual gear pump is that it is possible to produce a variable output gear
pump, run at constant speed, by means of suitable control with unloaders; in this case one
pump may be the prime flow provider and the other the prime pressure provider.
Hybrid pumps are where different types of pump are joined together, such as a fixed
displacement vane pump with a variable displacement piston pump. Such a design would
ensures compactness for both single use and dual use; one example would be that quoted
by Denison Hydraulics where the hybrid is used on a mobile machine with steering and
with separate hydraulic cylinders
An even more exciting tandem arrangement is shown by the Sauer-Sundstrand triple
pump unit consisting of two axial piston pumps and a fixed displacement gear pump all
built into a single unit. (Figure 2A.1.17B) The pump was designed for the specific
application of a bobcat. The two axial piston pumps are built in a back-to-back configu-
ration with a common centre section - reducing the attachments necessary. The auxiliary
dri ve flange for the gear pump is machined directly into the rear housing of the axial piston
pump assembly, providing the gear pump with a unique drive shaft arrangement, and
giving an exceptionally short length to the overall unit.
Pump performance
The delivery of a positive displacement pump is directly proportional to its displacement
and cycling speed (i.e. rev/min in the case of rotary pumps and strokes/min in the case of
reciprocating pumps). Displacement for any pump is normally expressed as volume per
revolution, or volume per stroke. Maximum delivery is thus governed by the maximum
speed at which the pump can be operated, although the majority of pumps are designed for
direct coupling to electric motors and the operating speed is fixed by the motor design.
Delivery available governs the speed of operation of the actuator in the system, true
power output then being the product of actuator speed and the load against which it is
operating. It is generally more convenient to determine system power in terms of hydraulic
power which is directly proportional to the product of pressure and flow delivery. Thus
Power (kW) = Pressure (MPa) x Flow (Umin) / 60
This directly represents the work capacity or maximum power rating of the pump,
mechanical output power equivalent being obtained by factoring by the actuator effi-
ciency. The maximum power rating of a pump is determined both by its design (e.g. pump
type) and mechanical considerations (e.g. elements of construction).
Typical values in this respect are given in Table 2A.1.3.
Pressure rating
Maximum pressure developed by a pump depends both on the pump type and its design
and construction. With certain types (e.g. vane and external gear pumps), practical
maximum pressures are limited. With others, such as piston pumps, there is no absolute
limit to maximum pressure, only that related to construction strength and other practical
parameters. Table 2A.I.I indicates normal maximum pressure ratings for various types of
hydraulic pump.
All positive displacement pumps need some form of protection by pressure relief
against excessive and damaging pressures being developed by accidental over-load on the
output side. This is normally provided by a relief valve, the operation of which must lie
within oron the power envelope. Certain designs, however, permit pressure compensation
to be introduced - this being a general characteristic of variable delivery pumps. The
suction pressure developed by pumps may not be sufficient to avoid cavitation, and hence
a boost pressure may be required on the inlet to provide a positive suction head.
The delivery flow of a gear pump is directly proportional to its speed, whereas the
delivery pressure is determined by the external load. With increasing load the pressure will
continue to rise, up to limits set by either a relief valve or the strength of the casing itself.
Some slip is inevitable, and hence, for a given speed, delivery will decrease with
increasing resistance - Figure 2A.I.18. Slip is virtually independent of speed, except at
very low speeds where it will normally tend to increase and volumetric efficiency decrease
quite rapidly. As a general rule, high efficiencies can only be achieved with gear pumps
by operating them at relatively high speeds. Where a pump is rated for operation over a
range of speeds, operating at the highest speed will give the highest volumetric efficiency.
A maximum speed of 3000 rev/min for continuous running is general for most modern
high-performance gear pumps, although this will also depend on size. Some smaller units
may be rated for running at speeds of up to 8000 rev/min, but 3500 to 4000 rev/min is a
more usual maximum.
Rated delivery may be expressed in terms of displacement per revolution, or actual
delivery given for some specified speed (often maximum speed for continuous running).
In the former case, delivery at any particular speed can be obtained by multiplying by the
speed in rev/min. In the latter case, delivery at any other speed follows from simple
proportion. Note, however, that delivery is normally determined empirically, using a low-
or medium-viscosity oil discharging freely (i.e. zero discharge pressure). Actual delivery
Efficiency
Overall efficiency of a pump is expressed as the ratio of hydraulic power output to power
input (normally multiplied by 100 to give efficiency as a percentage value). Overall
efficiency also includes mechanical losses (i.e. friction), and so actual hydraulic efficiency
or volumetric efficiency may be quoted separately. Table 2A.IA summarises some typical
values for different pump types. It should be noted, however, that the actual pump
efficiency achieved in a system can be affected by many operating factors. Figure 2A.I.19
shows some typical efficiency curves for the various types.
As a general rule, piston pumps have higher overall efficiencies than other types,
although pressure balancing or precision manufacture of gear pumps with fixed end
clearances has resulted in efficiencies comparable to piston pumps being achieved. On the
other hand, the overall efficiency of a simple, inexpensive gear pump can be quite low.
These figures are a general guide only. Actual efficiencies achieved depend very much
on the detail design and manufacture of an individual pump, and also to a large extent on
the size of the pump (smaller pumps tend to be less efficient than larger ones of the same
type).
The question of the duty cycle involved must also be taken into account. Thus where
there is considerable variation in demand it is usually more efficient as regards both
operating and running costs to use a variable-delivery pump, although the initial cost of
the pump will be higher. In many cases a variable delivery pump may be virtually essential,
as this form of delivery regulation is more efficient and easier to arrange than a variable-
speed drive; however, in certain pump designs (e.g. in-line piston pumps with multi-stage
units) it may be possible to tap off different deliveries and different pressures.
Efficiency may be very important (e.g. where large volumes of fluid are being pumped);
or relati vely insignificant (e.g. in a light duty system where ample input power is available
from an inexpensive driver, or an 'over-size' pump is to be used for a particular reason).
Power regulation
Power regulation may be necessary in systems with fluctuating loads so that the power
output of the driver running at constant speed can be fully utilised. This can be done with
a horsepower regulator which automatically reduces delivery as the load increases, and
vice versa, to maintain a constant value of hydraulic power in the system.
A theoretical power curve for a positive displacement pump is shown in Figure 2A.I.20.
The pressure level is determined by the external load, the maximum value being set by the
mechanical strength of the pump itself. Maximum flow is established by the maximum
permissible running speed. Maximum power is thus developed at maximum pressure and
delivery, the corresponding point on the curve being called the 'corner power point'.
In practice the envelope is not quite rectangular, since increasing load will produce
increasing slip, modifying the envelope to the form shown in Figure 2A.I.21. The value
of slip depends to a large extent on the precision with which the pump is manufactured and
may be as low as 3% to 4% at maximum pressure. The effect of slip is, of course, to reduce
the overall efficiency achieved.
Pumps capable of operating continuously at maximum pressure and maximum delivery
can be fully defined for power rating by the form of curve shown in Figure 2A.I.21. They
can obviously be operated anywhere within the envelope, or right up to the corner power
point provided the loading is steady.
Many duty cycles will, however, involve transient loads, in which case these must be
taken into account, otherwise an immediate load may place a demand on the pump beyond
either its available power or its maximum pressure rating. The usual method of rating in
such cases is to plot the envelope as a constant power line, as in Figure 2A.I.22. This now
allows for transient loads to be accepted within the corner power point, and also defines
the safe limit for transient loads. The over-load on the driver can also be determined,
relative to any momentary displacements above the constant power point line.
Power:weight ratio
The weight of an hydraulic pump is generally comparable with that of any other power
HYDRAULIC PUMPS 71
generator of the same power rating. In certain applications, such as aircraft hydraulic
systems, installed weight and bulk can be an important factor and the power:weight ratio
becomes a critical feature. Figure 2A.l.23 shows an example of an exceptionally high
power:weight ratio where both the motor and the pump are involved in an integral fluid-
cooled ac design; instead of 0.5 kW/kg, the ratio is as high as 2.0 kW/kg with two variable
displacement in-line piston pumps and two inlet boost pumps. Industrial counterparts are
becoming more and more available, somewhat de-rated in performance, but less costly.
Electrohydraulic control
The control of pumps - as regards flow, pressure, velocity and power- has conventionally
been undertaken by separate components. For smaller pumps this is quite acceptable,
however, by integrating the control with the pump, as one module, considerable gains can
be achieved for the larger pumps. For instance, response times are much reduced and
accuracy is improved. Typical applications would be injection moulding, drives in press
control systems or testing machines, cylinder drives in machine tools.
One example of a radial piston pump with electrohydraulic control is shown in Figure
2A.1.24. Three additional components are included:
Positioning control for the stroke ring of the pump (including a position detector,
a servo solenoid valve and an electronic amplifier)
Electric pressure control (flow is controlled to maintain the set pressure)
Electric circuit for leakage compensation (to compensate for the pressure- depend-
ent volumetric losses of the pump).
Hydraulic pump selection

Operational parameters to be considered in the selection of a hydraulic pump are:
System pressure requirements
Delivery volume requirements
Circuit requirements (fixed or variable, closed or open loop, need to be a motor as
well as a pump, etc.)
Drive speed
Cost effectiveness (including pump efficiency)
Fluid compatibility
Bulk and weight
Noise and vibration
The first three items effectively define the duty requirements.
System pressure
In the case of positive displacement types the pump operating under steady conditions
cannot develop a pressure greater than the resistance offered by the system to which it is
connected, i.e. the pressure developed is extremely dependent on the load. In a closed
system, this load can (theoretically) approach infinity (or the failure pressure of individual
components so that maximum working pressure is normally set by a relief valve).
In general, high pressures can only be achieved by pumps with seated valves, although
the inherent limitations of rotary valves or ports can be overcome to some extent by
refinements in design detail. For the very high pressures needed for heavy presses and
HYDRAULIC PUMPS 73
similar applications, only multi-piston pumps with seated valves are likely to be suitable,
when pressures of the order of 70 MPa (700 bar) can be achieved continuously.
Figure 2A.I.25 gives an overall view of typical regions of appropriateness as regards
pressure and displacement. Whilst minima are not shown, the average and maxima show
the range covered by gear, vane and piston pumps. It will be noticed that although gear
pumps can vary considerably in output at a given pressure, the vane pumps have the
opposite characteristic with a wide variation in pressure for the same displacement.
For lower system pressures the choice of pump type becomes much wider. Thus, for
general industrial hydraulics, where system pressures may range from 3 - 3.5 MPa (30-
35 bar) up to about 10 MPa (100 bar), virtually any type of positive displacement pump
can be used, selection of type then being based on other specific requirements, or power
rating. However, higher system pressures are becoming commonplace, so that 17.5 MPa
(175 bar) can also be considered an 'industrial' rating; this is beyond the limit of simple
vane pumps unless they are coupled as two-stage units. The general tendency to uprate
system pressures for higher working efficiency has extended the performance require-
ments beyond the capabilities of certain basic (and simple) types of pump, calling either
for the use of alternative pumps or further development of such basic types to meet new
pressure requirements.
Delivery
Basically for anyone type of pump, the delivery volume required governs the size of pump
necessary. However there are practical limits to volumes achievable with different pump
types. Where very large deliveries are required a screw pump may have to be considered.
The alternative is to consider the relative cost-effectiveness of providing the demand from
a number of smaller pumps or using dual or triple pump designs.
Delivery is also related to the speed at which the pump is driven. For directly-coupled
pumps this is normally the standard speed of electric motors. Higher speeds may be
obtained via belt drives or gearing, which may also prove advantageous in increasing the
efficiency of the pump. However, it is usually more efficient and less costly to drive a
pump by direct coupling rather than through gearing as this must involve some power loss.
Also there are practical limits as to the maximum speed at which individual pumps can be
driven.
An important parameter in this respect is the maximum power rating of the pump, as
representing its effective work capacity.
Drive speed
There may be a restriction due to the driver that is available. Occasionally portable units
are battery driven and speeds are limited.
Cost-effectiveness
Cost-effectiveness embraces a study of initial costs, operating costs, system reliability and
maintenance. This can finally decide the most suitable type of pump where there are
alternatives which can equally well meet the duty requirements. Here, cost factors may be
in opposition. Thus piston pumps tend to offer the highest overall efficiencies with high
reliability, but higher initial cost. gear pumps are generally less expensive than vane
pumps and more tolerant of adverse operating conditions (e.g. contaminated fluid), but
have lower efficiencies than vane pumps capable of the same duty.
Fluid
Mineral oil fluids present no problems in any of the types of hydraulic pump, provided the
viscosity is suitable for the design of pump or the system, and the fluid is of sufficient
cleanliness (see chapter on Contamination Control). Viscosity requirements are usually
specific and determined by the pump manufacturer's recommendation. The majority of
hydraulic pumps are designed around a fluid of a particularly viscosity. Any departure
from this viscosity will modify the performance of the pump, resulting in lowered
efficiency and, in extreme cases, even calling for re-rating of the maximum speed.
All pumps are likely to be harmed by the presence of solid contaminants in the fluid,
some types being more susceptible to damage than others, e.g., precision made pumps with
close tolerances and sliding surfaces. At the other extreme, however, slow speed in-line
piston pumps are comparatively contaminant insensitive.
The use of fluids other than mineral oils may seriously affect the life of the pump, mainly
as regards inadequate lubrication of rubbing surfaces and incompatibility with metals and
seals. Most piston pumps are life rated according to the bearings used; gear pumps would
fail due to teeth scuffing; and vane pumps due to sliding vane wear. In each of these
situations the oil used is significant. (The bearing effect is even greater where hydraulic
imbalance is present causing high bearing loads.)
A derating factor is given for the different pump types and oils used in Table 2A.1.5,
taking mineral oil as the standard. Note, however, that in some pump types, such as in-line
piston pumps, adequate lubrication for the pumps bearings can often be provided by a
pressurised oil system, and they are thus independent of the lubricity of the fluid being
pumped.
Pumps designed for use with water, or high water based fluids (HWBF), require special
materials to avoid damage from corrosion and excessive wear due to the poor lubricity.
Leakage is also likely to be greater due to the lower viscosity.
Bulk and Weight

No general comparisons can be drawn between power:weight ratios of different types of
HYDRAULIC PUMPS 75
pump, since so much depends on the individual design and the constructional materials
used. Comparative power:weight ratios can, therefore, only be extracted from manufac-
turers' specifications.
A further point here is that where weight is an important factor, it is usually the total
system weight rather than the weight of the pump which is the major parameter. Optimum
solutions are usually obtained by utilising system pressures of the order of 21 - 28 MPa
(210 - 280 bar) using smaller sizes of components throughout and also minimising fluid
volume.
Noise and Vibrations

Noise and vibrations are generally required to be limited. Excessive noise may be
disturbing, or a nuisance, excessive vibration can aggravate wear. Noise generated in
pumps is largely the result of sudden pressure changes between the suction and outlet side.
Thus pumps which produce high localised pressures are likely to be noisier than those
providing a more gradual pressure change. No specific data can be given since the actual
noise generated also depends on the mass and rigidity of the pump body, as well as the
materials of construction, where and how it is mounted, and whether or not it is isolated
from its mounting and delivery line.
In general, however, all piston pumps tend to be relatively noisy in operation. Gear
pumps maybe noisier than vane pumps, although they can be quietened by various means
such as, using helical gear forms, increasing the number of teeth (and reduce their size),
phasing the mesh of two gear pumps in tandem, reducing backlash, etc. Virtually all types
of pump can be made quieter by suitable modification of design detail without a loss of
efficiency. Again, as an alternati ve, a positi ve way of reducing pump noise is simply to use
an over-size pump and run it at a much lower speed to give the same delivery.
50 MPa (500 bar) - the new benchmark?

The upper limit for continuous pressure rating for off-the-shelf pumps has been around 35
MPa (350 bar) for a good number of years. However, there is a growing demand for more
compact power which requires higher pressures at similar efficiencies with no increase in
noise and cost. Special units have been available to order, particularly for small
displacements, but it is the introduction of the 50 MPa (500 bar) standard unit which moves
hydraulics into a new stage. Figure 2A.l.26 shows the axial piston pump produced by
Denison Hydraulics which can run continuously at 42 MPa (420 bar) with a maximum
intermittent rating of 50 MPa (500 bar).
77
PUMP DRIVERS AND COUPLINGS
IN THE great majority of industrial applications hydraulic pumps are driven by electric
motors. However, internal combustion engines are frequently involved, particularly
where mains electricity is not readily available; air motors are another type of driver. The
electric motors are, however, the most compact, efficient and quietest form of driver;
direct coupling is favoured with the pump being driven at a constant speed.
Except in the case of variable-capacity pumps, the pump drive speed governs the actual
delivery. Operating speed limits are set by the design of the pump, whilst desirable running
speeds are set by the normal running speed ofthe driver - it is more efficient and less costly
to drive the pump by direct coupling, rather than through gearing.
Normal maximum operating speeds for various types of pumps are summarised in
Table 2A.2.1. Specific proprietry pumps will be rated for the maximum speed at which
they can be run, either continuously or with separate speed ratings for continuous and
intermittent operation. These recommendations should not be exceeded.
Where direct coupling is used and the driver speed is lower than the maximum rated
speed for the pump, then some sacrifice of potential capacity must be accepted and the
pump size selected accordingly. Pump deliveries are usually expressed in terms of
displacement per revolution, or displacement per 1000 rev/min, so determination of
delivery at any particular driven speed follows by simple calculation. With variable speed
drives the delivery at any speed can be calculated on a similar basis.
For low speed pumps, such as multi-piston in-line pumps, it is likely that a gear
reduction will be necessary between the electric motor and the pump rotational axis.
Worm-and-wheel reductions may be used, or epicyclic gearing. Reductions of the order
of down to 20%, or less, are in order, for example from a 1470 rev/min prime mover to 286
rev/min pump.
Electric motors
There are numerous features which have to be considered by the manufacturer of electric
motors. Most of such detail is not relevant to this current book although it should be pointed
out that electric motors may be DC, AC inverter, ac vector, or servo, each with its own
advantages and disadvantages. The AC motor is more widely used, but dc has its place in
instrumentation and the smaller hydraulic systems where battery power is used. What is
important, though, is to examine the functions which relate directly to the hydraulic
application and its location, as listed below
i) Power - (from flow and pressure)
ii) Function - (working cycle)
iii) Speed - (range and direction of rotation)
iv) Start-up load - (no-load or load mass moment of inertia)
v) Control - (accuracy required)
vi) Braking - (required? if so, torque curve)
vii) Installation - (dimensions, foundation, coupling)
viii) Environment - (temperature, altitude, vibration, hazardous)
ix) Supply - (current, voltage, frequency, fluctuations, power
factor)
x) Regulations - (national, climatic, local instructions)
For optimum costs the sizing of the electric motor needs to be just sufficient to run the
pump at the required pressure and flow. Providing the flow and pressure are mainly
constant there is no problem in relating the power in, to the power out (with suitable
account of the losses incurred). However, if the system operates over a varying pressure/
flow cycle it may not be necessary to match the power at the maximum conditions
providing the torque is sufficient - a smaller motor could well be adequate, averaging out
the power requirements.
The majority of electric motors manufactured in the United Kingdom are produced to
British Standard requirements for electric design and dimensions, the most important
being:
BS 5000 Rotating electrical machines
[from 1978 onwards with numerous parts]
Electric motors are also specified by service ratings covered by BS and DW Standards,
VDE specifications, and others. Various other authorities also lay down their own
regulations and standards. Type of enclosure is likely to be of importance, as seen in Table
2A.2.2.
Smaller applications, including test equipment, (up to about 20 kW), may be driven by
single phase motors; however, the majority of industrial electric motors are of the
induction type, used with 3-phase supplies. Depending on the type and application, motors
The torque generated by 3-phase motors varies widely between zero speed and
synchronous speed. For adequate acceleration starting torque will exceed the break-away
torque of the hydraulic pump by a sufficiently large amount and during the acceleration
to working speed the motor torque will always be greater than the load torque. Minimum
requirements for pull-up types of fixed-speed motors are 0.5 times the starting torque and
not lower than 0.9 times the rated torque for motors of less than 50 kW.
Limit of overload capacity is determined by the break down torque. Induction motors
should be capable of delivering a torque up to 1.6 times the rated torque at rated voltage
and frequency for 15 seconds (VDE 0530).
Generally a high speed motor is cheaper for a given torque value.
Permissible motor-on times are governed by the permitted loading, otherwise the
motors may over-heat. Permitted loading is also related to the duty cycle. A 100% duty
cycle corresponds to continuous running with constant-rated load, or a cyclic duration
factor of 100%.
For periodic rather than continuous operation, the cyclic duration factor is given by:
Example: Cycle time is 10 min, with interval 6 min, (i.e. the motor is running for 4 min,
followed by a 6 min idle period at no-load, then restarting); the cyclic duration time
= 4 /10 x 100 = 40%.
With a low duty factor (i.e. ts appreciably greater than tB) the motor has sufficient time
to cool down between the intervals of operation and can be rated accordingly. With a
higher duty factor the rest intervals may be insufficient for the motor to cool off. Motor
characteristic curves show the permitted cyclic duration factor for different duties.
Figure 2A.2.2 shows a single combined AC electric motor/pump unit (a more complex
arrangement is discussed in the previous chapter on Pumps). There are several advantages
of such a design including the use of the hydraulic fluid to totally cool the electric motor.
Because the unit has no air cooling, and it may be completely enclosed, the sound level
is significantly reduced: a combined unit at 15 kW 3.5 MPa produced 58 dB which is some
16 dB lower than a conventional 1800 rev/min air-cooled motor when driving a vane
pump. Leakage is also reduced.
FIGURE 2A.2.2 - An integrated

motor/pump [Vickers Systems]
PUMP DRIVERS AND COUPLINGS 81
Asynchronous electric motor

The slip-ring rotor type of three-phase asynchronous motor is more expensive, not only
as regards the purchase price but also the need for servicing of the slip rings and brushes.
It is, therefore, primarily only used for those applications where the squirrel cage motor
is inadequate. Its advantages are:
Starting current is low
Starting torque is high
Speed changing is easier (loss-free or dissipative)
High switching frequency is possible
Efficiencies
The efficiencies of various types of electric motor have been quoted as shown in Table
2A.2A.
If the motor is to be run at constant speed but not at its full rating, then an electronic
controller significantly improves the running costs by reducing the voltage supplied to the
terminals. This may be undertaken on a continuously variable basis, is in some of the latest
variable speed AC inverter drives, or by switching from delta to star winding connections
on star-delta motors. Although DC motors with electronic control show the best efficiencies
below 50% full load, AC induction motors are the more common because of lower costs,
lower maintenance and wide availability.
Internal combustion engines

Where an hydraulic system is associated with an engine-driven system or machine (as on
mobile machinery) it is normally logical to drive the hydraulic pump from the main driver
via a V -belt or gearing to match engine speed to optimum pump speed. Off-loading of the
PUMP DRIVERS AND COUPLINGS 83
pump is essential with a continuously running engine. On some systems, a clutch may be
fitted to engage/disengage the hydraulic pump as required.
Where the hydraulic power demand is low it can be more advantageous to drive the
hydraulic pump by an integrally mounted DC motor worked off a battery which is
normally charged by an alternator associated with the main driver. Alternatively on larger
installations, e.g. ship's hydraulics, the hydraulic pumps, may be driven by electric motors
drawing current from the electrical services supply. The advantage of this is that indi vidual
pumps can be located close to actuator points.
Diesel engines have relatively high efficiency (35-40%) with good torque controllabil-
ity. A low speed is best for continuous duty, but short durations at higher speeds, say, 750
and 1500 rev/min, are possible. Larger diesel engines can be designed to operate on dual
fuels - diesel fuel for ignition and gaseous fuel for operation.
Other drivers
It is possible for hydraulic pumps to be driven by any power source. With smaller hydraulic
pumps, air motors are feasible. Water turbines are another possibility. For emergency use
in aircraft, the RAT (Ram Air Turbine) - see Figure 2A.2.3 - is an important back up
arrangement.
It must be remembered that the output from whatever the driver must match the input
to the hydraulic pump as regards
speed
start-up torque
continuous power
Dimensions must also be appropriate.
Couplings
The coupling between the pump driver and the hydraulic pump is as important as the two
components it connects. Not only does the coupling have to transmit the required torque,
it must be balanced, aligned and it has also to cope with axial motion and the torsional
oscillation in the combined system. In practice, however, the coupling may not only
'cope', it may also be able to improve the situation by preventing axial or torsional
oscillations from the driver (particularly internal combustion engine drivers) from
reaching the pump and causing excessive forced or resonant stresses. (The oscillatory
forces could exceed the steady output torque of the drive by a factor of as much as ten.)
As mentioned previously, direct connection may be possible in proprietary units but
normally the driver and pump would be separate units requiring a coupling and possibly
a gear train. Where a flexible coupling is used it is possible to arrange the stiffness so that
the critical natural frequency ofthe drive train occurs outside - usually below - the normal
operating speeds.
Because hydrostatic pumps usually have low inertias, it is advantageous to use a stiff
flexible coupling to connect them to a diesel engine. This has the effect of shifting the
critical rotational speeds into speed ranges higher than those of the engine. The coupling
is usually built into a housing between the diesel engine and the hydraulic pump. The
coupling, wherever possible, is bolted directly onto the diesel engine flywheel. One
example of a coupling suitable for mid to low power diesel driver to hydraulic pump
applications is shown in Figure 2A.2.4.
coupling, wherever possible, is bolted directly onto the diesel engine flywheel. One
example of a coupling suitable for mid to low power diesel driver to hydraulic pump
applications is shown in Figure 2A.2.4.
85
HYDRAULIC MOTORS
WHILST MANY types of pump will work as motors (particularly axial-piston pumps),
the most efficient hydraulic motors are designed specifically for the one purpose.
Combination pump-motors involving valves need special modifications, and even other
types, although working, would normally need small changes in detail if the efficiency and
noise levels are to be maintained.
Particular differences which may be observed between some types designed specifi-
cally as motors rather than pumps are modified port timing and the provision of case drains
to protect shaft seals; also vane motors usually differ from vane pumps in having spring-
loaded vanes to ensure good starting torque being generated.
Hydraulic motors work by accepting pressurised fluid at their inlet and converting
pressure energy into rotational energy (torque and speed). Basically torque matches the
operating pressure and rotational speed is dependent on the flow rate through the motor.
Classification
Hydraulic motors may be classified by type, much the same way as pumps - e.g. piston
motors, vane motors, gear motors, etc. They can also be classified by their performance
characteristics. A general classification for the latter is:
i) High-torque motors-designed to provide high output torque with maximum torque
(usually) available from starting up. Motors of this type are normally low-speed
machines, generally of radial-piston or axial-piston configuration.
ii) High-speed motors - designed for high operating speeds with low torque. Charac-
teristics may vary widely with different types and individual designs. Chief type of
high-speed motors are gear motors, axial-piston motors and vane motors.
iii) Medium-torquelmedium-speed motors - designed to provide good torque with
higher operating speeds than high-torque motors. A widely differing range of
performance may be offered by radial-piston, axial-piston, vane and gear motors in
this category.
iv) High-moment motors - designed specifically to provide good start-up torque with
higher speed of operation than high-torque motors. They are usually of radial-piston
type.
Piston motors
Initially the axial-piston configuration was the primary design for hydraulic motors, see
Figure 2A.3.1, however, the radial design has developed considerably and is now probably
the most common for the larger applications.
Axial-piston motors
The axial-piston design has the advantage that it is usually a true combination pump-
motor, i.e. it will operate equally well in either mode. The same specifications usually
apply for pumps and motors. Some axial-piston motors are specially designed for high-
torquenow-speed applications, or as specific matches to particular pumps for hydrostatic
drives. Varial displacement axial piston motors are also the main type used for infinitely-
variable-speed drives. These characteristics are discussed in the chapter on Hydrostatic
transmissions. See also Table 2A.3.1.
HYDRAULIC MOTORS 87
Radial-piston motors
The basic principle of the radial-piston motor is that the movement of the pistons radially
in a cylinder block either causes a rotation of the cylinder block (and an outside cam remains
stationary), or the outside cam rotates with a fixed cylinder block. Figures 2A.3.2a and
2A.3.3b show two designs. The force on the cam ring is converted into torque by the
reaction of the linear force on the cam roller against the cam ring; the cam roller can be seen
as a simple hydrostatic bearing. It is usual to have an even number of pistons to maintain
the balance in the motor, the greater number giving the smoother rotation (e.g. 12)
The radial piston carn ring design of hydraulic motor, covers a wide range of sizes from,
say, 1 litre displacement and 160 NmlMPa (16 Nm/bar), up to 150 litre displacement with
24 000 NmlMPa (2400 Nm/bar). This enables direct hydraulic drive of many types of
machine, and does away with the need of reducers and gearboxes. Without high speed
elements and gear trains, a very low inertia drive unit exists which offers a high power to
weight ratio with a high resistance to shock loading. The torque output is constant from
zero speed to maximum, so the machine can be started under full-load conditions without
trip-outs and time restrictions; good controllability is possible in all four quadrants -
be overcome by using spring-loaded vanes, or providing additional porting to pressurise

the bottom of each vane. In general, however, a vane motor even with such modifications
is not suitable for low-speed running, although, again, loss of performance at low
rotational speeds can be offset by increasing the number of vanes.
An alternative approach is provided by the so-called rolling vane or rotary abutment
motor. The rotor containing the sliding vanes is concentric with the housing in this case,
and pressure is applied continuously to two vanes to produce continuous rotation. The
rotary abutment gears provide a seal between the rotor and casing, as well as timing of the
rotor and rotary abutment.
Although considerably more complex than a simple sliding-vane unit, this form of vane
motor operates with very low friction and has good low-speed performance, and it has
been operated over a wide range of speeds.
Gear motors
See Figure 2A.3.6. Whilst some external gear pumps can also work as motors, certain
features need careful design. There is a need to provide a high volumetric efficiency and
at the same time to minimise internal leakage. Occasionally some mechanical efficiency
is sacrificed in order to improve volumetric efficiency, but this is more applicable to a gear
pump design than a gear motor design. Ideally, for easy starting up and smooth running,
friction should be kept to a minimum with a gear motor and this normally calls for some
degree of pressure balancing of the gears.
Conventional methods of pressure balancing applied to gear pumps to mini mise internal
leakage normally favour uni-directional running and may not always be applicable in the
case of a motor suitable for general use. The construction of a high-performance gear
motor with fixed side clearances involves a manufacturing cost penalty because of the
greater precision required. Low-cost gear motors, therefore, may have certain limitations
as regards overall efficiency and the maximum fluid pressures they can accept without
generating excessive bearing loads and friction. On the other hand, a well-designed and
HYDRAULIC MOTORS 91
precisely made gear motor can run at speeds of 5000 to 10000 rev/min and be capable of
working at pressures up to 21 MPa (210 bar) - Table 2A.3.3.
Internal gear motors are usually limited to low powers, but can offer specific advantages
in such applications. The internal gear with (typically one tooth less in number) drives
round the external gear and shaft, producing a shaft speed higher than the rotational speed
of the two gears. Thus a relatively high shaft speed can be achieved with relatively low
gear-to-gear sliding speeds.
Lobe-rotor motors
The lobe-rotor pump-motor is a form of gear motor where internal and external gear 'teeth'
are specially shaped to provide sealed 'pockets' of fluid in the same manner as a lobe-rotor
pump. The pressurised fluid fed to the motor acts directly on the exposed internal gear
'tooth' via appropriate porting or a distributor valve. The inner gear is thus caused to rotate
relative to the stationary outer gear. Various configurations are possible and the mechani-
cal output can be derived through intermediate gearing, usually arranged to provide a drive
reduction. Internal gears of this type are normally used for low-speed motors capable of
generating high torques - see Figure 2A.3.7.
Applications
Typical applications of direct hydraulic drives are injection moulding machines, mixers,
mills, winches, shredders, feeders and bulk handling machines such as bucket wheels and
ship unloaders. Figure 2A.3.8 shows the vast Marathon motor on a bucketwheel rec1aimer
(note the small size of the personnel handrail).
93
HYDRAULIC HAND PUMPS
MANUALLY-OPERATED hydraulic pumps - generally known as hand pumps - are

almost invariably of piston type. Limited use has also been made in the past of gear pumps
hand driven using a winding handle via step-up gearing, but much higher deliveries (and
pressures) can be achieved by piston pumps for the same effort.
Hand-operated piston pumps may be either single-acting or double-acting. Two-stage,
or two-speed, pumps provide for high delivery and minimum pumping time at light loads,
the pump output being reduced at a predetermined level in order to reduce the load required
to build up to maximum pressure. The change-over is normally based on a maximum effort
required of the order of 250 N. Figure 2A.4.1 shows a small two-speed lightweight hand
pump with a maximum handle effort of 400 N and able to reach 70 MPa (700 bar) with
1.5 mL per stroke ..
Alternative methods of providing two-stage operation include mechanical systems to

alter the leverage (and thus the stroke of the pump) and the use of two separate cylinders,
one providing high output at low to moderate pressures and the other low output at high
pressure Change-over is accomplished by unloading the low-pressure cylinder either
manually or automatically.
A typical two-stage design would incorporate two double-acting cylinders, one having
twice the stroke of the other, When the pressure rises to a given point, the load
automatically cuts out the longer stroke cylinder, allowing pumping to proceed with
reduced output at the higher pressure.
Because most of these pumps are used as portable devices, their construction usually
contains as much aluminium as possible with the higher pressure regions made from steel.
Hand pumps are used for such items as - portable pressure calibrators; delivery pumps
to over 110 mL/stroke; back-up fluid supply up to 150 MPa (1500 bar) with small flows
(5 cm3/stroke), personnel hoists, etc. Foot operation is also used.
Another typical example is shown in Figure 2AA.2; this hand pump is used for applying
a constant displacement of oil to a system, and is shown with a manual changeover level
to work a double acting actuator.
Hand pumps are also of considerable value as a back-up in case of the failure of mains
power in critical systems, such as ship steering and aircraft landing gear.
95
HYDRAULIC CYLINDERS
HYDRAULIC CYLINDER is the most commonly used term to describe the device which
gives both a linear force and output movement, although more generally itcan be classified
as a "linear actuator". It has many other names including ram, jack, power cylinder and
displacement cylinder. Some of these terms are industry specific and not very widely used.
To understand more easily some of the distinctions between the different terms used for
actuators it is better first to define the types of cylinder which are commonly available.
Classification and definitions

All hydraulic linear actuators have a fluid filled volume contained within the body of the
device, the tube or barrel, and a component which moves within the cylinder called a piston
or piston rod. When an actuator is moving to increase in length ie. the piston is moving out
of the cylinder body, it is said to be extending. When the piston is moving into the cylinder
body it is said to be retracting. In the simplest devices the piston is a cylindrical rod which
transmits the output force directly to the load and this is shown schematically in Figure
2A.5.1. --'-~ ..
FIGURE 2A.5.1 -
A simple ram actuator
(displacement
cylinder)
These devices are also given the name ram, plunger or displacement cy linder. They may
be used in heavy duty industrial applications egometal roIling mills but there are also many
smaIl diameter and short stroke devices which fit this category. A device like this, which
can only move under the action of supplied fluid in one direction (extend) is also caIled
single acting. Such a device requires an additional force to return the piston/plunger in the
retract direction when the fluid pressure is reduced, this could be provided by a gravity load
as described in the chapter on Basic circuits. There are also cylinders with a return stroke
operated by a spring which may be either internal or external. A seal set is required around
the piston rod as indicated and it is the cross sectional area of the rod which determines
the cylinder performance since there is a significant clearance within the cylinder bore.
An alternative type returns the piston under the action of a fluid flow and is caIled a
double acting cylinder. A double acting device has a rod connected to the load which is
a significantly smaIler diameter than the cylinder to aIlow a ring area between the outside
diameter of the rod and the cylinder bore for fluid to act on the other side of the piston
(which is the fuIl bore diameter). This gives an area exposed to fluid on each side of the
piston but these are different, and this type is sometimes referred to as an unequal area or
single ended actuator. This type requires some sealing between the piston outside diameter
and the cylinder bore as weIl as on the piston rod.
The term jack can be used to describe a single-acting cylinder employed for lifting or
'jacking' actions butit is also used to describe double-acting cylinders in certain industries
(eg. in the aircraft industry).
The two devices so far described are the most commonly used. The other basic variation
in geometry gives a design with a rod at both ends. This may be caIled a through-rod,
double rodded, double ended or equal area actuator. The rods may be the same diameter
giving equal ring areas each side ofthe piston and equal force capability in both directions.
HYDRAULIC CYLINDERS 97
It also gives 'two-bearing' support for the rod at each end ofthe cylinder and hence greater
rigidity. Some have different rod sizes, usually used to operate control switches. It requires
an additional sealing arrangement for the extra rod. All three of these basic types are
illustrated in terms of their standard symbols in Figure 2A.5.3.
Other cylinder designs

There are a number of other designs which, although less commonly used, can be
considered as standard alternatives and these are summarised in Table 2A.5.1. Perhaps the
most common of these is the telescopic cylinder which is made from concentric tubes of
progressively increasing diameter fitting inside each other. Each successive stage in this
design forms the function of both piston and cylinder tube except for the innermost (just
a piston) and outermost (just a tube). This design gives an operating stroke length which
is many times the closed body length of the largest diameter tube, the "cylinder" length
itself. These are mostly used in single acting form with a gravity load to give the force to
close the assembly. In hydraulic lifts two or three stage cylinders are common but for tipper
trucks from five to ten stages could be used.
Note that because of the larger area available with the largest diameter tube, it will tend
to extend first at a lower pressure and complete its stroke before the next stage starts to
extend. This means that the whole device automatically extends in the most sensible
sequence with an increase in pressure for each stage if the load is constant. However, it also
means that the smallest section must be sufficiently large to support the load and that all
the others will effectively be oversized. Note also that if driven from a constant flow source
the extend speed is likely to increase with each successive stage. This assumes ideal
operation and may be modified by seal friction.
Double acting telescopic designs are produced but they have a relatively small area for
the retract movement formed by the annulus between each tube section. The sealing and
hydraulic connection arrangements are also quite complex which make them expensive
devices.
Basic equations
There are two fundamental sets of equations relating the actuator dimensions to the
variables associated with the fluid itself, and these determine the actuator force and the
speed of movement of the piston relative to the cylinder. These equations are developed
for the double-acting, single-ended cylinder and substitution of the appropriate areas will
give the equivalent equations for the other types. The actuator diameter is the principal
dimension determining the full piston or cylinder bore area and together with the rod
diameter in a double acting device determines the ring or annulus area on the rod side of
the piston. Manufacturers usually specify the actuators with a table for the cylinder bore
In a single acting device this will also be the extend force available with the working
area either that of the piston as above or in the case of the true displacement cylinder of
Figure 2A.5.1 the rod area should be used. In a double acting device if return line pressure
is included producing a pressure on the annulus area of pA, then the force balance
becomes:
Thus a higher pressure on the piston would be necessary to give the same output force
(and perhaps a higher supply pressure). If the external force, opposing the motion as
shown, is graduaIly reduced then the piston pressure would also reduce to maintain the
balance. When the external force is zero then the two pressure forces must balance even
though this will require pp to be smaller than PA. If the external force becomes negative
it is then assisting the motion and for the equation to balance the annulus pressure pA, must
be considerably higher than the piston pressure pp. All these conditions are acceptable but
suitable valves must be used to control the return flow out of the actuator and establish the
necessary pressure at PAwhen this is controlling the motion. This is disc used further with
the examples in the Linear circuits chapter.
The maximum force which can be produced by an actuator is usually called the stall
thrust. In the extend direction this wiIl be obtained when the annulus pressure is zero and
the piston pressure rises to its maximum limit determined by the supply pressure ps. The
stall thrust will then be the product of ps with Ap. The term stall implies that the piston is
brought to rest by an increasing external force but a low friction actuator can produce a
thrust of almost this value even when moving if the supply pressure is maintained.
A force balance for movement in the retract direction wiIl also give this last equation.
The relative magnitudes of pp, and PA will again depend on the size and direction of the
external force foIlowing the same arguments as given above. The maximum force in the
retract direction will again be a stall thrust and now equal to the product of ps with AA.Thus
this will always be the lower value for an unequal area actuator.
A consistent set of units must be used with all these equations but this is relatively
straightforward. Note that a pressure specified in units of MPa implicitly includes a term
106, whereas a pressure in bar should be multiplied by 105 to convert to N/m2.
Pressure loss effects will, of course, also be present together with friction in both piston
and rod seals. Such performance losses may be small, but can appreciably modify the
theoretical performance in practice, and usually cylinders are sized on theoretical
performance, with a nominal aIlowance for such losses. These aspects are discussed in
more detail in this chapter under the Selection & peiformance heading.
In cases with a large number of cylinders in the extended position at the same time a
significant volume of fluid may be transferred from the tank and 'stored' in the cylinders.
The hydraulic power necessary to provide this will be greater. There will be a friction
loss in the actuator and pressure losses in the flow. These can be estimated or calculated
more accurately if the actuator and circuit details are known. These are discussed below
under the Selection & peiformance heading.
Cushioning
Many cylinder applications demand fast movements, if a large mass is being moved then
a large shock load can occur if the piston is stopped abruptly by hitting an end cover when
it reaches the end of its stroke. To eliminate, or at least substantially reduce the shock loads,
a more gradual deceleration is required as the piston approaches the end of its travel. This
is achieved by modifying the piston and end covers to provide cushioning. There are many
physical arrangements to provide this effect but they are all based on the same principles.
A cushion slows the final movement of the piston by restricting the fluid escaping from
the outlet port. The restricted flow provides a pressure force to give the necessary
controlled deceleration. A typical arrangement is shown schematically in Figure 2A.5.6.
The end cap provides an extension of the cylinder but with a substantially reduced bore.
The leading face of the piston is fitted with a reduced diameter piston extension (plunger,
pin, spigot or spear) which enters the cushion bore as the piston approaches the end of the
stroke. This effectively seals off the main path for the fluid flow to the outlet port. The fluid
trapped by the cushion piston escapes partly through the clearance between the piston and
bore but also through an orifice which is usually an adjustable needle valve. As the flow
is throttled high pressure rapidly builds up and decelerates the piston and load. The orifice
adjustment may be preset (for fixed throttling), or left adjustable for the user to set the
degree of throttling required to suit a particular application.
To initiate return movement, provision must be made to by-pass the orifice, otherwise
the fluid flow to the annular face of the piston would be so restricted that the initial
movement would be very sluggish. This by-pass usually takes the form of a ball type check
valve which provides a parallel path for the initial return flow. Not until the cushion piston
has left the cushion bore will the full flow at the inlet pressure be available to the piston.
This simple orifice cushioning produces the highest decelerating pressure just after the
spear has fully entered the bore. Peak pressures of over 40 MPa (400 bar) can easily be
produced and excessive pressures may cause fatigue problems in repetitive cycling. The
piston will decelerate most rapidly at first, following an exponential shaped curve as
shown in Figure 2A.5.?, until the cushion pressure force establishes a balance with any
driving force and the piston will then continue at a constant low speed until it reaches the
end of its stroke. There is usually some lead-in before the parallel section of the cushion
which will slightly reduce this initial peak pressure. However, the cushion itself may be
tapered for a significant length to provide another flow path through the gap between the
cusnion piston and the bore itself which will be in parallel with the flow to the orifice. Peak
pressures then occur later in the engagement and are lower, although the average pressure
will be maintained at a higher level for longer. Some designs refine this with axial grooves
on the cushion piston or step changes in the diameter. The small clearance annular spaces
between the cushion piston and the bore may well give a laminar flow regime for the fluid
which means that the pressure produced will be dependent on fluid viscosity and hence
temperature. Designs try to minimise such effects to give consistent cushion performance.
An alternative is to pass the flow out of the cushion section through a relief valve. This
must still be set to a relatively high pressure to give the necessary deceleration but it will
give a constant retarding force and hence more moderate action during the initial part of
the cushioned stroke. Since this relief valve is set to a high pressure the piston may well
stop before the end of its stroke. This can be overcome by keeping the restrictor in parallel
with the relief valve. Yet another alternative is to use a "choke ring" construction which
can also be modified to give an almost constant pressure through the cushion stroke. An
example is given in Figure 2A.5.8.
Tapering of the spear will take up small misalignments between the piston and cylinder
tube but alignment is also ensured by use of floating components to give a self aligning
feature. These may be either floating pistons on the spear itself or a floating bushing ring
in the end cap.
Cushioning may be applied to one, or both, ends of a double-acting cylinder and
standard size hydraulic cylinders are commonly available with or without the cushion
option. In most standard ranges of cylinders the addition of a cushion will increase the total
body length. There can be considerable differences in the cushion length between the
various designs and some manufacturers maintain the same build length for cushion and
non-cushion versions. In addition different cushion lengths and designs may be available
for the same cylinder to suit specific requirements.
Cushioning may be used most successfully on cylinders where speeds are over about
0.1 m/s but it may be necessary with a very large mass at lower speeds. To be effective,
the pressure developed in the cushion chamber must be higher than the fluid pressure
driving the piston. In adjusting the needle valve to give a rapid retardation it is possible
to set the throttle to give excessively high fluid pressures within the cushion chamber. This
can be even more damaging than not having a cushion at all and could cause a premature
fatigue failure. There are therefore some dangers in using cushioning with very high loads
or speeds where a controlled deceleration with external valve arrangements would be
more appropriate. It is really the initial kinetic energy which is important rather than a
speed. Manufacturers produce guidance charts indicating the maximum cushion capabil-
ity in terms of initial energy and driving pressure as shown in Figure 2A.5.9. The energy
can be calculated from:
Line Connections
In relatively few applications is a hydraulic cylinder completely static (even with rigid
mounting the cylinder itself may have motion as a unit). This means that the most common
method of connection for inlet and outlet lines is by flexible hose. Provision is usually
made for this by female tappings in the end covers at each port position capable of
accepting standard couplings, or a male stub pipe accepting a proprietary sealing ring and
nut. The same tappings can be used with either flexible or rigid lines. SAE flange versions
are also a possibility. The physical positions of the inlet and outlet ports are separated by
the length ofthe cylinder but can usually be orientated differently around the cylinder axis.
To make for a neater (and often less vulnerable) assembly, both line connections are
sometimes brought to the mid-point or one end of the cylinder.
In order to encourage air out of a cylinder the input connections should logically be at
the top, or facing upwards. This may not always be possible and although the fluid
movement usually entrains air and carries it out of the cylinder within relatively few
strokes there are cases where an air bleed is necessary or at least a desirable precaution.
These are easily incorporated into the end covers.
Components and construction

At first sight an hydraulic cylinder may appear a very simple and crude device but there
is an enormous quantity of expertise, specialist knowledge and technique in their design
and construction to give trouble free operation. This includes aspects of the materials used
and their manufacturing processes as well as the construction methods and sealing
systems. An introduction to these features is given below.
Cylinder tube
The early methods used for the production of high pressure cylinder barrels were casting
and forging, and these are stiII used for some larger sized high pressure cylinders. Cast,
forged and also extruded cylinder bodies have the advantage that one end-cover can be
formed integral with the cy linder tube, thus saving one joint and a source of leakage. The
disadvantage of construction with an integral end-cover is that it is more difficult to finish
machine the bore. It is too easy to leave machining marks in the blind end of the tube, and
honing is equally difficult in a blind hole. In view of the importance of surface finish for
satisfactory seal performance, cast cylinders may be cast as open tubes so that they can be
finished by through machining and polishing.
These days cylinders are constructed using either DOM (drawn over mandrel) or
seamless tube. The latter may be referred to as Cold Drawn Seamless or Cold Finished
Seamless, where the material is formed from a solid biIlet and the last stage in the
manufacturing process involves working the material when cold (Note: it would not be
cold to the touch). DOM tube is made from flat strip seam welded to form the tube and then
drawn to size. This process has cost benefits with usually very high quality. Cold working
increases strength and also tends to promote better machinability of the material, but can
increase the tendency towards distortion when machined. All cold-drawn tubing is subject
to some degree of ovality, however, very close tolerances can be held by suitably
controlled production methods.
The internal surface finish of the bore may be adequate with a fine finish drawing
process giving Ra values down to 0.1 11m. Honing typically produces surfaces finishes
within an Ra range from 0.1 11mto OAllm but adds to the cost of manufacture. A skived
and roller burnished tube is cheaper and may have a smoother finish below 0.1 /lm.
However, this may be too smooth since fluid trapped in micro-grooves can help with
lubrication, reducing friction, especially after a period with the piston stationary. Even the
surface finish measurements are not a sufficient guide because the production methods
give completely different characteristics. A honed surface, with a cross-hatched pattern
of fine grooves will give the best sealing and a longer seal life.
With water content fluids some form of protective treatment may be considered
essential to combat corrosion, even though the fluid may contain corrosion inhibitors.
Bore surfaces may be protected by nickel or chromium plating (subsequently polished) or
by other protective treatments. This is seldom necessary in the case of cylinders intended
for use with oil fluids, although plating or nitriding may be used to give a harder and more
scratch resistant bore surface.
For working in corrosive surroundings the outside of the cylinder tube may be protected,
either by surface coating or treatment, or by jacketing in a corrosion resistant material.
Austenitic stainless steels can be used for the tube and give a complete stainless steel
cy linder assembly. The use of plastics for tube construction is also a possibility. Provided
that the cylinder is reasonably small and the working pressure not too high, otherwise the
lower modulus of plastics can cause significant problems for the seals with the higher
radial strains giving excessive clearances. Plastic cylinders have been used successfully
for marine applications, as an alternative to stainless steel or a resistant aluminium alloy,
where the cylinder is exposed to salt spray. Double-tube construction has also been used
to protect the main (inner) cylinder from impact damage, or where 'armouring' offers
obvious advantages.
Strength of cylinder tube

The cylinder wall thickness usually depends on the pressure rating of the cy linder but may
also be influenced by the form of construction and/or the application. For heavy-duty or
arduous use a heavier wall thickness may be preferable for rEliability and life and hence
may be the most economic proposition in the long run.
Where the diameter to thickness ratio of the cylinder tube is greater than 16: 1, the stress
produced in the wall material due to internal pressure can be determined from the simple
formula for uniformly distributed hoop stress:
Pistons
Most pistons are made from either cast iron or steel, but other materials are used incl uding
brass, bronze, aluminium alloy, and also sintered iron and steel. Ferrous metal pistons may
be bronze faced, or hard chrome plated and polished. For use with water-based fluids
bronze pistons are a common choice, although plated or surface-treated iron or steel may
be employed as well as stainless and corrosion resistant steels.
The majority of small size pistons are a single piece design but this depends on the type
of piston seal used and the manufacturer's preference. The type of seal will normally
dictate whether the piston has to be designed as a one-piece, two-piece or three-piece
assembly in order to fit the seals. Although most manufacturers will supply alternative seal
arrangements for a particular cylinder these are not necessarily interchangeable. There is
a standard for seal cavities ISO 7425 but the alternative seal sets may be packaged very
differently. The choice of seals will in general be a compromise between friction forces,
maximum speed, seal life and cross seal leakage. However, some pressure activated seals
now provide a good combination which fulfils most applications.
Single piece construction can be used when pistons are fitted with simple elastomeric
pressure energised ring seals or cast iron (automotive type) piston rings. The former
normally .providing the simplest and least expensive form of piston for smaller cylinders
but are usually limited to velocities below 0.5 mls. Cast iron rings are noted for their long
term durability and give a higher speed capability but the leakage makes them inappropri-
ate for supporting loads in a fixed position for a long time when isolated from the hydraulic
supply. Some pressure energised seals with a low friction PTFE (Teflon) bearing surface
as shown in Figure 2A.5.lO, give very low friction, are capable of operating at lmls and
have very low leakage. This figure also shows a bearing insert to prevent metal-to-metal
contact between piston and tube bore.
Cup seals or chevron seals are generally preferred for larger sizes or heavier duties with
full stroke operation. These again are pressure activated with moderate frictional resist-
ance and speed limitations of about 0.5 m/s. Chevron seals have good long term reliability
but higher friction. To fit these to the piston it needs to be constructed in several pieces to
allow the individual seals and headers to be mounted in position in matching grooves.
These seal types are only active in a single direction and so two sets have to be mounted
as opposite pairs to provide sealing on a double-acting piston.
Special seal packs are available which give higher speed operation. These can give
speeds of over 2 mls but are only available as specials in consultation with the manufac-
turers. Seal sets are usually available as service spares if necessary. It is very important to
make the correct choice of seal materials in relation to the working fluid. Compatibility
is the subject of a standard, ISO 6072, but detailed discussion with the manufacturer is
recommended for anything that is in any way non-standard. There is further discussion on
compatibility in the chapter on Seals.
In addition, the type of seal used (both piston and rod) is usually the factor which
determines the operating temperature range of the cy linder. Normal limits are from -20C
or -30C to a maximum between +80C and + 100C. These limits can be extended with
special low temperature seals to typically -45C and with high temperature seals to
typically + 150C.
There are two common methods of attaching the piston to the rod, and in both cases the
rod is threaded. In one case the piston has a matching internal thread and screws directly
onto the rod. This is then locked in place with both a thread locking compound and some
form of pin. The other main alternative is to machine a shoulder on the rod with a smaller
diameter threaded end. The piston then locates on the shoulder and is held in place by a
nut, a cushion piston assembly or for a double ended cylinder the other rod may be used
to clamp the piston assembly. For heavier duties the piston may be welded to the rod, and
for lighter duties simply located by circlips.
Rods
Piston rods are normally made from high tensile steel bar, with a hard chrome plating
followed by polishing. Induction surface hardening of the bar reduces the severity of
accidental damage to the rod which might cause lifting of the chromed surface. It is used
in high quality cylinders and is necessary in some applications where the rod is exposed
and vulnerable. A distorted or dented rod will also reduce the life of the rod seals. Plating
is highly desirable to prevent corrosion, which again would result in mechanical damage
to the rod seals and severely reduce life. Plating depth is usually of the order of 25 /lm. A
high quality surface finish is desirable in order to minimise wear on the rod seals and Ra
values are usually between 0.15 and 0.5/lm. Higher grade nickel chrome is also used.
The quality and integrity of the surface finish of the piston rod contributes significantly
to the overall reliability of the cylinder and the system as a whole. It must be preserved even
in adverse environments. Polished stainless steel (chromed) is sometimes used for rods to
provide maximum resistance to corrosion, particularly for use with other than oil fluids.
There are also coating materials which can be useci, including ceramics. This is provided
on larger cylinders for use in inhospitable environments where sea water or chemicals can
splash onto the exposed rod. It offers the unusual sight of a piston rod which is black in
colour see Figure 2A.5.11. The other way to protect a rod is to use a rubber bellows cover
as shown in Figure 2A.5 .12, but the 'pumping' action of the bellows can also draw dust
in.
Rod sizes for the common sized production cylinders are normally about one half the
bore diameter, giving a ratio of piston to annulus areas of 4:3. With such geometry back-
pressure effects are generally negligible for ordinary working. Cylinders are also usually
offered with alternative sizes of rod: smaller in diameter for light duty or fast return double-
acting cylinders, where the ratio is usually of the order of7:6; larger in diameter for heavy-
duty application where the area ratio may range from about 7:4 to 2: 1. ISO 3320 specifies
both bore and rod diameters. The full range is available but particularly for smaller
diameter cylinders the combinations vary considerably between manufacturers. There is
more commonality for sizes over 100 mm bore.
End Covers
The end covers perform a number offunctions within the cylinder assembly and can house
"extra facilities". They must provide a structural pressure tight seal at both ends, with
hydraulic connections and possibly cushioning and bleed facilities. Additionally at the rod
(or head) end there are also the rod sealing and bearing arrangements. Frequently the
cylinder mounts are also incorporated into the end covers. End covers are usually, but not
invariably, made from the same material as the cylinder tube. Steel, sometimes phosphated,
and spheroidal graphite cast iron are the most commonly used materials. The choice may
also be dependent on the method of attachment to the cylinder tube as discused below.
The cy linderrod has to be both supported and sealed by the rod end cover (or both covers
in the case of a through-rod cylinder). A simple plain bearing is adequate, and is frequently
a bronze or SO cast iron insert into the end cover. There is a provision for two sealing
elements. The inner one providing a sealing effect to prevent fluid leaking out, generally
a pressure energised design (seal types are discussed in more detail in the chapter on Seals
and sealing). The outer element will be acting as a cleaning device, a wiper or scraper ring
to prevent dirt and other contaminant entering the cylinder. Abrasive dust collected by
the rod could otherwise be drawn back into the main seals and damage both rod and
seals.
The rod bearing and seals are usually removable and are available as service spares.
These can be changed provided the rod itself is in good condition. To simplify mainte-
nance, they may be combined in a single removable section as shown in Figure 2A.S.B.
This is generally referred to as a cartridge gland unit.
Standard production of cylinders may offer different types of rod seals (fitted in
alternative end covers), and even different bearing lengths, to suit different duties. It is also
possible to get end covers with "oversize" ports when high speed operation could cause
excessive pressure drop in the standard size. Another option is a head cover which
incorporates a drain connection between the main pressure seal and the wiper ring. In very
sensitive applications this should prevent any minor leakage from the main seal escaping
along the rod. A second fluid seal can also be included. Note that as a "drain" port it should
be connected directly to the tank via a line which will not experience any back pressure
effects. The normal return line is not suitable.
It is also possible to build into the end covers both air bleeds and adjustable mechanical
stops to control the limits of the piston travel. These latter are useful in some applications
where the full stroke is used and the final position needs to be set with some accuracy.
There are several different design implementations available depending on the accuracy
and frequency of adjustment. These are usually called stroke adjustors and the adjustment
range might typically be 2 to 3 times the bore diameter.
Assembly
There are several variations in the method of fitting the end covers to the tube each with
their advantages and some limitations. The main categories are:
(i) Tie-rod
(ii) Screwed-on ends
(iii) Flanged-and-bolted covers

(iv) Welded construction
Tie-rod construction, shown in Figure 2A.5.14, is probably now the most widely used for
small and medium sized cylinders. It offers a simple and inexpensive solution and length-
for-length compares favourably with other types. Particular advantages include - there is
no weakening or distortion of the tube wall; it allows a very wide range of alternative
mounting styles and optional features from a standard range of components. Tie-rods also
increase the bulk and do not provide the 'clean line' required in some sectors, such as food
processing. They may also be liable to damage when used on long cylinders. With a four
tie-rod assembly, end covers are commonly made square so that each corner forms a
convenient point for fitting one tie-rod parallel to the cylinder tube and the flat faces
provide suitable surfaces for port connections. Heavier duty designs employ more tie rods
with circular covers.
Cylinders with screwed ends have an advantage in diametral size over other types
(except welded), but usually require a thicker cy linder to accommodate the screw thread
without weakening the ends of the tube. They are commonly employed on smaller
cylinders where there is ample cylinder tube wall thickness for an internal thread, which
produces a very compact result as shown in Figure 2A.5.15. A leak-tight joint can be
provided by the inclusion of a simple ring seal in the cover. One disadvantage of this
method is that it is difficult to ensure correct angular alignment of the two heads after
assembly or servicing. Where an external tube thread is used it requires larger diameter
end covers but they do provide hoop support to the tube ends as shown in Figure 2A.5 .16.
Where the tube wall is of sufficient thickness the cover may be attached directly by
means of tapped holes in the cylinder tube and machine screws as in Figure 2A.5.17. This
method, although neat, can be costly on a long cylinder, and is strictly limited in the
diameter of threaded hole which can be accommodated without seriously weakening the
cylinder tube (alternatively it may call for an excessive wall thickness for the pressure
rating concerned).
All-welded cylinders are generally cheaper to produce, but are basically non-service-
able. This method results in a minimum depth of cover and eliminates the need for static
seals as shown in Figure 2A.5.18. It does, however, have certain limitations, in particular
the choice of material for covers which must be easily weldable and mild steel is normally
used. They are most widely used on mobile equipment.
Figure 2A.5.19 shows the use of flanges or tube rings, welded to the cylinder tube with
bolted-on covers and is one of the strongest forms of construction. Its disadvantage is that
it is more costly to produce and extreme care has to be taken during welding to avoid
distortion of the cylinder barrel. Cast cylinders, however, can be produced with flanges.
This type of construction is still used for some heavy-duty larger size cylinders.
An alternative flush flange fitting can be produced by other means of attaching the ring
to the cylinder tube. It may be screwed on an external tube thread and locked in place as
shown in Figure 2A5.20. A similar system which is more economic to produce employs
a tube ring positioned by a wire or circ1ip situated in a groove cut in the cylinder wall as
in Figure 2A5.2Ia. Yet another method employs a wire which fits into matching semi-
circular grooves machined in both the cylinder tube and cover as in Figure 2A5.21 b. The
wire is introduced through a hole, drilled tangential to the groove after the parts are
assembled, the parts being rotated to bring them into the correct position and, if necessary,
secured with a set screw. This type of fixing makes it difficult to remove the cover in the
event of corrosion of the wire and to avoid this either stainless steel wire or heavily plated
wire is preferred.
Hydrostatic bearings
Hydrostatic bearings are an alternative to plain bearings and since the number of dynamic
seals is also minimised then this design reduces frictional forces to a minimum. Hydro-
static bearings are designed to produce a self centring action which can eliminate any
physical contact between the bearing and its journal bore. They are, however, relatively
expensive and use is restricted to applications where very precise motion or position
control is required, such as machine tools, testing machines and flight simulators. The
principle of a hydrostatic bearing can be illustrated with the simple single tapered journal
shown in Figure 2A.5.22. This could represent a conical piston in a cylinder bore.
If the pressure on side A is higher than the pressure on side B, then a flow will pass in
the annular space between the piston and the bore in the direction shown. The pressure will
fall from P A to PH along the length of the piston but the majority of the pressure drop will
occur in the narrowest space on the left of the piston. If the piston is in the centre of the
bore then the pressure distribution acting around the piston circumference at any section
along its length will be the same. This is demonstrated by the radial arrows shown in Figure
2A.5.23a, all having the same length. If the piston is offset in the bore then this will modify
the pressure distribution around the circumference, the pressure falling where the
clearance is increased, and rising on the opposite side as shown in Figure 2A.5.23b. This
produces a force on the piston causing it to move back to the centre line - the self centring
effect. Note that if the pressure at PH is the higher then the reverse effect will occur ie any
slight offset will cause a de-centring force and the piston will lock against the bore.
A single cone bearing is also susceptible to a limit on this centring action if it is moving
axially in the bore. This effectively superimposes another flow term and if the piston
shown in Figure 2A.5.22. is moving to the left at a velocity greater than the velocity of the
fluid in the annular space then the self centring effect would be lost. The application of this
principle in practice then leads to the double cone design for both the piston and the rod
bearings as shown in Figure 2A.5.24. This gives a centring effect in all three bearing
regions whichever direction the piston and rods are moving.
It should be noted that there is a permanent suppl y to the rod bearing with a drain to tank,
and also a drain taken from the centre of the piston bearing. There is a continuous flow
through these paths to "activate" the bearing action. This can usually be kept in the region
FIGURE 2A.5.24 - A
practical application of the
hydrostatic bearing principle
of 2-4% of the maximum flow taken by the cylinder. The minimum clearance is kept as
small as possible, with suitable tolerances, within manufacturing limitations. The taper
slope on the bearings is of the order of I in 700. If this type of cylinder is used to hold a
load in a fixed position then fluid must be supplied to make up the bearing flow. These
actuators are usually used in servo systems with position feedback and the control loop will
automatically open the control valve to supply this flow.
Cylinder mounting
The mounting of the cylinder body is an important factor which interacts with both the
sizing, and life and reliability of the installation. Mounting is critical in determining the
buckling strength of a compression loaded cylinder as described later in this chapter. The
mounting style is also influenced by the type and construction of the cylinder. Mounting
may be either rigid, where the body does not move relative to its mounting structure or
pivoted where rotation relative to the structure is permitted. Pivoting mounts are
normally, but not invariably, mounted on the centre-line of the cylinder.
Mounting types have been classified in ISO standards according to the pressure rating
of the cylinder (ISO 6020 and 6022). The basis of the classification system is summarised
in Table 2A.5.3. An example of some of the typical styles and their designations is given
in Figure 2A.5.25.
It should be noted that some of these mounting designs react the load forces along the
cylinder centre line whereas the side mounting alternatives will react with an offset. These
latter are subjecting the piston and rod to a side load which increases wear on all bearing
and sealing surfaces and considerably reduces the buckling strength of the cylinder. The
load offset is worst for large diameter and short stroke devices and an alternative mounting
style should be sought unless the rod end can be guided. The manufacturers may impose
restrictions on the maximum pressure for the use of this type of mount. However, these
mounts are also available with an extension to the mounting plate which gives a thrust key
function (as shown in Figure 2A.5.25) to give positive location, react loads and prevent
movement of the cylinder body.
There are fewer alternatives for rod attachment. Rod ends are sometimes screwed (ISO
4395) which allows a number of fittings to be attached or alternatively those may be
welded. The usual alternatives are plain, flanged and a pinned end which may be either a
plain eye tongue, a clevis fork or a spherical eye bearing. The latter are capable of taking
a small degree of mis-alignment, up to 3 or 4, without showing undue wear or imparting
excessive side loads to the cylinder. Some possibilities are shown in Figure 2A.5.26.
The majority of cylinder manufacturers aim to provide as many alternative mountings

as possible on standard production cylinders so that most common applications can be
catered for without need of expensive special designs (which are usually also available).
Choice of mount
The choice of the best mount sty Ie can have a considerable effect on the life of the actuator,
although frequently it may be determined by the physical constraints in the application.
Rigid mounting may be necessary where the cylinder acts as a rigid structural member of
the machine in which it is incorporated otherwise a pinned mount may be preferred to
simplify alignment and provide a certain tolerance to subsequent misalignment. Some of
the factors affecting selection of suitable mountings are listed below:
(i) Cylinders with mountings not on the centre-line will have offset loads resulting in
extra wear, shorter life and reduced buckling strength.
(ii) It is vital that the structure holding the cylinder must be strong enough to react the
maximum load without excessive deflection. Mountings apply high concentrated

loads and possibly bending moments which must be reacted by the structure.
(iii) When the object moved by the piston rod travels in an arc then a pinned type
mounting must be used, but a fixed mounting type cylinder should normally be used
when essentially linear motion is required.
(iv) Buckling strength must be considered against stroke length (see below). For long
strokes cylinders with fixed mountings can usually have smaller diameter rods
without danger of bending compared with pivot or centre trunnion mounted
cylinders.
(v) The maximum load applied must result in a maximum tension or compression force
in the rod but forces in the cylinder body can be avoided. For thrust loads
(compression) the cap end flange reacts the load directly through the end cover. In
tension a head (rod) end flange mounting will react the loads directly.
(vi) The possibility of misalignment between a cylinder and whatever it operates must
be considered. If the misalignment is primarily in one plane the simple pivoted
centre-line mounting will give the necessary compensation. A spherical eye
mounting adds a further allowance.
There are several other features of mounting which should also be considered. Fixed
mounting cylinders should be keyed or pinned, provision being made at the design stage
of the machine. If the appropriate member of a machine is thick enough to take key-ways,
cylinders with integral key-mounts can be used. These should be at the correct end of the
cylinder. Only one end should be keyed to the machine to avoid losing the advantages of
cylinder elasticity. Temperature and pressure effects should also be considered. Locating
pins can be used instead of keys to take shear loads and maintain alignment, again at one
end or the other but not both.
Critical Rod Lengths

Cylinder rods can be stressed as rigid rods, provided the length of the rod does not exceed
ten times its diameter. The stress formula in this case is:
For general working the maximum permissible material stress may be based on the
ultimate tensile stress of the material and a suitable factor of safety usually between 3 and
4 depending on the anticipated application duty. This will normally give adequate rod
strength either in tension or compression.
If the rod length exceeds ten times its diameter, then it may be subject to buckling under
compression loading. Adequate strength in tension can then no longer be taken as an
indication of adequate strength in compression. The case of compression loading must be
analysed separately, when the rod is considered as a column. The material stress then
depends on the method of end fixing, viz.
Although it is possible to apply these equations manufacturers usually provide a more
convenient and quicker approach for selection. Manufacturers specify this detail in tabular
or chart form, the latter being more compact and, although appearing complex, is quite
easy to use. A typical set are shown as examples in Figures 2A.5.27 and 28. The first of
these gives an idea of the importance of mounting style in determining the buckling limit.
The latter then gives safe rod length and diameter information for steel rods, assuming that
there is no eccentric loading. In the case of horizontal cylinders, these may need supporting
to counteract bending movements. Where bending loads are present in the rod as well as
a compression load, the critical length is reduced substantially. Alternative formulae can
be used to calculate the required strength but this should be done in consultation with the
manufacturer.
The chart shows two slopes for each rod diameter line. The flatter slope at higher loads
is based on the basic rod strength with an increasing factor of safety as the stroke length
increases (a constant safety factor would give a horizontal line). The steeper portion of
each line at the longer lengths is determined from the buckling formula.
There are many ways to use such charts depending on the starting constraints of the
design. The most likely situation is when maximum force and the required stroke are
known as well as the preferred mounting style, and an adequate rod diameter is to be
determined. Figure 2A.5.27 should be used first to calculate an equivalent length LB, for
a mounting style and stroke length. This value can then be used with the required
maximum force as the coordinates in Figure 2A.5.28 to determine the minimum rod
diameter. NOTE this is an example chart and manufacturer's own data sheets should be
used for selection purposes.
The way in which the cylinder is mounted and the manner in which the rod is supported
at the extremity of its movement will determine whether or not a stop tube can be
advantageous. At first sight this may not seem to be an advantage since the full stroke
cannot be used, however, this can be compensated by selecting a slightly longer initial
working stroke. The result is structurally preferable but here again it is advisable to seek
advice from the manufacturers.
The bearing length of a cylinder may be increased, when fully extended, by the fitting
of a stop tube as shown in Figure 2A.5.29. An alternative is the use of a dual piston, which
effectively splits the piston with a space between the two halves and increasing the bearing
length.
Cylinder selection and performance

For both single and double acting cylinders the first stage in any selection is to determine
the performance requirements. These should include the required maximum compression
and tension forces, the operating stroke and either a maximum speed or a cycle time (to
calculate average speed). In addition values for maximum system pressure, or any limits
on this, and the maximum mass (or weight) which must be moved. Obviously any
peculiarities of the working fluid or the operating environment will also constrain the
alternatives and should be considered at the outset. A recommended procedure to be used
in conjunction with manufacturers' data on available sizes and performance limits would
be:
i) Determination of working areas from the required forces and maximum available
pressure to give an initial estimate of bore and rod diameters.
ii) Establish flow requirements from the areas and speeds.
iii) Check seal packs available and their suitability for the application (eg. speed, fluid
& temperature).
iv) Select mounting style including rod end fixing.
v) Check minimum rod area for buckling constraints, consider change of mounting or
use of stop tube.
vi) Establish requirement for cushioning and select as appropriate.
vii) Consider any other special requirements ego larger ports.
As with all such design procedures iteration within these stages may be necessary to
produce the best result.
The basic formulae quoted earlier in this chapter give theoretical values for speed and
force output. The following formulae can be used for 'back of the envelope' estimates of
cylinder diameters to meet the force requirements, with an allowance for a 'typical'
pressure drop and frictional losses:
where F = design force (N)

P = available pressure (MPa)
d = rod or piston diameter depending on single acting design (mm)
dp = piston diameter for double acting cylinder (mm)
dR = rod diameter for double acting cylinder (mm)
Pressure drops in both supply and return lines will occur at high flows, and although
these are sometimes neglected they may contribute in the order of 1 to 4 MPa (10 to 40 bar)
or more, which with seal friction could be significant in critical designs. For more exact
working the theoretical formulae should be used, with pressure drop more accurately
determined and back-pressure effects also included for double-acting cylinders. Friction
is also important, and this is discussed below.
Friction in cylinders
Details of the frictional forces in actuators are not well documented partly because there
is considerable dependency on factors which are difficult to estimate. The dominant
contribution to friction undoubtedly comes from the total seal pack including the wiper or
scraper ring. Friction depends very much on seal type, surface finishes and operating
circumstances, and there is significant variability associated with temperature and wear
related effects to further confuse the issue.
There are two situations which must be considered- stationary and moving. The friction
forces in starting a movement from rest are likely to include significant stiction type
effects. Stationary seals tend to lose the lubricant film which exists between the seal and
the sealing surface, particularly if the seal is pressure activated and under load. The time
spent at rest will also affect this. Once a seal is moving then the fluid will help to lubricate
and reduce friction. When moving there may still be a load dependent effect ie. friction
force dependent on pressure, but there will also be a speed dependent component or
viscous friction. The seal type and materials play a part in all these effects but particularly
the pressure dependent terms. F]uid viscosity also plays a role and can introduce
temperature dependency. The wear and running-in of both sea] and surfaces can also
interact to lower the friction. Tab]e 2A.5.4 gives an estimate ofthe range offrictionalloss
present as a proportion of the maximum theoretical force when the actuator is moving. In
starting from rest some seal types will give additional frictional resistance.
The effect on friction of the various physical factors discussed above can be predicted
qualitatively and used to indicate which end of the above scales to bias any estimate.
Additiona] points include:
i) The scraper ring can playa significant part in the friction levels, hence the double
ended designs have higher friction. Low friction scrapers are available.
ii) Larger bore pistons tend to have a lower frictional loss in relation to the theoretical
thrust since the sea] contact length increases in proportion to diameter whereas the
theoretical force increases with the square of diameter.
iii) For the same bore diameter the largerrod diameter will have higher friction because
of the larger surface contact length.
iv) For all double acting cylinders, although the friction force may be the same in each
direction, the pressure required to produce this for retract movement will be greater
because of the lower area.
These considerations can lead to an allowance in calculations for combined hydraulic
system pressure loss and actuator friction loss. Any such allowance should be checked
when circuit details are finalised but even then actuator friction is likely to remain an
estimate unless specific details are available from the manufacturer. Some manufacturers
now give sufficient catalogue detail for friction to be estimated depending on component
diameters, type of seal pack and operating pressure.
For the circuit aspects of cylinder performance, see the chapter on Actuator Perform-
ance in Section 4. Further details on the types of seal and their contribution to friction are
given in the chapter on Seals.
125
ROTARY ACTUATORS
ROTARY ACTUATORS are devices which produce a high torque output and have a
limited rotary movement. They fall into two main categories:
(i) Vane type actuators where the torque and motion is all produced in a rotary sense.
The construction limits the rotation to less than one revolution.
(ii) Piston type actuators are essentially linear attuators mechanically connected to
translate the linear force to produce an output torque and rotational movement.
These devices are capable of providing an output motion of one revolution or more,
but not continuous rotation.
All types give bi-directional output motion, and most produce the same torque in both
senses. Also output torque is generally constant throughout the stroke. The particular
advantage offered by hydraulic rotary actuators is direct rotary output from a compact
hydraulic device with elimination of linkages and lost motion associated with cylinder/
crank rod arrangements. The symbol for a hydraulically operated rotary actuator is shown
in Figure 2A.6.1.
Vane Actuators
The construction of a simple single vane actuator is shown in Figure 2A.6.2. It takes the
form of a cy lindrical housing through which passes a central shaft to which the vane is
rigidly attached. The housing itself has a second vane or abutment shoe fixed to its internal
diameter and extending to the output shaft, thus dividing the interior space into two
chambers. Hydraulic fluid is supplied to these chambers through connection ports close
to the fixed shoe. Pressure difference between the two chambers and acting on the moving
vane area produces the torque directly on the output shaft. Flow into and out of these
chambers causes the vane and shaft to rotate. Design geometry normally limits the rotary
movement of a single-vane actuator to about 2800 maximum.
126 THE HYDRAULIC HANVtlUUK
It is possible to modify the design to have two vanes fixed to the output shaft 180" apart.
and two fixed shoes in the casing providing two separate operating halves each with two
chambers as shown in Figure 2A.6.3. Each half of the actuator is connected to the supply
so that they operate in parallel. This gives twice the maximum torque output of a single
vane device for the same supply pressure. Obviously the maximum angle of rotation is
reduced and because of the second shoe only 100 is usually possible.
These designs produce a very compact body size for the torque available. Body
mounting can be either end face or side fixed. Shafts can be keyed or splined and the drive
may be taken from one end or with a through shaft, from both ends of the actuator. It would
be possible to fix the shaft and allow the body to rotate, with suitable provision for the
hydraulic connections which could be taken through the shaft. A typical design is shown
in figure 2A.6.4.
the inflow or outflow. Speed is seldom an important factor for such devices, which are
normally used for high-torque outputs. High speed operation can introduce problems with
high inertia loads similar to the cushioning requirements in linear cylinders.
The most critical feature of the vane actuator is the length requiring sealing around the
vane between the end faces and the internal bore of the casing. Since this is a moving seal
there is the usual compromise between good sealing with low leakage and resulting higher
friction. The sealing surface with the bore is also at the largest radius and operating at the
highest sliding speed and so good design is key in this area. Seals are usually pressure
activated and may give a non-linear torque relation with pressure. In addition to the vane
seal, further seals are required on the shaft where it emerges from the body of the actuator
and between the shaft and the fixed shoe.
The problem of leakage may also be evident where the actuator has to be locked with
the load held by a pressurised fluid column as in a linear cylinder. The lock will not be
positive due to internal leakage unless provision is made to supply extra fluid to
compensate for the leakage.
The shaft support may be either plain bearings or roller bearings (usually needle roller
type), depending on the application and the design bearing loads. Friction is responsible for
the difference between starting torque and dynamic torque. As a very rough guide the
starting torque can be estimated as 80% of the dynamic torque in good designs, but can be
improved upon if necessary. Overall efficiency of vane actuators is between 70% and 90%.
Piston-type Actuators
All piston based rotary actuators benefit from the use of components used in the more
conventional linear actuator. Many of the details of construction are shared and hence
similar features easily available. There are two main methods used to translate the linear
motion - rack and pinion or a screw thread. In both cases the normal characteristics of these
mechanisms apply with the mechanical losses associated with this translation process.
There are also some designs based on the scotch yoke which gives a non-uniform torque
characteristic.
Rack and pinion piston actuators

The basic construction of a rack and pinion actuator is shown in Figure 2A.6.5. This shows
ROTARY ACTUATORS 129
two pistons connected by a piston rod with teeth cut to form the rack. The pistons move
in the bore of the actuator body as in a linear cylinder type of actuator. The force on the
rod from pressure acting on the pistons is transmitted to the output shaft through the pi nion
teeth.
The relationships for force (and speed) used in the chapter in Hydraulic cylinders can
be modified to rotary motion through the effective pitch radius reff,of the pinion. The
output torque T is given by:
The use of a rack and pinion introduces the potential for some backlash between the
teeth which can affect the repeatability of positioning particularly with loads changing
direction or inertial loads giving overshoot. This backlash can be reduced by pre-loading
the pinion and rack together but with the introduction of greater friction. The shaft bearings
can be easily specified to provide support for side loads and shaft end loads. The shaft is
sealed but this is not subject to hydraulic pressures and both key and splined shaft
connections are obtainable.
Rack-and-pinion actuators incorporating a twin rack with a shared pinion reduce side
thrusts on the shaft and are thus particularly useful for heavy-duty applications. The use
of a second piston also increases the torque capability in a compact package as shown in
Figure 2A.6.6. There are many proprietary variations of such layouts including techniques
to eliminate backlash by pre-loading the racks using only one operating piston on each.
Standard rotations commonly available include 90,180 or 360. The rack-and-pinion
layout offers the possibility of greater rotary motion, in practice piston strokes capable of
generating up to five or six complete revolutions are quite possible.
Operating pressures up to 21MPa (210 bar) are normal with torque capacity ranging
from 20 Nm up to 70 kNm. As with the vane type the torque rating depends on the
application and rapidly fluctuating load cycles will have a lower limit.
The sectioned view in Figure 2A.6.7, shows the similarities in construction between
these actuators and linear cylinders. The end cover arrangement provides hydraulic
connection, bleed screws, stroke adjusters and cushioning with the appropriate piston.
Electrical limit switches are easily included and there is the possibility to incorporate an
electrical position sensor in an end cover extension.
Screw thread piston actuators

The other common design of a piston type actuator has the rotary output provided by the
piston and rod itself. Figure 2A.6.8. shows a schematic view The piston slides on a spline
along the piston rod which is also the output shaft. The piston extension has an external
helical thread which meshes with an internal thread on the ring fixed to the actuator body.
Thus application of pressurised fluid to either side of the piston will generate axial motion
ROTARY ACTUATORS 131
of the piston, which then rotates in the fixed thread with a corresponding rotary motion of
the shaft.
All designs producing movement through a helical thread tend to suffer from relatively
low mechanical efficiency. However, there is the advantage that this type of mechanical
linkage provides extremely good rigidity and irreversibility, using square threads giving
very good load holding capability. Standard rotational output is again 90, 180 or 360.
The length of the cylinder can also be increased, if necessary, to provide more than one
complete revolution of output movement. Operating pressures are again up to 21MPa (21 0
bar) and output torques cover the range from 50 Nm to 50 kNm. Cushioning at the end of
stroke and stroke adjusters are options in several designs. Figure 2A.6.9 shows a sectioned
view of a typical actuator.
A variation on this layout is a double-piston design with a double-helical thread on the

spindle and a third (central) pressure chamber. This has the advantage that the pistons are
always working in opposition and thus thrust loads on the shafts are eliminated. Another
configuration uses a re-circulating ball and thread mechanism.
Many actuators in this category were originally developed to operate plug, butterfly and
ball valves in process industry applications, frequently under completely immersed
conditions. Position indication can also be incorporated, either by a pointer attached to the
top of the output shaft, or by integral electrical switches. There are some designs
employing only a single working plunger with a return spring. This provides for single-
line operation, and also automatic valve return (automatic valve closure) in the event of
a failure of the hydraulics.
Scotch yoke actuators

The arrangement is similar to the rack and pinion designs as shown schematically in Figure
2A.6.10. This shows a guided piston rod connected to a drive shaft through a scotch yoke
which can be considered to give a low friction slider. This means that no significant force
can be transmitted from the rod acting along the length of the yoke (radial to shaft pivot).
The consequences are a high side force on the rod and the torque characteristic shown in
Figure 2A.6.11.
133
HYDRAULIC VALVE CLASSIFICATION
The diversity of hydraulic valves and their combinations for both function and construc-
tion leads to several alternative structures for a classification system. It is possible to
categorise valves:
as either on/off or capable of continuous regulation;
in their broad function as either pressure or flow control devices;
by the moving control element in the valve as either positive seating or overlap
sealing;
in terms of actuation methods there are three principal alternatives: mechanical,
hydraulic or electrical.
Anyone of the above categories could provide a starting point for a 'universal
classification'. Frequently, however, categories overlap, with cross connections confus-
ing any pattern. A 'functional' classification is possibly the most useful since it corre-
sponds most closely with the layout of manufacturers' catalogues and this is given below.
It does tend to exaggerate the importance of some categories at the expense of others and
gives no idea of construction.
The details given in the following chapters concentrate on the construction and
operation of valves with reference to applications but more detail is given on applications
and circuit performance in Section 4.
The valve as a hole

A useful starting point, at which commonality exists, is the concept of a hole. At the
simplest level all valves can be considered as a hole - an opening in the flow path through
which fluid can pass. This hole is more formally called an orifice and its size determines
a flow area, which in many cases defines the amount of flow. It is normally assumed that
the equation for orifice or turbulent flow can be applied to valve orifices as defined in the
chapter on Basic hydraulic theory.
In many valves the size of the orifice can be changed to give a changing flow area from
a maximum to a nominal zero. In some valve types (overlap sealing) the flow area may
never actually be zero even when 'closed', since a small but finite minimum value is
necessary to give clearances for movement. When this occurs the flow pressure relation-
ship through these clearances may be better described by the equations for laminar flow
where pressure drop is proportional to flow. The relevant equation is given in the chapter
on Basic hydraulic theory.
There are two fluid related variables in these equations (pressure and flow) and the way
in which they are interpreted is associated with the principal use of a valve and hence its
design characteristics. If a valve is supplied with a fixed flow then the size of the orifice
will determine the pressure drop, and hence this can lead to a pressure control function.
If a valve is supplied with fluid from a constant pressure source then, in conjunction with
the downstream pressure to give an overall pressure drop, the size of the orifice will
determine the flow rate through the valve. In this case we would consider the valve to
control flow. Most hydraulic circuits use valves in both these functions.
Valves are not always designed to be specific and in some cases the physical difference
between a flow control and a pressure control use of the same valve may be very small.
Either of the above equations can be used in this way although the turbulent flow, orifice
equation or 'square law', is the most commonly assumed. There are two main methods to
provide the orifice, one representing the seating group (the poppet) and one the overlap
sealing group (the spool or more precisely the linear spool). Between them these cover all
principal valve designs and probably over 95% of all valves made. The orifice area and flow
path formed in a poppet valve is shown in Figure 2B.I.I. The poppet moves under the action
of pressure and spring forces to give an opening area for flow increasing with lift.
The principle of the linear sliding spool valve is illustrated in Figure 2B.I.2. A collar
area for flow is formed between the edge of the spool land and the annular groove in the
valve body. The axial position of the spool gives a proportional flow area.
In both cases the metering effect in flow path encourages turbulent flow with a reduced
flow area and sharp corners or edges, and hence the application of the orifice equation.
Functional classification
Table 2B.I.1 gives the functional classification. This is intended to indicate the principal
effect of the valve in a circuit to control either a flow or a pressure. Note that flow control
valves may often be pressure operated. The next level of sub-division separates on/off
valves from those giving continuous regulation or modulation, the latter may be preset or
adjustable remotely in operation. The method of operation typically available for these
valves is indicated against each type. The categories in the centre of the table, giving
control of both direction and flow, cover an extensive and growing range of products.
There are two important aspects of val ves which are missing from this classification -
valve construction (introduced above) and valve connections.
Connection and mounting methods include:
(i) pipe or line mounting;
(ii) surface, sub-plate or manifold mounting;
(iii) stacking or sandwich mounting;
(iv) cartridge insert mounting.
These are described in more detail in the next chapter.
Terminology
There are some definitions appropriate to valves which are generic. The number of ways
indicates the number of functional flow path connections in a valve thus two-way or four-
way. Frequently this corresponds to the number of main flow connections or external
ports, and these are often denoted alphabetically A, B etc. The supply connection can be
denoted P for pressure, and the return line connection T for tank. The valve design may
limit the pressure level at the T port to a lower value than the P port and so they should not
HYDRAULIC VALVE CLASSIFICATION 137
be considered as interchangeable. There may well be other, subsidiary connections, which

are denoted X (usually an external supply), Y (usually an external drain) etc.
Valves are specified in terms oftheir maximum flow and pressure capability. These are
obviously related to their size and the strength of the valve housing respectively. The
maximum, or rated, flow wiII usually be quoted with an acceptable pressure drop
depending on the type of valve but often about IMPa (10 bar). The valve can of course be
operated at other pressure drops. The flow rate can be estimated using the orifice equation
above, scaling between the pressures in the ratio of their square roots.
There is also a classification of valve sizing which is frequently referred to as nominal
size or NG. The number following this designation acts as a guide to the size of the flow
bore diameter in units ofmm. It can give an approximate indication of the flow capability
of the valve egoNG 4 is capable of20 Llmin and NG 30 is capable of 600 Llmin. However,
it is very approximate, NG 10 valves may have a maximum flow capability anywhere from
50 to 150 Llmin depending on their function and mounting style.
139
VALVE CONSTRUCTION &

CONNECTION
MANY FACTORS influence valve manufacture and construction including the working
fluid and the application area as well as the functional aspects outlined in the previous
chapter. Materials for seals, valve housing and the moving element must be selected to be
compatible with the working fluid. This has most significance when the fluid is water or
has some other special requirements. The application area influences the performance
specification as well as materials, the design and manufacturing tolerances. The valve
manufacturer will produce a product based on a compromise between these factors and
cost.
Valve working elements
As outlined in the previous chapter, valves can be constructed with either sliding surfaces
or seating components to control fluid flow. The requirements for valves include both
regulating or modulating controls (giving continuous change between levels for either
flow or pressure ) and on/off control. Sliding and seating types are used in all cases but there
are some more suitable for certain applications. Seating valves generally give a very low
level of leakage, frequently zero, when closed and are usually less sensitive to solid
contaminants. A summary of valve control elements which are used in power hydraulics
is given in Table 1. The principal designs are discussed in more detail below.
Linear spool valves

A linear spool valve, almost universally called just a spool valve, comprises a cylindrical
rod or spool moving axially in a cylindrical bore or barrel. The spool has sections of its
length at the full diameter, called lands, and sections at reduced diameter making the spool
stem. External connections are made via holes through the barrel and fluid is directed by
the position of the spool lands relative to these holes. This is illustrated by the series of
views given in Figure 2B.2.1, which shows the spool in three positions - directing flow
to A, closed, and directing flow to B. Spools are most frequently held in position by springs
and require an actuating force to move. All methods of actuation described below are used
with linear sliding spools. Three-position spools are usually spring centred and two-
position spools sprung to one extreme position. Sometimes a detent function is available
which holds the spool in the selected position under the action of a spring loaded ball.
When the spool is in the central closed position then a 'seal' is obtained by the lands
blocking the outlet ports. However, the radial clearance between the spool outside
diameter and the bore will give a flow leakage path. The amount of leakage will depend
on the clearance, and if it is predominantly laminar flow, also fluid viscosity and the
overlap length of the land with the body. Elastomeric or other seals are not normally fitted
because they would introduce excessive frictional forces when moving the spool. The
holes in the barrel are most likely to be annular grooves or counterbores, sometimes
described asfull annular porting, and the flow area when the spool is in an open position
will be given by the collar area:
Flow area = 1t X d x x
Where x = spool axial opening
d = spool outside dia. (bore dia.)
The flow rate of a valve is proportional to the port area and, as shown above, is linear
with the movement of the spool. Some port connections through the body to the spool are
of circular section and if the spool travel is small compared with the port diameter, the flow
rate is then approximately proportional to the valve travel raised to the 3/2 power.
However, in systems where linearity is of prime importance and fully annular ports give
too much flow gain then rectangular sectioned ports can be used. The latter are generally
more difficult to manufacture and various methods are used. These include broaching,
spark erosion and fabricating the sleeve from a series of rings, the latter often being brazed
together to form the valve sleeve. Valves having flow rate with spool travel characteristics
other than those described previously can be made by using ports having the necessary
geometry and this is described in more detail in the appropriate chapter below.
Valve Construction
The barrel may be formed directly by machining in the valve body or it may be a separate
sleeve inserted into the body. In the most precise valve designs the sleeve and the spool
may be produced as a matched pair. The materials used for the barrel or valve body are
usually high grade SG cast iron or steel (cast or machined). For light weight aluminium
alloys may be used and for corrosion resistance stainless steels, are also used in some cases.
The spool is usually a hardened machined steel with a turned, ground or honed finish
depending on the tolerance requirements and sometimes coated. Solid ceramic spools are
now available for use with water and hard coated anodised aluminium is used in some
mobile applications. Typically, diametral clearances between the spool and the bore are
usually below 1O!-lm and may approach 1-2 mm in some cases.
Since all spool valves operate on a sliding principle, design normally follows the basic
requirements of all slide valves, i.e.
(i) Pressure-balanced ports are required, so that there is no net pressure force acting
axially on the spool.
(ii) Valve diameter should be a minimum consistent with suitable stiffness.
(iii) The valve body or sleeve must have adequate rigidity.
(iv) Friction forces must be minimised, and are related to material selection for
rubbing/sliding parts.
(v) Radial flow should be symmetrical around the circumference, in order to avoid
radial unbalance forces, which could increase friction.
(vi) Forces arising from changes in fluid momentum, must be minimised.
Parameters (v) and (vi) are largely controlled by the detail design of the spool.
There are many valve installations requiring control of several services and it is
effective to construct a multi-spool valve block or bank. There are two basic forms of valve
body construction, one being made of sections each containing its own spool and service
ports. When a multi-spool valve is required, these sections are bolted together. The second
is known as a monoblock construction in which the valve body is cast with all passages,
spool bores and service ports in one piece.
Sectional construction has the advantage that various standard sections can be stocked
to provide the valve configuration required for a specific application. The user can then
add (or subtract) sections to make up a valve to suit requirements. Equally, faulty sections
can easily be replaced. There are, however, disadvantages with sectional construction. The
faces of individual sections require sealing when they are bolted together. Another
disadvantage is the critical torque range required when tightening up the assembly bolts.
If they are not torqued tight enough, leakage can occur between the sections. If over-
tightened and distorting the body and barrel sections, problems can arise with the spools
sticking. This can also cause trouble when a multi-spool valve bank is mounted on a
machine since it is possible to create sufficient twisting in the housing to cause the spools
to stick.
Sliding plate valves
Sliding plate valves may be preferred to spool valves for particular applications, although
this type tends to be more expensive to produce and is prone to higher leakage. A sliding
CONSTRUCTION AND CONNECTION 143
plate valve may be likened to an unwrapped spool giving only two dimensions for the flow
path as shown in schematic form in Figure 2B.2.2.
Although this design overcomes the difficulties of manufacturing perfectly cylindrical
bores and spools it substitutes the difficulties of making perfectly flat and parallel plates.
It also introduces problems with unbalanced pressure forces and various novel methods
have been proposed to overcome the additional friction that this introduces.
Some rotary plate designs have also been used and these offer easier compensation of
pressure forces and opportunities for more complex patterns of interconnection.
Rotary spool valves

Although far less commonly used than linear spools, rotary spool valves share some of the
same variations for directional control of flow. They are confined in standard form to onl
off control, and manual operation although in principle other methods of actuation could
be used. The construction is based on a central spindle which can be rotated within the
valve body to line up with a pattern of holes connected externally as shown in Figure
2B.2.3.
Rotary valves rely on close contact being maintained between the rotating spool and a
back-up member and are rather mor~ difficult to produce with adequate sealing for high
pressures. The spindle shaft also requires sealing. In some designs good inter-port sealing
gives virtually leak free operation when closed but this can lead to high frictional forces
to operate the valve, especially when under pressure, unless the elements are pressure
balanced. They are, therefore, sometimes considered only for lower-pressure systems.
Some designs are made such that each port is duplicated and these two openings are
diametrically opposite. This eliminates any radial, out of balance, hydraulic forces, but
nevertheless the rotational resistance of the spool can still be high. Figure 2B .2.4 shows
a sectioned view detailing the construction of a rotary valve which has pressure balanced
seal seats reducing this problem. Standard flow ranges tend to be limited but the design
shown is capable of operating up to nearly 70 MPa (700 bar).
Alternative rotary valves include the rotary ball valve (described below) and the tapered
plug valve often preferred for use in the process industry.
Poppet Valves
Poppet valves have the advantage of high response and relative insensitivity to contami-
nation. They are also well suited to high-pressure duties and so may be preferred in such
applications. They have low leakage, can be made to seal for long periods of time and are
relatively cheap to manufacture. Elastomeric or other seals can incorporated into some
designs if required. They are suitable for large valve sizes but the large opening forces can
require the use of a pilot-valve system. Small poppets may be unguided Figure 2B.2.5a,
relying on the supporting spring for location, whereas larger poppets usually have a larger
diameter section guided in the body or a sleeve as in Figure 2B.2.5b.
The poppet lift, its face angle and diameter dictate the flow area available. This can be
illustrated with Figure 2B.2.6, which shows a poppet in the open position. The minimum
or throat area is determined by a perpendicular dropped from the seat to the surface of the
poppet, whose length is the lift multiplied by the sine of the face angle. The flow area
between the seat and the poppet is then given by the surface area of part of a cone defined
along the length of this perpendicular.
The pressure in region 1 will act on the full area beneath the poppet and that in region
2 will act on the annular area outside the seat diameter. Either of these can be used to
provide sufficient force to open the valve against a spring. The cavity above the poppet,
giving in fact the largest area, can be connected with either region 1 or 2 if appropriate.
It is more frequently used as a control input derived from either an internal pilot or an
external signal and used to close the valve or allow it to open.
When the poppet is opened, as in Figure 2B.2.8 the areas change from a diameter
determined by the seat, to a reduced diameter determined by the flow pattern coinciding
with the section where the flow area is a minimum as in Figure 2B.2.6. As a result there
is a change in the hydrostatic force as the poppet opens. If the poppet has been opened by
a pressure at 1 acting on the full area then this area will be reduced as the valve opens. This
reduction of force can be an advantage but can also lead to an oscillatory operation. If the
poppet is opened by pressure acting at 2 on the annular area then this is increased as the
poppet moves up and it will open even more against the spring.
FIGURE 2B.2.8 - Areas on which hydrostatic forces act in open poppet valve.
In the preceding figures the seat has been shown as a sharp right angle whereas in any
practical valve the seat will have some finite width. If the seat and poppet face angle are
very similar then the upstream edge of the seat can be taken as the sharp edge considered
above. If the angles are different then this can affect the flow patterns and give changes
at different poppet lifts.
There are several area ratios used as standard by manufacturers which give greater
flexibility in control for different applications. The shaping of the "nose" of the poppet can
also vary, and for metering of flows a long nose with notches can reduce the rate at which
flow area is increased with poppet movement. There are many design alternatives for
poppet valves particularly when used as logic elements.
Needle valves
The term needle valve is normally used to describe a finely tapered moving element, the
needle, moving axially relative to a concentric circular port. This type is usually manually
adjustable to a preset position and would give an orifice type flow characteristic. Three
basic configurations are shown in Figure 2B.2.9, giving different overall flow paths
through the valve. The threaded spindle would normally incorporate some seal, unless the
valve is for use at very low working pressures.
Ball valves
There are two types of ball valve - rotary ball and seating ball. They offer completely
different characteristics and application areas but both give a good leak free seal when
closed.
Rotary ball valves are very similar to rotary spool valves but with more limited porting
configurations. Although these are sliding element valves their good sealing is obtained
by the use of seals as shown in Figure 2B.2.l O.They may be categorised asfull or reduced
passage depending on the size of the hole through the ball relative to the valve ports. As
with other rotary valves these are usually manually operated and made with different port
connection possibilities.
Seating ball valves are similar to poppet valves but tend to occupy only the smaller size
range. As with many poppets they are normally pressure operated but some electrical
solenoid operated ball valves are available as pilots. They are also most appropriate in onl
off applications and are less frequently used for either flow or pressure regulation. Their
advantage is in the simplicity and availability of the ball element itself.
Contamination
The contamination considered here is the effect of solid particle contaminant in the fluid.
Control of contaminant levels is important for good trouble free valve operation, the level
depending on clearances and hence on the application area. Contamination has three main
effects: one is to reduce the reliability of the valve by clogging the flow path, another is
to jam the movement of the moving element through silting or wedging of large particles
and the third is to reduce the valve's life by increasing the wear rate at the metering edge.
Although an improvement in the valve's ability to operate in relatively dirty oil can often
be obtained by suitable re-design or materials selection, adequate filtration is still essential
to reduce wear. Wear occurs mainly in the fine metering orifices where high fluid
velocities occur. This wear results in increased leakage flows, modified flow character-
istics around the valve neutral position (in zero lapped valves) and can give a general
deterioration in performance.
Operating forces in valves

The moving elements require forces for their actuation and these may be mechanical.
pressure or electrical in their origin. Table 2B.2.1 also gives data on the method of
operation available for the valve types included. Mechanical forces come from manual
operating levers as well as a variety of cam and roller mechanisms. These usually act
directly on the control element most commonly spool valves.
Pressure actuation may be derived automatically internally as part of the valve design
or from external signals. It can provide both the signal for on/off operation and the
resolution necessary for modulation of the valve area in regulating applications. Most
poppet and seating ball valves are pressure controlled, and so are many linear spool valves
either as part of a two stage design or derived from a remote control location.
Electrical operation is most normally via solenoid devices for on/off switching, and
these may be either AC or DC. In general DC devices are slower acting than AC but less
prone to burn-out if stalled, since with an AC device the increased inductance when closed
is necessary to limit the maximum current. DC voltages are usually either 12 V or 24 V
and currents in the region of 1 or 2 Amp. AC devices follow usual industrial practice
operating at 115 V or 240 V AC with a similar power rating. For a modulated control then
the proportional solenoid, force motor and torque motor are all widely used and discussed
in later chapters. Also both DC motor drives and stepper motors have been used in
commercial valve designs to give a modulated control. Motor driven valves usually have
the capability to hold a pre-set position even without electrical power when they are de-
energised. In most valves there is a limit to the maximum economical force obtainable
electrically and so large valves may have an electrical stage operating a pilot valve to
operate in turn a main control valve.
The standard symbols used to indicate each type of operation are given in Figure
2B.2.ll.
Other hydraulic forces in valves

In addition to the hydraulic forces described above to operate a valve, there are two other
mechanisms where fluid can interact to give forces on a moving valve element. One of
these initially produces side forces but through increased friction these can affect axial
movement, and the second has a direct effect by introducing an axial force as a result of
flow.
Hydraulic lock
The first of these is frequently described as hydraulic lock where a spool (or a guided
poppet) can experience a pressure distribution causing a radial force pushing the spool
against one side wall of the valve body. Decentring forces of this type can be produced a
number of ways but the easiest to understand is with a spool tilted at an angle within a bore
as shown in Figure 2B .2.12a. A flow from the left end of the land towards the right causes
pressure distributions which are different on the upper and lower surfaces as shown in
Figure 2B .2.12b. These arise because the principal loss of pressure occurs at the narrowest
section. This results in a net force pushing the spool upwards but as it moves so the
difference in pressure becomes greater and pushes the spool against the valve bore causing
a frictional lock.
There is however also a restoring couple produced which corrects the initial tilt and so
the 'solid' lock may not occur if this is greater than the initial couple causing the tilt. It is
worth noting that a purely parallel piston, aligned perfectly in a parallel bore has a uniform
pressure distribution which does not produce any side forces even if it is off-centre.
There are other causes of hydraulic lock. Tapering of the spool (or bore) with flow
taking place from the small clearance end will produce a de-centring force unless it is
exactly centred. Note that this is a similar effect to the principle of a hydrostatic bearing
operating the wrong way around (see chapter on Hydraulic cylinders). A burr on the spool
land at the upstream end or a large particle of contaminant trapped at the upstream end can
also cause de-centring.
The usual method to reduce these side forces is to encourage a circumferential flow
around the spool from high to low pressure regions. Since the initial pressure distribution
was flow induced this tends to correct the original imbalance. This is easily achieved by
adding grooves on the land to carry this flow. The usual solution is either to minimise the
land length by relieving the diameter along the majority of the length except at the ends
for metering, or to use 3 grooves, one central and one close to each end of the land. This
latter has been shown experimentally to reduce the force required to break-out from the
locked condition to less than 10% of that for an un grooved spool. Similar arguments apply
to reducing locking effects in guided poppets and they also normally have pressure
balancing grooves.
An alternative solution is to consider tapering of the spool to act like a hydrostatic
bearing in producing a centring force. This will only work if the pressure difference across
the land is always maintained in the same sense. Other methods tried for reducing the
locking force include the grinding of a small taper or series of steps on the lands and the
addition of a spool-rotating device to the valve design. In electrically driven valves it is
also possible to prevent lock occurring by the use of a small and relatively high frequency
input signal such that a small amount of spool 'dither' occurs.
Flow forces
In addition to friction, axial forces may also be set up due to the changes in fluid momentum
through a valve. Often described as Bernoulli effects or Bernoulli forces they are more
accurately called momentum or flow forces. Since these are directly additive to the
actuating forces then they can have a noticeable effect on valve operation for both spool
and poppet valves.
The effect occurs whenever fluid is accelerated either to form a jet or to change its
direction. Since the fluid has mass this requires a force Gust like accelerating a solid object)
to change its momentum. In hydraulic components this force must be provided by a
pressure distribution within the fluid and reacted on the solid boundaries containing the
flow. It is this reaction of the pressure on the moving elements of the valve which is called
the flow force. The force required to hold a fire hose is a good example of the effects of
this flow force. A theoretically derived equation is used as the first approximation to give
flow forces:
Experimental evidence has shown this equation to be acceptably accurate for many
applications but it should be applied with care. It also indicates that the flow force is
directly proportional to the pressure drop across the valve (obtained by substitution for Q
and v), and hence the operating condition where the effect is most pronounced.
Figure 2B.2.13 shows a section of a spool between two lands with a metering orifice
formed on the right hand side. In Figure 2B.2.13a, flow leaves this chamber as a jet and
so the fluid has acquired a component of axial momentum with a reaction causing the spool
to move to the left, closing the orifice. The angle of the jet leaving the chamber, given as
e above, is taken as 69. This is again a theoretical angle (proposed by von Mises) based
on a large downstream volume. The pressure distributions reacting on the spool are
indicated by the arrow lengths on the spool lands as shown. The lower pressures on the
right are explained by the higher velocities in this region according to Bernoulli's energy
equation, and hence the alternative name.
Figure 2B.2.13b, shows the fluid entering the chamber through the metering orifice as
ajet. It now has to lose the momentum associated with the jet and this deceleration requires
a force on the fluid to the right and hence reaction on the spool to the left. This is again a
closing force as indicated by the similar pressure arrows on the end faces ofthe spool land.
A variety of designs have been proposed to try to balance these flow forces which
otherwise will tend to close valve spools. The most effective design is the so called
negative force port shown in Figure 2B.2.14. This is frequently built in to just the return
section of a four-way spool since it can produce a net opening force as the name implies.
Similar affects are seen when the flow through a poppet valve orifice is examined.
Figure 2B.2.15 shows flow in both directions through a poppet.
When flow is occurring upwards through the poppet orifice shown then the highest fluid
velocities will occur downstream of the minimum area where the flow tends to stay in the
form of a jet. Upstream of the minimum area the velocities are lower since the flow
converges rapidly. If the component of velocity parallel to the axis of the poppet is lower
in this entry region than in the exit region then the fluid must have accelerated in this
direction. To have accelerated upwards then there must be a force giving the fluid
additional momentum in this direction. This implies a downward reaction force on the
poppet itself, a closing force on the poppet.
When flow is occurring downwards a similar argument will suggest that the fluid again
increases in velocity as it passes the poppet and since this increase in momentum is
downwards then the poppet experiences an upward or opening force. Thus this case is the
only one which produces an opening force on the valve element. Methods of compensation
are also possible and Figure 2B.2.16 shows one alternative for each flow direction. In both
cases the philosophy is to deflect the flow radially and hence eliminate any axial changes
in momentum.
Connection and mounting

The external connections to the valve housing have important practical significance. The
correct choice can make system assembly and maintenance very much easier. The
physical attachment of a valve housing to the supporting structure requires some care to
ensure that it is not twisted or distorted, since bore distortion may cause spools to jam.
Multi-section valve bodies are more vulnerable in this respect. There are four main sub-
divisions for types of valve connection/mounting as introduced in the previous chapter.
Line mounting
In pipe or line mounting pipe fittings are connected directly to tapped holes in a valve body.
In some cases these may be tapered fittings but a threaded connector with a faced seal ring
or bonded washer are widely used. This type of connection reduces initial cost and in some
cases of small two way valves, the pipes themselves may substitute for a valve mounting.
Care must be taken in assembly, particularly if more than two way, to ensure that neither
fittings nor the valve body is unduly strained. This could cause leakage or malfunction of
the valve. Maintenance and servicing may be made more complicated if pipe runs have
to be disassembled for access.
Sub-plate mounting
Face or surface mounting a valve body allows hydraulic connections and mounting to be
incorporated together. The valve can be connected in this way to either a sub-plate or a
manifold block. The mating surfaces need to be flat and the a-ring seals normally used can
be incorporated in suitable faced grooves in the valve body. This method gives very good
serviceability since with sub plate's pipework need not be disturbed but they may add cost.
Manifolding will reduce the amount of pipework required but machining manifolds on a
small scale can also be costly. There is a standard layout defining the different configu-
rations and sizes which, although now covered by ISO 4401 or 6264, are still frequently
referred to as a CETOP mounting or size. Some typical shapes are shown in Figure
2B.2.17.
Sandwich mounting
A variant on face sealing which tends to be categorised separately is sandwich mounting
or valve stacking. Here the fluid is taken through intermediate valve blocks to make
connections with valves higher in the stack as shown in Figure 2B.2.18. The interface and
connection dimensions are usually based on the same standards as those used for face
mounting valves although there is also a five port connection pattern. A limited range of
adaptor plates are available between some adjacent sizes. This system allows the
advantages of building up more complex valve functions without special manifold
machining and again reduces pipe work inter-connections. Serviceability is also good and
system performance is not compromised provided that the stack is close to the service
supplied. Most valves are designed for vertical stacking but some horizontal stacking
systems are also available.
Cartridge valves
Cartridge valves are cylindrical inserts which contain the main control components of the
valve but lacking any external body. They are inserted into a housing which contains
suitable internal cross drillings to make the necessary fluid connections internally.
External connections to the housing itself can be threaded, giving an in-line assembly but
with good serviceability by replacing the cartridge. Other alternatives include sub-plate
mounted orthe useof amulti-valve manifold block. The latter gives the greatest advantage
in compactness and reduced component count but the number of assemblies required must
justify the higher cost in manifold design. This cost is reducing with more computer aided
design allowing the three dimensional galleries to be visualised and then the machining
instructions produced directly. The housings may be steel but are frequently machined
from aluminium alloy for lower pressures.
The cartridge may be either a slip-in insert with a retaining cap or a threaded insert.
Seals, usually a-rings are used to isolate the different sections and the more complicated
valves with more connections are progressively stepped in diameter. At the simplest a
cartridge may be a poppet with its sleeve as shown in Figure 2B.2.19, giving only a two
way function and these are available in a very wide range of sizes giving high flow
capability (over 10 ()()()Umin). A number of standard area ratios are available defining
the working areas as discussed earlier in this chapter. These enable different switching
functions to be implemented with different pilot sources and this family of valves is
sometimes referred to as logic elements.
In the smaller screw-in sizes clever design allows most of the valve functions in Table
2B.!.1 to be realised. These more complex but very compact multi-valve functions are
usually limited to about 100 Umin. Cartridge elements are available most commonly in
steel but also high grade aluminium alloy is used. Pressure ratings for cartridges up to 30
or 40 MPa (300 to 400 bar) are widely available but the housing material may limit the
working pressure below these values.
A typical manifold assembly is shown in Figure 2B.2.20 combining cartridge and face
mounted valves.
156
]57
Directional & Flow Control Valves
THIS CHAPTER describes valves designed to give control of fluid flow. These range
from those with simply an on/off action, through valves capable of directing flow via a
number of alternati ve paths, to val ves which may also give modulation or regulation of the
flow magnitude. There are two further chapters (Proportional valves and Servovalves)
which describe other families of flow and directional control valves.
Directional control valves

Directional control valves or selectors provide the means of changing the flow connec-
tions in a circuit. They are usually spool valves although poppet and rotary valves are also
capable of providing this function. Directional control valves (DCV) are normally
classified by the number of positions and the number of ports or 'ways' in the valve body.
These are designated by letters indicating the appropriate connection - P for pressure, T
for tank, and A & B for connections to the output service. Linear spool valve designs do
not permit the P and T ports to be interchanged. Standard configurations and symbols are
shown in Figure 2B.3.1. Valve symbols comprise a number of squares, side by side, each
square representing a possible position for that valve. Thus a two-position valve is
represented by two squares, and a three-position valve by three squares. Internal connec-
tions are indicated in each square for each position. The valve symbol also (usually)
includes an indication of the method of actuation of the valve and how it is returned to its
normal position. Essentially, therefore, selectors can be considered as two-way, three-way
or four-way.
Two-way valves
A two-way or two-port valve is basically an on-off switching element and can be simply
described as a shut-off cock. A variety of designs can be used to provide this function -
ball, rotary, slide, poppet and spool valves - with the features described in the chapter on
Construction and connection. Methods of switching the valve also include all of the
possibilities described in that chapter. The simplest type of valve is frequently operated
with spring return. For a 'normally open' valve, the spring holds the valve in the open
position; and for a 'normally closed' valve the spring holds the valve in the closed position.
DIRECTIONAL AND FLOW CONTROL YALYES 159
usually spool type or piloted poppets swi tched to give the correct opening pattern as shown
in Figure 2B.3.3. At the time of the last edition of this book, it appeared that poppet valves
may supplant spool valves in many DCV applications. They are widely available as
alternatives but tend to a higher price than the stand alone spool valve. Poppets provide
the advantages of positive seating and a higher maximum flow capability together with
greater switching flexibility as described below for the 'centre' position.
Five-way valves
A five-way selector is, basically, a modified four-way directional-control valve in which
the two sets of return- to-tank ports in the valve bore are usually taken to separate external
ports on the body. This provides independent speed control of exhaust flow, without the
need for separate non-return valves, or can provide for dual pressure operation of an
actuator. They are also used for interlocking circuits or circuit unloading. They are a more
common variant in pneumatics.
Six-way valves
A six-way selector is, basically, a four-way directional-control valve providing pump
unloading in the centre position. It is a type which has been developed specifically for
mobile equipment applications, and is described below under the heading of Central by-
pass valve.
DCV centre position alternatives

A brieflook at a manufacturer's catalogue will indicate the range of alternatives for four-
way valves as shown in Figure 2B.3.4. The main types are described here in relation to a
four-way valve but similar alternatives exist for three-way valves.
FIGURE 2B.3.4 - An example of manufacturer's catalogue altemati Yes.
i). An open-centre valve connects all ports together, the supply, the reservoir and the
service ports in the centre position, thus reducing power consumption when the
system is idling and allowing free movement of the output actuator.
ii). A closed-centre valve blocks off both the supply and service in the centre position.
thus holding the system pressurised both on the supply and service sides.
iii). A tandem centre blocks the service ports but connects the supply to the reservoir
port and thus off-loads the pump but holds the load pressure.
iv). Afloat centre connects the service ports to the reservoir, allowing actuators (even
single ended cylinders) to move freely.
DIRECTIONAL AND FLOW CONTROL VALVES 161
An intermediate position is sometimes used to define the connections during the

transition condition as the spool moves between one of the extreme positions and the
centre position. This position is only connected during the change-over and is referred to
as a transitory position. It ensures that one side of a system is connected first in order to
begin a flow, and probably a pressure change in that section before the main flow paths
are connected. This can prevent sudden pressure changes in systems giving smoother
operation. It is shown as an extra box in the valve symbol diagram but with a dashed line
between adjacent boxes or positions.
All these remarks are appropriate for both spool and poppet valves but pilot operated
poppets offer even greater flexibility in centre switching. If each poppet can be switched
independently and the switching logic is available then different centre positions could be
selected in different parts of a cycle. They can also have different transitory connections
for opening and closing and independently control the rate at which the poppets open and
close.
Spool valves are available with end stops to allow a maximum opening (flow) to be set.
It is also possible to limit the maximum rate of spool movement which gives a more gentle
flow metering action in the transient positions. This helps reduce shock in systems which
can be caused when a high flow is directed to a system. Many spool valves also have an
extension rod which can be used to manually operate the spool against a return spring.
Flow through any valve will be accompanied by an associated pressure drop and it is
this which effectively determines the appropriate size when selecting a valve. A typical
setof catalogue characteristics are shown in Figure 2B.3.5, for one nominal valve size. The
diversity arises from the different spool configurations and for the different flow paths,
some may have a restriction deliberately included.
FIGURE 2B.3.5 - Typical flow characteristics for a NG 6 directional control valve.
Other on/off valve designs

Check valves
Non-return valves or check valves are used in circuits, or combined in the body of other
valves, to provide flow in one direction only. The simplest type is the spring-loaded ball
valve which is widely used in the less demanding applications. Poppet valves form the
basis of the more sophisticated designs. In high pressure services good sealing is essential,
and it may be necessary to design the valve with a resilient seating seal. Damping may also
be provided to prevent the seal being damaged by the impact of sudden flow reversal. A
further consideration for high flow rates is that the sealing member must offer the
minimum flow restriction in the valve-open position. Also, of course, in high pressure
valves with resilient seals, the seal inserts must be designed to prevent extrusion or
displacement of the seal.
FIGURE 2B.3.6 - A simple ball check valve, cartridge mounted.
A typical design of a simple non-return valve is shown in Figure 2B.3.6. Spring rates
can be selected to give a range of pressures required to open the valve, typically from 10
kPa to 0.5 MPa (0.1 to 5 bar). A stronger spring may deliberately be used to provide higher
back-pressures for specific applications, ego to maintain a minimum back pressure in pan
of a system. If a valve is going to be mounted vertically then the moving element can be
seated by gravity forces which will give the minimum pressure to open the valve.
Check valves are used in a circuit to eliminate actuator movement and to maintain it in
a 'hold' position without creeping, which might occur due to spool leakage. The most
usual type for this particular duty is the pilot-operated check valve shown in Figure 2B.3.7.
Most pilot-operated check valves are supplied with a pilot ratio of at least 2: 1with the main
piston seat area. This means that to open the valve a pilot pressure is required equal to half
the pressure being applied to the primary side of the val ve. If the pilot -operated check valve
is being used on the annular side of the cylinder, then, with a 2:1 ratio pilot, the cylinder
area ratios must be less than 2: I,otherwise it will be impossible to open the valve. When
DIRECTIONAL AND FLOW CONTROL YAL YES 163
there is no pressure on the primary side ofthe cylinder, the pilot-operated check closes and
allows no flow from the annulus side of the cy linder, therefore retaining the load in a fixed
position. When pressure is introduced to lower the load, the check val ve is opened by pilot
pressure to allow free passage of the oil from the bore side of the cylinder. In the reverse
condition, to lift the load, the pilot-operated check operates as a normal check valve. This
type of pilot is known as a pilot to open check valve. It is also possible to get pilot to close
check valves.
Shuttle valves
A shuttle valve is a pressure operated valve which will connect one common port to one
of two alternative service ports. The symbol is shown in Figure 2B.3.8, which gives a good
idea of its operation. Whichever of the service pressures is higher will operate the valve
to make a connection to the common line. Shuttle valves are made with a single moving
element as standard but they can also be formed by connecting two check valves together.
Transfer valves
Transfer valves are used to short circuit an actuator, to allow for rapid mechanical
movement without the 'braking' effect which might otherwise be caused by the fluid. This
could apply, for example, when initial movement of an actuator load was under gravity or
mechanical force, with the final movement accomplished by hydraulic pressure. To
increase the speed of movement above that obtained by hydraulic flow alone, the suction
caused by over-speeding opens a valve to short-circuit fluid from the other side of the
actuator, making for sufficient volume to prevent cavitation.
A prefill valve performs a similar function in allowing a large flow to occur with a very
low pressure drop. They are mainly used in press circuits to allow the main actuator
cylinder to be filled directly from the system reservoir whilst being driven down with
auxiliary actuators.
A 'dump' valve also performs a similar function, but in this case is used with a single-
acting actuator. To speed up the return stroke, the throttling effect of the fluid being forced
out through the return line is relieved by a separate valve opening to 'dump' fluid at a
higher rate into an appropriate part of the return circuit. In this case the valve is pressure-
operated rather than suction-operated, and can be similar in form to a pressure-relief valve
with suitable locking characteristics, or even be a simple cock with mechanical actuation
derived from the return movement.
Sequence valves are described under the chapter on Pressure control valves because of
their similarities in construction.
Hydraulic fuses
Hydraulic fuses are intended to shut off automatically any line in which failure has
occurred to prevent a load falling or excess fluid loss from a system. The simplest type
works as a flow-sensitive valve, closing and sealing the line should the flow rate exceed
a pre-determined amount, as it would in the case of a burst or massive leakage. Such fuses
are generally characterised by low sensitivity, so that they will not respond to smaller
leaks; and they are also set to a value above the maximum flow rate for the system, which
is not necessarily representative of average working conditions (further reducing their
sensitivity to leaks).
The usual form of such a fuse is a hollow piston, with an orifice, spring-loaded in a close-
fitting cylinder. The piston is extended in the form of a stem, with the end shaped to provide
a seal in the outlet end of the cylinder at full piston travel. At fluid pressures up to the
maximum setting, the piston remains in the 'open' position under spring pressure. Higher
pressures result in a differential pressure across the piston orifice greater than the spring
pressure, causing the piston to move until the stem engages in the valve seat and shuts off
the flow.
Alternative types offuse are also used for quantity measuring and fluid sampling. With
a quantity measuring device, the body capacity is arranged to be slightly in excess of the
demand of the system protected, and the fuse automatically empties on the return stroke.
Any greater volume entering the fuse operates a valve inside it to shut off the flow, this
additional volume being detected as a leakage. The fluid-sampling fuse works as a two-
stage unit. First, pressure is applied to the protected circuit through the fuse with the circuit
inoperative. Any flow resulting must then be due to leakage, in which case the fuse is
'triggered' to shut off the supply. Only if there is no leakage does the fuse remain 'open'
to accept normal full flow through the second stage to the circuit.
A particular advantage of hydraulic fuses, apart from shutting off a damaged circuit and
minimising fluid loss, is that they enable other circuits on the same supply to continue to
be operated with the faulty circuit isolated by its fuse.
Flow control valves

Simple flow-control valves work on the basis of restricting the flow with either a fixed or
variable orifice. In the latter case flow control valves may be either simple throttle valves
or include pressure compensation. Table 1 provides a summary of various types of flow
control valves and their applications
An inherent disadvantage of simple restrictors is that the flow through them is
dependent on pressure drop across the valve and thus will vary with changes in load. As
a consequence their application is limited to those systems where the load is constant, or
where variations in flow rate, and thus operating speeds, are permissible. Fixed restrictors
may be made from a drilling in a section of the valve body or a replaceable plug. Calibrated
DIRECTIONAL AND FLOW CONTROL YALYES 167
constant flow characteristics, a combination of two restrictors can be used, one fixed and
the other automatically variable. These two elements are normally combined in a single
unit to produce a pressure-compensated flow control valve. There are both two-way and
three-way methods to combine these elements. A schematic of a two-way, series
compensated or restrictor type is shown in Figure 2B.3. I 1.
An orifice size to give the nominal flow is set by the needle valve and the pressure drop
across this maintained constant by the compensating spool. If the flow through the needle
valve is too high then the excessive pressure drop produced will act on the spool against
the return spring to close the metering edge of the spool. This will reduce the flow through
the complete valve.
The main limitation of a pressure-compensated flow restrictor is that whilst constant
flow characteristics are provided, independent of load, the controlled flow is throttled and
surplus flow must be directed through another valve, resulting in the pump working at full
relief setting continuously. To overcome this loss of efficiency, the alternative pressure-
compensated by-pass, spill-offorthree-way regulator valve can be used as shown in Figure
2B.3.12. Here the surplus flow is by-passed through the valve at working pressure, and at
the same time the controlled flow has low throttling losses. The use of a three-way
pressure-compensated by-pass regulator, therefore, enables much higher circuit efficiencies
to be achieved. Its main limitation is that, unlike the pressure-compensated restrictor, it
cannot be used in parallel configurations. Although it must be used with a pressure relief
for protection, the system pressure will not normally reach this limit.
The simplest designs of these valves come with pre-set fixed orifice settings. These are
useful to control the return flow from a gravity loaded cylinder to limit its descent speed.
One-way working can be provided by incorporating a non-return valve in the unit to
provide free flow in the other direction. Although manufacturers design the orifice section
to encourage the flow to obey the orifice equation, compromises necessary may mean that
this is not always so. In such cases the pressure drop across the orifice is likely to be
viscosity sensitive. This can be overcome by including temperature compensation to
adjust for changes in fluid viscosity. These methods of pressure compensation may also
be incorporated with directional flow control capability in pressure compensated propor-

tional valves, and these are described in the chapter on Proportional valves.
Deceleration valves
When working a cylinder at high speeds, it may be necessary to provide some means of
slowing it down before the end of the stroke, so as to prevent shock, in place of or in
addition to cushioning. If the mass is not too great, this can be done effectively by a simple
deceleration valve such as that shown in Figure 2B.3.13. The tapered spool plunger is
depressed by a cam on the moving part (piston rod) so that the amount of throttling can
be progressively increased as the valve moves down. Some designs incorporate an
adjustable sleeve to set the initial deceleration rate. This design also incorporates a check
valve to give free flow in the other direction. The absence of the check valve would limit
the acceleration which may be beneficial in some systems.
If an attempt is made to decelerate too great a load with this type of valve, the pressure
may build up dangerously and it is then necessary to employ a braking valve. The simplest
form of braking valve is merely a relief valve which comes into operation at the appropriate
point of the stroke. The energy is then dissipated by blowing through the relief valve.
Flow dividers
A flow divider valve is used where an input flow needs to be split into two outputs, in a
pre-determined ratio. A simple form giving an equal division is shown diagrammatically
in Figure 2B.3 .14, where the input is fed to two orifices of equal size feeding two output
ports. This has the same characteristics and limitations as a fixed restrictor in each outlet
line, with equal division of the inlet flow. This could be used as a differential lock of two
motors in a transmission or for the operation of two actuators at the same speed.
A more satisfactory form is the pressure-compensated flow divider where two spools
are used, mechanically connected, with the total flow passing across metering orifices in
each of the spools. An example of a cartridge type divider is shown in Figure 2B.3.15. If
DIRECTIONAL AND FLOW CONTROL YAL YES 169
the flow to one inlet increases, pressure drop across the spool is increased, causing it to
move against its spring and so provide more throttling effect. If the flow to one service
reduces to zero then the mechanical connection transmits movement to the other spool to
shut off the second service. The result being that the two outputs are substantially self-
adjusting to the output loads. Many such valves can be used during reverse flow, so that
the flow divider can also be used for combining two input flows, thus controlIing speed
on the return stroke of the actuators. Typical flow proportions from 50:50 to 90: 10 can
be obtained with both pre-set and variable types. Maximum flows available as standard
are in the region of 300 L/min.
FIGURE 2B.3.15 - Two spool cartridge mounted, flow divider valve [Sterling Hydraulics]
Flow dividers have certain inherent limitations, notably a tendency to become slightly
out of phase, which effect can be cumulative unless they are re-phased at the end of each
actuator stroke - i.e. loss of synchronisation could grow in a series of part-stroke
movements. Control of flow balance is best at the nominal setting of the valve where
differences of within +/- 3% may be obtained. At low flows there can be considerably
greater divergence. Synchronisation can also be affected by the number and manner in
which further flow dividers may be connected in the circuit, ego to produce synchronisa-
tion of more than two actuators. Gear type flow dividers are also available.
Priority valves
A priority valve is a variant on the flow divider in construction and function. It may be used
in asingle pump circuit where it is essential that one service in a multi-service system, must
always have priority on flow irrespective of other demands. A valve with a fixed priority
flow setting is shown in Fig 2B.3.l6. The hollow spool acts as a flow regulator by keeping
the pressure drop across the orifice constant, the surplus flow capacity being by-passed to
a subsidiary output. If the subsidiary pressure becomes higher than the priority service
pressure, the spool is forced down to the lower seat and a metering annulus is then formed
by the lower end of the plunger.
FIGURE 2B.3.16 - Priority type flow divider cartridge mounted: fixed setting [Danfoss]
The valve will only function when there is a constant tlow through the main outlet,
which pre-supposes an open centre directional-control valve being used. If the priority
flow is completely stopped, there is no pressure difference to force the spool down and the
valve will go to a completely closed position.
Central by-pass valves

Central by-pass valves are multi-spool valve banks giving control of both the magnitude
and direction of the flow. These valves are most widely used in mobile applications where
several services may be operated simultaneously. They are an open centre type design
which allows flow to pass through the valve with minimum flow restriction when not
required by a service saving unnecessary waste of power in the case of a continuously
running pump. These valves may be of a mono-block construction within a single casting,
or in a modular form with each spool in its own machined section. Frequently, they contain
other features, such as a system relief valve, within the same block.
Flow is supplied to the valve from a fixed capacity positive displacement pump as
shown in Figure 2B.3.17. The figure represents just one section of a valve spool and
indicates that the pump flow can be directed either to tank as QB, or to the service as Qs.
When the spool is in the central position the orifice a, to the service, will be closed, and all
the flow will be directed through the wide open orifice aB, to tank. As the valve is first
moved in the direction X, orifice aBwill reduce in size and since it is still taking the full
pump flow it will cause an increase in the pressure Ps. Further movement of the spool will
also begin to open orifice as and allow flow Qs, to pass to the output actuator. The
magnitude of this flow will depend on the pressures Ps and PL, and on the size of the orifice
as. The spool movement progressively increases Ps via orifice aB,and opens orifice as It
is not necessarily a linear relationship between the orifice areas and spool position but
obtaining a good modulation of flow to the service is the objective.
FIGURE 2B.3.17 - The principle of the central by-pass spool valve.
A complete mono-block, twin spool valve is shown sectioned in Figure 2B.3.18. This
shows clearly the by-pass gallery running through the centre of the valve bank supplying
both spools. There are also check valves built inside each spool to prevent back-flow when
the valve is opened. This would otherwise occur when a pressure is required to hold a load
and before the spool has moved sufficiently to build the supply pressure. There is also a
relief valve built in closest to the supply port. Frequently other functions are included for
each service including counter balance and cross line relief valves meeting all the
installation requirements from a single valve assembly. There are also many alternative
spool port connections and some spools with a fourth position which is frequently afloat
position which allows an actuator to move freely.
In multi-spool valves, there are alternatives in the way in which the supply flow is made
available to each service spool appropriated for different applications. Three common
valve configurations are: tandem, parallel and series circuits.
Parallel circuit
This type of valve is designed to make full system pressure simultaneously available to all
spools. An example of a circuit symbol is shown in Figure 2B.3.19. The valve is able to
divide the available flow between any number of services and provide full pressure at one
or any number of them, but at a proportionally reduced flow rate determined by the number
of services being supplied. If more than two spools are fully selected simultaneously the
service demanding the lowest pressure will receive the available flow rate. The flow will
always take the line of least resistance. It is, therefore, necessary to throttle the flow
passing to the lowest load to generate sufficient pressure to operate the higher loads. By
careful spool selection it is possible to operate a number of services simultaneously, all of
which demand a different supply pressure.
This configuration of valve provides the most effective and efficient work cycle for
many types of machine and is widely used on mobile diggers where four services, slew,
boom, dipper arm and bucket are operated simultaneously at different speeds.
Tandem circuit
The tandem circuit valve (sometimes referred to as series parallel) is intended to direct all
the available inlet flow to one service at a time. Consequently the nearest spool to the
supply takes priority over all down-stream spools and the most important service is
connected to the number one spool. An example circuit symbol for the valve is shown in
Figure 2B.3.20.
FIGURE 2B.3.20 - Tandem circuit connection for the central by-pass valve.
If the first spool is only partly selected to limit the flow passing to the selected service,
the remaining flow will pass into the centre gallery. This can then be directed by any down-
stream spool to its service but at a reduced pressure determined by the degree of restriction
created by the first spool. This configuration of valve is generally used on applications
where it is required to operate one service at a time, such as a front end loader which has
two distinctly separate services to operate, i.e. raise or lower the bucket arms and rotate
the bucket.
Series Circuit
The series circuit valve is designed to provide simultaneous and relatively fast operation
of a number of comparatively lightly loaded services or the fast operation of one high load
service. An example circuit symbol for this type of valve is shown in Figure 2B.3.21. In
the tandem and parallel circuits, the return flow from each service port goes directly to
tank. However, in the series circuit the return flow from the first valve becomes the supply
FIGURE 2B.3.21- Series circuit connection for the central by-pass valve.
flow to the second valve. Thus high speeds are possible for simultaneously selected
services. Although, the down-stream flow rate is dependent upon the type and dimensions
of the upstream actuator. If, for example, the actuator is an hydraulic motor the flow
passing to the down-stream will be nominally the same as that supplied to the motor. If the
actuator is single ended, double acting then the down-stream service will receive either the
rod side (annulus area) or the cap side (full bore area) discharge, depending upon the
direction of operation.
It follows that the outlet pressure from an upstream actuator must at least equal the
required operating pressure for the down-stream service, ignoring flow resistance pres-
sure losses. Hence the available useful work for each service is a function of the pressure
difference across the actuator. The operating pressure for the last service is the system
supply pressure minus the back pressure on all the preceding upstream services i.e. the
supply pressure has been divided between all services but the flow is available to them all.
To use a series type valve to best advantage it is essential to study carefully the duties
to be performed, the sequenced and combined operations, the flow and pressure range
required for each actuator and priori ty demands before deciding on the order in which each
service is connected to the valve. An example application for a series valve is in earth-
moving grading machines.
175
PRESSURE CONTROL VALVES
PRESSURE CONTROL is usually associated with limiting the maximum pressure in a

part of a hydraulic system. This may be done by allowing fluid out from a section of the
circuit through a restriction as in a relief valve or by limiting the flow into a section from
a higher pressure source as in a reducing valve. These two categories cover the major
design variants in pressure control and they are both modulating type controls. Modifica-
tions to these basic designs appropriate to other applications, some giving on/off control,
are also described in this chapter. Electrically modulated pressure control val ves also share
many features with the basic designs and are described in the chapter on Proportional
valves.
The maximum required pressure for a circuit may be determined in different ways
depending on the application. It can be used to limit the maximum force or torque from
an output actuator or motor. It may also be used to limit the maximum pump torque in order
to stay within a power limit for the pump driver. Obviously it can also be set in a safety
sense to protect the system components for operation within their designed strength.
Pressure relief valves

The most common type of pressure-control val ve is the pressure relief valve, fitted to limit
the maximum fluid pressure which can build up in part of a system. The simplest design
is shown in Figure 2B.4.1, which is a spring loaded poppet.
This valve opens when the pressure force at Ps overcomes the spring pre-load and moves
the poppet from the seat allowing flow to be throttled as it passes to the return line, usually
to tank. This pressure is known as the cracking pressure and this pressure is maintained
in the system by the throttling provided that the flow is made up on the inlet side. If there
is an increase in Ps then this will produce a higher force and push the poppet further open
until the spring has compressed to balance the forces again. The wider opening allows
more flow to pass. This characteristic is referred to as pressure override and is illustrated
with a typical pressure/flow graph in Figure 2B.4.2. Accurate testing of a valve usually
reveals some hysteresis and the reseating pressure to be lower than the cracking pressure
as shown. This is not normally evident from a pressure gauge.
This design is called a single stage or direct acting relief valve. The low mass of the
moving parts and short stroke movements give a rapid response. The spring in these valves
has to be relatively stiff if higher pressures are to be set and in basic form it can exhibit a
quite steep override slope with an increase up to 150% of the cracking pressure at the
maximum flow. This slope is also influenced by the reduced area and flow force effects
described in the chapter on Construction and connection. The compensation method's may
be used and this leads to substantially flatter override characteristics in many commercial
products. These can vary from a slope giving 2MPa to 5MPa change in a nominal20MPa
cracking pressure. Many commercial designs also include a damping piston attached
directly to the poppet to prevent a fluttering instability which can be excited by pump flow
pulsations. This piston also acts as a guide for the poppet.
Pressure override can give a region for operation along the override slope where flow
is split between the supplied service and the relief valve. If the service load increases
slightly then it will slow as the higher pressure opens the valve. This may be an attractive
characteristic for some applications but it should be remembered that all the pressure
energy in the flow passing through the relief valve is being dissipated as heat. Large
override slope can also lead to the necessity to pass the full pump flow through the valve
in order to set the 'correct' maximum pressure.
Maximum flows may lie between 5 Umin up to 300 Umin, and standard pressures from
PRESSURE CONTROL VALVES 177
3MPa to 40MPa (30 to 400 bar). Usually valves are adjustable within a limited range
through a screw setting of the spring pre-load ego 7 to 14 MPa but some are preset. The
lower limit may be flow dependent and not control high flows at low pressures. It is also
possible to use either a ball element or an unguided poppet but usually for lower flow rates.
One alternative to overcome the high spring stiffness is the differential relief valve
which has high pressure applied from above the poppet as shown in Figure 2B.4.3. This
then acts on the smaller annular area between the seat diameter and the full diameter of the
guide plunger and requires a lower stiffness of spring and offers lower override as a result.
The flow area is still established by the full seat diameter, however this flow direction
requires the use of a momentum spoiler as described in the chapter on Construction and
connection.
Pilot operated reliefvalve

As an alternative to these single stage designs it is possible to use a direct acting poppet
as a pilot stage to operate a second poppet passing the main system flow. Two stage designs
of this type are normally of integrated design with internal pilot connections and a
simplified section is shown in Figure 2B.4.4.
The upstream pressure is still acting on the poppet area as in the simple direct acting
valve but it is also connected to the upper surface of the main poppet acting on the largest
area. The annular area outside the seat diameter is connected to the tank side of the valve
and is at low pressure. The balance of these pressure forces holds the poppet firmly shut
even in the absence of the relatively light spring. The pilot stage operates as described
above and opens to pass flow to tank when the cracking pressure is reached. This flow must
pass through the fixed orifice in the connection to the upstream supply and this causes a
small pressure drop across the orifice. This pressure drop is not enough to allow the pilot
stage to close but is sufficient to tip the force balance on the main poppet because of its
large area. This allows the main poppet to open to an intermediate position against the
spring and relieve the main system pressure by throttling flow back to tank. Modulation
of the opening then occurs for any small increase orreduction in pressure to adjust the flow
to keep a constant pressure.
A number of useful options become available with pilot operated valves. It is relatively
simple to make connections, even between the stages, as shown in the cross section of
Figure 2BA.5.
FIGURE 2B.4.5 - Cross section of pilot operated valve [Rexroth]
An external drain for the pilot stage to a 'Y' port for connection directly to tank
independently of the main return, ensures that flow induced pressure drops which could
occur in the main return do not influence the valve setting. There is no need for the pilot
relief valve to be integral with the main valve and it is often convenient to connect to a
remote pilot valve. This connection may be in addition to the internal pilot and a selector
valve can be used to transfer control between the pilots. This is indicated in the circuit of
Figure 2BA.6 but will only work if the internal pilot is the higher pressure setting. This
can be extended to include additional selectors and pilots to enable several pressures to be
used during a complex operating cycle. Furthermore it is possible to connect this pilot line
directly to return through a selector valve (number 3 in the figure) and this will cause the
main valve to open at a low pressure, by-passing the pump output back to tank. This
method is often used for unloading a pump, making a second main valve unnecessary.
Two stage designs are probably the most widely used because of their flatter override
slope and greater flexibility. The override is normally in the region I to 3MPa (IO to 30
bar) on a nominal 20 MPa (200 bar) cracking pressure. A small positive slope is preferable
and deliberately retained since a negative slope would be unstable. The majority of main
valve stages are based on poppet valves benefiting from their zero leak characteristic.
However, there are designs which use a spool main stage. In general these give a lower
response rate and will give some leakage in the closed condition. The pilot stage may be
a ball or poppet as described above and the construction is often modular so that this pilot
can be used with alternative main stages including slip-in cartridges. Maximum pressures
up to 35MPa (350 bar) are commonly available with a wide range of maximum flow
capability. Relief valves with all the connection and mounting alternatives discussed in the
earlier chapter are available as standard.
Pressure reducing valves

Reducing valves meter flow from a high pressure supply to control a lower pressure in part
of a system. So called three-way reducing valves will also meter flow out of the lower
pressure circuit, acting like a relief valve. Reduced pressure is useful in controlling a
maximum torque or force from a particular actuator (maybe because of a buckling limit
in a cylinder) or to protect a group of lower pressure components. In deriving a supply
through a reducing valve it should not be forgotten that it is a throttling action which is
wasteful of energy. Hence it may be better to use a two pump system with separate
operating pressures where there is a large pressure difference or a sustained high flow is
required at the lower pressure.
Reducing valves operate with a pressure acting against a spring but are normally open.
moving to a closed position when the pressure in the controlled part of the system reaches
the required level. As with relief valves there are both direct (Figure 2BA.7) and pilot
operated alternatives. Reducing valves are normally sliding spool valves to give more
accurate metering. Although, as a result there will be a small leakage from high to reduced
pressure, this is not usually a problem. Reducing valves always have a tank drain
connection gi ving a pressure to act as a datum or reference level for the reduced pressure.
Direct acting valves exhibit similar characteristics to direct acting relief valves as the
spring is compressed giving a change in force with valve opening. In this case the pressure
reduces with increasing flow to the reduced service and this is often called underride. A
typical characteristic is shown in Figure 2BA.8. For this reason direct acting valves are
usually limited to flows under 100 L/min. Three-way designs are available which prevent
pressure in the reduced service from exceeding the set level by continued movement ofthe
control spool beyond the closed position, compressing the spring further and opening up
a flow path to return. It is also possible to incorporate a non return function to allow free
return flow if the reducing valve is used downstream of a directional control valve in
supplying an actuator.
Better pressure control is available with a pilot operated valve which has a lower
underride and higher flow capability. There are two methods in use for deriving the pilot
pressure. A simple extension from the direct acting valve and the two-stage relief design
is the simpler of the two. It takes a connection from the reduced port via a fixed orifice to
a pilot stage which is a spring loaded poppet. A typical schematic is shown in Figure
2BA.9. The force balance from the pressures acting at the ends of the spool holds the valve
towards the closed position when the pressure is at the required level and the pilot is open.
PRESSURE CONTROL VALVES 18)
The main spool must always maintain a small opening in order to provide the pilot flow
even though the load may not be taking flow. Versions of this design can have very low
underride giving pressure changes of under 0.5 MPa (5 bar) through the full flow range.
It is also possible to modify some designs to give three-way function and relieve excess
pressure from the reduced circuit. However, there is a definite maximum flow capability
for this design when flow forces become significant in comparison with the main stage
spring.
on the lower surface. The spool thus moves down to open the main metering orifice and
up to close it, in order to keep the reduced pressure at the level set on the pilot stage. A small
check valve is usually built into the design (in the spool in this case) to allow a low flow
to be returned from the reduced port if necessary. This design is not affected by flow forces
and can give good reduced pressure control at higher flows.
Pilot operated reducing valves can give the closest pressure control from a mechanical
valve and to achieve greater accuracy would require an electrical feedback system. Some
designs are however, affected by the supply pressure and this will change the reduced
pressure. Reducing valves are also available in all of the mounting and connection variants
described in an earlier chapter. Pressures available cover the full range up to 30MPa
(300bar) and maximum flows of around 500 Umin.
By-pass valves
A by-pass valve function gives control of the pressure difference across a part of a circuit
by bleeding off unused flow, generally back to tank. This is shown in the circuit of Figure
2B.4.11 controlling the pressure drop across an orifice. The valve element used may be
a spool or poppet like the main stage of a relief or reducing valve with a relatively light
spring. this spring determines the pressure drop and this is usually in the 0.5 to 2.5 MPa
(5 to 25 bar) range. This can be used to give a higher efficiency system than with a fixed
relief valve setting when only part of the pump flow is used.
Counterbalance valve
This is a special name given to a compound valve which includes both a relief valve and
a check valve in parallel as shown in Figure 2B.4.12. This combination allows free flow
in one direction through the check valve and restricted flow in the other direction throttling
PRESSURE CONTROL V ALVES 183
through the relief valve when sufficient pressure difference is established. It is used
frequently and called by a number of other names some indicating the application, these
include locking valve, overcentre valve, load control valve and brake valve. The use of
counterbalance valves is generally associated with the effects of a moving mass in a system
and the more common applications are described in the chapter on Linear circuits. They
are widely available as single bodied units in all mounting forms including the screw-in
cartridge. There is also a very common variant which has a pilot operated check valve to
allow free flow in both directions when appropriate.
FIGURE 2B.4.12 - Counterbalance valve: combined relief and check valves.
Sequence valve
A sequence valve, as its name implies, ensures that actuation movements take place in a
defined sequence. It can be used where it is necessary to ensure that sufficient pressure has
built up in one circuit before fluid is admitted to another circuit. A typical use is shown in
the circuit of Figure 2BA.13. When flow is switched to the circuit cylinder 1will receive
the full flow and extend fully. When it reaches the end of its stroke the supply pressure will
increase until the sequence valve is actuated. This will open the flow through to the second
part of the circuit and cylinder 2, which will then also extend.
In its simplest form it comprises a spring loaded spool valve with primary and secondary
ports, the spool being normally positioned so as to shut off the secondary port. The primary
pressure acts on the end of the spool, against spring pressure. When sufficient pressure is
present, the spool is moved against the spring, opening a connection through to the
secondary port. The valve also has a throttling action, which prevents the primary pressure
falling suddenly, but opens fully when the working pressure is reached. A check valve may
be incorporated in the same body to provide free return flow of fluid in the non-controlled
direction.
This simple sequence valve is similar to a direct acting pressure relief valve but the
normal return port must be capable of working at system pressures. A pilot operated
pressure relief valve may also be adapted for this purpose by connecting the drain from
the low-pressure side of the pilot section to the tank. Spool valves are, however, more
normally used. Some sequence valves operate with the main valve either fully open or
fully shut without intermediate metering to provide flow to a second circuit without a
pressure drop. Pressure may not be maintained in the primary circuit in this case.
For the return operation of the above circuit it may be necessary to add extra valves if
a particular sequence of movements is required. Otherwise the lightest loaded cylinder
will retract first and take all the flow.
Unloading valve
The unloading valve is based on a pilot operated relief valve construction but combined
with a full flow check valve as shown in Figure 2BA.14. The external pilot connection is
taken downstream of the check valve to sense the pressure in the service side of the circuit.
When this pressure reaches the cracking pressure of the pilot stage the valve will open and
allow flow to pass to tank through the main stage. The main stage is designed to open fully
and not to throttle the flow. Flow forces may be used to advantage in some designs by not
having a momentum spoiler. This off-loads the pump completely which can continue to
run but with a minimal resistance circulating the fluid directly back to the tank. The
PRESSURE CONTROL VALVES 185
pressure in the service circuit can be maintained at pressure because the check valve
prevents flow returning through to the pump side. It is normal to include a smaIl
accumulator as shown to maintain the pressure since the pilot stage ofthe valve wiIl take
a smaIl flow while it is open.
This is an energy efficient solution where flows are required intermittently by a number
of services but the pump cannot be stopped. When flow is taken by the service the pressure
falls and the pump flow is rapidly re-directed through the check valve as the pilot shuts.
It is also used frequently with multi-pump systems supplying a common ring main with
each pilot set at a slightly different pressure. The identical configuration is also called a
charge valve when used with a larger accumulator which meets some of the service flow
requirements.
Burst disc
Although not strictly a valve, a burst disc certainly has a pressure control function but in
a decidedly on/off manner. It can be used as an extreme safety device where there may be
reasons that a conventional relief valve may not be guarantied to operate. A burst disc is
a specially shaped and machined diaphragm of material which is designed to rupture at a
pre-determined pressure level. This will then dump the high pressure flow and off-load any
system which will faIl to a low pressure. This may cause gravity type loads to drop if other
precautions are not taken. Once fractured the disc must of course be changed. They are
available in different materials for compatibility with different applications and the
smaIler sizes are available in a wide range of pressure settings.
187
PROPORTIONAL VALVES
Proportional valve technology is a relatively recent innovation in the development of

hydraulic systems and is principally associated with electrical proportional technology.
The applications for proportional valves have grown to fill a gap which existed between
the requirements for a high performance closed loop system and the simpler electrically
operated on/off system. They have fulfilled requirements for improved quality, higher
productivity and greater flexibility in many manufacturing applications and this has been
achieved through more precise control of flow and pressure.
The expansion of proportional technology has however, been intimately bound-up with
the parallel developments taking place in digital electronics. The introduction of easily
programmable digital controllers for sequencing operations has provided the necessary
input for use with proportional valves. Without the capability provided by Programmable
Logic Controllers (PLC), the potential of proportional hydraulics would have been
considerably more limited.
Proportional valves were originally intended to provide an alternative control method
to that obtained by the manual setting of a system's valves. A single proportional valve and
its control electronics can replace quite complex combinations of flow and pressure
control valves, and give better circuit performance. The increased flexibility and ease of
adjustment meant that the potential market sector expanded as the potential was realised.
Such valves now prove an excellent solution in many open loop control applications
(either manual or computer controlled) and are also capable of fulfilling the requirements
of the less demanding closed loop systems.
It is important to remember that the term proportional valve is very difficult to define
in a precise way. In selecting any valve it is the matching of the specification to the
application requirement which is of most importance.
Proportional solenoids
The basis of many but not all designs of proportional valves is the proportional solenoid.
Indeed it is this device which probably gave its name to this series of valves. The
construction of a typical proportional solenoid is shown in Figure 2B.5.1. It is a device
which gives an output force proportional to an input DC current, and relatively independ-
ent of the position of the armature within the working stroke. This force is uni-directional
and attracts the armature to the pole piece at one end irrespective of the direction of current
flow.
In using the solenoid in a valve design this force must be balanced in some way. In
designs producing a pressure output signal the force may be directly balanced against the
required pressure acting on a known area. In a flow control valve the force is more
frequently balanced against a compression spring to give a fixed position of a valve and
hence a valve opening providing an orifice to meter flow.
Proportional solenoids also fall into a category which at one time was referred to as wet
solenoid. In such designs the armature moves in the working hydraulic fluid inside a sealed
pressure tube. The coil of the solenoid is a sealed unit fitted around the core tube. In some
cases there is an extension push pin on the end of the armature which allows the valve to
be operated manually.
Pressure control valves

Proportional valves can be built to operate as relief valves or as pressure reducing valves.
The electrical input stage is similar in both types and gives a pilot controlled pressure
signal but with a relatively low flow capability. This is used with a second stage which is
acting to give the main control function. Figure 2B.5.2 shows a typical pilot pressure relief
valve where the input solenoid force is balanced directly against a pressure force on the
tics to determine the output pressure, since other inaccuracies are reduced by the closed
loop controller. The improvements are obtained in steady state accuracy, hysteresis, and
rejection of external influences. Typical characteristics are shown in Figure 2B.5.4.
Operating limits in terms of flow and pressure ranges are very similar to the ranges
covered by conventional relief and reducing valves because of the similarity of the main
stages. The ability to control the pressure setting in a system at different stages in an
operating cycle has considerable benefits on system efficiency and life. The ability to
change pressures relatively gently in a system can also reduce shock loading with benefits
for all components limited by fatigue, and hence also for system reliability.
Flow control valves

The electrical inputs to flow control proportional solenoid valves follow similar construc-
tional features. In addition they may be used to control not just flow and hence actuator
or motor speed, but also the rate of change of flow and hence accelerations and
decelerations. This can be very important in controlling shock and pressure transients in
systems with considerable benefits in reduced noise, reduced leakage and improved life
and reliability.
Several types of proportional flow control valve are commercially available. These
range from two way cartridge mounted designs to multistage high flow manifold mounted
valves. A sectioned view of a cartridge valve is shown in Figure 2B.5.5. This uses the same
force control principle but acting on a pilot valve to control the pressure balance at the ends
of the main spool which in turn moves to meter the main flow through the valve.
It is possible to obtain a wide range of 2-way flow control valves with a proportional
solenoid input, particularly in slip-in or screw cartridge form. The majority of character-
istics are common with those for 4-way valves but with a higher maximum flow capability
(2000 Llmin possible). Designs may be single, two stage, or even three stage.
the maximum flow implies that it is a two stage design, the flow limit tor a smgle stage
valve is in the order of 100 L min-! because of the flow forces involved.
Four way designs are probably more common and show the development of the
proportional valve from the older directional control valve background. A sectioned view
of a force controlled design is shown in Figure 2B.5.7, with typical characteristics in
Figure 2B.5.8. It is necessary to use two solenoids since they can only provide a force in
one direction.
There is a deadband given by the spool overlap which is usually in the region of 10 to
25% of total spool travel. This is included to keep manufacturing costs low, despite the fact
that spool diametral clearances can be as little as 5 to 811m. The "apparent" deadband can
be reduced by increasing the amplifier output in the null region gi ving only lor 2% residual
effect. This deadband can of course be an advantage in open loop operation since the zero
flow condition will not be susceptible to drift or minor changes in null setting.
Solenoid based proportional valves have a maximum flow limit when the solenoid
reaches its maximum output force. It represents a saturated condition when the solenoid
can no longer hold the spool in the required position against the flow forces. The shape of
this characteristic can vary but it typically resembles a power limit as shown in Figure
2B.5.9. In many of the simpler designs there is an element of flow force compensation
since for a fixed input current, flow forces will cause the valve spool to close slightly as
the flow increases. Two stage designs are required to cope with flows of over about 80
Llmin and an example is shown in Figure 2B.5.1 O.
Flow valve using the proportional solenoid in the stroke controlled mode are also very
common. These have a position feedback sensor built into one end cap of the valve. As
with the pressure controlled valves these have improved characteristics. It is necessary
with stroke controlled designs to use the manufacturer's drive amplifier since this will
contain all the settings for correct closed loop operation of the complete valve. A two stage
design is shown in Figure 2B.5.11, with electrical feedback from both the pilot valve and
the second stage spool. This then has two position control loops and represents the most
sophisticated alternative. The amplifier supplied by the manufacturer for use with the
valve includes all the signal copditioning for the transducer and may also include PID and
more sophisticated control loops. There is a trend towards the in~gration of the electronics
and the valve as a single assembly. This improves inter-changeability since there is
consistent performance between valves.
Valves with electrical feedback have generally better steady state characteristics.
Hysteresis for a force control valve, typically in the range 4-8%, can be compared with <
2 % for a valve with electrical feedback and an equivalent flow range. Repeatability is also
improved, although flow force effects on the spool position may in fact become more
PROPORTIONAL VALVES 195
exaggerated. The spread in step response is quite small with the majority of rise times in
the range 25 to 45 ms but some of the larger valves have settling times of nearer 100 ms.
These times are often quoted for operation in one direction only. Where the rise time for
-90% to +90% of travel are quoted by the manufacturer, it is approximately double the
equivalent single sided time. Typical step responses are shown in Figure 2B.5.12.
Bandwidths may not always be quoted and could be as low as 2Hz but more frequently in
the region of 10 to 20 Hz and up to 50 Hz is possible. A typical frequency response is shown
in Figure 2B.5.13. It should be noted that the dynamic response of single stage valves is
not dependent on the supply pressure.
FIGURE 2B.5.13 - Typical frequency response curve for a proportional valve.
The addition of electrical feedback has less effect on the dynamic response. However,
there are some quicker valves which extend into the range considered as servovalve
speeds, and these generally use electrical feedback. They are normally of a sleeved
construction where the spool moves in a separate bushing as in most servovalves, and they
have zero lap. The low amplitude 25%) bandwidth may be in the region of 120Hz and
with a very low hysteresis of < 0.2% quoted. Such designs are intended for closed loop
applications.
Proportional valves are generally manufactured in monoblock form some with multiple
spool banks which are primarily for mobile applications. They commonly have standard
interfaces and can therefore be used with sub-plates, manifolds or to terminate stacking
systems. Mobile valves are more likely to have a direct threaded connection. The
maximum flows range from 10 to 500 Umin for a 10 bar pressure drop with the majority
of valves rated to at least 300 bar. Some valves take relatively large currents of 2 to 3A,
whereas many have an 800 mA maximum. Spool alternatives include metering on all four
lands, only on the meter out lands, and the matching of metering area for a 2: 1actuator area
ratio. The shaping of the notches on the spool may be parallel, vee or curved. This shape
defines the relationship between spool movement and flow area.andhence the shape of the
flow characteristic as in Figure 2B.5.6.
Although some proportional valves are not as appropriate for closed loop systems as
servovalves they can certainly be used in the less demanding applications. Due care must
be taken of their slower dynamic response and the effect of the dead band on the accuracy
of the controlled output. These'aspects are discussed more fully in the chapter on
Servosystems.
Other designs
There are many more variations within the proportional valve category both for
electrohydraulic operation and control. In addition to some alternative pilot stages, there
are also some valve designs which are marketed as proportional valves using force motor
or torque motor input devices. Since these are more associated with the higher perform-
ance servovalves they are described in that chapter. Designs based on both stopper motors
and DC servomotors have also been used. These motors can give operation such that a
valve will hold its position despite a loss of electrical power. A two stage design capable
of operating up to 124 MPa (1240 bar), with or without position feedback, is shown in
Figure 2B.5.14.
Other novel pilot stages have the similarity of controlling the pressure on the ends of
a conventional spool to control its position and hence the main flow. In one example
currently available the spool position is sensed with a transducer and used as an electrical
feedback signal to give closed loop control. Operated as pairs in opposite arms, these can
connect the spool end caps either to a pressure supply or to tank. When energised, the spool
can be moved in either direction to give the required position. A set of on/off solenoid
valves are used arranged in a bridge circuit as shown in Figure 2B.5.15. These valves are
intended for operation in the mobile and industrial sectors with an appropriate response
time.
FIGURE 2B.5.15 - Pilot circuit using on/off solenoid valves [Danfoss].
Two-way valves are also widely used for flow metering, particularly for the higher
flows. One novel design for position control of a poppet as a two-way flow control valve,
with compensation for changes in pressure drop has been made commercially with the
name Valvistor. The principle of the valvistor is shown in Figure 2B.5.l6
The proportional plot stage opens to provide a flow path from the top surface of the
control poppet and with a reduced pressure this will move up. As it does so, it opens a flow
connection through a slot shown on the right side of the poppet. The poppet finds a position
where the pilot flow is balanced, and so the higher the pilot flow the wider the poppet will
open. A four-way valve function can be constructed using these elements with other logic
valve elements if necessary to give a variety of valve functions.
Load independent proportional flow control

All the proportional flow control valves discussed above set an orifice area with the
position of a spool (or poppet). For an orifice the flow rate will depend on the pressure drop,
and so can vary with the operating conditions of the load. It is possible to add pressure
compensation as described in the chapter on Directional and flow control above. This
combines the features of electrical modulation to give magnitude and directional control
of flow but without load sensitivity. This can be achieved by the addition of a sandwich
valve slice which fits between the proportional valve and its manifold mounting. These
operate as series compensators, intended to work from a nominal constant supply pressure.
They are available for operation giving either meter-in or meter-out control as shown in
Figure 2B.5.17. The flow control obtained is not as accurate as for a two-way pressure
compensated valve but a wide range of flow sizes up to about 700 Vmin is available.
FIGURE 2B.5.17 - A sandwich module containing two meter-out compensators.
There is, however, a family of valves with the capability for combined directional and
compensated flow control. These were mainly developed for mobile use but are appropri-
ate in many industrial applications. The valves are constructed in multi-spool assemblies
with other features also associated with the mobile central by-pass category including a
built-in relief function. The concept can be illustrated with Figure 2B.5.18 which shows
a by-pass type compensator connected across a proportional spool valve orifice. The by-
pass valve acts to maintain a constant pressure drop across the spool valve, and will open
if this pressure drop is too large to correct the situation by reducing the supply pressure.
This has the effect of keeping the supply pressure at a fixed increment above the load
pressure. This increment is pre-set by the spring in the compensator to be in the range 0.3
to 1.2 MPa (3 to 12 bar), and can usually be adjusted. This means that such systems are
operating with a good efficiency maintaining the pump pressure only a little higher than
the pressure required by the load. The surplus flow is of course still by-passed to tank at
this same pressure, potentially representing the major loss.
This concept can be easily extended to give this compensation to the other service port
as shown in Figure 2B.5.19. It also easy to extend this to a second spool as an alternative.
The check valves shown will switch the pilot connections to allow the higher pressure to
take control of the compensator. The lower pressure service will have to throttle the
additional pressure drop since it is not possible to obtain compensation on two services
simultaneously with this design. An alternative scheme is shown in Figure 2B.5.20 where
this is possible.
FIGURE 2B.5.20 - Twin spool series compensated proportional valve.
This version includes a compensator for each spool of the valve bank, working as a
series type compensator. However, there is also a by-pass compensator controlling the
supply to the valve and hence the system pressure. This is pilot operated from the service
requiring the highest pressure which again is automatically selected by the shuttle valves.
The by-pass valve will give the required compensation for this service and the series
compensators will step down from the pressure set on this valve to give the required
pressure drop across their own spool. This is a compromise in efficiency terms but is an
obvious improvement over a fixed supply pressure system.
One further step can be made based on this design if a variable displacement pump is
used for the supply. In this case the return flow from the by-pass compensator can be
passed through an orifice and the pressure drop used to activate the pump displacement
controller to minimise this "waste" flow. This concept is setting both the system pressure
and the flow produced by the pump to match the requirements of the valve bank and is
hence a power matched system. Very high power transfer efficiencies can be obtained
with this level of sophistication. An example is shown in the chapter on Mobile and
agricultural hydraulics.
Multi-spool valve blocks are available for all these designs extending these ideas to
operate with more services.
201
SERVOVALVES
THE TERM servovalve defines valves which are specifically designed to operate as the
control element in a servo system. This does not preclude other valve types being used in
servo systems. There is a chapter devoted to Servo-systems but a working definition is
appropriate here. A servo-system is one in which a low powered command signal is
compared with an output and the error augmented by an auxiliary power source to give a
high power drive to control that output. The power density of hydraulics makes them
highly appropriate for this use. There are many examples of purely mechanical servo-
systems where the command is produced as a result of an error signal between a required
value and the actual output. However, there are considerable advantages for making an
electrical closed loop system as discussed in the chapter on Servo-systems.
A servo-valve thus has the appropriate characteristics to give a continuous variation of
flow with changing input signal. In a purely mechanical servo-system the valve is moved
directly by a mechanism or linkage. An electrohydraulic servo-valve can be defined as an
electrical input servo-control valve capable of continuous control for use in closed loop
servo systems. The control element is almost universally a spool. There are various
constructional features that are common to both mechanically operated andelectrohydraulic
spool valves, and these are discussed below, thereafter this chapter concentrates on aspects
of electrohydraulic servo-valves.
The servo-valve spool

Servo-valve spools are precision made, with fine clearances and limited spool travel,
perhaps only 0.3-3 mm. They may be either 3-way or more commonly 4-way valves. A
typical configuration for the main spool of a two stage valve is shown in Figure 2B.6.1.
Note that the four connections are not arranged as they would be in directional val ves, and
that S I and S2 indicate the output service ports.
The spool and its sleeve or bushing are both made from hardened steel, frequently a
hardenable chromium steel is used. The bushing is manufactured in finished condition and
a spool is then matched to an individual bushing. This involves final grinding to correct
diameter and then face grinding of the metering edges to give the required land length to
match the internal metering dimensions of the bushing. The diametral clearance between
the spool and bushing is typically of the order of 3 /lm, and good roundness is essential
for both the spool outside diameter and bushing bore.
The relationship between the land length and the bore dimension is referred to as valve
lap. A zero lapped valve has the land metering edges exactly adjacent to the annular
opening in the bushing as shown in Figure 2B.6.2b. An overlapped valve, as with most
other spool valves, has a spool land which is longer than the adjacent gap in the bushing.
Overlap is not frequently used in servo-valves and would usually be less than 5% of the
total spool travel. The third alternative is an underlapped spool where the land length is
shorter than the metering connection as in Figure 2B.5.2c. Again only underlaps of up to
10% are usually used but a fully underlapped spool is feasible.
The differences in lap have a significant effect on the flow areas around the null region
as the valve opens, and give the overall flow characteristics as shown in Figure 2B.6.3. An
overlapped valve will give no flow until the metering edges are adjacent with the bushing
passages. An exactly zero lapped valve gives a linear relationship between spool
movement and flow which is highly appropriate for operation in many closed loop
systems. Since exact zero lap is difficult to guarantee on a production basis, a standard zero
SERVOVALVES 203
lapped valve has a tolerance which may err on the slightly underlapped or overlapped side.
An underlapped valve has a higher flow gain whilst passing through the underlap region.
It also has a steady leakage flow whilst at null which again can give benefits in some closed
loop systems as described in the chapter on Servosystems.
As in proportional valves it is common to reduce the effective flow area to less than the
full collar area opened by the spool. In servo-valves this is usually achieved by machining
slots in the wall of the bushing bore to give the connecting flow area. These are usually
parallel sided and produced by Electro Discharge Machining methods. Standard valves
are made with symmetrical opening areas in both flow paths and both directions of travel
but this need not be so. Some applications use valve flow areas matched with the actuator
piston to bore area ratio ego flight simulators.
Contamination effects in servo spools

The smallest clearance in a valve is generally between the spool and bushing. This can
sometimes act as a filter in a system if the spool is stationary in its null position for a long
period. Fine contaminant can build up in the radial clearance between the spool and
bushing and this is known as silting. This can increase the forces required to move the spool
and in severe cases may cause the spool movement to become erratic. To overcome the
increased stiction due to silting up, the trend has been to increase the power capacity of
the device moving the spool if the resulting increase in weight is permissible.
Another factor related to the forces moving the spool is referred to as a chip shearing
action. If a large particle of contaminant (the chip) becomes jammed between the metering
edges then a large force capable of shearing the chip is applied to the spool. It has been
argued that this self-clearing action is not always a desirable feature may result in transient
disturbance of the system whilst the chip is being sheared and damage to the hardened
metering edges. However, the alternative of a jammed spool may have far worse
consequences.
Over long periods of operation, even in relatively clean fluid, erosion of the metering
edges may occur. This erosion produces an effective underlap of the valve, increasing the
flow gain and leakage around null. The effect of a given degree of erosion is dependent
on the stroke of the valve: the longer the stroke of the spool, the smaller the effect of the
erosion. This is because a longer stroke means a narrower port; hence, less leakage and less
change in flow gain. A longer stroke also reduces the proportion of the valve's operating
range over which erosion effects occur.
Direct acting electrohydraulic servo-valves

There are servo-valve designs with electrical devices acting directly on the spool, and
others, the majority, where there is a pilot stage operating the spool. Direct acting valves
require enough electrical power to act directly on the spool and move it with precision,
hence they are confined to the smaller size and flow ratings. They have very good dynamic
response but cannot achieve the highest response levels of two stage designs. The response
of direct acting valves is nominally independent of the supply pressure but for high flows
and pressure drops flow force effectds can be significant.
Solenoid operated
This design style can be considered in either the proportional valve or servo-valve
categories but has been included here because its characteristics and applications are more
closely related to this chapter. A sectioned view of a typical design is shown in Figure
2B.6.4. It is a sleeved spool construction, using only a single proportional solenoid and
with electrical position feedback of the spool position. It has an interesting arrangement
in the de-energised position which is at one extreme of the spool travel and can give
different flow connections to the normal operating centre. When energised the solenoid
pushes the spool from the end position to a conventional centre position and then
modulates about this point.
As with all other solenoid valves with position feedback, these valves must be used with
the manufacturer's closed loop amplifier if the full performance is to be obtained. Some
have been developed with digital closed loop control giving very robust characteristics.
Force motor operated

Force motors operate with a DC current in a coil producing an interaction with a magnetic
field, usually a permanent magnet, and these days usually a rare earth magnet, giving an
output force. Reversing the direction of current flow wi II change the direction of the force
and hence they are bi-polar devices (unlike solenoids). A cross section of a force motor
type industrial valve is shown in Figure 2B.6.5.
SERVOVALVES 205
Force motors have lower moving mass and larger driving forces than proportional
solenoids. Therefore they have potential for more rapid response and greater chip shear
capability. They also have a lower coil inductance and this again helps dynamic response
by reducing the electrical delay in changing the current. It is the advent of rare earth
magnetic materials at a reasonable cost that has allowed the development of commercial
designs for industrial use with good performance characteristics.
Aerospace valves
The use of direct acting valves has been most common in military aircraft applications
where they are replacing the two-stage designs. The multiple redundancy in operating
flight control surfaces can be obtained in an attractive compact package with such designs.
The stringent safety requirements require high quality products which are expensive in
industrial terms. Figure 2B.6.6 shows a typical design.
Pilot stages for electrohydraulic servo-valves

The majority of servo-valves are two or more stages with a second or main stage spool as
described above. The term pilot stage is used here to cover the conversion from an
electrical signal to a useful pilot pressure and flow acting on the spool to move it. The
power magnification in the pilot stage is usually very large and so electrical input powers
may only be of the order of I or 2 w. It is however possible to use the direct acting designs
described above to pilot larger spools for high flow applications.
Torque motor
The most commonly used input device is the torque motor as shown in Figure 2B.6. 7. This
comprises a moving armature which is able to rotate between the pole pieces of a
permanent magnet. There are two coils wound on formers around each arm of the armature
and a current flow in these magnetizes the armature producing a North pole at one end and
a South pole at the other. This causes both the armature ends to be attracted to the pole
pieces (one up and one down) producing a turning moment or torque. Reversing the
current direction will change the sense of the torque produced.
Torque motors are relatively small low powered devices with input currents of below
200 mA. The figure illustrating the design can be taken as approximately full size. The total
movement possible at the ends of the armature would normally be less than 0.5 mm. The
torque produced on the armature is proportional to the coil current for small deflections
of the armature. The armature is usually connected to a flapper, and both are supported by
a tube of non-magnetic material called the flexure sleeve. This both supports the armature
allowing sufficient movement and also provides a hydraulic seal separating the magnetic
components from the hydraulic circuit connected rigidly to the armature from below.
A change in the nominal zero position of a servo-valve can occur and this is called a null
shift. The null stability of a servo-valve is controlled principally by the stability of its
torque motor. Null shifts are usually caused by small internal mechanical movements in
the torque motor from stress and temperature changes and sometimes by the stress changes
in the magnetic circuit. To correct any null shift in service a number of methods are used
including a bias spring or magnet acting directly on the torque motor or mechanical
adjustment of the main stage bushing.
Nozzle flapper
The principle of a nozzle flapper valve is illustrated in Figure 2B.6.8, which shows the
more common double nozzle configuration. Each side of the system consists of a fixed and
a variable orifice in series, the variable orifice being composed of a nozzle with a flapper
plate very close to it. The flapper is rigidly attached to the armature of a torque motor. Flow
is metered from the high pressure supply through the fixed orifices and between the flapper
and the circumference of the nozzle. In mid position the control pressures at A and B will
be the same and equal to about halfthe supply pressure. A current in the torque motor in
one sense applies a torque to the flapper to move it towards one nozzle and away from the
other. This increases the restriction one side and decreases the flow through that nozzle,
causing a decrease in pressure drop across the fixed orifice and an increase in the control
pressure on that side. The control pressure on the other side will reduce by the same amount
because ofthe higher flow . A maximum swing in pressure from about 15 to 85% is possible
giving a high power gain.
The double nozzle flapper arrangements typically have a nozzle diameter about 500 IJ.m
and a fixed orifice diameter a little smaller at about 200 IJ.m. Whereas, the clearance
between the nozzle and the flapper is only 25 to 50 IJ.m. This clearance limits the
contamination particle size which this double nozzle version can handle without malfunc-
tion. Usually a filter element is built into a valve design to prevent particles of critical size
entering the pilot circuit. However, if one half of the circuit is blocked it can cause the spool
to move to one extreme position. During normal operation the flapper should not touch the
nozzles. Contact can occur if the valve is run when dry or the coils are overdriven and this
may cause damage.
Single nozzle flapper valves are available which operate on the same principle. The
movement of the flapper away from the single nozzle is not tightly constrained and this
type is, therefore, generally more tolerant of contamination. A single nozzle configuration
is unbalanced and susceptible to null shift which may occur due to changes in supply
pressure, temperature or other operating conditions. The double nozzle configuration is
not inherently subject to null shift with temperature and supply pressure variations.
These systems require a flow even in the null position, called a quiescent flow. It
represents a power loss but is usually relatively small. It is possible to use either a single
or double nozzle flapper valve as a single stage device with maximum flows of up to 10
Urnin.
Jet pipe valve
The first stage of ajet pipe servo-valve consists of a torque motor, ajet pipe and a pair of
receivers as shown in Figure 2B.6.9. The jet-pipe is rigidly attached to the armature and
rotates with it. A small flow of high pressure fluid is fed by a flexible tube to the jet pipe.
As the high velocity oil flows out of the end of the jet pipe, it impinges upon the face of
the receiver mounting. The connections to the receiver volumes are through two small
diameter holes located side by side in the receiver plate. With the jet-pipe centred over the
two holes, the flow splits equally and the kinetic energy of the jet causes equal pressures
to be developed in both receivers. When the torque motor causes the jet-pipe to rotate off-
centre the flow is directed preferentially to one receiver and the pressure recovery from
this jet causes an increase in pressure on that side. A similar decrease occurs in the other
receiver giving the control pressures to move a spool.
A jet-pipe/receiver combination is generally more tolerant to contamination and is

relatively insensitive to null shift arising from changes in temperature and supply pressure.
If the jet itself is blocked then the valve spool will simply stay in the null position. Control
pressure changes of between 10 and 90% are possible. Long term wear of the receiver plate
under the jet action can reduce the pressure gain. The jet pipe design also requires a
quiescent flow when in the null position but usually less than an equivalent nozzle flapper.
Two stage valves

Both nozzle flapper and jet pipe pilot stages produce a control pressure difference which
can be applied to the ends of a spool giving sufficient force to move it. In mechanical
feedback designs there is a connection from the spool back to the pilot and the torque motor
armature called a feedback spring or wire. This is shown in Figure 2B.6.1O for a nozzle
flapper valve but a similar arrangement is used in jet pipe valves.
The feedback spring between the spool and flapper applies a force to the armature to
return it towards the null position. When the feedback force and torque motor forces are
SERVOVALVES 209
equal, the flapper is restored to its null position, the spool forces are balanced and the spool
stops. As the spool displaces from null, it deflects the feedback spring, developing a force
counter to the input torque. Spool displacement is thus proportional to the torque input and
for practical purposes, a linear relationship exists between spool position and input
current. This flow-control characteristic is fully reversible and continuous control of spool
position can be applied from one extreme to the other. For the alternative pilot stage, the
feedback spring applies a similar restoring torque which returns the jet-pipe to null and the
spool again comes to rest in its new position.
The feedback spring is usually connected to the torque-motor member (jet-pipe or
flapper) with a ball on the end, fitting into a slot in the spool. This requires an extremely
close match between ball and slot. If a clearance develops, then spool drift can occur with
degradation of threshold, resolution and repeatability. There are some designs with
jewelled bearings which are used to prevent wear. There is also the possibility of closing
the loop between the torque motor and the second stage spool with electrical feedback. The
spool position is detected by a transducer and used as a feedback signal. This offers the
potential for considerable improvement in steady state accuracy as well as higher dynamic
response.
Two-stage designs can give maximum flows from 2 to 200 Umin and this rated flow
is usually quoted for a pressure drop of7MPa (70 bar) across the valve. This means that
there is a symmetrical 3.5 MPa pressure drop across each metering path in the valve. The
movements are small, even a valve passing 60 Umin may have a spool displacement of
only 1 mm. The same valve body size is available with a range of different flow ratings
obtained by changing the width of the flow metering slots in the bushing. Maximum inlet
pressures are in the range 20 to 35 MPa (200 to 350 bar). There is usually a wide range of
choice for the input current corresponding to maximum opening (the rated current).
Values typically from 10 to 200 mA being quoted for one valve model, and this gives a
very high power gain particularly ifthe low current variants are used. The very low input
current required can also have advantages in hazardous environments where an intrinsi-
cally safe device is necessary. Mechanical feedback valves typically have hysteresis less
than 3 to 4%, although low cost versions may exceed this. Electrical feedback valves have
hysteresis usually below 2% and some versions are less than 0.5%.
It is in terms of dynamic response that servo-valves show their best performance and
typical step responses are shown in Figure 2B .6.11. Rise times can be less than 2 ms and
settling times frequently under 5 ms, with even large valves responding within 20 ms.
Larger step sizes give a progressively slower rise time due to flow limits of the pilot stage
as shown in the figure. These step responses correspond roughly to a -3dB bandwidth of
20 to 150 Hz, but are again dependent on the test amplitude with manufacturers frequently
quoting data for both 25% and I00% amplitudes. All dynamic responses are limited by
the supply pressure as indicated in Figure 2B.6.12. Higher pressures again give higher
pilot stage flow and a more rapid response.
FIGURE 2B.6.12 - Typical frequency response for two stage servo-valve.
Direct acting spool valves can also be used as pilot stages and can control larger main
stage spools. An example is shown in Figure 2B.6.13 with a proportional solenoid pilot
valve. Such valves can give flows of 2000 Llmin at a IMPa (10 bar) valve pressure drop.
Responses are of course slower and also supply pressure dependent with large valves
having step responses in the 50 to 100 ms region at normal operating pressures.
SERYOY AL YES 211
FIGURE 2B.6.13 - Two stage, direct acting pilot high response valve [Robert Bosch Ltd].
Three stage valves

Higher flow capability can also be obtained with three stage valves. These use a two stage
servovalve as a pilot for a third stage spool and hence give higher flow capacity. Feedback
may be direct electrical position sensing of the third stage as shown in Figure 2B .6.14 or
by indirect spring. Some of these valves give a very high flow capability up to 700 Llmin
at 7MPa (70 bar) valve pressure drop. Although the dynamic performance is reduced it is
still relatively rapid, with rise times of under 40 ms quoted.
Valves with integrated electronics

There is a trend towards the use of valves with electronic position feedback of the main
spool. Many designs which are available commercially also incorporate the necessary
signal conditioning for the feedback sensor, the loop closure circuit and the torque motor
drive amplifier as an integrated electronic package. This gives the interchangeability
between items and avoids the complication of re-tuning a drive amplifier is a valve is
changed. There are also likely to be benefits in meeting EMC requirements. A sectioned
view of a design is shown in Figure 2B.6.15. This type of valve can now give a better
performance to cost combination than the more conventional mechanical feedback valve.
In particular the threshold and hysteresis are lower than most mechanical feedback valves.
Versions are also available with the mechanical feedback spring retained, which ensures
the spool position will return to null in the event of an electrical failure. The direct drive
valves described above are also available with feedback and an integrated electronics
package.
213
ACCUMULATORS
AN ACCUMULATOR for an hydraulic system is a specific device designed for the

storage of liquid under pressure. At the same time it can be effectively used to provide a
number of hydraulic system improvements, such as
i) Compensating for peak flows and pressures, by storing energy and thus economis-
ing on pump size.
ii) Providing a standby power source in case of emergencies such as pump failure or
fast shut down
iii) Compensating for leakage losses
iv) Damping out of system surges, shocks and vibration
v) Reducing pump ripple effects
vi) Compensating for volume changes due to temperature and pressure variations
vii) Aiding suspension damping of vehicles.
BasicalIy it is a container which includes a member which is resilient and alIows a
predetermined quantity of fluid, at a finite pressure, to be stored. The hydraulic fluid has
low compressibility, so some other means of resilient energy storage is required.
The resilience within an accumulator is usualIy provided by a compressible gas or
flexible load, such as a spring: The change in fluid volume within the accumulator will,
of course, lead to a change in the accumulator pressure. In a gas-loaded accumulator the
gas is normalIy separated from the hydraulic fluid by a physical barrier; in which case no
gas-saturated boundary layer is involved and the percentage capacity of such an accumu-
lator is high. They are the most common types of accumulator in use today, particularly
as they can be produced in compact form.
The gas-separator type accumulators can be broadly classified as having either flexible
or rigid separators. Flexible separators include diaphragms, bladders and belIows con-
straining the gas charges and mounted within a suitable pressure vessel. Rigid separators
take the form of a piston mounted in a cylinder.
Before describing the different types in detail it is valuable to appreciate the variety
which is currently available. Table 2C.l.l gives a brief outline of the main types, and Table
2C.l.2 gives an indication of the volumes and pressures with which each of the types is
normalIy able to operate.
Gas - rigid-separator-type accumulators
(Note that the gas normally used in gas-charged accumulators is nitrogen.)
Piston-type accumulator
A basic design of piston-type accumulator is shown in Figure 2C.1.1. Normally the piston
is 'free', but in some cases it may be connected to a conventional piston rod. Design and
construction are relatively straightforward and the type can be made in a wide range of
sizes. Cost is, however, relatively high. It is particularly suited for high-pressure systems
since cylinder stresses are readily determined and standard hydraulic quality cylinder
tubes can be employed for the barrels. Construction can follow that of hydraulic cy linders,
with tie-rod assembly for higher pressures or heavy duties. However, it is necessary to
make provision to prevent disassembly of the end covers when either the fluid or gas side
of the accumulator is under pressure.
The main disadvantage of piston-type accumulators is associated with the piston seal.
In normal operation there is no problem, but the task of maintaining an adequate seal when
the fluid system is shut down with the gas end still under pressure is difficult. Special seals
have been developed to eliminate roll-over of the seal lands (blades) to ensure as complete
as possible separation of oil and gas under the most severe operating conditions. A number
of designs include a liquid seal for additional protection.
With a liquid seal the fluid port is closed by a probe attached to the piston entering the
end cover at the end of its stroke and trapping a certain amount of fluid in the liquid end
of the cylinder. This amount of fluid is pressurised by gas pressure on the piston, but the
area on which it acts is less than the piston area. As a consequence the pressure generated
on the trapped fluid is greater than the gas pressure, preventing gas leakage into the liquid
end. Liquid seals also prevent hydraulic shock in the event of the fluid content of the
accumulator being fully discharged.
An alternative method of 'cushioning' the end travel of the piston is shown in Figure
2C.1.2. This is a dashpot-type piston-accumulator, where the fluid side of the piston
carries an extension of reduced diameter entering a cushion chamber in the cylinder end.
Should the fluid level fall to such an extent that the piston nose enters the cushion chamber,
dashpot damping is provided over the remainder of the stroke.
FIGURE 2C.1.2 - Dashpot piston accumulator with auxiliary bottle.

Solid piston accumulators are used where any sudden gas loss would be catastrophic.
They are often used in conjunction with back-up bottles, the additional gas volume
producing less pressure change. The response time of such devices is frequently as fast as
3 milliseconds. Any gas loss from these units will be relatively easy to determine and it
is therefore usually possible to plan their maintenance and the replacement of seals to
coincide with machine maintenance intervals.
ACCUMULATORS 217
Virtually all piston-type accumulators suffer to some extent from gas leakage which
may develop in use since no effective seals can be entirely free from wear. Such conditions
are aggravated by contamination of the fluid, or corrosion, which could affect the bore
finish or seal material. Periodic topping up of the gas charge is therefore normally
necessary in order to maintain a minimum working pressure.
Piston-type accumulators are useful for handling special duty fluids which may attack
conventional low-cost elastomers used with flexible separator accumulators; however,
fluid cleanliness is important in order to minimise seal wear. They would normally be
vertically-mounted, but horizontal or angular mounting is not necessarily excluded.
Whilst being ideal for storage of fluid at very high pressures and for providing damping
of system pressure and flow fluctuations, they are not so suitable as other types for shock-
absorbing duties, due to the inertia of the piston and the friction of the piston seals;
however, it should be noted that piston design and material has progressed towards a
lighter weight component than earlier designs.
Some solid-piston accumulators are designed with a tail rod which indicates the fluid
volume stored at any time. There are, however, other means of showing position, for
instance, for stainless steel cylinders, a magnet may be inserted in the piston and its
position detected externally, or, for ferrous designs, a spring loaded pulley system on the
gas side may be geared to a potentiometer. This is an advantage which can rarely be
obtained from a membrane type unit, and even then, not accurately. In addition, by
know ledge of the original precharge pressure and the use of a gauge connected to the gas
side of the unit, it is possible by correlating the position of the piston to the pressure shown
on the precharge gauge to ascertain that the unit has not lost any gas pre-charge. This may
be done at any time, i.e. it is not necessary to discharge the unit of liquid totally before
checking the precharge pressure to ascertain whether any of it has been lost.
The position indicator is frequently used to actually cut in or out of circuit one or more
pumps and this can be done eithernon-electrically by directly tripping the unloading valve,
or by tripping switches. The result is that it is not necessary to have more than about 0.65
MPa (6.5 bar) between one pump cut-in and the next or between pump cut-in and cut-out,
even when working at 35 MPa (350 bar) when these indicating-type units are connected
to gas bottles. This amount of control is extremely difficult to obtain even with the use of
the most sensitive pressure switches, and it is this degree of controllability which has
caused these units to be classified as 'pump controllers' rather than accumulators.
A special duty tandem type piston accumulator is shown in Figure 2C.1.3. It is also
called a self-displacing accumulator and comprises an accumulator combined with a
pressurised reservoir; it is thus capable of maintaining a constant volume of active fluid
in the hydraulic circuit. The gas precharge displaces the tandem piston to fill the low
pressure cylinder with fluid When the system is pressurised, the high pressure (hydraulic)
side ofthe accumulator is filled with fluid and the gas compressed. The fluid to fill the high-
pressure side is drawn from the low-pressure side. With the system working, any fluid
withdrawn from the high-pressure side is simultaneously replaced on the low-pressure
side, thus maintaining a constant volume of fluid both in the system and the accumulator
(apart from very slight fluid compressibility).
Annular-piston accumulator
An annular-piston accumulator is of conventional construction embodying a cylinder and
liner, the piston moving within the liner and the outer annular volume providing additional
gas volume - Figure 2C.IA. This has the advantage of providing a substantially larger gas
volume without increasing the length of the cylinder or employing a separate gas bottle.
In construction it has the advantage that the liner can be relatively thin since the gas
pressure on the outside is equal to the fluid pressure on the inside, and they thus offset each
other. Because the outer envelope does not house the piston, it may be constructed oflight
weight composites rather than a metal.
Membrane-piston accumulator
A membrane-piston accumulator incorporates a hollow piston constructed like a dia-
phragm - Figure 2C.1.5. This is a stronger form of separator than a diaphragm or
membrane and so such units can be designed to accommodate flow rates up to four times
as great as bag-type accumulators. The diaphragm within the piston also gives an
advantage over pure solid piston type units because the dynamic seals on the piston do not
have to move for every minute pulse or fluctuation in the system caused by pump- or servo-
oscillation. This advantage enables its dynamic seals to outlast those of a pure piston type
accumulator many times over.
The arrangement is such that the ports to the chamber are on a pitch circle diameter and
the ports to the diaphragm inside the piston are on the centre line. This makes it impossible
ACCUMULATORS 219
for the diaphragm to escape down the ports of the system and, as the piston moves at a much
lower pressure differential than the pressure required to extrude the diaphragm, the piston
itself acts like the anti-extrusion valve at either end of stroke. As a result of this
arrangement there is no anti-extrusion valve and in consequence the diaphragm's
sensitivity, unhindered by an anti-extrusion valve, is capable of dealing with high
frequencies from 10Hz to 1000 Hz. This combination piston-cum-diaphragm unit is also
used where the whole of any additional gas volume is required to be stored in additional
gas bottles.
Gas - flexible-separator-type accumulators

Bladder-type accumulator (closed bag)
Improved techniques in fabricating gas-containing bags by bonding a number of moulded
components together have resulted in bladder-type accumulators becoming the most
popular type. A typical design and operation is shown in Figure 2C.l.6; it consists of an
outer pressure vessel, normally cylindrical in shape with hemispherical ends, (but
sometimes a plain cylinder), and with drawn or forged construction to eliminate welding
or mechanical joints. Inserted into this pressure vessel is a bag of elastomeric material,
chosen for compatibility with the hydraulic fluid and temperature being used (normally
nitrile rubber for general use). The bag is drawn into position during assembly through the
wide opening which accommodates the fluid port assembly. The bladder is retained in the
shell via an integral gas valve assembly at its upper end and clamped in place on the top
of the shell with an external nut.
FIGURE 2C.1.6 - Bladder type accumulator [Robert Bosch].
The oil port is assembled in the other end of the shell, the joint commonly being closed
with an O-ring and the design adjusted so that the lower mouth of the shell will spread at
a pressure below the design pressure of the shell as a safety precaution. The val ve opening
is large to allow an un-restricted flow of oil. The large opening also allows the bag to be
removed for inspection or replacement should this be necessary. The poppet type valve
provides high volumetric efficiency with its main function to prevent extrusion of the
inflated bag when the fluid side is depressurised, or should all the fluid be drawn off. Other
safety factors normally include some safeguard to prevent removal of the fluid discharge
plug while there is any pressure remaining inside the bag.
In use, the bag is pressurised (with nitrogen) to the specified precharge pressure and
fluid is pumped into the main chamber to compress the bag. The gas precharge pressure
is invariably much lower than the fluid pressure and compression ratios of up to 5: 1 may
be achieved, according to the particular requirements of the installations. Nominal
maximum working pressure with this type of accumulator is commonly 21 MPa (210 bar),
although the same type can also be designed for higher rated working pressures, for
example, 35 MPa or even 69 MPa (350 bar or 690 bar). Wide pressure fluctuations occur.
The flexible bag is normally pear-shaped or similarly tapered, allowing it to reshape
more easily. Some modification of optimum shape may, however, be dictated by the
material used and the method of construction. Bag failure is unusual with modern designs,
although this can happen at a dangerously low level. This type of failure can occur where
an external gas bottle is used and the total gas volume falls appreciably due to a
considerable drop in temperature (as may happen over-night). If there is a risk of excessive
pressure ratios during operation, a portion of the bladder may be filled with oil; this
provides protection against damaging deformation but it will reduce the effective volume.
Another cause of failure can be a too fast a rate of precharge causing the bladder to become
overstretched and split: to avoid this a special transfer accumulator is sometimes used.
Bladder type accumulators are particularly suitable for use with oil fluids, but can also
accommodate other types of fluid provided that the bag material is compatible. Used with
water or water-fluids it is generally necessary to pre-treat the steel shell to prevent
corrosion. Epoxy lined interior finish for the shell is a typical treatment.
In the use of a bladder type accumulator it must always be remembered that if the bag
is punctured, the gas loss is likely to be sudden and total. For this reason, in order to prevent
the bag being damaged by contacting the bag anti-extrusion mechanism, the flow rate
capabilities of this design are limited. If used with gas bottles to increase capacity, it must
be remembered that in order to prevent the bag from being forced into the piping system
which runs to the bottles, it is necessary to reduce the ratio between the volume of the
accumulator and the total volume of the gas bottles. Additionally, to prevent bag damage
caused by violent changes of gas temperature entering the bag through small apertures or
sintered plugs, the flow rate of bag type accumulators connected to gas bottles must be kept
to the minimum. It may even be necessary to increase the number of bag type accumulators
connected to the gas bottles.
Bag-type accumulators are the most versatile of all types and are equally effective for
energy storage, shock absorption and 'holding', and 'reservoir' duties (fluid leakage
make-up and fluid volume compensation due to temperature changes). They also provide
effective damping of pump pulsations and are widely used for such other duties as gas
springs, pressurised fluid dispensers and transfer barriers between two fluids. Note that
they are normally installed vertically with the hydraulic connection at the base; however,
horizontal installation may be possible, but flow rates will be reduced.
ACCUMULATORS 221
Bladder-type, or cylindrical-type, accumulator (open-bag / open-top)

An open-top accumulator is shown in Figure 2CI. 7. Such devices are frequently called
open mouth, or cylindrical, units. They are low cost units, sometimes being manufactured
in bag materials which are glueable, they also have servicing, inspection and replacement
benefits as the complete cross section of the tube is laid open by the removal of the header
plug. This header plug is also used to seal the open mouth of the bag and because the bag
has an open mouth the internal shape mould former is easily removed from the moulding,
thus allowing the bag to be produced in a single operation. The header plug of the assembly
therefore provides a surface through which charging valves, venting valves, gauge
installation with isolation valves and gas back-up bottle connections can easily be
introduced in conjunction with safety over-load vent gas depressurisation rupture discs.
The units are specially designed for high volume, low cost, applications where it is more
economical to replace the unit (or the bladder) rather than refurbish it after long service.
Diaphragm accumulator
The original form of diaphragm accumulator was a spherical vessel accommodating a
moulded flexible membrane separating the chamber into two equal volumes. Today
diaphragm designs and materials may vary considerably. Construstion may be a convo-
luted form in synthetic rubber, the convolutions providing minimum creasing with
maximum flexibility of movement, and, in particular, maintenance of flexibility at lower
temperatures where elastomeric materials tend to harden. A rather simpler form of
diaphragm is shown in operation in Figure 2C.l.8.
The gas volume is charged at a lower pressure than the fluid volume. When the
accumulator is initially charged with gas to the required pressure the diaphragm is fully
flexed with the gas occupying the full volume of the accumulator. Fluid is then pumped
into the high-pressure side, compressing the gas and establishing a balance with equal
pressure on each side of the diaphragm, and therefore no stress (other than pure
compression) is placed on the diaphragm material.
A safety device would be incorporated to ensure that should the fluid chamber become
completely exhausted the diaphragm cannot extrude through the fluid port under gas
pressure.
Accumulators of this type are compact and light, but best suited to systems where
demand is intermittent and the fluid volumes required from the accumulator flow are not
very large. Mounting is usually possible in any desired orientation
Gas - non-separated-type accumulators

Non-separated gas-loaded accumulators are less common, but for the rare cases they are
used they have been included here. They take the form of a cylindrical pressure vessel or
air bottle mounted vertically, with a fluid port at one end and a gas port at the other. Fluid
is first introduced into the vessel and then pressurised by a precharge of gas to the required
minimum pressure. Further fluid can then be pumped in, compressing the gas and
increasing the pressure. Because of the high compressibility of gases, a large volume of
ACCUMULATORS 223
fluid can be accommodated, capable of being delivered from the accumulator above the
minimum pressure specified (Figure 2C.1.9).
The particular disadvantage offered by such a simple system is the mixture of fluid and
gas at the interface. This limits the amount of fluid which can be drawn off without
incurring the danger of drawing off gas into the hydraulic system. Also the cylinder must
be mounted vertically. On the other hand such accumulators are simple and comparatively
inexpensive and well suited to handling large volumes of fluid; and thus find widespread
application. Storage volume can be readily increased by multiple installations.
The shape of the pressure vessel is invariably tall and narrow so that the contact area
between gas and fluid is small. Even so, probably not more than two thirds of the fluid
volume can be used without the danger of gas being drawn out into the hydraulic circuit.
The pressure vessel may be connected directly to a high-pressure gas compressor
operating against a level regulator, and with pressure regulator valves and controls.
Controls are normally automatic, and could include monitoring of the actual fluid level
inside the liquid bottle. This could be by continuous pressure measurement; or by floats
coupled indirectly to external indicators; or electronically.
Once precharged, no gas is actually used up during cycling, other than small amounts
that may be lost by absorption. Thus no periodic topping up of the precharge is necessary,
unless there is a definite discharge of the gas into the fluid circuit through operational error
or malfunction of the accumulator.
Non-separated accumulators of this type for general industrial applications cover
pressure ranges from 3.5 MPa (35 bar) to 42 MPa (420 bar). A typical installation would
comprise a single liquid bottle with three auxiliary gas bottles of similar size and the
control system already shown in Figure 2C.l.2.
Float-Type Accumulators
Float-type accumulators were developed to overcome the interface mixing of gas and fluid
inherent in non-separated accumulators by reducing the area of fluid in contact with the
gas. Construction is considerably more complex, but such units find application in very
large installations where the cost of making a mould tool for the manufacture of a large
membrane to separate the gas from the liquid is prohibitive, and where the cost of honing
the bore of a large pressure vessel in order to run a seal piston separator in it is equally
prohibitive. They are also still used in installations which cannot easily obtain spare parts,
e.g. for blow-out preventer standby hydraulic energy on oil rigs at sea; and systems in less
developed semi-industrial countries where seals and rubber technology are not readily
available.
Non-gas accumulators
Accumulators which are not gas-loaded are also uncommon. There are still, however, a
few examples of their use and hence, they are described below.
Spring-loaded accumulator
Spring-loaded accumulators can perform the same duties as a weight-loaded accumulator
where volume demand is small. A basic design is shown in Figure 2C.l.l O. This comprises
224
a free piston sliding in a cylinder, with a compression spring (or springs) mounted in the
blind end ofthe cylinder. The accumulator is charged by pumping fluid into the cylinder,
displacing the piston towards the blind end and compressing the spring. The fluid is thus
pressurised by the spring force.
The pressure generated by a spring-loaded accumulator depends on the initial loading
and spring rate of the spring, and thus is not constant throughout the stroke of the piston.
Pressure will vary from a maximum when the spring is fully compressed, to a minimum
when the spring is fully extended, unless constant-rate springs are employed.
The particular advantage of a spring-loaded accumulator is that it is compact in design,
light in weight and can be used in any attitude. Basically, however, it is only suitable for
relatively small capacities and low pressures. It is also not generally suitable for high
cycling rates because of the limited cycling life of mechanical springs.
Weight-loaded accumulators
The basic design of a weight-loaded accumulator is shown in Figure 2C.1.11. A heavy
walled cylinder is mounted vertically on a substantial base and carries a ram. A cross-head
ACCUMULATORS 225
is attached to the top of the ram, from which is slung a weight box. This is filled with any
high-density waste, such as ballast, iron scrap, concrete, etc. Alternatively, in the case of
smaller units, specially made weights may be slung from the ends of the crosshead.
There are two main types, depending on the method of constraining the weights or
weight box. On a self-guided design the weight case is provided with internal guides. On
externally-guided designs the weight case is constrained against radial movement by
external guides or channels, usually mounted on a steel structure. The latter type is
normally preferred to large high-pressure accumulators to minimise bending stresses.
The ram is raised by pumping fluid into the cylinder. Once raised, the fluid in the
cylinder is pressurised by the combination of the weights and ram acting on the cross-
sectional area of the fluid column.
Full theoretical hydrostatic pressure should be available from a weight-loaded accumu-
lator, less a nominal allowance for seal friction. However pressure variations are likely to
occur with differences or variations in falling speed. Thus a pressure variation of 5% is
likely to be experienced with a maximum falling speed of 0.3 mis, but may be higher with
higher falling speeds. Momentary peak pressures may also be higher or lower than the
nominal pressure by an appreciable amount, depending on the rate of deceleration or
acceleration of the ram, respectively.
Falling speed can be controlled by the stroke/bore ratio of the ram. A stroke/bore ratio
of between 10 and 15 is commonly adopted for accumulators working up to 10.5 MPa (105
bar), although higher ratios are generally to be preferred for higher pressures. This,
however, increases the problem of obtaining mechanical rigidity and also increases the
overall height of the accumulator. This could make it unsuitable for indoor installation. As
a rough guide, the overall height of a weight-loaded accumulator is at least twice the stroke.
Cast iron cylinders are commonly employed for accumulators working up to 10.5 MPa
(105 bar). Cast steel or forged steel cylinders are used for higher pressures. Honed bores
are required, although satisfactory performance may be obtained with rougher bores using
leather seals. Rams may be made from cast iron (the original choice and still widely
employed), but preferably chrome-plated. Stainless steel or alloy steel rams are more usual
on smaller sizes of modern weight-loaded accumulators.
Weight-loaded accumulators continue to be used to meet heavy industrial requirements
and large units usually employ water as the fluid. The large weight-loaded accumulator
offers the advantage of extremely high capacity at relatively low cost per unit volume.
Construction is rugged and d~rable, and the units are capable of accommodating shock
loads. Only simple control gear is necessary.
The disadvanta~s of a weight-loaded accumulator are:
i) The accumulator is extremely bulky and heavy and thus could not be considered
where space- or weight-saving is an important factor.
ii) Pressure output is not quite constant, largely due to the effects of seal friction and
inertia; however,it is much better than the gas-type.
iii) Certain restrictions are imposed on delivery, largely due to limitations on falling
speed to minimise hydraulic shock
iv) The seals themselves may pose problems, both in providing adequate sealing with
low friction when they are used with such a low-viscosity fluid as water, and when
expected to give long seal life. Where such an accumulator is used as a central
source, failure of the seals would result in loss of supply to all the hydraulic
machines on the circuit.
Accumulator Sizing
The choice of the size of an accumulator is based on the system requirement. Initially,
therefore, it is essential to determine the exact reason for including an accumulator, and
what the range of its use will be. Basically the idea is to be able to supply a quantity of fluid,
within a certain pressure range, over a given time. The accumulator chosen will not only
need to be of a certain capacity but the precharge pressure will have to be known. Before
looking at some examples, it is important to appreciate the basic equations associated with
gas-type accumulators.
The nomenclature used in these equations is as follows:
Po = Gas pre-charge pressure [MPa]
PI = Minimum hydraulic fluid operating pressure [MPa]
P2 =
Maximum hydraulic fluid operating pressure [MPa]
Vo = Rated volume for the accumulator [L]
VI = Gas volume at PI [L]
V2 =
Gas volume at P2 [L]
n = Index for gas expansion
(Strictly speaking the ratio of the specific heats of the gas,
i.e. nitrogen - isothermal n = 1 and adiabatic n = 1.4)
For accumulators of the bladder type, the pre-charge pressure must be just below the
minimum hydraulic pressure, in order to avoid the bladder coming into contact with the
poppet valve. It is therefore customary to take
Po = 0.9 PI
In addition there must be a limitation on the maximum hydraulic pressure in order to
avoid permanently damaging the properties of the fabric of the bladder or diaphragm. Thus
P2 :54 Po for bladder-type accumulators
P2 :5 8 Po for diaphragm-type accumulators
The volume of hydraulic fluid which may be accumulated in the accumulator is the
difference between the maximum and minimum gas volumes. The exact value of these
volumes depends on the manner in which the accumulator is used. If the process is very
slow, such as with leakage compensation and the maintenance of a constant pressure in
a system, then the ratios are isothermal and
Po Vo = PI VI = P2 V2
If, however, the process is rapidly changing, which is the more usual operation, then the
gas changes temperature and the expansion is closer to adiabatic, and
The actual accumulator used would need to be some 11/2 to 3 times greater than the
calculated rated volume because of the restrictions mentioned earlier.
Figures 2C.l.12 and 2C.l.13 enable a quick assessment to be made from a know ledge
of the working pressure range and the oil volume; these charts give the values for either
isothermal or adiabatic operation. To use the charts, decide the range of pressure over
which the accumulator is to operate (PI to P2) and determine the precharge pressure (Po),
read-off the accumulator model which provides the appropriate volume. Take, for
example, the need to make up fluid loss during one part of a machine operation - say 5 L
over 3 seconds within a range of 10 MPa to 15 MPa, and a precharge pressure of 7 MPa
is acceptable - because of the rapid time this can be taken as an adiabatic operation and
Figure 2C.l.13 indicates that a nominal accumulator size of 36 L is suitable.
Where the pressures are above 20 MPa slight correction will be needed due to the
nitrogen gas used deviating from an ideal gas; the correction increases with increase in
pressure and may reach as low as 0.7 at 40 MPa (i.e. multiply the ideal volume by the
correction factor to obtain the real volume).
Where the storage in the accumulator has been rapid there is a rise in temperature; this
means that when the temperature drops to ambient, there wiII also be a drop in pressure.
Compression ratios are generally in the range 1.5: 1to 3: 1, depending on the application,
with 2: 1a typical average figure. This may be further modified, and the pressure difference
over the working range reduced, by coupling the accumulator to an auxiliary gas bottle.
The choice of size and type of piston accumulator is largely dictated by the particular
application. Thus relatively large capacities are required to cope with continuous opera-
tion and high demand. A much smaller size could be used where the accumulator has only
to supply peak demand or is worked only intermittently, or is mainly employed as a shock
absorber. Where the accumulator is employed solely as a source of emergency power the
size can be calculated on the flow demand.
For continuous operation with piston type assemblies the pre-charge pressure (Po)
should ideally be equal to the lower or cut-in value of the system working pressure (PI) as
this will gi ve the greatest swept volume over the working pressure range and thus mini mise
the number of pressure cycles. For intermittent use, or where the accumulator is used as
a source of emergency power, lower inflation pressures and consequently highercompres-
sion ratios can be used.
Other Accumulator Duties

Damping of pump pressure surges and ripple
Pressure surges caused by pump pulsations, or if the circuit is suddenly closed, are
common causes of shock wave generation in hydraulic systems. The severity of the
pressure pulsations and their frequency depends on the type of pump and its speed. All
positive-displacement pumps generate pulsating flow. Piston pumps are the worst in this
respect, and the fact that they are the type most used for generating higher pressures
aggravates the problem. The use of several piston pumps discharging simultaneously into
a pipeline may create very severe pressure surges liable to cause damage or failure due to
ACCUMULATORS 231
resonance. Pulsations and pressure surges such as these can be minimised or eliminated
mainly by the use of a bladder type accumulator whose absence of inertia or friction
permits the extremely rapid response essential for effective pulsation damping. [See also
the chapter on Vibration and noise.]
Pressure holding and leakage compensation

In a closed system where pressure must be held against the work by a holding ram for long
periods while further duties in the operating cycle call for pump capacity, the use of an
accumulator to replenish oil lost through leakage is advantageous.
The accumulator in a blocked circuit eliminates the problems of holding-pressure
variations created by the varying demands of branch circuits on the pump in open-centre
systems. In addition, system leakages, which are normally present or which develop over
a period of time, are automatically taken care of.
When lengthy holding times are required, two or more hydraulically-operated presses
can be run economically with the use of accumulators. External or internal leakage through
ram packings, valves or seals, results in piston creep and variation of the load on the work.
The accumulator compensates for such leakage, maintaining the correct loading for the
required period of time. Providing each press and accumulator is isolated during the
holding cycle, the system pump is free to meet the volumetric requirements of the other
presses.
Volume compensation
Accumulators are now making possible many uses of hydraulic mechanisms which
hitherto have not been feasible. This is especially true where the danger of increased
pressure due to thermal expansion of the fluid in closed systems would cause rupture of
the lines. The installation of an accumulator, pre-charged to the normal working pressure
in the line, readily takes up the expanded volume and, what is more important, feeds it back
into the line as thermal contraction takes place.
Hydraulic push-pull control mechanisms have been greatly complicated and limited in
use due to thermal expansion problems. The application of accumulators with a high
precharge simplifies them and extends their use.
Gas-loaded spring
The installation of an accumulator in a rigid hydraulic system introduces hydro-pneumatic
springing which can be used to advantage in many applications.
For instance, steel mill rolls or sugar mill cylinders (Figure 2C. 1.16) are required to
exert a constant pressure as material passes between them. If foreign matter or over-size
material is introduced the rolls must move apart to prevent damage and automatically
resume their normal positions at the required pressure. This springing action is accom-
plished by an accumulator, or a series of accumulators if necessary, of sufficient capacity
to absorb and release displaced fluid at almost constant pressure.
In all springing applications, the rapid action of bladder type accumulators due to lack
of inertia, friction and 'stiction' is advantageous, particularly where movements are small
and, even more important, where pressures are low.
Control
Two (or more) accumulators in a circuit can be used to provide synchronisation of
movement of hydraulic cylinders, constant velocity or constant pressure operation, as
required.
Holding devices
An accumulator is a convenient source of pressure to operate a holding device at constant
pressure, regardless of demand from other sections of the system, or for maintaining a high
working pressure on a workpiece during a long standby period. A particular advantage is
that this is accomplished without power being absorbed or heat generated, as would be the
case were the 'hold' maintained by a continuously running pump.
Dual-pressure circuits
An accumulator is a ready source of high pressure or increased capacity on a dual-pressure
circuit and in such cases may be charged by a separate pump. Thus a large-volume low-
pressure system demand could be met by a low-pressure pump (with its own accumulator,
if necessary) and the second high-pressure service met by an accumulator charged by a
small, high-pressure pump.
Emergency power, or energy storage

An accumulator can be used as an emergency source of power in the event of failure of
the primary power source (for example, failure of the pump driver). They are widely
employed in this way in aircraft systems for maintaining operation of emergency services
such as undercarriage operation, and also on large road vehicles utilising hydraulic brakes
and steering systems, but they also have industrial applications. The main requirement is
ACCUMULATORS 233
that the accumulator should be of a suitable size to meet the emergency demand. Reversion
to emergency operation can be fully automatic or manually selected, as required.
Safety Design Codes
Most countries have stringent requirements for gas loaded accumulator manufacture, but
these are not necessarily applicable to all such vessels. The chapter on Legislation and
safety has further information. A number of standards and guidelines both international
and national are given in the chapter on Standards.
The Pressure Systems and Transportable Gas Containers Regulations 1989 have a
number of highly important demands on accumulator manufacture and use. The key
features may be briefly summarised as follows:
i) Safe operating limits. Checks are necessary to ensure that the maximum system
pressure does not exceed the maximum working pressure of the accumulator and
that all safety devices are correctly set and in operation.
ii) Relevant Records. All certificates of conformity, test certificates, material certifi-
cates and maintenance records must be retained for inspection.
iii) Correct marking. The manufacturers need to mark the shell or pressure envelope
with their name, unit number, date of manufacture, design standard and, very
importantly, the maximum working and test pressures.
iv) Maintenance procedure. Particular procedure is required varying on the product of
system pressure (P - MPa) and accumulator volume (V -litres). The changeover
point is at PV = 25, and where the PV is greater than 25 a written scheme of
examination is necessary outlining the parts of the pressure system subject to
examination and the type and frequency of examination ..
The BFPA P54 gives a much greater explanation of the regulations. BS 7201 is also
applicable.
It should be pointed out that whilst every accumulator 'shell' is tested to full test
pressure (equivalent to 11/2 times working pressure) before despatch, the complete unit is
not tested in every case.
Accumulator-type devices
Special designs of gas-loaded hydraulic accumulators may be used to perform specific
functions such as pulsation dampers, liquid borne noise silencers, line surge alleviators,
pump controllers, etc. They are built and sized along the lines already described, but their
purpose is solely for the specialised application.
There is yet another area in which the original accumulator was used but where
specialist manufacturers now find it necessary to supply purpose built products. These
products are shock absorbers for liquids and they will be described in greater detail in the
next chapter. However, the surge damping accumulator is described below.
Surge damping accumulator

An accumulator has the inherent property of eliminating pulsations of a frequency greater
than the cut-off frequency of the device. The cut-off frequency of an accumulator can be
calculated from the geometry, fluid density and pressure.
235
SHOCK ABSORBERS
SHOCK ABSORPTION relates to both the absorption of shock within the hydraulic
system by the use of an accumulator device connected to the fluid system, and to a specific
hydraulic component designed to be attached to some other system or equipment to absorb
energy within it. The latter, hydraulic shock absorber, may even be used for pneumatic
systems where impact forces and vibration are prime causes of fatigue failure if the
damping is insufficient.
The absorption of hydraulic fluid shock

The products suitable for shock absorbers for liquid systems fall into three basic
categories:
i) Surge Collectors - These devices allow large amounts of fluid flow kinetic energy
within a pipe to be decelerated over a substantial period of time by the increasing
resistance caused by the compression of a pre-charge gas. The units are generally
designed to have very large entry orifices but restriction on outlet, preventing the
volume accepted surging back into the system.
ii) Shock Preventers - These are specifically preventers or 'alleviators'. These are
installed in a system so that their capacitance and impedance prevent a shock from
occurring by allowing a deceleration time for the fluid, but in nearly every case
these units have entries which are substantially as great as the diameter of the vessel
itself.
iii) Shock Absorbers - These units are capable of absorbing a shock, i.e. absorbing a
standing wave which has already been created and is travelling down a pipe. They
SHOCK ABSORBERS 237
Industrial hydraulic shock absorbers

The industrial shock absorber has the requirement of totally absorbing the kinetic energy
of a moving component. Kinetic energy can be dissipated by transferring it into heat
through friction - as with the automotive brake - or by the deformation of a substance -
as with a rubber bung, a spring, or passing a fluid or air through an orifice.
Whilst any of the above suggestions may be used for the simpler or lower energy
applications, hydraulic shock absorbers use a relatively incompressible fluid and hence
can be designed to generate very high reaction forces that can absorb extremely high
energy levels. In addition, the maximum efficiency and smoothest damping will occur if
the deceleration is linear, and this the hydraulic shock absorber can supply. This means that
maximum absorption should occur at the initial stage (when the kinetic energy is highest)
and then gradually decrease as the energy is absorbed. Such absorption is the complete
opposite to that experienced in a spring, where the absorption increases, rather than
decreases, as the spring is compressed.
In order to achieve true linear deceleration in an hydraulic damper (with an orifice) the
size of the orifice would need to be gradually decreased as the velocity of the object
reduces. One method by which this is achieved, particularly for small and medium size
shocks, is by using many orifices initially in a cylinder, and then, as a piston moves into
the cylinder the orifices are gradually covered. Figure 2C.2.3 shows the cross section of
one such orifice metering device. (See 'cushioning' in Hydraulic cylinders.)
I - Return spring 2 - Piston rod 3 - External adjusting ring

4 - Piston head 5 - Metering orifices 6 - Hydraulic fluid
7 - Elastomer foam
FIGURE 2C.2.3 - A linear deceleration piston shock absorber [Ace Controls International]
The deceleration of a mass may also be effected in two stages. For instance, it may be
required to have a soft contact with the minimum of initial reaction for low impact forces,
but when there is a change to a heavier shock pattern, the absorber needs to automatically
compensate and apply the necessary higher reaction. Figure 2C.2.4 shows examples of
shock absorbers which provide both soft contact and self compensation in one package.
There also two other methods which can provide excellent metering - these are annulus
metering and pin metering. The annulus metering is like a dashpot but with a tapered bore
to the cylinder to vary the restriction as the piston moves in; it would normally only be used
on very small applications. The pin metering idea uses a hollow rod with a tapered needle
which projects into the hollow from the base; this is able to cope with very large shocks.
FIGURE 2C.2.4 - Dual deceleration control shock absorbers [Ace Controls International]
The range of shock absorbers is considerable, from tiny units as small as a few
millimetres up to units able to dissipate lOO,OOONm of energy per cycle. The choice of
the size of shock absorber must cover not only the energy of the moving object but also
that of the propellant behind it; it is normal to oversize in order to cope with at least a further
25 % of energy dissipation. Figure 2C.2.5 shows the example of a miniature non-
adjustable hydraulic shock absorber with a maximum shock force of 700 N (plus
propelling force of 89 N), with a velocity up to 2.54 mls
FIGURE 2C.2.5 - A miniature non-adjustable hydraulic shock absorber

[International Motion and Control, formerly Enidine Group]
239
COOLERS
REA T GENERA nON within a hydraulic system is always present. If the system is only
run intermittently, or at a very low power level, additional cooling will not be necessary.
Certain applications also have a high natural air or liquid flow over the hydraulic
components, and again, there will be no need of a cooler. However, in the great majority
of applications the heat generated needs to be removed if the temperatures are to be kept
to acceptable levels.
Temperature levels become serious if they present a human hazard, such as bums due
to touching a surface above 80C, a fire hazard, oil evaporation, excessive drop in oil
viscosity, seal damage, or breakdown in oil properties. Most hydraulic systems should
operate at a maximum temperature of 65C.
Sources of heating
The principal sources of heat generated in hydraulic systems are the power losses
associated with pumps, relief and control valves Heat is also generated by friction losses
in the pipework and other components so that self-heating is proportional both to flow rate
and pressure drop. The whole system may also be subject to additional heating (or cooling)
from external sources.
In practice only heat generated by the pump, pressure drop through valves and hydraulic
motor power losses are likely to be significant. Heat losses generated by line friction and
cylinders are usually dissipated from the surface of those components. In most small
circuits heat losses are balanced at an acceptable temperature level by natural heat transfer
from the fluid to the pipes, actuator and reservoir, and subsequent dissipation to the air.
In some cases the source of heating can be a localised one (such as with bearings); this
is easily overlooked. An overall guideline is: the difference between the power input to a
hydraulic circuit and the mechanical power output developed by the actuators represents
the energy transformed into heat. Where this difference is less than about 1.5 kW, and
provided the whole system is reasonably well ventilated or exposed to open air, overheat-
ing troubles are not likely to occur. For larger differences, or where it is necessary to hold
working temperatures to specific levels to ensure consistent operation (e.g. in the case of
machine tool actuators), the cooling requirements may need analysing in detail. Perhaps
a simpler guideline relates to the motor size used - if this is 5 kW or above, then cooling
should be included.
Cooling
In most systems some degree of cooling is automatically provided by the reservoir, and
also by heat radiated from pipework (hoses with a larger surface area, and black, tend to
dissipate more than pipes). Where additional cooling is required, a heat exchanger (oil
cooler) can be incorporated in the system.
Basically there are two types of cooler. There is the older more conventional water-
cooled type requiring a separate water supply, or the air-cooled type requiring a fan driven
by an electric motor or some other rotary power source such as an hydraulic or air motor.
The cooler is generally placed in the return line to the reservoir on the low pressure side
of the relief valve.
Reservoir cooling
The cooling provided by a reservoir is primarily dependent on its surface area and
emissivity (see Table 2C.3.I), its location and the difference between the fluid temperature
and the ambient temperature. The effective coefficient of heat transfer is not just that of
the reservoir walls. It is modified by the presence of a boundary layer of fluid and the
degree of turbulence present inside the reservoir. It will also be modified by any airflow
past the exterior walls of the reservoir.
Whilst it is good practice to design the reservoir to contain two to three times the volume
of fluid being circulated per minute, there are applications, particularly in the mobile field,
where there is insufficient room or the weight penalty is too great. In such cases an
effective way of increasing the surface area of a reservoir without increasing its internal
volume is to add fins. Increasing the proportion of the vertical walls to the horizontal will
also improve the situation because the air flow is normally considerably more in the
vertical plane. The cooling performance of a reservoir will also be improved by good
ventilation.
To ensure adequate dissipation in this manner the reservoir should be free standing with
the sides and, possibly, the base, fully exposed and preferably in a position where it
receives ventilation by air draught. Actuators should be similarly placed, to avoid
localised overheating of the fluid at these points.
Water-cooled heat exchangers

The efficiency of a cooler depends on the difference between the oil temperature and the
medium used to provide the cooling (air or water). Thus, for the same thermal perform-
ance, the cooling provided by water always results in a smaller size unit than air-cooling
COOLERS 241
because heat extraction is more efficient. Air coolers need larger fin areas and greater
turbulence. The possible disadvantages of water-cooling are the increase in water costs
and the fact that it may be inconvenient to pipe water to the installation.
Two types of water cooler are generally available, one is the bundled tube type (tubular-
type) and the other is the plate-type. The plate-type will be described later in this section.
In the case of the tubular-type, the coolant, i.e. the water, is passed through the central tubes
of the cooler and the oil flows around the tubes but in the counter-direction. Figure 2C.3.1
shows a typical nested arrangement of copper or copper alloy tubes. The design of the
tubular water-cooler varies in the use of different materials, arrangement of tubes (or spiral
wound tubing) and addition of fins; a greater heat transfer is achieved by the use of fins,
as shown in Figure 2C.3.2. A small number of baffles may also be present which force the
oil to move from one side of the bundle of tubes to other several times whilst passing
through the cooler.
High pulse or peak pressures can exist in the system lines, and although relief valves can
be fitted within the cooler, they may not always be able to react faster enough to counteract
these. However, if external relief valves are not fitted, these internal relief valves can help
to prevent excessive blockage to the flow at cold-start or with deterioration over a
considerable time.
In big plants it is generally best to provide an auxiliary circuit for feeding the cooler (and
possibly the filters). The auxiliary circuit can operate independently from the main circuit.
In choosing a water-cooled cooler, the starting point is a knowledge of the heat which
is needed to be dissipated from the hydraulic circuit. This can be determined from a test
or calculation without a cooler fitted and measuring the rise in temperature which occurs,
i.e.
Examples of a range of coolers available from one source is shown in Figure 2C.3.3. The
second graph shows the pressure drops experienced in the cooler versus the flow.
Plate-type coolers consist of maybe 200 pattern-embossed plates in stainless steel so
arranged that a lattice of contact points are formed between the plates, with water and oil
passing between alternate plates. The points are brazed together to form a compact
pressure resistant cooler package with virtually all the material utili sed for heat transfer.
In this way several advantages accrue over other types:
Turbulent flow is maintained even at extremely low flow rates and velocities
Excellent distribution of flows minimises poor flow regions
Air blast coolers
Air blast coolers operate by forcing 'cool' air through a radiator through which oil is
passing. Although copper or aluminium radiators provide the best heat transfer, steel is
also used as it is more robust and able to cope with higher pressures. Although the air blast
type may be more bulky than the water type, it offers a simpler system because an air-
cooled heat exchanger normally requires only two pipe connections to fit into the circuit
concerned. It can also be designed to provide efficient cooling to within soe of the
surrounding atmosphere, or even less if required, but a higher efficiency occurs with
greater differences. Air-cooled heat exchangers do, however, require electrical, mechani-
calor hydraulic power to drive the fan, but can very often be made self-contained on the
hydraulic power pack. Figure 2C.3.5 shows the members of a typical air-blast cooler.
FIGURE 2C.3.5 - The components of an air-blast cooler [Fawcett Christie / Oiltech]
The cooling air is usually provided by a rotating multi-bladed fan. Fans, however, can
be noisy and in order to reduce the noise level, it is normal to use a low speed fan or 2-speed
fan which only operates at high speed when extra cooling is required. An alternative
approach, to improve the cooling capacity, is to use two slow speed fans, one either side
of the heat exchanger; this also has an improved efficiency, and one fan can be switched
off if less cooling is required (Figure 2C.3.6). Because the oil wetted surface is much
smaller than the air side surface, it is important to include in the oil side design turbulators
(or finning).
The power source for the fan is usually provided by an electric motor for stationary
FIGURE 2C.3.6 - The low noise high efficiency twin fan air-blast cooler [Occo Coolers]
COOLERS 247
Service comparisons
Air blast cooling has the advantage that the only likely malfunction - the fan not running
- is immediately apparent and caIls for remedial attention. It also aIlows quick visual
checking for correct functioning and the state of any fouling. By contrast, particularly in
hard water areas, it may be necessary to strip and de-scale the tubes of a water-cooler every
three months or so.
Considerations of maintaining oil flow are the same for both forms of cooling, although
many larger systems and power packs are now fitting fan blown coolers incorporating a
circulating pump, thus making the cooling function 'off-line' , and independent of the main
system flow fluctuations. This arrangement has the definite advantage of removing the
cooling elements from the rapid pressure peaks and 'hammering', which is often found in
the return line of large complex systems or servo systems.
If the oil passage waIls suffer from fatigue or rupture, the air blast cooler has
maintenance advantage. The leak can be seen and causes no lasting damage, since oil
escapes to atmosphere. The consequences of an oil/water rupture can be much more
serious, especiaIly if water reaches the pump intake. The resulting damage and rectifica-
tion costs may even involve a complete system strip and rebuild.
249
HEATERS
THE MAJORITY of hydraulic systems see heat exchangers solely as coolers. These are
discussed in the chapter on Coolers. However, there are a number of reasons why it may
be important to heat or preheat systems to improve the efficiency of the operation.
The most common reason is to reduce the loads associated with high viscosities in cold
climates. (This also applies to fuels and oils as well as the hydraulic fluids.) There is the
initial 'cold-start' and there is also the continuing or intermittent use where the system does
not have adequate time to reach an acceptable working temperature. The start condition
requires preheating because there is no initial flow. The continuing operation condition
may make use of the flow with suitable heaters through which the oil passes.
Effects of viscosity
Control of fluid temperature is highly desirable to avoid large viscosity changes when
starting from cold and running up to working temperature. A knowledge of the precise
working viscosity of the fluid means that the pump receives adequate lubrication and
operates at high volumetric efficiency. Changes in oil viscosity can also seriously affect
machine efficiency and performance. The higher the system pressure the more heating the
fluid is likely to receive once the system is working, and the greater will be the change in
viscosity. The volume of fluid in the system can also be a critical factor, the lower the
volume the more heat received per unit volume for a given amount of work (a proportion
of which is inevitably converted into heat). Marked changes in ambient temperature can
also affect the actual working temperature of the fluid; this is a problem more usually
confined to aircraft hydraulics than industrial hydraulics.
Preheaters
Most preheaters are electrical. They may consist of a surface heating tape/jacket/blanket
or an immersion heater, powered by mains electricity or by internal batteries (particularly
in the case of mobile machinery in cold climates).
Immersion heaters are less prone to accidental damage and there is no direct heat loss
to the surrounding atmosphere because they are sunk into the fluid. For oil, the heater
elements are al ways much larger than for water due to the limit of 12 W/cm2 on the heating
surface flux density.
Surface heating has the advantage that it can be retrofitted and causes no contamination
or disturbance to the oil. Also, unlike immersion heaters, there is no local high temperature
which may break down the oil properties.
Other types of heater may be used such as electric fans, induction heaters, infra-red and,
in rare cases, microwave. Figure 2CA.I shows a range of possible heating methods
including air heaters and a full trace heating system suitable for a reservoir.
FIGURE 2C.4.1- A range of heating devices suitable within and about systems [HaweD].
Oil heat exchangers

The conventional tubular or plate oil 'cooler' may be used in certain circumstances to
provide heating. The only difference from the cooler application is that the supply of water,
instead of being cool, is at a temperature higher than the oil. One particular type which is
used more exclusively for heating rather than cooling is the tube-in-tube type shown in
Figure 2CA.2; this may be low cost but provides good thermal efficiency because of the
counter flow technique with no turbulence.
251
RESERVOIRS AND TANKS
THE DESCRIPTIONS reservoir and tank are used synonymously to describe the vessel
used in a hydraulic circuit to store the working fluid for the system. However, the volume
of fluid in the tank is not likely to be constant. Fluid volume changes are produced by
thermal expansion/contraction of the fluid and system components, and also by unequal
area cylinders and other fluid storage devices like accumulators. The tank, however, has
greater potential than just fluid storage and if correctly designed should perform other
functions including: simplify maintenance actions; provide access for the filling or
topping-up the fluid in the system; remove entrained air from the fluid; remove solid
contaminants in the fluid by settlement; remove water (for non water-based fluids) and
provide some cooling effect. These aspects are discussed in more detail below.
Reservoirs can be divided broadly into two types - vented or sealed. The former act
purely as storage tanks and are open to atmospheric pressure. Sealed tanks are sealed with
respect to atmospheric air and may be either pressurised or unpressurised. In the former
case they may also have an ability to act as accumulators in addition to the 'tank' duties.
They can combine the duties of both reservoir and accumulator in a single unit where
weight saving is important and 'accumulator performance' demand is relatively light. A
vented type would normally be used for most applications.
Basic configuration and design

Reservoir design is far from standardised, and even size (volume) is based on largely
arbitrary lines. Designs tend to follow accepted practice for an application area, and tanks
may be incorporated as part of a machine structure, completely free standing or be pump-
mounted. Mounting the pump on the reservoir is quite common and in some cases the
pump may be submerged in the fluid. A selection of designs are shown in Figure 2.C.5.1.
Medium sized tanks tend to be box shaped, which in industrial systems are generally flatter
rather than higher. This gives less change in the oil level as the fluid volume changes, and
more importantly gives a larger top surface area to aid air release and base area for
contaminant settling. It is less appropriate in mobile applications where a higher sided tank
is more appropriate to reduce fluid movement and "sloshing" about. It also gives greater
heat exchange from the vertical surface area in contact with the fluid. Very many small
systems (below say 40 L) have cylindrical tanks with a vertical axis providing a complete
"power pack" installation and are again widely used in mobile applications. Very large
systems with capacities in the 1000' s of litres, again frequently have cy lindrical tanks but
with a horizontal axis.
Recommendations on tank sizing are very broad. Guide-lines are usually based on the
maximum flow taken from the tank by the pump or pumps. If this is quoted in Umin then
for general industrial applications the fluid volume (L) of the tank would be in the range
3 to S times this flow in one minute. In mobile applications this would reduce to the value
of the flow in one minute (with more frequent service intervals to change the fluid).
Somewhere between 10 and 20% of this volume should also be included as air space above
the fluid. These are approximate values and working practice in different industry sectors
may provide more definitive sizing. A check should be made on the volume of fluid taken
by long stroke or large diameter unequal area cylinders. In conjunction with the operating
cycle this should give the extreme possibilities for volume interchange between tank and
system. Charging or discharging of accumulators may also need to be considered.
A reservoir design which is based on the American HC (Joint Industry Conference)
recommendations for vented reservoirs in industrial applications is shown in Figure
2C.S.2. This provides a useful indication of important features and additional components
to be included but there is obviously considerable freedom with the layout. It should
perhaps, be noted that it is generally preferable to design the reservoir as a separate rather
than an integral tank since this will improve both accessibility and the ability of the
reservoir to act as a cooling radiator.
RESERVOIRS AND TANKS 253
The majority of industrial type, vented reservoirs are made of mild steel with welded
joints. Material thickness is typically 3 mm. However, if the pump is mounted on the
reservoir then considerable care should be taken to provide adequate support, and isolating
mountings should be used for the pump to prevent pump vibration being transferred to the
structure. Pump location below the tank, or alongside, in the L shape design, see Figure
2C.S.I, is preferable. These increase the fluid pressure head at pump inlet and reduce the
mechanical link path for vibration between tank and pump. Smaller tanks may be cast from
aluminium alloy or formed from pressed steel.
After fabrication the tank interior should be thoroughly cleaned, using shot blasting if
appropriate to remove scale and surface corrosion. Both the inside and outside of the tank
should then be painted with a suitable resistant finish. Special paints are available but the
choice of a suitable paint may be strictly limited where the reservoir is for use with
synthetic fluids, and one finished with an oil-resistant paint would be quite unsuitable for
such services.
A primary requirement of any reservoir design is that it should be easily cleaned both
initially and as part of planned servicing. A drain plug must be fitted to the bottom for
draining the reservoir completely, this should be at the lowest point in the base. A concave
base is intended for the tank in Figure 2C.S.2 but an angled base plate as in Figure 2C.S.3
is an equally appropriate method in the box design. It should also be remembered that the
tank base should be raised sufficiently above floor level to facilitate draining. Removable
end covers, top or side plates are also desirable for access to the interior to clean out sludge
and contaminant. Regular draining from the tank base is good maintenance policy, since
contaminant will settle with water from condensation and may form a sludge at the lowest
point in the tank. Preventing any accumulation from re-circulating is obviously beneficial
to system reliability.
There is potential for the use of reservoirs in providing cooling but it can be quite
variable. The heat transfer is complex involving transfer from the hot fluid to the bulk of
the fluid, the bulk to the contacting tank surfaces, and from the external surface to the
surroundings (convection and radiation). As an example, a tank of about 100 L may
dissipate from 0.5 to 2 kW when at 40 C above ambient, depending on these conditions.
The hydraulic power input could be about 3.5 kW based on a 20 Llmin pump at 10 MPa
for such a system. Design choices can help by maximising the contact area between fluid
and wall but installation is also significant in ensuring good air circulation and encourag-
ing radiation (the principal mechanism). The topic of heat rejection and cooling is
discussed more fully in the chapter on Coolers.
Instrumentation
A fluid level indicator should be part of every reservoir design. This may simply take the
form of a transparent window to act as a fluid level sight or an external sight tube. More
sophisticated level indicators may be fitted to some reservoirs, or a simple dipstick marked
with high and low levels. These levels would need to take account of fluid interchange
between system and reservoir. The inclusion of some temperature indicator is also
appropriate and may provide useful diagnostic information.
Filling, breathers and venting

All types of reservoir require a filler and a vent to prevent pressure build-up during filling.
A breather is also required to allow air in and out of the tank as the fluid volume changes.
In some designs this may be replaced by a single filler/breather or in the case of a sealed
reservoir, by incorporating a suitable vent opening in the filler hole. The filler opening
itself should be protected with an intake filter, commonly fitted inside the reservoir.
Access must then be provided so that filter strainers or elements can be removed for
cleaning.
The air breather must also contain a filter to prevent induction of atmospheric dust
particles as contaminant. The size of the breather must be sufficient to accommodate all
air flows whilst maintaining atmospheric pressure in the tank. It must allow for the
discharge of air from the tank at the maximum rate of fluid level rise, and induction of air
at maximum fluid level fall. It should then also accommodate the less rapid effects of
escape of entrained air which is released from the fluid surface in the tank and those due
to thermal expansion and contraction.
Separation of the filling and breathing functions offers the best solution in reasonable
sized systems (say 100 L upwards) or where top-up would be frequent. The best form of
filler is through a quick release coupling to a pipe discharging below the fluid surface. This
eliminates spillage problems and allows filling to take place with either a hand or small
electric pump direct from the supply drum through a system quality filter. The fluid
entering the tank can then be considered as clean. It is always best to fill or re-fill a system
using this method.
Line connections
Line connections to the reservoir are particularly important. These will include return
flow and pump suction lines, and possibly also some drain lines. Used correctly they can
aid air release, encourage contaminant separation and prevent contaminants re-entering
the system in typical industrial tanks.
Return pipes must always discharge below the lowest possible fluid level in order to
prevent aeration but there is some disagreement on the termination of this pipe. Opinions
are divided on whether contaminant should be allowed to settle on the base of a tank,
although it will always tend do so. Some return pipes are aimed to scour the tank base with
a jet to encourage the contaminant back into the system to be removed by the filter but this
will only ever be partially successful. It seems a good principle that contaminant which
is no longer carried in the fluid is a better objective, and hence it is preferable not to disturb
the settled layer. The danger in this may be that some extreme pattern of operation causes
the sediment to be disturbed by unusual flow and swept back to the system in larger
quantities than with the jet scouring policy. These remarks are confined to stationary tanks
and not relevant for mobile applications where both smaller volume and movement inhibit
sedimentation.
A good design for return line termination with a diffuser is shown in Figure 2C.5.4. flow
visualisation tests have shown that this encourages a well distributed low velocity flow
around the complete circumference towards the fluid surface. This leaves the tank base
undisturbed and any air is carried to the surface. One fundamental error is the frequent
recommendation to cut an angle on a return pipe with the idea that it has an effect on the
direction of the fluid jet. A simple consideration of fluid momentum will indicate the
fallacy.
The suction pipe is also located as low as possible to prevent drawing air in with the
fluid, but not so low that it can pick up contaminants which may have settled. Protection
in this respect can be given by fitting the suction pipe with a strainer. This may be mounted
with a vertical axis and solid base but if mounted horizontally and close to the tank bottom,
the lower third of the element closest to the bottom is best shielded to reduce contaminant
pick-up. It is preferable that provision should be made to remove the suction line strainer
so that it can be attended to without having to drain the tank.
Baffles
Baffles are used to prevent fluid just returned to the tank from passing directly back to the
pump inlet. A longer transit path is considered beneficial for a number of reasons. It
encourages better air and contaminant separation, better heat conduction from the fluid
and better mixing with the bulk fluid. This is usually obtained by separating the inlet and
return lines by as long a flow path as possible. A single baffle design as shown in Figure
2C.5.5 is common. This has a weir effect to encourage flow near the surface but it is always
necessary to leave a flow path near the base for draining, as shown. The use of an additional
longitudinal baffle would allow the flow path to be increased by 50%.
Baffles may also be necessary to minimise surging of the fluid when the reservoir is used
with mobile equipment. In this case the baffles would normally be arranged at right angles
to the direction of the 'surge' motion, virtually dividing the tank into a number of smaller
volumes with small inter-flow. An alternative is to arrange the baffles in 'maze'
configuration so that end-to-end (or side-to-side) flow is tortuous and thus heavily damped
during surge motions.
The use of angled baffles may also be helpful in removing entrained air bubbles. In this
case the baffle normally takes the form of a fine mesh screen mounted at an angle (45 to
60 to horizontal) but with its highest point below the lowest fluid level. Such a screen can
be effective in constraining bubbles on the inlet side of the tank but it may also encourage
foaming. Wire mesh baffles are also effective in breaking up droplets in water-emulsion
type fluids and such baffles are not desirable in reservoirs where these fluids are used.
Sealed reservoirs
Sealed reservoirs are isolated from contact with atmospheric air and may be used for a
number of reasons.
(i) On mobile equipment, aircraft, etc., where movement could cause spillage from
a vented tank.
(ii) To exclude contamination when working in a heavily contaminated environment
beyond the normal ability of a breather/filter.
(iii) To provide an accumulator effect.
Sealed reservoirs are generally larger than vented reservoirs for the same fluid volume
because a sealed reservoir requires a fairly generous air volume in order to minimise
changes in pressure as the fluid level changes. Sealed reservoirs may work at atmospheric
pressure, or be sealed under elevated pressures oftheorderof70- 140 kPa (0.7 to 1.4 bar).
A relief valve must be fitted above fluid level to blow off excess pressure above the
nominal filling pressure and a vacuum relief valve may also be fitted to protect the
reservoir should the pressure fall too low. Considerable care must also be taken to avoid
overfilling, since this will reduce the air volume and produce wider changes of pressure
during working. The sealed type must, therefore, be considered as a more specialised
design.
With a pressurised sealed reservoir, direct pressurisation of the air above the fluid in a
simple tank can lead to a greater than normal volume of air being dissolved in the fluid and
subsequently released in some other part of the system. Nevertheless, such pressurised
tanks have proved suitable for a wide variety of services, including aircraft installations.
An alternative solution, now generally preferred, is to fit a sealed reservoir with a
'breathing bag' to accommodate changes in fluid volume. The design of the reservoir is
then essentially similar to that of a bladder-type accumulator, which could be used
although nothing like as strong a construction is required. The use of such a design with
physical separation of air and fluid also eliminates the question of super-saturation of the
fluid under pressurisation.
Aircraft reservoirs
Reservoirs for aircraft systems pose special problems, principally being the loss of
atmospheric pressure with increasing altitude. The pump cannot produce a suction
pressure greater than the barometric pressure in an open tank, and thus a sealed tank is
virtually essential to provide a net positive pressure at pump inlet. This is exacerbated
because allowance must be made for fluid temperature to be reduced considerably at
altitude. A pressurised reservoir can thus become essential to avoid pump cavitation. Also
weight considerations influence the tank material and the volume of fluid is minimised.
In the case of many aircraft, feed must be provided by the reservoir at a wide range of
attitudes of the aircraft. One alternati ve, both for aircraft and industrial hydraulics, is the
use of a spherical tank with suction and return pipes near the centre, and sufficient capacity
to ensure that the reservoir is always more than half full of fluid under all conditions in
order to give uniform coverage of the pipe entry. Another solution, sometimes used in
aerobatic aircraft, is the use of a 'klunk' tank where the delivery pipe enters roughly at the
middle of the tank and terminates in a flexible pipe with a weighted end. Gravitational and/
or acceleration forces displacing the oil content of the tank will similarly displace the
weighted end ofthe pipe, so that the end always remains in the fluid as in Figure 2C.5.6.
A particular advantage of this type of tank is that it usually permits a smaller size of
pressurised reservoir to be used.
The same principle can be adopted for industrial machines which have extensive
motions. In this case the klunk tank principle can be used with both vented and pressurised
reservoirs.
259
TUBES & PIPEWORK CALCULATIONS
IN HYDRAULIC systems the term pipework is used to describe the interconnecting

system of tubes and fittings used to convey the fluid within a system. A tube can be
considered as a homogeneous, smooth bore, hollow line, either flexible or rigid, with small
bore dimensions in relation to its length. The word tube is most frequently used to describe
the rigid alternative and hose (described in a later chapter) to describe flexible reinforced
tube. The fittings are a means of connecting two tubes together or a tube to other system
components such as manifolds, pumps or filters. The term pipe is often considered to have
---a larger bore diameter than a tube and for some materials will have a higher bore surface
roughness than tube. The use of pipes in hydraulic systems is gradually decreasing but the
term pipework is still used quite commonly. Pipes and tubes are sometimes called
conductors. The reader should be aware that this terminology may differ between different
sectors of industry.
Selection of tube is based on three fundamental criteria:
(i) tube material;
(ii) tube bore diameter to carry the flow;
(iii) tube wall thickness to withstand the internal pressure.
The sections of this chapter deal with the relevant material aspects of rigid tube, aspects
of fluid flow in a conductor and tube strength. The material aspects of flexible tube are
described in the chapter on Hoses.
Tube (and pipe) for hydraulic use is almost always specified by its external diameter.
Some larger sizes are specified in terms of wall thickness and bore diameter, and may be
in Imperial dimensions. There is also an American convention of using dash sizes for tube
outside diameters, where each dash number represents an increment of 1/16 th inch. Thus
-8 (read as dash 8) would be 1/2 inch and -32 would be 2 inch.
Bore size tubing is, theoretically at least, the more useful in critical applications since
it enables the tube size to be chosen on the flow requirements, which are determined by
the actual bore size. In other words, it enables closer selection of an optimum hydraulic
size. It is also more realistic for rating high-pressure tubes due to the greater variation in
wall thickness possible. The equivalent hydraulic size of tubing based on outside diameter
can only be determined from a knowledge of the actual wall thickness used.
Standard sizes of tubing produced for hydraulic lines are normally produced in two wall
thicknesses, nominally rated as light (L) and heavy (S) gauge. Diameter to thickness ratios
range from 6: 1 to 10: 1 in the former case; and 4: 1 to 7: 1 in the latter, both increasing with
increasing diameter. Corresponding pressure ratings for light tubing range from about 42
MPa (420 bar) to 30 MPa (300 bar); and for heavy gauge tubing from 55 MPa (550 bar)
to 45MPa (450 bar).
For lower pressures, 'nominal' pressure ratings are sometimes adopted for pipes and
tubes, expressed in terms of a given maximum working pressure. This is usually
substantially below the theoretical value possible. Sometimes these tubes may be
produced in three different wall thicknesses, specified as light, normal and heavy gauge.
These ratings have no precise meaning, unless related to definite pressure ratings or
diameter to thickness ratios.
For very high pressure working, the small bore sizes necessarily associated with a low
diameter to thickness ratio normally call for the use of special high strength alloys or
composite construction in order to minimise wall thickness. The manufacture of such very
high pressure tubing is critical, since all flaws must be eliminated and it is inspected
internally, externally and through the section for cracks and other defects. An alternative
method of construction for thick-walled, very high pressure tubing is to fabricate the
section from two or more coaxial tubes in carbon steel and alloy steel, assembled as an
integral tube. Tubes of this type are capable of accommodating pressures up to 1400 MPa
(14000 bar).
Rigid tube materials

Choice of rigid tube materials is usually governed by
(i) the external environment of the application;
(ii) weight considerations in the application;
(iii) a balance between these two points and cost since the lightest, most corrosion
resistant materials are generally the most expensive.
Hence in the absence of special considerations low carbon steel of hydraulic quality is the
most common choice. Other materials commonly used are special brass alloys (principally
tungum) and stainless steel; less commonly used are copper, aluminium alloy, titanium
and plastics.
For use in hydraulic systems tubes are usually classified as precision tubes and their
quality is covered by a number of standards. For example the DIN standard 2391 covers
the dimensional and technical delivery aspects of steel tubing. The dimensional section
specifies preferred outside diameter sizes and wall thickness with their tolerances. The
delivery section includes the quality and grade of steel, supply conditions, length and
inspection documentation. Standard lengths, with different tolerance bands up to 7 mare
defined. Considerable care should be taken not to mix components and tubes from
different standards, and obviously metric and Imperial dimension systems.
Cast iron pipes

Cast iron pipes have a limited application for large bore hydraulics and in some water-
hydraulic systems. In the case of cast iron pipes rounded-off pressure ratings are normally
TUBES AND PIPEWORK CALCULAnONS 263
given, diameter to wall thickness ratios being selected to give a range of different pipe sizes
with the same working pressure rating. Such pipes are normally specified in terms of
nominal bore size, when actual bore size and actual overall diameter may differ appreci-
ably from nominal values.
Steel pipes
Steel pipes may still be used in some existing systems although they are less likely to be
used in new systems. The most useful standard covering steel pipes is the American
National Standards Institute pipe schedule given in Table 2C.6.1. The nominal size given
refers to the external diameter with the increased strength of the heavier schedules given
by increased wall thickness and hence reduced bore diameter. This system has arisen
because pipes are usually connected with an external thread, usually tapered, to their
mating part.
Steel tubes
Most tube for hydraulic use is manufactured as Cold Finished Seamless (CFS) sometimes
called cold drawn seamless. A chart showing the stages in a typical manufacturing process
is shown in Figure 2C.6.1.
A number of low carbon steel specifications are appropriate for hydraulic system use
varying in ultimate strength from around 400 MN/m2 to 600 MN/m2 Mild steel tube has
the advantage of being easily workable for bending or deformation in some coupling
operations but it is of course subject to corrosion. It is usually phosphated but may also be
chromated, zinc or even plastic finished. In most applications external paint protection is
recommended. Tube is typically available up to 42 mm diameter and an example of a
standard size range is given in Table 2C.6.2. Tube above about 50 mm is available in the
same welded, cold drawn form that is used for cylinder bodies (Drawn over Mandrel). This
is generally available to the American ASTM specification A513, but is not considered
suitable for bending.
Copper
Copper tubing is attractive for small bore, limited pressure services, because of the ease
of manipulation, as well as being resistant to general corrosion. It is not suitable for high
pressure lines, due to the limited strength of the material, and the fact that it is susceptible
to work-hardening and reduced fatigue resistance if subject to vibration. There have also
been some (rare) problems with mineral oil additive packs reacting unfavourably with
copper. Nevertheless, copper tubes may be rated for pressures up to 2lMPa (210 bar) in
standard production sizes. It is also necessary to bear in mind that the strength of copper
decreases with increasing temperature, so 65C is about a maximum for working at 7MPa
(70 bar) or above. The strength of copper is halved at a temperature of 200C.
An alternative, sometimes preferred to copper for higher pressure working, is alpha
brass, offering far better strength and fatigue properties.
Tungum
Tungum is brass based, comprising an aluminium-nickel-silicon-zinc-copper alloy (cop-
per >80% ) with an alpha type structure. It combines high tensile strength, good ductility
and ease of manipulation, freedom from cracking and brittle fracture, good fatigue
properties, and excellent corrosion resistance. It has a better strength to weight ratio than
mild steels but is more expensive on a weight for weight basis. It is, however, available
in a wider range of wall thickness than steel tube and for low pressure return lines may not
be more expensive overall. It is equally appropriate for high pressure lines, where high
working pressures and great reliability are required, standard production sizes are
available up to 70 MPa (700 bar).
Tungum has been widely used in aircraft applications because of its attractive combi-
nation of properties. Its excellent corrosion resistance, particularly in salt water or salt
spray, which could attack steel tubes, makes it equally well used in marine applications.
Because of its high tensile strength and corrosion resistance, tungum alloy can certainly
be regarded as an economic alternative to stainless steel and often to mild steel.
Aluminium
Aluminium alloy tubes have been used in aircraft hydraulic and servicing systems, and in
other applications where weight saving is important. The use of 61S-T6 alloy yields a
material strength almost directly comparable with that of low-carbon steel. In general,
however, aluminium is not regarded as suitable for pressure lines subject to vibration or
pulsating pressures because of the relatively poor fatigue characteristics of the material.
It may, however, be selected for low-pressure or return lines on a system where weight
saving is important, or as a cheaper alternative to copper.
Stainless steel
Because of their high cost, stainless steel pipes are normally only used for specialised
applications, where resistance to a corrosive atmosphere is required, or mechanical
TUBES AND PIPEWORK CALCULATIONS 265
strength has to be maintained up to very high temperatures. They are also appropriate for
use in systems which have to meet high standards for external cleanliness or hygiene. The
strength/weight of stainless steel tubing is superior to that of any other material, with the
exception of titanium. Alloys type 316/321are normally preferred for high-duty pressure
lines, with a maximum permissible material stress rising to 350 - 460 MN/m2 in the fully
hard condition. Stainless steel tubes can, therefore, be stronger than low-carbon steel
tubes, permitting reductions in wall thickness and weight for a given bore size and pressure
rating.
Titanium Tubing
Titanium offers an exceptional strength/weight ratio, excellent resistance to corrosion,
and a maximum service temperature of the order of 500C (930F). Tensile strength is of
the order of 860 MN/m2, which can be raised to 1100 or even 1400 MN/m2 by alloying.
Flow velocity considerations

Pipe sizes commonly used for hydraulic lines typically fall in the range from 6 mm to 40
mm, with both larger and smaller sizes available for more extreme flows. Flow Velocity
is the primary factor governing friction in flow through pipes, and is usually selected to
maintain an acceptable pressure drop. The flow velocities given in Table 2C.6.3 are
usually adopted as giving acceptable frictional losses, and are appropriate for both pipes
and tubes. In delivery lines it is also considered acceptable to err towards the higher
velocities in higher pressure systems, for systems over 25 MPa, a velocity of? m/s or more
would be acceptable. Whereas lower pressure systems of under 10MPa should be
maintained below the average quoted. The more stringent constraints on suction lines are
to reduce the possibility of cavitation but this section of a system's pipe work may require
more careful scrutiny if long lines are used or fluid has to be raised in level.
The flow rate used should be the maximum steady flow in that section of the system.
Higher transient flows are possible in delivery lines.
A more convenient guide to pipe sizing follows from re-casting the formula to
determine minimum bore diameter for a selected velocity:
This method can be particularly tedious in the case of turbulent flow and additionally
requires a fairly exact knowledge of fitting losses, etc. Line sizing by pressure drop,
however, has the advantage that system performance can be more closely predicted than
with line sizing by arbitrary flow velocity limits. It may also be found that using fully
acceptable pressure drop values as a design figure the resultant line size is considerably
smaller than that given by flow velocity selection, giving a saving in weight and cost of
the pipe work.
Pressure drop calculations

It is also possible to evaluate the pressure drop for a given pipe work system. This
calculation can be made to check a critical design, perhaps where selection has been based
on velocity considerations. As above, this more detailed evaluation must distinguish
between laminar and turbulent flow. In both cases, as described in the chapter on Basic
hydraulic theory, the pressure drop through frictional losses can be expressed by the
equation:
Pressure drop at bends

The previous solutions apply only to flow through straight pipes of uniform cross section.
Losses at bends do not lend themselves to definitive mathematical analysis and solutions
are invariably based on empirical data. There are changes depending on the radius of
curvature of the bend (R) relative to the tube bore diameter (d). As a generalisation, there
is no effective loss in bends of RId of 8: 1 or greater with turbulent flow. In the case of
laminar flow, some loss will be experienced at values of RId of20: 1or less, increasing with
decreasing RId ratio. As an example at an RId ratio of 3: 1, approximate practical values
are:
equivalent straight length of pipe = 2 x bend for turbulent flow
equivalent straight length of pipe = 5 x bend for laminar flow
Pressure losses atfittings

Losses at fittings are determined empirically and can be expressed in terms of 'resistance
coefficient' or equivalent pipe length. In the former case an 'average' or 'typical' friction
factor is assumed which does not take into account the variation with flow of Reynold's
Number and can lead to considerable error. In fact, there are three separate ways of
rendering 'resistance coefficients' which can lead to serious errors in interpretation. In
practice it is generally satisfactory, and considerably easier, to work with equivalent pipe
length i.e. the length of straight pipe of the same diameter as the fitting which would have
the same resistance or pressure drop as the fitting. Again, such data can only be determined
empirically, but can be expressed as a typical Lid value for various types of fitting although
such values are subject to scale effect as well as minor variations in geometry with different
sizes. A typical range of equivalent lengths is given in Table 2C.6.5.
Total pressure drop

The total pressure drop in a complete pipework system can then be estimated by rendering
all the bends, fittings, etc. in terms of equivalent pipe lengths, adding these to the actual
lengths of pipe involved.
This will then give a single value (LT) to enter in the appropriate formula forcalculating
pressure drop. Note, however, that further pressure drops may be present at the inlet port
to a cylinder; and through a control valve.
Pressure streugth of tube

Selection of tube is usually based on the working pressure but in many mobile applications
weight is also important. Pressure ratings for tubes are normally taken from the manufac-
turer's figures but can also be calculated from first principles. Where the diameter to
thickness ratio of the cylinder tube is greater than 16: 1, the stress produced in the wall
material due to internal pressure can be determined from the simple formula for uniformly
distributed hoop stress:
Provided the maximum material stress figure is taken within the limit of proportionality
of the material, this simple formula is valid. It does not hold true for higher stress values,
and thus will not accurately predict bursting pressures. The simple formula is thus
restricted to thin-walled tubes frequently appropriate for low and return line pressures. It
will over-estimate the pressure rating for thick-walled tubes (diameter to thickness 16: 1
or less), which includes the majority of high pressure lines. In such cases an alternative
formula must be used:
TUBES AND PIPEWORK CALCULATIONS 271
leakage as well as reduced reliability. If access to a leaking fitting is difficult it is less likely
to be repaired. If any maintenance operation is awkward or time consuming then it is not
likely to be done well if at all.
Positioning of components and routing of pipelines must therefore be planned to allow
access for all foreseeable maintenance and service operations. This should anticipate the
obvious such as access to a filter element but also predict the less obvious service and
replacement requirements of all other components. Filter access must also anticipate any
local draining or isolation requirements as well as allow sufficient space for the filter bowl
to be removed. It is also desirable that all pipe fittings should be accessible in case of
deterioration and leakage. If a failing tube connection is not accessible then it will not be
re-made. Attention should also be paid to the possibility of completely draining a system
or parts of a system, and suitable valves and access points included.
This may seem an idealistic wish list but the more attention that is paid in this direction
can only improve reliability and safety of hydraulic systems as well as minimising running
costs.
Pipe bending
The use of cold bends in tubing is beneficial if it reduces a connection (by reducing a
potential leak point) and may reduce pressure losses. There may also be a cost saving if
the appropriate bending equipment and expertise is available. The bend radius (R) of a pipe
or tube is normally defined with reference to the tube centreline and the tube overall
diameter (D). As a general rule of thumb a minimum bend radius of 5D will allow
reasonable benefit from the smoother flow path. However, there are practical recommen-
dations which place a limit at 3D in order to meet space limitations but working practice
may well give bends even tighter than this.
In the case of 1800 bends (V-bends) where there is the likelihood of appreciable
reduction in the wall thickness on the outside of the bend, the following formula is given
by ASTM A-566-65 to specify the minimum allowable wall thickness after bending:
Any bend will tend to reduce the fatigue strength of the line at that point; and any
mechanical or geometric faults in the bend will affect the normal line strength. Early
failure at bends can often be traced to ovality and the actual geometry of finished bends
is worth checking for this possible fault.
Supporting clamps
Pipework is subjected to numerous sources of loading inducing stresses in the tubes
themselves but more importantly in the connecting fittings. The aim of the pipework
designer and installer should be to minimise these loading effects on the fittings where they
are likely to cause premature deterioration resulting in failure and leakage. Sources of
loading include vibration, either internally generated or externally generated, thermal
loads (expansion and contraction), pressure induced loads, and externally applied loads.
The solution is both to support the pipework adequately and to avoid long straight pipe
runs. Support reduces the effect of vibration and can prevent external loads being applied
to fittings. The introduction of bends or even loops in long lines with appropriate support
prevents thermal and pressure stresses being applied to fittings.
Recommended spacing for supporting clamps of high-pressure lines is summarised in
Table 2C.6. 7. This is intended as a general guide and may be overruled by local geometry.
Closer spacing is necessary: at bends where support should be provided at the straight run-
out on both sides; where a number of individual lines feed out from a common group; or
where a hose joins a rigid line which should be supported locally. The main thing is to
provide adequate support for all lines and special attention should be given where
horizontal pipelines might be used a ladder rungs. It should also be remembered that
pipework can be very rigid and it should not be used to connect together two parts of a more
flexible frame where a short length of hose would allow the relative movement to take
place.
Various proprietary pipe clamps are available, some have an elastomeric lining bonded
to the inside to grip the tube without locking it rigidly in place. Others may be
manufactured with a semi-rigid plastic body in a variety of materials. It is also possible to
obtain multiple clamps which allow a number of parallel lines of different diameters to be
assembled closely.
In addition to providing normal support for the lines, pipe clamps also assist in damping
vibration which might set up oscillations in the tube run, again emphasising the impor-
tance of an elastomeric or vibration-insulating layer between the line itself and the fixed
rigid clamp. In certain systems, egoaircraft systems, all piping is required to be electrically
bonded and provision is often made for this in the design of the clamp. A small tongue or
similar shape is positioned so that it will contact the tube wall and effectively 'earth' the
pipe through the clamp.
Designing pipework installations

It has been strongly recommended that the designer should produce a pipework drawing
TUBES AND PIPEWORK CALCULAnONS 273
as well as a circuit diagram. Summarising the aspects to be considered in planning the

layout of pipeline runs:
(i) Pipework lengths should be kept to a minimum to reduce friction losses as well
as cost.
(ii) Long straight runs should incorporate at least one bend to accommodate thermal
expansion/contraction. Straight runs should be avoided as these can set up severe
tensile or compressive forces in the line and also make it difficult to couple to
fittings where the tube has to be sprung back to assemble or dismantle.
(iii) The number of bends should be reduced to a minimum consistent with the basic
geometry of the layout, employing generous radii as far as possible. The tube
should always approach end fittings as a straight length, adjusting the radius of
any adjacent bends to terminate accordingly.
(iv) All tubes, and particularly pressure lines, should be properly supported, especially
before and after bends on high-pressure systems. Any sudden interruption of flow
will tend to produce a straightening out of the line at bends, leading to 'whipping'
if the line is not supported. Supporting clamps should not, however, restrain the
tube rigidly but allow sufficient freedom of movement for thermal contraction or
expansion.
(v) Accessibility for maintenance purposes. Filter elements must be changed and it
is better to encourage access to filters with sufficient space, so that they will be
changed when scheduled or required. Similarly other components may need to be
changed or serviced without dismantling a system. Allowance should be made for
filling and draining systems and flushing on installation. All couplings should be
accessible in case fittings begin to leak.
(vi) Consideration to the use of lengths of hose to provide vibration or load isolation
between sections of a system.
(vii) Correct selection of tubes, hose and fittings for compatibility in accordance with
appropriate standards.
See also chapters on Pipework couplings and fittings, Hose, and Hose couplings and
fittings.
275
PIPEWORK COUPLINGS & FITTINGS
RIGID TUBE connections can be made in a number of ways and their importance for
reliable, leak-free system performance should not be under estimated. A number of studies
carried out in the last ten years have confirmed the high incidence of leakage from
connectors and fittings, at about 50% of the total sources ofleakage. The realisation of the
importance of good connections has lead to considerable development in this area in recent
years with many new designs, and some of these have now developed reputations for good
performance.
These developments mean that some previous classification systems are not now as
appropriate. Fittings can still be classified as either permanent or breakable, where the
jointing method in the former case does not allow for any disassembly. However, this
chapter is arranged with fittings described under the groupings associated with the
preparation of the tube itself, tube gripping where minimum tube preparation is required;
formed tube where some forming action to change the shape of the tube before assembly
is required; and a brazed or welded connection.
Some texts have distinguished between the terms fitting and coupling, such thatfittings
can be attached to tube ends and then the two fittings mated together with a coupling. The
term connector is a further alternative and is frequently associated with hose joints. No
rigid distinctions have been made in this book.
It should also be said that the best way to avoid a leak at a connection is to avoid the
connection altogether. Bending of tubes is one method but careful circuit and pipework
layout can also avoid unnecessary joints. Adopting different component mounting styles
should also be considered with, for example, manifold mounting of components a valuable
way to maintain accessibility and yet reduce the amount of pipework.
A tube connection must perform two major functions: provide a leak free joint and have
sufficient strength to hold two sections of tube together under a variety of loading
conditions. These loads were described in the previous chapter on Tube and pipework
calculations, and do not relate simply to pressure loads. There are other desirable features
for an ideal fitting which should be easy to assemble correctly, impossible to assemble
incorrectly, and be capable of breaking and remaking without deterioration. This chapter
is concerned both with the connection of tubes to other tubes and to other system
components such as valves, pumps and actuators.
Tube gripping fittings

Fittings under this heading make a mechanical connection to the outside of the tube to grip
it and give the joint the necessary mechanical strength. The sealing mechanism may be
either through metal to metal contact or by use of an elastomeric material. Both of these
mechanisms are referred to as a compression coupling. It should be noted that to provide
an effective seal with metallic contact the components must be pre-strained under an
elastic stress field. This pre-loading can reduce either through a loosening of the joint or
through fatigue and relaxation of the metallic components and then lead to leakage. The
same argument applies to elastomers but of course small changes in strain are less likely
to reduce the contact stress significantly and most elastomeric seals are designed to be
pressure energised. However, when using elastomeric seals then their compatibility with
different fluids must be considered for each case.
The simple bite coupling

Bite type compression fittings have the advantage that only the pipe ends need to be cut
true and square, when the fitting can be assembled in place and tightened up. Components
for such a fitting comprise a body, usually of hexagonal shape to accommodate a spanner,
a union nut, a hardened steel sleeve (sealing ring, compression ring, or ferrule) with a
cutting edge as shown in Fig 2C.7.1. The body has an internal cone angle of 24, to ISO
8434.
As the nut is tightened, the front edge of the sealing ring cuts into the pipe wall, throwing
up a small ridge of displaced metal, which provides a metallic interlock. The depth of cut
is controlled by the proportions of the sleeve and the final position, where there is a surface
of freshly cut metal, with the leading edge of the ring firmly embedded in the tube wall both
ring and tube deforming plastically. This provides positive sealing with mechanical
locking against axial displacement, the back edge of the ring also grips the tube. The ring
also relies on a certain 'spring washer' effect between the nut, coupling body and tube, to
maintain sealing under conditions of vibration. The joint can be broken, with the ring
restrained axially on the tube but able to rotate if correctly assembled, and itcan be re-made
a limited number oftimes without affecting its efficiency.
The DIN 2353 standard covers dimensional and material requirements of basic bite type
tube couplings. It is, however, not good practice to mix components from different sources
PIPEWORK COUPLINGS & FITIINGS 277
even though they should be compatible and work to this standard. There are both low
pressure (designated L) and high pressure (designated S, since it is a German standard)
variants. Low pressure sizes from 6 mm (25 MPa) to 42 mm (10 MPa) typically available,
and from 6 mm (63 MPa) to 38 mm (25 MPa) for the high pressure range.
There are problems with this basic coupling, one is the deterioration with service life,
either through loosening or strain relaxation. Loosening can easily occur since it is difficult
to prevent a slight torque loading in the tube, if this is the last joint made, and vibration can
then help to loosen the nut. Fretting may occur between the tube and the back edge of the
sealing ring. This may be caused by poor assembly procedures since different manufac-
turers have different recommendations and this could lead to different pre-loading ofthe
joint. Finally there is a limit to the number of times a connection can be re-made, either
after deliberate breaking or through re-tightening. Eventually over-tightening will cause
terminal failure and the only remedy will be replacement of the affected section of tubing.
There are, however, a number of variants and derivatives of the bite fitting which offer
better prospects.
Second generation bite type connectors

Developments of the bite type coupling have taken place successi vely over the last ten
years with a number of improvements producing' a second generation'. These have been
concerned mainly with improved reliability by separating sealing and gripping functions,
and prevention of over-tightening on assembly. Figure 2C.7.2 shows a cutting ring with
two biting edges and a third ridge to prevent over tightening.
A newer development is shown in Figure 2C.7.3, with a so called profile ring. This
provides two cutting edges but gi ves a larger contact region with the tube and is less prone
to relaxation. The square shoulder on the front outer edge of the ring also provides a 'stop'
when it contacts the connector body to prevent over tightening. The back edge shoulder
region is also relatively flexible radially and hence easily deforms to grip the tube giving
further support.
FIGURE 2C.7.3 - A profile cutting ring tube connection [Walterscheid].
Another alternative adopted by a number of manufacturers has been to keep the bite
action for tube gripping but to separate the sealing function with an elastomeric compo-
nent. A number of designs using this principle have been produced and an example is
shown in Figure 2C.7.4. This has been designed with easy assembly in mind, and gives
PIPEWORK COUPLINGS & FITIINGS 279
good sealing even in conditions with high pressure peaks and vibration of the pipework.
Such designs usually allow the elastomeric part of the seal to be changed if the coupling
is re-made.
These are three examples of a number of alternative designs produced in recent years
which can operate with DIN standard components. The nut thread is usually pre-coated
with lubricant for use with modern fittings.
Sealing ring couplings

Sealing ring couplings also separate sealing and gripping actions, and most frequently use
an O-ring for the elastomeric seal. The components required and the assembly sequence
are shown in Figure 2C.7.5 for a typical coupling.
A special coupling body (1) is required which with the support ring (3) gives the correct
sized cavity for the O-ring itself (2). The split collet (4) has a serrated inner edge with teeth
which can grip the tube. The inside of the nut has a cone shape which pulls the split collet
tightly onto the tube to gi ve the mechanical strength. Again a number of alternati ve designs
have been produced which have a similar component set. These again have good resistance
to vibration and pressure peaks, and the O-ring can of course be changed. These couplings
can usually be completely disassembled, leaving the tube unmarked and the components
can then be re-used.
Permanent tube grip fittings

The bite and sealing ring fittings can all be disassembled and re-made but there are some
grip type fittings which are permanent and two designs are described. One of these swages
the fitting body (a simple internally serrated sleeve) and the tube together in a press. It is
possible for the fitting to be swaged onto the tubing if it is thick walled, or for the tubing
to be swaged outwards into the fitting. Another gives a similar end result but by using a
shape memory alloy (nickel-titanium). This is initially machined to have a bore slightly
under the required tube size, and there are again serrations within this bore. The fitting is
cooled in liquid nitrogen and stretched over a mandrel whilst at this low temperature. It
then slips easily over the tube sections to be joined but when it returns to room temperature
it also returns to its original undersize dimension and shrinks onto the tube. This so called
cryogenic fitting is said to gi ve very good leak-free operation but is a relatively expensive
type and obviously requires special handling and assembly conditions.
Fittings with formed-tube

This group includes breakable fittings of the compression type but all require some pre-
forming of the tube before the connection can be made. This pre-forming operation is best
(and sometimes only) carried out using a purpose-built machine or press. However, it is
possible to use the majority of tube materials discussed in the previous chapter.
Flared coupling
The basis of the flared coupling is the expansion of the tube end in a cone shape or flare.
The inner surface of the cone provides a sealing region and the expanded diameter
provides a shoulder to give the mechanical strength of the joint. There are three variants
of the flared coupling as shown in Figure 2C. 7.6. The inverted flare type requires the tube
to be withdrawn the full length of the body for disassembly, whereas the disengagement
movement of the tube is much lower with the other two. The 3-piece, or sleeved fitting,
is preferable since it does not impart any twisting action on the tube when assembled, and
gives greater support to the tube in aligning the assembly.
Any flare between 30 and 90 may be used and there are a number of standards for flare
angle. British and European practice has favoured 300 but the most common American
standard is the SAE 37 flare (formerly called the JIC flare). This is also compatible with
hose connections designated as 74 (included angle). In all other respects flare angles are
NOT interchangeable. There is another common American SAE standard using a 45
angle but this is not often used in hydraulic applications except for some low pressure
automotive systems.
To produce a good quality flare in a tube requires some skill. The outer diameter of the
tube end after flaring must not be too large, when it might interfere with the threads in the
body, nor too small when structural integrity might be impaired. The tube thickness will
PIPEWORK COUPLINGS & FITTINGS 281
also have a significant effect on the result. The torque levels on the nut necessary to
produce a good joint are surprisingly low, small diameters can operate successfully finger
tight. Hence the break and successful remake potential for flared couplings is very good.
The limitations for flared coupling are the material ductility to withstand the deforma-
tion of flaring and the maximum wall thickness that can be used. Most of the materials
described above can be flared but mild steel, tungum and aluminium are the most widely
used. A wall thickness in steel up to about 3 mm for an 18 mm outside diameter tube can
usually be flared successfully. However, pressure ratings are often kept to medium levels,
below 20 MPa.
Other formed-tube couplings

There are two other notable pre-formed, proprietary couplings available for a wide size
range of steel tube. Both reproduce a type of end fitting which would normally have to be
welded to the tube. Figure 2C. 7.7 shows what is effectively a 90 flare but intended for use
with an O-ring seal on the end face. This produces a stand pipe end to the tube and allows
flanged couplings (described later in this chapter) to be used.
Welded and brazed coupling

The welding of steel or brazing of tun gum tubing can form some of the most effective
connection methods. Tubes can obviously be permanently joined together but also end
fittings of several types can be attached to tubes to make connections to other components.
The disadvantages of welding/brazing are the need for a skilled and qualified work-force
and the necessity to descale and clean tubes after welding to avoid disastrous contamina-
tion problems. Suitable procedures and guide-lines are available for both and the
techniques are very successful when correctly applied. This does, however, increase the
cost of such systems. Obviously appropriate tube materials must be used.
Welding may use socket type components as shown in Figure 2C. 7.9 or fittings for butt
welding. In general butt welding will be used for the larger sizes over about 40 mm
diameter tube. Weld fittings will be pressure rated and there are both light and heavy
PIPEWORK COUPLINGS & FIITINGS 283
ranges. A variety of end fittings are available and Figure 2C.7.1O shows the termination
of a tube connected to a manifold block. It includes a welded end nipple with 74 cone and
integral a-ring.
Flanged couplings
A flange type coupling can provide a stronger and more leak-proof than screwed fittings.
The most useful standard in this respect is the SAE flange series, which is of course
specified in inch sizes and requires an adaptor to metric tube sizes. The adaptor is welded
or brazed to the end of the tube and then provides an end face seal with the mating half as
shown in Figure 2C.7.11. The seal is made with an a-ring which fits into the groove in the
end face of the adaptor or fitting. The usual version, as shown has a split flange with two
bolts in each. An alternative has a single piece flange which usually incorporates the 0-
ring groove.
FIGURE 2C.7.11- An SAE flange coupling [courtesyof Hydraulicsand PneumaticsMagazine].
Some care is required in tightening such a flange fitting to make sure that all four bolts
are tightened evenly. The commonest form of failure comes from incorrect assembly
procedures when the flange can tip. SAE flange sizes start at 3/8 inch (10 mm) upwards.
Split flanges are available rated to 2lMPa up to 2 inch (50 mm) and there is a higher
pressure range series.
Screwed unions
Screwed unions are a 'traditional' form of pipe coupling still used in some existing
installations. They afforded a simple, compact form of joint which can readily be made and
broken down again, as required. Steel unions are employed for higher pressures and may
be rated up to 35 MPa (350 bar) more normally under 20 MPa.
In all threaded joints, the male thread was cut on the outside of the pipes to mate with
a female thread in the union or sleeve fitting. Tapered threads are best for screwed
connections, the most efficient seal being given by having a slightly greater taper on the
male than on the female part of the joint. Screwed unions have been used with PTFE tape
wrapped around the threads as a seal but this tape is also a notorious source of
contamination in systems and is best avoided. Modern fast-setting anaerobic type thread
sealants are free from the tendency of early sealants to degrade and contaminate the system
and can be advantageous with threaded couplings. In addition to improved sealing, such
sealants increase the strength of the connection and protect it against corrosion or
loosening due to shock and vibration. Manufacturers instructions should be followed.
Two examples of union type couplings are shown in Figure 2C. 7.12. Both incorporate
elastomeric seals, which also have the advantage of providing vibration damping as well
as sealing. Where the service temperature is too high to use an elastomeric seal, metal
wedge seals can be used.
285
HYDRAULIC HOSE
HYDRAULIC HOSE is the description given to a flexible tube designed with a reinforced
structure for the containment of hydraulic fluids under pressure. This structure is
sometimes referred to as composite hose. Hydraulic hose is the most convenient, if not the
only solution, for hydraulic lines connecting components or parts of a machine which may
move relative to each other. Other applications of flexible lines include isolation of the
pump from the pipework for noise and vibration damping; use as a shock absorber; and
to solve a variety of awkward installation problems where the use of rigid lines would be
inconvenient for installation, or maintenance for example to provide easier coupling or de-
coupling oflines where it would be difficult to withdraw a length of rigid tubing. However,
some considerable care has to be taken in specifying the correct hose and fittings for an
application if the hose is to be used correctly.
Flexible hose generally has a wide range of maximum pressure ratings with working
pressure up to 40 MPa (400 bar) quite possible as standard. It is a general characteristic
that for a given type of construction, a smaller diameter will have a higher pressure rating,
and vice versa. Hence, pressure ratings for a particular type of construction are given as
a range, when the actual pressure rating will be specific to a diameter size. Terms such as
"high pressure" or "ultra high pressure" may be used to describe such ranges. Although
these terms help in understanding the layout of a manufacturers catalogue it is of course
the rating of a particular hose which is important.
Production sizes of hydraulic hose range from 4 mm (3/16 in) up to 100 mm (4 in) inside
diameter, except in the case of very high pressure hose which is normally only of small
diameter. Sizes of hoses have commonly followed SAE standards where these also specify
established hose identification details as discussed below.
Construction
Basically a flexible hose consists of an inner tube, a reinforcement and an outer cover as
shown schematically in Figure 2C.8.1.
The Inner Tube - This retains the fluid and must be flexible, withstand high and low
temperatures without deterioration, be compatible with the fluid and have a smooth
surface to ensure free flow. Many types of synthetic rubbers are available to accommodate
the possible range of fluids.
The Reinforcement - This strong, flexible, reinforcing, and supporting member pre-
vents the tube from bursting under pressure, or collapsing under suction. Reinforcement
materials are knitted, braided or spiral-wound of natural or synthetic yarns and fibres,
metal wires, or combinations of these materials.
The Outer Cover - This provides protection for the reinforcement from abrasion,
corrosion, heat, light, moisture, oils, other chemicals and weathering. It also provides
identification either by marking, colour or corrugations.
Tube materials
The inner tube and outer cover materials are normally synthetic rubber although some
forms of nylon and other plastics are increasingly being used. It is not necessary for the
inner tube and cover materials to be the same and they may be selected to give the best
overall durability. Cost varies considerably (lOx) between rubber compounds and can
playa significant part in the choice. Natural rubber is not appropriate but there are a wide
range of synthetic rubber compounds available with many proprietary variants, known
widely under theirtrade names. Features of some polymer compounds, including the most
commonly used hose materials are discussed below.
(i) Nitrile-butadiene rubber (NBR) - Nitrile is a copolymer of butadiene and
acrylonitrile compounds. The proportion of the latter varies from 18% to 48% and
is often classified as low, medium and high Nitrile. It has excellent resistance to
petroleum based fluids and oils (increasing with Nitrile content) and will handle
fire resistant oil-in-water and water-in-oil emulsions. High Nitrile content
compromises the temperature range and typically it will operate in the range -
40C to 120C. It is the most popular material used in hose applications, however,
NBR is unsuitable for use with phosphate ester and less suitable for vegetable-
based oils. Nitrile rubbers have good mechanical properties but are highly
susceptible to ozone attack and so should be stored with care.
(ii) Polychloroprene (CR) - Widely known as neoprene, it is a compound with an
acetylene base and a moderate resistance to oils, but its main characteristic is its
great ability to withstand abrasion, sunlight and weather ageing. It is, therefore,
widely used as an outer cover for hose.
(iii) Ethylene propylene (EP) - these compounds are based on petroleum raw
materials, and hence are incompatible with petroleum based oils. They generally
have good mechanical properties and are weather and ozone resistant. This group
is mainly used to carry phosphate-ester based fluids and automotive brake fluids.
The diene modified form (EPDM) is widely available and has largely replaced
Butyl (IIR) rubber.
(iv) Fluorocarbon rubber (FKM & FPM) - fluorocarbons are the most appropriate of
a group of fluoro-elastomers, one of these has the trade name Viton. They offer
HYDRAULIC HOSE 287
a range of rubbers which can withstand chemical attack and are good at elevated
temperatures, operating in the range -20 to +240C. They are resistant to mineral
oils, oil-in-water and water-in-oil emulsions, water polymers, and aryl phosphate
esters but not alkyl phosphate esters. They generally have excellent weather
resistance.
(v) Polyurethane (AU & PU) - this ether based compound has strength and abrasion
resistant properties which are superior to most other polymers, with good flexing
and hard wearing characteristics making it suitable in harsh vibration environ-
ments. It is quite suitable for use with mineral oils and has good ozone and weather
resistance but has poor performance with hot water. It is rather expensive and
although it has good low temperature properties it has a limited upper temperature
capability to around 80C.
(vi) Polyester elastomers (YPBO) - this group covers a range of compounds which
have outstanding abrasion resistance with a resistance to most chemicals includ-
ing mineral oils. They are used for inner tubes as the basis of an all thermo-plastic
hose using polyester fibre reinforcement and either polyurethane, nylon or PVC
outer cover. In the smaller sizes this construction is capable of good working
pressures and can tolerate working temperatures up to 120C.
(vii) Silicone (MQ) - silicone rubbers have poor mechanical properties and have a low
resistance to many fluids particularly under pressure. They are not suitable as
hose materials but have particularly good low temperature flexibility, and high
temperature capability (-70 to +240C) which makes them appropriate for seals.
(viii) Polytetrafluoroethylene (PTFE) - is a non cross-linked polymer which is
virtually chemically inert. It has a high elastic modulus but low strength.
Manufacturing involves either sintering or moulding the basic shape but it can be
extruded into an inner tube, which needs to be thin in section. It provides
minimum resistance to flow and is compatible with most fluids normally carried.
PTFE will operate under a very wide temperature range -100 to +250C and up
to 20 MPa (200 bar). It is supplied in different grades with alternative properties,
and needs correct selection for a specific requirement. PTFE is widely used in
seals.
Hose manufactured from the synthetic rubber compounds, usually has the inner tube
moulded in the raw state on a mandrel. The layers of reinforcing, with intermediate fabric
or elastomer layers and then the cover are added before vulcanising and curing. Hose is
thus produced in finite lengths. Some discussion of rubber compounds and comparative
Tables may also be found in the chapter on Seals.
Reinforcement
Wire braid reinforcement is the type most commonly used for hydraulic applications
because it provides the best service life/cost economics. A basket weave wire braid
provides good dimensional stability and ease of connection to a coupling with the other
main alternative at higher pressures being spiral winding as shown in Figure 2C.8.2.
Spirally reinforced hose is extremely resistant to high impulse pressure conditions. This
is because there are no cross-over points (as in braided construction) and hence no local
places of rubbing action. Hoses of this type have greater impulse life performance and are
capable of operating at higher working pressures than braided construction. Each group
of reinforcement wires is called a 'plait' and each wire an 'end'. The number of ends in
a plait will vary between similar sizes and types of hose depending on the intended use for
the hose. During manufacturing, a thin gauge of compounded rubber formulated to
provide high adhesion, is applied between each layer or 'ply' of reinforcement. Tension
produced in the braiding process pulls the braid into the bonding material, forcing the
material through the braid interstices. This provides adhesion between reinforcement
layers and helps the finished hose to resemble a multi-ply composite.
HYDRAULIC HOSE 289
longitudinally and radially acts at a neutral angle of 54 44'. When the force vector angle
formed is greater than the neutral angle, the hose elongates and the diameter decreases.
When the force vector angle formed is less than 54, the hose contracts and the diameter
increases. If the hose has been manufactured to this neutral angle, pressurisation produces
no change in the net length or radial dimensions. It may be desirable to design the braid
angle to produce a change in volume to improve the ability of the hose to damp pressure
fluctuations. Not all hose is thus made with the reinforcement at the neutral angle. Some
examples of hose constructions are shown in Figure 2C.8.3.
Cover variations
For industrial use hose normally has a rubber type cover and can also be armoured with
an additional external spiral wire protective cover. Rubber covers that protect against
environmental conditions require resistance to temperature variations, solvent resistance,
as well as weather-resistant (ozone, ultraviolet, humidity). Normally, hydraulic hose
systems containing petroleum based fluids utilise neoprene as cover materials.
Hose standards
Hose construction is covered in various National and International standards including
ISO, BSI, SAE, DIN, API and CETOP. Most proprietary hoses conform to one standard
or another, the SAE standards having being the most widely followed (see Table 2C.8.1).
These dictate that the entire length of all hose is legibly marked on the outside cover,
parallel to the longitudinal axis, using a stripe or stripes that show the respective SAE hose
specification number. The internal size designation is repeated at least once every 228 mm
(18 in) and at the manufacturer's option, dash size designation may be included. Dash sizes
are designated in sixteenths of an inch by using just the numerator of the fraction. Thus,
- 12 is 12/16 or 3/4 inch in nominal size (the outside diameter for tube). They were
introduced to provide a compatibility standard for common connections between tube,
hose and fittings. Thus dash 8 components should all connect together. Note that this
ensures dimensional compatibility and does not relate to the pressure rating of the
components used, which may not be the same. A coloured yarn incorporated into the cover
identifies the manufacturer.
Similar designations are now used for ISO and DIN standards but the minimum
specification for the similar construction levels in these is now often higher than the SAE
requirement.
Pressnre rating
Most hydraulic hoses are constructed to meet established pressure rating standards. When
designing hydraulic hoses, three different hose pressure ratings are considered: maximum
operating pressure, proof pressure, and minimum burst pressure. The maximum operating
pressure is normally a quarter of the minimum burst pressure rating, giving the hose a
safety factor of 4: 1. Specification of these data may vary slightly with different manufac-
turers. Some classes of hose may be rated for a nominal working pressure with a higher
than usual safety factor for general applications, and a maximum working pressure rating
for approved applications where the safety factor is reduced to 3.
Logically, with good manufacture, a safety factor of3 should be satisfactory for systems
working at substantially constant pressure. Where the system is subjected to surge
pressures or other conditions generating pressure pulses, a factor of 4 (or higher) is
recommended. Where hose is used above rated working pressure it will result in shortened
service life and premature failure.
Required behaviour under impulsive conditions can only be determined empirically
and is described in various standards, ego ISO 1436. This specifies the minimum number
of cycles to failure at peak pressures of 125% or 133% of the mean pressure at a rate of
35 pulses per minute with a square type pressure waveform. A square waveform is more
capable of reproduction and easier to control within the parameters of the testing
equipment available. Further, the results from this wave are much more consistent with
regard to experimental scatter within a random sample. Other waveforms may still be used
for certain government, military and defence specifications and also in some national
standards where historical data is still very relevant.
The proof pressure, twice the maximum operating pressure, is used during non-
destructive inspection testing. Minimum burst pressure rating of a hose assembly is
normally four times the maximum operating pressure. Actual burst pressure is the pressure
at which rupture ofthe hose will occur.
Thermoplastic hose
Thermoplastic hoses have improved greatly in quality and performance in recent decades
and are now commonly considered for hydraulic applications. Many now have pressure
ratings comparable to that of 1- or 2-wire braid conventional hose. Particular advantages
of thermoplastic hoses are lower weight (normally less than half that of conventional wire-
braid hose), better flex impulse life and better resistance to abrasion. Properly constructed
a thermo-plastic hose is also capable of expanding under pressure and will still retain a high
burst pressure, an advantage if compliability is being sought.
Homogeneous semi-rigid nylon tubing can be rated for maximum working pressure of
the order of9 MPa (90 bar) and may, therefore, be considered suitable for low to moderate
pressure systems. It is, however, more often employed for flexible lines with a braided
reinforcement, where it is directly competitive with conventional flexible hoses.
The tensile strength offered by various grades of nylon ranges from about 43 MN/m2
up to 85 MN/m2 With glass braided fibre reinforcement, a maximum tensile strength
of the order of 170 MN/m2 may be achieved. A particular advantage offered by nylon
is that it does not deteriorate with fatigue, and also undergoes relatively little creep or cold
flow at elevated temperatures. On the other hand, the actual strength is markedly
temperature dependent and decreases rapidly. Tensile strength figures are normally
quoted for 20C and maximum service temperature for nylon pressure tubing is about
60C, at which level the maximum permissible working stress has been degraded by up
to 40%. Some thermoplastic hose is made with very high modulus aramid fibres which
produce very little strain or volume change. Conversely there are some low modulus fibres
which give lower working pressures but a greater expansion and are intended to reduce
pressure surge.
HYDRAULIC HOSE 293
All-metal hose
Metallic flexible hose normally takes the form of a convoluted metallic tube close braided
with steel wire as shown in Figure 2C.8.4. Depth and pitch of convolution and braid
tension are critical factors since these govern the change in length of the tubing under
pressure (which should be reduced as far as possible to minimise fatigue characteristics)
and also the degree of 'fretting' likely on the convolution crests. As well as strengthening
and protecting the outer surfaces of the convoluted tube, the braid layer or layers also serve
the useful function of dampening vibration and resonance, whilst containing the end loads
imposed by internal pressure.
The principal advantage offered by all-metal tubing of this type is the high working
temperature possible, typically in excess of 400C with seamless stainless steel (convo-
luted) tubes and braid. Such hoses are, of course, flame-proof. Pressure rating can be quite
high, although this tends to decrease rapidly with increasing bore size.
A more suitable form for high pressure applications has the convolutions in three-ply
stainless steel tube with an outer stainless steel braid. The outer braids are designed to
contain extreme pressure end loads and the hose is intended to be used with welded-on
fittings. Flexible metallic hose of this type retains the high operating temperature
characteristics of conventional convoluted stainless steel pipes (for example, up to
500C), with pressure ratings of up to 42 MPa (420 bar) possible, depending on bore size.
All-metal flexible hoses can be expected to have a higher flow frictional resistance than
plain bore hoses because of the convoluted inner form. This can be reduced by incorpo-
rating a flexible smooth bore lining, although the service temperature of the composite
hose will then be reduced to that of the liner material. Also the liner will expand between
the convolutions under internal pressure so that the bore diameter will still tend to take the
form of a series of ridges. The cost of the hose is increased appreciably in any case, and
with a high temperature liner material, such as PTFE, the increase in cost can be
considerable.
A more practicable form of construction where the use of a smooth bore PTFE liner is
required is semi-metallic hose, comprising alternate layers of PTFE tube and steel wire
braid. Although this reduces the maximum working temperature possible to the limit for
PTFE such a construction does permit pressure ratings up to 28 MPa (280 bar) with 4.7
mm bore hose; and the hose is substantially flame-proof.
Coiled tubing
Coiled tubing could be used for high temperature applications in place of metallic flexible
hose or swivel fittings. An important factor to consider in such cases is the fatigue effects
resulting from mechanical flexing and pressure cycling. Also a stack of coiled tubes is
highly susceptible to sympathetic vibration as a spring mass system, although this can
usually be eliminated or reduced by friction damping. Coiling can take the form of close
winding of the coils so that they bear against each other, or interleaving the coils with
friction bands. In practice the form of coiling would appear to affect the fatigue
characteristics quite drastically with pressure cycling, depending on the amount of tube
flattening produced. The smaller the degree of flattening the better. With a deflection of
7 per coil a fatigue life in excess of 200 000 cycles can be expected for stainless steel
tubing working at temperatures up to 100C.
Coiled nylon tubing hose may also be used for pressure damping, although it is more
usual to find it employed in the form of recoil hose. That is, the hose is coiled in the form
of a 'spring' length which can accommodate extensive axial movement, merely expanding
or contracting as a 'spring' with self-retracting characteristics. This can provide a much
neater and less troublesome installation than using a flexible hose to accommodate the
same degree of movement.
Hose selection & installation

In the case of hydraulic hose, the working pressure is generally the primary parameter,
which will normally determine the type (construction) of hose required. In terms of a flow
requirement hose is selected on the same basis as tubing to provide a sufficient diameter
to give an allowable pressure drop along the hose. It can be considered as smooth bore
tubing, and flow velocities should be maintained as recommended in the chapter on Tubes
and pipework calculations. Material compatibility is another feature for both the inner
tube and the external cover and must be matched to the fluid used and the environmental
requirements. Other factors to consider are the frequency and intensity of surge pressures,
both fluid and external temperatures, and the amount of flexing which will be required.
Finally protection of the hose should be considered.
As with any pipework installation careful planning and routing from port to port is
critical to a successful installation. The correct selection of the hose must be extended to
include the most appropriate terminations and the path followed by the hose itself.
Problems of abrasion and minimum bend radius are especially critical in flexing applica-
tions remember that hose end fittings are not flexible. Avoid twisting of the hose, the
installation should be oriented so that when relative motion is present the hose will bend
rather than twist. Add protection against external hose damage where necessary to both
hose and fittings. The following points are of particular importance in the installation of
hydraulic hoses and should be read in conjunction with Figure 2C.8.5 which gives
appropriate examples:
HYDRAULIC HOSE 295
Minimum bend radii - Bends are very common in hose installation and it is important
that manufacturer's recommendations on this matter are followed closely especially on
lines experiencing high surge pressures, constant flexing or vacuum conditions. Too tight
a bend also tends to cause wear at the joint between the hose and any fittings, which may
cause a fitting to blow-off. The simple remedy is, during installation, to allow sufficient
hose to cope with flexing movements. The correct choice of end fitting is critical to giving
a good installation and angled elbow fittings can be used to reduce bend radii. Note that
the hose should be bent in the same plane as the relative motion of the parts connected.
Correct line of hose - Just as important as minimum bend radii. Although the quickest
route between two points is a straight line, no hose installation should contain straight taut
hose. If it does, high operating or surge pressures can cause wear and eventually blow-offs.
Compensate for shortening effects by bending the hose or leaving slack in straight runs.
Try to avoid hose bending in more than one plane and avoid twisting on installation.
Abrasion -Usually caused by contact with moving parts; sharp edges; criss-crossing of
hose lines or improper use of clamps and poor assembly of elbow fittings adaptors. All of
these can cause a severe wearing away of the outer cover of the hose, also weakening its
reinforcement. Criss-crossing is one of the most common of abrasion problems, any
vibration, no matter how small, produces a sawing action, eventually wearing away the
protective cover of both hoses. This can easily be avoided by correctly applying a clamp
at the point of cross-over, thus efficiently separating the two lines. Hose can also be
clamped to the machine parts which are being connected (if carried out correctly it does
not invalidate the use of hose). This can reduce flexing between the hose and fittings, and
if a hose bends in more than one plane it can be clamped where it changes direction.
High ambient temperature - Exposure to high ambient temperature can drastically
shorten hose life, this excessive heat being transmitted through the outer cover to the inner
tube, causing breakdown of the material. Hose lines should therefore be routed away from
heat sources eg hot exhaust manifolds, or protective measures taken to reduce the effect
on the hose line, i.e. baffle or protective sleeve where re-routing is impracticable.
297
HOSE COUPLINGS AND FITTINGS

HOSE END fittings are an important part of hose selection. They offer great versatility and
the options available are capable of fulfilling a wide range of uses. However, this means
greater choice with more importance on making the correct choice if the best reliability
and performance is to be obtained. End fittings can be considered in two stages: the
connection from the hose to the fitting and the style of connector at the termination ofthe
fitting. The same sealing and mechanical strength requirements considered for tube
fittings apply equally to hose.
The greatest choice of options is at the end termination which is a breakable connection.
One problem here is ensuring compatibility between the number of different systems
particularly in terms of the range of cone angles which are used. These are not interchange-
able although adaptors are usually available.
The hose connection side has traditionally fallen into two broad groups - permanent or
re-usable types. Re-usable fittings were introduced half a century ago when the hose itself
was less reliable and replacement was a more common occurrence. These days the
reliability of manufacture of hose makes this less of a problem and flexible hoses are
commonly supplied as complete assemblies with permanent end fittings. This also
eliminates the problem in selecting the correct pressure range of fitting to match the hose.
Fitting connection to hose

There are many designs of both permanent and re-usable hose attachment fittings but they
all share some common features. In all cases the hose tube is gripped between an insert
nipple which fits inside the inner tube and forms part of the terminal connector, and an
outer ferrule (sleeve or socket) which fits on the outside of the hose to clamp it to the nipple
as shown schematically in Figure 2C.9 .1. In most cases these components are made of steel
and the intrusion of the nipple slightly reduces the bore diameter through the connector.
The seal is obtained between the material of the inner tube and the outer surface of the
nipple.
The mechanical strength is enhanced by a lip or a series of teeth on the nipple or on the
ferrule which, when the fitting is tightened up, provide a positive grip against axial loads.
In some cases the outer cover material is removed, exposing the outer layer of reinforce-
ment so that the sleeve obtains a 'bite' on the wire braid. This process is known as skiving,
and skive and non-skive fitting designs exist in both permanent and re-usable forms. The
outer cover must be removed with care to prevent damage to the braid, particularly if this
is a manual operation.
Reusable fittings are screwed or clamped to the hose end, and can be removed and used
again if the hose has to be renewed. Two examples of re-usable fittings are shown in Figure
2C.9.2. In the screw fitting, the nipple is first inserted in the hose and the cover socket is
then screwed forward over the hose to hold it in place. The segmented type fitting also
shown is appropriate for larger size (over 25 mm) spiral wound hose.
HOSE COUPLINGS AND FITIINGS 299
Many fittings of this type have the advantage that they require no special tools
(segmented type excepted) and can be assembled by hand, as weU as being re-usable. They
enable hose to be stored in bulk to be made up to length on the job, instead of having to
work to stock lengths of hose with integral end fittings. Some designs of re-usable hose
fittings require the use of a mandrel to ensure correct assembly without damaging the inner
tube. Reusable fittings are generally not available for the higher pressure hoses and are
diminishing in popularity in many sectors.
Permanently attached fittings have a similar nipple insert but are finished by means of
a ductile sleeve which is crimped or swaged over the hose and end assembly. Crimped
fittings tend to have a creneUated appearance whereas a swage sleeve has a smooth outside
sheU as shown in Figure 2C.9.3. In both cases the nipple has gripping teeth to hold the hose.
The concept of permanence means of course that if the hose is damaged then the end
fittings must also be discarded and the whole assembly must be replaced.
Choice of Type
The choice between permanent and re-usable fittings is often one of cost. On average, re-
usable couplings cost more than the permanent type, but the attachment procedure usuaUy
takes longer. When large numbers of hydraulic hoses are replaced, maintenance priorities
dictate the permanent sty Ie fitting. Many large users of hose will have crimping or swaging
machines which are not prohibitively expensive and aUow use ofthis cheaper alternative.
It is more common these days to order hose with specified fittings as an assembly. Re-
usable fittings still have their place for field replacement of hoses.
Fitting termination
The choice of fitting termination will usuaUy be dictated by the requirements of other
components or compatibility with other parts of a system. The advantage of aU hose
couplings is that they are capable of being broken and re-made. In many cases it may be
this feature which has led to the selection of hose. It should not be forgotten that high
pressure hose is only flexible in a relative sense and may not be very easy to handle. The
range of fittings available is indicated in Figure 2C.9.4.
The majority of couplings are screwed but flange fittings to SAE standards are also
widely available. The sealing method varies but the metal to metal cone seal is common
with all the national preferences for cone angle, noting that these are not interchangeable.
There are also some fittings incorporating an O-ring seal on the cone or as the face flange
seal. If the correct fitting is not available then frequently adaptors can be used and Figure
2C.9.5 shows some termination alternatives to give some help in identification.
The use of angled connections is recommended to improve the hose layout and reduce
bend radii. It is also important not to twist hose on connection. It is also possible to get some
swivel type couplings which can be rotated about their fixing.
Self-sealing couplings
Self-sealing couplings are particularly useful where some part of the system connections
may require periodic removal, change of position or as in many agricultural applications
a change between different implements. They are based on providing a two part seal; the
first provides a seal between flow and the surroundings when the coupling is in two parts,
and the second a complete seal for through-flow when the coupling is fully assembled.
Most two-way shut off couplings have male or female threaded bodies to accept standard
threaded hose fittings, allowing either the hose or coupling to be changed independently.
couplings are available in steel (surface coated), stainless steel and brass, the latter suitable
for all-water systems.
Such couplings incorporate a self-sealing valve in either the plug or socket connections
of the coupling, or in both. In the first two cases, a poppet valve is normally used; in the
last a poppet valve in one half and a sliding sleeve in the other, or a poppet valve in each
half as shown in Figure 2.C.9.6 - this usually being the cheaper solution.
Connection of the two components may be by screwing, bayonet fitting, or push-pull

quick release action. The main requirement is that they should be easy to part and re-
assemble. The majority of couplings have ball locking mechanisms, in which aring of steel
balls in the socket half of the coupling locates in a groove machined in the plug half. A
sleeve, spring loaded into the locked position is operated by hand to allow the balls to
retract for connection and disconnection. With high pressure systems, it is usually
necessary to lower the line pressure before uncoupling or re-coupling, to reduce hydraulic
load on the seals and cut down the effort needed to complete the coupling (which would
otherwise be considerable in the case of a large coupling). This limitation can be overcome
by hydraulically balancing the coupling design so that the sealing valves in each coupling
half are subject to only small hydraulic thrusts, regardless ofthe actual line pressure. Such
couplings can be broken and re-made at full system pressure with a minimum of manual
effort.
Flow must be immediately shut off when the line is broken, and immediately re-
established when the coupling is assembled. With a fully satisfactory design, no fluid is
lost and no air is introduced into the system. This is not usually possible with one-valve
or double poppet valve designs, and so the poppet valve associated with a sliding sleeve
may be preferred. The bore of a self-sealing coupling is usually made larger than the
corresponding line diameter, to leave a free area around the valve, equal to the full bore
area of the line. This reduces the pressure drop through the coupling although this is still
generally large in comparison with most other types of connection.
Flat face coupling

An alternative to the above is theflatface coupling which fulfils much the same function
but with an improved design. This was originally introduced to overcome problems of
Brinelling of the ball retaining mechanism when operated with high pulsating pressures
in hydraulic breaker hammers. The design has further evolved as in Figure 2.C.9.7, and
FIGURE 2.C.9.7 - A typical flat face self-sealing coupling [Holmbury Ltd].
is now available in a wide rang of fitting sizes, styles (including bulkhead) and pressure
ratings.
The flat face design has the considerable advantage of reducing the ingress of
contaminant into the system which may other wise occur, particularly through the female
half in a conventional design. It also avoids any oil spillage on disconnection and
eliminates the chance of air ingress on connection. It has an easier flow path than many
poppet designs and hence has a relatively low pressure drop. Many designs are flow bi-
directional but some are not. It still retains the ability to withstand high pressure pulses.
Although couplings from different manufacturers may not always be interchangeable
there are now some adaptors which convert between flat face and conventional self sealing
couplings as shown in Figure 2C.9.8.
305
SEALS
GOOD SEALING in hydraulic systems is an obvious requirement. The working medium

must be contained within the system components both for good safety and environmental
reasons, and for good efficiency. Seals within hydraulic systems may be classified as
either: static, resisting fluid pressure alone; rotating dynamic, resisting relatively low
pressure but with relative movement between contact surfaces; and reciprocating dy-
namic, where both surface sliding and pressure effects may occur. This latter case,
although not particular to hydraulic systems alone, is the most demanding and where many
developments have been made.
External leakage from hydraulic systems is a problem principally for safety and
environmental reasons rather than for power efficiency. The amount of flow lost through
leakage is rarely large enough to represent a power loss except for the important exception
of internal leakage in pumps and motors. External leakage affects the operating efficiency
mostly through the cost of replacement fluid and any consequent cleaning. If the fluid
replacement costs are too expensive then it is certainly a leakage problem. However, the
possibility of external leakage can prevent the use of an hydraulic system solution to an
actuation problem if the fear of the consequences of leakage is too great.
Seal materials
The materials most widely used are the elastomeric synthetic rubber compounds because
of their inherent elasticity. The most common is Nitrile rubber because of its compatibility
with mineral oil fluids but the wear properties of polyurethane make it the most popular
reciprocating seal. The non-elastic polymer (plastomer) which is also widely used is PTFE
(polytetrafluoroethylene). The mechanical characteristics of all polymer materials can be
difficult to characterise because of their rate and temperature dependency.
Nitrile rubbers are designated low, medium and high nitrile, depending on ascending
acrylonitrile content. Medium nitriles are most widely used since they combine excellent
compatibility with a useful working temperature range. The lower the acry lonitrile content
the better the low temperature performance but the poorer the resistance to mineral oil and
other hydrocarbon fluids. Nitrile rubbers have good abrasion resistance but are susceptible
to ozone ageing.
PTFE is noted for its almost complete resistance to chemical attack, wide service
temperature range and extremely low coefficient of friction when in contact with a steel
surface. However, it has low mechanical strength and low elasticity making it difficult to
'stretch fit' , hence the use of split rings. PTFE is usually used as a surface rubbing ring or
a support back-up ring in seal applications but it may be used as a component to reduce
friction in fabricated seals. It is also possible to obtain PTFE a-rings, which may be used
in the same groove size as the more conventional elastomeric a-ring.
Hardness of seal material is measured in Hardness Degrees, sometimes called Shore A
Hardness. Rubbers with low hardness are more flexible (softer) materials and will
generally have more conformity and seal more readily on rougher surfaces but are more
susceptible to wear, abrasion and extrusion. Abrasion resistance is generally better for
higher hardness elastomers. Hardness also influences frictional effects of seals and a
lower hardness will reduce breakout friction but higher hardness reduces running friction.
The hardness of rubbers can be modified by compounding, and the possible range for some
rubber materials is given in Table 2C.I 0.1. For the majority of seal applications hardness
values in the range 70 to 80 are used.
The elastic properties of seal materials are obviously important in maintaining the seal
contact. Elastic modulus is frequently used as a quality control test in manufacture. The
elastic recovery ability is also of importance in maintaining the seal contact. This refers
to the ability of the material to return to shape after deformation. It may have particular
importance in giving good sealing qualities when pressure fluctuations are present. In
cases where the elastomer may not have sufficient flexibility itself then composites of
rubber with fabric reinforcement can be used. These retain the stretch of the rubber but
without loss of strength and flexibility obtained from the fibre.
Temperature has a significant effect on the characteristics of elastomers. Decreasing
temperature tends to make elastomers harder, less flexible and reduce their elastic
recovery. They also have a so calledfreeze point, below which the material becomes brittle
and will break if flexed. At high temperatures elastomers lose strength, becoming softer
and more flexible. This is normally reversible but changes can be permanent and continued
high temperature exposure may exacerbate ageing. Operating temperatures for some
elastomers are also given in Table 2C. 10. 1.
Ageing refers to the progressi ve and permanent deterioration of the seal material with
time. There are many factors determining ageing but elevated temperature and oxidation
are usually the most significant (ozone for Nitrile rubbers). It is for this reason that a seal
'stored badly' for a long period before use may deteriorate and fail prematurely in terms
of service life. Ageing is usually characterised by an embrittlement of the rubber and a loss
of elastic properties.
Compatibility
When an elastomer is in contact with a fluid and it tends to absorb some of that fluid, it will
increase in volume or swell. This will modify the mechanical properties notably hardness,
strength, resilience and abrasion resistance, generally making the material more flexible
with reduced hardness. These conditions are to a certain extent inevitable and are taken
into account by the seal manufacturer in their design. A static seal may swell up to 50%
whereas a dynamic seal would not normally swell by more than 15% when used with the
appropriate fluid. The seal may also be subject to chemical attack by the fluid perhaps
causing embrittlement of the seal material. It is these features which determine the
compatibility of a seal material with the application fluid.
Mineral oil fluids are compatible with a wide range of synthetic rubbers, but Nitrile
rubbers are the normal choice of elastomer for seals, unless high-temperature working
demands the use of more exotic synthetics. The maximum working temperature of Nitrile
rubber is, however, higher than the maximum working temperature normally acceptable
for oil fluids. For proofed-fabric materials both Nitrile and Neoprene compounds are used
as standard. Polyurethane compounds feature outstanding tensile strength and abrasion
resistance, and are suitable for a wide range of mineral oils and fuels. Table 2C.10.2
summarises some features of elastomer compounds inlcuding fluid compatibility. Com-
plementary details of some of these compounds are given in the chapter on Hydraulic hose.
ISO standard 6072 also covers aspects of compatibility between elastomeric materials and
fluids.
Static seals
Static seals fall into the category of contact seals as do the majority of dynamic seals in
common use. Sealing contact is established by an initial pre-load in compression seals, in
addition some seals may be pressure energised. The majority of static-seal requirements
in hydraulic components can be met by O-rings and these serve to demonstrate these
features as shown in Figure 2C.1 0.1. This shows an O-ring initially squeezed in the correct
sized groove and then deformed under pressure to increase the sealing effect. Most seals
in hydraulic use are physically pre-loaded to a small degree and then pressure energised
with increasing system pressure.
O-rings are assembled in standard grooves, dimensioned to match the ring section
diameter (see BS 1806 or 4518). The pressure rating of the ring increases with reducing
clearance gap but is limited to about 10 MPa (100 bar). A possibility of extrusion exists
SEALS 309
at higher pressures which can be prevented with a back-up ring (usually PTFE). Provided
the mating surfaces are reasonably smooth and the squeeze is applied uniformly around
the circumference of the ring, further reduction in clearance gap (i.e. tightening up of the
joint) up to the limit of nominal metal-to-metal contact can effectively seal pressures in
excess of 700 MPa (7000 bar) with no extrusion of the ring.
The bonded washer, shown in Figure 2C.10.2, is another example of a static seal
commonly used on systems operating at pressures of the order of 20 - 30 MPa. Bonded
washers for use in oil-hydraulic systems have the sealing element in nitrile rubber (about
90" hardness) with a cadmium plated steel outer ring. They are available for phosphate
ester fluids with the appropriate rubber compound but not bonded since the phosphate
ester would destroy the bond.
Gaskets
Only limited use is made of conventional gaskets for static hydraulic seals. There are many
developments in modern gasket materials and designs, however, which can make them
worth considering for cover plate and other particular duties. Also there are many types
of all-metal gaskets suitable for high-temperature services.
Printed gaskets
Printed gaskets are a type of flat seal part gasket and part elastomeric seal. Using this
technique a gasket can be produced with the sealing medium exactly where required
incorporating the correct thickness, resilience and environmental resistance to suit the
joint under consideration. Further, it is possible to extend the role of such a gasket to
encompass other functions such as orifice sizing. More complex configurations than are
possible with a-rings, are also easy to produce. A single printed gasket could be used to
replace a multiplicity of a-rings and associated grooves as shown in Figure 2C.1O.3.
Printed gaskets for hydraulic services are normally made with selective deposits of
polyurethane based elastomer or similar polymers on a metal, phenolic or composite
substrate and have a pressure rating up to 40 MPa (400 bar ), with a maximum service
temperature of 90C. A metal substrate would be the normal choice for high-pressure
hydraulics, when slots and holes for valve porting, can be sealed with sealant beads on both
sides of the plate.
Rotating dynamic seals

Rotating shaft seals are used widely beyond hydraulic systems. The standard form for a
typical lip seal is shown in Figure 2C.1O.4, although there are many design variations. The
spring provides the compression load required when the seal is fitted to the correct shaft
diameter.
When running, a small oil film is maintained between the seal and the rotating shaft,
perhaps 0.25 11mthick. Provided this film thickness is not exceeded then the meniscus
SEALS 311
formed on the air side will not break and will prevent fluid escaping. This may be obtained
with a sharp edge as in Figure 2C.I 0.4 or a relatively large contact face as in figure 2C.1 0.5.
Basic seals of this type are capable of operating up to less than 50 kPa (0.5 bar). Higher
pressures, up to 700 kPa (7 bar) are possible by strengthening the support rim and
stiffening the web to prevent the sealing edge deflecting too much.
FIGURE 2C.I0.5 - Rotating shaft seal fluid film.
Seals of this type are capable of running at the typical shaft speeds required in most
hydraulic pumps and motors. Design variants are many but include the use of raised ribs
on the air side flank of the seal (item 5 in Figure 2C.1O.4) which actto return any escaping
fluid back to the sealing meniscus. This can considerably improve reliability with
vibrating shafts or those with large radial clearance. It is also possible to include a second
lip for dust exclusion, which may be necessary in some applications.
Reciprocating dynamic seals

There is a wide choice of high pressure hydraulic seals for reciprocating dynamic duties
and the selection depends both on the application and size of component involved.
Relatively simple seal rings may provide satisfactory performance on precision-made
components at pressures up to 35 MPa, but prove unsuitable on larger industrial
components operating at much lower pressures where the working clearances and finishes
are less favourable. The main compromise lies between good sealing effect and low
friction and wear, although many composite seals now give both low leakage and low
friction.
It should also be remembered that there are very many factors which influence seal
performance. In addition to the design of the seal itself and the materials used, the fluid
characteristics can also be important. The surface finish and mechanical clearances and
tolerances of the mating components add a further set of variables before the actual
operating conditions of pressure and speed are included.
The most basic form of reciprocating seal is the cup seal but it is suitable only for low
pressures. Cups have low friction, especially on starting, and they can be used satisfacto-
rily in cylinder bores which would cause early failure with other types of seal because of
excessive clearance between piston and cylinder, or generous tolerances on internal
diameter. To perform as double-acting seals cups must, of course, be mounted back-to-
back in a single assembly as shown in Figure 2C.1O.6.
Also it is worth noting that O-rings can be used both as rod and piston seals, although
they are most suitable for small-diameter short-stroke cylinders and moderate system
pressures. They are, however, often used for higher pressures by certain industries,
notably the American aircraft industry, in which they have established in the past a
particularly good reputation.
V-Ring seals
The V-ring, in its most basic form, is one of the earliest designs of flexible pressure-
energised lip seals. It has since developed considerably and is now produced in numerous
varieties oflip shape and section shape, see Figure 2C.1O 7. The main drawback with the
basic form is the vulnerability of the heel area to both wear, nibbling and ultimately
extrusion. The rounded heel sections shown and the use of back-up rings help overcome
this.
V-rings are used singly or back-to-back for double-acting seals. Simple V-rings may be
supported and located by positioning rings (lantern or retainer rings), especially where
open groove mounting or large sections are involved. For cylinders subject to rough usage,
shock loads or high pressures, the type of V-ring fitted up with an anti-extrusion ring is
recommended.
V-rings may be homogeneous elastomer or a composite with graduated flexibility or
even an integral back-up ring. They have the advantage of taking up very little space,
require only a simple machined housing. and are amenable to generous mechanical
SEALS 313
tolerances. In general U-rings are inexpensive, efficient seals for medium-duty hydraulic
cylinders at pressures up to about 35 MPa (350 bar) but higher operating pressures are
obtainable.
Among the numerous variants on the basic design, there are energised U-rings as shown
in Figure 2C.l0.8, often called wedge action seals. These generally combine the harder,
better wearing contact material with an insert of softer more elastic material to maintain
the compression at the sealing surface.
Another variant of note is the double lip polyurethane U-ring, which is widely used as
a rod seal in the 15 to 25 MPa range. The second lip, appearing as an innerridge in Figure
2C.1O.9, performs several functions. It stabilises the seal against rotation and holds the
outside heel firmly in the housing bore. It thus presents ingress of contaminant both along
the rod and around the seal via the housing.
FIGURE 2C.I0.9 - Double lip polyurethane rod seal [Hallite Seals Int. Ltd.]
V-Ring seals
The V-ring or chevron seal has a more robust section than a U-ring and can better withstand
both axial and radial loading. It is invariably used in sets, together with headers (or a header
and a gland ring), as in Figure 2C.l 0.1 O. They can act as both compression and pressure
activated seals and are usually used with a light pre-load applied on assembly to improve
static sealing. There are again a variety of detailed shapes and designs as shown.
The V-rings can be made from a variety of materials most frequently homogeneous
rubber, rubber impregnated fabric or leather. Some seal sets alternate between plain rubber
and fabric reinforced rings and the headers may be of completely different, usually more
rigid, materials including PTFE impregnated plastics. Some fabric reinforced rings are
split to aid assembly (joints should be staggered) but rubber rings are invariably one piece.
Typical pressure ratings increase with the number of rings in a set and rubber fabric is
capable of 70 MPa (700 bar) with leather rings giving up to twice this.
V -ring sets may be classified as medium or heavy duty and are appropriate both for
piston and rod seals, on a variety of sizes of components. Table 2C.1O.3 shows typical
overall dimensions for a range of sealing diameters. Their chief limitation is that they
require a fairly substantial gland length to accommodate them but the radial dimension is
relatively small.
Combination ring seals (Composite seals)

This description embraces a two or more part sealing ring produced as an integral unit,
usually an elastomeric ring with additional, more rigid components. These designs are
intended to address the compromise between friction and leakage and many of them
achieve this very successfully. It is common to have a PTFE or other hard plastic as a
rubbing surface in addition to an elastomer. A typical form is a square or rectangular
elastomeric ring in combination with a thinner ring ofPTFE to produce a composite seal.
The PTFE ring is on the inner diameter in the case of an internal seal, as shown in Figure
SEPlLS 315
2C.1 0.11, and on the outer diameter in the case of an external seal. Composites may have
all the components co-axially located as in Figure 2C.l 0.11, which also shows a stepped
ring for the internal seal, which acts with pressure in a single direction and can totally
prevent leakage, and the un stepped version necessary in most piston seals. Alternatively
the components may be axially located as in Figure 2C.1O.12.
Such seals can be located in a simple groove in a similar manner to O-ring fitting, but
with groove dimensions specific to the size of composite ring. This governs the degree of
squeeze and thus the radial pressure on the PTFE ring, and also locates this latter ring
axially. Beside providing a bearing surface this PTFE ring also confines the elastomeric
section within its groove, eliminating any possibility of extrusion so that no back-up rings
are required, even at pressures of the order of 35 MPa (350 bar). Numerous integral
combinations have been developed as proprietary seal rings, the majority of which are
designed to be accommodated in simple grooves. Groove dimensions are specific to the
geometry and construction of the particular section in order to provide the required amount
of squeeze on the elastomeric section for satisfactory static sealing at low pressures.
Sealing at higher pressures then depends on further deformation of the elastomeric section
under the applied pressure maintaining a seal contact pressure greater than the applied
pressure.
Proprietary seals of this type range from quite simple compact sections to large heavy-
duty sections, and cover the full range of working pressures normally required in hydraulic
cylinders and other components. They are produced both as single-acting and double-
acting seals. In the latter case the elastomeric section forms the middle part of the ring, with
back-up rings bonded to it on either side. These also act as bearing rings although some
designs incorporate additional bearing rings, examples of two further constructions are
given in Figure 2C.1O.13.
A particttlar advantage of composite ring seals is their compactness, reducing the size
of housing required. Piston seals incorporating bearing rings can also enable the length of
the piston to be reduced. Many seals of this type are also designed for simple assembly on
a one-piece piston head which is both stronger and more cost-effective than the traditional
method of assembling piston seals on a built-up piston head. Special PTFE compounds can
be stretch assembled but require adequate recovery time.
Wipers and scrapers

Piston rod seal assemblies usually include a wiper or scraper to prevent unwanted
contaminant being drawn back into the main seal region by the rod when retracting.
Wipers may be considered at the gentle action end of a spectrum which includes scrapers
at the other extreme to remove ice and caked mud. Both types may be included as a partial
seal (wiper/seal or scraper/seal) but this should not be taken as a replacement for the main
seal.
The action of an ideal wiper should be to remove dust and dirt particles adhering to the
oil film on the rod but without penetrating and removing this film. It should also prevent
air entering the system via the rod. This is a demanding specification and generally
achieved with a good degree of success in most industrial environments. The wiper must
have a reasonable flexibility but also be hard wearing in order to achieve this. A simple
chamfered hard elastomeric ring will achieve the required wiping action but may require
higher contact pre-loads with adverse effects on both wear and friction. There are many
designs for lip type wiper seals which generally achieve a better performance compromise.
A typical cross section of a notched type wiper seal is shown in Figure 2C.1O.14.
The action of a scraper needs to be more aggressive and requires harder materials both
to remove the contaminant and to have an acceptable life. Harder materials, plastic and
SEALS 319
variation of friction coefficient with time for an elastomeric seal, tending to the 'dry' value
shown as 10 times greater than the lubricated value.
However, these considerations cover only one aspect of the variation of the friction
coefficient and non-linear functions with both seal differential pressure and velocity can
also be defined. As an example, Figure 2C.l 0.17 shows the variation of friction coeffi-
cients with seal operating pressure for different seal types .
.Whereas Figure 2C.l 0.18 shows the frictional force variation with speed for a similar
range of seals but in combinations of rod and piston seals.
Seals lose their ability to function because of normal wear of the sealing edge of the
material. The greatest wear often occurs when starting from rest when the fluid film is at
its thinnest. Fluid erosion can also exacerbate wear as will particulate contaminants within
the sealing region. The surface finish recommended to minimise wear for cylinder bore
seals lies between 0.25 and 0.65 Ilm (Ra). It has been found necessary to adopt finer
finishes for rod seals because of the presence of abrasive dust particles, and recommen-
dations lie between 0.16 and 0.4 Ilm (Ra). At higher operating pressures the oil film
between the seal and surface is thinner and the friction is greater. It is then better to select
surface quality approaching the lower values quoted.
As a final example some actual friction values derived for a representative range of
seals, varying with operating pressure, are given in Figure 2C.1O.19. The losses due to
friction are expressed as a percentage of the maximum thrust of the cylinder in which they
were tested and are shown for the rod seals and piston seals separately.
In addition to all these considerations on friction, and the dimensional factors outlined
earlier in this chapter, the properties of both the seal material as well as the fluid are highly
dependent on temperature and will have an influence on the actual friction experienced in
service.
High-temperature seals
The maximum working temperature of elastomers is limited, depending on type. Seals in
Nitrile rubber are generally usable at temperatures up to 1000 C. Fluorocarbon is the most
satisfactory material for use in the temperature range 100 - 2500 C and has excellent
compatibility with hydraulic fluids, but lacks 'elasticity'. Silicone rubbers offer an even
wider range of working temperatures with good elastomeric properties, but are weaker
mechanically and have poorer compatibility with some hydraulic fluids. Either material
can, however, prove quite satisfactory for high-temperature static seals.
In the case of dynamic seals, automotive-type metallic piston rings are an obvious
choice for high-temperature piston seals. If necessary these can be energised by wave
springs. Alternative designs to consider are:
(i) Wedge seals based on a split PTFE ring bearing on a continuous PTFE ring,
energised by a metallic spring ring.
(ii) Metallic cup seals energised by a metal garter spring or finger springs.
321
PROTECTIVE COVERS
THE IMPORTANCE of cleanliness in hydraulic systems has been emphasised in the

chapter on Contamination control. One of the most important aspects of this is the
prevention of contaminant entering the system. Every time that a connection is left open
then anything from airborne dust to swarf can collect in the exposed opening, with the risk
that it will then be sealed into the system when the connection is made. This includes the
obvious valve, manifold, pump, motor and cylinder connections but should be extended
to include any other component including tube (before use and part assembled) and the
tank openings.
The requirement for protection is found at every stage of the manufacturing and
assembly process. Items in component stores departments need protection against dust,
moisture and corrosion to avoid rejection at the next production stage. Following
manufacturing processes, capping and plugging are common activities to ensure that air
and oil channels are kept clean and that exposed studs and threads are protected against
damage. Speed and ease of fitting and removal is critical, which explains why in many
cases plugs and caps have replaced masking tape and cardboard as the most cost effective
protection. The need for protection continues in the dispatch department where compo-
nents subsequently damaged in transit may be returned and cause costly delays.
Covers are usually made from thermoplastic elastomers including PVC and polyethylene
with materials resistant to higher temperatures also available. These include Thermoplas-
tic Rubber (TPR) suitable for use at temperatures up to 190C and high temperature PVC
capable of operation up to 230e. Compatibility is not usually a problem with the most
common hydraulic fluids but a check should be made. For example the grade of PVC
employed should be resistant to water based fluids. For extra resistance to attack by acids,
solvents and oils some ranges, maybe with limited sizes, are available in high density
polyethylene. See chapters on Hose and Seals for more information on elastomers.
Plain cover styles

The simplest sty Ie is a plain plastic cap as in Figure 2C.II.I a, which can be used to provide
protection for tube or stand pipe ends and are available in a wide range of internal
diameters. There are other variants with ribbed internal design, as in Figure 2e.11.1 b,
which provide a good grip but are easier to remove.
The hydraulic industry's most widely used style is the tapered cover, shown in Figure
2C.11.2, which can be used either as a cap or a plug. This is employed most frequently as
an interior plug for pipes, fittings and housings, however, it can be used as a cap for
protecting exposed studs and openings. As a cap, the dimension marked as d4 should be
smaller than the tube to be covered by at least 0.5 mm for every em of tube outside
diameter. If used as a plug, then the dz dimension should be at least 0.5 mm per cm larger
than the internal diameter to be covered. Versions of this style are available with larger
flanges to aid removal when used as a plug. Designs of parallel sided plugs are also used
with hydraulic components and versions with easy release pull rings are available. An
important advantage of the tapered design is that small variations on the size of the cover
can be accommodated by the taper.
Threaded cover styles

Tapered caps and plugs offer a very cost-effective solution to the majority of protection
problems. However, they rely on an interference fit and there is always the possibility of
them being accidently displaced or knocked off. A wide range of covers with moulded full
form threads (BSP, UNF and metric) is available which may prove more reliable,
including the most widely used internal and external threads. A recognised drawback in
using threaded covers is that screwing them into place is a time consuming operation and
any means of reducing both fitting and removal time wouid be advantageous.
Several ranges are produced in flexible PVC, which enables the cover to be push fitted
and quickly removed. Another advantage ofPVC is its high resistance to shredding. This
is particularly important for hydraulic components since any material removed from a
cover could subsequently cause problems as a system contaminant.
SECTION 3
PROTECTIVE COVERS 323
Threaded covers may incorporate either screw slots or a hand grip for assembly and
removal. There are also many partly threaded i. e. ribbed or ridged, covers which are easy
fitting, some being a push fit. Once pushed into position these snap-fit devices will give
a tight fit with only a single turn. Some other alternatives have only two or three threads
for ease of use: fitting and removal. Some covers feature a tear-away tab which aids
removal since when the tab is pulled a strip is removed all the way up the body and across
the top of the cover.
Covers which may be frequently re-used on the same item ego some forms of quick
release coupling, can be fitted with covers which have a retaining strap or ring to prevent
them being lost. These are generally push fit and removal devices.
325
HYDRAULIC FLUIDS
IN HYDROSTATIC systems the prime requirement of the fluid is to transmit pressure,

hence low viscosity and low compressibility are highly important features. Other factors,
such as those discussed in the chapter on Properties of Fluids, are also relevant, their
importance depending on the particular application in which the fluid is used.
The viscosity level of an hydraulic fluid is normally selected on the basis of the speed
and discharge pressure of the pump which must be satisfactorily lubricated for optimum
performance. Too Iowa viscosity will cause pump leakage and inadequate lubrication,
whilst too high a viscosity will cause overheating and cavitation. Probably one of the most
common problems associated with hydraulic equipment, after contamination, is leakage,
which is the prime cause of high fluid consumption. This is due, amongst more mechanical
reasons, to high pressures within the system and seal failure from both poor quality and
incompatibility. A further problem area is cross-contamination with other fluids, particu-
larly when the hydraulic fluid supplier or type is changed, or if there is access to soluble
cutting oils.
Types of fluid
Hydraulic fluids are not necessarily 'better' because they are more expensive. Different
types of fluid are appropriate for different applications and it is important to choose from
the application point of view. Water, for instance, has very low cost and it could be
adequate for low pressure-low-speed systems; at the other extreme it may also be
appropriate for high pressure pumps and systems designed specifically for that medium.
High costs are incurred, however, with aerospace hydraulic fluids where they need to
operate at high altitudes, low temperatures and must not ignite should there be a crash.
The purpose of hydraulic fluids has already been discussed in the chapter on fluid
properties. Now, in this chapter, the different types of hydraulic fluid are examined in
detail. Basically there are six distinct types, namely.
i) Mineral oils - either uninhibited or treated with additives;
ii) Water - either raw or sea water
iii) Emulsions - water-in-oil and oil-in-water;
iv) Water-based glycols - water containing alcohol, glycerine, etc.;
v) Synthetic fluids - e.g.
a) phosphate esters, b) carboxylate esters,

c) poly-alpha-olefins, d) polyglycols,
e) silicones.
vi) Ecologically acceptable fluids - including some in the above categories.
All of these have some fire resistance except the mineral oils and the synthetic fluids
(c), (d) and (e). The fluids have a designation rating given by ISO 6743 as shown in Table
3.1.1.
Mineral oils
Mineral oils are the most common choice for industrial hydraulic systems, with the
advantage of offering nearly all the requirements of an 'ideal' hydraulic fluid except for
fire-resistance. Straight mineral-oil lubricants are only suitable for hydraulic systems
working with low fluid temperatures and in highly clean systems. As a general rule,
however, such oils are compounded with special additives to produce hydraulic oils
specifically intended for use in all practical hydraulic systems. The moderate cost of good
quality hydraulic oil is more than justified and it should be regarded as a normal choice.
Care has to be exercised not to exceed maximum temperatures as the oil will deteriorate
and cause system corrosion and loss in performance; environmental pollution may occur
with spillages.
Special care has to be taken where mixing may occur. Thus the hydraulic system
associated with an internal combustion engine might employ the same fluid as specified
for the engine crankcase. Also on certain types of hydraulic equipment, particularly those
associated with a difficulty in having the correct fluid to hand, such as agricultural
machines, the designer has to allow for the possibility that the hydraulic system may be
filled by the user with almost any lubricating oil available. Specialised oil-hydraulic
equipment is, however, invariably designed around an oil of a specific viscosity (primarily
to suit the requirements of the pump or motor), and an hydraulic oil is always implied in
such cases.
Additives for mineral oils

Almost all hydraulic oils are compounded with additives, notably oxidation inhibitors,
corrosion inhibitors and anti-foam agents. Some oils may have less and others more (e.g.
film strength improvers or anti-wear additives can be advantageous where high bearing
loads are involved, and pour point depressants for fluids used in systems operating at very
low working temperatures, or starting up from cold in very low ambient temperatures). A
separate additive is also commonly employed to improve the viscosity index of the oil (see
later) .
The main cause of deterioration with a straight mineral oil is oxidation. The rate of
oxidation is increased by heating (e.g. high working temperatures for the oil), agitation
(which is present in most hydraulic systems to some extent or other) and the presence of
contaminants which can act as catalysts (notably metal particles).
Apart from the loss of lubricating properties, the onset of oxidation is accompanied by
the formation of soluble and insoluble degradation products, the latter being deposited in
the system in the form of sludge. The oil also loses its ability to separate from water and
air, both of which contaminants are invariably present; and will tend to become increas-
ingly acid, which can lead to corrosion.
Oxidation inhibitors work by showing a preferential absorption for oxygen and thus
remain effective as long as there is active additive remaining. Some, such as the
phosphorous and sulphur compounds, also possess marked anti-wear and anti-corrosive
properties and are thus multi-purpose additives.
Oxidation additives are usually added in concentrations of up to 5%, this being the
maximum figure for which such additives are fully effective. Higher proportions will not
normally give any increase in oil life and may even have undesirable effects.
Marked improvements in oil life have been achieved with specially treated base oils and
oxidation inhibitors, compared with figures realised less than a decade a go. It must be
emphasised, however, that ultimate life in a particular system will still depend largely on
the operating conditions, particularly the oil temperature and the cleanliness of the system.
Oxidation can be tested in different ways. Figures 3.1.1 and 3.1.3 show two examples.
Corrosion inhibitors are essentially rust inhibitors capable of adhering strongly to
FIGURE 3.1.1- Thermal stability test showing an oil with suitable additive versus an oil with
inadequate additive. (Tested by weight loss on copper and steel rods, discoloration of rods and
sludge-in-oil weight.)
metallic surfaces and 'passivating' the surface, or isolating it from contact with air and
moisture. The selection of a suitable additive is quite critical, however, both to meet the
service conditions concerned and avoid interaction with other additives. In particular,
certain types of rust inhibitors have a degrading effect on oxidation inhibitors, whilst
others may have a secondary effect of working as an emulsifying agent tending to emulsify
any free water present in the oil.
Anti-foam agents are added to ensure effective release of entrained air from the oil
surface in the tank without excessive foaming developing at the surface. Basically they are
'foam breakers', causing an early disruption of the air bubbles as they appear. By this
HYDRAULIC FLUIDS 329
means, air normally dissolved in mineral oil and released at lower pressures, or any
entrained air, is released with no adverse effects on the working of the system.
Anti-wear additives are both film-strength and metal surface improvers. They greatly
assist in maintaining the full lubrication properties of the oil and also lay down a deposit
on the component metal surface; this surface protection occurs either through physical
adsorption or by chemical reaction. One type of anti-wear additive is zinc dialkyl
dithiophosphate (ZDDP) in proportions up to about 1%, but because this has the
disadvantage of attacking silver within a component the oil manufacturer is expected to
mention this in the description of the product. The ZDDP combines an antioxidant with
its anti-wear performance
Another type of anti-wear additive, with similar basic properties to the ZDDP, is
sulphur/phosphorus (SIP); another is trial phosphate. The SIP, whilst being more expen-
sive, has some significant advantages over ZDDP in that it not only does not attack silver
but it has a better filterability in the presence of water. (Hydrolysis is possible with ZDDP
which leads to oil insoluble breakdown products blocking fine filters.)
The main wear test is the Vickers one shown in Figure 3.1.2. Another is shown in Figure
3.1.4.
Viscosity index improvers are additi ves to reduce the drop in viscosity with temperature.
The additives used are normally polymerised methacrylate esters, or butane or styrene
olefins, in proportions of from 4% to 8%. All such additives are susceptible to shear
breakdown and so the initial VI index achieved is seldom maintained in practice, the extent
of the break-down being dependent on the rate of shear experienced by the oil. In general,
an initial loss may be expected during the first few hours of working in the system, after
which the viscosity index should remain appreciably constant through the useful life of the
oil, unless continually subjected to shear stresses in a particular part of the system.
Modern polymeric viscosity index improvers show even better viscosity retention at
high shear rates than their earlier counterparts but still have limitations at extremely high
pressure, e.g. 35 MPa (350 bar). For very high pressure applications where no shear loss
can be tolerated, synthesised hydraulic oils are to be preferred as they do not use polymeric
thickeners and show no shear loss.
Water
Water, as a fluid power system fluid, has many strong advantages, but also many strong
disadvantages. It is the cheapest fluid, the most easily available and causes the least
problem when it leaks from a system, but then, because of its low viscosity, it is the most
likely to leak. On the other hand the low viscosity means that the water flows with the
minimum of energy loss through pipes and bends.
Compressibility is negligible, which ensures that less energy is absorbed and the time
to reach design pressure is rapid with no 'spongy' feeling to the dynamics. But then it may
cause corrosion, and its lubricity is very low and will demand special materials for rubbing
surfaces, particularly in pumps and actuators. Biocides may be necessary to deal with
fungus, and antifreeze will certainly be necessary if below-freezing conditions are likely
to be met.
Whilst the amount of water hydraulics is beginning to increase, particularly because of
its being totally fire resistant and non-polluting, in its basic form it is still somewhat rare
as an hydraulic fluid. More common are the emulsions which include a certain proportion
of oil or chemical.
Emulsions
Water-in-oil and oil-in-water emulsions are similar in general behaviour, the particular
difference being that in the case of water-in-oil, water is distributed in droplet form
through an oil medium, and in the case of oil-in-wateroil droplets are distributed through
a water medium. As a consequence, oil-in-water emulsions (with a water medium) are
rather more fire-resistant than water-in-oil emulsions. On the other hand, water-in-oil
emulsions (with an oil medium) generally have better lubricating properties
Both oil-in-water emulsions (HFA) and water-in-oil emulsions (HFB) have been
extensively used in underground mining and steel manufacture, but the separation
problems (water from oil) associated with HFB may be serious. The water content can be
up to 60%, fire-resistance being directly related to water content.
Similar additives may be included as for mineral oils, notably oxidation inhibitors, anti-
wear and anti-corrosion additives, and also emulsifying agents to maintain the emulsion
in stable form. Viscosity index improvers are not used since HFB emulsions are non-
Newtonian fluids with no viscosity index as such. Their actual viscosity is dependent on
the rate of shear, and at very high shear rates it reverts to that of the oil content itself. This
generally limits their application to systems or components which do not produce high
localised rates of shear, e.g. such emulsions would generally be unsuitable for use with
high-speed vane pumps or with rolling bearings.
Basically, HFB emulsions are most reliable in relatively clean systems; they also have
strictly limited working temperatures, in common with other water-based fluids, and the
possibility of water loss through evaporation and subsequent modification of the fluid
make-up.
In general, if there are no leaks in the system, any loss of fluid volume in aHFB emulsion
can be replaced by topping up with water. However, if there is fluid leakage, topping up
must be done with the correct water/oil mixture, otherwise the balance will be upset and
the lubricity of the fluid may be adversely affected, (or the fire-resistance adversely
affected if topped up with oil). Manufacturers of this type of fluid generally supply a
topping-up concentrate (which will usually also contain an anti-corrosion additive) and
specific instructions for its use.
Oil-in-water emulsions (HFA) are essentially water (also termed High Water Based
Fluids), containing approximately 2 - 5% of an emulsifier oil to provide limited
lubrication and anti-corrosion properties. Their use is limited to massive systems, i.e.
those which discharge to waste; and in pressure transmitters such as pit-props where
lubrication is not very demanding. Their main limitation has been their inability to
satisfactorily lubricate pump bearings. The percentage 'oil' requirement is normally
quoted as being 5% because if there are losses (and quick top-ups with water) then the
critical level of2% will not be passed. If the level drops below 2%, corrosion and bacteria
formation may start occurring.
Because of the low viscosity of HFA fluids - I cSt, or the same as water - hydro-
dynamic bearing films in pumps and motors are much thinner than with mineral oils.
Metal-to-metal contact can thus be more frequent, aggravating wear and also generating
contaminants. Modification of the detail design of the pump may be necessary to
accommodate the thinner films; and filtration down to 10 /lm in pressure lines and 25 /lm
in return lines is recommended to remove wear products. Filters and elements must be of
suitable type. Corrosion and erosion can also be a problem and most HFA fluids are not
compatible with zinc and cadmium plating, aluminium (unless anodised), cork gaskets,
and normal paints and sealants. Stainless steel or GRP is recommended for reservoir
construction. For seals, nitrile, neoprene and Viton are suitable elastomers.
Piston pumps in particular seem prone to increased wear and leakage rates operating
with HFA fluids; also showing lower volumetric efficiency (partially offset by slightly
higher mechanical efficiency). Many standard piston pumps will operate satisfactorily
with HFA fluids however, without modification at moderate speeds and pressures
provided pump inlet pressure is maintained at or above atmospheric pressure to eliminate
cavitation, e.g. using a flooded inlet with no filter or strainer on the inlet side.
Standard vane pumps, in general, are unsuitable for use with HF A fluids because oflow
volumetric efficiencies and rapid wear due to adverse vane tip loading. These limitations
can be overcome with special designs, e.g. with hydraulically balanced vanes with special
tip shapes and modified timing and porting. Such pump designs, however, are not suitable
for working as motors on HFA fluids.
Standard gear pumps are not suitable for use with HF A fluids, nor are they readily
modified for such duty.
HFA fluids can also produce problems with valves, notably leakage due to the low fluid
viscosity. In this respect, poppet valves are better than sliding spool valves; but all types
of standard valves can give acceptable service (except those with wetted components in
aluminium).
Theoretically, at least, an increase in leakage should also result in an increase in heat
generation, although this is more than offset by the higher specific heat and thermal
conductivity ofthe fluid. Thus HFA systems tend to run cooler than mineral oil systems
in actual practice.
Stability- HF A fluids are stable from +soC to +SOc. Below OC freezing occurs which
can separate the fluid. Above SOC evaporation is accelerated.
Lubricity - HFA have limited lubricating properties as indicated by the low viscosity.
Additives can improve this so lubricity therefore varies with the fluid manufacturer.
Acidity - To retard growth of bacteria HF A fluids should be alkaline at about 8 to 9.S
pH.
Bacterial contamination
Surprisingly, mineral oils and oil-water emulsions are prone to contamination by micro-
organisms. The necessary environment to introduce bacteria is free water plus a nutrient.
Water is always present in oils to some extent and the necessary nutrient may be provided
by the additives used in hydraulic oils. Basically, as the number of organisms multiplies,
additives are consumed, decreasing their percentage and effectiveness. Replacement of oil
losses with fresh oil again provides nutrient promoting further multiplication of the micro-
organisms.
In a hydraulic system contaminated with bacteria the oil becomes turbid and a slimy
sediment is also formed, further increased by depletion of anti-rust and anti-wear
additives. Under static conditions the organisms will become concentrated in the interface
between the fluid and a water layer. When the system is working, with the fluid in
circulation, the bacteria have the effect of acting as an emulsifying agent providing a
water-in oil dispersion. The bacteria themselves can also accumulate in certain parts of the
system to cause blockage in valves and filters.
The most effective way of removing bacteria from a system is to drain it completely,
flush through with an oil soluble biocide followed by conventional flushing oil and refill
with fresh sterile fluid. Filtering or batch filtering is no cure, for whilst this will remove
solid particles it will not remove bacterial infection and water.
Sterile oil is produced by re-refining the original oil to include complete sterilisation,
removal of products of oxidation and fine filtering. Provided this is done comprehensively
it can be fully effective and the cost can be as little as one half that of new oil. Naturally,
any depletion of additives would have to be restored.
Water-based glycols
Water-glycol fluids originated as straight water-glycerine mixtures, with the glycerine
content adjusted to give the required degree of protection against freezing in water-
hydraulic systems. The glycerine content employed can range up to SO%. A secondary
advantage offered by such mixtures is a raising of the viscosity of the fluid and an
improvement in viscosity index. Cost, however, is relatively high, nullifying one of the
basic advantages of using water as hydraulic fluid. Thus oil-hydraulic systems are
preferable to water systems for low-temperature applications for all general uses, even
where a large bulk of fluid is involved.
Water-polyglycol mixtures have, however, been further developed as industrial fire-
resistant fluids, mainly around Hydrolube H-2, offering superior protection to water-in-
oil emulsions, and lower cost and minimal compatibility problems compared with
phosphate esters. Nitrile rubber seals are equally suitable for mineral-oil fluids and all
water-based fluids. The lubricating properties of these mixtures are greatly improved by
the incorporation of anti-wear and load-carrying additives to provide satisfactory lubrica-
tion under boundary film conditions; and their viscosity is increased by the addition of
polymer type thickeners, which also provide a high viscosity index. They are reasonably
stable, although they do need close control and regular checking of water and alkaline
content.
The water content controls the fire-resistance of the fluid (increasing with increasing
water content). Evaporation and loss of water are likely during service, more especially
at higher system temperatures, and so the system design should aim to minimise such
losses. Topping-up is normally done with a pre-mixed water/glycol solution, following
specific recommendations given by the fluid manufacturer. Water alone, or glycol alone,
should not be used to top up a water/ glycol mixture to compensate for volumetric loss.
The use of a water-glycol fluid inevitably calls for bearing loads to be de-rated, a typical
figure being about one-third of the rating for an oil-lubricated bearing. The use of such
fluids is not generally recommended with rolling bearings, or for gear pumps operating at
pressures above 3.5 MPa (35 bar), or components with close clearances relying on
boundary film lubrication. Also the maximum service temperature of water-glycol fluids
is generally low in order to avoid evaporation and loss of water content, with the frequent
need for checking and topping up.
Synthetic fluids
Potential advantages of synthetic fluids are superior oxygen stability, high viscosity index,
lower viscosity ,lower pour point and good lubricating properties (considerably better than
water-based fluids, but not necessarily better than mineral oils). Synthetic fluids are
normally able to operate over a wider temperature range than mineral oils and they are
generally more environment friendly.
Phosphate esters are the best known and the most widely used type (HFD). The
performance of phosphate ester fluids is more or less directly comparable with that of
mineral oils, particularly as they are available in a wide range of viscosities. Their viscosity
index is lower than that of mineral oils, but can be enhanced by viscosity index improvers.
Bulk modulus is higher, however, which means that phosphate ester fluids are superior to
mineral oils as regards compressibility effects at higher pressures.
The chief disadvantage of phosphate ester fluids is their very high cost, followed closely
by their complete lack of compatibility with conventional elastomers and paint finishes.
Butyl was originally used for elastomeric seals and packings, with possible alternatives
in the more expensive Viton and silicone rubbers; however, ethylenepropylene (EP)
rubbers are now the standard choice elastomer for use with phosphate ester fluids.
The only paints suitable for reservoirs, which are compatible with phosphate ester
fluids, are epoxy based. More conventional paints used for external finishes are readily
stripped by spilt or leaking phosphate ester fluids.
A minor disadvantage of phosphate ester fluids is their higher specific gravity compared
with mineral oils. Maximum service temperature is generally higher and phosphate ester
fluids can be worked at temperatures up to about 150C without degradation of the fluid.
If a change is made from a system using a mineral oil fluid to a phosphate ester fluid,
a complete change of seals is necessary, as well as a change of paints used on the reservoir.
Fluid manufacturers can specify the change-over procedure necessary.
Silicone fluids
Silicones are another class of high-cost fluids - prohibitively so for all but highly
specialised applications. Their chief attraction is their suitability for working at high
service temperatures up to the order of360 - 370C, with the added virtue of an extremely
high viscosity index so that reasonable viscosity values are ma~ntained up to very high
temperatures.
The performance of silicones as lubricants has been considerably enhanced by the
introduction of improved silicone fluids, although these may show some slight loss of high
temperature properties. They remain the sole commercial fluids available for working at
temperatures in excess of 150C.
All silicone fluids are, of course, fire-resistant, but would not normally be selected on
this basis alone because of their high cost, compared even with phosphate esters.
Ecologically acceptable fluids

These fluids are becoming more common due to the greater awareness of the need for
biodegradable materials and the problems of toxic waste. Mineral oils are potentially
biodegradable, but because ofthe modern formulation - to give extended life - the rate of
molecular breakdown is extremely slow. Environmentally acceptable fluids need to
degrade relatively quickly when released into the environment. Agriculture and food
applications are initially the most relevant for such fluids. For instance, agricultural
tractors and forestry machinery using hydraulic equipment should always use biodegrad-
able fluids so that oil spillage causes the minimum of ecological damage.
Of the fluids already discussed, synthetic esters, polyglycols, dilute emulsions and
water glycols are already acceptable in most cases. The new hydraulic oils, developed
from plants such as rape seed and sunflower seed, are natural sources which have some
inherent additives. Figure 3.1.5 shows the biodegradable rate of a Shell Naturelle HF fluid
in soil, compared with a mineral oil.
A further important function of such fluids is in the food trade. Here the type of fluid
used must not only be biodegradable but also acceptable for human consumption. Perhaps
this is one of the fastest growing developments in fluids at the present time. For instance,
Shell Oils have produced their range of Cassida Fluids HF multi-purpose oils for the food
industry; in this case there is a careful blend of synthetic hydrocarbons and selected
additives which make the fluid both useful as a general purpose lubricant as well as an
hydraulic fluid.
Application snitability
There are a number of features which must be taken into account when considering the
most suitable fluid for an application. As suggested earlier, most general fluid power
applications would work happily with a mineral oil, as long as its viscosity (and additives)
are appropriate. The other hydraulic fluids would be chosen for specific reasons such as
compatibility, fire resistance, or ecological demand.
Compatibility
Possibly the most common problem, is that of compatibility. A new oil may have been
agreed with a supplier, perhaps to cover a number of new applications, and it is used with
existing systems for which it is not suitable. Immediately, there is both the problem with
compatibility with the seals and with the metals used in the various components. (Not only
the base oil type must be considered, but also some additives may attack certain metals.)
Specifically, metals to be avoided in contact with different types of fluids are:
mineral oils - none
water-in-oil emulsions - cadmium, zinc
water-glycol - cadmium, zinc
phosphate esters - aluminium (bearings)
chlorinated hydrocarbons - aluminium (bearings)
water - iron, zinc, aluminium
Lack of compatibility, in the case of additives, is usually discovered as corrosion of

bronze parts, e.g. in vane pumps and bronze piston slippers in piston pumps. ZDDP, as
mentioned earlier, is not compatible with silver plated components.
Fire resistance
Fire resistance is a very definite requirement with certain systems, but all could benefit
from some fire resistance; indeed, there are grades of fire resistance. The specific
applications for fire-resistant fluids are in high-pressure industrial hydraulic systems
operating in close proximity to naked flames or high-temperature sources, or specifically
in hazardous surroundings where fire anc\/orexplosion risk must be reduced to a minimum.
The other important field of application is for aircraft hydraulic systems, where system
pressures are normally higher than in industrial practice, and the consequences of fire even
more serious. The primary risk, in all cases, is that failure of, or damage to, a pressurised
system can lead to the release of a fine spray of fluid which can be ejected a considerable
distance; and in the case of an oil fluid this spray is highly combustible.
The distinction between a 'non-flammable' and 'fire-resistant' fluid is not exact, since
only water is a true non-flammable fluid. Most water-based fluids are fire-resistant rather
than non-flammable, with a strong 'snuffing' action. Synthetic fluids range in character-
istics from being substantially non-flammable to having limited flammability. All such
special fluids are thus most accurately classified as just being fire-resistant, rather than
non- flammable.
Whilst the advantages of using a fire-resistant fluid in such cases are obvious from the
safety point of view, there can be disadvantages in their adoption, notably higher cost,
some loss of lubricating properties and problems of seal compatibility, compared with
mineral-oil fluids. Their application is, therefore, still far from universal even in systems
where a fire risk is present, although they may be mandatory in some situations.
Fire-resistant fluids fall into two distinct categories:
i) Water-based fluids where fire-resistance is derived from their water content.
ii) Synthetic fluids where fire-resistance is derived from the chemical composition.
The main types of fire-resistant fluids are:
i) Water-based: oil-in-water emulsions,
water-in-oil emulsions,
water-glycol mixtures,
water
ii) Synthetic (no water): phosphate esters,
phosphate ester mixtures,
halogenated aromatics,
silicones.
Specific advantages of aqueous-based fluids are ready availability and relatively low
cost. Water-glycol mixtures are appreciably more costly than oil/water emulsions, but
have the further advantage of higher viscosity and excellent viscosity-temperature
characteristics. Disadvantages are generally poor lubricating properties, working-tem-
perature limitations imposed by the water content and the possibility of phase separation.
Synthetic fluids offer a nearer approach to oil-like qualities and availability in a wide
range of viscosities, but may have other disadvantages apart from high cost. Thus
halogenated aromatic compounds and silicones are poor lubricants (without additives)
and chlorinated aromatics are toxic. However, phosphate esters are suitable for high
pressure pumps of all types. Products of combustion may be obnoxious but are unlikely
to be toxic.
Changing fluid types

In order to avoid cross-contamination, it is important to rigorously prepare the system for
the new fluid. Basically, this consists of removing all traces of the previous fluid (including
any gums that might have formed on, say, valves) and checking that the components are
compatible. The routine could be as follows:
Completely drain the system, preferably when warm; ensure that low points and
blank ends are disconnected.
Using a lint-free cloth, wipe all oil-wetted open surfaces, such as reservoirs.
Using appropriate solvents, thoroughly clean surfaces where impregnation may
have occurred, such as with paints
Check composition of all seals, i.e. that they are compatible with the new fluid.
(Change if necessary.)
Check which metals could be susceptible to the new oil, and examine the system
to see if they are present. (Change if necessary.)
Replace filter elements with appropriate type.

Fill with the new fluid and flush through at low pressure for a minimum of 30
minutes. If the change is from mineral oil to water-glycol, the flushed fluid can be
left for 2 or 3 days in the system to let any mineral oil come to the surface (and be
removed).
Remove the flushing oil and fill with the new oil again, and run at operational
conditions for several hours to endeavour to bring into solution the remaining 1%
or so of the previous fluid.
Drain and fill with the fluid to be used.
It may be necessary to mark the filler positions with the type of oil now installed.
If water or a water emulsion has previously been used, it is important to remove as much
of the water as possible.
341
FILTERS AND FILTRATION
HYDRAULIC SYSTEMS operate at considerable pressure, and in order to provide the

efficiency at such pressures, pumps and motors have close clearances, and control valves
and servos require even smaller clearances. It is not surprising, therefore, that small
particles within the system fluid - the solid contamination - may well cause damage and
blockage within the system. Other contaminants, such as water (in mineral oils) and
biological growth (in water based fluids), also have a serious effect on the system. Air, too,
may be a contaminant. It is, therefore, important to include some means of separating the
contaminant from the working system in order to achieve trouble-free operation and long
component life.
Filters and filtration are the major components in contamination separation and control.
This chapter, therefore, introduces the subject of such control by looking at the most
important aspect, before the next chapter goes into the more intricate details of the
complete subject. However, in order to fully understand the filtration mechanisms, it is
valuable to have a basic understanding of contamination; this is provided by the brief
following section.
Contaminants
Suffice to say, at this stage, contaminants are not just confined to the original build.
Contamination is constantly being generated and ingested, and hence filtration is an on-
going requirement.
The types of contaminant likely to be present, or generated, in hydraulic systems can
be generally classified as:
i) 'Soft' particles - soluble, or non-soluble
ii) 'Hard' particles - metallic or non-metallic (e.g. silicaceous)
iii) Liquid - water or chemical
Solid non-soluble contaminants, smaller than the clearance spaces, can silt up the clear-
ances, resulting, for example, in valves 'sticking'. Larger non-soluble contaminants can
produce 'blocking' so that flow is restricted in orifices. Abrasive contaminants of the same
sizes as the clearances can lead to high rates of wear. Very much smaller solid particles can
also lead to erosion in parts of the system having high fluid velocities. It is therefore
essential to incorporate filtration in any closed hydraulic system in order to remove solid
contaminants of all types above the particle size considered to be seriously detrimental to
the system. These particles can be governed by the general 'cut-off' rating of the filters
employed. This is further emphasised by the fact that the presence of solid contaminants
in the system will inevitably generate more contaminants, often at an accelerating rate.
Contaminants within a typical hydraulic system may be derived from both external and
internal sources, and may be in the original build or ingested later. Table 3.2.1 gives an idea
of the range of contaminants which may be present:
Size of contaminant particles

Particles in a system may vary from the infinitely small to possible a few millimetres
across. To attempt to remove the submicron particles from hydraulic systems would not
only be difficult, it would be extremely costly due to the number of very fine filter elements
which would have to be changed at frequent intervals. Generally for hydraulic systems,
it is considered, that particles are either
Coarse particles - wear and silicaceous dust (greater than 5 /lm)
Fine particles - silt (5 /lm or less)
The silt particles are mainly responsible for clogging if allowed to build up; they also cause
erosion and general degradation of the system if the particles are abrasive. Wear particles
are usually the larger particles which can cause catastrophic failure, or be the result of
catastrophic failures, e.g. by forcing open a check or relief valve or causing binding in a
pump or motor; the silicaceous dust would be ingested in such as quarry environments, as
well as many other situations. It may be possible to deal with only one of these size types,
but most systems would need to control both.
Water contamination
Water can seriously derate the effectiveness of hydraulic mineral oils. It can also
cause corrosion. (See chapter on Fluids.) Where the content is large, most of the water
can be removed by draining-off from the bottom of reservoirs (after a static period).
FILTERS AND FILTRATION 343
Smaller percentages can be absorbed by appropriate 'water absorbent' filters or vacuum

dehydration.
Filters
There are two schools of thought as to the most efficient means of removing solid particles
from the fluid (it is also possible to advocate contaminant insensitive components)
The particles should be allowed to settle, for removal later
The particles should be encouraged to remain in suspension, so that the filters can
work to their maximum efficiency.
Whilst a choice may be possible with some static systems, all mobile equipment must
attempt to deal with particles by seeking to remove them from a flowing system, i.e. by
filtration. The general use of smaller reservoirs also gives more rapid circulation of fluid
and thus less opportunity for particles to settle out.
It has been assessed that over 70% of hydraulic system failures are due to contamination
or poor fluid condition. In the case of mobile hydraulics failure rates due to fluid
deterioration may be even higher. Essentially, therefore, filters are necessary in hydraulic
systems to provide a particular or specified level of contaminant removal. This can vary
with the type of system, types of components involved, application and duty cycle.
Filter ratings
Filters for solid contaminants are rated according to their ability to cope with the
contaminant. The ratings cover the size of particle filtered out (and efficiency), the
pressure drop experienced, the pressure rating of the element and housing, and the amount
of contaminant which can be retained.
The efficiency of particle size fIltration

So-called 'nominal' filtration size ratings are no longer acceptable in high precision fluid
power systems. Such a rating is open to considerable abuse, as there is no standard of that
type which is sufficiently closely controlled to ensure consistency.
'Absolute' filtration is, according to the definition, where 100% of a certain size of
particle (and larger) is retained by the filter. Whilst this may be possible and true where
beads are being retained by an accurately machined porous sheet, it is not practicable for
the real particles (of considerable shape variety) in a 'randomly' constructed fibres filter.
Another possibility is where the 'efficiency' of the filter is quoted in relation to the
proportion of contaminant in a certain test fluid which is retained by the filter. Again this
is open to abuse, because this proportion would vary depending on the concentration of
contaminant, the flows used and the time over which the test has been conducted. It is, thus,
an arbitrary value determined by the filter manufacturer and oflittle use in comparing one
filter (from one supplier) with that obtained from another.
The only currently internationally acceptable rating is that prepared by the International
Standards Organisation (ISO), and quoted under ISO 4572 - Hydraulic fluid power -
Filters - Multi-pass method for evaluating filtration performance. Two Multi-pass test
stands are shown in Figure 3.2.1.
dependent only on the size of the filter housing, and remains constant for a constant flow
rate) and partly by a viscous loss (which depends on fluid viscosity and the permeability
of the filter medium). This latter loss increases as the element becomes more clogged,
leading to a gradual rise in pressure drop as contaminants collect on the element.
Eventually the pressure drop will rise sharply after the fashion shown in Figure 3.2.2. The
apex of the 'knee' so formed by such a characteristic curve determines the useful life of
the element, in terms of an 'acceptable' level of differential pressure increase, and the
period over which this increase is substantially low and linear in characteristics. The
differential pressure can be expected to rise very sharply after this with further testing.
The filter bypass would normally be set to this 'knee' pressure, or to the safe limiting
pressure for the strength of the element, whichever is the more significant. Generally
speaking, a 'safe' pressure would not be more than about 50% of the 'bursting' or
disrupting pressure of the element, although the bypass would normally start to open
below this figure to give some differential pressure release.
Whilst oil viscosity does increase slightly with system pressure, changes in the differential
pressure developed across the filter element are mainly due to flow. This explains why
relatively weak filter elements, with a disrupting pressure of perhaps 0.7 MPa (7 bar ) or
less can perform satisfactorily in systems operating at a pressure of 21 MPa (210 bar ) or
more. In the ideal situation with no pressure surges, the element would never be subjected
to more pressure than the actual differential developed across it, unless it becomes clogged
and is not pressure-relieved by a bypass valve. The inertia of such a valve may, however,
make it susceptible to damage under rapid surge pressure conditions mainly because of the
instantaneously modified flow rates and changes in velocity head.
The limiting differential pressure depends on the filter medium construction and may
vary widely with different types. As a general rule for filters of the same construction, the
finer the filtering the lower the permissible differential pressure, which in turn means the
larger the size of filter likely to be required.
The relationship between pressure drop and flow rate is largely determined by the
design and size of the complete filter (see Figure 3.2.3), and by the fluid viscous losses
through the element. Characteristic curves in this case can be calculated for a specific fluid
and specific fluid temperature (i.e. specific fluid viscosity). Performance curves of this
type are normally available from manufacturers for each filter type and size.
The general relationship between pressure drop, flow rate and fluid viscosity can be
expressed for the element in the form:
Integrity and strength of a filter

Integrity tests need to be undertaken to ensure that the filter will be safe and provide
reliable service in the chosen system. Collapse and burst resistance tests are part of the
suite ofISO standard tests. A full list of the ISO tests is given in the chapter on Standards.
Compatibility
Compatibility with the system fluid must relate to the system temperatures involved to
ensure that no degradation of the element or its seals occurs during its normal service life.
Degradation can occur through:
i) Absorption of fluid into the filter medium or binder rising causing swelling
(increased pressure drop and choking)
ii) Hardening or embrittlement of the filter element which can cause cracking and
breakdown of the material.
iii) Disintegration of the element.
In general glass micro-fibres and wire mesh are fully compatible with all hydraulic fluids
(provided the complete filter does not include parts in aluminium, cadmium, magnesium
or zinc which are attacked by water-in-oil fluids). Cellulose media tend to swell in water
and are not generally suitable for water-in-oil and water-glycol fluids. Filters with active
media cannot be employed as these are capable of removing additives commonly used in
hydraulic oils.
Compatibility with other system components needs little comment, other than that the
filter should be readily fitted and coupled to existing units and the fact that filters of the
required size are available to fit standard line sizes, etc. It is also desirable that the form
of the filter is such that it is readily accessible for removal of the bowl or body and element
for cleaning or element replacement.
Basic filter construction

The 'filter' consists of several parts, as shown in Figure 3.2.4. Not all features are included
in every filter, and there are other parts which relate to special filters. The key feature of
the filter is the 'element', which may be integral with the housing (spin-on canister), or a
separate item which can be replaced. Obviously, for expensive housings used for high
pressure systems, it is more cost effective to have a replaceable element.
Types of hydraulic filter media

Examples of different types of filter element used in hydraulic filters are given in Table
3.2.3. Paper (cellulose media) and wire mesh are widely favoured but can only provide
partial silt control. Finer filtration involves the use of glass fibres, and these are now
extensively used in fluid power applications.
Sintered porous metal and glass fibre media are capable of providing full silt control.
Glass fibre covers a wide range of size; it offers good pore size distribution, greater open
area than cellulose or wire mesh media because of the smaller fibre diameter, and better
dirt holding capacity. Basically it may have low strength, but this can be enhanced with
resin treatment. It is normal practice to support glass fibre elements on both the upstream
and downstream sides.
Cleanable versus disposable filter elements

In view of the difficulty of removing all trapped contaminant, the cleaning of filters is
generally not undertaken. Inefficient cleaning may actually cause the filter to become a
hazard, because debris may be shaken out of the filter pores but not removed - it then is
ready to enter the system as soon as full flow is reached. In some cases the cost of re-
cleaning may be as high or higher than the replacement element.
Where cleaning is considered a viable option, and is economically acceptable, then the
more robust media - e.g. wire mesh and sintered porous metal elements - are the most likely
to be used as they are more economically re-cleaned, and may be used a number of times.
Three to seven cycles of cleaning and re-use are typical, but this depends on the service
conditions and cleaning facilities available. The most suitable cleaning system for woven
mesh is an ultrasonic bath; metal fibre elements usually require chemical cleaning.
Electrostatic fIlters
Electrostatic 'filters' are used off-line. The 'filters' are alongside the flow, and hence, do
not become clogged in the same way. The idea of such a filter is that when particles in the
fluid enter a high voltage field, then each particle becomes charged (either positively or
negatively). The charged particles then move across to adhere to the side of the 'filter'
sheet. Large quantities of contaminant can therefore be trapped without in any way
increasing the pressure drop through the 'filter' .The idea of the electrostatic filter is shown
in Figure 3.2.6.
Such filters are not suitable for water based fluids due to the high electrical conductivity.
Only a small proportion of particles are trapped at each pass, and at start-up there is a
certain shedding of particles; this means that such filters cannot be used in the main fluid
line, but over long periods of time may be very effective off-line.
FILlERS AND FILTRATION 351
Filter location
Filter location in a system is equally as important as the type of filter. General practice is
to locate filters either as a protection of a susceptible component (such as a servo-valve),
or to trap the debris generated by a wearing component, e.g. on the pressure side of the
pump. A filter on the suction side of the pump can be less robust and less costly, but would
have to be of large size and relatively coarse to avoid the possibility of restricted flow
starving the pump and causing cavitation; a suction gauze in a reservoir may be used. To
keep the general level of contaminant to an acceptable standard, off-line filtration may be
more acceptable (or an extra); very low levels of contaminant overall can be achieved by
this means. (Such low levels may also be achieved in-line in specific regions, by suitable
fine filtration.)
So possible positions for filter location (see Figure 3.2.7) are:
i) In the delivery line immediately downstream of a pump.
ii) At the inlet point to each critical component, if necessary
iii) In the return line to the reservoir.
iv) In the pump-case drain line.
v) Off-line and run independently.
Pressure-line f'dtering
A pressure-line filter is located on the delivery side of the pump and is thus exposed to full
system pressure. It will protect the system downstream from pump-generated or pump-
passed contaminants, but not from any contaminants generated downstream of the filter.
Three possible filter configurations are shown in Figure 3.2.7. Locating the filter before
the relief valve gives constant flow through the filter. Located downstream of the relief
valve the flow through the filter will depend on system demand; and in off-load periods
will have leakage flow or full flow depending on whether the control valve is of blocked-
centre or open-centre type, respectively. Such positioning thus makes it more difficult to
estimate the varying flow rates to which the filter may be subjected and so the former
Filter location
Filter location in a system is equally as important as the type of filter. General practice is
to locate filters either as a protection of a susceptible component (such as a servo-valve),
or to trap the debris generated by a wearing component, e.g. on the pressure side of the
pump. A filter on the suction side of the pump can be less robust and less costly, but would
have to be of large size and relatively coarse to avoid the possibility of restricted flow
starving the pump and causing cavitation; a suction gauze in a reservoir may be used. To
keep the general level of contaminant to an acceptable standard, off-line filtration may be
more acceptable (or an extra); very low levels of contaminant overall can be achieved by
this means. (Such low levels may also be achieved in-line in specific regions, by suitable
fine filtration.)
So possible positions for filter location (see Figure 3.2.7) are:
i) In the delivery line immediately downstream of a pump.
ii) At the inlet point to each critical component, if necessary
iii) In the return line to the reservoir.
iv) In the pump-case drain line.
v) Off-line and run independently.
Pressure-line f"dtering
A pressure-line filter is located on the delivery side of the pump and is thus exposed to full
system pressure. It will protect the system downstream from pump-generated or pump-
passed contaminants, but not from any contaminants generated downstream of the filter.
Three possible filter configurations are shown in Figure 3.2.7. Locating the filter before
the relief valve gives constant flow through the filter. Located downstream of the relief
valve the flow through the filter will depend on system demand; and in off-load periods
will have leakage flow or full flow depending on whether the control valve is of blocked-
centre or open-centre type, respectively. Such positioning thus makes it more difficult to
estimate the varying flow rates to which the filter may be subjected and so the former
system is normally preferred. In this case a bypass across the filter is essential to eliminate
excessive pressure build-up against the pump should the filter become clogged.
Additional protection for the system can then be provided by further filters preceding
critical components, or point-of-use filters. Filter requirements can be selected in a
number of different ways, depending on how critical the protection is for each component.
If the first filter (following the pump) provides the necessary fine filtering, the first
component in the system is protected. Subsequent components needing protection from
contaminants which may be generated by the first component can be preceded by an
additional fine filter. A component with more generous clearances not needing such
protection need not have a point-of-use filter preceding it.
Return-line filtering
A return-line filter is located downstream of the last working component in the system, but
upstream of the reservoir. It thus removes all contaminants (down to its rating level)
ingested or generated by the pump and system components before the fluid is returned to
the reservoir. It has the advantage that it is not likely to be subjected to large pressure surges
as can occur in pressure lines, but it can be subject to unsteady flow conditions. It thus
needs to be robust enough to accommodate flow surges. However, in general its cost will
be significantly lower than a pressure filter. Figure 3.2.8 shows tank mounted filters.
Off-line filtration
An off-line filter may be located in a separate loop between the hydraulic system pump
and the reservoir. It operates independently from the main system with its own low flow
pump; it may be fitted directly in and out of the reservoir. Its purpose is to provide a means
of maintaining the cleanliness of the fluid contained in the reservoir only (Figure 3.2.9 is
a typical unit). Where the loop enters the system upstream of the main system pump, it
provides an additional pressure boost for the pump. Fluid cooling may be included.
FILlERS AND FILTRATION 353
Its particular use is for overall fluid cleaning at suitable maintenance intervals. It can,
if necessary, be operated when the main system is in use. It does not, of course, dispense
with the need for a filter or filters in the main system since it only cleans the amount of fluid
present in the reservoir. In general the filtration system used, such as electrostatic
filtration, can bring the level of contamination down to very low levels given sufficient
passes of the fluid. Main line filtration has a reasonably high efficiency withjust one pass.
Suction fdters
The only provision usually made for filtering on the suction side of the pump is a simple
sump strainer fitted to the suction line in the tank or reservoir. The suction line connecting
to the tank should be located as far as possible from the turbulent flow induced by the return
line, and the two should be separated by baffles providing a sharp-edged weir barrier, if
possible, so that sediment cannot be carried across from the 'return' side to the 'suction'
side of the tank. Sometimes a strainer may also be fitted on the return line to prevent coarse
particles being released into the tank, but this should not be necessary in the case of static
installations as effective 'separation' can be achieved by good tank design. A major
problem with such filters is the difficulty in servicing, and many designers no longer
include reservoir based filters.
If a filter is used on the suction line, rather than in the reservoir, then only a coarse filter
should be used with a large housing to reduce pressure loss as much as possible. A bypass
is also essential.
The choice of fdter rating and size

There is a very large range of filter manufacturers and suppliers, each of whom has
available a selection of possibilities. Some of the larger manufacturers have available a
Guide for Selection, either in printed or computer form (CD ROM). The BFP A (British
Fluid Power Association) provides an independent guide (P5) which outlines a compre-
hensive process for filter choice.
There are different approaches to filter choice. For instance, one can consider the
hydraulic circuit and choose a filter which is simply related to the type of 'system' or
industry (Table 3.2.4). However, this is very basic and may be highly inaccurate because
each system contains a range of components which vary in contaminant sensitivity from
one system to another. Another approach is to break: the system down into regions and
consider the most sensitive component in each region that a filter has to cover.
Ultimately, the approach is to select appropriate filters to achieve the required
cleanliness level; this is how the BFPA P5 is arranged. To abbreviate this process would
not help; however, some indications of what is involved can be appreciated from the next
chapter - on Contamination Control.
A suitable size of filter is selected from the flow-rate/pressure drop characteristics,
bearing in mind that these must be corrected for fluid viscosity if it is different from the
empirical figures or quoted viscosity. Where size and weight are not important it is
generally best, for service life, to use an oversize rather than a marginal size filter, unless
the cost factor is critical. Alternatively, maintenance is easier by using 'duplex' filters,
with two or more filters in parallel so that one can be isolated for removal whilst the other
takes the full flow; this will increase both the number of connections required and the
pressure drop through the fittings. An oversize unit will give a lower pressure drop for the
same flow rate and a longer period between servicing.
Magnetic fdters
Magnetic filters range from simple magnetic plugs normally intended for fitting in
reservoirs, to versions of more or less conventional filters incorporating a permanent
magnet element. An example of the working of a magnetic filter is shown on an in-tank
insertion filter in Figure 3.2.10.
A magnetic filter will attract and collect ferrous metal particles, such as wear particles,
and polarised non-ferrous particles. A proportion of other particles may also be retained,
however, by a mechanical entrainment process (although performance in this respect is
unpredictable). Because the magnets are upstream of the main filter in the housing, any
particles shed during cold starting, when viscous forces are at a maximum, will be retained.
The inclusion of the magnets extends element life.
FILlERS ANDFILTRATION 355
Magnetic filters are usually only regarded as secondary filters for the specific purpose
of removing ferrous metal particles or initial wear products. On their own they in no way
replace the media type of filter for hydraulic circuits, but designs such as Figure 3.2.10,
providing a combination of both types in a single unit, are used in reservoirs either for
suction or return-line filtration.
Duplex fIlters
In situations where continuous flow is essential over long periods of time, a duplex filter
arrangement may be fitted. This allows a filter to be changed without flow cessation. See
Figure 3.2.11.
Filter testing
The chapter on Standards (the sections sub-headed Filters) lists a number of tests which
can be undertaken on filters and their assemblies. The testing is essential in order to be able
to maintain consistency, or to compare different filters (see Figure 3.2.1). More detail on
these tests are given in the Filters and Filtration Handbook and the Handbook of Filter
Media published by Elsevier.
357
CONTAMINATION CONTROL
CONT AMINA nON CONTROL is considered by many as the essential key to reliability
in hydraulic systems. The greater the emphasis on contamination control, the longer the
life of the system.
The function of contamination control is that of keeping the levels of contaminant in the
fluids to an acceptable level. The use of filters, as described in the previous chapter, is of
considerable importance in such control, but filtration is not the only factor which needs
to be considered. For example, even if the filtration built into the system was correctly
designed and fitted, and it worked efficiently when the system was commissioned, poor
control and maintenance of the system thereafter may totally overload the filters and
prevent their being of any value.
Contamination is constantly being generated and ingested, and hence filtration is an on-
going requirement. Contamination is not only solid, it can be liquid or even gaseous.
High fluid-operating temperatures can mean reduced oil viscosities and less protection
against wear, contributing to increased contamination generated within the system.
Erosion can be a very real problem in high-pressure systems as localised fluid velocities
may be as high as 175 mls or more. This can lead to rapid wear on hardened and polished
surfaces adjacent to high velocity fluid streams when particles of only 3 to 5 Jlm are present
in the fluid.
The environment is also a source of external contaminants during the working of a
system, typical entry points for such contaminants being:
i) Air breathers,
ii) Rod seals on hydraulic cylinders
(particularly in dirty atmospheres and/or with increasing seal wear),
iii) Access plates and other detachable items.
Contamination control can never remove all contaminant, but it does need to bring the
contamination level down to what is considered acceptable for the particular system.
The measurement of contamination

The level of solid contamination in a system can be described in a number of ways, e.g.
Gravimetric level - total mass of particulate in a volume of fluid
Membrane optical density - appearance of particles on a membrane through which

a quantity of fluid has passed
Count of particles - at two or more sizes, per volume of fluid
Assessment of particle size concentration - at two or more sizes.
There are advantages and disadvantages for each, but all present the operator with a
measure of the seriousness of the contamination in a system.
Gravimetric level
To obtain the gravimetric level a membrane (say, 47 mm diameter, 0.8 f.lm pore size) is
weighed new, and then 100 mL of fluid passed through. After drying the membrane, it is
re-weighed to find the increase due to the solid debris. Certain components in the fluid may
affect the membrane and lead to plugging, so sometimes a second membrane is used,
below the first one, to find the weight change solely due to the fluid rather than the debris
and fluid.
The gravimetric level is reported as
f.lg per 100 mL of fluid
Such a technique may not be adequate for modern systems due to the low levels of
contaminant actually collected - requiring very high accuracy in weighing if only 100 mL
of fluid is examined. It should also be stated that some 'plasticisers' or additives in the oil
could be taken out in the top membrane and hence give a misleading result.
Membrane optical density

The advantage of an examinz.tion of a membrane is that the particles will be viewed by eye
(through a microscope). Tbe overall intensity of the debris on the membrane can be
compared with a 'standard' intensity and a 'better' or 'worse' conclusion drawn; the
appearance also enables a trained operator to sense the type of debris present. For instance,
he or she will be able to answer such questions as
Are the particles shiny or dull - metallic or paint, etc.?
CONTAMINATION CONTROL 359
Are the particles smooth or rough - atmospheric or fatigue, etc.?

Are the particles transparent or translucent - quartz or rock, etc.?
What colour are the particles - rust or coal, copper or bronze, etc.?
However, the process is long and dependent on the experience and feeling of the operator
of the day. The intensity level may be reported as a direct comparison with an accepted
class, such as:
between Class x and y on a certain comparison chart appropriate to the industry, e.g.
Conpar, RN, etc.
Figure 3.3.1 shows two photographs of contaminant on membranes. One is of a relatively
dirty system (ISO 4406 Code 21/19/17) and the other of a good cleanliness system (ISO
4406 Code 16/14/11). (The ISO codes are explained in the next chapter.)
Count of particles
A count of particles is possible on a membrane The particles are sized and counted and a
value expressed as
x particles > Y I!m per 100 mL
However, there are much faster means of counting (although not giving the visual
appearance), such as automatic optical counting. These techniques may be used on-line,
and hence give an immediate display of the contamination level. Counts can be made at
a number of sizes, depending on the counter used. A disadvantage of optical counters is
that they may be confused by air or water in an hydraulic fluid - counting the break up of
water as particles. With such counters, it is important not to exceed their particle
concentration limits.
Because of the difficulty in appreciating very large numbers, the count of particles is
usually reported as a class, or classes, although the actual count should also be given.
possible classes are discussed in the next chapter, but the most common is the ISO 4406
which would be quoted, for example, as
ISO 4406 Class 21/16/12
Assessment of particle concentration

Whilst the optical devices count, or endeavour to count, individual particles one at a
time and add them all up for the volume analysed, the assessment devices are able to give
a 'count' from the final situation. One example is the filter blockage technique; this can
assess particle 'count' at the sizes of the fitted pore filters used within the instrument. It
does not have the restrictions of the optical devices and can be used in much more rugged
situations - even where fluids are mixed or are totally opaque. Figure 3.3.2 shows one
example of a filter blockage contamination monitor.
This type of contamination monitor is ideally suited for the class description mentioned
above. It is able to give the currently most valuable two sizes, that of 51!m and 151!m, and
to display the ISO 4406 as, for instance
ISO 4406 Class -/16/12
Target cleanliness levels
The level of acceptability depends on three features
the contaminant sensitivity of the components
the operational conditions of the system
the required reliability and life expectancy.
In nearly all systems, expectancy of a long life with reliable functioning, is of paramount
importance. (Target aircraft might be one exception.) Because of such a requirement it
would be the intention of the designer to try and match the expected environment and
operational condition with components which would last.
Slow speed, low pressure, systems, such as in swing bridges, have large clearances in
their control components and particles rarely would block up or jam mechanisms. In such
a case the need for tight contamination control is less critical.
On the other hand, high pressure precision operations, or those having serious
consequences should failure occur, have the requirement of the very best cleanliness
levels. Aerospace and refined servo controls would be part of this requirement.
Table 3.3.1 gives on idea of the initially quoted target levels of cleanliness for various
hydraulic components. Whilst it might be possible to give a general indication for a
system, it is preferable to know the contaminant sensitivity of individual components
within the actual system, before deciding on the overall acceptable level (however, this is
not often feasible).
In particular, component manufacturers must take great care before quoting a cleanli-
ness level requirement for a component. If the level is too clean, the user will find he cannot
achieve it, and will be reluctant to purchase the component for fear of losing his warranty.
On the other hand, if the requirement is too lenient, then the component may well fail in
use, and the manufacturer or supplier will find that he is having to replace too many
components free of charge. It is probably best for the supplier and user to agree a
cleanliness level before it becomes a requirement of a warranty.
Incidentally, it is not always easy to ensure that cleanliness is maintained in practice -
unless very regular checks are instituted with an automatic contamination monitor. It is too
easy for a filter to become blocked and remain unchanged because oflack of maintenance.
If this happens the fluid will begin bypassing the filter, or will begin shedding debris into
the system, and cause a greater problem.
Steps in contamination control

Contamination control is a terotechnological feature - it must be exercised over the total
life span of the system from design to scrap. Individual steps will now be discussed.
Design
Every hydraulic system must be designed with cleanliness in mind. There are six features
which should be considered in the design
Level of reliability and life required
Target cleanliness level
Reduction in contaminant ingression (seals, reservoir, etc.)
Component insensitivity to contamination
Sampling points
Ease of maintenance.
Required reliability and life - vary with systems. But it is important that steps are taken
to ensure that the anticipated life is achieved by suitable contamination control. Where
long lives are expected, this will involve appropriate filters, but also the programme of
regular maintenance; suitable monitoring of contamination - preferably on-line - and a
choice of components with good contamination insensitivity (for example, poppet valves
rather than spool valves).
Target cleanliness level - has already been mentioned in the previous chapter, and
greater details are given in the next chapter.
Ingression - occurs primarily at the reservoir, but also at seals and on the occasions that
the system has to be broken for maintenance. The section later on reservoir flexible
separators gives some ideas for improving the reservoir design where maximum cleanli-
ness is required.
Contamination insensitivity - means that components are less likely to fail should
contamination enter the system fluid. For instance, poppet valves are less likely to wear
or jam than are spool valves, because there is less sliding involved. Surfaces which receive
a jet of fluid, such as in control valves, may need to be hardened. Clearances need to be
as large as possible.
Sampling points - are essential for determining the contamination level at a particular
position - sampling from the reservoir is not good enough. Such points need to be in
turbulent regions so that the fluid, and contaminant, are thoroughly mixed, e.g. immedi-
ately after a bend. There are several sampling point fittings available, Figure 3.3.3 shows
the HSP fitting. It might also be considered that instead of a proprietary sampling point-
with a screw-on contaminant-sealing cap- that a small flat faced coupling be used instead;
this would not cause so much trouble where there is a high likelihood of fitters not
replaciD1! the cap on the 'proper' sampling point.
Ease of maintenance - usually means that it is easy to get to the parts needing
examination or change. Filters, for instance, are prime candidates, because their elements
have to be changed when nearing blockage. The pressure drop indicator must also be
visible, so that a fitter knows when the filter change time has arrived.
Manufacture of the system

The manufacture may be undertaken according to the design, but is each component
thoroughly cleaned and protected from the ingestion of debris during transit? In fact, are
the components tested to ensure that they meet the required cleanliness standard? Capping
of open cavities and pipes will be essential at this stage. In smaller manufacturing
workshops, it is quite common to have welding in the same building as the assembly; this
is a disaster unless complete shielding and dust extraction is provided, because the weld
'sparks' carry many metres in the air and become lodged within components.
Build and commissioning

Components need to be carefully unwrapped and fitted with the minimum of possibility
of atmospheric debris entering the system. Once built the system must be flushed through,
either with a lighter flushing fluid followed by the correct fluid, or by the correct fluid
replaced once or twice. Flushing can take many hours, even days, in the complex systems;
individual components may need to be flushed separately or protected by extra filters (or
looped out of the flow) during the flushing. It is important to have adequate filters in the
circuit-even extra ones for the flushing operation - in order to prevent contaminant being
passed round the system and dumped in the reservoir.
Because filter elements may block up quite rapidly during flushing, it will be likely that
several replacement elements will be necessary during the flushing cycle. Flushing should
continue until the contamination monitor indicates that the cleanliness level required, has
been reached.
One feature of the build, which also affects later operations, is that of filling with the
hydraulic fluid. Although 'as manufactured' fluid may be of the required cleanliness, once
it gets into a container or drum the level of contamination will go up; the fluid must be
pumped into the system through a fresh fine filter.
Operating
A regular routine must be arranged whereby the filters are checked for blockage (or near
blockage), and then replaced. Any leaks which are observed must be reported and
corrected.
Care must be taken to ensure all caps and covers are replaced, particularly on the
reservoir, and all oil is pre-filtered as it enters the system. The reservoir must be maintained
at its correct level. Air breathers, also, must not be allowed to choke up.
Care must be taken to ensure all seals are replaced if they become damaged, because,
again, they are prime places for the ingestion of contaminant.
Note must be taken of any unusual circumstances which may affect the fluid or the
system, such as, excessive temperatures or flows which could dislodge particles or oxidise
the fluid.
Contamination control requires frequent checks on the contamination levels of the
particulate within the system fluid, i.e. by taking samples from the sampling points. Should
this rise, then the system must be checked for faults. It may be due entirely to extra wear,
in which case a greater level of filtration may be required (a finer filter or more of them);
if however, it is due to ingression then the seals should be examined for wear, or the
maintenance programme improved.
Water contamination
It is possible that liquid contamination other than water is present. Certain chemicals may
have seeped in, or cross contamination between different oils may cause a residue.
However, the great majority of liquid contamination is water, except, of course, in water
based fluids; however, excessive water content may be serious and prevent adequate
lubrication and reduce viscosity. Conversely oil in excess may be a contaminant in water
based fluids causing a sticky scum which usually floats on the water causing blockage of
filters, etc.
Water is present in all mineral oils, but it is only acceptable up to about 50% of the
saturation level. Above that level corrosion may occur, and the fluid may possibly
becoming acidic; lubrication properties are also reduced
The water may be present due to condensation in reservoirs or unfilled pipes and hoses,
or may be ingested through seals and breathers.
Water content, as mentioned earlier, can be reduced through settling in reservoirs, but
if full control is necessary vacuum purifiers should be used (or possibly, water absorbing
filters ).
Air contamination
At atmospheric pressures, hydraulic oils may contain up to 8% of their volume in
dissolved air. Such dissolved air is no problem. However, as the pressure increases, the
amount of air which can be dissolved increases, and then becomes a problem when the
pressure is reduced, such as downstream of relief valves.
The presence of air bubbles (released from the dissolved state) will not only cause
erratic operation due to compressibility, but encourage oxidation of the oils and increase
corrosion. Air bubbles in the suction line of certain pumps also cause damage. Cavitation,
although not strictly air contamination, may occur where low pressures cause implosion
and excessively high surface pressures.
The control of air in a system is best undertaken by ensuring the design is correct and
there is less opportunity for air to enter. The reservoir is a key problem area to be addressed
because of turbulence, low levels and flow restrictions causing below atmospheric
pressures to exist upstream of the system pump (see chapter on Reservoirs); the flexible
separator mentioned below can provide some improvement.
Reservoir flexible separators

Reservoirs are usually fitted with air breather filters. However, such breathers provide
only limited protection against atmospheric borne contaminants entering the tank since
their cut-off point is normally no better than 15 - 20 J.Un. Certainly particles smaller than
this entering the fluid may be removed by the system filter or filters, but in systems
requiring ultra-clean fluid the conventional tank breather may be replaced with a flexible
separator to ensure that no contaminant enters the system at this point.
The principle of the flexible separator is to provide a permanent flexible non-porous
barrier between the atmosphere and the system fluid, without affecting the operational
functions of the system components. The flexible separator takes the form of a synthetic
rubber bag, totally enclosed, save for a metal stem giving access to the bag interior. The
separator is introduced into the fluid reservoir with the stem protruding through the tank
lid (see Figure 3.3.4), and, providing the tank and the breather are effectively sealed, as
the fluid level rises and falls air flows in and out of the separator only. Thus all air and its
associated contaminant is prevented from contacting the fluid.
Alternatively, where internal installation is impracticable, the separator may be
connected externally to the reservoir. In this case, air trapped in the reservoir flows in and
out of the separator as the fluid level rises and falls. Thus all air and its associated
contaminant is eliminated.
In both the described methods, the fluid in the reservoir is always at atmospheric
pressure.
A further possibility is to seal the reservoir, which is then slightly pressurised, but in this
case, the separator, installed in the reservoir, is completely enclosed and contains a slight
positive pre-charge pressure. The reservoir then functions in a similar way to a low-
pressure accumulator. This system has the added advantage of giving the pump a positive
head, but of course must be stressed with a positive pressure in mind.
For non-pressurised systems, the separator is sized on the normal fluid displacement
from the reservoir plus an additional capacity to compensate for system losses; this
additional leakage compensation should normally be not less than 25% of the displaced
volume.
In practice, the separator volume is usually quite small in relation to the tank capacity.
In hydrostatic applications, for example, the volume displaced by the cylinder rod needs
to be considered. If accumulators are being used, then their normal storage capacity
should, of course, be known.
Having over-sized the separator capacity by 15%, there is enough additional capacity
available to cater for normal system leakage. However, in the case of serious maintenance
neglect, where a system is not topped up over a long period, or in the event of catastrophic
failure when the fluid level drops dramatically, the maximum separator volume could be
exceeded. In this event, as no air can enterthe reservoir and the separator is fully expanded,
a depression in pressure within the reservoir will result, which could cause the pump to
cavitate. There are two simple and practical solutions to this problem. The first is to employ
a low level alarm system; this may consist of an audible or visual warning, which if not
acted upon is followed, after a further drop in fluid level, by automatic pump switch-off.
The second is the use of a normally sealed filter cap which has a built-in suction-relief
valve opening at 35 mbar (0.035 bar) depression. Thus, when the fluid level drops
catastrophically, air is allowed into the reservoir.
All these sytems have the disadvantage of not permitting harmful air (or gases) to be
released into the atmosphere. It may, therefore, be necessary to include a relief valve to
open when a certain pressure is exceeded.
367
CLEANLINESS CODING SYSTEMS
THE CLEANLINESS level of a fluid may be described either in comparative terms or as

an absolute. For a system which has operated successfully for some while, it may be
sufficient to compare the level of contaminant with that which has been detected in the
past. However, a higher accuracy and a greater consistency is achieved by the use of
absolute standards based on particle counts.
Counts are usually made of particles within the size range from 211mto 100 11m.In order
to cover the range perhaps six or eight sizes may be set for counting in the sense of number
of particles greater than the size. Because the numbers vary from the odd one or two up
to millions (at the smaller sizes) a coding is often adopted to provide an immediate
appreciation of the cleanliness level; the class number, or code, is then all that needs to be
expressed.
Although the ISO 4406 codes are the most quoted for hydraulic systems, both they and
the AS 4059 are internationally accepted codes. There are other standard classes which are
used occasionally.
ISO 4406: 1996

Figure 3.4.1, is the ISO 4406 - Code for defining the level of contamination of solid
particles. The objective of this two- or three-number coding is to show the level of solid
contamination at two or three sizes of particle, i.e. at (211m), 5 11mand 15 11m.Each scale
number of the code is increased by 1 when the level of particle count approximately
doubles. Table. 3.4.1 gives the actual particle counts associated with each scale. Originally
the count was associated with 100 mL, and the count was considered 2" where n was the
scale number, but this was changed to per 1 mL in 1987 and hence mathematically has to
be multiplied by 100.
Thus
Count 'up to and including' x 100 = 2"
where n = scale number
In fact the count has been rounded off and hence the numbers given in ISO 4406 (Table
3.4.1) are not precisely those obtained from the equation.
Because not every application requires all three scales, or perhaps the monitor is unable
to measure them, there are three variants on the three figure code:
a) e.g. */19/14, the * meaning that there are too many particles equal to
or greater than 2 !!m to count.
b) e.g. -/19/14 , the - meaning that there is no requirement to count
particles equal to or larger than 2 !!m.
c) e.g. 22/19/14, meaning all three sizes have been counted.
CLEANLINESS CODING SYSTEMS 371
It is recognised that at the very clean levels, the number of particles being counted, in even
100 mL, are too few to be statistically reliable. Therefore if less than 20 particles are
counted, the actual quantity of fluid used and the number counted should be quoted, thus
"2 particles counted in a total of 200 mL".
NAS 1638 (National Aerospace Standard) - SAE AS 4059

The NAS 1638:1964 contamination classes, modified (or extended) by AS 4059:1990,
cover much more detail than the ISO 4406 codes. (AS refers to the Aerospace Standard
of the Society of Automotive Engineers.) Although they do not cover such a wide range
of bands - there are only 15 classes - the counts are made from 2 11mto 50 11m. They are
also stated as counts per 100 mL. Table 3.4.2 shows the counts associated with this
standard in the same way as the ISO 4406, i.e. in number of particles greater than the size
given. (The original NAS 1638 standard gave the number of particles per band, such as
between 15 11mand 251lID and it went from 5 IlIDto 100 11m.)
The graph of these NASIAS classes is shown in Figure 3.4.2. The International
Standards Organisation has endorsed the SAE AS 4059 in its one page ISO 11218: 1993
standard.
SECTION 4
373
LINEAR CIRCUITS
THE OBVIOUS aim of hydraulic circuit design is to provide the functions required from
the system in an unambiguous way. It is equally important that the efficiency of the circuit
should be as high as possible. This principally involves a match of pump flow and/or
pressure to be as close as possible to the circuit requirements at any instant. It is also
preferable to avoid any potential sources of excessive fluid friction and pressure drop. At
the same time components must be 'protected' where necessary.
The majority of the circuits described in this chapter give a fixed configuration to
perform specific tasks and give appropriate solutions to a variety of circuit control
requirements. There are also many proportional and electro-hydraulic circuit solutions to
fulfil these requirements but with greater flexibility to adapt to provide solutions to other
tasks with the same hardware. The examples quoted here are all based on linear circuits,
with an hydraulic cylinder as the output actuator. Many of the ideas are also appropriate
for rotary actuator output in either the semi-rotary or motor form.
Directional control of movement

The control of travel direction, extend or retract, is the most basic requirement in linear
circuits. The DCV (directional control valve) used for this has only an on/off function and
cannot modulate the amount of flow. The DCV function can be performed either with a
spool valve or a combination of poppet valves.
Single acting cylinders

Single acting cylinders may be extended with hydraulic flow but require another source
of force to provide movement in the retract direction. Basic control circuits for a single-
acting cylinder are shown in Figure 4.1.1, using 2- or 3-way, 2-position valves (DCV).
Figure 4.1.1 a also shows some common alternative means to operate the valve to extend
the piston with a spring return action. The cylinder is also sprung loaded to retract, whereas
in Figures 4.1.1 band c, an external force is required ego a gravity load. If the piston needs
to be held in an intermediate position, then a 3-way selector valve with 3-positions must
be used as shown in Figure 4.1.2. This could be substituted in the circuits for both Figure
4. 1.1a and b but not that of Figure 4.1.1c.
position is required, as in Figure 4.1.3, or a 4-way, 3-position selector (DCV) for
intermediate hold as in Figure 4.1.4. It should be noted that the flows either side of the
piston will not be the same but in the ratio of the piston and annulus areas.
If the main control valve is a spool design then further refinement of the circuits above
may be necessary if the piston has to be held in position under load for long periods because
of cross-port leakage in the valve. Check valves (pilot opened) could be included to hold
a position. Alternatively these control functions could be carried out by using pilot
operated poppet valves throughout, as shown in Figure 4.1.5. Since poppets are positive
seating valves they give the necessary sealing and if the pilot operation allows (four
independently controlled) they may be switched to give alternative sequencing depending
on the requirements. r--------------------------------------------- ----- ;
! fun._-- -...---.-; r-__
nn_nn .hu __
nn __
: :
Cylinder speed controls

Simple speed control of cylinders can be realised with an adjustable restrictor or flow-
control valve in either the supply or exhaust line as shown in Figure 4.1.6. Speed control
is effective for both directions of motion but forward and return velocities will vary in
relation to the different effective piston areas and the load pressure across the piston
determining the restrictor pressure drop.
If the restrictor is placed in the return line after the directional-control valve, the piston
is 'braked' between the supply and the return line pressure, preventing the piston running
away. This is called meter-out control. Meter-in speed control, Figure 4.1.6b, can be used
where the cylinder is always under load on the controlled stroke. If the load is highly
variable or has a tendency to run away, then meter-out speed control is generally preferred.
Speed control can be applied to either the extend or retract stroke, the metering valve being
on the appropriate 'in' or 'out' line respectively. However, meter-out control on the rod
side can lead to very high pressures occurring as a result of the pressure intensification
effect of the piston areas. Figure 4.1.7 shows a combined throttle and check valve giving
meter-out control for the retract stroke and free flow in extend. The flow through the valve
and hence piston speed will still be dependent on the pressure drop across the metering
orifice, and if the load is changing this may not be satisfactory. The use of a pressure
compensated flow control will overcome this since a constant pressure drop is automati-
cally maintained across the metering orifice.
These straightforward control systems suffer from relatively high energy losses (and
high fluid heating) since the pump is operating at maximum pressure all the time and
excess flow is discharged through the relief valve. Such losses can be minimised by
adopting bleed-off or by-pass speed control .
In the bleed-off circuit in Figure 4.1.8, the throttle valve is located in the line to the
cylinder, by-passing to the reservoir. This valve is significant only when that particular
line is pressurized, and in this case all throttling losses are linked to this bleed flow and the
pump pressure determined by the load. The main disadvantage of such a system is that flow
control (speed control) is indirect and will vary as the actual pressure drop across the
restrictor varies with load. For this reason the by-pass flow control (sometimes called 3-
way flow control) valve of Figure 4.1.9 is commonly preferred where losses are to be
minimised. A variable restrictor is used in the main line to sense the flow rate and the by-
pass flow set accordingly to maintain a constant pressure drop across the restrictor.
This gives the same positive speed control as meter-in or meter-out. Both the bleed-off
and by-pass circuits are basically meter-in systems. They would have no effect on cylinder
speed if connected to the exhaust line from the cylinder. They can only be used where a
single service is being operated at one time since the pump pressure is set by the load and
can only take one value.
Rapid-motion valves
Rapid-motion valves are designed to provide a change-over between rapid and slow
forward motion automatically as piston load increases. One possible circuit is shown in
Figure 4.1.10, where at low pressure both sides of the cylinder are connected together by
the rapid motion valve. This uses the return flow from the rod side in addition to the pump
flow to give a high speed but with a lower force capability since the pump pressure acts
on the differential piston area. After a pre-determined pressure has been reached the
control piston of the rapid-motion valve switches to the through-flow position and full
pressure is applied to give the piston working stroke at the pump flow. This through flow
378
position must be maintained to retract the piston. There are various possible modes of
working depending on the rapid-motion port inter-connections.
Synchronisation
When two or more cylinders are connected in parallel to the same selector, the cylinder
with the lighter load will move first, the pump flow always finding the easier lowest
pressure path. When the first cylinder has completed its stroke the next lightly loaded
cylinder will then start to move, and so on. Even if the loads are meant to be equal they
are never likely to be exactly so, and this form of sequential motion will always occur.
Synchronisation of cylinders can be achieved mechanically (very rigid structure), electri-
cally (with feedback to electro-hydraulic valves), or hydraulically by suitable circuit
design. Only hydraulic synchronisation will be considered here.
Where the cy linder loads do not differ appreciably, satisfactory synchronisation may be
achieved simply by incorporating restrictors in each of the cylinder lines, adjusted to
provide synchronous speeds of operation as in Figure 4.1.11. The accuracy of synchroni-
sation obtained depends on the loads remaining constant and even then may not be good.
Series connection of cylinders is another simple method of synchronisation which

offers an improved degree of synchronisation. The alternati ve connection possibilities are
shown in Figure 4.1.12. It is best if the cyIinders are both of the through-rod type, otherwise
synchronisation will only be achieved if the differential (annular) area of the first piston
is equal to the full piston area of the second cylinder. A basic limitation with this
arrangement is that each cylinder in the system must be proportionally smaller in size.
Better synchronisation of two equal-size single-rod cylinders is possible if one cylinder
is retracting when the other is extending as in Figure 4.1.12c.
Synchronisation can also be achieved with flow dividers and two typical circuits are
shown in Figure 4.1.13. In all such circuits, even if they are initially synchronised quite
accurately, progressive loss of synchronisation will develop due to internal leakage. It is
thus necessary to include some method of re-synchronisation in the circuit. This can be
manual, via two additional valve positions enabling each cylinder to be bottomed-out in
turn, or automatic. An automatic re-synchronization circuit is shown in Figure 4.1.13b.
Here the cylinders are synchronised at the end of each stroke when the mid position is
selected.
operating at the lower load and, being mechanically coupled, provide equal flow in each
cylinder line. Additional restrictors switched across each motor permit flow to pass once
one cylinder has bottomed-out, thus allowing the cylinders to re-synchronise at the end of
stroke.
Pressure control
There are a number of pressure control valve options which are beneficial in linear circuits.
Figure 4.1.15 shows the use of a pressure reducing valve to limit the maximum pressure
experienced at the second cylinder. It should be remembered that deriving large flows via
a reducing valve is not to be recommended because of the inherent inefficiency as a result
of the pressure drop across the reducing valve itself.
The other widely used pressure control valve is the combination of check and relief
valve known as a counterbalance valve. Figure 4.1.16 shows two counterbalance valves
with remote pilot operation of the check stage. In normal operation from the DCV the
check valves are both opened allowing free flow in both directions. If there is a change in
the direction of the action of an extemalload, such as might occur if a linkage goes over-
centre, then the cylinder might have to restrain the movement. When this occurs the
LINEAR CIRCUITS 381
pressure on the drive side will fall and the piloted check valve will close; if extending in
the figure then the pressure at A falls when the load starts to pull and the check valve on
the rod side will shut. The load must then work against the pressure set on the relief valve
to continue to move and this gives a braking effect.
This circuit also provides a useful function when a load with considerable momentum
is brought to rest at an intermediate position. This momentum might arise from the load
speed or its mass or their combination. If such a load is brought to rest very suddenly when
a DCV is shut, then excessive pressures might result. In the circuit shown a high pressure
on either side of the piston will cause the relief valve to open and limit the pressure rise.
The load will over-run but no damage will occur. Note that it is also necessary to provide
flow into the other side of the cylinder when this over-run occurs to prevent cavitation or
aeration. This is achieved with the check val ve function provided there is a source of flow
available as with the open centre valve.
Sequencing
A common requirement is for two or more cylinders to operate in a pre-determined
sequence to ensure one motion is complete before the next starts. There are various ways
in which this can be done. If the sequences are relatively straightforward then mechanical
sequence valve switches or pressure sensing valves can be used. Figure 4.1.17a shows a
single flow-sequencing circuit which provides for both cylinders to extend simultaneously
but delays retraction of the upper cylinder until the lower has completed its retraction
stroke and operated the sequencing valve via the mechanical trip. The sequencing valve
could be placed in the return line, instead of the forward line to the upper cylinder, the
upper cylinder would then be locked in the extended position until released by retraction
of the lower.
A simple system of double-sequencing is shown in Figure 4.1.17b, again with the
sequencing valves in the forward flow lines. This would allow, for example, a sequence
operation of undercarriage doors opening and closing, with under-carriage retracting and
extending. Again the sequencing valves could control return flow rather than forward flow
for positive (hydraulic) locking of one cylinder.
Pressure sensing sequencing circuits avoid the mechanical and hydraulic connections
necessary for the above circuits. However, they are more prone to malfunction if an
unexpected 'high' pressure trips the sequence prematurely. Two examples are shown in
Figure 4.1.18 to give the operation of a second cylinder once pressure on the first has
reached the preset level. This need not correspond to a fixed position of the first cylinder.
More complex multiple sequencing is usually more conveniently controlled by electri-
cal switches operating solenoid valves or via a PLC. Mechanical sequencing may be
preferred, when mechanical locks can readily be incorporated as part of the system rather
than as separate items.
Energy efficient pump circuits

There are a number of techniques which can be used with basic fixed displacement pumps
to improve energy efficiency. There is also always the option to use a pressure compen-
sated variable displacement pump if more appropriate.
A simple circuit which uses twin pumps to match the combination of high flow at low
pressure with low flow at high pressure is shown in Figure 4.1.19. The lower pressure
pump is automatically off-loaded to tank when the system pressure reaches the unloading
valve setting. The remaining pump will deliver its flow to the system until the main relief
valve setting is reached. In a typical sequence the piston will move with high speed until
the load begins to increase and the LP pump is off-loaded.
If the product of total flow at the unloading pressure is the same as the product of the
HP pump flow at the relief pressure then the system will be matched in power terms to the
pump driver attwo conditions. This idea can be extended to three or even four pumps using
different unloading valve settings.
Accumulators can also be used for energy saving and a basic circuit is shown in Figure
4.1.20, where the accumulator is first charged by the pump and then the pump shuts down.
Circuit pressure is supplied from the accumulator until the pressure drops to a predeter-
mined minimum, when the pump is cut in again. The control element used is a pressure
switch which senses the accumulator pressure and controls the electric motor switching.
This stop-start circuit can provide substantial savings in energy where the hydraul ic circuit
is used very intermittently.
Another energy-saving circuit is shown in Figure 4.1.21 where a 6-way, 3-position
selector valve allows the accumulator to pressurise the circuit with the selector in the
neutral position. This can provide 'hold' conditions for long periods, during which the
pump can be off-loaded as shown. In extreme cases it could be combined with the previous
circuit for intermittent duties i.e. the pump re-charging the accumulator during idle periods
and then being switched off by a pressure switch.
385
ACTUATOR PERFORMANCE
THE BROAD definition for hydraulic actuators covers linear actuators (single- and
double-acting cylinders); semi-rotary actuators; and hydraulic motors (full rotary actua-
tors) used specifically for actuator duties as opposed to motor drives. Components
normally involved are hydraulic cylinders and the treatment given in this chapter
concentrates on these devices. The ideas are also generally appropriate for application to
both semi-rotary actuators and motors with the modifications indicated at the end of this
chapter.
As well as extending the theory used elsewhere in this book this chapter acts as a
summary of the major formulae associated with actuator performance. Unlike the other
sections in this book reference is also made to other unit systems.
Linear actuator basic speed relationship

The movement of a piston will be determined by the flow supplied if transients associated
with pressure changes are neglected (this qualification is discussed later in this chapter).
In the case of a cylinder supplied from a pump with a constant flow Q, then the speed U,
can simply be defined:
It follows that this relationship also provides a means of calculating the pump delivery
required to achieve a given speed or time of operation. Working formulae which are cast
to give the operating time of a cylinder are given in Table 4.2.1. Suitable rearrangement
will give the pump delivery necessary. It should be noted, however, that these are ideal
figures and may be modified in practice as described below.
This assumes that the acceleration (and deceleration) transients are negligible, which
is generally true if the system has low mass, low maximum speed, long stroke, or an
appropriate combination of these. The speed is thus controlled by the pump flow and may
be calledflow limited. In the flow limited case increasing the relief valve setting would not
change the piston speed. Nor would a reduction in setting unless it was reduced to a very
low level when the piston would slow as some flow passed through the relief valve in
which case the flow limit assumption is no longer valid.
There is an alternative to the actuator speed being established by a flow limit, this is
called pressure limiting. This is the more usual case when the speed is controlled by
significant pressure drop across a throttle or a proportional valve. Other pressure drops in
the flow path may be considered including the throttling effect of the inlet and outlet ports
of the cylinder, back-pressure in the return and other pressure losses in the pipework.
Reducing the relief valve setting as described above reaches the lower limit as a flow case
when the relief valve pressure becomes equal to the pressure drops within the circuit.
There is simply insufficient pressure to push the flow around the circuit and the fluid takes
the easier path through the relief valve. It then becomes a pressure limited case.
The above analysis can easily be modified to determine the maximum speed when
pressure limited, by calculating the maximum flow and substituting this for the pump flow
This shows clearly that the speed can be changed by both the throttle valve opening and
by the supply pressure. It is in this sort of circumstance that "fiddling with the relief valve"
can change the piston speed but this is not the recommended method of adjustment once
an operating pressure has been selected. This equation is appropriate for either meter-in
or meter-out control. The example in the introductory chapter on The hydrostatic principle
and basic circuits includes both flow limiting (in the extend direction) and pressure
limiting (in the retract direction) where the "supply" pressure is established by the load
weight.
Note that the case of a pressure compensated flow control valve is analogous to the
pump flow case where a 'fixed' flow will be delivered irrespective of load and other
pressure drops. This then can be considered as the flow limited case within the range of
normal valve operation.
Basic linear actuator force analysis

The equations above are modified by the inclusion of a pressure difference across the
actuator piston but the principles established remain unchanged. The maximum capability
of a linear actuator may be represented by a constant load moved with constant velocity
over the whole of the stroke, with the resulting load-travel diagram shown as the horizontal
line in Fig 4.2.1. This is seldom realised in practice since the usual motion includes
acceleration from rest which mayor may not be followed by a period of uniform velocity
and possibly deceleration approaching the end of the stroke with a cushion or other means.
An actual load is more likely to take the form of the curve shown in Figure 4.2.1. The
area under this curve represents the actual work done, which is obviously less than that
under constant load conditions. In other words the actual work done by the cylinder is
usually less than its potential capacity.
The force necessary to accelerate (or decelerate) a load is obtained from Newton's
Second Law as the product of the mass M, and the acceleration a:
F=Ma
There are other components of force which must also be considered and strictly the force
above is the net force balance; a sum of all the components along the cylinder axis.
Friction will frequently be present, which will be added to the total force required to
accelerate a load, or subtracted from the total force required to decelerate a load. The
frictional forces inherent in hydraulic cylinders are usually small as discussed in that
chapter, and frequently only the frictional forces associated with movement of the load
itself are significant.
Where the load is moved vertically, the load itself represents a further force to be
overcome (for upward movement) or assisting the actuator force (for downward move-
ment). Since the friction force mayor may not be of the same sign as the weight, the four
possible conditions are detailed below for clarity:
Load movement Force equation
accelerating upwards Fh - Ff = M (a + g)
decelerating upwards Fh + Ff = M (a + g)
accelerating downwards Fh - Ff = M (a - g)
decelerating downwards Fh + Ff = M (a - g)
Where Fr = the friction force

Fh = the hydraulic force (i.e. normal cylinder force)
and consistent units are used.
Where the subscripts 1,2 depend on the direction. This gives the force in N if the
pressures are in MPa and the areas in mm2
One further case is sufficiently common to be considered. That is where hydraulic force
generated by the actuator continues to be applied during deceleration of the load. This is
the case with a cushioned cylinder approaching the end of its stroke or where separate
cushioning devices are used to decelerate the moving load. For horizontal movements, the
total force involved during deceleration then becomes:
Cushion force = Ma - Ff + Fh
For vertical movements this is again modified by the weight of the load as above. This
total force represents the force present in the cushioned side of the cylinder. The cushion
ACTUATOR PERFORMANCE 389
stroke is usually quite limited (and fixed); the higher the initial speed the higher will be
the required deceleration and may reach the limits imposed by the construction of the
cylinder. Excessive velocities may be damaging, or may affect cushioning suitability.
Thus, in general, if the intended operating velocity of a hydraulic cylinder exceeds 0.5
m/s the supplier should be consulted regarding the suitability of the cylinder and cushion
(where applicable) for these piston speeds.
Combined analysis for linear actuator

The hydraulic force acting will depend on the pressures acting on the piston and this will
depend on the pressure drops as a result of flow in the other circuit components. This in
turn will change as the load speed (and hence flow) changes. This transient situation can
be accounted for with an equation which also includes the pressure drops. This is again
developed for valve control but could be extended to include any other pressure losses.
Figure 4.2.2 shows the general configuration and defines the symbols used.
Compressibility 'delay'
The effects of fluid compressibility have been neglected in the analyses presented so far.
In many instances this is quite valid since the transient times associated with load
accelerations are usually long in relation to those associated with compressibility and even
these can frequently be neglected. A more detailed analysis can obviously include these
effects but it is beyond the scope of this book. However, an estimate can simply be made
of the delay occurring before pressures in a system have reached a new steady value. This
can be compared with other time intervals occurring and will indicate whether a more
detailed consideration is required.
In this case there is only one oil column which changes pressure as the piston moves
since the rod side is always atPs. The final case is the equal area, through-rod design, where
the above equation still applies but with Al =A2. In this case stiffness will be a minimum
when V I = V 2, and the piston is in its mid position, giving:
Strictly speaking the volume of the fluid column lengths in the lines to the cylinder
should also be taken into account. This may be significant if the control valve is remote
from the cylinder when the total volume should be used. The connecting stiffness between
both the body of the cylinder and its mounting, and between the rod connection and the
load may also be significant. If these are too 'flexible' then they may form a lower limit
to performance than that of the fluid column stiffness.
Semi-rotary actuators
The formulae for semi-rotary actuator designs based on geared cylinders can be adapted
from those given above. It is possible to modify the equations to deal with torque and
angular motion or to refer load detail back through the effective radius and deal with the
linear piston motion as given. In either case some estimate should be included for the
frictional losses in the internal gearing or linkage. In the case of vane-type semi-rotary
actuators, the output torque is proportional to the effective vane area, the only frictional
losses are those of the seals (usually negligible) and the output shaft bearing (s). The
equations are most appropriately cast in the rotational form.
399
HYDROSTATIC TRANSMISSIONS
THE PRINCIPAL components of an hydrostatic transmission are a primary unit (a pump)

and a secondary unit (a motor) with pressure energy transmitted between them. Pressure
energy is converted into torque in the motor and as rotary motion takes place so flow is
taken by the motor and power transmitted from the pump driver to the motor load. Both
the pump and motor are positive displacement machines, they may be fixed or variable
capacity, and most of the designs of hydrostatic units described in Section 2 could be used.
The power transfer uses the pressure energy of the fluid by hydrostatic reaction, with
no significant change in the velocity of the fluid. This is in direct contrast to a hydrody-
namic (or hydrokinetic) transmission such as an hydraulic coupling or torque converter,
where it is the change in velocity of the hydraulic fluid in the circuit which is responsible
for the transmission of energy. A further general distinction between the two is that, for
a given power, the energy is transmitted by a small flow of fluid at high pressure in the case
of hydrostatic transmission, and by large fluid flows at comparatively low pressure in the
case of hydrodynamic systems.
The potential advantages offered by hydrostatic transmissions are listed below. The
extent to which these characteristics will be achieved will depend on the configuration and
detail design of the transmission.
(i) Stepless change of ratio between pump drive and motor output.
(ii) Zero output speed and start from rest without a drive clutch.
(iii) Forward and reverse control.
(iv) Dynamic braking and load over-run control.
(v) Simply set limits for acceleration or deceleration.
(vi) Simple and reliable method of stall protection via pressure-relief valve.
(vii) Potential to optimise the speed and power range of the pump drive.
(viii) Considerable flexibility in the positioning of pump (s) and motor (s).
(ix) Good speed control including creep speeds.
(x) Rapid response to demand changes.
(xi) Shock and vibration isolation between drive and load.
All hydrostatic transmissions have a pump supplying flow to a motor but there are two
fundamental ways in which the return flow from the motor can be connected and this gives
rise to two very different categories of transmission. When the motor outlet is returned to
the system tank it is an "open" configuration as in Figure 4.3.1, but when it is returned back
to the pump inlet then it is called a "closed" configuration as in Figure 4.3.2. In some texts
the terms "open loop transmission" and "closed loop transmission" have been used. This
is ambiguous, potentially causing confusion with the meaning of open and closed loop in
the sense used by control engineers. The current trend for greater use of electrical closed
loop operation, with some form of feedback signal, in association with hydrostatic
transmissions increases the potential for confusion. In keeping with current practice, and
with the previous editions of this book, the terms "open-circuit" and "closed-circuit" are
preferred for use here.
units could build-up causing seal or even structural failure. It also shows a relief valve from
the high pressure line to tank. This performs its usual safety function of limiting the
maximum pressure in the system which determines the maximum torque on both the pump
and motor. The pump torque limit can be used to prevent the prime mover stalling. The
limiting motor torque will determine the maximum transmission output and is available
through the operating speed range. However, there may be a minimum low speed from
which this torque is available without excessive ripple. This will depend on the motor type
and its design, some providing close to the limiting value from rest.
Open circuit transmissions are most used when the motor is working against high
friction or purely dissipative loads with low inertia, such as a fan drive. They are also
appropriate in systems where a single pump is used to supply multiple services, maybe
some linear outputs as well as rotary. Each service would have its own control valves, and
systems can be designed with sufficient pump capacity to operate several services
simultaneously.
The main limitations for open circuit transmissions come either from an over running
inertial load or from a requirement for braking. In both cases the motor will try to rotate
faster than its supplied flow permits. This will cause low pressure in the motor inlet line
and either aeration or cavitation or both. In such circumstances it is necessary to increase
the pressure on the motor outlet above that on the inlet and this will give a torque opposing
its rotation It is possible to modify the circuit to include restrictors or braking valves to
provide this. It is also possible, by connecting the pressure line between the pump and the
motor through a check valve, to allow the motor with its drive inertia to freewheel, either
to rest or down to that speed matched by the pump flow. Care must be taken to prevent a
surge in the drive - a possibility to which this form of transmission is more prone.
Cooling may not be necessary if adequate heat exchange is provided with the reservoir
and if it is of a suitable size. If not, then a cooler can be included generally in the return
flow to tank from the motor. This is also the most likely location for a full flow filter with
a strainer on the pump inlet. Full flow filtration on the pump outlet is another possibility,
HYDROSTATICTRANSMISSIONS 403
requiring a suitable high pressure filter element and housing and this would provide
greater protection for the motor.
Closed-circuit hydrostatic transmission

In a closed-circuit hydrostatic transmission one pump is frequently associated with a
single motor, the two being directly connected in a closed hydraulic circuit (Figure 4.3.2).
As in the open circuit design, fixed displacement units can be used to give a fixed ratio
drive. However, it would be possible for either line 'A' or line 'B' to be at high pressure
and thus the system will cope with the over-running and braking conditions.
Additionally, with a closed system, the circuit is symmetrical and flow could pass from
the pump to the motor via line 'A' or via line 'B', and hence drive the motor in either
direction. If a variable displacement pump is used which is of an over-centre design, as
shown diagrammatically in Figure 4.3.4, then both the speed and direction of the motor
can be controlled from a pump rotating in one direction. For this to work then the high and
low pressure pipes, or lines connecting pump and motor must again be inter-changeable.
It also follows that if the primary and secondary units can operate either as a pump or
motor, their roles can change and dynamic braking is possible as discussed below as part
of dynamic response. Thus the specific advantages of a closed-circuit over an open-circuit
transmission are that it gives the features of reversibility and braking more readily. It is
thus more suited to loads with significant inertia or where loads may be over-running and
change direction from opposing to assisting the motor rotation.
It is still necessary to protect the pump drive and the components in the circuit with a
relief valve. If it is likely that either line of the transmission could be at high pressure then
both must be protected and cross-line relief valves must be used as shown in the circuit
diagram of Figure 4.3.5. Note that the flow through one of these relief valves passes back
to the pump inlet side and is not lost from the circuit by returning back to tank. A single
valve would be sufficient when it is known that only one line will be at pressure.
The equations quoted above for motor speed and torque are still appropriate with the
proviso that it is now a pressure difference across the motor which balances the load
torque. It is also still necessary to include the efficiency terms. However, there is a further
consequence of leakage in that this now represents a loss of fluid from within the closed
circuit and obviously this cannot be sustained for any useful operating period. A make-up
flow must be supplied to ensure that the transmission lines and both pump and motor are
filled. This is done by the addition of a boost or charge pump connected from the tank to
supply the lower pressure line of the transmission. Since this may change between A and
B, the circuit used allows for this and automatically connects the supply to the lower
pressure line. A typical circuit, shown in Figure 4.3.6, uses a fixed capacity pump working
with a relief valve to provide a constant pressure supply usually with Pboost at about 1 to 2
MPa (10 to 20 bar). If line A is at high pressure then the boost circuit will supply flow to
line B via the check valve to keep the pressure at a minimum value I'boost. Ifline B is at high
pressure then the lower check valve will be shut and the boost supplied to line A. The
pressure difference, determining the motor or pump torque, is thus (PL -I'boost).
Closed circuit transmissions are inherently sealed systems and this can give advantages
for operation in some special circumstances. High humidity environments with dust and
dirt can be accommodated without special precautions but operation in some corrosive
atmospheres or even totally immersed is possible without extensive modification. Most
transmissions meet the requirements for Intrinsic Safety and many pump and motor
designs can be used with fire resistant fluids and hence in hazardous environments.
Transmission configurations
Hydrostatic pumps and motors can be coupled together to form a transmission in the
following basic combinations:
(i) a fixed-capacity pump and a fixed-capacity motor;
(ii) a variable-capacity pump and a fixed-capacity motor;
(iii) a fixed-capacity pump and a variable-capacity motor; and
(iv) a variable-capacity pump and a variable-capacity motor.
HYDROSTATIC TRANSMISSIONS 405
The first two types listed have been discussed above in the context of both open and
closed circuit layouts. The simplest fixed ratio transmission, type (i), confers only some
of the advantages in the list at the beginning of this Chapter and is really suited to an
application where continuous running occurs for relatively long periods. In any starting
transient whilst the load is accelerating excess flow from the pump must pass through a
system relief valve. In this acceleration phase the motor will experience a constant torque
but once at speed then the load torque will establish the required system pressure (provided
that it is below the relief setting). This fixed ratio transmission is not appropriate for stop
start type application. Also the majority of benefits in the initial list require the inclusion
of variable capacity units. The type (ii) transmission offers the ability to give variable
speed, a similar constant torque limit when operating at relief valve pressure, clutching and
a reverse ratio making it suitable for a wide range of applications and hence the most
widely used of these four.
A simplified circuit for the type (iii) layout with a fixed-capacity pump and variable-
capacity motor is shown in Figure 4.3.7. The variable displacement setting of the motor
again gives an adjustable ratio. It is usually designed with a larger capacity motor than
pump, which means that when at maximum displacement the motor runs at a slower speed
than the pump. Reducing the displacement setting of the motor increases the output speed
in inverse proportion to the setting but decreases the output torque in proportion i.e. at a
fractional displacement of x= 0.5, the speed will have doubled and the torque hal ved. Since
the product of these is power then the result, neglecting losses, is constant power over the
motor speed range. This range for continuous operation is usually limited by a minimum
displacement setting for the motor of around 0.25. The system becomes rapidly more
inefficient for lower settings. This form of motor or secondary unit control is very relevant
in high efficiency ring main type systems described in more detail below.
The fourth variant a variable-capacity pump and a variable-capacity motor is the most
versatile, and a simplified circuit is shown in Figure 4.3.8. This arrangement combines the
constant power and the constant torque characteristics of items (i) and (iii) with the
clutching and reversing characteristics of item (ii). It is especially suitable for drives
requiring a wide range of speed and torques. The range is again limited by a minimum
displacement setting for the motor of around 0.25. The output speed from this system
would normally be operated through the full range of displacement settings for the pump
(from 0 to 1) with the motor at its full displacement setting. Once the pump reaches its
maximum setting then the motor displacement can be reduced whilst transmitting full
power with increasing speed and reducing output torque. The characteristics of these
configurations can be compared as shown in Figure 4.3.9.
Typical specification and practicalities

The overall efficiency of a transmission can be determined from the ratio of the output
shaft power to that at input. This is the same as the product of the individual efficiencies
for the pump and motor as follows:
It should be noted that all these terms vary with speed, pressure and displacement setting
and that very few transmissions will operate under constant conditions. It is therefore
misleading to quote a single efficiency figure, the efficiency will depend very much on the
operating conditions and will change through the duty cycle of the transmission. A typical
efficiency characteristic for a transmission is shown in Figure 4.3.10 on an output torque

speed map.
Boost supply
A closed circuit transmission requires a boost or charge pump. It will be sized not only
to make up the anticipated leakage from the main pump and motor but also to provide an
interchange of fluid between the transmission circuit and the reservoir. This provides both
cooling and cleaning functions and the flow is encouraged by a purging circuit which
bleeds off flow from the low pressure line of the main transmission circuit. An example
is shown in Figure 4.3.11. The power consumed by this circuit should also be included in
FIGURE 4.3.11 - Typical purge circuit to improve fluid interchange with reservoir.
any estimate of the overall efficiency of the transmission. For this reason the size of boost
pump and the purge flow are kept to a minimum, consistent with system reliability.
Typically a boost pump will provide from 15% to 30% of the full primary pump flow. It
is usually the more arduous duty cycles with significant periods at high power which
require the highest flow, where the boost power consumed is then a low proportion of the
transmitted power. Strictly the boost supply should be sized on the worst case total leakage
for all the hydrostatic units (pumps and motors) connected in the closed circuit summed
with any other bleed or auxiliary outflow. A useful rule ofthumb is to take 10% of the total
summed capacity of the hydrostatic units but this may not be the worst case and this
condition should be checked.
The boost pump itself will commonly be a gear or gerotor type pump or sometimes a
vane type. It will often be built into the body of the main pump operating from a common
drive shaft with the necessary valves for the boost circuit and also cross line relief valves.
It is common to include the necessary purge circuit components but packaged within the
motor casing.
Fluid conditioning
It is also usually necessary to provide a heat exchanger to cool the oil returned to the
reservoir. This flow may first be passed through the casings of both pump and motor
providing both cooling and contaminant flushing. This cooling flow is easy to provide and
prevents any local overheating at the point of actuation, an advantage in some applications.
The cooler may be either of the radiator or water heat exchanger type as described in
Section 2.
The main system filtration is also provided in the return flow path sometimes via the
purge flow and sometimes from the boost circuit relief valve. In both cases it would usually
be placed upstream of any cooler in order to protect the tubing within the cooler. Another
alternative is to filter all the flow on the boost pump output. It is also usual to operate the
boost with an inlet strainer to prevent larger particles entering the circuit. It would be very
unusual to put a filter in the main transmission loop even though appropriate filters are
available. The aim is to maintain a low contaminant level overall and it should not be
necessary to filter all the fluid on a continuous basis.
Hydrostatic unit selection

There are choices to be made when selecting components for a transmission between the
various designs of pump and motor. The choice between swash plate and bent axis pumps
was discussed in Section 2. In many cases, the final drive requirements for speed and
torque will determine the type of motor required. In applications calling for high torques
at slow speeds, however, there is often a choice between a slow-speed high-torque motor
and a high-speed motor driving through a reduction gear. Low-cost gear boxes sometimes
render the latter financially attractive. Generally, however, it is more economical to
employ slow-speed high-torque motors and confine high-speed motors to high-speed
applications. The main reasons are higher overall efficiency, more effective break-away
torque and less space taken up. Both types are capable of rotation at very low speeds of
1-10 rev/min, with very little torque fluctuation.
There are many additional features which can be relatively easily included with a
transmission. For example in mobile systems it is possible to ensure that the pump is off-
loaded during an engine start and there are many control options as described below.
Variable unit control

The main description for variable displacement control of pumps is gi ven in Section 2 but
factors relevant to transmissions are highlighted here. Aspects of control are reviewed
initially in terms of the physical control mechanism and then considered in terms of the
control strategies.
Direct operator control of the displacement setting of a pump is possible through a lever,
hand wheel or lead screw and this is common for smaller units. For larger units it is usual
to provide servo-assistance with a position control system to give the required displace-
ment setting. The input may be either mechanical, pilot hydraulic, pilot pneumatic or an
electrical signal. For the first three the position control feedback is most likely to be
mechanical. This is also possible if the demand is electrical but it is more likely that there
will be a position transducer giving a feedback signal for electrical closed loop operation.
In all cases it is possible to tailor the shape of the characteristic between the input signal
and the resulting pump displacement. This may be linear giving an approximately
proportional output speed or non-linear with lower gain and sensitivity at the low control
inputs giving better "feel" to adjust low speeds more easily. Frequently all of these
possibilities will be available from the manufacturer but there are many alternative
methods used by the different manufacturers to implement these functions.
The main control strategies relevant to transmissions include - constant pressure,
constant flow, load sensing, and constant power. The first two of these may seem to offer
only the features that are easily obtained from a fixed displacement pump, however, they
achieve these functions with potentially greater efficiency if the duty cycle is appropriate.
Constant pressure control using a pressure compensated variable pump gives the same
stall protection for the prime mover but with a reduced flow (and power) consumption to
match the load. The constant flow alternative is usually obtained by sensing the pressure
drop across an orifice in series with the pump output, and using this as a feedback to a type
of pressure compensator. There is an inevitable small loss of pressure in the flow path in
the order of 1.0 MPa (10 bar). It is used where the prime mover may change speed but a
constant output speed is required from a fixed displacement motor.
The constant power options give a wide range of implementations from crude approxi-
mations to more sophisticated characteristics. It is possible to achieve a characteristic
close to a constant power curve as shown in Figure 4.3.12. This requires a mechanical lever
and valve system. Alternatively a simpler single valve system with two spring rates can
give the straight line approximation ofFigure4.3.13. The second spring is shorter and does
not begin to compress until the valve has de-stroked the pump through approximately
2h of its displacement.
It is also possible to combine these constant power options with both maximum and
minimum pressure over-ride as shown in Figure 4.3.14. These variants all allow the
maximum prime mover power to be used over a wider range.
In addition, combinations of these controls are possible so that, for example, a constant
power control can be obtained with a maximum output limit, or a constant power
characteristic with an over-riding manual control.
Electrohydraulic controls range from the electrical remote input signal referred to above
to the far more sophisticated, digital fully-integrated system controller. Essentially all the
control strategies described above are available but with additional facilities. The greater
flexibility with electrical adjustment or through software changes, makes combinations of
different operating settings and over-riding control in different regimes much easier to
implement. Additional features include electronic anti-stall, acceleration and deceleration
limits, load speed control and automotive control (where the pump displacement is
determined by prime mover speed and hence the driver has a single accelerator pedal
control). There are also interface possibilities with other control or supervisory systems.
For example in vehicles through the standard CAN (Control Area Network) link, the
transmission operation can be integrated with other vehicle systems for optimisation of
various aspects of performance. Controllers offering these facilities are now relatively
widely available.
The physical mechanisms for motor displacement controls follow a broadly similar
range, with the same possible input signals. Control strategies are more limited because
there is less need for other than speed control. Electrical input to the motor opens up the
possibilities as described above.
The use of dual displacement motors is worth a mention as a simplified case of variable
displacement motors. A single dual displacement motor drive gives the opportunity for
two torque/speed ranges at the output. If two motors can be used with an additional valve
to isolate one motor, then this increases the possible number of ranges to four: both motors
full displacement; one full and one on half displacement; one alone on full; and one alone
on half displacement.
Vehicle configurations
There are a number of ways that transmissions can be configured and their use in vehicle
propulsion is a convenient way to describe them. The principal alternatives are shown in
Figure 4.3.15, but the following comments are not confined to vehicles and are relevant
in many other drive applications. If the pump and motor are connected back-to-back then
the system will have the variable speed advantages and the highest dynamic response but
will still require mechanical driveline connections. The next stage requires a motor for
each drive axle with a conventional differential drive to the wheels. This gives much
greater freedom to the location of components and freedom of movement on the axle itself.
Locating a motor at each wheel further extends this freedom giving greater ground
clearance possibilities. As shown these are all fixed displacement motors but variable or
dual displacement units could be used.
It is possible to operate a multi-motor system on this basis, although this will impose
certain performance limitations depending on the method of connection. Thus with 'N'
equal size motors in parallel, and assuming an equal load on each, the motors' ideal speed
will be proportional to pump output flow divided by N. This output will be reduced as a
result of the volumetric efficiency of the pump and motors giving a total flow loss for the
circuit. It also relies on the load torque on each motor being equal. If one motor is very
lightly loaded in comparison with the others or able to rotate freely then it will be able to
take the majority of the flow at low pressure. The others will not rotate and the system will
behave like a conventional differential geared system. The use of flow dividers would
ensure a supply to all the motors. With series connection of the motors, motor speeds
would be proportional to pump output but there would be a successive loss of speed
through each motor in the chain due to progressive flow loss in the previous motors.
Although the torques from each motor need not be the same, the pump must supply the total
pressure load and the motors be capable of operation with both ports pressurised.
The torque/speed range requirement largely determines whether a fixed pump/variable
motor, variable pump/fixed motor or variable pump/variable motor provides the most
economical system.
Dynamic response and characteristics

An appreciation of the dynamic behaviour of transmissions is important, in particular the
type of response which might occur and the principal factors which determine this
response. There are two aspects of dynamic response which may be considered - large
changes in pump displacement setting, and the general dynamic behaviour for small
changes about an operating point. These are described here in the context of a single
variable pump and fixed motor in a closed circuit transmission. The arguments can be
extended to the other configurations.
For a large increase in pump displacement the pressure in the delivery line will increase
rapidly as more flow is supplied by the pump than is taken by the motor. The rate at which
the pressure changes will be determined by this flow difference, the volume of the delivery
line and the bulk modulus of the fluid. This increasing pressure will increase the motor
torque and accelerate the motor shaft and load. It is likely that for a large change the
pressure will reach the relief valve setting before the load speed has increased to take all
the extra flow from the pump. The excess flow will pass across to the low pressure line and
return to the pump inlet. The motor will continue to accelerate whilst the pressure is higher
than the load torque, and its speed may increase above the final steady value before the
pressure falls from the relief valve limit and a speed overshoot will have occurred.
Obviously a reduction in the pump setting before this is reached, can bring the flows back
into balance and the motor will reach a new speed smoothly without overshoot.
A large decrease in pump displacement implies that the motor will take more flow than
is supplied by the pump and the pressure on the pump delivery line will now rapidly fall.
In an open-circuit system this causes problems as discussed above. In a closed-circuit the
minimum pressure will be limited by the boost system. In addition the motor will now be
returning more flow to the pump inlet than is required by the pump and the pressure in this
side of the circuit will increase. This causes a change in roles between the high and low
pressure lines and the initiation of braking of the load. The rates of pressure change are
governed by the same equation in all these cases.
The volume will depend on which side of the transmission circuit is being considered,
although in most cases the line A and line B volumes will be very similar. It will comprise
the obvious component of the pipe-work between the pump and motor but should also
include some allowance for the internal volume of the pump and motor themselves. A rule
of thumb figure allows half the volumetric displacement of the units as an approximation
to include a component for both the swept and dead volumes. This allowance should be
refined for a close coupled pump and motor where their volume will be the dominant term,
If flexible hose is used to connect the pump and motor then this will reduce the stiffness
in this pressurised volume as discussed in Section 2. This can be taken into account by
using an appropriately reduced Bulk Modulus. This could be significant and give an
effective value for Beff=O.25B for the fluid alone, if for example, twin braided hose is used.
The alternative, in which a transmission is disturbed about an operating point, could be
caused by changes in the displacement setting of either the pump or motor, the pump speed
or the motor load torque. A detailed analysis for all these cases is beyond the scope of this
book but a typical result is worth consideration. The case for a pump displacement
disturbance for a system with a fixed displacement motor is presented here in transfer
function form. This method is a useful shorthand way to represent the differential
equations for dynamic systems and shows a characteristic second order response:
Strictly speaking this transfer function describes the transient change in motor speed
following a small change in pump displacementxp. The standard response shapes (see
l"llM,
texts on control theory) can be obtained for different standard inputs and are largely
determined by the damping ratio. Figure 4.3. I 6 shows typical non-dimensional responses
for a step change. Mathematically a damping ratio of unity is the limiting case of "critical
damping". As can be seen for a damping ratio ofless than about 0.7 an oscillatory response
occurs with a frequency of oscillation close to the natural frequency. A larger damping
ratio gives a smoother response but may be slower to reach the new speed.
HYDROSTATIC TRANSMISSIONS 4] 5
Values for both damping and natural frequency can be established numerically as a
function of the system physical parameters. A lossless analysis indicates the principal
factors as follows:
This gives a clear indication of the physical factors which determine the dynamic
response of a transmission. The principal component of the inertia term J, is usually the
load inertia itself. Most motors have a low inertia in relation to their torque capacity. The
damping term comes from the frictional loss torque on the motor shaft to include the load
friction. This has been modelled here as a viscous friction term which is usually found to
be the most appropriate. Flow losses in the transmission also contribute to the damping
ratio as described below.
The volume of the transmission is the volume which is at the operating pressure PL, and
does not include the volume at boost line pressure. As above it includes the pipe-work
between the pump and motor and the allowance for the internal volume of the pump and
motor. Thus, remembering the effect of hose, the connection between pump and motor is
one which has a significant influence on the system dynamics and is an area where the
designer may have some freedom of choice.
In a transmission with a variable capacity motor it is quite acceptable to make an
allowance for operation at a reduced motor setting by using a proportion of the full
capacity. This means replacing DM in the formulae above by XDM.
Dynamic analysis modified to include transmission efficiency

A more detailed analysis can take account of the frictional and flow losses in the hydraulic
units. Previously in this chapter these had been considered in terms of volumetric and
mechanical efficiencies. However, in this analysis it is more convenient to include them
as either flow losses or friction torque losses. This is a general approach which can be
applied to any hydrostatic unit and is sometimes referred to as the "Wilson Model". The
equations used will depend on whether it is operating as a pump or a motor, since as a loss
it must always subtract in the sense of power flow. For flow from a pump the ideal output
is reduced by a loss term to give the actual flow output. For a motor the flow converted
to speed will be reduced from the inlet ideal flow by the loss term. These losses are usually
considered as pressure dependent since they are generally associated with a leakage flow
through the fine internal clearances in the units. Pressure difference across the unit is the
simplest term but there have been many alternative more complicated proposals in
attempts to generalise this approach. Hence in a dynamic analysis, this dependency
reduces to those variations occurring in the high pressure line, since the boost pressure is
constant. Also in practice this leakage is likely to be viscosity and hence temperature
dependent. The friction torque loss can also be generalised for either a pump or a motor
in a similar way. However, in this case it is normally considered that there are two
components: a viscous type loss which is speed dependent, and a Coulomb type loss which
will be load and hence pressure dependent.
Application of this approach to the transmission analysis introduces the flow loss for
both pump and motor, which can be accounted for with a single coefficient (the sum ofthe
individual motor and pump coefficients). The pump torque losses do not need to be
included for the transfer function considered. The motor speed dependent torque term is
already included in the coefficient Fv, and so just one additional coefficient is required.
This modifies the transfer function coefficients given above as follows:
It is generally found that these effects are small but they may need to be considered in
some applications. Note should be made of dependency of these coefficients on fluid
viscosity. Also there may be some differences in acceleration from rest or low speed where
there may be extra friction torque in the motor (and load if appropriate) associated with
stiction or break-away torque. This would not be included in a dynamic analysis.
Secondary control
Although the term secondary control can be interpreted simply as a variable motor control
type (iii) (see description of transmission configurations above) it is mainly used to
describe a specific type of ring main based system with more than one output service. Its
importance in giving a high efficiency drive with good dynamic characteristics and
accurately controlled outputs requires special consideration in this chapter. Secondary
control is most appropriate in applications where a number of rotary services can be

operated with a combined supply and usually involves electrical closed loop control in
each service. It also offers high efficiency solutions using energy recovery and this may
be equally appropriate in single service operation.
As explained throughout this book many hydrostatic systems, including transmissions,
work in a way in which flow is supplied to a circuit and the operating pressure is established
by the load. This can be called flow-coupled. The principle of a secondary controlled drive
system is operation from a constant pressure source by taking only the required flow, and
this can be called pressure-coupled. When the output is via an hydraulic motor then this
type of drive achieves high efficiency. By controlling the motor displacement setting, an
adequate torque can be obtained from the constant pressure network to balance a load
torque without additional throttle losses which would occur with more conventional valve
control.
The basic components in a circuit are shown in Figure 4.3.17 for a single service. The
pump is pressure compensated and maintains a minimum pressure in the ring main. The
accumulators can be used to store pressure energy above this level when flow is fed back
regenerative1y into the circuit by any of the services. The service shown comprises
principally of the hydraulic drive unit with its displacement setting determined by the input
to the proportional valve. Note that the symbol indicates it as a pump/motor unit.
The principle of operation can be established by considering the torque balance on the
pump/motor shaft. The hydraulic torque is set by the displacement setting of the unit since
it is operating with a constant pressure drop between the high pressure and tank
connections. If the displacement is set to balance a steady load torque, say due to a gravity
load, then the unit can hold its rest position. If the unit setting is increased slightly then the
extra torque produced will exceed the load and it can accelerate, raising the gravity load.
As the unit begins to rotate as a motor so it will begin to take flow from the supply. The
higher the displacement setting of the unit the more rapid will be the acceleration.
Reducing the unit setting back to its original position will bring the load back to rest. A
further reduction in setting will mean thatthe load torque is larger than the hydraulic torque
and the load can accelerate in the reverse direction, now lowering the gravity load. When
it does this then the unit will be acting as a pump, transferring flow from the low pressure
tank line to the high pressure line. In this condition other services may take this flow or
the accumulators could be charging. In either case the potential energy from the load can
be recovered, and the pump can remain off-loaded. Kinetic energy can be recovered in a
similar way from decelerating an inertia load.
Although these torque balances underlie the principle of operation, the displacement
setting is usually controlled to give either a demanded position or speed rather than torque.
A feedback signal from the speed sensor shown in Figure 4.3.17 will be fed into an error
amplifier to be compared with a demanded value. The error signal generated will be used
to drive the proportional valve. Analogue and digital controllers are both available, some
with simple proportional controllers, others with more complex algorithms.
An hydraulic unit (acting as a pump) is sometimes used in place of the electrical
tachogenerator with a connection direct to the operating cylinder giving hydro-mechani-
cal closed loop control. This has advantages in some safety critical regions of operation
but is prone to slight drift errors as a result of leakage in this unit.
Such systems offer efficiency benefits both by avoiding throttling losses in controlling
the output and by energy recovery from the load. However, one major hydraulics company
has compared the potential dynamic response capability with equivalent power rated
electrical drives. This has shown that secondary control hydraulic motors have approxi-
mately 10 times the free acceleration capability of either AC or DC electric motors in the
100-200 kW range.
419
DESIGN CONSIDERATIONS
HYDRAULIC SYSTEM design includes both the selection of the most appropriate circuit
to perform a function and the selection and sizing of the components within that circuit.
Such a design must be supported by suitable calculation and if necessary computer aids
(discussed in the next chapter). In many cases the designer may not have a 'free hand' in
selecting a solution since previous practice, industry sector, customer or company policy
may restrain the options and dictate features of the design.
It is not the purpose of this chapter to review all possible options but to make suggestions
for the major considerations in both circuit and component choice. Many design aspects
have been covered in the relevant chapters and these must be consulted for more detailed
discussion and information.
Since hydraulic circuits involve fluid flow, it is also important to identify "where the
fluid goes". This is perhaps the most useful concept for the designer to bear in mind at all
times together with a continued emphasis on simplicity and cost.
A design procedure
The design process must start with a specification for what is to be achieved. The more
detail at this stage the better, it may well be constraining but will provide a better
framework for the subsequent detail and should produce the most appropriate solution.
This specification may be more than simply performance information and may include
requirements for safety, reliability, maintenance, available space, weight, operating
environment and other possible constraints. If not available then certainly this is the sort
of information which should be sought. A study should also be made of any relevant
standards specific to the application area. A suggested approach is given in Table 4.4. I but
it is recognised that different priorities will exist in different application areas and modify
this approach. Direct cost has not been listed as a consideration because it has universal
relevance.
From the performance specification as much detail as possible should be identified on
the load forces and speeds required throughout the system operation. This may be cyclic
or more random. It may need to deal with both average loading and worst case. There may
be some freedom here for the designer to develop alternative load cases. For example if
a cycle time is the specified goal then this may be achieved by alternative strategies as
shown in Figure 4.4.1.
As shown either a rapid acceleration to a moderate steady speed, or a more gentle
acceleration to a higher speed could complete a cylinder stroke in the same time. The first
requires higher forces to give the acceleration; in hydraulic terms a larger area or higher
pressure. The second requires lower forces but higher speeds; which in hydraulic terms
might imply a higher flow but not if the actuator area is smaller. The restrictions of actual
component sizes will limit the useful solutions from such considerations but still allow
alternatives to be developed.
DESIGN CONSIDERATIONS 421
Duty cycle
One of the most useful concepts to develop in conjunction with load and speed information
is the typical duty cycle, since this can help in selection between alternatives. The
predictability of such a cycle varies greatly with the application area but even the use of
typical and extreme cycles will be found helpful, and in particular an idea of their relative
occurrence. Graphical presentation is the most appropriate and a typical duty cycle is
shown in Figure 4.4.2. Many cycles are event triggered but their presentation on a time
base as shown is still a valuable design aid.
This is shown in terms of force and velocity variation with time, the addition of power
variation on the same base is also possible. A useful extension to this, for purposes of
solution comparison, is to translate these axes into flow and pressure which is possible for
a particular actuator area. In addition the duty cycle should be examined for extreme flow
or pressure maxima, with the idea of eliminating the worst peaks. If this cannot be done
then the information is still useful in selecting the most appropriate source to supply the
hydraulic power.
A high peak flow for a relatively short duration would suggest an accumulator to
supplement the pump flow during the peak transient. Longer periods spent without the
system taking flow would suggest some form of pump off-loading. If there are require-
ments for both high flow at low pressure and high pressure but at low flow then these can
be met by a dual pump system with unloading of one pump. A less predictable but power
expensive requirement might justify the installation costs of a pressure compensated
pump. The clue to potential choices lies with a detailed examination of the duty cycle.
System efficiency
The overall efficiency of a hydraulic system has implications beyond the running costs and
unnecessary provision of energy. Principally any waste heat must be rejected, and this
adds further to the installation costs if coolers are required. There are many methods which
have been developed to improve the efficiency of hydraulic circuits. There are two aspects
which should be considered: flow matching and pressure matching. Flow matching
involves the pump circuit whereas pressure matching circuits may be achieved with
appropriate pump or valve selection. If both flow and pressure matching can be achieved
then the solution has been power matched, achieving possibly around 90% peak transmis-
sion efficiency.
In valve terms, the alternatives which should be considered include:
Open/tandem centre valves
Central by-pass valves
By-pass flow control
Pump unloading valves
Switched or modulating supply pressure
In power pack terms, the alternatives which should be considered include:
Off-loading pump when not required
Multiple fixed displacement pumps with sequential unloading
Accumulators to provide peak flow demands
Variable displacement pumps, pressure compensated
Accumulators to provide pressure with off-loaded pump
Power matched valve and pump control
Secondary control, regenerating absorbed power
The best choice must be made in consideration of the balance between installed cost and
estimated running costs for the typical duty cycle.
423
COMPUTER AIDS AND SIMULATION
Hydraulic systems are no different from any other area of engineering in the influence and
potential use of computer tools. There are both generic codes and purpose written
programs which can help with anything from component design and selection to complete
system dynamic simulation. Finite element methods have been used in the design of pipe
fittings, CFD methods can predict the effects of flow patterns in components but the
emphasis in this chapter will lie in the area of system design and concentrate on programs
which help in component selection and system operation.
These methods still span a wide range but have the common feature of enabling various
design choices to be compared. The alternatives which exist in many system designs are
frequently not difficult to compare but require repetitive calculation of the same functions
with different parameters. The use of a computer approach allows more design iterations
and a better optimised solution. However, both the validity and accuracy of the results as
output from any program are always dependent on the data and modelling base which are
the input. Even when reputable and proven programs are used it is still possible to produce
a nonsensical output if the basis of the techniques and their range or limitation of
application are not fully appreciated.
Categories and requirements

To be useful as a design tool any computer program should permit investigation of
alternative solutions. In some cases this may simply be a comparison between different
size units to optimise component selection. Ideally, however, the output data should be
available in a flexible way such that comparisons can be made of say cycle time, power
consumption, dynamic response, or accuracy for the alternative methods of solution. It
would for example be useful to make comparisons between an open loop and a closed loop
solution to the same problem but this is likely to be difficult to achieve except with the more
sophisticated packages.
Steady state solution packages are available from many of the manufacturers and these
provide very useful design tools for a wide range of applications. They are, however,
essentially aids to help select the most appropriate sized component from that manufac-
turer's range. They will not give any indication of potential problems in transient
behaviour, which is frequently the area of difficulty in systems aiming to benefit from the
rapid dynamics which hydraulic systems can give. Although the product range is very
similar from most major manufacturers these programs do not normally allow compari-
sons with another range of products. Work is under way under the auspices of the BFPA
to try to standardise the format for presentation of information on components so that they
may be imported and used more widely in all computer packages.
Full dynamic simulation is undoubtedly the best method to predict performance and
hence avoid problems. This requires good dynamic modelling of the components in the
system and the correct choice on the part of the user in selecting the most important
dynamic effects. The models will be cast in differential equation form and the simulation
itself will integrate these equations and maintain compatibility for the values of the
variables during this process. The process of numerical integration can be illustrated with
an example for the variation of velocity v, and position y, with time t, in the form:
Thus the change in position can be evaluated for a small time step ~t, for a given value
of velocity. This process must be carried out a large number of times with small steps in
time to accommodate any changes in velocity which might occur. The real change in
position in a sensible time interval is then the sum of all these small changes.
This is a simple, so called single step integrator, in fact an Euler integrator. This works
well with mathematically continuous functions (eg. straight line or a smooth curve) when
reasonably large time steps can be used. However, it will not work well with rapidly
changing, non-linear or discontinuous functions such as will occur when a piston reaches
the cylinder cushion or a relief valve opens or shuts. If a simulation is to run in an
acceptable time, these require more complex integration handling which can recognise the
situation and adjust accordingly. There are many such alternative methods, the most
effective involving a variable length time step.
Basic component selection

Many manufacturers now produce their product range on CD-ROM. Several also produce
design programs which help the user in making the selection between circuits as well as
components. The data available on components is usually very detailed, thus replacing a
catalogue. In many cases it is possible to down-load a standard format drawing for
incorporation into the design, together with data for a parts list.
Most of these programs are arranged in a tutorial format which takes the user through
an appropriate design sequence for the circuit chosen. This may allow checks to be made
in various critical areas, such as a check for pressure intensification on the rod side of a
piston when dealing with an overrunning load. These will of course contain the data for
the products of that company in suitable format to be used in the selection. Some programs,
however, allow the entry of user defined data to explore the 'what if alternatives more
fully.
COMPUTER AIDS AND SIMULATION 425
COMPUTER AIDS AND SIMULAnON 427
The output available from these programs is more limited than with the general purpose
packages described below, particularly in graphical display of system behaviour. How-
ever, a print or file output of the full details of the chosen component (s), including fully
dimensioned drawings, is undoubtedly very powerful and efficient.
Other hydraulic specific programs

There are a number of programs which are available to simulate various aspects of
hydraulic system performance, and again these cover a wide range. They have the
advantage of not only describing the performance of typical hydraulic components with
appropriate dimensions and parameters but also being user friendly to a hydraulic
designer. They usually provide data in a suitable format for a selection of real products
which can be used in the design or simulation. These may be as 'examples' only but there
is normally the flexibility to add to this database as the user builds a collection of favourite
products.
The other advantage with such programs over the more generic approaches below is the
layout and use of more familiar fluid power symbols in describing the components and
circuits. There are also all the obvious building blocks available as standard to enable a
circuit to be constructed rapidly from these existing modules. A full range of graphical and
tabular data are usually available as output, and such things as efficiency comparisons
between alternative pump supplies can be obtained.
Generic computer methods

There are a large number of computer programs which can be used to solve general
engineering problems which are equally appropriate to solving the equations for hydraulic
systems.
Given sufficient knowledge of the equations describing the behaviour of a system or a
component it is always possible to write a special purpose program to predict performance
and examine different component sizes or specifications in fulfilling the overall perform-
ance requirements. There are many program levels at which this can be successfully
carried out.
It is possible to use a standard high level language such as Fortran, Basic or C. The code
becomes very specific and is more difficult to edit; it is less transportable between users
and more cumbersome to make system modifications. Thus this approach is only likely
if the system studied can justify the programming time, and it is very expensive, has very
specialised requirements or is one of a large number of similar systems. The complexity
of the modelling can reflect the requirements and may vary from simple steady state
descriptions of components to more complex fully dynamic equations.
Standard spread sheet packages are now very powerful and will easily cope with the
steady state modelling of even quite complex systems. Although a specific circuit will
have to be constructed it is possible to produce this in a modular form which allows
modifications or use in different system analyses. Some of the more tedious calculations,
such as pressure losses through pipework can easily be incorporated and dimensional
changes investigated. It is also possible to include some of the non-linear features of
hydraulic components with conditional execution of alternate blocks. Most of these
COMPUTER AIDS AND SIMULATION 429
packages now also produce graphical output and so the results can be presented in an easy
form for comparison. The disadvantage is the difficulty in knowing what equations have
been embedded in the blocks without a detailed and time consuming interrogation of the
spread-sheet.
There are a number of more mathematically based environments where equations can
be written in a conventional form as they might be hand written on paper. These solvers
are also extremely powerful, again allowing non-linear and conditional statements to be
included. Data sets can be consulted as in a sub-routine and 'modules' constructed which
are easily copied from one program to another. The information display in this style of
program is very effective if the user is familiar with an equation format. If the equations
are well arranged, with some commentary, then this mimics the layout which might be
used by a designer and can communicate very effectively. Obviously any data can be
easily changed to investigate alternatives. Graph plotting is also very straightforward and
many mathematical functions are built-in.
There are also more advanced equation solvers based on the manipulation of matrices.
These can be used in both a simple and more complex way. In simple form they can be
programmed as described above and again have good graphical output facilities for the
display of results. They may also be used as the basis for time or frequency domain analysis
if the appropriate parts of the program are available. In this form they again can include
non-linear and conditional features which often occur in hydraulic components. The usual
method of constructing a system simulation is through building blocks in a block diagram
type of format. Depending on the complexity of the modelling approach the diagrams can
become very busy and do not necessarily have an obvious relationship with the system
components, although it is possible to condense blocks in a more user friendly form. For
time domain simulation the standard integration routines are rather limited in relation to
the demands of many hydraulic circuits when non-linear features are included.
Packages for control system design in themselves cover a wide range and most can be
used with hydraulic system models. At the simplest end are packages which will accept
a transfer function description of a system and will display the step or frequency response
of the system (the latter in Bode, Polar or Nichols formats). The transfer function can be
combined with different forward path control or feedback path functions to investigate
their effect on stability or closed loop response, and the results superimposed. These are
useful design tools to investigate alternative controllers but the limitations of linear
transfer function modelling limit their precision in predicting performance.
Most of the matrix solution packages can also be used for control system design. Even
here the control system investigation is likely to involve a linearised version of the
hydraulic system. However, some of the control options available are very varied
including many advanced ideas such as H-Infinity design methods. The output can again
be displayed in terms of either step or frequency response for comparison.
There are a number of specialised time domain simulation programs. These will contain
the building blocks to help construct the hydraulic components and they will certainly
have all the features required both for non-linear modelling. It is also likely that they may
have some basic hydraulic components as standard together with other common electrical
and mechanical components. Frequently these have been based on a particular application
area which a standard model library will reflect. However, the commonality of a
differential equation allows their use to be extended. They have the advantage of more
sophisticated integration methods which help produce a system simulation in an accept-
able run-time.
SECTION 5
431
SERVOSYSTEMS
A DEFINITION of an hydraulic servosystem is a system which provides both power
amplification and automatic correction for deviation from the required value of an output
variable. As such it is a closed loop system with feedback of the output to be compared with
the required value and used to correct for any errors. The output in hydraulic servos is
usually position but speed, load or pressure are equally possible alternatives. Sometimes
a combination of these is used. The power levels in hydraulic systems can be high and
require only relatively low power signals (mechanical or electrical) to direct them, they
are thus suited for servo applications.
Some examples of hydraulic servosystems include the flight control surfaces on most
larger civil and military aircraft; the control of the force between the rolling cylinders in
metal rolling mills; actuation of movements in robotics; and the movement of the cabin
in a flight (or games) simulator. These applications all involve electrical control systems
with a transducer producing the feedback, an electrical controller and an electrically
modulated valve controlling the hydraulic output. Mechanical servosystems are also
possible but tend to be limited in their application, and mostly used for servo assistance.
Examples include power assisted steering in wheeled vehicles, brake servos, and in
hydraulic pumps themselves, many swash controllers have a built-in mechanical position
control servo. There is far greater potential with an electrical rather than with a mechanical
servosystem where design options are relatively limited. For example injection moulding
requires position and speed control in different parts of the cycle and switching of the
control variable is possible electrically.
This chapter will try to introduce the reader to some of the more basic aspects of control
in the context of hydraulic servosystems and highlight the most important features. This
is not a text book on control nor on the specialised aspects of hydraulic servosystem design.
The main explanation is given in terms of an electro-hydraulic position control system,
with a double ended (through-rod) cylinder. However, the ideas are equally appropriate
to any positioning actuator, including rotary, and need to be considered for any hydraulic
servosystem.
Closed loop basics

As stated above the basis of a hydraulic servosystem is referred to as closed loop control.
There is a standard method to represent control systems, known as block diagrams and a
typical electro-hydraulic closed loop system is shown in Figure 5.1.1a. This can simply
be considered as having a block or box to represent each of the major components in a
system, shown schematically in Figure 5.1.1 b. The valve could have been built integrally
with the cylinder and it would be possible to draw these as a single block which has exactly
the same meaning in control terms. In a similar way the feedback amplifier (or signal
conditioning) has been incorporated into a single feedback path block with the transducer
itself, although it could have been shown separately. The block diagram also shows a
summing junction, where the feedback signal is subtracted from a demanded signal to
produce the error signal. This error indicates the difference between the actual system
output and the required output and is then used to operate the valve to supply hydraulic
fluid to a cylinder to move the output and correct the error. If the error signal is just
multiplied by a constant value before passing to the valve then this is called aproportional
controller, with aproportional gain constant which is selected in design or on final test and
commissioning.
t:'~_._.. _l _
The lines between the blocks have significance as intermediate variables in the control
loop. For example the input to the valve block is shown as the current to the valve i, and
the output from the same block is the valve spool position x. The demand signal Vi, can
be considered as a voltage level which must be set external to the loop, perhaps by a
manually adjustable potentiometer or from a supervisory control computer or PLC.
The advantages
In general a closed loop system will give more accurate control of the output, a better
dynamic response and be less sensitive to changes in system parameters, e.g. frictional
coefficients. The accuracy of an open loop system relies on a complex link through a
number of components between demand and output (the forward path in the figure is an
example). The use of a closed loop gives the possibility for better performance and more
SERVOSYSTEMS 433
precise and accurate control of the output (within the accuracy limitations of the feedback
transducer). This means that if any drift in the output occurs, maybe through internal
leakage or temperature changes in the forward path components, this will be detected and
automatically corrected. It would also be possible for the effect of any external loads
applied to the output to be detected and a correction made but the error would not
necessarily be eliminated. This depends on the controller used as discussed below.
The problems
The system output will not always be the required demanded value in a simple 'propor-
tional control' system. In a situation where the output actuator has to generate some force
to overcome any external load (perhaps a gravity load) then there will have to be a pressure
difference across the actuator piston. A pressure difference can only be maintained if the
valve is partially open (if it is exactly centered then the pressures will equalise). The valve
can only be open if it receives an input and hence for a proportional controller there must
be a difference between demand and feedback, i.e. an error. Errors of this sort are usually
called steady state errors. It should be noted that in order to give a particular value of valve
opening, the error can be kept small if the proportional gain is large.
There is an aspect missing from the simple block diagram given above and that is the
lag in components giving delays in time between the input and output from each block. All
components will experience these delays, even the electrical amplifiers. However, it is the
slowest components within the loop which have the most significance, and this will
generally be the combination of load and actuator. A re-consideration of the events when
an error occurs will indicate the importance of this delay. If the correction for an output
disturbance is delayed through the loop then it may arrive too late to make the required
correction. Worse than this if the disturbance is fluctuating then by the time the correction
has propagated around the loop it may actually arrive back at the output to be in-step with
a component of disturbance in the opposite direction and add rather than subtract. A signal
like this can actually chase itself around the delays in the loop, gradually increasing in
magnitude. This condition is known as instability, and is a property of the loop itself.
A larger forward path gain will increase the speed of correction within the loop but
makes the unstable condition more likely. Thus the main problem is the choice of forward
path gain to optimise between the requirements of steady state accuracy and avoiding an
unstable response. If an acceptable compromise is not found then design changes or more
sophisticated controllers are required.
The system transfer function

A more mathematical approach can quantify the characteristics described above as well
as describing the other response features of servosystems. One way to achieve this is to
add appropriate mathematical models for the various components into the block diagram
and this is shown in Figure 5.1.2 for the same system.
The usual representation used in such diagrams is a lowercase's' to be the equivalent
of differentiation with respect to time (dJdt). The component models are thus differential
equations obtained either from a suitable analysis to obtain an equation of motion, or from
testing and matching the results with standard transfer function responses. All of these
equations can be called a transfer function, which relates an output to the appropriate input
as the following example for the valve shows:
There are only two blocks shown with such dynamic models, the electrical blocks are
shown with a simple constant or gain value as a transfer function. These latter could be
described more fully by similar differential equations but the time lags produced by these
would be so quick that they could not be seen even in the most rapid of transients. This can
be demonstrated by a more detailed comparison of the valve and the load equations used.
As given above the valve is described by a first order equation often called afirst order
lag, there is only one s-term. In this equation the speed of dynamic change is described by
the numerical value taken by the time constant i, the smaller it is the more rapid will be
the response. The cylinder and load have been shown as a third order equation (highest
power S3), this combines the spring-mass second order effect, described in the chapter on
Actuator performance with the natural integration effect between flow and position (a
flow into a cylinder gives a velocity, and so must be integrated to give position). The speed
of response of this equation is governed by the natural frequency term (On but in this case
the higher the frequency the faster the response. It is therefore easier to make comparisons
if the reciprocal of the natural frequency 1/(On, is used. Ifa valve is chosen which has a time
SERVOSYSTEMS 435
constant 5 times quicker than the load i.e. t> 5 x I/ron then the valve will have virtually
completed its response before the load has moved. This is shown in Figure 5.1.3 where the
step response for the load and valve are compared. Note that this shows only the second
order part of the load response. In fact most analyses will neglect the valve response if it
is this much faster than the load, or rather it may be better to select a valve with a faster
response. The electrical circuits are likely to have time constants which are 100 times
faster again and so may legitimately be neglected from the outset. The only occasion this
may not be valid is with very low frequency electrical filtering. If a more accurate analysis
is required with this filtering and the valve dynamics included, then it would also be
preferable to use a second order model for the valve itself.
This approach confirms that the load itself is the limiting component in the dynamic
response of the system. There are two ways in which this information is useful- it can be
used to estimate the maximum response rate likely from the system and also it can be used
to help analyse the stability of the system as described in the next section on frequency
response. The minimum response time will ultimately be limited by the load natural
frequency. The best step response time possible is in the order of 5/ron and this assumes
that there is sufficient power available for the acceleration.
It is also possible to simplify this block diagram to an equivalent form which gives just
one transfer function, the so called Closed Loop transfer function. This is in fact the best
mathematical form to predict the overall system response. The effect of the proportional
gain term Kp, could then be examined for its influence on the dynamic system response.
The typical response curves are shown in Figure 5.104 for increasing values ofKp As can
be seen the response gets quicker but increasingly more oscillatory which indicates that
instability is being approached.
In fact it is the total gain obtained by multiplying the numerators of all the blocks in the
loop which is important. This is sometime called loop gain. The controller gain is the only
term which is likely to be adjustable, everything else will be selected for other criteria. It
is sometimes thought that the gain of the feedback transducer can be adjusted to modify
the loop gain. Although it plays its part, it is better selected on the basis of a sensible output
voltage (usually 2V but up to 10V) corresponding to the full range of output.
SERVOSYSTEMS 437
The significance of frequency response

An alternative method which is used, very effectively, to describe dynamic response of
components and systems is in terms of their frequency response. Frequency response
describes the output from a component when it is excited from a sinusoidal input of
constant amplitude but over a wide range of frequency (which can be considered as a
gradually increasing frequency). The output waveform is compared to the input and two
measures of the relative response can be identified - the size of the output waveform
establishes the amplitude ratio; and the relative position in time establishes the phase
change (usually a phase lag). Figure 5.1.5 shows these two characteristics and the most
common way to represent them on a logarithmic frequency scale as a Bode diagram.
The higher the frequency at which the amplitude reduces and the phase lag occurs then
the better will be the system dynamic response. The frequency responses for the load and
valve described above can be compared as in Figure 5.1.6. This also shows the character-
istic difference between a first and third order transfer function.
This approach allows a more detailed explanation of the stability problem in terms of
sinewaves. As a reminder, the unstable condition occurs when the signal being fed around
the loop from the output adds to itself when it reaches the output again. This could be
formalised for a sinewave by saying that it is in-phase again when it has passed around the
loop. Since the signal passes through all the components in the system before returning to
the output, at any particular frequency the total phase change which occurs around the loop
can be summed. The summing junction itself contributes a 180 phase change (like
inverting the wave), and so the unstable condition can occur when the other components
contribute another 180, and the signal has shifted a total of 360 and is in-phase again.
However this is not the only aspect, the signal must also be increasing in amplitude as it
passes around the loop for the output to increase in an unstable manner. A signal slightly
reducing in size around the loop i.e. an amplitude change just smaller than one, will be
stable but give such an oscillatory response that it could not be sensibly used. This
conforms to the Nyquist Criterion of stability, described in most control textbooks as an
amplitude ratio equal to unity at a phase change of 1800 for what is called the Open Loop
transfer function (which is simply the product of all the blocks in the loop).
It helps explain why increasing the proportional gain makes the system unstable since
this causes a larger amplitude in the signal as it is passed around the loop. It also explains
why the valve should be chosen to have a faster response than the load. A slow valve will
have a lower frequency response and consequently give too much phase shift at too low
a frequency, although there are compromises possible as it also reduces the amplitude.
Frequency response analysis can be used as the basis for system design and the
application of more sophisticated controllers as described below. However, it has its
limitations, principally since it is a so called linear technique operating with system
equations which are linear (the differential equations in Figure 5.1.2 are linear). The
behaviour of hydraulic systems can be explained in these linear terms but they are in reality
non-linear and the coefficient values in these equations will change (for example the
natural frequency with piston position). Hence the design of systems is reliant on good
estimation of the coefficient values to use, and frequently it is better to use 'worst case'
values.
Characteristics of hydraulic components

Component selection can obviously influence system performance. Various parameters,
such as load mass and required stroke, are going to be fixed by the design specification.
The coefficients in the transfer function equations can relate the physical characteristics
of the system to its components for a double ended cylinder as follows:
The natural frequency was also discussed in the chapter on Actuator peiformance but
its importance here requires additional emphasis. An increase in the volume term reduces
the natural frequency and response. This volume comprises an inevitable volume in the
actuator itself plus any volume in the connection from the valve to the actuator body. For
this reason it is better to keep the valve as close as possible to the actuator and minimise
this volume component. If the connection volume can be considered negligible then a
useful simplification can be made, which will also be true generally. The minimum natural
frequency (the worst case) occurs when the piston is in the mid-position and hence the
relevant terms in area and volume become:
SERVOSYSTEMS 439
Where L = total piston stroke

This emphasises the importance of actuator area in the dynamic response even though
it is within the square root, and explains why short stroke, large area actuators have a high
natural frequency. Area is normally chosen in order to give the required force either at stall
or dynamically. A larger area will have a consequence in an increase in the flow
requirement for the same maximum velocity and it is likely that this will also require a
correspondingly larger valve. Note also that an increased area also reduces the 'gain' of
the cylinder/load transfer function unless the valve size is also increased.
The conventional analysis for a linear actuator system is based, as here, on an equal area
cylinder. This is much the simplest case mathematically and is the choice for many
servosystems because the performance will be symmetrical. However, there are also many
systems with an unequal area actuator but there are some characteristics which may limit
the maximum performance. An unequal area cy linder operated from a symmetrical valve
(the same metering restriction on both ports) will have a higher velocity in the extend
direction than in the retract direction (unlike the case when operating from a constant flow
supply). In control terms this translates as a higher gain in the extend direction which will
provide the limiting case for stability. The ratio of extend to retract gains is the square root
of the ratio of the piston to annulus areas. The performance in the retract direction will
obviously be inferior unless some form of directional compensation can also be included.
The other limitation relates to the pressure levels necessary to balance the forces across
the piston when at rest. If used with a symmetrical valve there is no unique solution for the
individual pressures, a number of combinations will provide the force balance, and it is
likely that the valve will have to adopt an off-zero position depending on the history of
previous movement. Analytically, there is a unique solution for the completely symmetrical
case and also if the area ratio of valve opening matches the area ratio of the actuator. There
is a good high performance example in the main support actuators for a flight simulator
where this flow matching is employed. However, many lower performance systems work
quite adequately with the unsymmetrical configuration and a symmetrical valve.
The value for damping ratio is often the most difficult to predict. Friction in the actuator
and the load both contribute but there is also a 'fluid damping' component arising from
flow through the valve as a restriction. This latter term will be dependent on val ve opening,
and will fall to a minimum value when the valve is closed. Cylinders for servo use are
generally chosen to have low friction seals to reduce stiction and any hysteresis effects but
this combination is likely to give a 'worst case' estimate for 1;which is quite low, typically
1;~ 0.2 is likely unless there is significant load friction.
The normal rule for selecting valve sizing is based on a maximum power transfer
calculation. This shows that the pressure drop across the valve should be 113 of the supply
pressure when operating at the designed maximum power. This gives 213 of the supply
pressure available across the load actuator. This may never be an actual operating
condition but itis still a useful sizing rule of thumb. A valve sized too small will give good
control but will waste energy. An oversized valve will not give as good control at low
openings nor will it waste energy but other valve performance features may become
significant including hysteresis and drift. There may of course be circumstance where a
better solution can be found by breaking the lhrd rule.
The valve also contributes to the non-linear aspects of the equations, mainly through its
behaviour as an orifice, since this shows that the flow rate is proportional to the square root
of pressure drop. Most electro-hydraulic valves also exhibit some amplitude dependency
in their dynamic response. The other feature of valve characteristics which is important
is the valve lap at null or zero opening. A zero lapped valve will give the most linear
characteristics with opening and is the most frequently used. A fully under-lapped valve
actually gives more linear performance in a mathematical sense but tends to be wasteful
of energy, it can also help improve system damping. Under-lapped valves have a higher
flow gain about the null but a reduced pressure gain and hence reduce the overall stiffness
of a closed loop system and affect its steady state accuracy.
The selection of the most appropriate val ve for a servosystem must be balanced with the
features and performance requirements of the system itself. In particular the dynamic
response of the valve must be sufficiently fast in relation to the load natural frequency or
time constants. A valve (or valve amplifier combination) which gives a linear relationship
between input and valve opening is generally easier to incorporate into a control loop.
However, dead-band or a large valve overlap is not a problem with most speed control
systems. A valve with low hysteresis is desirable since significant hysteresis can introduce
steady state errors with proportional control. Although these characteristics are most
frequently met by servo-valves it is equally appropriate to consider other valve designa-
tions for servosystem use if their characteristics are appropriate.
Pressure control
In systems where pressure is the controlled variable then there are some additional or
alternative considerations. Since pressure in a system changes very rapidly then more
attention needs to be paid to the valve dynamic response which should be specified in as
much detail as possible. It must either be selected sufficiently high relative to the pressure
response of the system as above or it can be selected to be the slowest responding
component and hence be the dominant feature of the dynamic response. It is also usual to
specify a characteristic of servo-valves which may be relevant and that is its pressure gain
characteristic. This usually shows that the full supply pressure change can be obtained
through a small proportion of the valve opening, as little as 5% total travel is usual.
Controller structures
The term controller structure refers to the way in which the control loop is built since
there are alternatives which allow modifications to improve the overall performance.
These may refer to the forward path of the loop alone or can include modification in
the feedback path. These modifications are sometimes called compensation in control
terms. The simplest controller to apply is the proportional controller but it has its
limitations in the compromise possible between good dynamic response, stability and
good steady state accuracy. The methods of compensation are therefore, alternative
SERVOSYSTEMS 441
ways to obtain the best performance from a hydraulic system design.

The most common alternative is the PID controller, standing for Proportional plus
Integral plus Derivative. This can be used in place of the simple proportional control in the
forward path. It means that in addition to operating the valve opening in direct proportion
to the error, additional components due to the derivative (rate of change of error) and
integral (summed total error) will be included.
The derivative term should improve the system response to sudden changes in demand,
which is not normally a problem with a hydraulic system. It could also help by improving
the damping ratio of the system but this will depend on the originalload/cy linder transfer
function (it is likely to help if this is second order dominant). Differentiation of any noise
on signals (particularly from feedback) can lead to problems and hence they may need to
be filtered.
The integral term is intended to improve accuracy by removing errors due to external
loads. The integrator will change the output until it sees a zero input, and the error is
'integrated to zero'. Unfortunately a position control system usually includes an integrator
implicit in the load/cylinder transfer function, and double integration leads to stability
problems. There is thus great difficulty in setting the integrator gain. Too low and it will
take a long time to remove the error and the output will apparently 'drift' towards the
required value. Too high a gain and the system will become oscillatory, higher still and
it will be unstable. Integral terms with even moderate gain may show a large, slow
overshoot which is evidence of what is called integrator wind-up. This occurs where the
output from the integrator is increasing whilst the error is large but unnecessarily since the
correction is already being made by the proportional term. There are standard solutions for
integrator wind-up.
Thus PID is not an ideal solution for hydraulic servosystems, although this will depend
on the exact form of the system transfer functions and most will benefit. Alternative
methods exist, both in the forward and feedback paths. These latter, sometimes called
feedback compensators, can be used principally to improve damping. A technique
effective for the normal position control transfer function is to use velocity and accelera-
tion feedback, sometimes called state feedback. However there are difficulties in deriving
the acceleration signal. The most useful alternative in hydraulic systems is to use an
additional feedback loop with a signal from a differential pressure transducer connected
across the load actuator. This acts in a similar way to the acceleration term when the mass
effects are significant. The signal should also be high pass filtered to prevent steady loads
causing additional errors in the steady state. These methods can be used with a propor-
tional plus integral forward path, which may be necessary since the integrator is the best
method to improve the steady state accuracy.
Digital control
A closed loop controller may also be formed with the inclusion of a digital computer or
microprocessor within the loop. This requires the feedback signals to be in a suitable
digital form, either directly produced or through analogue to digital conversion. The
output from the digital control calculation will also be digital and will need analogue
conversion to drive the valve amplifier. The main advantages come from the possible
SERVOSYSTEMS 443
control calculations which can be implemented and these are sometimes referred to as the
control algorithm. The flexibility of control algorithms offer improvements for even
simple controllers, with the further opportunity of very much more sophisticated control-
lers which can maintain consistent system performance despite many parameter changes.
As simple examples the problem of integrator wind-up can be greatly alleviated by
disabling this part of the calculation when large errors exist from a sudden input change.
Another technique is known as gain scheduling, where different controller gains could be
used in different operating zones such as different parts of a piston stroke or for different
directions of travel for an unequal area cylinder. Another task made much easier with
digital control is the transfer from one control mode to another such as in the three way
transfer from position to speed to load control in different parts of the operating cycle for
some injection moulding machines.
Further discussion on aspects of both open and closed loop digital control is given in
the chapter on Computer control.
Position feedback in cylinders

There are many types of transducer which can be used with electro-hydraulic servosystems,
too many to be described here. However, there are some transducers which have been
produced specifically, or specially adapted, for use in linear actuators (hydraulic cylin-
ders) and information on these is less widely available. Measurement of speed, load and
pressure for feedback signals can be carried out with more conventional transducers.
Pressure transducers are discussed in the chapter on Instrumentation, but for details of
other devices consult an appropriate instrumentation handbook, bearing in mind the
requirements discussed above.
The reality ofincorporating a transducer into a cylinder as an integrated unit is relatively
recent butit is an eminently sensible idea operationally. There are potentially maintenance
complications if there is a transducer failure and access is required to replace it. However,
the added mechanical protection for the transducer should help to protect from damage and
improve reliability. In an external installation for the transducer it should be remembered
that the piston and rod are often free to rotate about the cylinder axis unless constrained
by the rod mounting, and this will complicate the transducer mounting.
The basic configuration used in the majority of designs is shown in Figure 5.1.7 for both
single ended and through rod cylinders. In both cases there is a transducer body and a rod-
like probe which are fixed relative to the cylinder body, and a modification to the standard
piston rod which includes the attachment of the moving part of the transducer. This
moving part varies with the operating principle of the transducer as described below. In
the single ended case the transducer probe will be immersed in the operating fluid and
subject to the operating pressures. The major features of the variants available are
summarised in Table 5.1.1 but it should be noted that these specifications are liable to
change as manufacturers develop their products.
Conventional transduction methods

There are variants of both hybrid track potentiometers and LVDTs (Linear Variable
Differential Transformer) which have been produced for mounting in hydraulic cylinders.
The principles of operation of both these devices are well established and hence have not
been detailed here. For the hybrid track potentiometer, a collar is attached to the piston rod
which makes contact with the resistive track itself, as the wiper, and connects through a
second sliding contact to a low resistance conductor to complete the return circuit. In the
case of the LVDT the conventional core of magnetically permeable material must be
attached to the piston rod and move within the transducer body.
Eddy current device

The probe tube contains a single detector coil which is supplied with a carrier frequency
in the order of 10kHz. The moving part of the transducer is in fact a tube of special material
(Anticorodal B) which is inserted in the hollow piston rod. The eddy currents produced
in this tube interact with the coil where the two are overlapped, to change the impedance
of the coil. This impedance change can be detected by a number of methods to give either
a voltage or current output. The impedance is also very temperature sensitive and so active
temperature compensation is built in as part of the design (in fact in some variants, a virtue
is made of this and the temperature signal is available externally).
Variable resistive vector transducer (VRVT)

This device has a single detector coil within the probe tube and again operates with a carrier
frequency supply. The moving part is a high magnetic permeability core which moves
within the coil and changes its impedance. The coil is supplied with a constant current
drive and the in-phase potential difference across the coil is detected. This is demodulated
to give a voltage output linearly proportional to the core displacement. The core may be
SERVOSYSTEMS 445
attached directly to the piston rod or supplied within a sleeve fitting inside the rod and
outside the transducer probe.
Linear inductive position sensor (LIPS)

The probe contains two coils, one a sensing coil with a relatively coarse pitch which
determines linearity, and the second a balance compensating coil. These are supplied from
a high frequency oscillator (1 - 2 MHz) with careful control of the supplied voltage. The
coil impedance is again varied by the length of overlap with eddy current effects in a
surrounding tube of conducting material. Aluminium or stainless steel may be used but it
is possible that the piston rod material itself may give an adequate signal and no additional
sleeve is required. The detection of the output is also novel and involves a sampling
technique with values taken equi-spaced about the voltage peak. The high frequencies and
special signal conditioning techniques have led to a solution with all the electronics built
into the transducer housing giving a self contained package somewhat like a DCDC
LVDT.
Magnetostrictive
This transducer is digital in nature but unusually this is as a result of the discrete time
interval inherent in the process rather than quantisation in the position information. The
probe contains an outer protective tube, an inner tube or waveguide, and along the
centreline a conducting wire as shown in Figure 5.1.8. A permanent magnet is located
within the piston, moving with it and external to the probe. A current pulse is passed
through the circuit comprising the central conductor and the waveguide. The interaction
of the magnetic fields from the magnet and the pulse produce a torsional strain on the
waveguide as a result of an effect called magnetostriction (like piezoelectric but a
magnetic rather than an electrical field). This torsional pulse (produced instantaneously)
travels along the waveguide at a known speed and when it reaches the transducer head it
rotates the sensing coils to give an output. The positional information comes from the time
taken between the pulse occurring and the torsional wave reaching the sensing head, which
indicates the distance from the magnet to the head. This information is available as a time
interval but can easily be processed to give a proportional voltage.
Linear resolver
This is again inherently a digital type device but the signal is processed into analogue form.
The end of the probe houses four relatively short length coils, two act as supply and two
as detector coils. The moving section is a series of tubes, located end on end, inside the
piston rod, which each house a self contained length of measurement coil (relatively coarse
winding pitch). These coils are energised by the alternating field produced in the supply
coils. The signal in the detector coils then varies as its position changes relative to the
measurement coil windings. The winding pitch in both the detector and measurement coils
define the resolution limits and the measurement coil defines the basic linearity. The
segmenting of the measurement coil reduces the power requirements since only a
maximum of three adjacent lengths will ever be energised at one time.
This transducer only gives incremental information i.e. the increment of distance
moved from a starting point. To know the new absolute position the starting point must
also be known. A modified measurement coil can be incorporated at any position in the
stroke to provide a datum which can provide a check on absolute location. It is also
relatively easy to build in a back-up battery to maintain the absolute position information
in the event of a power failure.
Integrated digital sensing

This system uses a completely different mounting style and has been available from
several manufacturers in the past. It relies on the machining of small grooves in the piston
rod itself which are then coated to give a completely smooth finish (chromium plated or
ceramic coated). The position of the grooves can be detected by magnetic proximity
sensors built into the rod end housing. Two sensors are usually used carefully spaced
relative to the groove intervals to improve the resolution. The resolution is limited by the
closeness and machining accuracy for the grooves. This is a very compact system which
is robust, not exposed to hydraulic pressure and, depending on the detail design, the
electronics and sensors may be accessible for easy service. This system, however, only
gives incremental information as described above and requires a datum.
447
ELECTRO-MODULATED HYDRAULICS
THE SUBJECT of electrohydraulics, where an electrically operated valve modulates an

output, has increased significantly in importance in recent years. There are aspects
associated with both open loop and closed loop control which have developed but
principally in the former area. There are many features which can now be incorporated as
standard in many electrical drive amplifiers even for the most basic modulating valve.
Some of these electrical aspects have been introduced to improve the basic performance
of the hydraulic valves and others to make full use of the potential of hydraulic actuation.
This chapter will concentrate on the aspects associated with the valve drive amplifiers
themselves and the consequences in terms of either valve or system performance. This is
not intended to replace an electronics or instrumentation text which should supplement the
material presented here. The chapter deals only with modulating type valves which can
regulate pressure or flow and not with on/off electrically operated valves. These are
broadly divided under the headings of servo-valve and proportional valves but some
common features are discussed first.
Common features of drive circuits

The drive circuits discused in this chapter are all variants under the broad heading of DC
amplifiers and are used to interface the valve with other electrical circuits. They all
conform to a general layout as shown in Figure 5.2.1, although there are many detailed
differences.
There are three principal connection routes: the demand, the supply and the output to
the valve itself. The demand is usually a voltage signal representing the 'opening' of the
valve to give the required flow or pressure. This may be derived from a number of sources
including a manually set potentiometer, via a switch between pre-set levels, a computer
system, or other electrical circuits. The supply is the power input to run the circuit itself
and supply the power requirements of the valve (depending on type between 15 mW and
50 W). The options for the supply are usually quite varied and they may be unipolar or
bipolar, which requires a plus and minus connection with a common centre ground level.
Voltages are typically in the range 10 to 30 V DC, and the power requirement will depend
principally on the valve. The output of the circuit is the connection to the valve coil (or
coils) and this must be matched with the valve. This usually requires consideration ofthe
maximum output current to correspond to the maximum valve input but voltage levels
must also be adequate.
The vast majority of commercially available valves operate with an electromagnetic
conversion from an input current to a force or torque. The current I, passes through a
copper wire winding which has a resistance R, and therefore there is a heating effect (PR).
The resistance of copper is highly temperature dependent, perhaps increasing by 30% in
normal operation, with a corresponding reduction in output if the coil is operated from a
constant voltage supply. This change could not be tolerated in an open loop system but
might be acceptable in a closed loop system since it could automatically be corrected by
the main loop feedback. However, it is a simple matter to provide the correction in the drive
amplifier directly. This uses a circuit with current feedback but this may also be called a
constant current amplifier or a voltage to current converter. The principle of such circuits
is to pass the coil current through an accurate resistor made with temperature insensitive
material and detect the volt drop across this resistor. A simple circuit showing the series
resistor, and how the volt drop is used as a feedback signal is given in Figure 5.2.2.
In this simple circuit the current through the detecting resistor is not just the current
through the coil but also includes the current through the feedback resistor. The relative
magnitude of the resistances involved means that the difference is usually less than 0.1 %.
Some of the supply voltage is also lost across this detecting resistor which represents a
small power loss. It is usually sized at about 0.5 to 1 n, in comparison with much larger
coil resistances of 10 to several 100 n (values depend on valve type), and so again this is
a small proportion. There are many variants of such circuits with various potential
improvements.
The other main feature common to many circuits is known as dither. This is a signal
ELECTRO-MODULATED HYDRAULICS 449
superimposed on the demand signal to improve hysteresis and frictional effects occurring
principally in the valve. It is usually a sine waveform, although this is not that significant,
at a relatively high frequency and at an amplitude (often adjustable) about 5 - 10% of the
maximum valve signal. The way in which this works is indicated in Figure 5.2.3.
.
This shows as an example a ramp increase in input signal starting at zero time both
without and then with the dither signal superimposed. A band around this input is intended
to indicate the hysteresis limits of the valve (shown exaggerated). Without dither there is
a delay in valve response and for any changes within the hysteresis boundary the valve
would not respond. The valve drive with dither shows that the response is improved and
will follow the input more accurately. In addition small changes in input will produce an
output in response to the sine peaks when they move outside the hysteresis band. The dither
frequency is high in comparison with the normal valve response, typically 5 times the
valve natural frequency, and is shown at a 'low' level for clarity. Although precision is not
required, the correct selection of dither frequency and amplitude can significantly affect
the success of its use.
Servo-valve drive circuits

Servo-valve circuits have been taken here to include those valves with a torque motor
input. These have two coils and are generally low power inputs in the range 15 mW to 250
mW. The word amplifier is often used to describe appropriate valve drive circuits but in
this case should not be interpreted as a large amplification in power level. It is usual to use
current drivers as described above for servo-valves but it would also be possible to use a
circuit with a voltage output. In both cases fairly conventional operational amplifier
circuits can be used quite successfully to drive such valves, although some regard has to
be given to the inductance ofthe valve with some operational amplifier output stages. The
valve manufacturers also supply suitable drivers matched to their products. These
frequently include additional circuits for the conditioning of feedback signals, providing
a dither generator and even forming a summing junction with basic loop closure gain
terms. These circuits require both positive and negative current output to match the bi-
polar characteristic of torque motors.
The normal maximum output for servo-valves is based on a rated value of current in both
of the coils. Such valves were historically operated with one coil as a demand and the other
with the feedback signal acting in opposition, and the whole providing the function of a
summing junction. It would therefore be possible to operate a valve with up to twice this
rated current in one coil alone to give the full output but this would now be unusual. It is
more normal to operate with the coils driven from the error or control signal and either
connected in series or in parallel. When in series the same (rated) current is flowing
through both the coils to give the full output but of course the steady voltage to provide
this current has to be based on the summed resistance. If the coils are operated in parallel
then the supply must be capable of providing twice the rated current (since it splits half and
half in each coil) but at half the voltage for the series connection. The latter alternative is
often favoured, since if one coil failed (not very likely these days) the second would still
provide valve operation. More importantly it provides the best opportunity of optimising
the valve dynamic response, at least electrically. Since it is easier with this configuration
to provide a higher over-voltage to help establish changes in the current through the coil
inductance more rapidly.
Proportional valve circuits

This section refers predominantly to valves which are operated using proportional
solenoids. These take higher currents than servo-valve coils, in the range 0.5 to 2 A, and
require a current controlled amplifier. The amplifiers would thus be typically rated in the
range 15 to 50 W. Use of the basic analogue circuit described above would give the major
volt drops in the amplifier circuit itself when operated at low outputs. This wastes power
but more importantly can give significant detrimental circuit heating. The reduction of this
higher power dissipation within the circuit has led to the almost universal adoption of Pulse
Width Modulation (PWM) output stages in the drive circuits which eliminates the power
wasted within the control circuit itself. Solenoids are uni-directional in their action and so
typical drive circuits are only required to provide one sense of current direction. Simple
flow throttle valves and pressure control valves will only have one solenoid but the
majority of four-way proportional valves have twin solenoid operation, each solenoid
driven by a separate circuit. Proportional solenoids are more often used in open loop
applications and many of the features incorporated into the amplifiers will enhance open
loop system performance.
The force controlled type of solenoid can be simply operated from any DC supply that
will deliver the rated current. A control resistor in series with a battery can give very
acceptable manual control of flow or pressure. There are more sophisticated joy-stick
levers available to give the similar manual input appropriate for a four-way valve.
However, all valve amplifiers work to supply a controlled current to the solenoid as
described above and they use a PWM technique.
A PWM signal is intended to obtain a mean level between upper and lower limits by a
timed switching between these limit levels. It will work wherever there is a delay between
switching a signal as an input level and the actual controlled signal finding a new level,
in other words there must be some filtering or lagging effect such as with the lag between
current and a supply voltage in a solenoid (time constant given by ratio UR). Figure 5.2.4
shows two different mean levels produced by PWM signals.
The difference between the PWM signals is the time spent at the high level in relation
to the time at the low level. The examples in the figure are shown at 25% and 75%, when
this time is split half and half then the mean value should be at 50%. This time split is
sometimes called a mark space ratio but a much better term is modulation ratio. This latter
relates the ratio of ON time to total period time, which gives a more useful linear
relationship with the resulting mean level. PWM frequencies are usually very high,
frequently in the kHz region. They should not be confused with, nor do they act as, dither
frequencies which are usually much lower.
The majority of proportional valves are constructed with some spool overlap which
gives an effect generally referred to as a deadband. Valve amplifiers usually incorporate
a deadband compensator which reduces the apparent deadband in relation to the actual
overlap. This is achieved by jumping the output signal to a value set at about 10- 20% of
its maximum for a small change in input around the zero value. This is shown in Figure
5.2.5 together with the effect on the spool opening.
The deadband compensator usually switches in at a fixed level of about 1-2%, whereas
the level of output is usually adjustable. It can be set appropriately for both the valve and
the application. There are many cases in open loop applications where a deadband is an
advantage.
Itis usually possible to adjust the gain between the input signal and the maximum output
of the driver. This does not affect the matching of the amplifier to the valve coil but rather
the value of input signal which gives the rated output. This allows interfacing flexibility
with a number of different source circuits, which may have from 1 to 10 V maximum
available. Another option with many drive circuits is to accept a current magnitude input
signal as an alternative to a voltage. There are several interface standards for conveying
information as a current, most commonly the 4 - 20 mA system. Information as a current
will not degrade if conveyed over large distances and the 4 - 20 mA system allows break
(open circuit) detection if the current falls below the 4 mA minimum.
As referred to in the chapter on Actuator peiformance hydraulic loads may behave like
ELECTRO-MODULATED HYDRAULICS 453
spring/mass systems in an oscillatory fashion. This is best avoided because large pressure
transients and noise are also usually produced. These oscillations are usually excited by
rapid accelerations or more frequently the higher decelerations possible for an actuator
and load. A deceleration (and acceleration) limit can easily be incorporated into valve
drivers and eliminates these oscillations. These limits are called ramp limits since the rate
at which a valve is ramped open or shut is controlled. In the case of flow control the rate
of spool movement gives a rate of change of flow, and since flow is equated with velocity
then this rate limit is effectively an acceleration limit. The effect of a ramp controlled
opening and closure of a valve on an oscillatory load can be seen in Figure 5.2.6.
Ramp control can be implemented as one setting, limiting all valve movements or as a
so called four quadrant controller. In this case opening and closure rates can be
independently set for two solenoids, effectively gi ving individual control of acceleration
and deceleration in both directions of travel. The ramp is also useful with pressure control,
for example a pump rapidly coming on-load can cause unnecessary noise and pressure
fluctuations. The more gradual changes possible with ramp control avoids such transients.
There are also many valves available which include position feedback - the stroke
controlled proportional solenoid options. The drive circuits for these valves not only
include the normal facilities but also the necessary supply and conditioning of the
feedback signal and the completion of the closed loop control of the valve. In most cases
this latter is likely to be more sophisticated than a simple proportional controller and may
even be a digital controller for robust performance.
Drive circuits for proportional valves are available in a wide range of configurations
from amplifiers built into plugs to sophisticated cards which include many optional
features. The range of alternatives and facilities offered with most amplifier systems is
very large and should meet most interfacing and open loop control requirements.
Additionally, monitor points and LEDs give a good indication of the correct operation of
the system and can assist with any fault diagnosis. The current actually supplied to a
solenoid, an indication of a cable break or the position of a valve spool may be accessible.
Other features available may be controlled from external logic signals, for example
external selection of demand voltage levels which are pre-set on the circuit or an inhibit
function to interlock with other safety logic. There is greater sophistication possible when
used with a PLC to set different options in different parts of a machine cycle.
Many of these features can be built into the connecting plug of the valve or on an on-
board driver circuit for either a force controlled or a feedback valve. The fullest range of
adjustment requires a rack mounted amplifier, some are very versatile including micro-
processor control (see also Figure 8.8.5). In addition there are DIN mounted modules or
individually housed amplifiers which give almost the same capability.
Electromagnetic compatibility
As discussed in the chapter on Legislation and safety, most hydraulic components and
some systems fall into the benign category of The Electromagnetic Compatibility
Directive and are thus exempt. The requirements apply to apparatus liable to cause
electromagnetic disturbance or to have its performance affected by such disturbance. They
set standards for both the emissions from any electrical device or installation as well as the
immunity of both these to interference. Perhaps the key lies in this last word. It is all about
"interference", and eliminating it between any two or more items of operating equipment
both by reducing interference output and reducing interference susceptibility of all the
items involved. The components mainly affected in hydraulic systems are the electro-
hydraulic valves (all types), their amplifiers, feedback transducers and switches, and
electric motors or power packs. Note that some of these components may also be covered
by other Directives as well. In the case of a complete installation then this may also include
the wiring and cabling.
In some respects this means that techniques for noise suppression that were used when
deemed necessary (or found necessary!) will now be routinely incorporated. This includes
the techniques of input/output protection, screening and filtering which can to some extent
be incorporated retrospectively in existing designs and products. However, the screening
particularly at cable terminations requires more stringent methods and even improved
connector types. Many of these ideas have generally been incorporated into existing
devices, however, it seems likely that future generations of products will have addressed
these aspects more fundamentally with ideas incorporated from the outset. However, most
manufacturers are developing an EMC strategy to introduce new and improved methods
including integration of control cards and transducers in valves which will improve
intrinsic immunity.
Demonstration of conformity to the Directive for most hydraulic products may take
place either through testing or through the production of a TCF (Technical Construction
File). The TCF includes a description of the product, with drawings, circuit or block
diagrams, and details of the EMC provisions with wiring layout, filters, screening and
installation details. It may also include any relevant test results. Testing requires a
specialist facility which is relatively expensive. However, it is likely that more in-house
facilities will be produced for development work, relying on specialist test houses only for
final conformance testing.
This is a situation which is still developing rapidly and likely to continue to change for
some time in the future. More detailed discussion is beyond the scope of this book and
EMC is thus best tackled by studying a specialist text.
455
PNEUMATIC LOGIC CONTROLS
THE POSSffiILITY of using pneumatic logic as a controIling technology is often

overlooked with many designers foIlowing the electrical digital route without recognising
the existence of an alternative. Pneumatic logic is undoubtedly limited in capability in
comparison with the PLC alternative but does offer some advantages. There are no EMC
(electromagnetic compatibility) problems with a pneumatic solution. There are many
cases where the intrinsic safety of pneumatic components makes them more appropriate
in hazardous or even wet environments.
There is no reason why pneumatic logic and sensing should not be used in conjunction
with hydraulic circuits although they might normaIly be used in an all pneumatic system
with pneumatic actuation. The requirement will be for valves which are pneumaticaIly
piloted as weIl as for the air supply.
Availability of logic
Logic circuitry is derived from the miniaturisation of more or less conventional valves and
often referred to as moving part logic (MPL). It uses elements operating at normal
pneumatic system pressure and capable of operating at high speeds. These now form the
basis of virtuaIly all practical pneumatic logic circuits, and have been adopted on a
relatively large scale. Properly designed MPL elements are reliable, positive in response,
and suitable for use with both dry and lubricated compressed air supplies. Also the
question of providing interconnections compatible with the size of these elements has
been solved by the development of modular construction and assemblies.
Basic logic functions can be performed by simple valves or combinations of
simple valves. Simple 2-way valves can be used in combination to provide AND or OR
logic response as shown in Figure 5.3.1. EquaIly a 2-way valve can invert an output, to
produce the NAND or NOR functions; or invert an input to provide INHffilTiON.
However, for control purposes it is also necessary to have YES and MEMORY functions
(to deal with sequential requirements). To produce a MEMORY function requires a
combination of 3-way valves, or a 4-way valve. A summary of the most commonly
available logic elements is given in Table 5.3.1. AIl MPL control systems require more
than the five basic elements listed earlier. It is possible to produce a number of auxiliary
devices from these basic gates.
PNEUMATIC LOGIC CONTROLS 457
The use of a fine restrictor valve having full flow in the reverse direction and a pressure
reservoir (or even the capacity ofthe interconnecting tubing) makes it possible to construct
simple RC time delays with good repeatability up to 30 J.ls.By combining these elements
with the NOT and YES devices normally-open (NO) and normally closed (NC) delay
functions can be achieved.
A reasonable range of pneumatic sensing devices is also available and most common
functions can be fulfilled. These include: limit switches, exhaust back pressure sensors,
bleed sensors, proximity sensors, pressure switches, vacuum switches and magnetic
proximity.
Pneumatics versus electronics

Pneumatic logic control circuits work on exactly the same principles as programmed
electronic controls. There was a time when a pneumatic logic control circuit of medium
complexity generally proved cheaper than its electronic equivalent. With the advent oflow
cost electronics, sensors and the PLC (Programmable Logic Controller) the position has
changed considerably. Apart from compactness, the advantages with a PLC is that the
programme can be re-written and sequences changed without "re-piping" the control
system. It can also be interfaced with other electronic control components, computers or
in a network. Generally pneumatic logic is appropriate for shorter distances and for the
simpler logic systems. These are discussed below.
The speed of pneumatic logic is slower than the equivalent electronic logic, however,
it is still fast in relation to the controlled system. The transmission of pneumatic signals
over longer distances does introduce significant lags due to compressibility, while the full
line volume has to change pressure. If the control line lengths are below 4 m, then these
delays will not be noticed and pneumatics could be more cost effective than a PLC. If the
line lengths are over 16 m, then pneumatic logic is unlikely to be fast enough. Between
these two distances then pneumatic logic should be seriously considered as an alternative.
Assessment of the relative complexity of an application can also be carried out. A
simple approximation can be based on the sequence (s) to be performed and the number
of inputs and outputs as follows:
Sequences = total number of individual operations in the most

complicated sequence variant
Inputs = total number of sensing elements
Outputs = total number of movement commands (eg. extend and
retract counts as two)
The complexity function can then be plotted against the total input/output number as
shown in Figure 5.3.2. This figure takes approximate allowance for features such as
installation and component cost, assembly, design and physical enclosure size and can be
used as a selection aid. If the point falls in zone A then pneumatic logic should be the best
choice. If it falls in zone C then, a PLC is the likely best solution. In between then a more
detailed examination is required unless the other factors discussed above may also come
into the considerations.
FIGURE 5.3.2 - Complexity function to select pneumatic logic or PLC [Parker Pneumatic].
Circuit design
Most MPL component manufacturers have developed their own system methods to make
logic design easier to understand and minimise circuit design time. This simply parallels
computer technology where special, and simpler languages have been devised to make
programming easier. In that case the language is tied specifically to a particular system,
which in turn is based on a particular choice of logic elements providing all the logic
functions necessary. This, like the language, can vary with different systems. It is virtually
impossible to generalise on this subject, different manufacturers of systems having
different methods of approach and component construction.
459
COMPUTERS FOR CONTROL

THE COMBINATION of hydraulic actuation and electrical control has often been seen
as an ideal partnership. The computational power and flexibility of the modern computer
systems can only improve the opportunities. Both open and closed loop control applica-
tions can benefit from the sophistication of computer control. The power and controlla-
bility of hydraulics provide the opportunities in both cases.
The computer may have a number of functions: generating the demand signal in
response to a pre-defined sequence or in response to other inputs indicating the state of
another part of a machine or as command signals; giving closed loop control; analysing
and recording data for manufacturing record or condition monitoring; communicating
with other computers or operators; acting as a design or diagnostic aid during commission-
ing or maintenance. There are a variety of ways to perform these functions offering
different levels of flexibility, cost and robustness to the industrial environment.
This chapter introduces some of the most basic concepts relevant to hydraulic systems
and gives an idea of the capabilities of computers for control. However, it should be
remembered that this capability may change very rapidly.
Basic input output interfacing

Information in the real world needs to be transferred into the computer and similarly the
result of a program execution needs to be communicated back to the real world. The
techniques associated with this interchange are usually classified as input output interfac-
ing. This can be further sub-divided into exchanging information which is essentially
digital in form or that which represents a continuously changing variable which is an
analogue signal. Digital information may represent a single item ego the position of an on/
off valve, the presence or absence of sufficient pressure, or it may represent the output from
a digital transducer or timer. Analogue information may represent the position of a piston,
its speed or the level of a demand from a joystick controller.
Obviously the internal representation of information within computers is based on
binary logic, and numbers are represented by a pattern of bits. The internal arithmetic of
a computer may work with 32-bit numbers but this resolution is not normally required
when dealing with real physical signals representing position or pressure. At the simplest
level on/off or go/no-go information can be represented by a single bit. The representation
of a full piston stroke typically may be anything from 8-bit to 12-bit, the latter becoming
more or less the standard and giving a resolution of I in 4096 or about 0.025%. This is far
more accurate than most transducers can produce an electrical signal.
The techniques of digital transfer usually depend on the number of bits of information
being handled simultaneously. If 8 or more bits are handled in parallel as representative
of say a time interval then standard interface chips can be directly connected together.
These usually work with TTL (transistor-transistor logic), in a particular logic family (eg.
CMOS) to give compatible voltage and impedance levels. The same technique can be used
for transferring single bits when they are handled together but usually these are divided
to connect to individual components and need to be established as either an input or output.
If an input, then the incoming signal must be distinctly either at a high or low level and not
wander around in between. It must be of the correct voltage levels and may also require
some noise or other filtering. The use of optical isolation is common, which prevents any
direct electrical connection between the signal source and the computer. For an output, one
of the most common available is known as an open collector, which refers to the collector
of a transistor, the base being switched by the computer. Although this is an 'output' it only
gives a ground, or earth connection when enabled, and so the device connected to the
collector must also be connected to a supply on its other terminal. There are switch type
(relay) outputs which can provide a supply but these are less widely available, except with
PLC devices.
The same distinctions must be made between input and output for analogue signals.
Outputs come via DA (digital to analogue) converters. These give a voltage output
corresponding to the digital bit pattern set on their input from the computer. The voltage
is well regulated to give the required stability but can only ever exist at the levels
determined by the bit resolution. A 1O-bit converter with a 10.24 V maximum output will
only give an output in increments of 0.01 V, which is usually more than sufficient. The
last value set is automatically retained until it is updated. The update rate being determined
by the computer.
The analogue input, AD (analogue to digital) process is rather more involved. There are
many alternative methods used, usually depending on the conversion speed required (see
specialist texts). The process is called digitisation and is considered in two parts, a
COMPUTERS FOR CONTROL 461
magnitude and a time related part as shown in Figure 5.4.1. The magnitude part or
quantisation of the analogue signal is determined by the resolution of the converter, and
hence the number of bits being used. The discretisation is the time interval between
subsequent conversions and depends on the conversion process used (and the number of
channels if a single converter is being supplied via a multiplexer with more than one
signal). This sample rate (inverse of sample frequency) can vary widely from a fls to up
to 10s of ms, and should be selected on the basis of the task to be carried out. For control
purposes this will be related to the controller update time and the number of control tasks
and calculations which are being computed.
Another aspect of sample rate is concerned with a phenomenon known as aliasing. A
digitised signal plotted on magnitude/time axes, would look like a series of points,
providing no information on what the signal does between these points. Normally the
resolution and sample rate are selected so that the points are very close together for the
signal which is of interest. However, this does not stop other signals interfering with the
process and also being sampled. This is a particular problem if higher frequency signals
(above the normal sample frequency) exist since these can reflect back down giving an
identical dot pattern with a lower frequency, as an alias. This is normally prevented by
including a low pass filter on the AD input, often called an anti-aliasing filter. Filters with
a very sharp cut-off are used.
Sequence control
The simplest control level is the logic controller which carries out a sequence of pre-
defined operations, using a PLC (Programmable Logic Controller). The majority of these
are general purpose devices but there are some hydraulic specific controllers which can
be used in this way generally giving much more than the basic features. PLCs usually work
with digital inputs and outputs requiring various actions to be completed before the next
step in the sequence. Many, including most hydraulic specific devices include the
possibility of some closed loop control as well as the basic sequencing function.
A basic PLC is a specially packaged computer system, to be both small and rugged,
intended for reliable operation in harsh industrial environments. It contains the essential
elements of any computer a central processor, memory, clock and interfacing, and may
operate from mains or a DC supply. The operating program is retained in memory,
commonly EEPROM (electrically erasable), and so disk dri ves are not required. The most
basic units may have only 6 inputs and 6 outputs but almost all devices have a modular
extension capability which can raise this to many hundreds of inputs and outputs. These
are usually digital but both AD and DA are supported by suitable modules with the
capability for either current or voltage signals. Most PLC ranges have other modules
including timers or counters, signal conditioning for temperature sensors, and communi-
cation modules including operator selection panels and displays. Most devices can be
connected in a local network to operate in conjunction with other PLCs, for programming
or for data transfer and logging functions.
The programming 'language' most widely used is known as Ladder logic, this was
originally devised for use with complex relay logic systems but has been retained.
Programming can take place as the entry of a sequence of instructions from a hand held
programmer or via a network interface. The program listing can usually be displayed either
as coded mnemonics or in the more visual ladder logic format as in Figure 5.4.2. Such
programs can usually be composed in more conventional computers (screen and key-
board), where operation can be checked and even a simulation system run. These can
include many features to assist the designer. Many PLCs will have sufficient memory
capability to contain more than one program where operating sequences are frequently
changed.
PLCs have many built-in functions including floating point arithmetic, mathematical
and statistical functions, real time calender and PID control loops. The most advanced
devices will be capable of carrying out many subsidiary monitoring, data analysis and
logging functions with considerable flexibility in their programming capability.
Closed loop control

The more sophisticated end of the PLC family above is capable of carrying out limited
closed loop control. This may be quite appropriate for some applications but there are other
possibilities including general purpose closed loop control cards, special hydraulic system
closed loop control cards and the use of a PC or workstation controller with suitable
interface cards.
Most of the control cards are referred to as motion control cards and may have the
capability to provide limited sequence control or interface with PLCs. They share the
construction features of most PLCs in being a cut-down computer but they have built-in
COMPUTERS FOR CONTROL 463
interfacing appropriate for continuous control including analogue output and either
analogue or parallel digital input. The general purpose cards frequently have parallel
digital input to receive standard rotary encoder incremental signals. The hydraulic
versions will usually have suitable interface modules which can drive the valves directly,
or include the signal conditioning for a wide range of sensors, including some of the more
common integral cylinder position sensors. An example of a dedicated card is shown in
Figure 5.4.3.
The control update times are usually faster than those posible with PLCs but this will
depend on the detail design of the card and the number of axes which are being
simultaneously controlled. The control algorithms are usually more versatile than a simple
PID, and this is certainly true for the hydraulic specials, where PID limitations have been
recognised. Nearly all of these systems will have the control programs down-loaded from
a conventional host computer (eg. PC). The language in which these are written is
particular to each manufacturer but is usually a mnemonic type form. The program can
usually be tested in the host before down-loading and sometimes on-line tuning is possible
with the control loop executing in the card. Once the card has been programmed then the
host machine is no longer required for normal operation.
The possibility of dedicating a PC, an industrial version of a PC (again packaged to be
proof for more than coffee spills) or even a single board PC, is appropriate if more
advanced control algorithms are to be written. This can have great benefits in tuning and
setting-up since there are many packages which can run identification and controller
design software even operating in a Windows environment. This means that a single
machine with a sophisticated software capability can be used to interrogate a system,
design the controller, simulate the controller and the system together and then actually
control the system on-line using exactly the same control code. This code can then be
transported to a minimal system capable of operating at the required speed. There are some
relatively inexpensive programs which will perform closed loop control and allow simple
oscilloscope type monitoring to investigate tuning alternatives. There are also some very
expensi ve systems which include more robust hardware and more sophisticated software.
The external interfacing for a PC is widely available in extension card format. The
typical card appropriate for control will contain something like 8 digital input/output lines
with two AD and two DA, l2-bit converters. The sampling rates possible should be
adequate for two-axis control if the system bandwidth is not too great and the control
algorithm is not too complex. It is possible with this type of system to implement adaptive
control schemes which keep a system's performance consitent despite wide parameter
variations. However, this level of sophistication would be better placed in a dedicated
processor and it is quite possible to use a self contained computer on an extension card
within the Pc. Such computers may be faster than the host machine and if special DSP
(digital signal processing) chips are used then the instruction set may be more suitable for
complex control algorithms.
SECTION 6
465
FAULTS AND LEAKAGE
HYDRAULIC SYSTEMS are not immune to failure. The high pressures and rugged
environments in which many systems operate are highly conducive to faults, fracture and
failure. Obviously, every effort is made to design components and systems to withstand
the expected terrain and conditions, but the unexpected frequently occurs and some means
of detection and maintenance is necessary to retain the required reliability.
This Section examines the whole spectrum of monitoring and maintenance of equip-
ment in order to provide both acceptable life in machine operation, and safety to personnel.
This first chapter sets the scene as regards the failures which may possibly occur.
Faults
There are many types of secondary fault, but basic faults can be divided into four types:
Those causing leakage
Those causing excessive temperature
Those causing malfunction of control or operation
Those causing complete stoppage of the system
These four types are explained in more detail in the following.
Leakage
Leakage of oil is not only a major expense as regards the replenishment of a costly
'component' (i.e. the hydraulic fluid), but it is also expensive as regards downtime of
machinery. Leaks have been common place on hydraulic systems for many decades, and
although the situation is improving, they are still a highly sensitive subject.
Minor leaks may not be very much in evidence, except for a fall in oil level in the
reservoir calling for unexpectedly frequent topping up, or a deposit of oil on floor or
machine surfaces. In the latter case the source is also indicated. Where the source is
obscure, it may be necessary to examine the whole run of the system logically starting at
the pump and following through the high pressure circuit and then the low pressure circuit,
back to the reservoir.
Massive leaks are generally instantly recognisable by the oil spray under pressure, or
the considerable pool of residual oil collecting under the system. If the system is working,
a large leak will also immediately indicate its presence by loss of system pressure and a
slowing down of the actuators. The pump will also become starved and cavitate as soon
as the oil level in the reservoir has fallen below its normal 'low' level. This will call for
immediate shut-down, as continued operation will only empty the system and could cause
mechanical damage to the pump.
The cause ofleaks usually involves couplings, or hose problems, or seals, or mechanical
wear - see appropriate separate chapters.
Couplings
The most common cause of leaks is faulty joints and couplings in the pipework. This
situation should improve in the future with a number of new connector designs now
available which are aimed to improve both assembly and reliability. Ifleakage occurs then
the solution will depend on the coupling design. Tightening may help to re-seal compres-
sion fittings where there is a metal seal (bite and flare most commonly) but over-tightening
a simple bite coupling will mean complete replacement of ajoint is necessary. Expedients
such as sealing compound or tape should be avoided since they are likely to cause
contamination, with far worse consequences. There are several coupling designs incorpo-
rating elastomeric sealing components and these can deteriorate or be damaged, and in
both cases should be replaced.
Hose problems
Hose leaks occur due to damage and fracture of the hose (or tube) material. This may
happen due to vibration and fatigue, or merely because of lack of protection from other
components which are moving. Occasionally failure occurs due to pressures exceeding the
maximum permissible for the hose.
Seal Damage
Leakage from actual components is usually due to damaged seals or packings, although
slight leakage may be normal from some components. If seal failure is premature, every
effort should be made to diagnose the cause of failure, as an alternative type of seal may
be indicated for improved life. Also when a seal has failed, necessitating disassembly of
the component, the condition of the rubbing surfaces on the component should also be
examined, e.g. cylinder bores. These could have been damaged as a result of seal failure
or perhaps been the primary cause of seal failure, due to deterioration or rusting. In either
case, such surfaces need re- finishing before new seals are fitted or, as an expedient to avoid
excessive shutdown time, an alternative seal material, more suitable for the degraded
rubbing surface, could be tried. Normally, however, seals should always be replaced with
similar devices and recommended materials.
External leaks are obvious and their source is easy to determine and deal with. Internal
leaks are often more difficult to detect and may involve tackling one component at a time.
In most cases it is possible to determine from the circuit diagram which ports should be
blanked off from the pressurised supply at a given condition. If such a port is opened by
disconnecting the line to it, and flow is evident from the port, then there must be internal
leakage to that port. The circuit diagram will also indicate which components are suspect
in providing a leakage path direct from the pressure line to the return line flow to the tank.
FAULTS AND LEAKAGE 467
Seal failure is less common as a cause of internal leaks. They could be due to scored
surfaces against which the seals rub, or oil viscosity being too low. In the former case the
cause of scoring should be determined if possible, and action taken to prevent a repetition
(e.g. changing the type of seal on a cy linderrod to prevent ingress of abrasive contaminants
if working in a dirty atmosphere). If low oil viscosity is the cause, this could be due to an
excessive fluid temperature.
Seal failure
Seal failure may be due to lack of compatibility of the elastomer, although this is unlikely
if specified seal materials are used. The most common causes of seal failure are:
i) Extrusion - caused by excessive pressure, lack of support and back-up for the seal,
excessive clearances, or faulty groove or gland geometry. Seals should always be
used within recommended limits for pressure (with due regard to intermittent peak
pressures). Extrusion or wedging can normally be eliminated by reducing clear-
ances and/or providing the seal with back-up rings.
ii) Cracking- may be due to age-hardening, physical deterioration, thermal hardening
at very low temperatures, excessive heating due to high friction (lack oflubrication
or too tight a fit) or abrasive wear. Age-hardening can occur during long idle
periods, particularly at low ambient temperatures. Abrasive wear is primarily
caused by the mating metal surface finish being too rough. Elastomeric seals
normally demand a surface finish of 0.4 Ilm or better for good life.
iii) Spiral twisting - normally limited to O-rings and usually caused by side loading
(e.g. reciprocal motions). This condition can often be relieved by the use of a back-
up or gland ring.
iv) Suiface damage - abrasion and wear caused by a rough mating surface or sharp
edges on grooves or back-up rings, or fluid contamination.
Where there is definite evidence of surface damage or roughening, the condition of the
surface on which the seal runs should be checked. Surface finish conditions for satisfactory
seal life are summarised in Table 6.1.1.
Static seals
Static seals are generally less troublesome than dynamic seals as regards development of
leakage, although couplings can be a source of trouble if pipe work is subject to marked
vibration. Couplings incorporating O-rings or toroidal rings have vibration damping
properties, but elastomeric rings are likely to deteriorate if subject to high temperatures
(above 100 DC) and can fail as a consequence. Where such temperatures are likely to be
realised elastomeric static seals can be replaced with metal wedge seals.
Bonded seals and bonded washers provide excellent, long-lasting seals in themselves
but centralising can sometimes be a problem if there is no centralising shoulder on the
coupling itself. Alternatively an internal washer is sometimes used, in which case it should
be in a resilient material, e.g. PTFE. A plain steel washer used between a union head and
a bonded seal will introduce a leakage path. There are other reasons why a bonded seal may
not seat correctly and thus lead to leakage. Such leakage problems are attributable to other
items than the seals or washers themselves.
Mechanical wear
Sliding of parts within hydraulic components is normally lubricated, e.g. within a pump
or valve, say. However, where the lubricant is starved or the loads excessive, wear may
result and lack of precision occur. This tends to be a gradual fault with an ever increasing
drop in efficiency of the component and system as the leakage builds up.
Excessive temperature
Excessive temperature may be considered to have occurred when an unprotected compo-
nent is too hot to touch, too hot to operate successfully, or when the oil is irreversibly
changed in property (for the worse!)
Steady operation of a system invariably requires a stable temperature perhaps between
40C and 80C although working temperatures of above 80C are possible.
Above 80C, the higher the oil temperature the more rapid the onset of oxidisation and
the development of degradation products, as the oxidisation inhibitors are used up.
Possible causes of excessive oil temperature are:
i) Use of oil with too high a viscosity - This causes an excessive load on the pump,
with lowered pump mechanical efficiency and an excess proportion of the input
power being transformed into heat. .
ii) Bypass or relief valve set at too high pressure- This allows the pump to overwork
and generate excess heat, which is transferred directly to the fluid.
iii) Pump not suitably off-loaded - This allows the full input power to be developed
in the form of fluid heating.
iv) Insufficient fluid in the reservoir.
v) Internal leakage - This may be due to wear on the pump. Equally, a fluid with too
Iowa viscosity will cause increased leakage flow, but not necessarily over-heating
unless lubricity is low and excess mechanical friction results.
vi) Unloading valve blocked - The unloading valve to the tank or accumulator may
be blocked, and not operating correctly.
FAULTS AND LEAKAGE 469
vii) Restrictions in lines - Damaged, kinked or partially blocked lines may introduce
excessive load on the pump and consequent over-heating.
viii) Output motions obstructed or stalled giving excessive flow through relief valves,
with high heating through these restricted flow paths.
ix) Excessive flow velocities - High flow velocities may promote high frictional
losses and over-heating.
x) Reservoir too small- Oil volume is insufficient to allow normal cooling during
cycling, hence there is a gradual build-up of temperature.
xi) Insufficient normal cooling - Lines, etc. should not be located in high ambient
temperatures where they may be subject to heating rather than cooling through
normal radiation.
xii) Inadequate inter cooler - Faulty, too small or badly positioned.
Malfunction of control or operation

Contamination in the form of tiny solid particles can cause considerable harm to a system
(see Chapter on Contamination control). Such debris can cause wear, but it can also result
in the blockage of small orifices in control valves, as well as eventually building up in
larger spaces to reduce flow and operation.
Sliding spool valves can be jammed solid and become totally inoperative simply
because of the presence of small silting particles in the hydraulic fluid.
Electrical (or electronic) problems may be considered in this category.
Complete stoppage of the system

Anyone of the above faults could cause a complete stoppage of a system; they, and others,
are summarised in Figure 6.1.1 and Table 6.1.2. However, in general, the fault is usually
(but not invariably) apparent before such stoppage. The complete stoppage means that
maintenance must be undertaken immediately - there can be no delay to a more convenient

time.
Complete stoppage also occurs due to mechanical faults, such as a dri ve shaft fracturing
or an accident puncturing a vital supply line.
471
INSTRUMENTATION
INSTRUMENTATION IN hydraulic systems is used not only to assess the output of a

process but also to determine the health of the system. This chapter deals with the steady
state monitoring of pressure, flow and temperature, and the calibration of instruments. The
following chapter on Testing systems and components looks at equipment and test circuits
which are used to test for the condition of systems and components after they have been
built. A third chapter Monitoring techniques discusses various means of assessing the
health of hydraulic applications.
Pressure
Pressure may be determined by means of a number of different devices. The whole
spectrum is best appreciated from Table 6.2.1.
Of those shown, the most common form of pressure sensing in hydraulic systems is the
pressure gauge. It is described in more detail below. It is visually easily understood and
responds with reasonable rapidity. For remote sensing, however, it is too cumbersome and
expensive - involving long lengths of piping - and it would normally be replaced by a strain
gauge sensor. Also a pressure switch may be more appropriate than a gauge.
For very low pressures the glass manometer could be used, but this is rare in hydraulic
power applications.
'Pressure', strictly speaking, is the difference in pressure between two places; in
hydraulic systems it is normally quoted as a gauge pressure. However, it may be a
differential pressure or an absolute pressure. These three are defined below:
PD - differential pressure - pressures taken from two places in the system, e.g. either
side of a filter.
PO - gauge pressure - pressure taken from the system on one side and from the
atmosphere on the other side (can be vented or sealed).
PA - absolute pressure - pressure taken from the system on one side and connected
to a vacuum on the other side (or sealed with a vacuum).
Pressure gauges
The most common types of pressure gauges used on hydraulic systems are the Bourdon
gauge and the spring-loaded plunger type gauge.
Alternative instruments with a higher accuracy, or higher range, than that offered by a
Bourdon gauge are the precision Bourdon gauge or Budenberg gauge, but both are rather
delicate. They would normally only be used as standard or test gauges, although their
future in this respect is undoubtedly rivalled by the electronic pressure sensor.
A typical Bourdon gauge is shown in Figure 6.2.1. The accuracy obtained is reasonably
good, although inclined to drift with use. British Standard BS 1780 specifies that the
accuracy of such a gauge when new should be within plus or minus I % of the maximum
graduation. In practice it is safer to assume that the likely accuracy of a Bourdon gauge
is of the order of plus or minus 21/2% of full scale.
The gauge used should have a suitable full-scale deflection consistent with the pressure
which is to be measured, i.e. the normal readings taken should lie at the upper end of the
scale. If not, the possible error will be proportionately greater. Also, gauges should be of
generous size. A minimum dial diameter of 50 mm (2 in) is usually recommended, with
a preference for 75 mm (3 in) or even 100 mm (4 in) diameter.
INSTRUMENTATION 473
The accuracy of the gauge will also depend on how it is installed, the conditions to which
it is subjected in service, and, particularly, how well it is protected against pressure surges
and mechanical shock. The latter can seriously 'shift' the calibration.
For continuous 'in circuit' use the gauge should ideally be mounted on a rigid surface
isolated from mechanical vibration produced by the pump, valves or machine, which
could affect the delicate mechanical movement. Mounting on a separate instrument panel
is a logical solution, but there is no basic objection to mounting the gauge on the machine,
or even on a rigid pipe, provided the mounting point is free from vibration. Complete
isolation, however, is preferable.
Normally Bourdon gauges are mounted with 'snubbers' or 'gauge savers' in the gauge
connection line. These are essentially throttling devices which restrict flow and thus
dampen pressure surges under conditions of fluctuating pressure. They may be pre-set or
adjustable by means of a needle valve or fixed orifice.
For complete protection against pressure surges a pressure gauge can be isolated from
the circuit via an isolating valve which holds the gauge at zero pressure. The valve is then
operated manually to bring the gauge into circuit when a reading is required. On release,
the valve returns to its isolating position.
Other types of valve are also produced where the gauge is normally left continuously
in circuit but automatically isolated by the valve on the appearance of surge pressure
peaks.
An alternative approach is to fit self-sealing couplings at points in the system where
pressure readings are required. A single gauge can then serve a complete system, being
itself connected to a matching coupling and simply plugged in at various points in turn to
obtain pressure readings. No protection is provided should a pressure surge occur when
the gauge is actually plugged in; a 'snubber' can be used in the gauge line.
The spring-loaded piston gauge is more rugged and largely shock-proof, i.e. it will
withstand pressure surges and shock loads. Also such types can readily be designed for
pressure up to 17 MPa (170 bar) and can be used anywhere for continuous measurement
of pressure. The accuracy of this type of pressure gauge is not as good as that of a Bourdon
gauge, however, nor is the ultimate pressure range so high.
Uses ofpressure gauges

Pressure gauges may be fitted purely as a nominal check that the system pressure is of the
required order, relying on the gauges to indicate the presence of some faults should the
readings fall or rise unduly. In a properly designed system, however, pressure gauges
should be incorporated (or at least provision made to plug them in) at all points where
pressure measurement can be of real value for system tuning or fault finding.
Pressure monitoring is particularly useful in the case of complex circuits. In this case
it will usually be found that the most important 'pressure points' are in the pilot system,
as, if a main valve fails to operate, they will immediately show the possible reason.
Pressure tappings can be taken from the line or passage between the pilot and main valve
or the main valve may have a special port which, when the valve operates, opens to the line
leading to the pressure gauge. Care must be taken to avoid the likelihood of flow jets at
the pressure point, if static pressures are required.
Calibration and checking of pressure gauges

To ensure continuing accuracy pressure gauges should have their calibration checked
periodically. The general recommendation is an annual check. Gauges should be checked
monthly, or at least every two months, if readings are to be relied on. This applies
particularly to gauges which are used for circuit 'tuning', whether or not they are in
continuous use.
The simplest method of checking is to use a portable hand-held pressure calibrator
(Figure 6.2.2). If its use is high then a dead weight tester is a good investment for the
INSTRUMENTATION 475
instrument or test shop. (The dead weight tester is used for calibrating the hand-held
calibrator.). The dead weight tester is the standard form of pressure gauge test apparatus
comprising a screwed ram and a piston weight platform assembly (Figure 6.2.3). The
instrument is filled with oil, which is then pressurised to the required level by placing
weights on the platform. The screwed ram is rotated until sufficient additional pressure is
applied to lift the weights, and the weights then spun to ensure freedom of error from
friction. With the weights spinning the pressure in the system is at the desired value and
the gauge can be read. The accuracy of dead weight testers can be up to 0.03% of reading,
the hand-held calibrator may have the reduced, but nevertheless high, accuracy of, say,
0.05% full scale. ~ .....
If a gauge while still on the machine does not return to zero when pressure is released,
it should first be checked by undoing the pipe connections at the gauge itself. If the pointer
then drops to zero the trouble is due to air trapped in the gauge tube and will probably
disappear in a short time. The readings at higher pressures are not affected. If the pointer
does not return to zero, the gauge should be corrected or re-calibrated. A common cause
of zero errors on hydraulic gauges is the sudden release of pressure which strains the
mechanism. If it occurs, one of the protective measures mentioned earlier can be adopted.
Flow
The measurement of flow in hydraulic systems is almost as important as pressure.
However, when it comes to the instrumentation available there are considerably more
possibilities than with pressure. The choice will depend on such features as the application,
retrofitting, accuracy required, available funds, type of fluid, range of flows to be covered,
etc. Table 6.2.2 lists and describes 18 different types of flow device.
INSTRUMENTATION 477
TABLE 6.2.2 - Flow measuring methods (continued).
Range Accuracy %
(Examples at at full
Method only) reading scale Comments
Oval gear 30 mUmin to 0.15 % Excellent meter for chemicals
wheel [PO] 16,000 Umin and batching
Paddle wheel 10: I 2% Low cost insertion type.
Velocity Tolerant of particles.
from 0.1 Me;;r~ flow ~Iocity in region
to IS mls of wheel.
Pelton wheel 10: I 0.3 % Ideal for ultra low flows.
(Hall effect) 2 mUmin to Non lin;;r.- - - - - -
20,000 Umin
Piston 1000: I 0.2 % Good for viscous fluids.
[PO] 1.0 mUmin Very accurate at low flow for
upwards petrol.
Rotating Up to 75 mm 0.15 % Very high accuracy.
vane [PO] diameter pipe Petrol pumps and aircraft.
Turbine 6 mm to 0.5 % Low viscosities preferable.
(JR, magnet 600 mm Fiscal meter.
or Hall effect diameter Damaged by particles
detection) > 100 !lm, and needs a strainer.
Ultrasonic Velocity 2% Easy to clamp onto pipe.
- Clamp-on from 0.02 Fluid needs bubbles or particles
(Doppler) mls upwards > 30 !lm
Ultrasonic Almost zero 2% Independent of conductivity.
- Clamp-on to 15 mls Can be retrofitted.
(Transit- Best with flows above QI mI~
Time) Temperature and pressure may
affect the result.
Ultrasonic See above I% See above
- Dedicated
Variable area 10: I I-6 Low cost for small bore pipes
(Rotameter) 0.2 mUmin (up to 75 mm)
to 750 Umin
To 75 mm
bore.
Vortex 10: I I% Good repeatability.
shedding 5 Umin High pressures.
upwards. No moving parts.
25 mm to Pre~re drop may be low:- -
0.8 m full Best with relatively clean low
bore, but can viscosity fluids with Re > 5,000.
be inserted.
PD = Positive Displacement. Below - - - Variable advantage. Below --- Disadvantage
INSTRUMENTATION 479
Temperature
Temperature changes slowly, so may be measured externally to the system fluid, e.g. on
the surface of a component such as a pipe (ensuring good contact), or if the fluid needs an
accurate assessment an insertion type of temperature sensor may be used. Table 6.2.3
shows a variety of possibilities with the range of temperatures which can be covered
currently with each type of device. Temperature sensitive paints or crystals can be also be
used for set values or maxima.
Other instrumentation
There are other steady state functions which are important to measure. These include the
level of reservoirs, the characteristics of the hydraulic fluid, rotational speed of the dri ver,
etc. For these the reader is directed to books on the condition monitoring of mechanical
and hydraulic plant.
Instruments which combine several functions in one unit are now extensively available.
The most common will be those which sense pressure, flow and temperature with one in-
line module - inserted in the pipeline carrying the critical hydraulic fluid. By means of
suitable interfaces or models it is possible to cover the complete industrial range of
pressure, flow and temperature - Figure 6.2.4 shows one example. Several sensors in a
permanent system, computer controlled and logged, are shown in Figure 6.2.5.
PC100 Data Aquisition System

for IBM compatible PC's Th C
e P 100 data aqulsillon
Provides: sysl8fll comprises:
16 pressure Inputs
12 flow Inputs Plug in boards D A signal COnditioning unit
Plug In bosrds lor IBM
compatible PC's
Flow, pressure, speed "
18fIlperature sensors
R;\\'-~~
~QQ ////
//,
...
Signal conditioning unit ~
~/ y~ ~~~
ormagnellc~ sens,
Flow
BI-dlrectlonal llow melers with
built In temperature sensor
Flows 0.2 10 800 litrealm In., pressures to 500 bar
Pressure
Pressure transducers
ll-520 bsr
Pesk 780 bar
r
FIGURE 6.2.5 - A sensor data acquisition system [Webster Instruments] .
Hydraulic System MUlTI-

FUNCTIONTESTERS measure
and display static pressure,
peak pressure, fluid
temperature and flow with
data logging if required .
Hydra-Test TEST POINT
EQUIPMENT allows random
checks on fluid pressure and
temperature or fluid sampling
anywhere in the system.
HYDRAULICSYSTEM
PRODUCTSLIMITED HYDRAULIC VALVES for flow
II
MoncktonRoad and pressure control; in-line,
I Wakefield,'
West Yorks,WF2 7AL
Tel' (01924) 364748
cartridge and mini-cartridge
series. Pressure gauge
prot~ct~on,.s~ubbers and
Fai: 101924) 290450 mulb-C1rcUl"so'ators.
481
TESTING SYSTEMS AND

COMPONENTS
THE TESTING of hydraulic systems depends on the type of system and its operational
characteristics. Each situation is different and normally will rely on the attachment of the
most appropriate instrumentation mentioned in the previous chapter; however, some
smaller complete systems can be tested by suitable rigs. This chapter looks first at the
possibility of rig testing, then at the testing of each component; finally, examples are gi ven
of systems under test.
Rig testing
Each rig can be designed to test specific components or complete systems.
A rig is normally specific to the individual system requirement, such as performance
data or endurance, and thus is both time consuming and costly to construct. It can,
however, be fully justified for prototype testing where
i) the system is critical either in terms of design or performance requirements, or,
ii) the cost can readily be recovered by standardising a proven design for economic
production of a sufficient number if it is not likely that further changes will be
needed in the light of service experience.
In the latter case, besides proving the system design, the rig may also be useful for the
evaluation of the prototype design of individual components.
Besides providing performance-proving data, such testing can also establish wear and
life characteristics of indi vidual components, as well as ultimate life of critical systems.
In many cases the latter can be governed by a single small component rather than by the
fatigue limits of the main structure or structural components.
A further use for such rigs, after prototype proving, is for the running-in of individual
components, although this can equally well be done on much simpler rigs for production
components. This has been found desirable in the case of many high-performance
components manufactured to very close tolerances which can sustain a high degree of
wear during the first few hours of use. Such components can be fully run-in, then, if
necessary, stripped down and examined before being assembled in the system proper.
Testing pumps
A simple set-up for testing the performance of pumps is shown in Figure 6.3 .1. A particular
point to note is that if the tester is inserted between the pump and its relief valve (as in
diagram b) it is no longer protected. The pump should not be started unless the tester
pressure control valve is open and pump pressure (as indicated on the tester) must not be
allowed to exceed the maximum setting of the relief valve. This test mode enables pump
delivery to be measured on no-load and maximum (pressure) load. The difference between
the two is an indication of the state of the pump.
Tester positions for testing the relief valve are shown in Figure 6.3.2. The tester load
valve should be in the no-load position before starting up. After a run up to system
temperature the load valve is closed until zero flow is indicated, when the pressure gauge
will show the system relief valve setting required. Irregularity in indicated pressure
reading will show a leak in the intake side of the pump, or cavitation. Flow rate indicated
on no-load should approximate to pump rated delivery. The flow rate at 50% pressure
setting should be the same - any marked difference indicating a relief valve fault.
Testing valves
Testing facilities for checking the directional control valve (selector) in the circuit are
shown in Figure 6.3.3. The flow rate with no-load should be equal to the rated pump
delivery. The same flow rate should be indicated when the pressure is adjusted to near
maximum system pressure. Any marked loss in flow rate means that there is a leaking
selector.
Testing cylinders
A relatively simple rig can be constructed to test hydraulic cylinders (see Figure 6.3.4).
Components required are a motor-driven hydraulic pump drawing fluid through a
reservoir, a directional control valve, filters, and a pressure control valve located in the
circuit between the discharge filter and the directional control valve. For testing, the
direction of working of the cylinder is repeatedly reversed by the directional control valve
using any suitable switching arrangement. Testing is then continued for the desired
number of operating cycles to be evaluated.
Similar considerations apply for the determination of actual operating rates, a series of
tests being taken at the different flow rates required. Practical measurement will determine
realistically the effects of friction and back-pressure on performance, and will establish
the true flow rate required. Further, a test rig provides scope for investigating the value,
or necessity, of fitting restrictors in forward or return lines, or both, in order to arrive at
optimum movement times, or to assist in synchronisation.
Testing pipes and tubes

The strength of pipes and tubes was evaluated in the past on purely empirical lines -
working pressure ratings were based on pressure test figures with a suitable factor of safety
applied This method of 'proof rating may still be used for non-ferrous metal tubes, non-
metallic tubes, and non-seamless tubes.
A distinction can be drawn between a pressure test to establish an ultimate bursting
pressure for a particular tube and a true 'proof test which is employed merely to 'prove'
that the tube can withstand a suitable higher working pressure than the normal working
pressure rating. In the former case the tube is tested to destruction and figures for a number
of different specimens will tend to show appreciable scatter. This is not necessarily
significant since once a mean bursting pressure figure can be established the maximum
working pressure rating will normally incorporate a safety factor of at least three, and
usually very much higher in the case of cast or fabricated tubulars.
'Proof' pressure testing implies that a particular specimen will be tested at some figure
above its maximum working pressure rating to establish that the safety factor value is at
leastthat of the ratio 'proof pressure/maximum rated working pressure. Theoretically, the
proof pressure selected could be anything between the maximum working rating and the
burst pressure, but if the component is to be used in the system after proof testing, it is
important that the proof pressure employed should be less than the yield point of the
material, otherwise permanent weakening may result. Yield point is clearly established in
the case of ferrous metals as the point where the stress/ strain curve departs from linearity.
Non-ferrous metals may not exhibit a definite yield point, in which case the limit is best
set by the O.I % proof stress for the material. In the case of non-metallic hoses where neither
of the aforementioned apply, the 'proof' or testing pressure is normally set at a specific
fraction of the maximum rated working pressure, the actual factor employed depending
on the construction involved. (See the listing of hoses and tubes in the chapter on
Standards.)
Testing complete hydraulic systems

Mobile system
As one example of a complete system which needs testing, consider an aircraft. This can
be tested by a mobile hydraulic system or a static test system. Figure 6.3.5 shows an
appropriate mobile hydraulic test system for an aircraft overhaul base, typically comprising
a static power installation with a 110 kW power unit (1)
supplying a motor/pump test unit (2)
with an electronic control bench (3)
which automatically displays and prints out test parameters and a component test
bench (4)
supplemented by mobile hydraulic test console (5) which are coupled to an aircraft
for system tests.
Mobile petrol, diesel or electrically powered trolleys (6) and (7) provide comple-
mentary test facilities inside or outside the hangar.
TESTING SYSTEMS AND COMPONENTS 485
All connections from the pumps to the rest of the trolley system are made by means of
flexible hoses to minimise the effects of pulsation and noise throughout the rest of the
circuit. This arrangement together with a pulse attenuator downstream of the main pump,
dampens amplitude of pulsation by about 90% with a reduction in noise of up to 19 dB.
Filters in use are rated 1 f.lmto 3 f.lmto conform with the high cleanliness requirement of
fluid used in such systems; they are of the non-bypass full strength element type fitted with
differential pressure switches to give visual indication at the control panel that the
elements require changing, and to shut down the trolley, automatically, when the elements
become blocked.
Pressure control over the trolley delivery outlets is achieved by pilot-operated pressure-
compensated reducing valves which enable the pressure at each outlet to be adjusted
within the pressure range of the trolley. Return line back-pressure in each line is adjustable
by means of pneumatic operated pilot back-pressure valves within a range of 0.2 - 1.7 MPa
(2 - 17 bar), the setting being governed by pressure requirements of the aircraft reservoir.
A pre-charged nitrogen cylinder supplies the pneumatic pilot pressure.
An example of a static power installation is shown in Figure 6.3.6 The power unit, the size
of which is unrestricted by the requirements for portability, is housed in an isolated, sound-
proofed power-house which contains filtering, cooling, and de-aerating units. Remote
control is provided in the hangar, and with the addition of automatic protective devices,
direct supervision is not required during operation.
A typical power-house would accommodate the following items of equipment:
A motor (diesel or electric) driving two axial-piston, constant-delivery, high-
pressure pumps.
A large main hydraulic fluid reservoir open to the atmosphere.
A smaller sealed tank to supply filtered, de-aerated fluid to the main pumps.
Hydraulic filters (l - 10 11m).
Electrically driven pumps to circulate oil and water through an oil/water heat-
exchanger.
A fan-cooled water-cooling unit.
A main control unit containing motor switchgear, electrical controls and system
indicating lamps.
Connection points for an auxiliary power source.
Hangar equipment would comprise.
A coupling point with self-sealing couplings installed in each of several isopits
(Industrial Services Outlet pits).
An additional auxiliary supply point mounted on the hangar wall, with self-sealing
couplings but without an accumulator boost unit.
Return tank and pump units which pump the return fluid back to the power unit.
An electrical control panel carrying a duplicate set of system controls and
indicating lamps.
Emergency stop buttons.
Mobile consoles which connect to the aircraft and the isopit connections and which
enable the test supplies to the aircraft to be individually controlled.
Fluid from the main reservoir is cooled and filtered by the oil-circulating-and-cooling unit
and then passed to the sealed reservoir of the power unit. Water for the cooling unit is fed
from a water-mains cistern which can be heated in extremely cold conditions by a I kW
heater, but is normally cooled in a cascade type cooler with an electrically driven fan for
additional cooling.
The power provides a high-pressure, high-flow supply to the wall-mounted connection
and the isopit, to which mobile consoles are connected by flexible hoses to interface with
the aircraft. If required, two consoles may be operated from one isopit by using a simple
flow-dividing manifold which has three pressure and three return self-sealing quick-
release connections. The consoles provide control, filtration and sampling of the fluid to
and from the aircraft's system. Fluid returning from the aircraft to the isopit is pumped
back to the main reservoir by two return tank-and-pump units.
A wall-mounted junction box consisting of two stop-valves and two connection
TESTING SYSTEMS AND COMPONENTS 487
manifolds may be used to connect with a servicing trolley, such as that used to service
helicopters, using a different type of fluid from that of the SPI. The helicopter servicing
trolley is powered by a hydraulic motor driven by pressure fluid supplied by the SPI.
Component test benches may be similarly powered.
489
MONITORING
THE MONITORING of hydraulic equipment and systems is not quite the same as the
monitoring Qfbasic mechanical systems. Mechanical systems, in general, produce much
'cleaner' signals of distress and are successfully monitored by basic vibration techniques,
wear debris analysis or simple steady state techniques.
The difference with hydraulic systems is the fluid. The fluid, be it a thick mineral oil or
water with its much lower viscosity, has the considerable effect of damping any
mechanical signals and vibration monitoring may not be the best. Fortunately, there are
many other techniques which are available, and some of these can be highly successful.
An earlier chapter, on Faults, has already highlighted the key areas of failure likely to
occur in hydraulic systems. This chapter now outlines several monitoring techniques
which may be able to detect the early signs of deterioration in a system - in time to prevent
any catastrophic failure occurring.
Monitoring techniques
The following ten types of monitoring technique are briefly explained in terms which
relate to hydraulic systems. The purpose of these brief descriptions is to enable the reader
to appreciate that there are a number of different possibilities, and by careful examination
of the system and consideration of the techniques available, then the most appropriate
monitor or monitors can be chosen. Greater detail about all these techniques is given in
more specific books, particularly in those covering the range in terms in hydraulics and
fluid power:
Human
Vibration
Ultrasonics
Wear debris analysis
Oil analysis
Thermography
Leakage
Corrosion
Steady state analysis
Performance
Human
For as long as there have been hydraulic systems, there has also been human monitoring.
The human being is able to detect 'changes' in the normal, particularly changes in the
MONITORING 491
sound level and its harmonic content. He or she is also able to use the other senses with
great success after sufficient training and experience. This is not to be scorned. It is highly
valuable, and a forman who decides to pass over the warnings of an experienced worker
who senses something wrong, may do so at possible considerable cost.
Consider Table 6.4.1. Many of these examples are frequently met in practice.
Vibration 1 Hz to up to 25 kHz)
Vibration monitoring may not be as effective in hydraulic systems as it is on mechanical
systems, but, nevertheless, it is still very valuable for certain aspects.
Vibration sensors detect alternating changes in position or pressure; they are thus
normally accelerometers or dynamic pressure sensors. Vibration may also be sensed from
other transducers such as velocity or displacement sensors (for lower frequencies) or from
torque transducers for shaft drives. The measurement of acceleration, however, tends to
give the best overall monitor.
The type of appropriate application in hydraulic systems is that which relates quite
strongly to mechanical components. Pumps, particularly gear pumps, but also piston and
vane, are possibilities, as are components which experience shock, such as cylinders and
valves.
Analysis of the captured signal can be quite simple, such as the rms value (i.e. the
general level of vibration - almost the sound level), or highly complex where many
components are analysed and sorted electronically in a kind of 'fingerprint' or 'signature'
analysis. Harmonic analysis involving a Fast Fourier Transform (FFT) is quite common
because this may identify the differences in a number of components at the same time
(provided the damping is not too high); Figure 6.4.1 shows an example of a pump
deteriorating where the 1st (fundamental) and 3rd harmonics rise with wear.
Ultrasonics (20 kHz to up to 100 kHz)

The use of ultrasonics has not been as extensive as it could be. Suitable sensors are able
to pick up these very high frequencies without being confused by the normal audio
frequency signals which tend to proliferate in any working system. The problem arises in
that many features may be showing signs of load (and fatigue) and not just one component
The following is a list of hydraulic components which may be detected in this frequency
band:
Valve and valve seat wear
Cavitation in pumps
Crack propagation
Pipe suction leaks
Intemalleaks within a component
Reservoir leaks
Liquid starvation
Complete pump failure - but that is somewhat late!
Figure 6.4.2 shows an example of an ultrasonic monitor based on a 40 kHz frequency
which is able to detect fluid leaks as low as a few parts per million for both gaseous and
liquid systems.
Wear debris analysis

The debris generated by a wearing component is quite often passed round the hydraulic
system in the oil. True, a filter may remove it but usually not all of it Because this evidence
is circulated, it is relatively easy to fit a monitor in one part of the system and be able to
detect the different evidence of faults in various places in the system.
The use of debris monitors has already been explained in the chapter on Contamination
control under the heading of 'assessment of particle concentration'; however, wear debris
MONITORING 493
is a slightly different subject because it involves not only non-metallic debris, but very
much the metallic debris. This metallic debris is usually classed as either 'ferrous' or 'non-
ferrous' because the detection method may use magnetic forces which are only influenced
by the ferrous part of the debris.
There are numerous instruments available for the examination of the hydraulic fluid for
debris. Some of these are able to work 'on-line', i.e. they have the ability to look at the fluid
as it passes through a pipe or hose. Others work 'off-line' by examining a sample of oil
taken from the system (like a blood sample in the human being). Whilst the on-line device
is less likely to give an erroneous result because there is no transfer of the fluid, the off-
line device is able to present the debris for deeper analysis such as by eye (microscope)
or elemental analyser. Figure 6.4.3 shows an arrangement for the examination of the oil
optically, electronically and chemically, looking at oil degradation, oil contamination and
wear debris.
The metallic debris is likely to be fine 5 11m) in the early stages of wear. This is
probably acceptable as a running-in process. However, when the amount generated begins
to rise, then action should be taken to correct or change the appropriate component. If
larger particles around 100 11mor greater are observed then fatigue may be occurring.
Oil analysis
The analysis of the hydraulic fluid is the next step after wear debris analysis. Although the
debris in the oil may provide the greatest evidence of faults occurring, or of wear
beginning, the oil can reveal many other aspects related to faults. Consider the possibilities
shown in Table 6.4.2.
The monitors used for detecting the various parameters shown in Table 6.4.2 are usually
off-line; however, some may be available on-line, such as viscosity, density and water
content. Temperature, of course, can only be satisfactorily monitored in the line of flow.
Thermography
Thermography is the analysis of a component by observing its surface temperature pattern.
A small, or large, area of the component is examined by a heat sensitive video camera set
up to operate over a certain range of temperature. The image produced portrays the range
of temperature by colour or grey scale.
Whilst thermography is more often used on electrical and mechanical systems, its use
on hydraulic systems is not ruled out. For instance, it should be able to detect 'hot spots'
caused by excessive friction in valves and pumps and actuators; it would do this by
showing the change in thermal pattern from the normal one to that currently being
experienced.
One great advantage of thermography is that it is used remotely - with no connection
to the system. Figure 6.4.4 shows its use in a marine application where large areas can be
inspected in one operation by 'panning' over the equipment in one pass. Images stored by
the portable video recorder are then downloaded as Polaroid photographs and compared
with previous records. Prime detection has been in effects caused by vibration and salt
water corrosion resulting in loose joints and leakage.
MONITORING 495
Leakage
The measurement or monitoring ofleakage is a varied subject. It can be as simple as noting
a puddle of oil on the ground, or as complex as detecting the internal leakage within a valve.
This range ofthe detection ofliquid leakage is indicated in Table 6.4.3. Not all leakage
is critical (as mentioned in the chapter on Faults) but the effectiveness of its detection
means that the quantity ofliquid leaking must be determined in order to assess if it exceeds
the cri tical 1imi t.
Corrosion
There are a large number of monitoring methods for corrosion, as listed below:
Crack monitoring by ultrasonics, eddy current, magnetic particle inspection (MPI),
magnetic flux leakage or liquid penetrant.
The weight loss method (with a coupon)
Thickness measurement (with or without a coupon)
Electrical resistance (ER)
Thin layer activation (TLA)
Zero resistance ammetry (ZRA, or Galvanic Coupling)
Linear polarisation, or DC polarisation (linear polarisation resistance measure-
ments - LPRM)
Electrochemical impedance monitoring (ElM)
Electrochemical impedance spectroscopy (EIS)
Electrochemical noise (EN)
Harmonic analysis (HA)
Field signature method (FSM)
Biological activity monitoring
Corrosion in hydraulic systems is mainly a problem where high water based fluids are used
(or rather misused) and in marine applications. Whilst it is not normal to fit corrosion
MONITORING 497
monitoring on an hydraulic system at build, itcan be very important when a problem arises.
The type used depends on the situation, but if a probe can be inserted then LPRM, or the
more advanced ElM, EIS and EN may be very suitable.
Steady state analysis

'Steady state' monitoring is the sensing of features which can be measured with basic
gauges. Pressure fluctuations are described in the chapter on Maintenance. Pressure, flow
and temperature have already been discussed in the chapter on Instrumentation: perform-
ance will be discussed in the next section here, but there are other possible steady state
sensors such as level, speed, load, torque.
These are all easily understood parameters, and if they should vary significantly then
action should be taken to correct the situation.
Performance
Ultimately any deterioration of the hydraulic system results in a drop in efficiency, i.e. for
the energy put in the output is less - the performance drops. In other words, both input and
output really need monitoring if performance is to be a measure of condition.
Performance, in itself, could be checked from a rotational speed, or pressure, or flow.
It could be assessed from the ease of manual control, or even the sound level. It could be
measured from the actual output of the system - the rate of production or accuracy of
finished product.
The input to the system may be the fuel used in terms of litres of diesel or current taken,
power absorbed, etc.
Efficiency of a pump has been measured by measuring the temperature rise of the
hydraulic fluid passing through the unit; the same has also been seen in a valve operation
where wear causes an increase in temperature.
Great care has to be taken in using this type of monitoring. The prime confusion results
from the measurements being taken at a different set up of operational parameters. Another
problem is that variations in a system do happen acceptably. Any comparison can only be
meaningful if the original and the current operation are identical in load and speed and time
from start-up and temperature and, possibly, atmospheric conditions.
Analysis of data
Monitoring does not just consist of tuning into the condition of the equipment. Some
means of recording or storing the information is necessary, such as data loggers or in-built
storage facilities in the monitor itself. The information gleaned then needs to be analysed.
Analysis usually follows one of two routes. Either there is an absolute value above
which, or below which, the system may not go - or if it does it implies a serious fault, -
or the monitoring parameter is trended. The trending needs care. There are always some
fluctuations in monitoring levels, but what is to be looked for is that change in parameter
value which is gradually, and ever increasingly, moving away from the original level.
Figure 6.4.5 shows an example of a deteriorating piston pump. It had been monitored
in various ways and some of the monitors had more fluctuations than others (as shown by
the width of the lines), but the tendency is the same with all the techniques. In this case the
wear has been gradual, but should it increase more rapidly the slopes of the trend would
increase markedly.
Choice of monitor
The monitor must relate to the feature considered liable to develop a fault. However, there
MONITORING 499
are many other factors which must be born in mind in the ultimate choice. Table 6.4.4 lays
out 15 features which should be considered, although some would need greater weighting
than others depending on the application.
501
MAINTENANCE PROGRAMMES
MAINTENANCE USED to consist of just a regular routine inspection. It was correctly

regarded as an essential part of the working of any hydraulic system. It ensured that the
system was working at a normal efficiency and instigated preventative maintenance as
necessary. With such a process, suitable schedules for routine maintenance are laid down
(and adhered to), based on intervals determined by experience or specific recommenda-
tions of pump and component manufacturers.
However, maintenance can be much more effectively undertaken - as regards time and
manpower and cost - if some technique of continuous monitoring is included. Figure 6.5.1
shows the range of types of maintenance which are available, and those which include the
cost effective 'monitor' in the system. It suggests that a combination of time-based
maintenance (i.e. the regular routine maintenance) and machine based maintenance (i.e.
the inclusion of a monitor) provides the best solution.
Maintenance of filters
Full flow filters do not necessarily provide protection from contaminated fluid since they
are normally provided with a bypass which operates when the filter element becomes
clogged. This is a necessary precaution to prevent further increase in differential pressure
across the filter element which could cause it to fail and break up, distributing particles
downstream. Thus, whilst a clogged filter is normally 'safe' as regards element migration,
the flow is now fully bypassed and any contaminants in the fluid will be circulated through
the system. This presents some difficulty since the 'life' of a filter element depends
primarily on the system itself and the degree of contamination inherent in the working of
the system. This emphasises the importance of establishing intervals for preventative
maintenance for filter checking on an empirical basis.
In the absence of specific recommendations from the filter manufacturers, a programme
of filter maintenance can be chosen from the following:
1. Estimate element 'life' on an hourly basis and change filters at such regular periods
as are found necessary. Or, better,
2. Use filters with clogging indicators, with periodic examination of the indicators at
intervals shorter than the anticipated 'life'. Here it should be noted that many types
of clogging indicator can be triggered by the higher fluid viscosities normal when
a system is started up from cold. Or, better,
3. Couple a clogging indicator to give a visual and/or audible warning when the
element becomes clogged and the bypass is opened. Or, better,
4. Use a differential pressure sensor across the filter to measure the pressure drop and
indicate when a change is required. This is only as reliable as the accuracy of the
sensor, and also depends on the signal being passed (electronically?) to a point of
decision.
The filter 'life', or clogging time, will decrease with degradation of the fluid and with
any unanticipated increase in the amount of contaminant in the system. It should also be
noted that any change in the system may well require several changes of the filter following
the change, due to higher levels of contaminant being present. Any marked reduction in
filter life, therefore, should call for an analysis of the fluid. It is particularly important that
any fluid sample withdrawn from the system should be taken from a suitable point on the
system when working and at normal running temperature - preferably through a sampling
valve. Withdrawing a fluid sample from a static system, particularly from inside the
reservoir, is virtually useless, except as a check for possible water contamination. In cases
of doubt, separate samples can be withdrawn from different parts of the system.
Maintenance of fluid
The life of the oil is dependent on the grade employed and the system operating
characteristics (particularly the service temperature). With any high-quality hydraulic oil,
properly matched to the system requirements, life should normally exceed 15,000 hours.
With more severe service conditions life will be shortened and periodic sampling
should be included in routine maintenance. It is also generally desirable to check and
MAINlENANCE PROGRAMMES 503
record the working temperature of the fluid at each routine maintenance interval as an
excessive temperature at any stage can indicate potential trouble and accelerated degra-
dation of the fluid properties.
'Workshop' examination of an oil sample is not a very accurate method of assessment.
The general state of the oil as regards sludging, etc. will be indicated by the amount of
build-up of contaminants on the filters. Thus any oil removed from a filtered system should
be clear at any stage of its life. Cloudiness will indicate possible water contamination. If
this does not disappear on standing, with the water separating out at the bottom, it is a sign
that the demulsifying agent is exhausted. Oil sampling is shown in Figure 6.5.2.
A 'workshop' test for water contamination is to place a drop of the oil on a hot-plate.
A normal oil will smoke or bum. An oil drop contaminated with water will 'hiss'. If water
contamination is suspected, the sample can be left to settle, and the water should separate
out at the bottom. This will not necessarily happen, however, if other contaminants in the
fluid are present to the extent that they act as effective emulsifying agents. Under such
conditions the sample will not clear and water will not settle out.
A lightening of the oil colour, relative to the original, may mean that the oil has become
contaminated or diluted with a thinner oil and its viscosity lowered as a consequence. The
mixed oil mayor may not have similar resistance to oxidation, and so could affect overall
oil life as well as viscosity. A colour change (lighter or darker) may also mean that a
different oil has been used for topping-up the system. Again this mayor may not affect the
useful life of the oil, depending on the quality and viscosity of the added oil. The best
safeguard with a colour change, if the oil sample is otherwise clear and smells 'sweet', is
a viscosity check.
Oil samples may be submitted to an independent laboratory for test, or to the oil supplier
for his laboratory report (where the latter provides this service). In the first case it is
necessary to know the original specification (or preferably a laboratory report from the
same source on the new oil) in order to interpret the test data. In the latter case the state
of the oil will normally be evaluated as part of the test report.
Maintenance of pumps
Pumps should be serviced at regular intervals, as recommended by the manufacturer. If
pump performance is at all suspect, this should be checked by an individual test. Apart
from deterioration of the pump itself, parameters which can affect pump performance are:
i) Low oil level in reservoir - temperature increase, cavitation
ii) Filters clogged, or restrictions in intake line - cavitation
iii) Air leaks in intake line - cavitation
iv) Entrained air drawn from reservoir - cavitation
v) Oil viscosity too high - power losses
vi) Oil viscosity too low - low lubricity
vii) Pump speed too low - inadequate performance
Several of these items will normally be accompanied by noisy pump operation due to
cavitation or lack of damping. Too much air could also facilitate oxidation of the fluid.
Malfunctioning pump
A suspected mechanical fault in a pump can be checked by disconnecting the pump from
its driver and turning over by hand. Any excessive tightness or non-smooth rotation is a
certain indication of internal mechanical damage to the pump or its bearings. If the pump
motion is smooth, but noisy when connected and run, this is most likely due to cavitation,
caused by starvation on the inlet side. This could be due to low oil level in the reservoir,
clogging of the inlet strainer, or air in the suction line.
A diagnostic summary of pump faults is given in Table 6.4.1.
MAINTENANCE PROGRAMMES 505
Maintenance of hydraulic cylinders

Particular points to check in cylinder maintenance are summarised under separate
headings:
Rod endfittings
Wear on these components can be the result of poor or even a total lack of lubrication on
moving parts. A misaligned cylinder can also cause damage, particularly to the mating
threaded parts if excessive bending moments are involved.
Gland end cap

Thread pick-up and damage here can very often be traced back to insufficient cleaning-
down of the whole cylinder from a previous strip-down. It is imperative that all dirt and
contaminants be thoroughly cleaned off the cylinder's external surfaces before any work
is commenced. All cylinder oil should be drained off and parts cleaned using a lint-free
cloth. A note should be made ofthe order of component disassembly to ensure that all items
are replaced in the right order. The correct spanners should always be used as damage
caused by poor tools can mean that end-cap removal can only be achieved by destructive
means.
Wiper seal housing

Can often be overlooked as a possible source of cylinder failure. Wiper seal seating should
be examined carefully to ensure that there is no surface deterioration which would allow
ingress of abrasive or corrosive materials behind the wiper seal.
Wiper seal
Because this is not a pressure seal its importance can be overlooked. All seals should be
renewed throughout as a matter of good maintenance policy. Re-fitting a used wiper seal
is uneconomical and, although not immediately obvious on re-instatement, it would
permit ingress of dirt leading to an early cylinder strip-down
Rod seal
This is the main pressure-retaining seal to prevent oil leakage to atmosphere. Once again,
it is important to check the seal cavity housing in the gland bearing assembly, as any seal
is only as good as the surface on which it works. Care should be taken when fitting new
seals, which must be well lubricated with system fluid, to ensure the correct orientation
to mating parts. Incorrectly fitted seals do not effectively maintain the hydraulic fluid
pressure.
Static O-ring seals

These can bed themselves into the surface of mating parts leading to seal deterioration and
possible corrosion in the seal housing and on the cylinder tube wall. These material
surfaces may need treatment or attention before seal replacement.
Bearings
If the gland and piston bearings are bronze, wear can be recognised by burnishing or high-
spots. Excessive wear at these points can destroy the strut principle of the cylinder, leading
to 'knuckling'. Score marks on the gland bearing could indicate possible shaft damage.
Scoring on the piston head indicates oil contamination within the cylinder, which can also
lead to damage on the internal bore of the cylinder casing. Oil contamination can only be
cured by flushing out the complete hydraulic circuit, with special attention to filters, etc.,
and re-filling with the correct grade of clean oil.
Barrel
Care should be taken to flush out thoroughly before examination. The internal bore should
be examined with a high intensity light but the polished bore can make damaged areas very
difficult to recognise. Casing damage must be removed by honing, where practicable
within the seal manufacturer's tolerances. External casing examination is visual and
damage may be obvious, as it is usually mechanical damage at the bearing points caused
by lack of lubrication or cylinder misalignment. Excessive damage will mean complete
casing replacement. Indentations on the outer casing wall can transfer through to the inner
wall and are recognised as concentrated burnishing on the piston head. However, this is
not readily seen on examination, owing to the reflected light. Piston head wear on the
cylinder bore can usually be recognised by extruded seals. Any external pipework
damage, such as indentations, can reduce oil flow and piston speed.
Rod
Damage can come in many forms including indentations, surface abrasions or even
bending. Slight bending may not be obvious visually but shows as high burnishing on the
gland bearing, burning or discoloration on the shaft high-spot, and sometimes wearing
through of the shaft plating. Excessive bending is obvious and usually requires complete
shaft replacement. Bending is caused by over-stressing or similar maltreatment. General
wear on the shaft-plated surface must be dimensionally checked for acceptability.
Corrosive conditions can be indicated by the extent of the wear and shaft re-plating to the
manufacturer's tolerances may be necessary.
Cylinder re-assembly
Absolute cleanliness is essential on re-assembly and the subsequent need for re- furbishing
can often be traced back to a lack of care in previous cleaning and re-assembly procedures.
The correct tools should always be used and only soft hammers to avoid damage to plated
or polished surfaces. All components should be freely lubricated with system fluid and
care taken with the correct order, alignment and fitting of all items on re-assembly.
Couplings for oil feeds should be tightened to prevent oil leaks and cylinder malfunction.
The incorrect coupling of oil feeds can cause burst feed tubes or cylinder casings due to
hydraulic lock-off within cylinders. Pressure intensification of 4: I or 5: I can be experi-
enced due to full bore pressure acting upon a locked-off annulus.
Maintenance of hydraulic hose

Maintenance problems are usually, but not necessarily, matched to the severity of the
application. The system design needs to take this into account to ensure that likely
maintenance points are readily accessible. Equally the maintenance schedule and fre-
quency should be matched to the severity of the application.
Any of the following conditions will require hose and/or fittings replacement:
i) Leaks at fittings or in the hose;
ii) Damaged, cut or abraded cover;
iii) Kinked, crushed, flattened or twisted hose; hard, stiff, heat-cracked or charred
hose; blistered, soft-degraded or loose cover; cracked, damaged or badly corroded
fittings; and fitting slippage on the hose.
Visual inspections will reveal when clamps, guards and shields need to be tightened or
replaced.
Handling flexible hose

For in-plant hydraulic hose service to continue on an uninterrupted basis, the engineer
must initiate a suitable maintenance plan. Well-planned maintenance helps ensure the
safety of personnel and enhance hose performance.
When hose is received at the plant, some applied common sense at the outset, in the
unpacking procedure, for example, can preserve hose life. The use of sharp or heavy tools
should be avoided as they have the potential to damage a hose, and, therefore, create an
unknown safety hazard. The hose should be stored in its original container until installed,
particularly pertinent if the carton or crate is durable. Also, protective wrapping should be
used both when handling hose and when placing it in storage.
Storage
Hydraulic hose stored for along time should be in a coiled-flat plane, not hung haphazardly
on hooks or spike-like protrusions. The storage area should be cool and dark and free from
dust, dirt, dampness and mildew. Things that attack hydraulic hose in storage include
temperature extremes, humidity, sunlight, oils, solvents, corrosive liquids and radio-
active materials. Also, insects and rodents are attracted by rubber and proper protection
should be provided.
Ozone oxidises rubber, so the hose must be stored away from electric motors and other
ozone generators. Hydraulic hose needs to be kept away from heated surfaces such as
steam pipes and radiators because excessively high temperatures can harden and crack
rubber surfaces. Recommended storage temperatures are + 10C minimum, to +38 C
maximum.
System maintenance
Pressure
In a working system there will inevitably be pressure variations, such as those due to
variations in the effective compression ratio of an accumulator and pressure surges,
produced by back-pressure effects, in the case of double-acting cylinders. Thus the actual
pressure level at a particular point in a system can only be established empirically, when
ideally pressure variations should not be more than about plus or minus 3% throughout a
working cycle. A figure of plus or minus 5% is more realistic, and about the limit of
accuracy for the majority of less expensive pressure gauges which may be fitted in the
circuit. Such pressure gauges are usually protected against surge pressures and thus will
not indicate momentary surge pressures. They will, however, give a reliable indication of
any marked change in average pressure, indicating a potential or actual fault, provided the
gauges are checked for calibration at regular intervals. The error which can develop on a
gauge which is not periodically checked can be well outside the 5% limit and present a
false indication of a 'fault'.
Loss of pressure is a general indication of a leak or partial pump failure, or the lesser
likelihood of the failure of a bypass or relief valve. A poppet-type valve opening to a
bypass can also 'starve' a particular line by lifting at a pre-determined pressure and then
failing to reseat until a very much lower pressure is reached, because of the tendency for
such a valve head to remain 'floating' in the fluid stream. Valves with a degree of inherent
modulation are thus preferred in such circumstances.
Pressure variations in the system

The causes of marked pressure variations in the system are most likely to be found under
the following headings:
i) Undamped bypass valve - The cure in such cases is to replace with a modulated or
constant pressure type.
ii) Clogged valves - Calling for cleaning or, if desirable, replacement by a more
suitable type with a 'self-cleaning' or non-clogging motion.
iii) Pump pressure pulsations - Which are more marked with some types of pump than
others. Undesirable pressure fluctuations of this type can usually be damped by an
accumulator; alternatively, a different type of pump could be used with more
constant pressure delivery characteristics.
iv) Excessive foaming in the reservoir, caused by air entrainment - This could be
caused by poor reservoir design, an air leak in the system, or similar design fault,
allowing air to become entrained with the fluid.
v) Air inclusion - Such as produced by air leaks allowing pockets of air to become
trapped in the system, or unsatisfactory bleeding when the system is filled. Air
inclusion will be marked by lack of rigidity in the operation of the system.
Additional bleed points may be needed in the system, or the circuit design revised.
Excessive system pressure

Excessively high system pressures may be caused by partial blockages, although these
would normally be relieved by relief valves in the system. Excessive flow through relief
valves, however, may starve the flow to the actuators, resulting in slower motions. An
excessive system pressure will also increase the fluid heating and may lead to excessive
fluid temperatures.
System malfunction seen at the actuator

The basic problem here is to determine whether the malfunctioning is due to one of the
more obvious faults described earlier, or to the failure of one or more of the individual
components controlling the operation of the system. The greater the complexity of the
circuit the more the inter-dependence of the control and behaviour of the various elements,
and the greater the number of individual loops or circuits which may be involved.
Diagnosis then relies on isolating the fault within a specific group, and from there
determining the basic fault responsible for the malfunctioning of that group. Treatment is
thus specific to the circuit and requires close study and attention to the circuit diagram as
well as full appreciation of the function of each individual component. The most
straightforward method is usually to 'read back' from the actuator involved to establish
the point at which the lack of behaviour or control is initiated, ignoring those components
which are not directly concerned with the function which is at fault.
In the case of simpler circuits more direct and straightforward action is usually possible,
particularly if the system pressure is first checked.
A. System pressure normal, lack of actuator movement

i) Absence of signal;
ii) Solenoid failure in the solenoid valve accepting the signal;
iii) Mechanical obstruction.
B. System pressure normal, loss of actuator speed
i) External oil leak;
ii) Internal oil leak in actuator or valves in actuator pressure line circuit;
iii) Faulty or badly-adjusted relief valve;
iv) Partially-blocked control valve;
v) Pump fault or blockage in delivery line reducing delivery, but not necessarily
pressure;
vi) Overheating of fluid causing loss of viscosity;
vii) Excessive wear on actuator;
viii) Excessive loading of actuator, caused by unduly high external load or eccentric
load loading;
ix) Increase in actuator friction, e.g. due to maladjustment of compression packings,
distortion through applied bending loads, etc.
C. System pressure low, loss of actuator speed
i) Loss of delivery through a pump fault or leak;
ii) Accumulator failure, or loss of gas pressure requiring a recharge;
iii) Faulty pump drive;
iv) Incorrect valve settings;
v) Faulty relief or bypass valves;
vi) Maladjustment of metering valves in the case of linear actuators, or faults or dirt
in valve;
vii) Dirt under cushion ball valve (linear actuators).
D. System pressure variable, variable actuator speed

i) Worn pump, or pump faulty, with failure to produce required maximum demand;
ii) Under-sized accumulator, or loss of compression ratio (accumulator may need a
gas recharge);
iii) Internal leakage in part of the system;
iv) Partially-blocked control or sequencing valve;
v) Faulty pressure-relief valve;
vi) Out of order cushion ball valve (linear actuators).
Irregular action
Irregular or erratic action is commonly caused by air entrainment. Check as aforemen-
tioned for 'Variable fluid pressure'. Other possible causes are:
i) Excessivefriction - Caused by seals or packings being too tight or incorrectly fitted
so that 'wedging' results. Extrusion of seals and 'wedging' may occur at high
pressures with a-ring seals, unless provided with back-up washers. Lack of
lubrication or binding on slides, etc., can be another cause of excessive friction.
ii) Misalignment of actuators - Misalignment of actuators, tables, slides, etc., can
cause irregular action.
iii) Compressibility effects - The compressibility of the fluid under high pressure can
affect precise movement and control. This is a normal fluid characteristic. A
synthetic fluid is better in this respect. Air entrainment also reduces the effective
modulus of the fluid.
iv) Lack of synchronisation - Exact synchronisation is difficult to achieve within
straightforward hydraulic systems and it may be necessary to apply synchronisa-
tion via mechanical linkages or the use of accumulators.
System pressure remains high with system unloaded

i) Faulty non-return valve isolating the unloaded part of the system;
ii) Badly adjusted unloading valve;
iii ) A combination of i) and ii ) .
Note: If adjustment of the unloading valve produces no change in system pressure, then
the fault lies with the non-return valve.
See also chapter on Leakage.

511
VIBRATION AND NOISE
ANY FLUCTU AnONS in movement, or variations in system pressure, can cause sound
to be generated. Hydraulic systems are not immune from such disturbances; indeed,
because they are tending towards greater compactness, the power densities are increasing
with higher cyclic forces in reduced volumes, with the resulting greater excitation levels.
Sound is assessed by the human being in terms of intensity and harmonic content. The
design of the ear is such that it is able to detect a considerable range of intensity; this means
that in order to define the intensity a logarithmic scale is normally used, i.e. the decibel
(dB) scale. Table 6.6.1 gives an idea of the dB range both in terms of the dB scale and the
actual sound pressure levels.
The harmonic (or frequency) content of a noise is also of importance. A young person
may at the very best be able to hear up to 20 kHz, but most adults would find 17 kHz or
less their upper limit; also detection ability changes with different frequencies. It is,
therefore, more common to use the dB (A) scale which takes into account the varying
degree of sensitivity from low to medium to high frequencies, and applies a higher
weighting factor to the 1kHz-4 kHz band and a lowerfactor to above and below this band.
Permanent damage can be caused to the hearing if the sound level occurs more than just
momentarily above around 90 dB(A).
Unusual harmonics may also be a cause of distress. Human beings may well describe
the noise in terms of dislike like 'howling', 'screeching', 'whining', etc. and if the level
of such sounds, although below 90 dB(A), is not further reduced illness may well result.
Vibration and noise can also be a serious feature when attempts are made to mask the
presence of equipment such as in naval warfare or fishing. However, the actual power
developed in the noise is relatively small and virtually no improvement in efficiency
results from a reduction in noise; for example, less than 1W of effective sound power level
may be emitted by a 10 kW pump producing 90 dB (A) of noise.
There are, fortunately, a number of ways of reducing the effect of vibration. The subject
of vibration and noise is best addressed by examining sources of excitation, transfer paths
and means of reduction - 'vibration' being the actual movement and 'noise' being what
the human hears. ('Resonance' occurs when the frequency of excitation matches the
natural frequency of some component or part of the system.)
Sources of vibration and noise

The following components are the most important as regards sources of noise in hydraulic
systems:
The pump and its driver
The pipes and hoses
Control elements
Possible reasons for the excitation are outlined in Table 6.6.2 which looks not only at
the normal source but also where the vibration increases due to faults developing in the
system or component. Although not normally a 'source', the reservoir is quite frequently
a sounding device which releases sound due to pulsations entering it; this is also discussed
later.
Transfer paths
The path which the fluctuations take in their journey to the ear, may be anyone or more
of the three media:
Structure-borne
Liquid-borne
Air-borne
The path may be quite complex, and an analysis based on cross correlation may be
required to determine the route. However, if that route has been correctly diagnosed then
the most efficient means of reduction can be applied. If the path is unknown, then a number
of exploratory tests may be necessary, with differen t types of reduction, to determine the
optimum. It must be remembered that more than one media may be involved and hence
more than one type of reduction may be necessary. Air-borne and, normally, liquid-borne
noise originate from structure-borne noise.
Reduction of vibration
Vibration may be reduced, as far as the listener is concerned, by
Damping - absorbing the energy
Detuning - increasing the frequency difference between the excitation frequency
and the natural frequency of the component
Reducing pulsations - dropping the levels of excitation (which may involve, at the
same time, changing the frequency of excitation).
These will be discussed below in relation to the individual components examined.
Pump drivers vibration and noise

Electric motors
The source of sound from electric motors is the cyclical changes in magnetic flux which
cause small dimensional changes and a resulting hum. Careful design can reduce this
slightly, but not completely.
The cooling fan is often totally enclosed (i.e. the totally enclosed fan cooled - tefc -
motor) and is designed so that rotation is in either direction. This means that it must have
straight blades which are notoriously noisy. If the motor is to be used in one direction only,
a quieter fan may be fitted.
Diesel drivers
Heavy diesel engines have a very characteristic diesel 'knock' and it may be necessary to
place the noisier engines in suitably damped enclosures.
Pump vibration and noise

The pump is the primary source of noise in an hydraulic system. In particular, the cyclic
pressure pulsations cause a sound to be generated at a distinct frequency and harmonics.
In addition there is the mechanical generation through the motion of bearings, flexing of
teeth or vanes, and the general sliding and forces associated with piston pumps. Although
it may be possible to reduce the noise output to the human listener to moderate levels by
suitable acoustic treatment (enclosure or mounting), much effort has been expended on
reducing pump excitation by subtle design changes. Another approach, particularly for the
larger system, is to use smaller, but more, pumps, or run the pumps at lower speed; it is
also possible that resonances can be avoided at a slightly changed speed. Mounting on a
resonant 'sounding board', like a reservoir, should also be avoided.
Pump design
External gear pumps usually have equalisation grooves milled in order to smooth the
pressurisation process around the circumference; even so, pulsations around 10-14% of
the delivery flow still occur. A greater number of smaller teeth could reduce this, but only
at the expense of performance. More recently other techniques have been adopted such as
offsetting by a half tooth a coupled gear pair which can drop the pulsations to around
3-6%, or by using helical gearing. (See chapter on Hydraulic pumps.) Another method,
shown in Figure 6.6.1 is that of using zero backlash; this results in a similar output to the
phased pair design as regards frequency but a slightly lower level of noise. Although such
a 'perfect' design would normally be prohibitive as regards cost, the manufacturers have
managed to achieve this mesh by incorporating an adapti ve degree of freedom to the gears'
centre distance.
Piston pumps, with a gradual phasing of the pulsations from each piston, can also
significantly reduce the sound levels which each individual piston would otherwise
produce. The radial piston pump design has the advantage of incorporating a reaction ring
which remains in constant contact; this constitutes a force cage in the sense of absorbing
a considerable amount of the cyclical forces generated by the pump.
The material properties of the casings used can also contribute a few dB to the reduction
in noise. With this, and careful design of the casings, one pump design may be
considerably quieter than another
Acoustic treatment (damping)

Pump/motor vibration can be minimised by mounting the pump and motor on a common
base (or mounting the motor integral with the pump) and isolating the complete unit on
a resilient mount. A general recommendation is that the natural frequency of the isolated
mount should not exceed one-quarter of the shaft speed (frequency), although it may be
permissible to approach one-third of the shaft speed if a stiffer mount is required.
If further acoustic treatment is required the whole pump/motor unit can be fitted with,
or within, a suitable enclosure. The majority of hydraulic pumps are driven by electric
motors, so no special problems are involved other than size and ensuring an adequate
airflow for cooling the electric motor. If necessary a forced draught ventilating system can
be used with a completely sealed enclosure, employing duct silencers of the absorptive
type.
A simple method of isolating or decoupling the pump from the delivery line is by a
flexible hose connection. Isolation can be further improved (if necessary) by using two
such hose lengths in close proximity and mounted at 90 to each other. Ideally, isolation
by flexible pipe should include bends in two mutually perpendicular directions with equal
distances between bends.
A further method of decoupling is the use of O-ring or similar elastomeric seals in a
suitable coupling. This is particularly effective for decoupling high frequency vibrations
but is less effective at lower frequencies. Couplings of a similar type are produced for
semi-flexible connections, allowing for a limited degree of movement or misalignment.
Such couplings are not effective isolators unless the resilient elements employed provide
complete isolation between the components connected (i.e. do not allow direct metal-to-
metal contact).
In general, isolation of the pump and motor from the tank by suitable mountings and
decoupling from the pipework will free the rest of the system from the transmission of
mechanical vibrations and the consequent possibility that these would be amplified. The
coupling between the pump and motor may need careful design and fitting to avoid out-
of-balance and a too solid a route for the sound; a flexible coupling is ideal.
Pipe and hose vibration and noise

Although pressure pulsations are always expected to be present in pipe lines, it is highly
desirable to isolate the lines from the hardware so that vibration generated by the pump
unit is not transmitted through the pipes, with the possibility of resonance occurring at
other connecting points. Hoses can be used in this process of decoupling.
Sound waves carried in pipe walls may produce flexural motion, longitudinal or
torsional waves, although it is the flexural waves which generate the majority of the air-
borne sound. The sound in the liquid is in the form of longitudinal waves which produce
audible sound at each discontinuity or bend. The waves may be pump generated (changes
in power and pressure, or varying amplitudes of pressure pulsations) or fluid generated
(flow instability, turbulence or simple fluid friction).
Fluid-generated noise in small bore pipes with low to moderate flow rates is generally
negligible, unless pressure pulsations or cavitation are present. Thus consequent radiation
of airborne noise, is usually due to the higher level of vibration generated by couplings.
Pipe resonance is the cause of the highest sound levels, where standing waves resonate
with particular supported lengths of pipe; this can be removed by changing the lengths of
pipe or the support points - see sub-section below.
Thus standard treatment for noise reduction is:
i) Damping by means of suitable isolating pipe supports. This also provides decoupling
for supporting structures
ii) Decoupling from other sources of noise or vibration in the system.
iii) Soundproof 'lagging'.
For the majority of systems only i), and to a lesser extent ii), should be necessary.
'Lagging' is normally only required when there are vibrations present which cannot be
damped or isolated by simple means. This is most likely to occur on pumped systems
employing thin walled large diameter piping, particularly on the suction side.
Sufficient damping for pipes is usually provided by suitable supports, or pipe clips
spaced at regular intervals, the supports having resilient linings so that vibration in the pipe
is not transmitted directly to the surface to which the supports are fixed (Figure 6.6.2).
All pipework installations should be designed on the basis of avoiding abrupt changes
of section which could lead to large flow velocity changes and generation of turbulence.
As far as possible, too, generous bend radii should be employed for similar reasons.
Suction line
The suction line is a first suspect in a normally acceptable hydraulic installation which
proves noisy, and where the noise cannot be directly attributed to pump or components.
Suction lines can generate noise if there is an excessive pressure drop when the pump is
VIBRATION AND NOISE 517
sucking below atmospheric pressure and drawing air out of solution (hydraulic oils
normally contain about 8% air in solution) and cavitation causing the oil to vapourise. The
resulting formation of bubbles, and their subsequent collapse, can cause a 'mechanical'
noise which is often erroneously diagnosed as a pump fault. This cavitation must be
avoided not just for the noise level, but for the erosion that will ensue in the pump.
Suction line noise can also be caused by a partially blocked or undersized suction filter,
poor placement of the outlet pipe in the reservoir (or possibly the reservoir being fitted
below the pump) or entrained air.
Standing waves
Optimum pipe clamp spacing can be analysed in terms of standing wave phenomena,
although this is seldom necessary. The case of axial standing wave is usually academic,
for practical lengths are usually substantially lower than the critical length, which is
defined by:-
Theoretically, at least, the distance between pipe supports should always be less than
this resonant or critical length. A more expensive way of reducing the noise effects is the
use of gas loaded accumulators or pulsation dampers (see below).
Actuator vibration and noise

Although some noise in generated by the motion and 'thumping' of the pistons, the
primary effect in cylinders is to radiate noise rather than produce it. Dampers can be fitted
both internally and around the cylinder.
Valve vibration and noise

Noise due to the operation of valves, regulators and control elements is transient and
related to the degree of turbulence or cavitation produced. In specific designs and certain
circumstances individual elements may be subject to vibration and generate a continuous

noise. The noise level of such devices is dependent on the design and the localised flow
velocities produced and also on the response time, where applicable. The latter effect can
be minimised by arranging that the response time is not shorter than that required by the
system. This will result in minimum 'hammer'. 'Water hammer', as it is termed, depends
on the switching velocity of the valve - i.e. on the spool-switching velocity in the case of
spool valves. Valves operated by dry solenoids have uncontrolled response and so often
produce 'hammer'. Wet solenoids are cushioned by the hydraulic fluid so move more
smoothly and open the valve passages more gradually (but at the expense of some loss of
solenoid power).
As a general recommendation, simple undamped ball-and-spring non-return and relief
valves should not be used. On the design side, every effort should be made to ensure that
the flow passages of valves are swept and free from sharp edges and comers as far as
possible. Directional control valves must also be carefully designed to prevent flow
instability occurring.
Cavitation in valves
Cavitation is a breakdown in flow caused by the localised fluid falling below the vapour
pressure of the fluid. Consequently, vapour bubbles are formed resulting in irregular and
noisy flow. Such a reduction in pressure can occur in regions of localised high flow
velocities, e.g. at restrictions to the flow path. Thus, the onset of cavitation is marked by
a critical pressure, which in turn is dependent on a critical velocity. The effective critical
pressure is also a function of temperature because this governs the vapour pressure of the
fluid. Accurate prediction of cavitation conditions is most difficult, and usually impossi-
ble, in the design of valves and fittings, and problems have to be tackled on empirical lines.
Although much has been done to design fittings which do not produce cavitation at normal
or recommended flow rates, it does not follow that this will be maintained over the full
range of operation. If the flow rate is sufficiently restricted, cavitation and noisy flow can
be expected. Thus a partially closed tap or valve is nearly always noisier than when fully
opened; also quite a small change in position, and thus flow rate, can cause a change from
cavitating (and noisy) to non-cavitating (and relatively quiet) flow. It is also a character-
istic of many valves, that for flow rates or valve openings which produce cavitation,
cavitation noise increases with increasing frequency; whilst for higher flow and large
openings, where flow is non-cavitating, noise does not vary greatly with frequency. This
also explains the considerable difference in interpreted sounds - higher frequencies being
more readily radiated and sounding louder to the ear.
In the case of high pressure systems, or valves subject to high pressure drops, it is
desirable to utilise flow paths designed to eliminate cavitation as this can cause physical
damage to the valve components as well as excessive noise. The problem, basically, is one
of preventing the pressure in the val ve throat from falling below the fluid vapour pressure
in order to prevent cavitation occurring. This requirement can be rendered in mathematical
form:
Pulsation dampers
Gas-loaded pulsation dampers such as that shown in Figure 6.6.3 are relatively expensive
but are highly successful at attenuating sound. This is particularly true of the straight
through type shown. Small bladder type accumulators can also be used with only perhaps
15 mL of gas to provide adequate attenuation. One disadvantage of gas-loaded devices is
the need to regularly top up the gas and the occasional need to replace the bladder or
diaphragm used.
Shock preventers
Shock preventers are pulsation dampers (or accumulators) characterised by having very
large flow inlet apertures which are partially closed off by liquid trying to flow back out
of them. They are not shock absorbers, as they prevent shock or surge occurring. For the
same reason, they do not attenuate shock.
Acoustic filters
Acoustic filters can be fitted to systems where pressure ripple is high. These are essentially
tuned silencers (like automobile silencers) which are critical in design and are usually
effective over only very narrow frequency bands; the attenuation achieved can be quite
high. They are particularly suitable for constant speed pump systems, but they may be
expensive as they are purpose built.
Untuned silencers simply comprise an expansion chamber with broader coverage but
reduced attenuation. An accumulator is, in effect, an untuned hydraulic acoustic silencer
and is most effective at lower frequencies. Dissipative-type silencers provide for dissipa-
tion of energy through viscous flow losses and, as a consequence, consume some fluid
energy. They may be combined with an untuned silencer, although the attenuation will still
be appreciably lower than that of the tuned type.
In general, wave cancelling filters are to be preferred since the frequencies involved are
low. If the pressure transients are narrow band, a Quincke Tube and expansion chamber
can be effective. A major disadvantage of this and other types of simple wave-cancelling
filters, however, is the relatively high pressure drop produced. The more usual form of
hydraulic silencer is the pressure-release type. This gives minimum pressure drop and
broad band filtering, but is pressure sensitive and needs regular routine maintenance.
See also chapter on Shock absorbers

521
LEGISLATION AND SAFETY
ANY PRODUCT or system is only acceptable in industry if it is safe. The yardstick by

which 'safe' is defined is formulated by various health and safety bodies, and in particular
the Acts of Parliament and the European Community directives as regards the United
Kingdom. Other countries or groupings also have their own distinctive regulations. This
chapter will highlight the current situation regarding the hydraulics industry in the UK and
Europe by looking first atthe type of accident which may occur, and then at the regulations
which are designed to prevent or avoid them.
Potential hazards
The distinctive accidents related to hydraulic fluid power are those associated with the
high pressures in the system. The medium, i.e. the oil or water, may also present hazards
when misplaced, such as slippery floors. There are, of course, serious consequences from
mechanical- e.g. rotating and reciprocating - faults, or from electrical faults. Tempera-
tures, too, may be over the permitted 'touch' level.
Table 6.7.1 is a list of some typical fault or failure situations causing actual bodily harm,
with some suggestions of what would have improved the situation. (These are only
examples to give an idea of the type of problem. Accumulators are notorious in that they
may leave a section of a system pressurised until they are fully discharged.)
Legislation
There are' Acts', 'Regulations' and 'Standards' with relative importance in that order. As
regards personnel, the Health and Safety at Work Act (1974) is still the principal
parliamentary act in the UK. All directives and regulations are covered by this Act in the
general sense, for instance, the Supply of Machinery (Safety) regulations (1992). Stand-
ards are lower down in the strata although they may have been formed at virtually the same
time as the regulations and acts; their value is in defining details which will help
compliance with the regulations.
All new machines in the European community supplied since 1 January 1995 are
required to comply with the Machinery directive 89/392/EEC (with amendments to date)
and must carry the 'European Conformity' CE mark.
There are two purposes for the CE mark:
It allows free movement into and within the countries of the European Economic
Area
It shows that the manufacturer has declared the product as meeting the minimum
safety requirements.
In addition, from I January 1997 it is necessary, according to the Provision and Use of
Workplace Equipment Regulations (PUWER), for companies with a work-force of more
than five people to undertake risk assessment of their existing older machinery. The risk
categories are basically:
LEGISLAnON AND SAFETY 523
Severity of potential injury

Frequency of exposure to the hazard
Possibility of the user being able to avoid the hazard
Likelihood of the occurrence of the hazard
Five safety risk categories, from B - safety devices and control systems generally
acceptable for domestic appliances, to 4 - any single fault must be detectable and up to
three faults should not lead to a failure of the safety function, cover the requirement of the
safety system.
One set of regulations particularly appropriate to accumulators is the Pressure Systems
& Transportable Gas Container Regulations (1989), which came into effect on the 1 July
1994. It requires three features:
the safety testing of products
the maintenance of records related to the safety of the products
the identification of hazardous products
The EMC Directive

The EMC Directive, i.e. the Electromagnetic Compatibility Directive, came into force on
the 1 January 1996. Basically it requires appropriate machinery (i.e. that which includes
electrical and electronic apparatus) to be immune to the common sources of interference
and also not to cause interference to other equipment. It is concerned with electronic
radiation, i.e. electronic noise, but this, whilst not being a primary feature of hydraulic
equipment, is still present in many parts of the hydraulic system; control, sensors and the
drives, all include this possible interference.
Such interference can not only cause an adjacent electronic product to malfunction,
which in medical or aircraft apparatus can be life threatening, it can also be a serious hazard
to personal health.
In fact, the directive is not an absolute requirement but rather a requirement not to
exceed a certain level of electronic noise - chosen to be that which will not cause a change
in the function of the adjacent 'apparatus' (including human beings!). Tests are thus
undertaken on the Radiated Test Level to an IEC (International Electrotechnical Commis-
sion) standard, for the class required for the apparatus and environment; there are also tests
to the European Electromagnetic Immunity Standard to test the susceptibility to radiation.
It should be noted that there are differences depending on whether the unit is actually
an 'apparatus' or a 'component'. An 'apparatus', which must carry the CE mark, is a
finished product having an intrinsic function. A 'component', which does not carry aCE
mark, has no intrinsic function. A 'system' includes a number of items of apparatus
designed to work together to fulfil a specific objective. Whether an apparatus or a system
needs to be tested depends on the final marketable functional product. It will be apparent
that hydraulic pumps, valves, etc. are 'components' rather than apparatus, and hence do
not need to be approved, but when they are built into a power pack, say, they become a
functional product and need to comply as a complete unit (see BFPA below).
The BFP A/P70 (1996) is a useful guideline to understanding the requirement of the
EMC Directive, and the BFPAlP68 (1995) provides the necessary 'EC Declaration of
Conformity for Machinery' documentation in the 13 languages of the European Commu-

nity. (The declaration must be in the language of the country where the apparatus is to be
used.). The BFPDAID 10 (1995) provides a range of manufacturer's test and conformance
certificates covering the following directives:
The Machinery Directive 89/392/EEC (amended by 91/368/EEC, 93/44/EEC and
93/68/EEC)
The Supply of Machinery (Safety) Regulations 1992 (SI 1992/3073)
It also describes and illustrates how the forms should be completed.
Other safety requirements

Safety is the reponsibility of both supplier and user. Legally, the designer of a system may
be liable. It is important, therefore, to sense all aspects of safety taking into account the
full spectrum of possible personnel who may be invol ved in the life cycle of the product
Full records should be kept at each stage by the person responsible, to show that all
reasonable care has been taken in the design, manufacture, build and operation of the
product, and that the disposal can be undertaken without hazard.
SECTION 7
525
MINIATURE HYDRAULICS
ALTHOUGH PNEUMATICS may be more commonly associated with miniaturisation,

suitable hydraulic components are also able to supply the same attraction from the point
of view of space and weight saving. For instance, miniaturised valves can form the
working logic elements in a control system, operating at very low power levels. Miniaturi-
sation of the main circuit, however, inevitably leads to restriction on power levels, mainly
on account of practical limits to the flow velocities which can be accommodated without
excessive pressure drop losses. For that reason reduction in size as such offers little
advantage except in applications requiring power levels up to about 1 kilowatt. Here
miniature hydraulic systems are directly competitive with, and can offer specific advan-
tages over, mechanical, pneumatic and electrical power systems. In particular, they
embrace applications involving the instrumentation and control of mechanisms, and the
transfer of control signals of limited output potential.
Power levels and pipe sizes

All the components of a typical hydraulic system are usually available in miniature form,
i.e. the pumps, cylinders, valves, etc., but offering overall a compact yet powerful circuit.
The power level obtained from any hydraulic system is directly proportional to the product
of fluid pressure and flow rate. Theoretically, at least, the flow level for any size of system
can be increased by increasing the pressure. Practical limits are set by the size of clearances
necessary to maintain satisfactory leakage levels. Material stress levels are less important
and become less significant still with miniaturisation. The typical maximum pressure set
by technological demands is of the order of 21 MPa (210 bar).
The pipe bore sizes associated with miniaturised hydraulic systems are usually within
the range of 1.8 mm (0.07 in) to 4 mm (0.16 in) - or say a nominal size of 3 mm (0.125
in) - although larger sizes (e.g. up to 6 mm) may be employed for high pressure lines.
Typical bore sizes for a similar pressure rating would be 1.7 mm for 3 mm o.d. and 4 mm
for 6 mm o.d. tubes. The main reason for selecting the larger tube size would be to reduce
flow velocity (and thus pressure drop) for a system with a fairly high demand from the
pressure line, e.g. for rapid cylinder movement. The upper limit of flow rate which can be
accommodated through a 3 mm o.d. line size is of the order of 3 Umin.
These parameters, i.e. P = 21 MPa (210 bar) max. and Q = 3 Umin max., set the output
power levels obtainable, viz.
Pmax= 3 x 21/61.2 = 1.03 kW
Miniature cylinder sizes and forces available

Specifically, to be described as miniature, a hydraulic cylinder needs to be smaller than
the smallest standard hydraulic cylinder, i.e. 25 mm bore. Logically other components,
especially val ves, should likewise be reduced to a matching size. Cylinder forces available
should then be of the order shown in Table 7.1.1.
There is little advantage in making much smaller cylinder sizes as lower output
requirements can be supplied more efficiently by working with reduced pressure.
Maximum thrust force from the20mm (0.75 in) size cylinder is thus 6600N at a maximum
pressure of 21 MPa (210 bar).
Cylinders may be single- or double-acting. In the case of double-acting cylinders
miniaturisation imposes certain limits on the output force available for the outward stroke
because of the size of the rod required to accommodate the inward stroke force. In other
words, the rod diameter may need to be larger in proportion to the cylinder bore than in
conventional hydraulic cylinders. Typically, for example, the rod diameter used may be
as much as 50% of the bore. In general, maximum piston speed likely to be achieved with
a 20 MPa (200 bar) system is of the order of 200 mm/sec.
Miniature self-compensating shock absorbers with ratings as high as 18 Nm per cycle
and an operating stroke of only 16 mm can be effectively used in the smaller applications.
Miniature valves
Control valve dimensions are largely governed by the method of actuation - e.g. the size
of the magnet in the case of electro magnetic controls. If low viscosity fluids are being
used, then it is often feasible to use normally pneumatic valves, providing the seals are
compatible.
A particular problem which has to be solved in the design of miniature valves is how
to reduce the hydrodynamic force. This force originates in the change of fluid stream
momentum when flowing into and out of the valve spool ports, and acts on the spool in
such a way as to tend to lock it in the closed-flow position. It is especially unfavourable
MINIATURE HYDRAULICS 527
in miniaturised flow distribution valves with small spool travel and high pressure drop on
the distributing edges The need to overcome this force may substantially increase the
relative dimensions of the control electromagnet, as seen in the example shown in Figure
7.1.1 where the solenoids represent more than 2/3rds of the total length.
FIGURE 7.1.1 - The WL43 4-way miniature solenoid operated directional control valve
[Mason Hydraulics Ltd].
The basic requirement for efficient valve operation is that the change of flow momen-
tum does not exert a closing force on the valve spool, provided that the fluid jet does not
act on a movable part of the valve. Thus the valve port geometry has to be designed to
provide a substantial reduction in the 'initial hydrodynamic' force present. In the case of
hydrodynamically operated valves, both inlet and return flows must be considered. Figure
7.1.2 shows an example of an insert pressure compensated flow control valve (31 mm
long).
FIGURE 7.1.2 - FLF A28l23lOD Pressure compensated flow control valve [Lee Company].
Miniature pumps and motors

Miniature external type gear pumps with as low as 0.25 mUrevolution output are
available, and able to operate at 21 MPa (210 bar). Miniature fixed displacement axial
piston pumps can provide even lower outputs; one design consists of a fixed cylinder block
and rotating angled swash plate driving three pistons, shown in Figure 7.1.3; output is as
low as 0.045 mLirevolution with a continuous pressure of 35 MPa (350 bar).
Miniature power packs may be operated by DC or AC electric motors up to 5.5 kW.
Miniature hydraulic motors are also available with displacements in the range 8 mL to 32
mL and torque values 1.3 to 5.1 daNm. These are particularly suitable for machine tools
and robots. (See Figure 7.1.4.)
Miniature circuit considerations

Whilst circuit design follows that of conventional hydraulics it is often possible with
miniature hydraulics to simplify the control system. Thus Figure 7.1.5 shows a circuit
where a cylinder can be controlled by a single-acting directional control valve.
The system is shown in the neutral position with the valve closed. The hydraulic fluid
is pumped from the hydraulic pump through the accumulator to the directional control
valve inlet and to the rod end of the hydraulic cylinder. The rod end of the cylinder is
therefore constantly under operating pressure. To initiate an operation, the valve on the
supply side is opened with the lever by means of a thrust pin. The hydraulic fluid then flows
into the cylinder head end. Upon releasing the lever the valve is closed by its spring.
When the other valve is opened with the same lever, piston return begins. Upon
releasing the lever the valve is closed by its spring. The stroke speed of the piston can be
controlled by means of a throttle screw in the housing. The entire hydraulic system is
protected from overloading during operation by a pressure relief valve in the pump.
Should the permissible maximum pressure be exceeded (e.g. through external forces
acting on the cylinder), the directional control valve connected with the return flow acts
as a safety valve for the cylinder in question.
Fields of application
The main fields of application for miniaturised hydraulics are:
Utilisation of the advantages of hydrostatic drives in instrumentation techniques,
and of their compatibility with electric control systems.
Replacement of mechanical parts oflow power output in both machines and various
instruments.
Replacement of electrical or pneumatic components where it is advantageous from
both the technical and the economic point of view.
Practical fields of application are numerous, and include:
Operation of windows, sun roofs, seat adjustments, fan belt stretching, lorry
mounted tail-lifts, etc., on road vehicles.
Operation of windows, doors, blinds in houses and commercial buildings.

Operation of dentists' chairs, operating tables, hospital beds, etc.,
Operation of feed clamping devices and similar mechanisms associated with
machine tools.
Automation of controls.
Source of servo power in radio control systems.
531
HIGH TEMPERATURE HYDRAULICS
EVERY COMPONENT has an upper limit as regards temperature. Usually in hydraulic

systems, the lowest of these 'upper limits' is found in the fluid, but it may be present in
the seals or an inadequately protected control system.
Conventional hydraulic fluids

Conventional hydraulic fluids begin to break down at relatively moderate temperatures,
and even below the break-down point will show accelerated deterioration. Mineral oils
have a somewhat higher temperature rating than water-based fluids and with certain types
can give satisfactory service at fluid temperatures of up to about 140C. Phosphate ester
fluids can be used up to about 150C; but neither type is particularly suited for working
at these high temperatures. Hydrolytic stability is often suspect at temperatures above
normal working figures. Many otherwise satisfactory fluids may break down under such
conditions; others become excessively acidic and promote corrosive attack or show an
excessive rate of oxidation.
All fluids, therefore, can be given maximum temperature ratings for continuous duty
consistent with a normal service life; and a higher short-term rating for intermittent duty.
The latter will result in reduced life, depending on the severity of the over-rating. Typical
values are shown in Figure 7.2.1
In general it is desirable to maintain fluid temperatures substantially below recom-
mended maximum operating values as this will have a beneficial effect on fluid life. Also
the use of synthetic fluids for higher temperature service may show unexpected limita-
tions. It is characteristic of synthetic lubricants that if they reach an excessively high
temperature (as could occur at some localised point in the system) they can vaporise
without leaving any residual surface film, i.e. leave surfaces dry and unlubricated.
Viscosity
Other characteristics of the high-temperature operation of fluids are poor stability and a
loss oflubricity (due largely to a loss of viscosity). It is virtually essential that the fluid
retains a viscosity of at least 2 cSt at the operating temperature if efficient lubrication and
sealing are to be maintained in other than systems designed for water hydraulics. This must
also be accompanied by good shear resistance, otherwise the inherent shear losses in the
system will still further reduce the effecti ve viscosity. In any case the shear forces present
in the higher-temperature system will probably be higher than in standard practice, due to
the employment of closer fitting components and finer tolerances to minimise leakage at
low fluid viscosities.
Vapour pressure
Another important factor to consider is the vapour pressure of the fluid at the operating
temperature. If the vapour pressure is unduly high the fluid will tend to evaporate, which
can lead to cavitation at the pump as well as seal failure, etc. It is necessary, therefore, that
to be suitable for high-temperature working a fluid should have an acceptable vapour
pressure figure at the system working temperature. Provided these various requirements
can be met, the ultimate choice of a high-temperature fluid can then be based primarily on
thermal stability.
Table 7.2.1 shows the vapour pressure of typical fluids at elevated temperatures.
HIGH TEMPERATURE HYDRAULICS 533
'Tailored' fluids
One of the advantages of silicones is that they have a high degree of thermal and hydrolytic
stability, i.e. they retain a high viscosity index and good resistance to shear breakdown.
Lubricity can be obtained by 'tailoring' the molecule rather than using additives (which
would be subject to thermal breakdown).
A successful form of 'tailoring' is incorporating chlorophenyl groups to improve the
lubricating properties of a basic silicone fluid and compounding with an oxygen inhibitor.
Fluids of this type are suitable for continuous service temperatures up to 290C and have
a life of 50 to 100 hours at 3 16C, depending on service conditions. Above 3 16C there
is a certain amount of degradation and molecular re-arrangement, the actual break-down
rate being dependent on the temperature and of the order of 4% per hour at 370C. The
break-down products are low-molecular-weight silicones which are volatile at the
temperatures at which break-down can occur and soluble in the fluid at lower tempera-
tures.
Cooling
In practice, where high service temperatures are involved, and consequently high fluid
temperatures, it is generally more reliable, and certainly less expensive, to provide the
system with a satisfactory method of cooling so that conventional fluids can be employed
and maintained at acceptable fluid temperatures. This has the advantage of offering the
widest choice of fluids and matching components, with known performance. It does,
however, place an absolute premium on the cooling system being fully effective as any
failure of the cooling system can lead to early breakdown, and possibly severe damage,
to system components.
Various methods of cooling exist (for more detail see the chapter on Coolers). The use
of inter-coolers can be satisfactory where the environment temperature is not excessive
and may be reduced to a single heat exchanger which ensures that the fluid temperature
at the pump never exceeds a design maximum temperature consistent with the type of fluid
employed. At higher environmental temperatures simple cooling will no longer be
effective and individual components themselves may require cooling - for example, by
water circulation through jackets or by water or alcohol sprays. This results in considerable
complication of the system, and cooling systems of this type tend to be heavy and bulky
and not always entirely reliable.
With a cooled system any of the conventional fluids may be employed (a fire-resistant
fluid, for example, in place of a mineral oil, if required, to reduce fire hazard) and the
design problem is simply one of providing adequate cooling. The choice of fluid may,
however, be influenced by the specific heat and thermal conductivity of the fluid which
govern the size of the heat exchanger needed to remove excess heat from the system.
The most satisfactory high-temperature fluids so far produced are of the silicone type,
although methyl silicones are poor as lubricants unless suitably modified. Disiloxanes are
also subject to hydrolytic instability. All silicones tend to have a low bulk modulus (that
is they are fairly readily compressed) and are thus somewhat 'elastic' in very high pressure
systems.
Seals
Where very high fluid temperatures are expected. and a suitable fluid is available (for
example, a modified silicone), the design of appropriate seals sets a specific problem. Not
only is the service temperature of elastomeric seals limited but any lack of marginal
compatibility is aggravated by high fluid temperatures. For very high temperature
systems, therefore, and particularly in the case of dynamic seals, it may be necessary to
design a 'sealIess' system, or one which employs metalIic seals only, in materials
compatible with the fluid at the working temperature.
Silicone (and similar) high-temperature fluids are of fire-resistant or 'non-flam' type
and thus provide an additional safety measure for high-temperature working. Their cost
would, however, prohibit their selection as a 'non-flam' fluid in conventional systems
operating at low or moderate temperatures. At such temperatures, too, their low bulk
modulus would be a distinct disadvantage in high-pressure systems.
The main limitation of silicones is lack of compatibility with conventional elastomers
and with certain metals at high temperatures. Aluminium, magnesium and copper are
attacked at high temperatures and thus stainless steel is normalIy recommended for high-
duty systems with silicone fluids. It is also recommended that alIjoints are welded in such
cases. Stainless steel is difficult to flare and flareless compression joints or screwed joints
may be prone to leakage. Conventional elastomers may be used with silicone fluids at low
fluid temperatures (for example, up to 95C, or 175C in some cases) but show some
shrinkage. For higher temperatures Vi ton is applicable up to 260C. Above this, no
elastomeric seal is likely to be suitable in any case since alI elastomers are limited in
service temperature. (For more detail see the chapter on Seals.)
Other components
The use of components and fluids compatible with the high service temperature required
represents a simpler, more direct solution, but at the expense of considerably reducing the
choice of fluid and placing a premium on component specification. The problem in this
case is two-fold -
first, the selection of a suitable fluid, and
second, the re-rating of components (or selection of special components) at the
higher service temperature.
In addition, the choice of suitable seals may be particularly restricted by the need to
achieve compatibility with the special fluid at the working temperature involved.
AII metallic components wiIl suffer a certain loss of mechanical strength with increas-
ing temperature, although this is seldom serious (and usualIy negligible) within usable
temperature limits. Thus tungum alIoy suffers a loss of some 20% at400C, compared with
its strength at normal temperatures. Light alIoys show a more marked loss and, in any case,
are not suitable for lines in high-pressure systems. Steels are less temperature dependent
and stainless steel can be considered to be unaffected by temperatures within the range
likely to be encountered in practice for any system. Nickel alIoy tubing is an alternative
choice for very high temperature systems and has some advantage in manipulation.
HIGH TEMPERATURE HYDRAULICS 535
Liquid-metal fluids
Liquid metals offer the possibility of being used as hydraulic fluids in highly specialised
systems working at fluid temperatures up to 650C. One of the most suitable metals so far
evaluated is a eutectic alloy of sodium and potassium (23% sodium, 77% potassium)
which has a melting point of -12 DC and a boiling point of approximately 705C under
atmospheric pressure. The density of this liquid alloy is comparable to that of water. A
further advantage is that the alloy shows no signs of degradation or deterioration with
either time or temperature over its whole liquid range.
The problems of utili sing such a fluid in a practical system are, of course, considerable.
Sodium-potassium alloy reacts violently with oxygen or moisture and so special tech-
niques and extreme precaution have to be taken to prevent contamination of the fluid. Its
lubricating properties are also poor and its tendency to provide fluxing action can promote
localised welding although the high thermal conductivity is helpful in reducing high
temperature in bearings and shaft seals.
Sealing presents a particular problem, demanding the use of metallic or ceramic rotating
face seals, or metallic rings and reeds for sliding seals. Both vane and centrifugal type
pumps with appropriate modifications and bellows-type shaft seals have shown them-
selves capable of handling the alloy at pressures up to 14.06 MPa (140.6 bar) and
temperatures in excess of 5400C. Bearings and seals lubricated by the fluid remain the most
critical items.
With a liquid-metal fluid a unique opportunity is offered for the design of servo-valves
utilising the conductivity of the metal (fluid), such as the employment of a small electro-
magnetic pump as a combination transducer and pilot stage.
High temperature testing rigs

Figure 7.2.2 shows a pack able to deliver a controllable high temperature hydraulic fluid
to a test system. This enables small systems, such as used in the automobile industry, to
be thoroughly tested at the required temperature and pressure likely to be experienced in
use. Temperatures between 15 DC and 135C (or even 150C) are ensured for system flows
up to a pressure of 21 MPa (210 bar) normally up to 13 L/min.
537
ULTRA-HIGH PRESSURE
HYDRAULICS
THE WORDS 'high' and 'ultra high' are not as absolute as many might like to believe.
Each decade changes the levels to suit the current technology. At the start of the 21st
century 'high' might be considered as above 35 MPa (350 bar), and 'ultra high' as greater
than 50 MPa or 100 MPa (500 bar or 1000 bar). At pressures above 30 - 35 MPa (300-
350 bar) the system experiences changes in dimension of both fluids and system
components, higher material stresses, and demand for closer working clearances and
tolerances all begin to cancel out the advantages expected of higher pressure working. (See
Properties offluids.) Thus for hydrostatic systems in general, maximum practical working
pressures might be set at 56 - 70 MPa (560 - 700 bar). The higher figure also represents
the maximum normally adopted for actuators.
Practical fluid systems can, however, be worked at very much higher pressures. Typical
examples are liquid springs where the working fluid performs as a compressible medium,
and chemical processing such as polythene production, where pressures of the order of280
- 350 MPa (2800 - 3500 bar) may be utilised. Ultra-high pressures are also used for metal
extrusion, hydraulic forming, the pressing and compacting of metal and other powders and
auto-fretting. Other fields include high-pressure testing and the further development of
fluid-pressure devices in the still relatively unexplored fields of metal ductility, chemical
reactions and changes of state under extreme pressure conditions.
The practical limit for fluid pressure devices yet developed is of the order of 40 - 50 GPa
(400 - 500 kbar). Within this limit, a current suggestion of the limiting pressures for
systems is shown in Table 7.3.1. and for components in Table 7.3.2 . However, it should
be pointed out that the components at the upper limits are specials, and the great majority
of available components are normally limited to much lower pressures.
Generation of ultra-high pressure

Pumps of the mechanical type are generally unsuitable for producing ultra high pressures
and so hydraulic intensifiers are normally used as the source of fluid power. In the case
of an all-hydraulic system the source of primary pressure is a pump, with pressure
ranging up to about 110 MPa (1125 bar). Final pressures up to 1.4 GPa (14 kbar) can
readily be achieved by such devices, although the majority are designed for lower pressure
working.
Measurement of ultra-high pressure

The normal method of measuring pressure up to about 1.4 GPa (14 kbar) is the free piston
gauge. The diameter of the piston is measured and the area calculated whilst the weight
it supports can be compared very accurately with the standard. When pressures are low the
area of the piston is fairly large and can be calculated with great accuracy. If a lubricating
oil is used as the pressure medium the diameter of the cylinder is made 0.005 mm (0.0002
in) larger than the piston and the effective diameter is then 0.003 mm (0.0001 in) greater
than the piston diameter.
With increasing pressure the piston diameter must be reduced, both to reduce the
supported weight and limit the stresses in the cylinder. The relative accuracy to which the
piston can be measured then becomes less, and also knowledge of the thickness of the oil
ULTRA-HIGH PRESSURE HYDRAULICS 539
film between piston and cylinder. As the pressure increases the piston and cylinder
diameters also expand, and the true value at any pressure must be known.
One method of evaluation is to compare the behaviour of similar free-piston units made
from materials with different elastic coefficients. Alternatively, the necessary data is
obtained from theoretical considerations.
The hydraulic circuit for a typical proprietary dead-weight tester is shown in Figure
7.3.1. The main pump supplies the intensifier through a directional control valve and a
pressure-loaded relief val ve. The springs of this valve give a relief pressure of around 0.35
MPa (3.5 bar), but when the needle valve is opened this pressure is applied to the spring
end of the plunger, automatically increasing the relief pressure. Thus the operator need
only turn this valve and the pressure rises automatically until the valve is shut. The
intensifier discharges through non-return valves and the high pressure circuit is released
through a special needle valve.
The low-pressure pump is connected directly to the hydraulic weight-revolving motor

so that all the oil pumped passes through the motor, or if it is stationary, through the bypass
valve. The motor exhaust is then led to a second pressure-loaded relief valve governing
the opening ofthe loading val ve, producing pressure to lift the weight off-loading cy linder.
As each successive weight is picked up the pressure increases and is locked in the cylinder
by the check valve. The weights are on-loaded by opening a release valve.
Dial gauges are also available for pressures up to 1.05 GPa (10.5 kbar), usually based
on a Bourdon tube with a high-tensile steel tube.
Fluids for ultra high pressure

The practical limit for working of conventional mineral oils is 350 - 400 MPa (3.5 - 4
kbar), above which they become so viscous as to approach 'solid' characteristics. The
working pressure range for such fluids can, however, be extended up to about 690 MPa
(6.9 kbar) by the addition of viscosity index improvers, provided the fluid itself is not
subjected to continuous high-shear rates which cause a break-down of the additive and
some loss of performance.
Water does not suffer from this effect, but due to its very low viscosity is extremely
difficult to seal. It is thus less satisfactory as a working fluid for high-pressure systems,
although it may be used as the fluid medium for ultra-high-pressure testing, particularly
destruction testing where the fluid is lost.
Castor oil has a superior performance to mineral oils as regards viscosity/pressure
characteristics and can be worked up to about 690 MPa (6.9 kbar) without excessive
thickening. Castor-alcohol mixtures, or similar castor-base fluids retain good fluidity up
to very much higher pressures - see Table 7.3.3.
Glycerine is one of the best fluids for ultra-high-pressure working, remains fluid up to
the highest practical pressures and has a very high bulk modulus. It is more generally used
in the form of water-glycol mixtures, however. The primary limitation of such fluids is
their tendency to cause rusting on ferrous metal components.
Synthetic lubricants are an alternative choice, although they tend to be expensive. This
can be partly offset by diluting the fluid with a compatible low-viscosity liquid. Many of
the synthetic lubricants also have high bulk moduli, which are desirable to reduce
compressibility effects, and low freezing points, which are essential to prevent solidifica-
tion if the fluid is rapidly expanded on the high-pressure side.
In addition to governing the 'elasticity' of the ultra-high-pressure system, the bulk
modulus of the fluid governs the efficiency of the intensifier. The more the fluid is
compressed in developing the ultra-high pressure (i.e. the lower the bulk modulus of the
fluid), the greater the wastage of power input in compressing the fluid. The bulk of this
power loss is stored in the compressed fluid (a proportion being transformed into heat),
and will be released when the pressure restraint is removed. Thus the higher the bulk
modulus of the fluid the lower the 'explosive' release of energy in the event of a break in
the system or a component, as in destruction tests.
Pipes, fittings and seals in ultra-high pressure systems

All fluid devices and their connections on the ultra-high side are potentially hazardous.
Special constructions may be required, and certainly special pipes and fittings. All designs
of ultra-high pressure equipment try to keep dimensions as small as possible to minimise
the bursting stresses involved. Tube bores are usually 3 mm or 1.5 mm, with a suitable wall
thickness to accommodate the maximum hoop stress within the maximum permissible
working-stress for the material. Composite tubes may be used for maximum strength with
minimum overall diameter. Intensifier high-pressure cylinders may be as large as 25 mm,
but would normally be less for pressures above 700 MPa (7 kbar).
Standard sizes of tubing have been established in both alloy and Type 316 ST. The
smallest size is 1.5 mm bore by 6 mm o.d. which is good for 690 MPa (6.9 kbar), whilst
for 1.39 GPa (13.9 kbar) a composite tube having a stainless steel core and alloy steel
envelope is preferred. The envelope is drawn down onto the core to pre-stress it.
Pipe joints are also standardised in principle if not in detail. The block is tapped with
an appropriate thread - 20 tpi for 6 mm o.d. pipe and the inlet hole coned at an angle of
60" (Figure 7.3.2). The pipe is screwed with a left-hand thread on which a stop sleeve is
screwed and coned at an angle of 59 so that the joint is first made at the small end of the
cone. The force tending to burst open the joint is obviously dependent on the joint area and
by this means it is kept to a minimum.
Larger joints are more of a problem and the details may depend on the general design.
The principle of unsupported areas first introduced by Bridgman is a useful guide; this
joint, in the form of a plug or piston seal, is shown in Figure 7.3.3. The force on the end
of the plug is transmitted to the packing - usually non-metallic - which, being annular, has
a lesser area than the plug. The pressure tending to flatten it is therefore about 50% greater
than if it had no hole and the packing is forced into the hole with correspondingly greater
force. The force can be so great that the neck of the plug is pinched off.
The same principle is applicable to outlets for cylinder ends (Figure 7.3.4). As the fluid
velocity is low, the outlet bore can be small so that the spigot is not seriously weakened.
The conical seat for the packing also tends to reduce the inward thrust as it is compressed
under pressure.
Provided that extrusion can be prevented, O-rings have a greater sealing capacity. One
method of doing this is to insert a triangular-section close-fitting ring behind the O-ring
(Figure 7.3.5). The O-ring under pressure behaves as a very viscous fluid and it is
reasonable to assume that, like all fluids, it becomes more viscous as the pressure
increases.
High pressure aircraft seals present a double problem in that not only do the seals have
to prevent leakage at the high pressure, but they also have to cope with the fire resistant
phosphate ester fluid which is used.
Auto-frettage
Auto-frettage is widely used for treating cylinders which have to withstand high pressures,
although the highest pressure for which it can be applied is limited by the fact that auto-
frettage requires application of a pressure appreciably higher than the working pressure.
Auto-frettage consists of pressurising a cylinder until the yield point is exceeded; it is
then given a low-temperature stress-relieving heat treatment. The effect is to produce high
compressive stresses in those parts of the cylinder which are subjected to tensile stresses
when it is under pressure. The result is that the safe working pressure may be doubled, as
the risk of fatigue, due to reversal of stress, is eliminated.
There is a limit to the pressures for which auto-frettage is applicable, simply because
it is impossible to produce the over-pressure. One method of overcoming this, in suitable
cases, is with taper construction, as shown in Figure 7.3.6.
FIGURE 7.3.6 - Taper construction for strengthening ultra-high pressure cylinders. The piston
packing of alternate washers of copper and leather is effective for high pressures.
Isostatic pressing
Isostatic pressing involves applying pressure uniformly over the whole surface of a subject
undergoing compaction, and is applied particularly to powdered forms of solids. Various
media may be used for applying the pressure, a high-pressure or ultra-high-pressure fluid
generally being the most convenient for low-temperature working. A normal requirement
in such cases is that the subject material and pressurising fluid must be separated, usually
by putting the material into some form of flexible container or flexible mould, which is
then sealed to prevent entry offluid. It is this principle of applying the compacting pressure
uniformly overthe whole surface being treated which distinguishes isostatic pressing from
conventional die pressing and forming. A further practical advantage is that the reactive
force available can be very high compared with that given by a conventional press, with
a consequent saving in bulk and weight of equipment.
A typical form of isostatic press is shown in Figure 7.3.7, comprising essentially a

heavy-walled pressure vessel with a detachable top plug for access to the working space,
with suitable seals. This is allied to a pumping system, which must be capable of generating
the required fluid pressure at a suitable flow rate to provide an economic pumping time,
and in particular allow for the considerable reduction in displacement volume of the
subject undergoing compaction.
The subject is loaded in the form of a 'tool' which comprises the flexible mould or bag
isolating the material from the fluid, mounted as an assembly with the powdered material
and a form of mould for producing the required shape or form. This is vented as necessary
to allow hydrostatic pressure to reach the outside of the flexible bag, compacting the
powder to the pre-determined shape set by the rigid members of the assembly.
The pressure required depends primarily on the pressure-density characteristics of the
powder being compacted. The isostatic nature of the system means that the surface of the
powder is subject to constant pressure, and thus implies constant powder density during
compacting, although the pressure system within the piece will depend on the transmis-
sibility ofthe powder. In practice, very constant densities are normally achieved, certainly
superior to those produced by mechanical pressing processes.
Rate of pressure application, dwell time and rate of pressure release are also important
parameters affecting the quality of the finished product. Rate of pressure application is
controlled by the rate of pumping, and rate of pressure release by releasing the pressurised
fluid through a metering valve, optimum times for all three phases normally being
determined by sample runs.
Isostatic pressing is also extended to sheet metal forming and other fabrication
techniques, where the pressure systems within the subject may be further modified. The
pressing of ready-formed solid materials, rather than compacting, is generally referred to
as hydrostatic pressing.
Liquid springs
The liquid spring consists essentially of a stout cylinder, stressed to withstand very high
pressures, enclosing a piston and rod assembly adequately sealed at the rod end by high-
pressure seals - Figure 7.3.8. The piston incorporates two orifices, one large enough to
permit rapid flow and the other acting as a metering orifice or recoil orifice. Opening of
the rapid-flow orifice is controlled by a suitable valve to give one-way opening. The
metering orifice provides for flow in either direction.
On the retraction stroke, the piston is driven up the cylinder and thus the volume
available for the fluid is reduced by that volume of piston rod entering the cylinder,
resulting in compression of the fluid. It is this compression to very high pressures (about
350 MPa) which provides 'spring' action due to the compressibility of the fluid and to a
lesser (and usually negligible) extent, the elasticity of the cy linder itself. At the same time
dashpot damping is provided by the main orifice being open and by flow through the recoil
orifice.
On the extension or recoil stroke, piston movement is energised by the expansion of the
fluid from its compressed state The main orifice is closed by the valve, but flow from one
side of the piston to the other can take place through the metering orifice. This flow
provides damping on the recoil stroke
Pressure build-up on the 'spring' stroke is progressive and directly proportional to the
stroke (governing the length of rod entering the cylinder, and thus the physical reduction
in fluid volume). Compressibility to be anticipated with a mineral oil fluid is of the order
of3.5% per 7 MPa (70 bar), although the relationship will not be linear, particularly when
higher pressures are reached towards the end of the stroke. Thus the spring action tends
to become progressively 'stronger' with increasing stroke, and, if the rod section is
particularly generous, may produce sufficient reduction in fluid volume to raise the fluid
pressure to a point where it becomes virtually solid. Primarily, therefore, the spring
reaction is determined by the diameter of the rod relative to the cylinder diameter, and
small changes in rod size are more effective in controlling the spring 'rate' than the
selection of fluids with different bulk moduli.
Hydraulically stressed bolts

Ultra-high hydraulic pressures of the order of 210 to 350 MPa (2100 to 3500 bar) are used
in special designs of high-duty bolts capable of being stretched longitudinally as a
preliminary to fitting. The resulting reduction in diameter enables the bolt to slide easily
into a specified size of hole. On release of hydraulic pressure, the bolt expands to provide
the correct degree of fit to maintain a definite pre-determined fitting force between the bol t
and hole. Similarly, such a bolt can easily be removed again, if necessary, by hydraulic
stretching.
This form of bolt obviates the need for hammering bolts into position where particularly
rigid or accurately aligned assemblies are required, and eliminates any possibility of
scoring the bolt or bolt hole surfaces.
The design comprises a blind drilled bolt with a reduced diameter screwed head which
can be assembled as a unit with a loose piston rod, piston, rubber 'tyre' and a hydraulic
head, in that order - Figure 7.3.9. Hydraulic pressure is applied directly to the centre of
the hydraulic head and is transmitted via the tyre and piston to the piston rod, generating
a pre-determined force sufficient to stretch the length of the bolt the required amount, but
without exceeding the maximum permissible stress in the bolt material. The reacti ve force
is contained jointly by the hydraulic head and the bolt head to which it is attached.
The total contraction which can be achieved on the bolt is of the order ofD/2 500, where
D is the bolt diameter. This is generally adequate to provide a radial gripping pressure of
the order of 21.2 MPa (212 bar) between the bolt and hole surface. Axial grip can also be
provided, if required, by controlling the run-up tightness of the nut.
547
WATER HYDRAULICS
THE USE of water as an hydraulic fluid has had a chequered history. It was the first and
only hydraulic fluid at the dawn of fluid power, and it continued unchallenged throughout
the nineteenth century. However, with oil and the discovery of suitable oil resistant seals,
the situation rapidly changed, and by the 1930' s almost all fluid power was oil operated.
(Some notable exceptions being heavy industry on large presses, steel rolling mills and
mining.)
The problem with water was not that the fluid was poor, but rather that the machinery
could not cope. Water, with its low compressibility and viscosity, is the best of all
hydraulic fluids for transmitting power. However, corrosion and leakage, particularly at
the required high pressures, forbade its use in most applications - unless much higher costs
were warranted.
Now, the design of specific water hydraulic equipment is beginning to change the scene.
Better sealing, improved surface friction capabilities and cheaper corrosion resistant
materials have combined to enable initial costs to be greatly reduced. The running costs
are significantly less than oil and the safety and environmental cleanliness features are
much appreciated; this all contributes to a highly desirable system.
100 % water has not taken over, nor will it totally take over, in all machinery. It is still
of higher initial cost than oil hydraulics, and there is the obvious problem of a high freezing
point meaning that it could be a source of complete failure in low temperature environ-
ments . Water based fluids, and those specifically with an anti-freeze included can be used
to lower temperatures but care must be taken that their limits are not exceeded.
In summary, the following are the main advantages of water based hydraulic applica-
tions:
Lower fluid purchase costs
Lower storage costs
Lower disposal costs
Reduced system pressure losses
Totally fire resistant
No-mess maintenance
Reduced insurance premiums
Cleaner industrial operations

Environmental friendly
No oil contamination of food products
Fluids
Fluids are discussed fully in the chapter on Types of fluid. However, it should be noted that
'water' in industrial applications is never 100 % H20; there are contaminants present
which could be corrosive (such as saline products), and bacteria may multiply in static
regions.
Water may intentionally have a proportion of another fluid, e.g. a lubricant; or an
antifreeze such as a glycol, or a biocide. The raw water (the normal term used) may be 'tap'
or 'sea' water, and the mechanical components in the system must be appropriate and
compatible for the water in use.
Pumps and motors

The design of suitable pumps and motors needs to cover the problems of leakage,
corrosion and lubrication.
One solution has been to divide off the water/pressure side from the oilllubricant side.
However, although this is used in, say, mining applications, it may allow a certain amount
of mixing of oil and water which could be unacceptable in closely controlled environments
such as food manufacture.
Pumps for high water base fluids such as the oil-in-water dilute emulsion (95% water
5% oil) are often similar to the mineral oil equivalent and, indeed, may just be derated oil
pumps with suitable seals.
So called 'pure' water pumps are usually fixed displacement and are available in in-line
piston, radial and axial piston designs - see a radial piston pump in Figure 7.4.1. The older
in-line piston design is slower running but able to provide the ultra high pressures
sometimes required; three or five pistons are quite common, and a reduction gear box
(epicyclic or worm & wheel) may be used to bring the speed down from the supply 1500
rev/min. A separate oil supply is used for the lubricant. The newer axial and radial designs
use the water as both the pressure fluid and the lubricant. Typical maximum conditions are
shown in Table 7.4.1.
WATER HYDRAULICS 549
Valves
The major problem associated with valves is the lower viscosity of the fluid; this causes
greater leakage than for oil with similar designs. Flow velocities are approximately double
those where oilis used. Hence it is common to use seat valves - poppet and ball seat - rather
than steel sliding spool valves.
For nominal sizes up to about 10 mm, ball seat valves are used, although ceramic spool
valves are now being introduced - see Figure 7.4.2. For sizes beyond this, it is more
common to use 2/2 way logic elements (either face mounting or cartridge type).
All the oil valve types are available for 'pure' water except the true servo-valve.
Proportional valves may, however, provide the service necessary.
Cylinders
Cylinders are exceptionally prone to corrosion. Coating, or the use of acid/corrosion
resistant stainless steels, is essential, with the better the bore finish the lower the corrosion
(it can be cleaned more easily) and the lower the leakage. Seals and bushes need to be
compatible with the water.
Pipes and hoses

Although it is common practice for the same design and size of pipes to be used for both
oil and water, the pressure peaks found in water systems are higher than those in oil
systems. As with oil hydraulics the tube material tends to be a ductile steel with non-
embritlling properties.
Hoses are less used in water systems due to the generally higher flow rates requiring
greater bore sizes.
Manifold blocks
The use of manifold blocks in small systems is highly advantageous providing the blocks
are strong enough. The requirement for pipes is then less, and the pressure peaks can be
contained. It is because of these peaks that the conventional cast blocks are not favoured;
hot forged quality steel (e.g. SAE 1044) is acceptable. Stainless steel blocks are more
expensive and are generally only used in 'pure' water systems.
Seals
There is a good range of seals available which are compatible with water. Typically these
are in nitrile, silicone and ethylene propylene (EP), although other materials are also
available. More detail is given in the chapter on Seals.
Reservoirs
A major problem in reservoirs is the fluid surface which is open to air. Static regions
greatly encourage biological growth and corrosion. Even so the reservoir tends to be larger
(l1/2X)than for oil systems because water absorbs and gives off heat more slowly than oil.
Although thorough painting may provide a corrosion resistant surface at the commence-
ment of a system operation, it is a short term expediency when examples of the paint have
been seen to peel off and cause secondary damage in a system.
Because the specific gravity of water is some 15 % higher than that of oil, contaminants
are more buoyant in water and hence they settle more slowly. See chapter on Reservoirs.
Filters
Apart from the restriction of not using paper filters (which generally swell) most oil filters
are acceptable. However, housings must be in stainless steel. As with all hydraulic
systems, filtration is a major component and must not be neglected - contaminant causes
failure in water hydraulics components just as it does in oil hydraulics components.
Circnit design
There are some distinct differences between the oil hydraulic system and the water
WATER HYDRAULICS 551
hydraulic system. The circuit is designed for higher velocities, say between 6 and 10 m/s
with valve flow velocity around 12 m/s. Valves, as mentioned above, tend to be seat val ves
and 2/2 way logic functions (as with large flows in oil hydraulics) are used. Because the
pumps are more reliable when of fixed displacement design (at present) there is a greater
use of accumulators - these have the helpful, secondary, function of smoothing out
pressure peaks.
Applications
Although most applications are suitable for water hydraulics - except for the low
temperature ones - there are some for which there is a significant advantage:
Mining (intrinsic safety)
Hot metal processing, e.g. steel, copper, aluminium (fire prevention)
Paper and wood-particle board processing (prevention of oil contamination of
product)
Food industry (contamination avoidance)
Textiles
Medical (contamination avoidance)
Subsea (especially sea water availability)
Rescue tools (safety)
Radioactive areas (safety)
553
HYDRO-PNEUMATICS
THE TERM hydro-pneumatics applies to systems, machines or devices employing both

a hydraulic component (liquid working medium) and a pneumatic component (gaseous
working medium). These include air-hydraulic cylinders and intensifiers as the most
common components. Other air-hydraulic systems include pulsation dampers, surge and
shock preventers and check units.
Air-hydraulic cylinders
Air-hydraulic cylinders combine the features of hydraulic control with the simplicity of
an air supply. A particular attraction of air-hydraulic cylinders is that they can provide the
rigidity and speed control normally associated with hydraulic systems without the cost of
an hydraulic pump and driver. They are only appropriate (and air-hydraulic systems in
general) where light output forces are required.
The air-hydraulic cylinder lends itself to integral construction, as in Figure 7.5.1,
even though the two circuits are quite separate. The single rod end of the combined
cylinder is usually made the air cylinder. Since the hydraulic cylinder is through-rod only
a nominal size oil reservoir is required, which is commonly of the spring loaded type.
Reversing the positions of the air and oil cylinder would nullify this advantage. Throttle
valves in the hydraulic lines provide for adjustable speed settings, with the same
Almost any desired sequence of speed control associated with linear motion can be
obtained by suitable design of the hydraulic circuit. Independent valves can be inserted to
provide a check at any particular point, or cam-operated valves can be used to provide skip
feeds. The hydraulic circuit affects only the speed of operation of the system. It has no
effect on the air circuit and its controls, except to prohibit the use of unloading or pressure-
relief valves sensitive to back pressures. The design of the air circuit can follow
conventional practice and be as simple or complex as necessary. Proprietary air-hydraulic
cylinders are normally available with plain hydraulic pistons (for controlled speed in
either direction via regulator valves in the hydraulic circuit) or with a one-way valve in the
hydraulic piston to give fast forward or return motion.
Separate air and hydraulic cylinders

Similar solutions can be obtained by mechanically connecting separate air and hydraulic
cylinders. Where only a proportion of the working stroke needs damping this has the
advantage that the air cylinder rod contacts the hydraulic cylinder rod only on that portion
of the stroke where damping is required. It is important in such cases, however, that there
should be sufficient stroke length on the hydraulic cylinder to accommodate any
mechanical adjustments required, as otherwise it might be possible to adjust the working
movement in such a way as to 'bottom' the hydraulic cylinder.
More complex mechanical systems may be used, normally with air and hydraulic
cylinders of equal stroke, the two locked and unlocked, as required, by mechanical
couplings and latches. If necessary movements may manipulate regulating valves in the
hydraulic circuit itself, and this is often preferred to mechanical systems where differential
speed control is required. Differential speed control can also be obtained by by-passing the
appropriate flow regulator in the hydraulic circuit to give unrestricted flow. Such a valve
could be mechanically operated by hydraulic piston movement.
The combined air-hydraulic cylinder has the advantage of convenience, only one unit
to be mounted, but it has two possible disadvantages compared with the use of separate
air and hydraulic cylinders. One is that the longer length may make it less suitable for
certain installations. The other is that the through rod relies on seal design to ensure that
no air can bleed through into the oil circuit, as this could lead to jerky checking actions.
The risk is generally eliminated by incorporating a bleed hole in the central section
between the two sets of rod seals. Any leakage from either cy linder will then be bled off
to atmosphere, rather than into the other cylinder. The presence of air or oil at the bleed
hole will also indicate seal leakage.
Intensifiers
Air-hydraulic devices of the type described above can also be used to convert from a
pneumatic to a hydraulic form of power transfer. The normal pressure levels in pneumatic
systems make such concepts of limited interest. However, intensifiers, also known as
(pressure) boosters, are units designed to provide a high pressure fluid (hydraulic) output
from a lower pressure fluid (pneumatic) input. Most types of intensifiers provide complete
separation of the low pressure and high pressure fluids such that different fluids may be
HYDRO-PNEUMATICS 555
used. Thus, compressed air (or gas), water or oil may be used on the low pressure side, with
oil as the high pressure fluid. Water is sometimes favoured as the high pressure fluid for
intensifiers designed for high pressure test work because of its low compressibility (less
than that of oil). For very high pressure testing, a fluid with even lower compressibility
may be used.
Intensifiers with a hydraulic output can be all-hydraulic or air-hydraulic. The working
principle is the same in both cases, the intensifier comprising two piston/cylinders of
different diameters mounted on a common rod as in Figure 7.5.2.
Low pressure is fed to the larger piston, generating high pressure via the smaller piston,
with pressure multiplication directly proportional to the ratio of the piston areas. The
pressure ratio achieved is given by:
Ideal pressure ratios are given in Table 7.5.1, assuming an efficiency of 100%. The
efficiency achieved is dependent on friction and internal leakage, and also the amount of
heating of the fluid. For simple' one-shot' or single-stroke intensification such losses may
be negligible, i.e. " = 100%. Where the intensifier is operated to give a continuous high
pressure output the actual working efficiency achieved may be reduced to 80%.
Continuous delivery can be provided by employing two intensifier cylinders operating
alternately. They can be separate units or, more conveniently, co-axial or tandem units,
where a single high pressure cylinder can be common to both 'working' cylinders. These
are generally known as continuous intensifiers. Figure 7.5.3 shows a double acting
used. Thus, compressed air (or gas), water or oil may be used on the low pressure side, with
oil as the high pressure fluid Water is sometimes favoured as the high pressure fluid for
intensifiers designed for high pressure test work because of its low compressibility (less
than that of oil). For very high pressure testing, a fluid with even lower compressibility
may be used.
Intensifiers with a hydraulic output can be all-hydraulic or air-hydraulic. The working
principle is the same in both cases, the intensifier comprising two piston/cylinders of
different diameters mounted on a common rod as in Figure 7.5.2.
Low pressure is fed to the larger piston, generating high pressure via the smaller piston,
with pressure multiplication directly proportional to the ratio of the piston areas. The
pressure ratio achieved is given by:
Ideal pressure ratios are given in Table 7.5.1, assuming an efficiency of 100%. The
efficiency achieved is dependent on friction and internal leakage, and also the amount of
heating of the fluid. For simple 'one-shot' or single-stroke intensification such losses may
be negligible, i.e. 11 = 100%. Where the intensifier is operated to give a continuous high
pressure output the actual working efficiency achieved may be reduced to 80%.
Continuous delivery can be provided by employing two intensifier cylinders operating
alternately. They can be separate units or, more conveniently, co-axial or tandem units,
where a single high pressure cylinder can be common to both 'working' cylinders. These
are generally known as continuous intensifiers. Figure 7.5.3 shows a double acting
intensifier designed for pressures up to 340 MPa (3.4 k bar). This embodies two high
pressure cylinders with a central low pressure cylinder. At the end of each stroke a micro-
switch is triggered by plungers to give reversal through the solenoid valve and to
supercharge the high pressure cylinder with low pressure fluid. This results in a consid-
erable improvement in volumetric efficiency. Air-hydraulic intensifiers are normally
designed to operate on shop air at a pressure of about 0.7 MPa (7 bar) and may have
pressure ratios as required.
559
STANDARDS
THREE SETS of fluid power standards are described in this chapter. A range of
Guidelines and data sheets from the British Fluid Power Association (BFPA) is also
appended. The initial selection of standards is taken from the International Standards
Organisation (ISO), and deals with standards for manufacture and use of fluid power
components and systems. The second set is a similar compilation but comes from the
American National Fluid Power Association (NFPA). The third set is from the Institute
of Petroleum (IP), and define the tests which may be undertaken on hydraulic fluids. Some
idea of the size of the standard can be ascertained from the number of pages within the
document - shown in square brackets after each title.
ISO standards
There are well over 100 ISO standards for hydraulic fluid power. The majority ofthese are
included in the list below. In addition there are some standards which relate to fluid power,
but which are not exclusive to fluid power - such as hoses and fittings, and for these only
a selection is given. The list (including 'Aerospace') is arranged as follows:
General
Accumulators
Brake systems
Cylinders
Filters (and contamination control)
Fittings (other)
Fluids (and leakage)
Hoses and tubes, connections and couplings
Pumps and motors
Seals
Valves
General
ISO 31: 1992, Quantities and units - [13 parts]
ISO 1000: 1992, SI units and recommendations for the use of their multiples and of certain
other units (Ed. 3) [22 pp]
ISO 1219-1: 1991, Fluid power systems and components - Graphic symbols and circuit
diagrams - Part 1: Graphic symbols (Bilingual edition) [40 pp]
ISO 1219-2: 1995, Fluid power systems and components - Graphic symbols and circuit
diagrams - Part 2: Circuit diagrams [21 pp]
ISO 2944: 1974, Fluid power systems and components - Nominal pressures [1 p]
ISO 4413: 1979, Hydraulic fluid power - General rules for the application of equipment
to transmission and control systems (Ed. 3) [17 pp]
ISO 5598: 1985, Fluid power systems and components - Vocabulary (Bilingual edition)
[90 pp]
ISO 5784-1: 1988, Fluid power systems and components - Fluid logic circuits - Part 1:
Symbols for binary logic and related functions [21 pp]
Symbols for supply and exhausts as related to logic symbols [5 pp]
Symbols for logic sequencers and related functions [10 pp]
ISO 5859: 1991, Aerospace - Graphic symbols for schematic drawings of hydraulic and
pneumatic systems and components [2 pp]
ISO 6771: 1987, Aerospace - Fluid systems and components - Pressure and temperature
classification [1 p]
ISO 9110-1: 1990, Hydraulic fluid power - Measurement techniques - Part 1: General
measurement principles [4 pp]
ISO 9110-2:1990, Hydraulic fluid power - Measurement techniques - Part 2: Measure-
ment of average steady-state pressure in a closed conduit [6 pp]
ISO 10448: 1994, Agricultural tractors - Hydraulic pressure for implements [2 pp]
Accumulators
ISO 5596: 1982, Hydraulic fluid power - Gas-loaded accumulators with separators -
Range of pressures and volumes, characteristic quantities and identification [4 pp]
ISO 10945: 1994, Hydraulic fluid power - Gas-loaded accumulators - Dimensions of gas
ports [2 pp]
Brake systems
ISO 611: 1994, Road vehicles - Braking of automotive vehicles and their trailers -
Vocabulary - Trilingual edition (Ed. 3) [45 pp]
ISO 3803: 1984, Road vehicles - Hydraulic pressure test connection for braking equip-
ment (Ed. 2) [2 pp]
ISO 3996:1995, Road vehicles - Brake hose assemblies for hydraulic braking systems
used with non-petroleum-base fluid [13 pp]
ISO 4925: 1978, Road vehicles - Non-petroleum base brake fluid [22 pp]
ISO 6120:1995, Road vehicles - Brake hose assemblies for hydraulic braking systems
used with petroleum-base fluid [13 pp]
ISO 6597: 1991, Road vehicles - Hydraulic braking systems - Measurement of braking
performance (Ed. 2) [19 pp]
STANDARDS 561
ISO 7308: 1987, Road vehicles - Petroleum-based brake-fluid for stored energy hydraulic
brakes [20 pp]
ISO 9128: 1987, Road vehicles - Graphical symbols to designate brake fluid types -
Bilingual edition [3 pp]
Cylinders
ISO 3320: 1987, Fluid power systems and components - Cylinder bores and piston rod
diameters - Metric series [2 pp]
ISO 3321 :1975, Fluid power systems and components - Cylinder bores and piston rod
diameters - Inch series [2 pp]
ISO 3322: 1985, Fluid power systems and components - Cylinders - Nominal pressures
(Ed. 2) [1 p]
ISO 4393: 1978, Fluid power systems and components - Cy linders - Basic series of piston
strokes [2 pp]
ISO 4394-1 :1980, Fluid power systems and components - Cylinder barrels - Part 1:
Requirements for steel tubes with specially finished bores [9 pp]
ISO 4395: 1978, Fluid power systems and components - Cylinders - Piston rod thread
dimensions and types [2 pp]
ISO 5597: 1987, Hydraulic fluid power-Cylinders - Housings for piston and rod seals in
reciprocating applications - Dimensions and tolerances (replaces ISO 5597/l) [9 pp]
ISO 6020-1: 1981, Hydraulic fluid power - Single rod cylinders - Mounting dimensions
- 16 MPa (160 bar) series - Part 1: Medium series [8 pp]
ISO 6020-2: 1991, Hydraulic fluid power - Mounting dimensions for single rod cy Iinders,
16 MPa (160 bar) series - Part 2: Compact series (Ed. 2) [16 pp]
ISO 6020-3: 1994, Hydraulic fluid power - Mounting dimensions for single rod cy linders,
16 MPa (160 bar) series - Part 3: Compact series with bores from 250 mm to 500 mm
[12 pp]
ISO 6022: 1981, Hydraulic fluid power - Single rod cylinders - Mounting dimensions-
25 MPa (250 bar) series [5 pp]
ISO 6099: 1985, Fluid power systems and components - Cylinders - Identification code
for mounting dimensions and mounting types (Ed. 2) [45 pp]
ISO 6195:1986, Fluid power systems and components - Cylinders - Housings for rod
wiper rings in reciprocating applications - Dimensions and tolerances [6 pp]
ISO 6547: 1981, Hydraulic fluid power - Cylinders - Piston seal housings incorporating
bearing rings - Dimensions and tolerances [3 pp]
ISO 6981: 1992, Hydraulic fluid power - Cylinders - Rod end plain eyes - Mounting
dimensions (Ed. 2) [4 pp]
ISO 6982: 1992, Hydraulic fluid power - Cylinders - Rod end spherical eyes - Mounting
dimensions (Ed. 2) [4 pp]
ISO 7181:1991, Hydraulic fluid power-Cylinders- Bore and rod area ratios (Ed. 2) [3 pp]
ISO 8131 :1992, Hydraulic fluid power- Single rod cylinders -16 MPa (160 bar) compact
series - Tolerances (Ed. 2) [4 pp]
ISO 8132:1986, Hydraulic fluid power- Single rod cylinders, 16 MPa (160 bar) medium
and 25 MPa (250 bar) series - Mounting dimensions for accessories [8 pp]
ISO 8133: 1991, Hydraulic fluid power-Single rod cylinders, 16MPa (160 bar) compact
series - Accessory mounting dimensions [12 pp]
ISO 8135:1986, Hydraulic fluid power- Single rod cylinders, 16 MPa (160 bar) medium
and 25 MPa (250 bar) series - Tolerances [2 pp]
ISO 8136: 1986, Hydraulic fluid power- Single rod cylinders, 16 MPa (160 bar) medium
series - Port dimensions [2 pp]
ISO 8137: 1986, Hydraulic fluid power - Single rod cylinders, 25 MPa (250 bar) series -
Port dimensions [2 pp]
ISO 8138:1986, Hydraulic fluid power- Single rod cylinders, 16 MPa (160 bar) compact
series - Port dimensions [2 pp]
ISO 10100:1990, Hydraulic fluid power - Cylinders - Acceptance test [2 pp]

ISO 2941 :1974, Hydraulic fluid power - Filter elements - Verification of collapse/burst
resistance [2 pp]
ISO 2942: 1994, Hydraulic fluid power - Filter elements - Verification of fabrication
integrity and determination of the first bubble point (Ed. 3) [4 pp]
ISO 2943: 1996, Hydraulic fluid power - Filter elements - Verification of material
compatibility with fluids [8 pp]
ISO 3722: 1976, Hydraulic fluid power - Fluid sample containers - Qualifying and
controlling cleaning methods [2 pp]
ISO 3723: 1976, Hydraulic fluid power - Filter elements - Method for end load test [I p]
ISO 3724:1976, Hydraulic fluid power - Filter elements - Verification of flow fatigue
characteristics [2 pp]
ISO 3938: 1986, Hydraulic fluid power - Contamination analysis - Method for reporting
analysis data [7 pp]
ISO 3968: 1981, Hydraulic fluid power- Filters - Evaluation of pressure drop versus flow
ISO 4021 :1992, Hydraulic fluid power - Particulate contamination analysis -Extraction
of fluid samples from lines of an operating system (Ed. 2) [6 pp]
ISO 4402: 1991, Hydraulic fluid power - Calibration of automatic-count instruments for
particles suspended in liquids - Method using classified AC Fine Test Dust contaminant
(Ed. 2) [6 pp]
ISO 4405:1991, Hydraulic fluid power - Fluid contamination - Determination of
particulate contamination by the gravimetric method [4 pp]
ISO 4406: 1987, Hydraulic fluid power - Fluids - Method for coding level of contamina-
tion by solid particles [4 pp]
ISO 4407:1991, Hydraulic fluid power - Fluid contamination - Determination of
particulate contamination by the counting method using a microscope [6 pp]
ISO 4572: 1981, Hydraulic fluid power - Filters - Multi-pass method for evaluating
filtration performance [28 pp]
STANDARDS 563
ISO 5884: 1987, Aerospace - Fluid systems and components - Methods for system
sampling and measuring the solid particle contamination of hydraulic fluids [14 pp]
ISO 7744: 1986, Hydraulic fluid power - Filters - Statements of requirements [5 pp]
ISO 11170:1995, Hydraulic fluid power - Filter elements - Procedure for verifying
performance characteristics [4 pp]
ISO 11217: 1993, Aerospace - Hydraulic system fluid contamination - location of
sampling points and criteria for sampling (/Draft endorsement of SAE ARP 4268) [I
p]
ISO 11218: 1993, Aerospace -Cleanliness classification for hydraulic fluids (Adoption of
SAE AS4059 Rev. A) [1 p]
Fittings (other)
ISO 4400: 1994, Fluid power systems and components - Three-pin electrical plug
connectors with earth contact - Characteristics and requirements (Ed. 3) [4 pp]
ISO 6952: 1994, Fluid power systems and components - Two-pin electrical plug connec-
tors with earth contact - Characteristics and requirements (Ed. 2) [5 pp]
ISO 9634: 1994, Aerospace - Fluid systems - Dimensions of plastic protective caps and
plugs [6 pp]
Fluids (and materials)

ISO 2592: 1994, Petroleum products - Determination of flash and fire points - Cleveland
open cup method (Ed. 2) [8 pp]
ISO 2719:1988, Petroleum products and lubricants - Determination of flash point -
Pensky-Martens closed cup method (Ed. 2) [11 pp]
ISO 3448:1992, Industrial liquid lubricants - ISO viscosity classification (Ed. 2) [4 pp]
ISO 4404: 1996?, Lubricants, industrial oils and related products - Family H (Hydraulic
systems) - Fire-resistant fluids (Categories HFA, HFB and HFC) - Determination of
corrosion resistance [13 pp]
ISO 4406: 1987, Hydraulic fluid power - Fluids - Method for coding level of contamina-
tion by solid particles [4 pp]
ISO 6072:1986, Hydraulic fluid power - Compatibility between elastomeric materials
and fluids [14 pp]
ISO 6073: 1980, Hydraulic fluid power - Petroleum fluids -prediction of bulk modulii [9
pp]
ISO 6743-4:1982, Lubricants, industrial oils and related products (class L) - Classifica-
tion - Part 4: Fire resistant fluids Family H (Hydraulic systems) [2 pp]
ISO 7745: 1989, Hydraulic fluid power- Fire-resistant (FR) fluids -Guidelines for use [I I
pp]
ISO 11218: 1993, Aerospace - Cleanliness classification for hydraulic fluids (adoption of
SAE AS4059 Rev. A) [1 p]
ISO 13357: 1997?, Hydraulic fluid power - Filterability of hydraulic oils
Hoses and tubes (including couplings)

ISO 4397: 1993, Fluid power systems and components - Connectors and associated
components - Nominal outside diameters of tubes and nominal inside diameters of
hoses (Ed. 2) [2 pp]
ISO 4399: 1995, Fluid power systems and components - Connectors and associated
components - Nominal pressures (Ed. 2) [I p]
ISO 5675:1992, Agricultural tractors and machinery - General purpose quick action
coupling hydraulic couplers (Ed. 2) [4 pp]
ISO 6149-1 :1993, Connections for fluid power and general use - Ports and stud ends with
ISO 261 threads and a-ring sealing - Part I: Port with a-ring seal in truncated housing
[6 pp]
ISO 6149- 2: 1993, Connections for fluid power and general use - Ports and stud ends with
ISO 261 threads and a-ring sealing - Part 2: Heavy-duty (S series) stud ends -
Dimensions, design, test methods and requirements [10 pp]
ISO 6149-3:1993, Connections for fluid power and general use-Ports and stud ends with
ISO 261 threads and a-ring sealing - Part 3: Light-duty (L series) stud ends -
Dimensions, design, test methods and requirements [10 pp]
ISO 6162: 1994, Hydraulic fluid power - Four-screw split-flange connections for use at
pressures of 2.5 MPa to 40 MPa (25 to 400 bar) - Type I metric series and type II inch
series [13 pp]
ISO 6164: 1994, Hydraulic fluid power - Four-screw, one-piece square-flange connec-
tions for use at pressures of 25 MPa and 40 MPa (25 to 400 bar) [11 pp]
ISO 6605: 1986, Hydraulic fluid power - Hose assemblies - Method of test [6 pp]
ISO 6772: 1988, Aerospace - FIuid systems - Impulse testing of hydraulic hose, tubing and
fitting assemblies [3 pp]
ISO 6773: 1994, Aerospace - Fluid systems - Thermal shock testing of piping and fittings
[2 pp]
ISO 6803: 1994, Rubber or plastic hoses and hose assemblies - Hydraulic - Pressure
impulse test without flexing (Ed. 2) [4 pp]
ISO 6805: 1994, Rubber hoses and hose assemblies for underground mining - Wire-
reinforced hydraulic types for coal mining - Specification (Ed. 2) [7 pp]
ISO 7241-1: 1987, Hydraulic fluid power - Quick-action couplings -Part I: Dimensions
and requirements [5 pp]
ISO 7241-2: 1986, Hydraulic fluid power - Quick-action couplings -Part 2: Test methods
[15 pp]
ISO 8434-1: 1994, Metallic tube connections for fluid power and general use -Part I: 24
compression fittings [28 pp]
ISO 8434-2: 1994, Metallic tube connections for fluid power and general use -Part 2: 37"
flared fittings [28 pp]
ISO 8434-3: 1995, Metallic tube connections for fluid power and general use -Part 3: 0-
ring face seal fittings [29 pp]
ISO 8434-4: 1995, Metallic tube connections for fluid power and general use -Part 4: 24
cone connectors with a-ring weld-on nipples [25 pp]
STANDARDS 565
ISO 8434-5: 1995, Metallic tube connections for fluid power and general use -Part 5: Test
methods for threaded hydraulic fluid power connections [7 pp]
ISO 8574: 1990, Aerospace - Hydraulic system tubing - qualification tests [6 pp]
ISO 10763: 1994, Hydraulic fluid power - Plain end, seamless and welded precision steel
tubes - Dimensions and nominal working pressures [3 pp]
ISO 11926-1 :1995, Connections for general use and fluid power- Ports and stud ends with
ISO 725 threads and O-ring sealing - Part 1; Ports with O-ring seal in truncated housing
[4 pp]
ISO 11926-2: 1995, Connections for general use and fluid power- Ports and stud ends with
ISO 725 threads and O-ring sealing - Part 2: Heavy-duty (S series) stud ends [11 pp]
ISO 11926-3: 1995, Connections for general use and fluid power- Ports and stud ends with
ISO 725 threads and O-ring sealing - Part 3; Light-duty (L series) stud ends [11 pp]
Pumps and motors

ISO 3019-1;1975, Hydraulic fluid power - Positive displacement pumps and motors-
Dimensions and identification code for mounting flanges and shaft ends -Part 1: Inch
series shown in metric units [8 pp]
ISO 3019-2: 1986, Hydraulic fluid power - Positive displacement pumps and motors -
Dimensions and identification code for mounting flanges and shaft ends - Part 2: Two-
and four-hole flanges and shaft ends - Metric series (Ed. 2) [13 pp]
ISO 3019-3: 1988, Hydraulic fluid power - Positive displacement pumps and motors -
Dimensions and identification code for mounting flanges and shaft ends -Part 3;
Polygonal flanges (including circular flanges) (Ed. 2) [5 pp]
ISO 3662: 1976, Hydraulic fluid power - Pumps and motors - Geometric displacements
[2 pp]
ISO 4391 :1983, Hydraulic fluid power - Pumps, motors and integral transmissions -
Parameter definitions and letter symbols (Ed. 2) [10 pp]
ISO 4392-1: 1989, Hydraulic fluid power - Determination of characteristics of motors-
Part 1: At constant low speed and at constant pressure (Ed. 2) [7 pp]
Part 2: Startability (Ed. 2) [10 pp]
Part 3: At constant flow and at constant torque [10 pp]
ISO 4409:1986, Hydraulic fluid power - Positive displacement pumps, motors and
integral transmissions - Determination of steady-state performance [15 pp]
ISO 4412-1: 1991, Hydraulic fluid power - Test code for the determination of airborne
noise levels - Part 1; Pumps (Ed. 2) [20 pp]
ISO 4412-2: 1991, Hydraulic fluid power - Test code for determination of airborne noise
levels - Part 2: Motors (Ed. 2) [20 pp]
ISO 4412-3: 1991, Hydraulic fluid power- Test code for determination of airborne noise
levels - Part 3: Pumps - Method using a parallelepiped microphone array [9 pp]
ISO 8278: 1986, Aerospace - Hydraulics, pressure compensated, variable delivery pumps
- General requirements [23 pp]
ISO 8426: 1988, Hydraulic fluid power - Positive displacement pumps and motors -
Determination of derived capacity [9 pp]
ISO 9206: 1990, Aerospace - Constant displacement hydraulic motors - General specifi-
cations [18 pp]
ISO 9632: 1992, Hydraulic fluid power - Fixed displacement pumps - Flow degradation
due to classified AC Fine Test Dust contaminant - Test method [8 pp]
ISO 10767-1: 1996, Hydraulic fluid power - Determination of pressure ripple levels
generated in systems and components - Part I: Precision method for pumps [22 pp]
ISO 12151: 1996?, Connections for hydraulic fluid power and general use-
Hose fittings - Part 1: Hose fittings with ISO 8434-3 O-ring face seals [11 pp]
- Part 2: Hose fittings with ISO 8434-1 and 8434-4 24 cone
connector ends with O-rings [11 pp]
- Part 3: Hose fittings with ISO 6162 flange connections [11 pp]
Seals
ISO 3601 -1 :1988, Fluid systems - Sealing devices - O-rings - Part 1: Inside diameters,
cross-sections, tolerances and size identification code (Ed. 2) [4 pp]
ISO 3601-3: 1987, Fluid systems - Sealing devices - O-rings - Part 3: Quality acceptance
criteria [7 pp]
ISO 3939: 1977, Fluid power systems and components - Multiple lip packing sets -
Methods for measuring stack heights [4 pp]
ISO 6072:1986, Hydraulic fluid power - Compatibility between elastomeric materials
and fluids [14 pp]
ISO 6194-1 :1982, Rotary shaft lip type seals - Part 1: Nominal dimensions and tolerances
[7 pp]
ISO 6194-2: 1991, Rotary shaft lip type seals - Part 2: Vocabulary (in seven languages)
[18 pp]
ISO 6194-3: 1988, Rotary shaft lip type seals - Part 3: Storage, handling and installation
[7 pp]
ISO 6194-4: 1988, Rotary shaft lip type seals - Part 4: Performance test procedures [3 pp]
ISO 6194-5:1990, Rotary shaft lip type seals - Part 5: Identification of visual imperfec-
tions [7 pp]
ISO 7425-1: 1988, Hydraulic fluid power - Housings for elastomer-energised, plastic-
faced seals - Dimensions and tolerances - Part 1: Piston seal housings [5 pp]
ISO 7425-2: 1989, Hydraulic fluid power - Housings for elastomer-energised, plastic-
faced seals - Dimensions and tolerances - Part 2: Rod seal housings [3 pp]
Valves
ISO 4401: 1994, Hydraulic fluid power - Four-port directional control valves -Mounting
surfaces (Ed. 2) [18 pp]
ISO 4411: 1986, Hydraulic fluid power - Val ves - Determination of pressure differential/
flow characteristics [11 pp]
ISO 5781 :1987, Hydraulic fluid power - Pressure-control valves (excluding pressure-
STANDARDS 567
relief valves), sequence valves, unloading valves, throttle valves and check valves -
Mounting surfaces [17 pp]
ISO 5783: 1995, Hydraulic fluid power - Code for identification of valve mounting
surfaces and cartridge valve cavities (Ed. 2) [4 pp]
ISO 6263:1987, Hydraulic fluid power - Compensated flow-control valves - Mounting
surfaces [17 pp]
ISO 6264: 1987, Hydraulic fluid power - Pressure-relief valves - Mounting surface [25
pp]
ISO 6403:1988, Hydraulic fluid power - Valves controlling flow and pressure - Test
methods [53 pp]
ISO 6404:1985, Hydraulic fluid power - Servovalves - Test method [23 pp]
ISO 7368:1989, Hydraulic fluid power- Two-port slip-in cartridge valves -Cavities [17
pp]
ISO 7789:1993, Hydraulic fluid power - Two-, three- and four-port screw-in cartridge
valves - Cavities [25 pp]
ISO 7790:1986, Hydraulic fluid power - Four-port modular stack valves and four-port
directional control valves, sizes 03 and 05 - Clamping dimensions [2 pp]
ISO 9461: 1992, Hydraulic fluid power- Identification of valve ports, subplates, control
devices and solenoids [3 pp]
ISO 10372: 1992, Hydraulic fluid power - Four- and five-port servo valves - Mounting
surfaces [7 pp]
NFP A Standards
The arrangement in this list is the same as the earlier ISO standard list, but Brake systems
and Fittings are missing and Reservoirs is included. Where the NFP A standard is identical
to the ISO standard, only the ISO standard has been given. Due to the large number of
standards not all have been able to be included, but the main ones are present. 'M' denotes
that the standard is to SI dimensioning. 'R' is a renewal date.
General
Accumulators
Cylinders
Fluids
Pumps and motors
Reservoirs
Seals
Valves
General
T1.21.1-1978 (RI983), Procedures for self-calibration by fluid power manufacturers [10
pp]
T2.lO.lM-1978, Metric units for fluid power applications [16 pp]

T2.1O.2M-1977, Survey on metric language by the U.S. fluid power industry [9 pp]
T2.12.1O-1993, Recommended practice- Hydraulic fluid power- Systems and products
- Testing, general measurement, principles and tolerances [18 pp]
Accumulators
T3.4.7M-1975 (RI980), Method for establishing and verifying the fatigue and static
pressure ratings, and conducting production tests of the pressure containing envelope
of a metal fluid power accumulator [26 pp]
Cylinders
T3.6.4M-1968 (RI992), Bore and rod size combinations and rod end configurations for
catalogued square head industrial fluid power cylinders [16 pp]
T3.6.5M-1968 (R1982), Static pressure rating methods of square head fluid power
cylinders - Part 1: Pressure containing components [8 pp]
T3.6. 7-1969 (Rl 988), Fluid power systems and products - Square head indented cylinders
- Mounting dimensions [49 pp]
T3.6.8M- I984 (R 1992), Fluid power systems - Cylinders - Dimensions for accessories
for catalogued square head industrial types [6 pp]
T3.6.11M-1972 (R 1988), Bore and rod size combinations, rod end configurations,
dimensional identification code, and mounting dimensions for 3/4 and 11/8 inch bore
catalogued square head tie rod type indented fluid power cylinders [34 pp]
T3.6.29M-1976 (RI981), Method for establishing and verifying the fatigue and static
pressure ratings of the metal tie rod or bolted fluid power cylinder [47 pp]
T3.6.3 lM-1976 (Rl 98 1), Telescopic cylinders and cylinders of non-bolted end construc-
tion - Pressure rating supplement [14 pp]
T3.6.36-1978 (R I984), Fluid power cylinders - Bibliography [6 pp]
T3.6.37- I 991, Hydraulic fluid power- Cylinders - Method for determining the buckling
load [2 pp]
T3.6.54M- I 986, Hydraulic fluid power - Cylinder ports - SAE straight through O-ring
and four-bolt flange ports - Heavy duty and light duty cylinders [6 pp]
T3.6.59-1993, Hydraulic fluid power - Cylinders - cushion performance [4 pp]
T3.19.12M- I981 (RI988), Method of testing, measuring and reporting test results for
reciprocating dynamic hydraulic fluid power sealing devices [12 pp]

T2.9.11M- I989 (RI994), Hydraulic fluid power - Method for detecting the particulate
count of an oil sample from a system (using liquid automatic counters) [10 pp]
T2.9.12M-1984 (RI994), Hydraulic fluid power - In-line liquid automatic particle
counting systems - Method of validation [9 pp]
T3. I 0.3-1967 (RI980), Glossary of terms for hydraulic fluid power filters and separators
[12 pp]
STANDARDS 569
T3.1O.4M-1991, Hydraulic fluid power - Filters and separators - Graphic symbols

(supplement) [4 pp]
T3.1O.5.1M-1976 (RI989), Hydraulic filter/separator housing pressure rating
(supplement) [12 pp]
T3.1 0.8.3-1990, Hydraulic fluid power - Method for determining the quantity of built-in
contaminant [7 pp]
T3.10.8.12M-1976 (RI986), Method of determining the pore size of a cleanable surface
type hydraulic fluid power element [10 pp]
T3.1O.8.18M-1977 (RI994), Hydraulic fluid power - Coarse filter element - Multi-pass
method for evaluating filtration performance [18 pp]
T3.10.12-1983, Filtration and contamination - Bibliography [5 pp]
T3.1O.14M-1992, Hydraulic fluid power - Determination of differential pressure/flow
T3.1O.18-1990, Hydraulic fluid power - Filter artwork universal symbols [8 pp]
Fluids
T2.13.2 (R3-1989), Hydraulic fluid power - Fire-resistant fluids information report on
company trade names [7 pp]
T2.13.3-1994, Index of non-proprietary hydraulic fluid specifications and selected
recommended practices [8 pp]
T2.13.4M-1994, Recommendations for conservation, maintenance and disposal of hy-
draulic fluids [8 pp]
T2.13 .5M-1991, Hydraulic fluid power - Industrial systems - Practice for the use of high
water content fluids [7 pp]

T3.8.11-1977, Tube fittings and conductors - Bibliography [11 pp]
T3.20.1-1989, Fluid power systems - Quick action couplings - Glossary [2 pp]
T3.20.7-1983, Quick action couplings - Bibliography [2 pp]
T3.20.8M-1975 (RI988), Quick action coupling pressure rating (supplement) [13 pp]
T3.20.15-1990, Hydraulic fluid power - Quick action flush face type couplings [5 pp]
T3.26.1-1977, Fluid power hose, hose fittings and hose assemblies - Bibliography [12 pp]
Pumps and motors

T3.9.13-1982, Hydraulic fluid power - Glossary - Pumps and motors [6 pp]
T3.9.17M-1990, Hydraulic fluid power - Positive displacement pumps - Method of
testing and presenting basic performance data [15 pp]
T3.9 .18M -1978 (RI994), Method of establishing the flow degradation of fixed displace-
ment hydraulic fluid power pumps when exposed to particulate contaminant [17 pp]
T3.9.20M-1992, Hydraulic fluid power - Pumps - Method of testing and presenting basic
performance data for pressure-compensated pumps [14 pp]
T3.9.21-l978 (RI985), Hydraulic pumps and motors - Bibliography [6 pp]
T3.9.22-1982 (RI988), Pump/motor pressure rating (supplement) [4 pp]

T3.9.25M-1977 (R1994), Hydraulic fluid power - Motors - Methods of establishing
speed degradation when exposed to particulate contaminant [8 pp]
Reservoirs
T3.I6.2M-1969 (RI987), Non-integral industrial fluid power hydraulic reservoirs [15 pp]
T3.I6.3M-I973 (RI987), Requirements for non-integral industrial fluid power hydraulic
power units [11 pp]
T3.I6.9 (R1982), Reservoirs and power units - Bibliography [7 pp]
Seals
T3.I9 .11M-1972 (R 1994), Groove dimensions for floating type metallic and non-metallic
fluid power piston rings [16 pp]
T3.I9.l2M-I98I (RI988), Method of testing, measuring and reporting test results for
reciprocating dynamic hydraulic fluid power sealing devices [12 pp]
T3.l9.I8M-1973 (R1994), Groove dimensions for fluid power radial compression type
piston rings [11 pp]
T3.I9.22-I982, Sealing devices - Bibliography [7 pp]
Valves
T3.5.IM-I984, Hydraulic fluid power - Valves - Mounting interfaces [41 pp]
T3.5.2M-1968 (RI988), Symbols for marking electrical leads and ports on fluid power
valves [9 pp]
T3.5.I4M-I982 (RI989), Hydraulic fluid power - Directional control valves - Methods
for determining the metering characteristics [8 pp]
T3.5.15M-I986, Hydraulic fluid power - Valves - Method for determining the internal
leakage characteristics [7 pp]
T3.5 .I6M-1991 , Hydraulic fluid power - Flow control val ves - Method for measuring and
reporting regulating characteristics [11 pp]
T3.5.26M-I977 (RI990), Hydraulic valve pressure rating (supplement) [20 pp]
T3.5.27 -1976 (RI982), Hydraulic valves - Bibliography [12 pp]
T3.5.29M-I980 (RI988), Hydraulic fluid power - Solenoid-piloted industrial valves-
Interface dimensions for electrical connectors [4 pp]
T3.5.33M-I985, Hydraulic fluid power - Cylinder actuator mounted valve - Standard
dimensions for mounting surfaces r8 pp 1
STANDARDS 57]
Institute of Petroleum standards

The following list of standard test methods are in the order ofthe fluid descriptions gi ven
in the chapter on Fluid Properties.
Viscosity (kinematic) IP 71
Grade IP 3448
Viscosity index IP 226
Density IP 59, IP 160, IP 365
Specific heat -
Compressibility -
Vapour pressure IP 409
Surface tension
Thermal expansion
Thermal conductivity
Aniline point IP 2
Pour point IP 15
Cloud point IP 219
Flash point IP 34
Fire point IP 36
Boiling point IP 123
Spontaneous ignition (auto ignition) ASTM-D 2155
Air release value IP 313
Water content IP 74
Water separability ASTM-D 1401
Neutralisation value (TAN) IP 1A, IP 139
Electrical resistivity (phosphate ester) IP 384
Lubricity
Filterability
Foam resistance IP 146
Dernulsification IP 19
Wear Vane pump test IP 281
Corrosion Rust preventing IP 135 A and B
Copper corrosion IP 154
For water-based IP 329
Stability IP 331
Oxidation characteristics IP 157
Biocide
Biodegradability CEC -L-33-A-93
Compatibility with seals IP 278
Load carrying capacity IP 334
The British Fluid Power Association guidelines and data sheets

The list below covers the majority of the guidelines in numerical order, with the latest
revision date. In order to locate the appropriate documents the list is preceded by a
summary contents table with P numbers only.
BFP A documents
Subject Relevant P numbers
General P3, P16, P53, P56
Accumulators P54
Control (and P41, P49
Electrohydraulics)
Cylinders P25
Distribution D2, D8, DIO [BFPDA]
Filters (and Contamination P5, P9, P48, P55
Control)
Fluids (see also Leakage) P12, P13, P57, P65, P66, P67
Flushing P9
Hoses (and Tubes, P7, P47
Connections and
Couplings
Leakage P59, P60
Legislation (and Standards P17, P29, P46, P58, P61, P68, P70,
and Contracts) D 10
Noise P4
Seals (and Plugs) P22, P52
Testing P28
Valves P27, P41, P49
BFPAlP3:1995 Guidelines for the safe application of hydraulics and pneumatic fluid
power equipment.
BFP AlP4: I986 Guidelines for the design of quieter hydraulic fluid power systems (Ed. 3).
[under review]
BFPAlP5:1996 Guidelines to contamination control in hydraulic fluid power systems.
BFP AlP7: I991 Guidelines to the selection and application of tube couplings for us in fluid
power systems.
BFP AlP9: I992 Guidelines for the flushing of hydraulic systems.
BFP AlP12: 1995 Hydraulic fluids - mineral oil data sheets.
BFPAlPI3: 1996 Fire-resistant hydraulic fluids - data sheets.
BFPAlP16:1994 Fluid power symbols wall chart.
BFP AlP I 7: I978 Code of practice for ordering scheduling contracts.
BFP AlP22: I 990 Industrial O-ring standards - metric v inch.
BFPAlP25:1981 Hydraulic cylinders - contract price adjustment clause and formula.
STANDARDS 573
BFPAlP27: 1993 Guidelines on understanding the electrical characteristics of solenoids

for fluid power control valves and their application in potentially explosive atmos-
pheres.
BFPAlP28: 1994 Guidelines for errors and accuracy of measurements in the testing of
hydraulic and pneumatic fluid power components.
BFP AlP29: 1987 General conditions for the preparation of terms and conditions of sale of
UK fluid power equipment manufacturers and suppliers.
BFP AlP41: 1995 Guidelines to hydraulic fluid power control components.
BFP AlP46: 1989 Product liability in Europe - a practical guide for industry.
BFPAlP47: 1994 Guidelines to the use of hydraulic fluid power hose and hose assemblies.
BFPAlP48: 1988 Guidelines to the cleanliness of hydraulic fluid power components.
BFP AlP49: 1995 Guidelines to electrohydraulic control systems.
BFPAlP52: 1992 Guidelines to the plugging of hydraulic manifolds and components.
BFPAlP53: 1996 Fluid power - at the forefront.
BFP AlP54: 1992 Guidelines to the pressure systems and transportable gas container
regulations 1989 and their application to gas loaded accumulators.
BFP AlP55: 1993 Guidelines for the comparison of particle counters and counting systems
for the assessment of solid particles in liquid.
BFP AlP56: 1995 Fluid power - engineers data book.
BFPAlP57:1992 Guidelines to the use of ecologically acceptable hydraulic fluids in
hydraulic fluid power systems. [under review]
BFPAlP58: 1994 The making of fluid power standards.
BFP AlP59: 1993 Proceedings of the 1993 BFP A leak-free hydraulics seminar.
BFPAlP60: 1994 Leak-free future for hydraulic connections.
BFP AlP61 :1994 A guide to the use of the CE mark.
BFPAlP65: 1995 VDMA 24 568 and 24 569 - rapidly biologically degradable hydraulic
fluids - minimum technical requirements and conversion from fluids based on mineral
oils.
BFPAlP66: 1995 BFPA survey on ecologically acceptable hydraulic fluids.
BFPAlP67: 1995 Ecologically acceptable hydraulic fluids - data sheets.
BFPAlP68: 1995 Machinery directive - manufacturers declarations.
BFP AlP70: 1995 The EMC directive and its effect on fluid power applications
BFPDA/D2: 1994 Technical guidelines for distribution of hydraulic fluid power equip-
ment.
BFPDA/D8: 1994 Quality control procedures and requirements for BFPDA distributors
for the manufacture of flexible hose assemblies for hydraulic fluid power.
BFPDA/DlO:1995 Hydraulic systems and power units - a user's manual incorporating
manufacturer's test and conformance certificate.
(Note: BFPDA is the British Fluid Power Distributors Association.)
575
TRAINING
THE SUBJECT of hydraulic systems is often considered to be so simple and basic that no
form of education or training is necessary. An advantage of hydraulic systems is that they
can quite simply be put together and made to work. However, this apparent simplicity is
also one of the biggest dangers with hydraulics and can frequently lead to poor perform-
ance and reliability. The knock-on effect after a bad experience may be to discount
hydraulic systems in future applications when in fact they may still provide the best
solution. Good education and training, giving adequate understanding of hydraulic
systems, is an obvious solution.
This should cover the complete range of engineering knowledge and skill levels.
Getting the best from the hydraulics involves the initial design, the correct assembly and
construction procedures, correct installation and commissioning as well as correct
maintenance schedules and their implementation. Understanding what a hydraulic system
should do and what it cannot do will help in many of these stages. A good understanding
of the implications of poor contaminant control and its effect on life and reliability are also
vital.
In recognition of these requirements and in view of the poor standards then attained the
British Fluid Power Association set-up an Education and Training Committee. This was
charged to work with the manufacturing and user industries as well as educational
establishments to encourage better standards. The development of a range of City and
Guilds modules, and the proposals for Certificate recognition of some courses are
outcomes of the work of this committee. It has also collected data on the educational
establishments actively involved in teaching aspects of hydraulics and pneumatics.
This chapter will review the types of courses available, and the qualification possibili-
ties in the broader fluid power field, including both pneumatics and hydraulics.
Short courses
There are many short course providers operating in this area. These include a number of
hydraulics manufacturers as well as colleges and universities. The term short here
generally means less than one week, and inevitably such courses must be free standing
with their own independent objectives. Some of them are simply that, whereas others may
form part of a coherent set leading to a complete training package. All short courses tend
to be intensive and so most provide a variety of activities including classroom and practical
laboratory exercises. Individual consultants, or fluid power establishments, may also
provide in-house training tailored to meet specific requirements for that company.
The range of topics may apparently have considerable overlap between the different
providers but there may well be a different emphasis on content depending on the provider.
Some establishment are more concerned with the design and mathematical aspects, and
some emphasise the implementation and practical skills. There is a strong need for both
of these if the hydraulics industry is to flourish. Depending on the level, some of these
courses now have approval to be included as part of the Chartered Institutions Continuing
Professional Development (CPD) schemes.
In addition to this broad range of courses there are specific training courses, given by
either OEM or component manufacturers, relevant to the use of their products, including
software. There are also an increasing number of courses in which both a manufacturer and
an educational establishment collaborate in organising and presenting the material. This
format provides a good balance in the content and its presentation.
Qualifications
There are relatively few qualifications specific to fluid power, these cover a wide range
but leave rather a large gap. The BFP A Education and Training Committee also surveyed
the content of a wide range of courses, including university degrees, to establish the fluid
power content but with rather disappointing results even in so called mechatronic courses.
However, there are fluid power components in other courses and again at a wide range of
levels.
The City and Guilds 2340 'Fluid Power Engineering Competences' scheme now
comprises a series of three competence based courses specifically for the fluid power
industry. The elements are:
Part I 'Pipe fitting & hose assembly systems'
Part II 'Pneumatics and hydraulics'
Part III 'Electrical/electronic control of fluid power systems'
The last ofthese was introduced in 1996, and in that year there were 36 centres approved
to run various parts ofthis scheme. Details of these and the scheme may be obtained either
through the City and Guilds or BFPA.
The other qualification area is at MSc/Diploma level for postgraduates and is a one year
full time course.
SECTION 8
577
MECHANICAL HANDLING
TillS CHAPTER serves as an introduction to most of the following chapters in this
Section. It indicates the particular versatility of an hydraulic fluid power system (as against
electrical or mechanical or pneumatic) in the process of moving a product. Such a subject
relates to robots, to cranes and fork lift trucks, to personnel lifts and articulated booms, to
vehicles and tractors, to bulldozers, loaders and excavators, and other mobile applications;
most of which will be covered in detail later.
Hydraulic pumps
The high conversion efficiency of hydraulic pumps - and in particular the develop-
ment of gear pumps with high volumetric efficiencies - has resulted in the extensive
application of hydraulics in mechanical handling. The most widely used source of
hydraulic power is the gear pump embodying pressure-balanced bushes or side plates to
minimise clearance gaps without the necessity for high side-plate forces. At the same time
lubricating systems have been developed for these pumps circulating low-pressure oil
through the bearings with negligible loss of volumetric efficiency.
In the interests of power saving and the elimination of hydraulic oil coolers some of the
larger diesel-powered mechanical handling vehicles employ multiple-pump systems,
often of the split-circuit type. Commonly a tandem gear pump comprising a large and a
small element has its combined capacity matched to the total power available for hydraulic
services at full system pressure, whilst the delivery from the individual pump sections is
sized to suit the services being supplied.
A further refinement, though more frequently used in earth-moving equipment, is the
dual pump off-loading system. In this case the combined capacity of the pump is matched
to the engine power available but at a pressure lower than maximum system pressure - say
14 MPa (140 bar) in a 20 MPa (200 bar) maximum-pressure system. Then, the capacity
of one of the individual elements is matched to absorb the same total power but at
maximum pressure. An unloading valve dumps one of the pump deliveries when the
pressure level in the system reaches the predetermined setting, thus giving high-speed
operation of the hydraulic services when at moderate load and a lower speed at full load.
The advantages of multiple fixed-delivery pump systems are heightened by the pressure
capabilities of some of today's more advanced gear pumps. Continuous operating
pressures of 25 MPa (250 bar) with a 20% allowance for pressure transients are now
available in pumps with individual displacements up to 92 cm3 trev (5.6 in3 trev), whilst
in the smaller capacities the continuous operating pressure can be as high as 27 MPa (275
bar ). Figure 8.1.1 shows a typical operating duty envelope for a high performance pump
of this type. In tandem or multiple pump arrangements the combined sections can operate
at main system pressures of around 20.7 MPa (207 bar), whilst individual sections can
operate up to the higher pressure levels where necessary to provide, for example, high
clamping loads with small, economical actuators.
Whilst the future of the efficient, low cost, fixed-displacement pump is assured, even
the versatility offered by multiple fixed units cannot match that of the variable-delivery
pump. Its fundamental advantage is the reduction of power wastage by its ability to reduce
delivery flow in accordance with the system requirement. This facility is achieved by a
swash control servo which can be arranged to respond to various circuit parameters.
Pressure-compensating control changes pump swash to maintain a predetermined con-
stant pressure level in the system. Constant-flow controls limit pump stroke to give a pre-
set constant delivery irrespective of pump speed change (ideal for large vehicle steering
duties). Power-compensating controls give constant multiples of flow and pressure along
a constant horsepower line drawn between the limits of maximum system pressure and the
pressure at maximum pump stroke. Manual control of swash gives an ideal means of speed
control for, say, hydraulic motors without unnecessary flow through a relief valve.
Hydraulic motors
Electric motors are sometimes cited as the most usable type of system motor for
mechanical handling, as the whole concept of electrical activity is rotary; the basic
behaviour of hydraulics is that of a linear motion with expansion or flow. Pneumatics, too,
are contenders for certain applications. However, there are distinct advantages in hydrau-
lic motors to that of other media types.
MECHANICAL HANDLING 579
Stall occurs without damage

Ideal for constant speed and very frequent start, stop and reverse actions
High power/torque density
Can be held at full torque without energy loss
Static hydraulic systems

Each static hydraulic system is usually designed and built for one specific and restricted
purpose. A manufacturer of such systems, in order to provide the required scope, must
have available a very comprehensive range of valves. This range will include a variety of
pressure-control valves, that is, relief, reducing, off-loading, balancing, dividing and
check valves. Most of these may be controlled by pilot pressure. Directional valves have
a great variety of spool configurations for obtaining the characteristics required. Control
may be by mechanical means, pilot pressure or solenoid.
In some cases a static system may be based on an accumulator. Thus, when the
maximum pressure is reached the pump is unloaded or the motor stopped, then, when any
circuit demand causes a sufficient decrease in pressure, the pump will automatically
reconstitute the full pressure. The fluid pressure may be converted back into mechanical
energy by using cylinders, semi-rotary actuators or rotary hydraulic motors. All these may
be used in static mechanical-handling installations.
Static mechanical handling systems are finding an increasing number of applications
in industry, civil engineering, commerce, etc. Examples include:
Scissor lifts
These come in a variety of types and sizes. The advantage is that a load may be lifted from
FIGURE 8.1.2 - An hydraulic scissor lift platform driven by electric motor [Simon Access Ltd]
a very low minimum height to several times that height. Sometimes automatic control is
used so that sheet material may be loaded or off-loaded a sheet at a time, and the top of the
stack will remain at a constant height. In the simplest form a feeler or finger is used to sense
the top of the stack, and this is connected to a val ve which controls the flow of oil into or
out of the supporting hydraulic cylinder. Such a device is ideally suited for feeding printing
machines or other paper-converting machines. BS 5323 and the new European scissor lift
standard PrEN 280 cover the safety of this device. An example of a scissor lift is shown
in Figure 8.1.2.
Passenger or goods lifts

Hydraulic power has many advantages, such as low initial cost when used only as far as
the first or second floors. In such applications the single-acting displacement ram is,
accommodated in a borehole directly under the centre of the floor of the lift cage. For
greater heights a hydraulic ram is mounted in the side of the lift shaft and operates a jigger
mechanism which, with a cable, doubles the movement. This hoisting system has the
added advantage that there is no mechanism at the top of the lift shaft. The motor, pump
and control equipment are usually housed in the basement, where any generated noise is
of least nuisance.
Bridge operation
Whilst most slow speed swing bridges can operate successfully with little complication,
there are applications where considerable care needs to be exercised. For instance, wind
forces may be significant; inertia effects may cause problems if a faster swing rate is
required, and overall stability could be at risk.
If a fast system movement is specified it is helpful to have controls as close as possible
to the cylinders; this reduces pipe lengths and hence raises the natural frequency of the
system (with the high bridge inertias).
High inertias may cause overrunning and valve control has to be matched very carefully
to the inertias and possible wind loads. Mannesmann Rexroth have successfully dealt with
the problem in one application by using an electronically controlled load-holding valve
which could control the back pressure by means of measuring the pressure electronically
and closing the loop around a proportional load-control valve. This allowed an adaptive
counter-balance valve to be designed and enabled stability parameters to be incorporated
into the electronics.
In highly complex bridge controls, as described above, it is now of advantage to use
PLC's to select the speed at which the bridge should be operated. This controls the relevant
meter-in proportional valve, monitors the cylinder pressure and takes the decision
regarding the setting of the back pressure valves. The PLC also provides the capacity to
monitor and control other features such as flow demand from the pumps, main valves and
general protection of the circuit; a link can also be made to a data logger to provide a record
of instrumentation output for each bridge swing.
Warehousing
Storage systems are in use from which any quantity of many hundreds of items may be
extracted automatically using hydraulic lifts. In this way the central store of a chain of
supermarket stores can be operated by a relatively small staff using a computer. Indi vidual
orders are computer processed in such a way that the automatic mechanism can produce
exactly the goods required.
Mobile hydraulic systems

Electric Fork Lift Trucks
Hydraulic power for most battery electric fork lift trucks is supplied by gear pumps directly
driven by compound wound motors which offer a light-load speed 50% to 60% greater
than when fully loaded. Maximum light-load speeds reach 4000 - 5000 rev/min, and the
most appropriate type of pump capable of efficient operation at speeds of this order is the
gear type. Motor speeds can, of course, be reduced, but this will add considerably to the
cost, size and weight of the truck at the design stage.
Most lift trucks feature a simple hydraulic circuit which consists of a reservoir, constant-
displacement pump and open-centre control valve. Standard trucks are fitted with a single-
acting spool valve controlling the hoist cylinder and a double-acting spool valve control-
ling the tilt function. Valves with up to six services are used on trucks fitted with
attachments such as a fork-sideshift device, clamps, or a rotating head.
An additional double-acting spool may be necessary for extending and retracting the
mast where the truck is designed to be used in gangways of varying width.
In order to effect a high degree of accuracy in load placing, the control valve in all lift
trucks must have first-class metering characteristics. It is also imperative that both sides
ofthe tilt cylinder are fully charged with oil at all times, since any voids which are allowed
to form will result in loss of control and consequent instability of the load or the truck itself.
Exhaust charging valves or pilot-operated check valves are normally fitted to prevent such
an occurrence.
The forks of a lift truck are usually lowered by gravity. The pressure generated by the
weight of unloaded forks is not high and the control valve and pipework must have a low
pressure loss if an acceptable lowering speed is to be achieved. Lowering forks which are
fully loaded requires great care in the use of the control valve, as too high a lowering speed
could be dangerous. A burst pipe or hose between the lift cylinder and the control valve
would give rise to the same problem. To remedy this a one-way flow-regulating valve can
be fitted directly to the lift cylinder in order to maintain approximately constant exhaust
flow at all pressures between that equivalent to light-loaded forks and fully loaded forks.
Gear pumps operate with high overall efficiencies but fixed displacement. Any
metering of the pump supply, such as occurs when stopping the load, tilting or side-
shifting, involves considerable power wastage. Variable-displacement piston pumps offer
the possibility of improved average operating efficiency. Service speed would be a
function of pump output and hence control lever travel, so that where necessary operations
would be carried out with a low pump flow and consequently absorb a minimum of power.
A further demand made by the battery electric truck application on pump design arises
from one ofthe truck's own main attributes. Its cleanliness, quietness and freedom from
fumes and quietness make it ideal for indoor operation but in cold-rooms, warehouses,
stores, etc., the noise made by the truck's hydraulic pumps can be the greatest source of
annoyance in the building. However, pump designers have made significant advances in
minimising both mechanical and hydraulically-generated noise by careful attention to
manufacturing standards on gears and the detail design of fluid relief and associated
porting in the pump. Equally, vehicle designers can minimise noise amplification by
careful attention to pump mounting and the location and attachment of pipework.
Battery electric trucks are limited in capacity by the present economics of battery
design, and find their greatest use in factories and warehouses where their silence and
absence of noxious exhaust gases are obvious advantages. They are not suitable for duties
involving long travelling distances, because oftheir relatively low speeds and the problem
of battery capacity.
Engine-powered lift trucks

Larger fork lift trucks are powered by diesel engines (or sometimes by petrol engines)
when the hydraulic supply can be provided by an engine-driven pump.
Increased lift speeds are achieved as a result of the more powerful pumps having higher
flows. Because adequate engine power is necessary for traction purposes and is not used
when the hydraulic system is operational, hydraulic power consumption does not usually
become a major factor in trucks below 5 tonnes (5 tons) capacity. For trucks of larger
capacity it is necessary to reduce hydraulic power consumption at high loads. A variable-
displacement pressure-compensated piston pump would achieve this.
Current systems use a two-speed lift approach which can be achieved by a double-acting
lift cylinder in which the annulus and head sides can be inter-connected by a valve. Low
speed is achieved by supplying only the head side of the cylinder and thus exhausting the
annulus side, whilst high speed is attained by inter-connecting the head and annulus sides
of the cylinder and therefore pressurising the annulus and adding its exhaust to the supply
at the head side. The lifting capacity is, of course, reduced in the proportion of the rod area
to the piston area and the lifting speed increased inversely. The ratios of high to low speed
are obtained at the design stage by the choice of the appropriate rod size. In some designs,
the valve inter-connecting the annulus and head sides of the cylinder is automatically
pressure-operated. Normally the valve is biased to the high-speed position, the maximum
load which can be lifted being a function of the system pressure and rod area. Loads in
excess of this result in a higher system pressure which shifts the valve, blocking the inter-
connection and diverting the annulus side of the cylinder to tank. Loads equivalent to
system pressure and the piston area can thus be lifted.
The circuit diagram of an alternative two-speed system is shown in Figure 8.1.3. Two
lift circuit pumps are used, one of which is unloaded automatically and dumped to tank
when the system pressure reaches a value equivalent to the maximum allowed input power
to the hydraulic pumps. The remaining pump can thus operate up to a higher pressure
before input power again reaches its maximum allowed level. This system has the
advantage that a normal single-acting lift cylinder can be used. In conjunction with a
bypass flow regulator, a third and smaller pump is fitted, the controlled outlet of which
supplies oil for power-assisted steering purposes. The bypass outlet supplies a separate
valve for controlling tilt and other auxiliary services where full system flow would be an
embarrassment and would require a permanent throttling device in a single-pump system.
A multi-pump system allows freer selection of pump capacities for the differing needs of
the services.
Power-assisted steering is necessary for all but the smallest lift trucks to ensure rapid
low-effort manoeuvrability in confined areas. Hydraulic power for this purpose is
sometimes provided by a separate engine-driven pump incorporating its own small
reservoir, flow regulator and system relief valve. Some manufacturers utilise the main
hydraulic system.
Articulated booms
Both electric and diesel power is used to drive the hydraulics in these personnel or material
carriers. Figure 8.1.4 shows the fold up design of a single boom arrangement which might
have a working height of up to 12 m with an outreach of over 6 m for 225 kg.
Side-loading trucks
Side-loading lift trucks are designed to carry long loads and complete containers on larger
size vehicles. The mast is mounted on one side of the chassis and moves across to pick up
or deposit the load outside the wheel-base of the vehicle. The load is moved in and lowered
on to the deck for transportation. To give stability during lifting and lowering, hydrauli-
cally-operated stabiliser jacks are used. The problems confronting the hydraulics designer
are very similar to those posed by front-loading lift trucks. Mast extension and retraction
is a standard requirement. Stabilisers can be powered by an auxiliary circuit. Power
steering is standard and the necessary pressure oil supply can be made available from
either a separate pump or bled from one of the main circuits by a flow-regulating valve as
described. A typical hydraulic circuit for a side-loading truck is shown in Figure 8.1.5.
Straddle-carriers
The straddle-type lifting vehicle grew out of the need to move standardised containers on
and off trucks and flat cars for marshalling purposes. To ensure that the rows of containers
can be marshalled closely together it is necessary to keep the supporting legs of the vehicle
as narrow as possible, which leads to the use of traction chain drive from the upper deck
to the wheels. The deck itself houses the driver's cab, engine, transmission and all
hydraulic power units.
Straddle-carriers are now more or less standard equipment at marinas for lifting and
launching craft up to about 10 tonnes (10 tons) dead-weight, although here the design is
usually less sophisticated with mechanical lift and hydraulics involved only for power-
assisted steering. Engine and cab (or steering position) are side-mounted. Straddle-
carriers of this type are generally known as travel-hoists.
Skip loaders
Skip loaders are designed for round 12 tonne lift capacity, and although there is a
considerable mechanical content in the arrangement, it is hydraulics which provides the
power for the operational functions.
In one example the vehicle engine power take off (PTO), through a gear box, provides
the pump drive with a flow of up to 75 Llmin at 21 MPa (210 bar). Lever mounted
monoblock valves control the legs and lift cylinders as shown in Figure 8.1.6. Over-centre
check valves ensure that a loaded container continues at controlled speed when the lift
arms move over centre when loading or off-loading containers. A regenerative valve in the
cylinder circuit enables the operator to quickly off-load empty skips.
FIGURE 8.1.6 - Skip loader hydraulic circuit with over-centre check valve and regenerative
valve [RJE Telehoist]
Mobile cranes
Most mobile cranes in the 15 tonnes (15 ton) class are hydraulically actuated and
considerable inroads are currently being made into the 30 tonnes and 60 tonnes (30 and
60 ton) and even larger classes. Hydraulics can provide infinitely variable speed actuation
at least as effectively as any other form of power transmission. It affords an extremely high
power to weight ratio and is now more competitively priced in relation to the alternatives.
Cranes demand greater control than any other materials-handling vehicle. Safety, as in
other handling equipment, is of paramount importance. Lift trucks, side-loaders and
container handlers use linear power output actuators (cylinders or rams) for most
functions. In addition, they require rotary actuators or hydraulic motors for some
functions, such as winch drive and slewing operation. Telescopic mobile cranes normally
have derrick cylinders, a telescopic cylinder to extend and retract the jib and hydraulic
motors for the hoist winch and slewing. Lattice jib or strut cranes will normally have two
or three hydraulically-driven winches controlling the jib and hook through wire rope
connections. Both types require stabiliser jacks or outriggers to confer the stability
necessary for working at extended radii round the machine.
Many hydraulic cranes are powered by constant-displacement gear pumps. Winches are
driven by gear motors through reduction gears or, in some larger sizes, directly by slow-
speed radial piston motors. Starting, stopping and speed regulation through the entire
speed range is handled by manual directional-control valves. Fine metering control is
essential. Starting a suspended load requires that the hydraulic motor develops a high
stalled torque efficiency (low torque or a 'running' start is not possible). Although gear
motors have excellent low speed torque efficiencies, there is nevertheless a considerable
variation as each gear tooth passes through its engagement cycle. Many times there is
sufficient 'slack' in the drive for the variation to be unnoticed - backlash in splines.
reduction gears, flexible couplings, etc. However, in a crane with a suspended load all
'slack' is taken up as the load must be supported by the motor immediately the brake is
released. A dual motor in which the tooth engagement in each section is phased so that one
section is at its most favourable torque position while the other is at its least favourable.
offers a considerable improvement, giving consistent low speed torque efficiencies in
excess of 90% of theoretical maximum torque output. A dual motor can also be arranged
for series-parallel operation, offering the possibility of a two-speed drive which can be
invaluable for crane winch operation where high, light-load hook speeds are desirable.
Piston motors also exhibit starting torque variations, depending on the number of pistons
used, for which allowance must be made when choosing the displacement of the motor for
the particular job.
Leakage across the internal elements of a fluid motor, whether of the piston, gear or vane
type, demands the installation of a friction brake on crane winch applications. The use of
a high-speed motor with reduction gears aIlows a relatively small brake to be mounted on
the high-speed input shaft. The load is generally lowered against hydraulic pressure
(imposed by a valve) sufficient to support it. Consequently the brake is not used
dynamically. Many crane winch motors now incorporate a built-in disc brake which is
spring-loaded to the 'on' position and provided with an automatic hydraulicaIly-powered
release mechanism.
Hydraulic cranes which utilise constant-displacement pumps are equipped with two,
three or even four pumps. Multi-pump combinations permit the incorporation of stepped
speed ranges providing, in combination with control valve metering, infinitely variable
speeds at the output units without excessive power wastage. A typical mobile crane
hydraulic circuit such as would be used on a 15 tonnes (15 ton) machine is shown in Figure
8.1.7.
Mobile cranes with extremely long lattice jibs are commonly used during the construc-
tion of high rise buildings, bridges and other civil engineering work. Considerable skill is
needed to position loads which may be more than 30 m (100ft) above the operator's cab.
High lifting speeds are necessary to keep cycle time as low as possible. However, high
lowering speeds with heavy loads are undesirable and dangerous. Hydraulic control gives
the possibility of improved performance in all these respects.
Pressure-operated interlocks can be arranged to prevent high speeds automatically
when high loads are lowered. The possibility of providing feed-back servo-controls to
allow predetermined lifting or lowering speeds to be 'dialled in' by the operator is also
worth investigating. Close control of a load so far away as to be almost 'out of sight' could
be ensured and many breakages eliminated.
There is no doubt that the use of closed-circuit variable-displacement hydrostatic
transmissions for winch drives is now economically feasible. Some large capacity crane
winches already incorporate this type of drive in preference to the more rudimentary open-
circuit gear pump-motor combinations. An extension of the use of true variable-displace-
ment hydrostatic drives can be envisaged, particularly to large power outputs where the
gains in average operating efficiency could be considerable.
The problems of providing effective manual operation of the control valves on many
materials-handling machines can be difficult to solve. Cabs are small with limited
accommodation for direct lever-operated control valves, especially those of large capac-
ity. Hydraulic lines of sufficient capacity for the large flows which are now common are
difficult to run to a central location, particularly if a considerable number of services is
involved. There is considerable virtue and, indeed, necessity in the case of large units in
siting the control valves in positions affording maximum accessibility and minimum pipe
runs. By intelligent siting, the length of pipes and hoses and the numbers of connections
and bends can be considerably reduced, offering a reduction in cost not only in pipe and
fittings, but also in installation, which may be of even greater significance.
Mechanical links between the operating levers and valve spools are possible in some
installations but may require excessive ingenuity and cost in installations with turntables
or elevating cabs. One solution is to use remote hydraulic power controls where small
pressure-control servo-valves can be hydraulically inter-connected to the spools of the
main control valves and so arranged that the displacement of the main spool is a function
of the movement of the servo-valve lever. A linear relationship equivalent to that which
would be obtained by direct actuation of the main spool is possible and can be improved
on. The servo-valve can be designed so as to give a large increment of controlled outlet
pressure for a small spool travel and thereafter small increments of controlled pressure for
the remainder of the spool travel. The main valve spool can be centred by multiple spring
packs giving a discontinuity of spring rate, and therefore force increment, at various points
in its travel. The combination of these two features enables a large proportion of the servo-
spool travel to be utilised while the main spool is traversing only a relati vely short metering
band. This results in better metering characteristics than if the main spool were directly
operated. The servo-operating pressure is of the order of 1.5 - 2 MPa (15 - 20 bar) so that
small bore plastic hose with low-pressure fittings can be used for inter-connecting the
main and servo-valves. Supply lines to and from the servo-valve would be similar.
Instantaneous flow rates of up to about 14 Llmin from the servo-valve are all that is
required to ensure a maximum response time of 0.2 sec in the largest size of main control
valve. The servo-valve is arranged for console mounting and designed for a low level of
operating force consistent with providing adequate 'feel'. Hold detents, if required, are
provided on the servo-valve rather than on the main valve and are arranged for pneumatic,
hydraulic or electrical release as may be appropriate. In addition to providing remote
control, the system also ensures extremely low-effort operation of even the largest control
valve. Operator fatigue is markedly lower with a consequent improvement in machine
effectiveness and the main control valve is totally enclosed and does not require any
dynamic seals.
Drives
Nearly all engine-driven lift trucks, side-loaders and container handlers are mechanically
driven through a torque converter, 'hot shift' gearbox and conventional differential axle.
All are ideal applications for hydrostatic transmissions. The speed range over which
maximum power must be transmitted is not great - very high speeds are neither necessary
nor practicable on load carriers of this type - but the greatest precision of control (i.e. the
ability to start and stop smoothly and to position the vehicle with great accuracy) is
absolutely vital. Hydrostatic transmissions can meet the control requirements better than
any other form of transmission.
Three types of output drive are possible:
i) High-speed hydraulic motor or motors connected through reduction gears to
conventional axles with differential units;
ii) Low-speed wheel motors;

iii) High-speed wheel motors with integral planetary reduction gears.
In theory, it would be desirable to dispense with axles, which are expensive, and impose
certain constraints on the layout of the vehicle. Low-speed wheel motors or high-speed
wheel motors with reduction gears obviate the use of axles. However, the wheel-motor
units have to be attached to the vehicle frame and one pair at least must be steered.
Provision for oscillation about the longitudinal axis of the vehicle may be necessary to
confer stability on uneven ground. Possibly the greatest drawback to the use of wheel
motors is that the most suitably shaped radial piston type is difficult to design as a variable-
displacement unit, while the variable-displacement axial piston unit - either as a low-
speed direct drive unit or as a high-speed/planetary gearing combination - has poor
physical shape characteristics for wheel mounting. Stepped displacement characteristics
can be designed into radial piston units by allowing a proportion of the pistons to be
disengaged when low displacement is required. Alternatively, series/parallel flow to a
number of constant-displacement wheel motors can be arranged to give steps of effective
displacement and, therefore, several speed ranges. The necessary valving is complicated
however, and although technically possible does not commend itself commercially. For
any vehicles whose required speed/torque characteristics are such as to need a variable-
displacement motor, it would seem that a high-speed motor or motors driving through
conventional differential axles is the most practical solution
Power steering
Most mobile mechanical handling equipment in use today employs power steering,
generally of the hydrostatic type without mechanical linkage. The metering units are
normally of the geared rotor or vane type and are used in conjunction with power steering
pumps having integral flow-control valves so that the steering circuit is supplied with
virtually constant flow irrespective of engine speed. In the simple spill type val ve any flow
which exceeds the pre-set requirement of the steering circuit is dumped back to tank or to
pump inlet; circuit protection is usually provided by a pilot relief val ve which, on cracking,
signals the main metering spool to dump flow to tank. Such systems are, of course, power
wasting, but having the advantage of simplicity and low cost are nevertheless very widely
used.
The true priority-flow valve is similar in function except that the excess flow from the
priority service, usually a large proportion of the total pump output, can be supplied to
other hydraulic services. These services can be used at entirely different pressure levels
from that of the priority service. Typical pressure levels are 10 - 14 MPa (100 - 140 bar)
for the priority (steering) circuit and 17.5 - 20.7 MPa (175 - 207 bar) for the 'secondary'
services. It is normal practice for this type of valve to include an integral full-flow relief
valve, with its own return line, to protect the priority service; the alternative of dumping
flow into the secondary line can be undesirable when other services are in use.
A further facility offered by some flow-control valves is the provision of a variable
adjustment for priority-flow settings, useful on applications like augers where differences
in material density may require a range of settings rather than one fixed flow level.
Control developments
Problems of size and siting of directional-control valves have led to the development of
remote operation of the main valves by various means. Hydrostatic systems have been
developed and used on applications such as manned work platforms, but more commonly
low-power pilot valves at the operator's hand servo operates the main control spools. Pilot
operation can also give advantages of control; manual operating loads are very light giving
accurate 'feel' and the servo-actuators can be designed to give a non-linear relationship
to the main valve spools to improve inching control of services.
A further variety of remote operation is the recently introduced electro-hydraulic
controller. These small units, connected to the main control valve only by electric cable,
offer proportional control (i.e. the control valve spool position, hence the load speed,
changes in proportion to the movement of the remote hand control), giving a high degree
of controllability and positioning accuracy. The freedom given by the cable connection
enables the manual-control units to be incorporated in a portable box, which can be worn
on a neck strap; machines can thus be operated in hazardous areas whilst the operator is
in complete safety.
Load sensing valves change the delivery of the pump in relation to movement of the
control lever. In this system the pump servo is arranged to give a basic constant-flow
control function and the spool valve opening is the variable orifice used to set the pump
flow level. In maintaining a constant (low) pressure drop across the spool orifice the pump
changes its flow output in relation to the orifice size and this principle gives the following
important advantages:
Pump flow is independent of system pressure and pump speed (for a set spool
selection)
Absolutely minimum power loss
Spool deadband is minimised and spool forces are small (due to small metering
pressure drop)
Precise control and feel (with service load speed proportional to valve movement
up to maximum pump flow).
591
INDUSTRIAL ROBOTS
ROBOTS ARE found in many applications and with a considerable range in complexity.
Not surprisingly, the automotive sector is the primary user, for such things as welding,
painting, assembly, etc. where large quantities of high cost operations are involved.
However, other sectors are becoming more interested as control improves rapidly, and
safety with security is paramount.
The reason for this advance in control is the better ability to detect the features which
surround the robot so that it can take the required action. 'Pick up and place' operations
are still of great importance, but to be able to make alternative decisions, depending on the
object and destination, opens up a totally new field of opportunities. The list below shows
the robot range:
i) Manual manipulators - which are basically simple tools, handling devices, etc.
worked manually from a remote position.
ii) Pick-and-place robots - low-technology robots limited to a range of simple
movements.
iii) Fixed sequence robots - capable of performing movements in a pre-determined
sequence, and where the set information cannot readily be changed.
iv) Variable-sequence-robots - where the programmed performance can readily be
changed for another.
v) Play-back robots - which repeat a programme initially 'taught' to them by a human
operator.
vi) NC Robots - programmed by numerical control data, i.e. basically the same control
mode as a numerically controlled machine.
vii) High-technology robots - which in addition to being programmable have a built-
in capacity to detect changes in the work environment or work condition and correct
accordingly.
Pick-and-place, and a proportion of fixed sequence robots may have relatively straight-
forward movements which can be mechanically, pneumatically or hydraulically operated.
Other categories with computer-type control are invariably based on servo-systems for
movements. On an overall basis non-servo-types are decreasing in proportion to servo-
types; and whilst pick-and-place robots represent the greatest number of robots in use, a
FIGURE 8.2.1- The six degrees-of-freedom commonly employed in industrial robots
high proportion are now servo-types. Non-servo-types now account for less than 25% of
all industrial robots
The fundamental components of an industrial robot are the mechanical system, the
power-drive system, the sensor and/or servo-system, and the functional control system.
Four different kinds of mechanical systems have been developed, each with a readily
identifiable pattern of movement.
FIGURE 8.2.2 - A typical robotics application - spot welding Peugeot bodies [Kuka Welding
Systems & Robot Ltd]
INDUSTRIAL ROBOTS 593
Polar - an extendible arm is mounted on a central pivot after a revolute joint.

Cartesian or Rectilinear - the gripper moves along by three perpendicular tracks
to achieve the required height, width and depth.
Articulated-based on the human form, the mechanical arm can bend at an 'elbow'
and swivel at a 'shoulder'.
Gantry - here an extendible arm is mounted on a central slide which can go up or
down and swivel on its mounting.
As to which is the best geometry is dependent on the application. Whilst the articulated
type is the most flexible, it is the most difficult to control. Flexibility and dexterity both
increase with the number of joints and the quality of control. The most popular for high-
speed assembly is the horizontal gantry type - commonly called the SCARA type,
(Selective Compliance Assembly Robot Arm) - whilst the articulated type is more common
in the automotive industry.
One example of a welding robot is shown in Figure 8.2.2. In this example two six-axis
robots are placed on each side of a conveyor, with the second set angled at 15, giving an
even greater level of accessibility. The robots detect the type of car body being presented
and then complete some 174 spot welds in a 102 seconds (including the body transfer). In
this case the hydraulics is used for balance control.
Drives
Three types of drive system are in wide use:
i) Pneumatics
ii) DC or AC Electrics.
iii) Hydraulics.
This is an hydraulics book, but it is important to realise that pneumatics and electrics
may be more appropriate in certain situations.
Pneumatics gives a cheap and simple power system but does not allow easy control of
either speed or position. It is a convenient system because modular builds can be undertaken
quick and simply. It is used for driving and effector tooling and feeding mechanisms.
Electrics are convenient and the most precise. Speed is excellent; for example, a small
ac or dc servo-motor with sensing via an optical encoder will move a mass of 60 kg at
speeds up to 1.8 m/s with repeat accuracy to 20 !lm and stroke lengths from 150 mm to
over 2 m. Earlier there were load limitations because of the weight of the drive motors
mounted in the arms, but that is less now due to lighter weight high-power motors which
can actually form part of an arm joint. 150 kg is quite acceptable.
Hydraulics is highly reliable and there are a number of positive advantages of this
system over the others, the most important being the greater overall power and response
capability; high stiffness and freedom from backlash are also present provided there is not
too great a use of elastomeric hoses. Hydraulic components also provide such valuable
features as the cylinder for linear actuation and accumulator for energy storage and
meeting peak demands. An example of an 100% hydraulic robot is shown in Figure 8.2.3.
FIGURE 8.2.3 - The Unimate robot

[VA Technology Ltd]
Hydraulics are normally used for applications where power requirements exceed 10
kW, indeed thrusts of up to 15000 N can be achieved enabling force-fit assembly to take
place. The most accurate systems will use servo valves but these will be expensive.
Proportional solenoid valves are cheaper and they may be sufficiently accurate and
responsive, however, additional digital drive electronics may have to be purchased. The
highest accuracy will probably require a reasonably stable temperature of the fluid -
possibly between 43C to 48 C - and the fluid may have to be preheated depending on
the environmental conditions.
Finally, for situations where there is a flammable atmosphere hazard (i.e. in paint
spraying) intrinsically safe electro-hydraulic systems can be used to ensure that electrical
power levels are kept so low as to eliminate the chance of ignition. This offers a very
considerable weight and space saving over flameproof equipment.
Controls
The sensor and servo-control system can be either quite straightforward or as sophisticated
as required. Simple pick-and-place robots use on/off controls to drive against mechanical
stops or limit switches with springs and dampers to stabilise motion. A smoother action
would be obtained with a suitably controlled proportional valve or servo-system. The
power supplied to each axis can be modulated by a dc power amplifier or by an electro-
hydraulic control valve. This allows the arm to be driven in a continuous and pre-
determined manner using, for example, a signal recorded in the functional control unit or
from a control handle. Combining the sensor system and servo-system allows feedback
control in which the control signal to the servo-system becomes the displacement of the
gripper as sensed by the instrumentation; this allows increased position accuracies to be
attained. Continuous path control can be provided which is really an extension of the
position control just described. The path is broken down into a sufficient number of
INDUSTRIAL ROBOTS 595
intermediate positions with point-to-point control between each. This leads to a teach
mode emerging in which, as the robot is moved through a trajectory, a recording is made
of the sensory position feedback. This recording can then be played back, as quickly or
slowly as required, as a drive signal to achieve the desired path.
The functional control system can be as basic as a programmable plug board, consisting
of a matrix where the rows correspond to each axis and a stepping switch energising each
row in turn to operate the robot through its cycle. PLC's are clearly the most powerful and
attractive means of control. Robots can be programmed to carry out specific tasks, take
alternative actions dependent on results from sensory feedback and, through communica-
tion links, interact with other robots.
Effect of Robot Size

A large, powerful robot normally works in noisier and less pleasant surroundings than its
smaller counterpart. A smaller robot will have an easier environment and probably a softer
job. Also, because of its location, it will be treated with greater care. All of these factors
make it possible to design the small robot with a smaller safety factor than that required
for a larger robot. As a result, the cost advantage of a hydraulically driven robot diminishes
with size.
With installation, maintenance and other operational costs to be taken into account, the
most cost-effective small robot is usually electric driven. The exact crossover point
between hydraulic and electric drives may vary with robot configuration and the robot's
intended use.
A main limitation which may be noticeable with electric-driven robots is that drives are
under-powered, necessitating the use of high gear ratios to obtain the necessary dynamic
performance. As a consequence they can be fast acting for small movements, but prove
very slow for large transfer moves.
A particular advantage of hydraulics in this respect is that energy can easily be stored
in an accumulator and released when an extra burst of robot activity is required. Thus it
is not uncommon to find some hydraulic robots which momentarily require a delivery of
say 220 Llmin being supplied by a pump of only a quarter of this capacity operating part
time in loading an accumulator.
Other particular advantages are the simplicity and reliability of a hydraulic cylinder for
providing linear motions; and the efficiency of a hydraulic motor in providing rotary
motions. Part-rotary motions are also readily provided by suitable designs of semi-rotary
actuators (i.e. vane-type actuators).
In paint spraying and other applications the environment may present an explosion
hazard and the robot must either be explosion proof or intrinsically safe so as not to ignite
the combustible environment. Here, the hydraulically driven robot has a great advantage
over the electrical since the electric energy from feedback devices and the energy to drive
servo-valves can be small enough not to ignite the explosive fuel/air mixture.
Similar considerations apply to pneumatic driven robots, with the advantage of faster
movements. However electric drives are widely favoured for paint spraying robots
because of their suitability for generating fast, small movements.
Other uses
There are a number of developments where the human is being replaced by robots for
safety reasons. Although these may be electric for convenience, they may also have
hydraulic content. For instance, bomb disposal, underwater operations (to replace divers
for investigation or build), pipe laying. Other considerations are movement of toxic items
and waste disposal.
597
MACHINE TOOLS AND AUTOMATION
MACHINE TOOLS are the basis of most industrial production. They are designed to
forge, cut, bend, press, roll and generally form and finish material into a required shape.
This chapter looks at the overall hydraulic content of such machinery with some detail
descriptions, but injection moulding machines and presses are discussed in later chapters.
Particular advantages of hydraulics applied to machine tools are:
i) It can provide a complete system for operating all machine tool movements from
a single power source (e.g. a single electric motor driving a single pump or pump
group).
ii) The elimination of costly lead screws and the necessary anti-backlash equipment
always associated with mechanical drives.
iii) Extremely smooth movement under infinitely variable speed control.
iv) The availability of proportional control response via proportional control valves
and electric signalling.
v) Ready adaptation to automation through manual control, or sequential or combi-
nation control, via logic control circuits.
vi) Simple and easily replaceable power cylinders or hydraulic motors, and accurate
positioning using servo-control systems.
vii) Long working life at optimum efficiency by use oflow pressure drops at high flows.
Basic system
A basic hydraulic system for a machine tool consists of an (electric) motor driving a pump,
which circulates an hydraulic fluid from a reservoir to the various control valves, and on
to cylinder/actuators. An hydraulic power pack is a convenient source (see chapter on Ring
mains and power packs). At least one filter in the system is essential to maintain the
necessary standard of cleanliness in the hydraulic fluid for the required machine reliabil-
ity. Piping and hoses are the links between the components, but by use of carefully
designed manifolds (for the control valves) these may be kept to a minimum. Another
feature which may be necessary to give the maximum accuracy to the machine operation,
is that of maintaining the temperature within a few degrees of the optimum; this is done
by means of an oil cooler such as an air blast cooler.
Motion control
Motion control can be by cams, limit switches and stops; from numerical or tape controlled
systems, working through solenoid-operated or hydraulically piloted valves, or sequen-
tial/combinational logic circuit designs. (Programmable logic controllers - PLC's -
improve the sophistication of the operations considerably.) Most systems require consid-
erable machine flexibility which can only be met by using CNC systems and position
controlled drives - proportional and servo valves are used (PVNC and SYNC) in
combination with both hydraulic cylinders and hydraulic motors.
A particular advantage of the hydraulic system is the range of speeds which are possible,
which would be very difficult to achieve with a standard ac motor without the complication
of complex electronics.
Actuators are simply cylinders with the piston mechanically connected to the moving
part, saddle, slide or table; in other cases, the cylinders are combined with the slide itself,
with the piston and rod remaining stationary. Normally, they are double acting, but in a
few instances it is possible to have a spring return, e.g. in work-holding clamps. For milling
or multi-drill heads, where torque requirements and speeds are high, hydraulic motors are
used to advantage.
The actuator rod diameter is simply determined from the designed maximum thrust to
be transmitted. The desired lowest slide feed rate (cm/min) is compared to the smallest
repeatable flow rate (mUmin) of the flow control valve to be used (typically around 30
mUmin) and an effective piston area calculated (mUmin divided by cm/min = cm2 ).
Knowing the effective area and the maximum thrust, the maximum oil pressure at the
actuator can be computed and, by taking control and other circuit element pressure drops
into consideration, the maximum working pressure at the pump is determined. Flow rates
are dependent on actuator size and speed of movement and the number of actuators
operating simultaneously, i.e. the machine sequence of operations. This information is
usually combined on a machine sequence diagram, from which the circuit elements
(pumps, accumulators, control valves, filters, pipe diameters, etc.) are selected.
The control elements, such as directional, pressure control and flow control valves, can
be mounted on individual subplates and piped together, or they can be mounted on a
module block with internal connections.
Connections
In the case of elements that are piped together, common problems are that the control
requires too much space, that the pressure drops through the resulting long pipe runs are
too great and that these are expensive and not always reliable. The result is a high price
to pay for the control. When the individual elements are mounted on module blocks and
these are then put together to form control systems, the expense of pipework is reduced.
However, if the pipe connections to the actuators and to the oil reservoir go out from
a particular side of the control, this can lead to awkward runs for the connecting pipework.
Moreover, any subsequent additions to the controls, for either of the aforementioned types,
can only be carried out with difficulty. It would be better in these cases, therefore, on
standard machine tools and standard machine-tool elements (e.g. unit heads), for design-
ers to consider proven module systems in the form of small compact blocks, produced in
large quantities and therefore favourably priced. These embody a number of piping
options to the actuators and oil reservoir and can be expanded easily and at any time for
specified functions.
An example of modular block controls for machine tool slides is shown in Figure 8.3 .1.
Here rapid motion forwards is obtained by energising solenoid 'a' of directional valve 4,
thus allowing pressurised oil entering the block at Port P (or P') to pass from P to B of valve
4, and over the check valve to the P port of valve 3, which has solenoid 'a' energised.
Pressurised oil thus passes from P to A and via port P (or A') in the block to the cylinder,
moving the piston from left to right. Displaced oil flows into the block through port B (or
B') to port B of directional valve 3 to tank via valve 4 and port T (or T') of the block.
During this time, oil has also been passing through the flow control valve and joining
the rapid transverse oil flow to the left of the check valve in the end plate, keeping the
pressure compensator 'live'. (Port A or A' signifies the port in the base or side of block,
respectively.) Thus a fine feed forwards is obtained by simply de-energising solenoid 'a'
of valve 4 and diverting displaced oil from the T port of directional valve 3 to pass over
the back-pressure valve 6.
In this example, a mechanical dead-stop is used to limit the length of cylinder stroke and
limit switches are used to trip from rapid motion to fine feed. Reverse feed is obtained by
de-energising solenoid 'a' of directional valve 3 as solenoid 'b' is energised and reverse
rapid motion by re-energising solenoid 'a' of directional valve 4.
Any combination of rapid motions and fine feeds can be obtained in either direction
depending upon the number of trips and limit switches that can be accommodated on the
slide of the machine tool.
Hydraulic workholding
Another versatile control module is the work-holding control shown in Figure 8.3.2. This
shows an hydraulic power pack, run from an electric motor, supplying hydraulic pressure
to control valves. The two valves on the manifold control the swing cylinder and the
threaded cylinder. A separate accumulator helps maintain pressure and provides smooth
system operation in the supply pressure line.
Depending on the type of work-holding, a sandwich pressure-reducing valve and
double-solenoid directional valve are fitted to the clamp or chuck cylinder. Displaced oil
is returned to tank via the directional valve and the T port of the block. Reverse operation
is effected by energising the second solenoid. The check valve gives the clamping system
MACHINE TOOLS AND AUTOMATION 601
a measure of protection against any drop in the supply pressure, especially when an
accumulator is fitted.
The versatility of mounting is similar to the feed-control blocks, having ports suitable
for either manifold or pipe connections on the underside and duplicate threaded ports on
the sides. Applications of these controls to machine tools are diverse, but include as
examples: fine boring with boring units, drilling machines, radial turning with boring
heads, turning and forming with single slides, and parting-off on lathes, and the operation
of checks and clamping systems of all types on various types of machine tools.
An example of the time savings using hydraulic clamping is given in Table 8.3.1 which
shows the times for clamping (with a 6 mm stroke) for both manual and hydraulically
operated via an electric driven pump unit.
Cylinders
Hydraulic cylinders for work holding can be split into three main groups (see also chapter
on Hydraulic cylinders):
Swing Clamps - designed for applications where the fixture workpiece area must
be free for workpiece loading/unloading without obstruction. Normally used for
clamping a component.
Linear Cylinders - designed to provide a straight linear push/pull force. Available
in many styles including threaded, block, hollow and standard cylindrical cylin-
ders or edge clamps.
Specialised Cylinders - designed for specialised applications, and include collet
locking cylinders, positive locking cylinders, etc. These cylinders can be used when
hydraulic supply has to be disconnected prior to machining operation, i.e. palletised
or progression machining.
For work-holding, cylinders can be fitted with an end-flange for rigid mounting with the
clamping device attached to the end of the rod, or the cylinder can have a threaded body
for mounting (the latter arrangement normally being restricted to miniature clamp
cylinders, although work capacities may extend up to 4-tonnes). Alternatively, a valve
cylinder can be used to power a suitable design of mechanical clamp. Clamp cylinders are
normally single-acting and fitted with spring-return; some double acting cylinders are also
used. An example of the routine of use of a swing-type work-holding cylinder is shown
in Figure 8.3.3.
MACHINE TOOLS AND AUTOMATION 603
The following list of guidelines are useful in setting up clamping cylinders:

Clamps should be applied through a supported or rigid part of the fixture
Reaction forces experienced by the cutting tool, must be taken up by a solid or rigid
part of the fixture - not by the clamps
Components should be held firmly but not to cause distortion
Clamps should be positioned on a supported area of the component, e.g. a fixed stop
or worksupport
Clamps must be positioned to avoid interference by the cutting tool.
Should extra support be necessary then hydraulic work supports can be used. These
provide an automatically adjustable support for the workpiece. This is accomplished by
one of two methods - one, the spring advance adjust work support, to contact the
component and then lock hydraulically, and secondly, the fully hydraulic version.
Automation
There are various degrees of automation ranging from semi-automation to a fully
programmed system, namely.
i) Semi-automation - typified by sequential control where an operator is responsible
for start/stop and also inspection and supervision.
ii) Automation - typified by sequential control with 'feedback' where an operator is
only necessary for stop/start.
iii) Full automation - or a programmed sequential control system using punched tape
or a computer.
iv) Fully programmed automation - where the control is fully programmed and is
capable of self-analysis, correction, etc.
Control Circuit Design

Control system design must start with a precise analysis of the movements which must be
performed, normally using a distance-time diagram. It is then necessary to allocate power
components (e.g. cylinders) to achieve these movements and select suitable control
components (e.g. robots). The major problem is then to plot the circuit to achieve correct
working with no possibility of lock-up and, preferably, elimination of all redundancies.
This can demand considerable skill and experience in circuit design and is normally best
done on a logic basis.
See also chapters on Workshop tools, Injection moulding and Presses.
605
WORKSHOP TOOLS
THE IDEA of an hydraulic workshop tool is that of performing a power action with a
device which is movable. Such a 'tool' is not necessarily restricted to a conventional
'workshop' but is seen more as a 'toolbox' item, much like a spanner or an electric drill,
but operated by hydraulic power and probably somewhat larger than normally carried in
a real tool box.
The most common application is that of raising and lowering a heavy product, such as
a vehicle, by means of an hydraulic jack. Whilst this is a very simple device consisting of
a hand pump and a cylinder with a plunger (with the all important valve) it does
demonstrate the concept of all hydraulic workshop tools.
The load supported by an hydraulic jack, or the force produced by any other type of
workshop tool, is calculated from the basic equation
Load (N) = Cross sectional area of the cylinder (m2) x Pressure (N/m2).
The majority of such jacking cylinders are single-acting, with load/gravity return,
although if used for horizontal thrusting they may incorporate spring-return. Alterna-
tively, a double-acting cylinder can be used in such applications.
Other operations which can be performed by hydraulic workshop tools include the
following:
Bending Clamping Compressing Cutting

Drilling Forming Gripping Lifting
Pressing Pulling Punching Pushing
Spreading Straightening Torque Work-holding.
Obviously each one of these can be undertaken manually, but the advantage of using
hydraulic power is the speed, accuracy and power which can be generated - each one
of which brings considerable cost saving if repetitive operations are considered. An
example of time saving with clamps is shown in Table 8.3.1 in the earlier chapter on
Machine tools.
Power source
The workshop tools may be individually designed forthejob (and integral with a cylinder),
or may be in the form of attachments designed to fit onto a single standard work cylinder.
In the latter case a single cylinder can serve a variety of work duties, although, of course,
only one attachment (tool) can be used at a time. The power source, however, is a separate
feature involving an hydraulic pump.
The simplest arrangement is where an individual tool can be connected to a hand pump
via a flexible hose. This gives a basic pump/cylinder set. Equally the hand pump can be
replaced by an electric (battery or mains) or engine driven pump set - Figure 8.4.1
illustrates this arrangement.
The pump needs to match the tool. In most cases it is the pressure which is needed, but
there are other applications where the requirement is flow; for instance, on a hand held
diamond core drill a flow around 20 Umin would be essential for maximum torque and
speed, and this can be provided quite adequately by a portable power pack.
Where the use of more than one tool is contemplated a complete system of individual
implements can be fed from a motorised pump connected to a manifold which acts as the
distribution point for individual take-off lines. This system can be quite simple (Figure
8.4.2) or incorporate additional controls for pressure and flow regulation, an accumulator
and additional valves in individual lines. The degree of control will depend on the
particular system envisaged. Valves may be necessary to control flow rate, direction,
pressure relief, shut-off and load-holding. Electric controls may be contemplated to limit,
govern or stop cy linder motions, or to provide automatic pressure control. Pressure gauges
may also be incorporated to provide a visual indication of pressure generated by the pump
at individual work points as a guard against over-loading the tool or workpiece.
WORKSHOP TOOLS 607
FIGURE 8.4.2 - Basic pump and cylinder set, using an electric pump and additional cylinders
Cylinders
The design and construction of the work cylinder needs to be more robust than that of
conventional hydraulic cylinders. In particular, the rods must be strong, with adequate
bearings to resist bending and eccentric loads without damage to the cylinder. Since
workshop tools are often used in unfavourable conditions, extra attention should be given
to the design of the rod seal pack to prevent the ingress of moisture, dirt or anything
adhesive. Figure 8.4.3 shows one design.
A single-acting cy linder can also be designed to provide both push and pull outputs, e.g.
by making it a through-rod type, or by using a hollow plunger design.
Double-acting cylinders have specific advantages for pushing, pulling and lifting
applications, but have the disadvantage of requiring two lines with a selector valve for
control. It is good practice to incorporate an integral safety valve in the cylinder to protect
it from over-pressurisation in the event of a return line being blocked.
In the case of plunger-type cylinders the output force is usually provided directly by
movement of the plunger, or from a saddle attached to the end of the plunger. If the surface
to which the force is applied is not exactly horizontal, the saddle can be of the tilting type
to relieve the cylinder of excessive side-loading.
Similar saddles may be used on the end of cylinder rods, although the variety of
attachments which can be used with 'force' or 'jacking' cylinders is considerable and
largely dependent on the ingenuity of individual manufacturers
Attachments - Clamps, Benders, etc.

The same principle can be extended to clamps, benders, pullers, etc., where the basic tool
includes a standard cylinder - some examples are shown in Figure 8.4.4
WORKSHOP TOOLS 609
The hydraulic vice is another useful workshop tool which can provide controlled force
of up to 4 tonnes or more in a very compact unit. This can be designed as a bench unit or
for fitting machine tables. Another particularly useful tool in this category is the hydraulic
collet chuck, used for holding individual workpieces for milling, drilling, slotting, etc.
FIGURE 8.4.5 - An hydraulic wedge attached to a Quarter Horse portable power pack [Power
Team]
A further device which is of considerable value is the powered wedge. Figure 8.4.5
shows an example of one attached to a portable power pack. The 12 VDC power pack
delivers 2 Llmin at up to 3 MPa (30 bar) down to 160 mLimin at up to 70 MPa (700 bar).
The wedge, or hydraulic spreader, can spread up to 100 mm max. at 1 tonne or 300 mm
spread at 0.5 tonne, with a minimum (closed) spread of only 30 mm.
See also the chapters on Machine tools and automation and Ring mains and powe rpacks
61 I
INJECTION MOULDING MACHINES
THE REQUIREMENTS of injection moulding machines are, in many ways, similar to

those in most other hydraulic machinery. Rapid production rate and efficiency is
paramount, and noise levels must be kept to a minimum. However, additional features
relate to the variations in the use of such machines.
The flexibility of the moulding machine is seen in its changing machine cycles and the
variety of material which may be used. Because of this need for variation, a modular design
is often preferred, and the setting up process requires a procedure which adaptable and
easily applied. Control may be either open or closed loop - the clamp circuit is likely to
be closed loop, whilst the screw circuit can be either open or closed loop.
(It is worth mentioning, too, that blow moulders require either open or closed loop
proportional directional control for each of the circuits used. There can be up to four
circuits, i.e. mould, mould carrier, ejector and blowing stem.)
Critical control parameters in injection moulding are temperature, injection pressure
and injection velocity. Adequate temperature control measures present no particular
problem. However, accurate control of pressure and velocity has only been possible since
the introduction of modulating valves. Such valves controIled by an electric pilot signal
can provide infinite variation of pressure and flow. They can be used on both open loop
and closed loop systems providing vastly improved system stability and precision. Other
operational advantages provided by electrically piloted proportional control valves are:
i) Smoother machine movement (including mould closure, carriage action, etc.).
ii) The elimination of ram shocks and with them the mechanical shock and vibration
sustained by the machine.
iii) The elimination of having to make hydraulic parameter changes physicaIly at the
board when they are required.
The introduction of PLC's enables more precise control of the variables in question.
Units for controlling flow and pressure

The most important movements in the process generaIly involve hydraulic assistance.
These are:
i) Screw rotation, which is usuaIly done by an hydraulic motor
ii) Opening and closing of the moulds, which is normally done by an hydraulic
cy linder.
iii) Injection action, which is also done by an hydraulic cylinder
iv) Ejector advance/retract cycling
v) Core advance/retract cycling
vi) Component removal
vii) Mould retraction
The use of variable flow pumps for the non-simul taneous feeding of the various separate
circuits, providing each with a different flow rate (and pressure) can be undertaken with
vane and small piston pumps working in the 15 - 20 MPa (150 - 200 bar) range. Dedicated
manifold design enables greater consistency in the circuit with appropriate pressure,
direction, flow and proportional valves. Modular construction also reduces pipework and
joints and potential leak points.
Mould actuation
As always there is a demand for faster cycle times, but these must not be included to the
detriment of the equipment. Sufficient damping must be included to prevent mechanical
shock and ramming in the hydraulic lines.
The ideal flow diagram for the closure cylinder is dependent on the mechanics and the
particular type of mould-closure mechanism. In all cases, however, the following
characteristic phases are present:
i) Initial movement - high pressure to overcome inertial forces and provide initial
acceleration to the moving parts.
ii) Middle phase -low pressure because the movement is at constant velocity with, in
effect, just frictional forces to overcome.
iii) Closure phase - high pressure required for closing movement of the toggle
mechanism and for holding the mould halves closed (end of cylinder stroke).
In the last phase, the pressure characteristics could require modification to the extent of
limiting the pressure for a very brief interval of time, just before the toggle reaches its
maximum extension, in order to avoid the possibility of structural damage occurring. The
pressure at the end of the stroke must, of course, return at its maximum value to guarantee
good mould closure and holding in that position. Flowrate control during the mould
opening and closing action is, therefore, of primary importance, while pressure regulation
at the end of the final portion of the closure stroke is advisable in order to avoid possible
structural damage.
Injection
The injection action is accomplished by means of cylinders and in this case, the significant
hydraulic parameters which need regulation are reduced to just one, i.e. pressure. The
injection cylinder moves at the same time that plastic material flows into the mould, which
INJECTION MOULDING MACHINE 613
occurs at a pressure of between 100 MPa (1000 bar) and 200 MPa (2000 bar) at the
injection nozzle. At least two hydraulic pressures should be provided during the mould-
filling operation.
An initial pressure level is needed to fill the mould up to 85 - 95% of its capacity, while
a second pressure level is needed to accomplish the final mould-filling operation and also
to offset plastic volume reduction due to in-mould cooling. More complex moulds and
plastic materials having particular characteristics require that three or four different
pressure levels be provided during mould filling. These hydraulic pressures are applied to
the cylinder in accordance with the mould-filling characteristics desired and are, therefore,
in proportion to the position of the injection cylinder. Since the load on the cylinder,
besides being high, is also, for all practical purposes, constant, there is a proportional
relationship between the hydraulic working pressure and the injection velocity until the
mould is filled.
Rotation of the plasticising screw

This is usually accomplished either by the use of a low-speed hydraulic motor or by means
of a geared-down motor; the latter, though, is not as acceptable. During this phase,
plasticising takes place and the injection chamber is loaded while the injection cylinder
automatically withdraws. It is generally agreed among the experts in this field that both
the plasticising and loading operations, for any particular material, should take place at
constant velocity. The parameter of prime importance, therefore, with motors having good
volumetric efficiency, is the rate of flow to the motor. In other words, the motor speed must
be independent of the load.
Circuit diagrams
The five figures give several arrangements which have been used successfully in the past.
Each one is a simplified diagram illustrating the important features of the hydraulic circuit.
Figure 8.5.1 shows the use of a programmable sequencer receiving signals from
proximity or micros witch end-stroke indicators.
Figure 8.5.2 shows control by end-stroke indicators providing hydraulic parameter
regulation by means of a proportional pressure relief valve and a three-way proportional
compensated flow regulator.
Figure 8.5.3 indicates the use of accumulators in the circuit. These preclude the use of
dissipating flow regulator valves. The mould actuation cylinders and injection cylinders
are pressure-regulated by means of proportional pressure-reducing valves. The screw
motor is regulated by a proportional two-way compensated flow regulator.
Figure 8.5.4 shows a feed screw plunger and toggle action mould separation and closure
mechanism. (Pressure and flow regulators are not shown.)
Figure 8.5.5 shows the flow and pressure being regulated by the use of a single
proportional valve capable of regulating flows up to a maximum of300 L/min and pressure
to 25 MPa (250 bar).
617
HYDRAULIC PRESSES
ALTHOUGH MANUAL presses are still used for small punching and fitting operations,
hydraulic presses are predominant in repetitive industrial machinery. Their size varies
considerably from small portable units, with hand pumps generating perhaps 70 MPa (700
bar) and a press force of 10 tonnes, up to enormous 20000 tonne dedicated forging presses.
Flow is dependent on pump operation and pressure on the reaction encountered by the
ram. Ram pressure builds up to overcome resistance until either a pre-determined
resistance or a pre-determined ram position is realised. Pressure is thus under positive
control, as is the ram speed.
Small hydraulic presses

Small hydraulic presses can be as low as 1tonne although 10 tonne is a more common basic
unit. These are produced as self-contained general-purpose units for factories, machine
shops and assembly shops. Figure 8.6.1 shows a simple 10 tonne bench type press which
is powered from a 70 MPa (700 bar) hand pump; it has a 257 mm total movement with 1.5
mm per stroke. Many are used in conjunction with automatic feeds. Control may be by dual
levers, dual knobs, single lever or by closing the guard. Push knobs should, however, be
avoided as they are most inconvenient to operate continually. The position of the top of
the stroke is easily adjusted by a hand wheel which operates a screw and nut stop. The
bottom of the stroke is usually determined by the resistance of the work, but is sometimes
controlled mechanically, especially on rapid stroking presses. A press with automatic
stroke control is shown in Figure 8.6.2. Pressure-control is often important and this is done
by adjusting a relief valve.
1 - ram. 2 - pump. 3 - electric motor.

4 - tank. 5. relief valve. 6 - directional
control valve. 7 - handwheel for
adjusting top stop. 8 - top stop.
9 - pressure gauge.
FIGURE 8.6.2 - Small self-contained bench press with automatic stroke control
The return stroke is usually hydraulically operated but the system is simplified if a
spring return is fitted. Pumps and motors vary with the individual makers but plunger
pumps, giving a pressure of 28 to 42 MPa (280 to 420 bar) are preferred for applying the
working pressure, and the speed can be varied by unloading one or more plungers. To
avoid overloading the motor the maximum pressure may be reduced as the speed is
increased.
If the main force has only to be applied for the last fraction of the stroke, a dual pump
unit has advantages, a gear or vane pump giving a rapid approach. The high-pressure pump
can then be comparatively small.
Medium-sized general-purpose presses

Both upstroking and down stroking presses are made but the preferred layout, except
where operational considerations conflict, is the downstroking press with all possible
hydraulic gear mounted on top of it, much like Figure 8.6.2. This arrangement keeps floor
space required to a minimum, although it does pre-suppose adequate roof height; the tank
should be immediately above the cylinder, so simplifying the prefill valve. The pump can
then be mounted inside, above or alongside the tank. (A slightly different type of press is
the waste compactor which operates horizontally with forces between about 20 tonne and
60 tonne.)
HYDRAULIC PRESSES 619
Double-acting rams avoid the complication of having additional return cylinders. Metal
piston rings may reduce maintenance costs and the consequent slight leakage on the return
stroke tolerated.
If a double-acting ram is used, gravity has to be relied upon to speed up the approach;
the weight of the suspended parts is normally sufficient to ensure this, and indeed the speed
may even need regulating.
A typical manual-control circuit for a medium-size press is shown in Figure 8.6.3. On

the down stroke the main cylinder (1) draws oil from the tank (2) through the prefill valve
(3). Oil from the annulus passes at first through valve (4), but as the ram reaches the end
of its stroke, valve (4) is closed by a cam and the oil must then pass through the restriction
of the reflux valve (5), and is slowed down. When the ram meets the resistance the prefill
valve closes and pressure builds up until the relief valve blows. On the return stroke the
oil from the cylinder passes through the main valve (6) back to the tank, and the oil to the
annulus passes freely through the check on the reflux valve (5).
Electrical controls are normally preferred and can result in some simplification of the
hydraulics on more complex working cycles. A typical electrical-control circuit is shown
in the neutral position in Figure 8.6.4. Oil is circulating back to the tank but restriction (8)
causes sufficient back-pressure to operate the reducing valve (6) which supplies the pilot
system for the main directional valve (5). In moving valve (5) to the 'down' position,
pressure is applied to both ends of the pull-back rams, the net result being a downward
force, the speed of descent being controlled by the reflux valve (3). The main ram draws
in oil from the tank (1 ) through the prefill valve (2). When the ram encounters resistance
pressure builds up, and when it reaches about 1.4 MPa (14 bar), valve (4) is opened,
connecting the pump directly to the main cylinder. The main relief valve (7) also protects
the pull-back circuit if, for example, the reflux valve (3) is set too closely.
For the up stroke, pressure is applied to the annulus of the pull-back rams and also the
pilot ofthe prefill valve, so that the main cylinder discharges directly back to the tank and
the top of the pull-back cylinder discharges through the main valve.
A limit switch wouldretum the main valve to the neutral position at the top of the stroke.
Larger presses
Larger presses are usually made for specific duties, for example, drawing presses, forging
presses, coining presses, extrusion presses and stretching machines, and these will be
discussed in more detail below. Some other types of special-purpose hydraulic presses are:
Body Panel Presses - large table-area presses for forming sheet metal panels
Typical maximum pressure 300-350 tonne.
Rubber Die-Forming Presses - used throughout the aircraft industry for the
production of wing and fuselage components.
Gap- Type Presses - for wheel forming, forging, straightening and bending.
Typical size 80-85 tonne.
Cold-Forging Presses - with capacities up to 1500 tonne for making components
such as gear shafts with minimum material usage, improved grain structure and
enhanced UTS.
Wheel Rim Presses - producing pressed steel wheel rims. Typical size range 220-
500 tonne.
Plastic Moulding Presses- for either continuous or single-cycle automatic control,

incorporating automatic platen heating and cooling control.
Laminating Hot-Plate Presses - for making plastic laminates. Typical maximum
sizes 2000-2500 tonne, steam-heated and water-cooled.
Board Presses - with variable pressure and dwell for the production of insulating
and building boards. Typical size 3000 tonne.
Baling Presses - for baling cotton, coir fibre, sisal and wool.
Dry Ice Press - for making blocks from carbon dioxide gas.
Hobbing Presses - for the die manufacturing industry.
Grinding Wheel Presses - hot and cold platen presses for the manufacture of
grinding wheels.
Powder Compacting Presses - for powder metallurgy production, etc.
Brake Lining Presses - typical capacity 220 tonne (see Figure 8.6.5)
Drawing presses
The superiority of hydraulic presses for deep drawing is well established. Presses
equipped for single-, double- and triple-action drawing are now made in a variety of sizes.
The conventional method of double-action drawing (Figure 8.6.6) uses a pressure plate
controlled by rams, usually four in number, clustered around the main ram. These are
arranged to apply the pressure plate before the main punch makes contact, so avoiding
puckering. The force on the pressure plate can be varied as pressing proceeds and made
proportional to the resistance to the punch ram. Suitably shaped pressings can be passed
right through the tool, making ejectors unnecessary.
It is often more convenient, however, to invert the punch and die (Figure 8.6.7) and the
blank is laid on the pressure plate. The main ram exerts sufficient force to overcome the
resistance ofthematerial and the upward thrust ofthe bottom ram. The bottom cylinder
can be connected to an accumulator or to the main hydraulic system.
With the conventional stationary bottom cylinder the main ram and press frame must
be suitable for a load equal to the sum of the forces required for drawing and for the
pressure plate. The size of press can be appreciably reduced by fitting the pressure plate
cylinders to the crosshead carried by the main ram (Figure 8.6.8). The pressure plate force
is now completely self-contained and the main ram need only be large enough to give
sufficient force for the drawing tool. With this method it is necessary to provide a separate
pressure supply to the pressure plate rams and they cannot be connected directly to an
accumulator. An alternative construction, giving the same effect, has a single ram below
the press tool supported from the crosshead by tie-bars.
Triple-action drawing tools enable a wide variety of intricate shapes to be drawn
including those with portions indented in the main surface. In one method the metal is first
gripped between the outer die and pressure plate (Figure 8.6.9) by rams clustered round
the main ram, and the drawing operation is done by the punch pushing the metal into the
bottom die in which a counter punch recedes against the thrust from the bottom rams.
FIGURE 8.6.10 - Triple action drawing tools for pressing the metal in the reverse
direction to the main draw
Another method draws the metal in the usual way and afterwards it is drawn in the
reverse direction by a reverse punch which is forced upwards by the lower ram into a
shaped cavity in the pressure plate (Figure 8.6.10).
The fluid cell press is a development of the drawing press, where the upper die is
replaced by an oil dome. The dome is connected to an electrically controlled pressure fluid
supply. A flexible rubber diaphragm is fitted to the dome. The dome is supported by
hydraulically clamping the upper and lower cylinders of the press together. As the draw
progresses the pressure in the dome is controlled relative to the draw ram position. The
pressure is increased or lowered to suit the draw conditions. The result is that very deep
draws can be made with constant product side wall thickness which is maintained even
where radii are formed in the wall. The pressure in the dome is controlled at pressures up
to 110 MPa (lIOO bar). Figure 8.6.11 illustrates the process of the fluid cell press using
two rams and an oil dome.
Forging presses
The efficiency of a forging press is measured by the time taken to produce a forging of the
correct size - the shorter the time, the less the number of re-heats required. This calls for
automatic stroke control and reversing the hammer at the exact point necessary to achieve
the desired thickness. A 3000 tonne forging press is illustrated in Figure 8.6.12.
Stroke control can be achieved mechanically or electronically. In a typical mechanical

system the position of the tool relative to the press frame is continuously measured by a
counting device, initially set to register the number of counter units representing the final
tool position. The counter then counts down as the tools converge, until zero count is
reached. At this point the return valve is triggered and the ram is reversed. Correction must
be incorporated for ram speed (usually by tacho-generator) and frame stretch (as a figure
proportional to pressure).
A circuit for a typical fast acting (planishing) forging cycle is shown in Figure 8.6.13.
In this example a special forge valve is used with LVDT feedback to provide suitable
interfacing with the electronic and microprocessor control. (The valve and its character-
istics are shown in Figure 8.6.14 - it is designed so that the flow is never reversed which
enables planishing rates of over 200 strokes/min to be achieved.)
Electric controls can be analogue or digital. The best accuracy which can be obtained
with an analogue system under industrial conditions is about I in 500. Greater accuracy
can be obtained by a digital system, but for a forging press, the accuracy corresponding
to 0.8 mm in 3250 mm or I in 4100 is sufficient.
A schematic of a microprocessor based control system is shown in Figure 8.6.15. Such
control systems, and possibly auxiliary PLC's, provide
Control logic to integrate manipulator and press
Electric power panel for proper command (ensuring full hydraulic system protec-
tion)
Provision of manual and automatic modes, with emergency push-button and

auxiliary essential devices for the user.
The main hydraulic content of the forging process includes
Main pump and motor assemblies and unloading manifolds (Pumps with flows as
high as 800 Umin at 50 MPa (500 bar) may be needed.)
Auxiliary pump and motor assemblies for forge valve pilot circuit
Accumulator push back power units for fast acting return stroke
Auxiliary power units for boost, filtration and cooling
Fluid main power supply tank and pipework
Forge valve assembly
Separate prefill and exhaust valve for mounting on press, with prefill tank, if
required
Coining presses
Coining presses are similar to forging presses in requirements, except that the piece is
cold-worked by stamping dies designed so that the area of the faces to be flattened is
reduced to a minimum. Forces between 300 and 1500 tonne may be needed, with precise
pressure-control and fast cycling times. Such machines are commonly fitted with an
automatic feeding device and hydraulic ejectors.
Extrusion presses
The size of extrusion presses varies with the type of product and the extrusion ram may
exert a force of from several hundred tonne for the smaller presses to several thousand
tonne for the larger; a 3000 tonne extrusion press is shown in Figure 8.6.16. As the
efficiency of the press depends to a large extent on the speed with which it can be fed with
heated billets and the dies changed, these items require special consideration and are often
operated pneumatically, or, on the larger presses, hydraulically. A diagrammatic represen-
tation of a typical hydraulic circuit is shown in Figure 8.6.17.
HYDRAULIC PRESSES 63 J
PLC control is ideal, connected to limit switches via encoders on the butt length, billet
length, burp cycle, etc. Figure 8.6.18 shows one example of a suitable electronic control
circuit.
Hydrostatic extrusion
Hydrostatic extrusion is done cold and invol ves the introduction of a liquid filling between
the ram and the billet so that the liquid supports the billet over its whole area and also acts
as a lubricant. Pressure as high as 30000 MPa (300 kbar) may be involved. Use is also
being made of a technique which employs both direct force and liquid pressure.
Figure 8.6.19 shows a billet with a high length/diameter ratio being extruded. The
method is equally suitable for tubes, using a travelling mandrel, clad materials and wires.
The latter are coiled before placing in the liquid cavity. Brittle materials are extruded into
a back pressure as this is found to improve ductility
FIGURE 8.6.19 - High pressure chamber on hydrostatic extrusion press
Stretching machines
Many types of rolled and drawn non-ferrous sheet and sections are treated by stretching
beyond the yield point. The material is straightened and its physical properties are
improved. Stretching machines are made in a wide range of sizes and are able to deal with
every size of drawn and rolled non-ferrous section.
The circuit shown in Figure 8.6.20 is for a 150 tonne machine with a 254 mm diameter
working ram and 89 mm diameter return ram. Main interest in this circuit lies in the
provision for preventing damage should the work break.
As it stretches the material, the main ram pushes the return ram back. The oil escaping
from the return cylinder is at first forced through the subsidiary relief valve RV2. As the
main ram pressure increases, the servo-pressure on valve SV is sufficient to operate it, so
that it bypasses the relief valve and oil passes back to the tank at a fairly low pressure. If
the material breaks, the pressure in the main ram will fall, valve SV will close, and the oil
forced out of the return cylinder will have to pass again through the relief valve, so
providing a powerful brake. The operator closes the main valve immediately.
"---
FIGURE 8.6.20 - Circuit for ISO tonne machine with safety feature should work break
Special hydraulic valve requirements

Special valves are required in hydraulic presses for dealing with the high pressures and
high flow rates encountered, and also for dissipating the energy stored in press frames and
cylinders and in the fluid itself due to compressibility effects. If such energy were released
through a conventional spool valve there would be severe shock forces generated. Also the
heat generated and the high oil velocity would lead to damage to the valve seat. Figure
8.6.14 illustrates one type of low pressure servo-controlled pilot circuit val ve which has
good dynamic response with high accuracy.
Relief valves for large presses

In large presses certain conditions can cause a rapid pressure rise, which can become
dangerous, before the ordinary relief valve has time to open.
A relief valve which is responsive to change of pressure obviates this difficulty, as it will
open if the rate of rise is rapid enough at a pressure below the ordinary relief valve setting.
Basically the valve is similar to a balanced pilot-operated valve, but with spindle balanced
by pressure on its ends. Figure 8.6.21. The capacity above the spindle is connected to the
main pressure line through a restriction and normally the pressures are balanced with the
valve held closed by a light spring. A rapid rise of pressure is delayed by the restriction
from building up in the capacity chamber, and, if the difference is sufficient the spindle
will open and in doing so will compress the oil which will act as a 'liquid spring'. This
valve can function as a conventional relief valve by connecting a pilot relief valve to the
capacity.
Amplifier valves
An amplifier valve which can be opened by the application of pressure is useful in some
circuits, and an example of this is shown in Figure 8.6.22.
FIGURE 8.6.22 - Amplifier, or unloading, valve operated by pilot pressure [Oilgear Towler]
The pressure is admitted at port B, and port A leads to the reservoir or other parts of the
system. The pilot connections are made through ports C and D. Normally the flow path
is closed by the main valve E, which is kept seated by the pressure from connection B
acting on the differential area of the valve. It will be noted that the upper portion of the
valve is larger in diameter than the valve seat; the pressure being equalised through the
internal passage F. The central hole is sealed by the pilot valve spindle.
To open the valve, a low pilot pressure applied to port C, acts on the underside of the
piston G and lifts the pilot valve against the load of the spring H. The fluid above the main
valve will escape past the pilot val ve seat through the drilled passage in the centre of the
main spool because the restriction to the internal passage F in the main valve is greater than
the restriction through the central hole. The lowering of the pressure on the upper side of
the main valve allows the differential area to act in the opposite direction, and the main
valve will lift with the pilot valve so passing fluid from port B to port A past the end of the
main valve. As the upper part of the housing is filled with fluid, port D must be connected
to exhaust to allow the trapped fluid to escape. The pressure differential between
connections B and A need only be 14 kPa (0.14 bar) to keep the valve open. The actual
maximum valve opening can be set by the adjusting screw J, when the main valve will
assume a position relative to the pilot valve in order to create a suitable pressure above the
main valve to maintain a force balance over the valve.
The valve is closed by connecting C to exhaust; the spring H pushes the pilot piston
down, closing the central hole in the main valve. Pressure then increases on the top of the
main valve, the differential area enabling the main valve to move toward its seat at the same
speed as the pilot valve. In order to fill the upper part of the housing, port D must be
connected to a fluid supply.
Pilot connections C and D can be connected through a directional control valve to
pressurise the appropriate side of piston G to achieve a more positive action.
The main valve will accurately follow the speed and position of the pilot valve. By
substituting the dimensional control valve in the pilot supply by a servo or proportional
valve, and adding an electrical position transducer, very fast and accurate control of the
fluid passing between ports B and A can be achieved. This feature is the basis of the forge
valve design described in 8.6.14.
Working pressures ofthis design of valve exceed 100 MPa (1000 bar) within ports A
and B. Pilot pressures are usually 3.5 MPa (35 bar) and up to 20 MPa (200 bar) when servo
pilot valves are used.
Prefill valves
For a given size of pump the advance and return speed of a press can be increased
considerably by fitting a fast switching prefill valve; this enables the oil to flow in and out
of the cylinder without having to pass through the pump. Usually these valves are used
with down-stroking rams, the valve being mounted on top of the cylinder, preferably in
an oil reservoir.
Prefill valves are also made with an integral decompression valve, which takes the place
of a decompression valve in the main control valve. On small presses the pilot opening
feature can be dispensed with and the cylinder exhausted through the main valve. An
example of a prefill valve is shown in Figure 8.6.23.
637
HYDRAULIC RING MAINS AND

POWER PACKS
IT IS often convenient to have available a separate source of hydraulic power and flow to
run a particular component, machine or rig. Where this is to occur within the boundaries
of a company, and is likely to be of value for some considerable time in a particular area,
the use of a ring main is ideal. If, however, the use is limited, or needs to be transported
to other sites, then a power pack would be suggested.
Hydraulic ring mains

The idea of an hydraulic ring main is similar to that of the household electrical ring main.
Here there is an identical supply available, both in voltage and current capacity, at any
point in the ring, irrespective of what else is happening in the ring. Likewise, the hydraulic
ring main is designed to be able to supply the same pressure and flow at any take-off point
within the system.
Basically the components of the ring include a reservoir, a pump (or possibly several
pumps), suitable control and relief valves and a distribution of twin pipework around the
whole area to be ringed. The twin pipes are the high pressure supply pipe and the low
pressure return pipe, each with a more than adequate supply of take-off coupling points.
Obviously, there is also a need for coolers and filters and one-way valves, and possibly a
bank of accumulators to accommodate wide variations in demand.
With the hydraulic ring main, careful design is required to ensure that sudden changes
in one part of the circuit do not significantly affect the other parts, such as might occur with
a sudden increase in flow demand. Pressure has to be held to a tight tolerance. One
technique is to run all the likely demands at the same time (although not actually requiring
them), but this is expensive. Microprocessor control enables a much more efficient system
to operate particularly where variable loading is expected.
One type of control, using a PC workstation and aPLC, is illustrated by the VDU display
shown in Figure 8.7.1. Here a considerable back-up is arranged, but which is not used
unless required by the systems under test or if something goes wrong. There are seven
general purpose test manifolds, four test cells and other rigs which may be in use, the fluid
being pumped around the system in 100 mm nominal bore pipes from a special supply
facility. The flow of 2280 Llmin at 21 MPa (210 bar) comes from two independent supply
systems involving six pumps each. The control side includes the motor control system
with a PLC, a PC workstation with an RS 422 link, and manifold distribution/isolation
control panels - the objective being to minimise the power used to supply the test cells.
Hydraulic power packs

Hydraulic power packs - also known as 'hydraulic power units' - are self-contained
hydraulic supplies normally comprising an integrally mounted electric motor/pump unit
with associated tank together with necessary valves.
The pump/motor unit may be mounted on the tank, or separately, and packs are usually
available in either horizontal or vertical configuration. In some cases, with vertical
configuration, the electric motor may be mounted on top of the tank with the hydraulic
pump within the tank. Petrol and diesel engines may be used where an electrical supply
is not available and a battery pack may be inadequate.
HYDRAULIC RING MAINS AND POWER PACKS 639
Relief and check valves may be mounted on a manifold (e.g. on the motor end of the
pump), or on the tank. The basic unit may then be piped to cylinders or actuators through
a suitable control valve; or tidily mounted in a console with line connecting points, and
control valves and any required instrumentation (e.g. pressure gauges) mounted on the
fascia. Hose assemblies are generally preferred to rigid piping for connecting the power
pack to actuators, but this depends on the circumstances. With hose assemblies, for
example, it is a simple matter to disconnect the power from one machine and transfer it
to another, should this be required.
FIGURE 8.7.2 - Compact power pack within a steel frame [Paratech (UK) Ltd]
Some power packs are specifically designed to be portable, i.e. fitted with lifting
handles or skid- or trolley-mounted in the case of larger sizes, whereas others are intended
for more or less permanent or console-type mounting. Figure 8.7.2 shows a Honda driven
power pack type which is able to supply up to 30 Llmin at up to 14 MPa (140 bar); it should
be noted that the reservoir capacity may be quite small and cooling and a separate supply
may be needed.
More permanent power packs can be built to specification and can include much larger
capacities and pressures. Maximum pressure rating is usually of the order of 21 MPa (21 0
bar).
At the other end of the scale miniature hydraulic power packs are also available dri ven
by 12 or 24 volt DC electric motors, or even smaller sizes for special applications. Compact
units are particularly valuable for small applications and one example is shown in Figure
8.7.3 of a pack designed for the lift of an ambulance, operated manually.
Other packs take the form of battery packs which can be carried on the shoulder;
although they can give very high pressures, e.g. 70 MPa (700 bar), the reservoir capacity
will be minimal. Figure 8.7.4 shows one example.
Selection
Power packs should be specified to include a number of different features, depending on
the application. For instance:
Hydraulic pressure- The system pressure requirement to provide, say, the correct
clamping force from a cylinder (which should not be exceeded).
Oil capacity - The oil reservoir capacity should exceed the total requirements of all
the cylinders in use.
ai/flow - The oil flow should be based on a flow rate which delivers the required
oil capacity within the required operating time (for clamping, for instance). It must
not exceed the maximum flow rate permitted in any part of the complete system.
Power source - e.g. Pneumatic, electric, diesel, etc.
Control valves - As required for the system holding, relief and flow; possibly
controlled manually, electric or pneumatically.
HYDRAULIC RING MAINS AND POWER PACKS 641
FIGURE 8.7.4 - A battery powered hydraulic power pack [Enerpac]
Applications
A representative list of power pack applications includes scissor lifts, access ladders,
tailgate lifts, lorry-mounted cranes, dump bodies, snow ploughs, paraplegic lifts, hose
swaging equipment, cutters and drills, and rubbish compactors.
AC power units are used in a wide range of machine tool applications and provide
hydraulic power and pressure for lubrication of bearings, and for the operation of clutches,
clamps and shuttles of all kinds. They provide the power to move and feed cutting
equipment such as drill heads, boring heads and other hydraulic-fed cutting-tool holders.
An especially important application in the machine-tool field is in the retrofit area. Users
who wish to retrofit tracer equipment, special clamp devices, special tooling, or other
typical machine tool features can power them with a separate hydraulic power unit without
having to tap into the main hydraulic system of the machine tool in question.
Hydraulic power units are widely used for rubbish compactors, and in the supply of
hydraulic pressure to thrust cylinders. These power units usually incorporate a Duplex
pump providing two fixed deliveries to achieve maximum output of oil at low pressures
for fast traverse and quick return of the cylinder. It also provides a high pressure low
volume supply for the compaction work stroke. This arrangement not only reduces heat
but also permits the use of smaller electric motors at full power output at both of these flow
and pressure requirements.
All the components for this operation can be incorporated in a Duplex pump located
within the reservoir. When the compaction cylinder meets the load resistance, the pressure
in the system rises until it overcomes the spring force setting on the unloading valve stem
in the high-volume pumping section. The valve stem moves, allowing the high volume
pumping section's flow to bypass at low pressure into the reservoir. An internal check
valve between the two pump sections prevents the high-pressure flow from leaking back
during the unload phase. When the pressure in the system is reduced below the unloading-
valve setting, the combined flow of the two pump sections is made available.
An externally adjustable relief valve is fitted in the hydraulic circuit to prevent
excessive pressure build-up. The electric circuit incorporates a differential pressure
switch taking pressure from the thrust delivery. When resistance is met and the pressure
builds up to give maximum thrust, the pressure switch operates. This provides an electrical
signal to a solenoid operated four-way spool valve to return the cylinder to the 'load'
position.
See also chapters on Workshop tools and Reservoirs.
643
MOBILE AND AGRICULTURAL

HYDRAULICS
THIS CHAPTER examines the use of hydraulics in the non-transport sector - automobile,
rail, marine and aerospace are covered in other chapters. Specifically, then, the contents
here relate to such items as earth movers, cranes, winches, forestry and farming; however,
lifts (and fork lift trucks) are discussed in the Mechanical handling chapter.
The principal advantages of hydraulics as the choice for powering mounted or remote
systems are:
High power output from a compact actuator
High power conversion efficiency
High power/weight ratio
Extreme flexibility of approach and control.
These characteristics are largely responsible for the fact that the hydraulic cylinder or
ram has become in general the standard choice for lifting, slewing, shoving, digging,
tipping, etc. on chassis mounted equipment.
Such equipment can be towed, or be self-propelled. In either case, the engine necessary
for movement can be utilised to drive the hydraulic pump with a considerably higher
efficiency than that achieved by mechanical take off.
The features which are special in mobile machinery are:
Power sources and pumps
Actuators (cylinders or rams)
Control valves
Circuit design, and
Hydraulic motors.
Each of these will be discussed in the following text, before more detailed examination
is made of earth movers and agricultural tractors.
Power sources and pumps

Mobile hydraulics implies the movement of the power source which drives the hydraulic
pump. This source is likely to have the dual role of driving the 'vehicle' as well as powering
the hydraulic system. Normally the greater power demand is from the transportation
system, particularly if the machine is large, and there may be cases where it is more
economic, or more convenient, to provide the hydraulic system with its own power source
so that it becomes a self-contained power pack. There is, of course, the possibility of using
an all-hydraulic system where mobility is also provided by a hydrostatic drive with
hydraulic motors - see chapter on Land-based transport.
Diesel engines are probably the most common source of mobile power, but petrol is also
used. In some smaller applications an electrical supply may be used when within buildings
or close to them.
Drive couplings
Where up to full engine power is used for hydraulics, the hydraulic pump is driven directly
from the engine crankshaft. Equipment carried on a self-propelled chassis commonly has
the hydraulic pump driven from the engine gearbox or crankshaft. Where the prime mover
already incorporates a power take-off, as in the case of tractors, a pump/gearbox unit may
be fitted directly to the power take-off, or it may even be shaft driven if the pump is an
integral part of the separate trailer-mounted equipment.
Pumps
Although many mobile machines are used in the open air, there is still a strong lobby
requiring the minimum of environmental noise. Pumps are therefore chosen which are
known for their low noise levels, such as vane types or the newer design of gear pump. (See
also chapters on Pumps and Vibration and noise.) The gear pump (and motor) is ideal for
low cost mobile hydraulics.
However, the axial or radial piston pumps, because of their high pressure performance
and versatility, are essential in some cases, and suitably lower noise outputs have been
made possible by careful design. Indeed, variable displacement axial piston pumps are
made by a number of manufacturers specifically for mobile operation because of demand
by users.
Actuators
Normally a cylinder is used for deriving all the linear motions required, operating as
directly as possible (i. e. with minimal mechanical linkage ). The lower the system pressure,
the larger the size of cylinder required to produce a given output force and the greater the
pump delivery required to achieve a given rate of working. Except in special cases,
however, the work done is seldom significant. It is the force output which counts, provided
this is achieved at an acceptable rate. This applies to most of the motions provided by
mobile equipment for mechanical handling, etc. The input power required by the pump
will, however, be directly related to its delivery (governing work rate of the actuator) and
pressure (governing force output of the actuator).
The highest working efficiencies can usually be achieved with higher working pres-
sures. Thus although 35 MPa (350 bar) has been about an optimum figure more recent
designs can cope with 40 MPa (400 bar) with pressure peaks as high as 60 MPa (600 bar),
although these higher values are used normally only in specialised systems such as aircraft
MOBILE AND AGRICULTURAL HYDRAULICS 645
hydraulics. For mobile equipment, 14 MPa (140 bar) is a more realistic optimum taking
into account practical considerations and cost.
The sealing of the cylinders has attracted considerable attention. Formally, even though
friction and cost were higher, seal sets of chevron type were favoured for piston seals
instead of single lip seals as they do have a longer life and lower leakage. However, more
recent designs use polyurethane; they include polyurethane rod and wiper seals as in
Figure 8 .8.1 and piston seals. Piston seals, unlike rod seals, have to work with loads on both
sides and hence a special design involving a fabric reinforced thermoset, filled thermo-
plastic or PTFE sealing element and energiser ring may be used; in the design shown in
Figure 8.8.2 there are in addition two supplementary radial grooves to prevent the 'blow-
by' effect during rapid pressure rises. (V-cups are common for single sided applications.)
For even greater pressures the rings have been redesigned to give further protection against
gas extrusion; for this, pressures in excess of 40 MPa (400 bar) are acceptable (Figure
8.8.3).
A 14 MPa (140 bar) system is capable of providing a force output of the order of I tonne
from a cylinder only 31 mm in diameter, or 10 tonnes from a 93 mm cylinder. A high
pressure mobile cylinder with another design of sealing is shown in Figure 8.8.4. See also
chapters on Hydraulic cylinders and Seals.
Control valves and selectors

The control of flow and pressure may either be direct through manual levers, or remote
through some form of solenoid activation. By using light-weight spools a greater finger-
tip sensitivity is achieved in manual lever operation, which an operator of say, a back hoe,
front end loader or heavy duty crane, finds is able to provide extra-fine metering.
Proportional control, particularly with electro-hydraulic systems, was earlier seen as
only available from servo-valves - and indeed, they do give high response, high stiffness
and are ideal for closed loop control. But such valves are expensive and do not behave well
in extreme environments, and there has been a move towards solenoid operation. The
proportional solenoid valve (where the valve spool is positioned in response to an
electrical signal) is best in open loop systems - normally without feedback - with less
positional accuracy. However, positional and pressure accuracy can be improved by using
the sensing of the piston on certain designs and the use of digital amplifier cards (Figure
8.8.5).
FIGURE 8.8.5 - Proportional solenoid valve with digital control [Hydraulik Ring]
Control valves and selectors for mobile hydraulics are invariably designed for stacked
assembly, with internal circuit configurations for parallel circuits, series circuits or tandem
circuits (series-parallel). Some offer a combination of these circuits in anyone valve block
to adapt directly to a specific application (instead of employing a combination of
individual valve types). For example, a tractor shovel requires a separate digging
operation, with dipper arm and swing services simultaneously. The digging service would
have a series-parallel circuit while a series circuit would be used for the other two. The
combined pressures of the two services, simultaneously operated, are generally less than
the pressure required on the digging service, so that the pressure limitation of the series
circuit would be of no significance in this instance. See chapter on Directional and flow
control valves.
Circuit design
Parallel circuit configuration (Figure 8.8.6)
In a parallel circuit, the valves have a common internal pressure gallery, which allows
simultaneous flow to any number of service ports. In theory, it is possible to select any
number of spools and thus provide full pressure to as many services as are required. In fact,
splitting the flow between a number of services means that the operational speed will be
lower than the maximum available. In practice it is rare that each service requires the same
pressure. If the spools are fully open then the oil will go to the lowest pressure first, so that
this will move before the other services. This can be counteracted by throttling the flow
on that particular spool, and thus creating sufficient pressure in the valve internal gallery
to operate other services at higher pressures. The effectiveness of this is dependent upon
the metering characteristics of the valve, which should be tailored to the application.
Series circuit configuration (Figure 8.8.7)

The series circuit also allows simultaneous operation of more than one service, but in this
case the exhaust from the spool nearest the inlet is the supply for the next downstream
spool operated. The disadvantage is that the operating pressure of the last service is
additional to the pressure required by the preceding upstream service, which again is
additional to that required by the valve spool further upstream. Although the flow to each
service is approximately the same, the total main relief valve pressure available is split
between the number of services selected. Consequently, although simultaneous operation
is available, the usefulness is restricted by the limitations of total pressure.
Series-parallel or tandem circuit configuration (Figure 8.8.8)

In this type the spool nearest the inlet has first priority. If two spools are operated at the
same time, the upstream spool closes off the through-bypass circuit. The flow will then
automatically go to the next spool downstream. It is therefore possible, by throttling, to
operate the spools simultaneously, but normally this circuit is used where it is required to
feed the whole of the pump supply to only one service at a time.
Spool valves are available either bolted together in banks, with sandwiching gaskets
between them, or as complete units containing encapsulated relief and check valves. See
chapter on valves.
Hydraulic motors
Hydraulic motors may well be considered for linear applications in certain fields where
cylinders have limitations - particularly long-stroke movements where side loads may
complicate cylinder design requirements and add to costs. A motor-driven rack and pinion
output, or an hydraulic motor-driven rack and pinion output, or an hydraulic motor driving
mechanical linkage, at a comparable overall conversion efficiency, may prove a reliable
alternative.
A particular advantage of the hydraulic motor is that it can give a starting torque
approaching that of its running torque and run very smoothly at very low speeds. Inertia
is also much lower than with most other forms of drive, so that starting and stopping can
be almost instantaneous. Shock loads, which are so often a cause of failure in gearboxes
and mechanical drives, are fully absorbed by the fluid. The use of hydraulic motors has
revolutionised the powering of slow-speed devices in many industries, notably winches,
conveyors and cranes. An 80% conversion of energy is a very conservative estimate (e.g.
the hydraulic pump required to drive a motor of P kW power would need a power
hydraulic of the order of 1.25 P kW ).
Like the cylinder, the hydraulic motor can be mounted at the working point, requiring
only fluid-line connections and giving the same sort of flexibility of operation as that of
an electric motor but with far greater flexibility of performance. Hydrostatic drives are,
in effect, merely an extension of this principle. And both hydraulic motors and hydrostatic
drives provide yet another field of further application for mobile equipment anywhere, or
in fact, for any service.
Earth-moving vehicles
The heavy duty demands of earth-moving vehicles have favoured the adoption of
hydrostatic transmission as well as hydraulic machinery. The latter most commonly
involves linear motions which can be provided by a cylinder or cylinders through a linear
mechanism. Where rotary power is required, or can conveniently be utilised, it can be
provided by high-speed low-displacement hydraulic motors or low-speed high-torque
piston motors. Torques in the latter case are considerably in excess of those provided by
any alternative machinery of comparable size.
Power for hydraulic machinery is provided by a separate pump driven by the vehicle
(diesel) engine. In the majority of cases this is of fixed-displacement type. The first
problem is coping with the power developed when all the services are inoperative: boom,
digger, slew motor and track drives. In addition, there is the problem when the actuator
reaches the limit of its stroke of limiting the pressure by some pressure-limiting device,
such as a relief valve, which can also serve as a spill for the unused flow from the pump.
There are two main solutions to the problem of 'off-loading' or 'idling' . One relies on
the use of variable delivery pump where the lever setting the pump delivery control is acted
upon by the pressure. As the pressure increases beyond some set value up to the maximum
the lever moves in such a way as to reduce the pump output to zero.
Alternatively, a fixed-delivery pump can be used with some form of unloading valve
which, behaving in a similar manner, dumps the flow back to the reservoir at a relatively
low pressure. The system is maintained at its high pressure by means of a small delivery
pump protected by a further relief valve. Both these systems maintain a live supply.
The two or multi-pump system is ofetn used in conjunction with the central by-pass
valves described above. These off-load the pumps when no load is required by any of the
services. An alternative series of valves also used are the pressure compensated valves
described in the chapter on Proportional valves. These gi ve good modulation of flow with
the additional features of a compensator which maintains a constant pressure drop across
the control spool and hence independence from pressure variations. Such valves can be
used to great effect when combined with a variable displacement pump to give a power
matched system. This offers the advantage of good energy efficiency with both the pump
pressure and its flow output matched to the service requirements and tracking any changes.
Figure 8.8.9 shows a typical circuit using closed centre valves. The load sensing (LS)
connection is used to control the pump output and maintains a flow level just above the
system requirements. If no valve is selected then the pump will give a minimum output
simply making up leakage flows.
With all line supply systems enough heat may be generated by the 'off-loaded' pump
to necessitate a small radiator or other heat exchanger. Again, in the neutral position the
service connections to the actuator can either be blanked, maintaining it in any desired
position for booms and buckets on loaders; or it can connect the service connections
together (and incidentally to the reservoir) on some track motor controls. It is also common
to incorporate check valves to prevent the inadvertent release of pressure on the operation
of a subsequent service.
Where multiple fixed-delivery pumps are used, for example three on a front loading
shovel, they are usually arranged so that one is normally directed to the power steering
whilst the others feed the bucket and boom rams. The power available from the engine is
limited and if one pump running at its maximum absorbs all the power, a special valve has
to be incorporated to sense this and limit the combined power of the pumps within the
engine capacity.
Excavators and diggers

Excavators may either be tracked or wheeled vehicles, but both have very similar
hydraulic content. Figure 8.8.10 shows six features which are common on these units.
There are many varieties of pump combinations and, in addition to the digger operations
of the boom, dipper, bucket and slew actuators, there is also the problem of providing
power for the track motors which normally drive through a gear pair to produce the
necessary low revolutions.
In oredr to increase the speed of rams when extending, or to transfer fluid under
following load conditions, user is made of a differential circuit. Here the annular area fluid
is directed into the full area to help the pump which now increases its speed as only the
piston rod volume has to be filled, although the effort is reduced as the pressure acts only
on this area. Units known as demand valves may be incorporated so that the necessary
services are given priority.
Pressurisation of the track motors also releases their brakes hydraulically and the slew
motor is operated by an open neutral selector which allows the assembly to rotate freely
to the desired position at which the operator applies a holding brake.
With variable-delivery pumps, in particular, it is possible in a very simple manner to
produce a constant horsepower control to avoid stalling an engine.
On crawlers of all sorts it is now quite common to drive the tracks by means of hydraulic
motors, using either high-speed units and reduction boxes, or low-speed piston motors and
a single pair of gears. Each track is then controlled separately by the operator to steer the
vehicle and the application of pressure to the motor releases the brake holding it.
When it is realised that these machines can climb gradients of 1:3 it is obvious that the
torque applied at the drives is of considerable magnitude and can really only be obtained
from a hydraulic transmission or torque converter, the former giving a much larger range
of speed control.
Another use for hydraulic motors is in the slewing mechanism, used on excavators,
which rotates the cabin, engine and digging assembly on a large bearing at high speeds and
requires high accelerations only obtainable from fluid power.
Agricultural tractors
The basic method on which the traction and position-controls of most modem tractors are
based still follows the original Ferguson system (of the 1940' s ?) where implements were
mounted directly on the tractor, lifted and lowered by hydraulic rams, with the further
refinement that hydraulics also provided a means of controlling the manner in which the
implement moved through the ground. By means of linkage and levers the traction of the
implement was used to signal to the control valve handling power to the lift, so that weight
could be transferred to the rear wheels of the tractor to increase its pulling power.
Today most farm tractors incorporate hydraulic power take off (HPTO) as standard, and
may have quite sophisticated systems for traction and position-control. With the increas-
ing use of the tractor as a mobile power source, cylinders are also used on tractor-trailed
equipment such as tipping trailers, and there has been an introduction of many services for
which hydraulic power has replaced human muscle power.
Front-end loaders and ditch digging and cleaning attachments are commonplace, and
there is an increasing use of hydraulics to power external and auxiliary equipment such as
hedge and verge trimmers, power saws, etc.
While the availability of the built-in hydraulic system for tractors has led to the idea of
powering external applications such as these, the power available is often insufficient.
This, coupled with the need for more sophisticated control systems, has led to the provision
of auxiliary hydraulic systems to meet the larger capacity or control refinement demanded.
Though the demand for auxiliary power is probably not yet such that fitting increased
capacity systems would be justified economically on all tractors, many manufacturers
now offer machines with such a system.
As the benefits to tractors from hydraulics became apparent, so their use in other farm
machines developed. For instance, the combine harvester where a single hydraulic service
replaced the electrical lowering and lifting of the cutting table some thirty years ago, now
has the pick-up reel raised and lowered as well as rotated hydraulically, and also
incorporates other hydraulically powered functions including power steering and hydrau-
lic disc brakes for maximum braking efficiency.
Another major farming machine is the beet harvester which can automatically harvest
up to six rows at a time (see Figure 8.8.11). This has a much more complex logistics
demand than the wheat harvester, for instance, it must cut off the tops first and then pull
the beet from the ground whilst moving in very precise rows at about 10 mph. The
hydraulic system involves two quite separate units - one to provide a transmission for the
chain drive and the other to perform the harvesting and control functions. Special 'feelers'
are used to sense the beet rows and a control loop alters the steering accordingly. Because
of the large number of dirt-sensitive components operating in an unfriendly environment,
each of the subsystems is protected with pressure line filters; it had been found that
cleanliness of ISO 4406 16/13 was acceptable in the transmission unit and 12111 in the
main system.
See also chapter on Mechanical handling.
655
LAND-BASED TRANSPORT
THE USE of hydraulics in road and rail vehicles tends to involve those features where
manual control is inadequate as regards strength. The well known aspects of hydraulic
braking and power steering are two which are very common in automobiles; there are also
the hydraulic suspension dampers. Larger vehicles may use hydrostatic transmissions
where all four wheels are driven by hydraulic motors, and power lifts, loading arms and
small winches may occur on vans and open trucks.
Basic automobile hydraulic brakes

Hydraulic brakes are standard in automobiles, the basic system employed having changed
little since the 1940's. A master cylinder of the order of 25 mm diameter is operated by
a foot pedal through mechanical linkage so that a maximum pedal pressure of the order
of 440 N will produce a fluid pressure of the order of 4 - 6 MPa (40 - 60 bar). This master
cylinder supplies fluid to four slave cylinders simultaneously through two separate
independent circuits, so that if one should fail the other will brake at least two wheels
Figure 8.9.1.
The hydraulic circuit is a simple one-way type with spring return of the slave cy linders.
No control valves are necessary since fluid flow is controlled by pedal pressure providing
master piston movement in one direction and by spring return of the slave pistons when
pedal pressure is released. A small reservoir is, however, necessary in the circuit both to
ensure that the system always remains filled and to provide a compensation chamber to
accommodate changes of fluid volume with temperature.
Forces produced at slave cylinders on hydraulic braking systems can be readily altered
by adjusting the size of the cylinders. Thus to increase brake-actuating force at, say, the
rear wheels, the diameter of the slave cylinders for these wheels can be increased; this
would also be necessary if disc brakes are used instead of drum brakes. Further
redistribution of effective braking can also be achieved by modification of the actual
mechanical units involved, e.g. to provide mechanical, or power assisted, 'servo-action'.
Achieving the correct front-to-rear balance in braking systems is obviously an impor-
tant handling and safety consideration. It is a particular problem in those vehicles which
experience a wide variation in weight distribution between laden and unladen conditions
such as light vans and estate cars. A brake pressure regulating valve can be used which will
limit the proportion of master cylinder pressure applied to the rear brakes. The valve is
mounted near the rear suspension and is mechanically linked to change its position as the
suspension is compressed when the vehicle loading is increased. This increases the
proportion of pressure and the relative braking effort at the rear.
The main requirements for the fluid are that it should have a viscosity capable of
providing sealing at all likely ambient temperatures (and the high temperatures seen in disc
brakes) and sufficient lubricity to lubricate the piston seals adequately. It also needs to
have good chemical and physical stability since it is in contact with air in the reservoir. Seal
requirements are compatibility with the fluid, low break-out friction (low 'stiction') and
high resistance to ageing and cold setting.
Wheel slip at braking, or locking of the brakes, can be dealt with in several ways
involving gravity pressure valves, or more complex computer contro\. Wheel spin is the
opposite concept, but again may be corrected by controlled hydraulic braking. For
instance, on Euclid Hitachi 60 and 85 tonne trucks the rotation of each wheel is constantly
monitored and a microprocessor used to control an electrohydraulic proportional valve so
that it sends a control pressure to a reducing valve in the braking system to stop any
offending wheel (Figure 8.9.2).
LAND-BASED TRANSPORT 657
Vacuum-hydraulic brakes
Where the brake-actuating forces required are greater than can readily be obtained
manually it is essential to boost the pedal input. This normally involves the employment
of a booster cylinder which supplements the pedal pressure above a pre-determined level
of input effort. Higher input pressures are then accompanied by 'boost' pressure, resulting
in a substantial increase in the force applied to the master cylinder and a consequent
increase in system pressure without requiring excessively high pedal loads. Such a system
also retains the same 'feel' as a simple system, with braking effect directly proportional
to pedal pressure. A typical vacuum booster unit is shown in Figure 8.9.3 which, used in
conjunction with an otherwise conventional hydraulic circuit, is correctly referred to as a
vacuum-hydraulic servo-system rather than just power assisted braking.
Servo-power action can usefully be extended to assist manual operation of mechanical
clutches. Since the engine is invariably flexibly mounted and the engine unit and clutch
actuating linkage can have differential movements, a hydraulic link is also effective in
both eliminating wear on all mechanical linkages and providing more precise response.
It is also possible to connect the clutch pedal and mechanical clutch via a simple hydraulic
system of master and slave cylinders which is unaffected by relative motion of the engine
1-vacuum valve seated. 2-valve disc.

3-atmosphere valve open. 4-plunger
and diaphragm moved forward. 5-f/uid
displaced to wheel cylinder. 6-ball-end
seated in piston. 7-piston moved
forward. 8-from master cylinder.
9-main diaphragm moved forward.
FIGURE 8.9.3 - A vacuum-hydraulic servo system in the applied position.
unit and chassis. Force multiplication, (that is, 'servo-assistance') mayor may not be
provided, according to the size of cylinder and operating force required.
Servo-operated brakes
For heavier vehicles a complete servo-power system is generally preferable for brake
actuation, utili sing the delivery from an engine-driven pump. A circuit of this type is
shown in Figure 8.9.4, employing a tandem power valve and two accumulators. The pump
draws hydraulic fluid from the tank and delivers it to a cut-out valve on the first
accumulator and a non-return valve on the second accumulator. Each accumulator is
connected to a unit on the tandem power valve which in turn connects to the appropriate
brake lines. One accumulator feeds the front and the other the rear brake lines.
When the brake pedal is depressed this operates the power valve admitting pressure
from the pressure lines into the brake lines, and actuating the mechanical braking system
in the usual manner. Release of brake pressure allows fluid to flow from the brake lines
to the return connections.
In power brake systems it is usual to arrange the lines so that in the event of a pipe
breakage, or similar failure, power braking is still available on at least one axle. Manual
reversion mayor may not be made available in the event of complete power failure. If
required, pressure switches can readily be incorporated into the system to indicate loss of
pressure in the system or in either circuit of a dual system.
A-tank. 8-filter. C-frame cylinders.

D-pump. E and F-accumulators.
G-frame cylinders.
2-cut-out. 3-non-return valve.
FIGURE 8.9.4 - Example of a power braking system circuit with tandem power valve.
Power steering
Power-assisted steering is now available for most cars, as an extra if not as a standard. All
but a very few (which are electric) are hydraulically assisted. In its basic form, it comprises
a hydraulic booster cylinder with one end fixed and the other linked to the steering arm,
fed by an individual hydraulic circuit with its own pump and control valves. A typical
system comprises a hydraulic pump (commonly of the vane type) driven by the engine,
an oil reservoir (usually integral with the pump), an actuating cylinder incorporating a
directional control valve, open-centre type valve, and flow-regulating and pressure-relief
valves. These two valves may be incorporated in the same block which may be integral
with the pump/reservoir unit. The actual arrangement of components can vary with the
type of vehicle and space available. Some alternative methods of mounting are:
Pump, valve and reservoir in a single block, requiring only two lines to connect to
the actuating cylinder. The cylinder directional control valve is then operated by
mechanical linkage from the bottom of the steering column.
Separate pump and reservoir. This has the advantage of allowing the pump to be
mounted at the front of the engine on an adjustable bracket. The reservoir unit may
incorporate the valve block; or the safety relief valve may be incorporated in the
cy linder.
Relief and directional control valves incorporated in the cylinder and the flow
regulating valve omitted. The pump is then associated with a simple reservoir
connected to the cylinder by two lines. This simplified system is suitable for use on
slow moving vehicles.
Power steering worked off an existing hydraulic circuit in the vehicle. In this case
the existing reservoir can be used although the power steering needs its own pump,
which can be separately mounted at the front of the engine. The valve block is then
separately mounted. On slow vehicles the flow-regulating val ve can be omitted and
both the safety-relief valve and directional-control valve mounted on the cylinder.
There are also more sophisticated systems used, e.g. to provide proportional response
and/or vary the degree offeedback force or 'feel' on the steering wheel itself. Typically,
using a spool valve as a control val ve, the amount of hydraulic assistance is then dependent
on the movement of the spool, which is in proportion to the effort applied by the driver to
the steering wheel. If the steering resistance at the road wheels is small, as on an icy road
or when making small corrections at speed, the amount of effort and valve movement will
be small, as will also the degree of power applied, thus avoiding over-violent response.
To obtain maximum advantage from power-assisted steering the ratio can be made to
vary progressively from a moderate ratio at the straight-ahead position to a numerically
lower ratio towards each lock. Thus the overall number of steering wheel turns between
locks can be reduced without having gearing too direct for the straight-ahead high-speed
position. It is then possible to turn the steering wheel from lock to lock with the car
stationary or when manoeuvring at low speeds with a very moderate effort at the steering
wheel.
Secondary systems
Secondary systems which may be operated by an engine-driven hydraulic pump include
door opening, hood raising, built-in jacking systems, etc. For jacking purposes, however,
a manual pump is more usual since there is a relatively limited demand for this type of
system. Ancillary systems ofthis type commonly employ a gear pump which actuates one
or more miniature cylinders through a control valve. Rotary actuators are occasionally
used. Hydraulic motor drives are also used to a limited extent on primary services in some
designs, for example, driving the cooling fan. In this latter example the flow rate to the fan
motor, and thus the speed of the fan, can be thermostatically controlled so that the degree
of cooling is matched to the engine temperature.
Auxiliary services
On certain classes of commercial and other heavy vehicles, hydraulic power may be used
for ancillary services, such as tipping motions on trucks, or the operation of special
attachments. Where the demand on the hydraulics is light or intermittent such services may
well be powered by the same hydraulic motor as the power-steering circuit, and an
accumulator may possibly be included in the circuit. Normally, however, such demands
are best met by a separate pump powering an entirely separate circuit; this pump may either
be a second pump in tandem with the first or be a complete power pack as the example
shown in Figure 8.9.5 for tail lifts and demount vehicles.
Engine starters
The high power necessary to start large diesel engines (up to 1500 kW) can readily be
provided by an hydraulic motor and hand-primed hydraulic system. This has the advan-
tage of making the starting system completely independent of battery or other external
sources. An example of such a starter system is shown in Figure 8.9.6.
Power is derived from a hydraulic accumulator pre-charged with nitrogen to 10.5 MPa
(105 bar). When oil is pumped into the accumulator the gas is compressed to 14 MPa (140
bar). This stored energy is released by opening the hand-operated starter valve allowing
a small quantity of oil to pass through the hydraulic starter motor giving slow rotation of
the Bendix unit to achieve full engagement with the flywheel. The valve then opens fully
allowing full flow to the hydraulic starter motor, thus turning the flywheel through nine
complete revolutions at 375 rev/min and starting the engine. The oil returns to the reservoir
where it is pumped back to the accumulator.
The system can be re-charged by an engine-driven pump in only thirty seconds after
starting. The system is protected against over-charging by an off-loading valve which,
when the pressure reaches 14 MPa (140 bar) re-directs the oil in an open circuit from the
re-charging pump back to the reservoir, where it will continue to circulate at low pressure.
Automatic transmissions
Automatic transmissions are another major application of hydraulics to vehicles. These
are highly-developed individual designs based on hydraulic torque converters, planetary
gearing and hydraulically-operated multi-disc clutches and band brakes. Power flow in
each gear ratio is achieved by locking various elements of the planetary gear set under the
control of hydraulic valves. The size of each transmission is determined by the torque
characteristics of the engines they are designed to match.
Hydraulic suspensions
For many years Citroen cars have been made with a self-levelling suspension which
accounts for load changes and leakage in order to ensure that suspension travel is
maintained regardless of use. This hydraulic system is not intended to influence the
suspension dynamic response to road or manoeuvre induced loads, and as a result the
power and flow requirements are minimal. Indeed the ride height control valve is
mechanically attached to the suspension via a flexible member in order to ensure that the
response to inputs over about 0.3 Hz is attenuated. In comparison, fully active suspensions,
first seen on race cars in 1981, seek to control all wheel movements by replacing the
conventional passive suspension spring and damper components with a load controlled
high response hydraulic actuator. Much higher frequency movements need to be absorbed
by this active suspension and a valve bandwidth of over 100 Hz is necessary. One such
fully active suspension system is shown in Figure 8.9.7
A requirement for the hydraulic pump (engine-driven by means of a toothed belt) is that
its delivery, even at low speeds, must be sufficient to cope with the high lateral forces that
can be applied to the car in violent manoeuvres. For all the installations so far developed
an axial piston unit has been the choice; with constant pressure control, the swash-plate
angle is varied automatically, according to the speed and fluid flow demand. This
arrangement minimises power consumption and therefore heat input to the fluid. The
hydraulic circuit incorporates an accumulator to damp peak pressures from the pump and
to help provide the initial flow surge necessary when lateral (or longitudinal) forces are
suddenly generated. Fluid pressure is applied to a control valve at each front suspension
and to a single valve at the rear axle.
It will be noted that, in the diagram, the front struts are integral with their gas spring/
damper units, whereas the rear ones share a centrally mounted spring with its own damper.
At the front, the struts are single-acting but at the rear they are double-acting, the smaller
areas beneath their pistons being connected to the tops of the diagonally opposite front
struts. Since the rear control valve is insensitive to lateral weight transfer, the pressure
above both pistons remains constant during cornering, but the underneath pressure in each
is modified to that of the inter-connected front strut: this modification opposes the lateral
weight transfer at the rear wheels by a couple proportional to that exerted at the front
wheels. The result is a constant ratio of front! rear roll stiffness which ensures consistent
handling qualities throughout the range oflateral acceleration. Appropriate proportioning
ofthe effective areas of the front and rear pistons enables any desired degree of under-steer
to be achieved in the interest of stability.
Although a great deal of research and development has been carried out on fully active
suspensions their use has been limited to racing cars, military and research vehicles. Even
with high response components, fully active suspensions have been criticised for harsh-
ness at high speed over rough ground, and this, together with cost and power consumption
limitations, has limited commercial exploitation.
A more promising active suspension design for passenger cars is the active roll system
based around the existing front and rear roll bars shown in Figure 8.9.8. The objective of
this system is to prevent body roll during manoeuvres without influencing the suspension
response to road imperfections. Retaining the suspension passive components, a valve
controlled hydraulic actuator on both the front and rear roll bars is used to artificially twist
the torsion bars, and thus apply a moment to the vehicle chassis to counteract that
generated by handling manoeuvres. The system is either displacement or pressure
controlled; it uses a single valve to control both actuators and receives a signal from an
accelerometer mounted laterally on the vehicle chassis. Since passenger car roll frequen-
cies are typically less than 2 Hz the active suspension dynamic response and associated
power requirements are limited. An average power consumption of less than 200 Wand
peak require of 2 kW is typical of such systems. The elimination of body roll during
manoeuvres partially overcomes the ride/handling compromise suffered by passive
suspensions and can improve handling behaviour by eliminating camber induced tyre
effects. It is likely that roll control suspensions, first available on the Citroen Xantia, will
be developed for use on off-road vehicles which suffer from excessive roll.
Hydrostatic transmissions
4-wheel hydrostatic transmissions are ideal for difficult terrain or complex manoeuvring.
Construction industry vehicles are extensive users covering from road rollers to access
platforms. An important feature is the individual wheel control, and flow divider
combiners can ensure that drive is maintained to two, four or more wheels regardless of
the terrain. Figure 8.9.9 shows one example. Where slippery conditions are likely to be met
there are several complex arrangements possible; three are shown below
The electronic solution (Figure 8.9.10) Sensors on each wheel monitor the speed
and instructs electro-hydraulic servo valves to make corrections where this differs
from the mean.
The dual motor solution (Figure 8.9.11) The two front wheels have standard
motors, but the rear wheels use specially ported motors which act as dual motors
- hydraulically independent but mechanically coupled. When front wheels slip,
pressure falls in the front wheel motors but rises in the rear; conversely when
braking is required and the motors are acting as pumps (as when going downhill)
flow is reduced to the front wheel motors increasing their braking effort.
The series motor solution - (Figure 8.9.12) This is similar to the second solution but
uses standard motors all round. In this case when a wheel slips it increases the
pressure differences across the other series connected wheel motor.
Rail hydraulics
The use of hydraulics on railways covers a number of features. For instance road/rail
barriers, train stops (buffers) and point machines. Track laying maintenance, including
cranes, also involve hydraulics. Hydrostatic transmissions are used for special purpose
low-speed drives to give good control.
FIGURE 8.9.13 - The clamp lock point mechanism [Smiths Industries Hydraulics Company]
An example of a point machine, hydraulically clamped, is shown in Figure 8.9.13. The

idea is that the unit moves, locks and detects both open and closed switch positions in a
smooth positive sequence; it is operated remotely and powered from standard AC or DC
voltages. (The unit consists of two clamp lock and detector mechanisms, a coupling bar,
two single acting hydraulic actuators with a centre mounting thrust bracket, and flexible
hose connection to a separately mounted electro-hydraulic power pack.) The mechanism
saves using long vulnerable connection rods, and works in 2.5 s with an operating force
of 3.3 leN.
See also chapter on Mobile and agricultural hydraulics
669
HYDRAULICS IN AEROSPACE
THE USE of hydraulics in aerospace is considerable. Not only are many control features
of aircraft operated hydraulically, but also ground support and test equipment and rigs
depehd heavily on hydraulic power. Passengers and luggage, too, are manoeuvred by
hydraulic ramps and gangways. Pilot training is also facilitated by complex flight
simulators involving rapid activation from hydraulic power.
Civil and military use of hydraulics in aircraft differ in detail because of varying design
philosophies and requirements. This chapter, therefore examines a few ofthe concepts in
which each are involved, and highlights some of the special features of military aircraft.
ARP standards (Aerospace Recommended Practice from SAE) are valuable for defining
a number of aerospace hydraulic features, for instance, ARP 4752:1994 - Aerospace-
Design and installation of commercial transport aircraft hydraulic systems.
General aircraft hydraulics

Hydraulics, with high power and high stiffness, is ideal as a medium for the operation of
landing gear and flight controls, e.g. ailerons, spoiler, elevator, rudder, etc. Small ancillary
services, such as door locking and opening, may also use hydraulics but it is the
manoeuvrability and landing control which is predominantly the field of hydraulics.
In the early days of powered flying controls applied to fairly large aircraft, input demand
seldom exceeded 11 to 15 kW. By the late 1950's 30 to 37 kW was quite common, and
represented a typical level for high performance aircraft of smaller size. Military aircraft
now see up to 750 kW.
Highly favourable power/weight ratios are attainable with system pressures of 21 - 28
MPa (210 - 280 bar); even higher pressures are also conceivable and are used as described
in a later sub-section under Military aircraft. The power/weight ratios have risen from the
early ratios of about 1.7 - 2.1 kW/kg to over 7.5 kW/kg with 28 MPa (280 bar) systems.
The very high power levels themselves, however, have added further problems, notably
the necessity for adequate temperature control.
The pressure of 21 MPa (210 bar) has been standard in most civil aircraft since the
1950's but a few, such as the Concorde, operate at 27 MPa (270 bar). Flow rates required
have also risen considerably - from about 9 to 22 Umin with earlier powered control
systems to 1236 Umin in the case of the Boeing 747 and 2045 Umin for the Galaxy.
Pumps
Variable displacement axial piston pumps are the normal choice, generally incorporating
their own valves, to reduce delivery and off-load the power to a low level, when the
pressure has reached some 90% of its maximum. Variable displacement axial piston units
are generally preferred, because of their more flexible control characteristics, and current
standard types can readily produce deliveries of 175 - 227 Umin running at 5500 rev/min,
although 140 Umin at 3700 - 4000 rev/min may be more common. The pressurised
reservoir required by the variable displacement axial piston pumps is commonly provided
by an air pressurised reservoir in ci vii aircraft; for instance the Airbus pressure is 0.35 MPa
(3.5 bar).
Systems
Actual systems used have tended to become more and more complex, basically because
of necessary system duplication on the primary controls. These cover all functions
necessary to keep the aircraft in the air - e.g. elevator, rudder and aileron controls. In the
case of smaller, low flying aircraft, where control surface loads are not excessive,
HYDRAULICS IN AEROSPACE 671
reversion to manual control may be acceptable in the event of hydraulic failure. At higher
input power levels, however, emergency conditions are normally provided by a second
system based on a separate circuit and, where possible, a separate source of hydraulic
power (including a second electric motor). Thus it is necessary to at least duplicate the
hydraulic lines, valves and actuators, although these are not seen as back-up on civil
aircraft but rather of equal status.
Triplex systems have become commonplace on larger aircraft, providing two main
systems with the third as standby. On a four-engined aircraft this may involve as many as
six pumps in a combination which provides power from at least two pumps in the event
of two-engine or two-pump failure.
Physically, duplication normally involves the use of two complete and separate
systems, each with its own actuators. (Either two tanks are required or the design must
prevent loss of fluid should a fracture occur on one side.) These systems may have separate
actuators (Figure 8.10.1) or tandem actuators (Figure 8.10.2). There is also the choice of
using both actuators simultaneously for normal operation (that is, a control powered by
both systems), or employing one ofthe pairs of actuators purely as a standby unit, normally
on open circuit, but to which backup power can be switched in an emergency. The choice
affects both the attainable rate and the percentage of maximum output which can be
achieved under emergency conditions.
Thus with a dual system (both systems working simultaneously under normal opera-
tion), characteristics may take the form shown in Figure 8.10.3. With a back-up system,
the operating characteristics may be of the form shown in Figure 8.10.4. Again, depending
on the specific arrangement installed, which will finally decide the most suitable
configuration for the performance envelope required, there is flexibility of choice.
Landing gear
There are two basic types of landing gear - cantilever and articulated, although the
FIGURE 8.10.5 - Landing gear for the A340 Airbus [Dowty Aerospace].
cantilevered type is the most common because it is cost and weight efficient. (Articulated
gear is used for aircraft with low ground clearance or where the storage room is limited;
however, they do provide a smoother taxi ride over bumpy runways and have easier
maintenance.) Figure 8.10.5 shows one of the largest cantilever landing gear designs
involving electro-hydraulic components. Valves and actuators are needed for the retrac-
tion and extension of the main and nose landing gear for both normal and emergency
conditions. Landing gear for all aircraft have similar requirements but are dependent on
the likely speed of landing and the load of the aircraft, indeed their extended height may
vary from less than 2 m to over 4 m and has to support from less than 20 tonnes to current
designs being considered over 450 tonnes - the A340 Airbus being about 260 tonnes
Military aircraft hydraulics

Intensive research in the US has indicated that 55 MPa (550 bar) may provide a more
efficient system without any major redesign of systems. However, as yet, the maximum
pressure used, and even then only a few special projects, has been 35 MPa (350 bar), with
27 MPa (270 bar) being more common. The desire to operate at higher pressure comes
through a need to reduce the size and weight of hydraulic components particularly for
fighter aircraft where thinner wings require a more compact component. There are several
problems with much higher pressures, such as
Fluid viscosity
Fluid pressure surge
Heat generation
Maintaining actuator stiffness.
Fluids need to be non-inflammable and compatible with the seals which have to be used.
This limits the possibilities although a fluid based on a low-molecular weight polymer of
chlorotrifluoroethylene (CTFE) may be possible. Heat generation is greater with higher
pressures, and, with less space, cooling may not be adequate; complex cooling systems
may be required, which rely on both air and fuel heat-exchangers to maintain the hydraulic
fluid at acceptable working temperatures. Actuator stiffness is changed because of the
lower quantity of fluid which is used in the smaller pistons; this causes a weaker fluid
'spring' and a considerable drop in the system's natural frequency.
In the 1970's military aircraft of the fighter/bomber category required a hydraulic
system output capability of around 75 kW. As designs have advanced since then,
considerable changes in requirement have occurred. For instance, in the case of variable-
swept wing aircraft the demand approaches 150 kW, with about 70% of this power utilised
for sweeping the wings. Size also plays an important part - thus in very large military
aircraft, up to 750 kW may be made available for the primary and utility systems (e.g. C5A
Galaxy), and even a smaller fighter aircraft such as the YF-22A requires 600 kW.
In military aircraft the pressurised reservoir is more commonly provided by engine air
bleed at 0.7 MPa (7 bar), or by a detensified reservoir pressurised from the high-pressure
supply. An example of this is shown in Figure 8.10.6, utilising what is commonly known
as a 'bootstrap' circuit. See further detail in the chapter on Reservoirs and tanks .
All controls requiring high-force, high-speed applications need the accuracy of the
electro-hydraulic capability where self-contained hydraulic units, controlled through
digital electronics, provide the best matching of performance and application. The EF A
(European Fighter Aircraft) involves considerable feedback from sensors to microproc-
essor control ensuring that aerodynamic stability is maintained whatever the role.
Helicopters when landing on aircraft carriers also need to have a means of securing them
to the deck, and for this a 'Deck-lock' is fitted to the helicopter allowing the helicopter to
rotate through 360". (Figure 8.10.7)
Some aircraft are able to integrate the landing gear hydraulics with steering and braking.
Figure 8.10.8 shows the circuit diagram for this multi-function operation. This has the
advantages of shorter flow paths with external connections (and hence leak possibilities)
reduced to a minimum. Maintenance and compactness are also improved.
On military aircraft, hydraulics are also used for the control of gun mounts to provide
a constant elevation and azimuth, and helicopters involve a refined landing lock system
for aircraft carrier landings.
contributions from pneumatics. However, it is primarily used for those applications where
either hydraulic functions on the aircraft have to be tested, or where power is needed for
lifting, jacking and steering. Hydrostatic drives are also used occasionally.
Mobile test stands

Service trolleys provide a power source and hydraulic fluid flow over the range of
pressures expected on the aircraft. Flow needs to be stable and hence pulse attenuators are
usually fitted immediately downstream of the pumps (this also reduces noise levels). It is
highly important to have the highest level of cleanliness of oil in the system, and hence fine
2 11mfilters are fitted and there is regular moni toring of this oil (Figure 8.10.9)
Mobile facilities are also used for testing components such as hydraulic pumps, motors,
valves and actuators.
FIGURE 8.10.9 - The RAP's MK XI hydraulic service trolley (HST) being monitored
[UCC International].
Testing rigs
Testing of aircraft components at every stage of assembly is vital for maximum reliability.
Flushing rigs are also used during the assembly process and the next stage not permitted
to take place until the approved level of cleanliness has been achieved. For most large scale
repetitive applications a ring main would be used. Figure 8.10.10. shows a circuit for the
flushing of aircraft wings.
Personnel and baggage gangways

Mobile gangways for passengers at airports are common place and have, conventionally,
been operated electro-mechanicaIly. However, the use of hydraulics for both small
airports and large, has shown certain advantages, particularly where high volume is
anticipated. An example of a smaller user is shown in Figure 8.10.11 with a passenger
stairs.
Another application concerns the wheel driven tubular passenger structure needed at
large airports. This has to be driven quickly and accurately and raised to the correct height
to match the aircraft door. By using a single electric motor driving a double axial piston
pump and attached gear pump, all the facilities are covered and controlled with precision.
The two axial pumps independently drive axial piston motors at the wheels (at a maximum
of 28 MPa (280 bar)) and the gear pump supplies the power for the swivel rotation and
height (at 16 MPa (160 bar)). The unit is shown in Figure 8.10.12.
FIGURE 8.10.12 - The drive and control unit on the undercarriage of a telescopic gangway
[Mannesmann Rexroth].
Pilot training
Flight simulators are a major component in pilot training. Simulators invariably require
six degrees of freedom and hence six cylinders are required in order to achieve any
combination of movement. For the system to be as realistic as possible the cylinder action
must be as friction free as possible, and hence hydrostatically supported pistons which
float on a fine film of oil to a high cleanliness level are used.
See also chaoter on Tubes & Pipework Calculations and Hydraulic cylinders.
679
MARINE HYDRAULICS
HYDRAULICS ARE used extensively in marine applications. A small marine survey

vessel, for instance, may have no fewer than fifty separate units of hydraulically driven
machinery including propeller and thruster drives, steering, cranes, davits, winches, etc.
Passenger ships and naval vessels will have even more.
On large commercial ships many functions originally operated by steam or electric
motors are now also commonly powered by hydraulics. These functions include:
Valve control - VLCC, ULCC and Product Tankers, OBO, Bulk Carriers, LNG!
LPG;
Hatch covers - OBO, Bulk Carriers;
Loading ramps - Ferries, Ro-Ro;
Deck machinery - All classes;
Cranes - Bulk Carriers and Specialised Ships;
Fire pumps - All classes;
Lifts - VLCC, U LCC, OBO;
Pumps - Specialist and refined Product Carriers;
Deepwell pumps - LNG, LPG, clean Product Carriers.
Marine system pressures

System pressure varies widely depending on the availability of suitable components
meeting the regulations of the Classification Societies. Generally open circuit systems
operate between 15 and 25 MPa (150 to 250 bar), and closed circuit up to 35 MPa (350
bar); this helps to minimise unit sizes and reduce flows. There are, however, advantages
in using low-pressure pumps and motors on certain types of ships since the components
can be made more rugged and running and maintenance costs are lower.
Pumps
A wide range of gear pumps, piston pumps, screw pumps and main pumps are used in
marine hydraulics ranging from very small sizes (e.g. to operate small cable winches) to
quite large units for working deck machinery. Vane pumps appear to predominate in the
smaller sizes. The main hydraulic pumps are commonly driven from the ship's engine.
Other pumps may be driven by an auxiliary engine or electric motor.
Actuators
Hydraulic motors are widely used as actuators in marine hydraulic systems and provide
a direct solution where a continuous rotary output is required. Semi-rotary actuators are
used in preference to mechanical linkages where limited rotary output movements are
required (e.g. in steering units). Linear output motions are normally derived directly from
hydraulic cylinders.
The choice of seals is particularly important. Sometimes the actuator is mounted
directly on to a valve which does not have a top seal, i.e. the actuator bottom seal is the
valve top seal. This seal must then be compatible with whatever cargo or ballast may be
carried. In refined product carriers this can present a problem. Thus, it is always wise to
mount, in this instance, an actuator having a twin seal arrangement and a vent between
seals. This will ensure that should one or other seal fail at any time there is no fear of
contamination of either product or hydraulic fluid.
Choice of actuator depends on the type of equipment to be operated, but in the case of
large valves used on modern tankers, choice may not always be obvious. The following
Table 8.11.1 is a general guide:
In general it requires a greater effort to open a valve than to close it. This is particularly
true of the wedge gate where the wedge is driven into its seat until it is sealed. As the sides
of the seat grip the gate a high degree of force is required to break it out again.
Butterfly, ball and bulkhead valves show this characteristic to a lesser extent. In the
closing operation it is essential that the moving element is able to travel far enough to
ensure tight sealing. The wedge gate travels until the valve resists the closing thrust exerted
by the actuator. In the butterfly valve there is normally some positive mechanical limit to
travel, set so that the paddle is then in its fully-sealed position. Before reaching this point,
the seat resistance may have been allowed to stall the actuator.
The ball and bulkhead valves rely primarily on accurate positioning of the moving
element to seal tightly and this is achieved with mechanical stops.
To stop an actuator in the fully-open position of the valve, either a positive mechanical
stop is used or the hydraulic power supply is cut off at a predetermined position in the valve
travel. The mechanical stops may be incorporated in the valve itself, in a gearbox mounted
on the valve, or in the actuator. Some degree of adjustment during commissioning is
generally required.
Pipes and fittings

For marine hydraulics the use of carbon steel (and even some types of stainless steel) pipes
MARINE HYDRAULICS 681
is ruled out because of corrosion problems. Any such systems will normally fail within
three to five years, ifnotearlier. In many combined systems, where cupro-nickel, tungum
or corrosion-resistant stainless steel (properly painted) has been used, correctly protected
heavy-duty carbon steel accessories have been shown to be good for about 10 years but
generally require replacement at that time, the pipe tolerances usually remaining within
reasonable limits. The most successful piping systems are those which use cupro-nickel
or tun gum (an alpha-brass alloy), and where possible all pipe to pipe connections are made
by brazing or welding using sleeve couplings.
Wherever the piping has to be connected to equipment, then the use of pipe accessories
and couplings using the compressible O-ring concept show the best results, since no
deformation of the base pipe material is caused and all hydraulic sealing is by means of
a flexible oil-compatible rubber ring. Such fittings in carbon steel, stainless steel, cupro-
nickel and tungum are available, all but the first one being expensive on a first-cost basis,
but necessary if very costly delays are to be avoided due to pipe system or connection
malfunction.
Given the simplicity of actuators mounted directly on the cargo or ballast valve, or
hydraulic motors on pumps, submerged in the tanks of the vessel, many vessels have
extensive amounts of hydraulic pipe in the tanks and here cupro-nickel or tungum is very
acceptable. Even with the inerting of tanks and ullage spaces, which should reduce the
corrosion rate (by reduction of the amount of oxygen in the atmosphere), the sulphurous
vapours of many cargoes attack the exposed piping underneath the deck and it is
recommended that such pipes be properly sheathed for 4 or 5 m below the deck
penetration. A very satisfactory alternative for valve control is to use reach rod drives.
A simple method of providing a more reliable hydraulic system is to design the complete
system so that an absolute minimum number of hydraulic fittings and lengths of hydraulic
pipe are required by integrating all the hydraulic equipment. This can be achieved by
'blocking' or 'manifolding' hydraulic equipment so that all the control functions are
simply carried out inside the block with only a pressure and return line being required at
the control station.
Steering
Whilst steering consists of basically using the ship's rudder linked to a set of double acting
linear actuators, the torques experienced may be as high as 1,000,000 Nm. Not only has
this effort to be supplied, but it has to be present with sufficient accuracy. Hydraulics,
through proportional control, provides the reliable and exact positioning required, and it
can be interfaced with the ship's control systems and auto-pilots (closed loop design).
A typical electrical control system for hydraulic steering is shown in Figure 8.11.1. This
is a high-pressure system with hydraulic pressure supplied by two electric motor-driven
pumps. Two completely independent electric control systems are present. The main
system provides synchronised movement in the manner of a proportional servo-system.
The standby system is push-button operated - one button for port and one for starboard.
Movement in either direction depends on the time a particular button is kept depressed.
The actual resultant rudder angle is displayed on a rudder position indicator.
One servo-system is shown in Figure 8.11.2. Again, this is a high-pressure system
supplied by a constant delivery pump which runs continuously. The pump delivery, and
thus the rudder transit time, is determined by the setting of a flow control valve with both
coarse and fine adjustments.
The three-position open-centre control valves are connected in parallel to a pair of

double-acting cylinders through pilot-operated check valves. The master and standby
control valves are solenoid operated from the electrical system. If the master valve fails
it is automatically isolated and the standby valve brought into the circuit. If the electrical
system fails the third valve (VI) becomes the control valve (manually operated).
Feed-back is generated by potentiometers at the wheel and rudder head, generating
signals which are fed to the switching unit from which the error signal is derived. Wheel
steering is normal but push-button controls may also be provided. Variable displacement
pumps may also be used.
Mooring
Steering when docking can cause considerable problems to large vessels, or smaller ships
in narrow rivers, etc. unless additional control is included. Thrusters are one solution;
these are basically marine propulsion units, with propellers, fitted below the water line on
the ship's hull. The propeller shaft in each unit is driven by bevel gears through a column
rotated by diesel engine, electric motor or hydraulically. The shaft may also be driven
directly by hydraulic radial piston motor. Figure 8.11.3 shows a port and starboard thruster
either side of the main propeller. Where hydraulics are used, the ease of control is
particular noticeable when a closed loop transmission system is used and electrically
controlled from the bridge - variable speed is available in either direction. Hydraulic
thrusters can easily be made to be retractable to cut down on drag during normal cruising.
Deck machinery
Other mooring facilities operated by hydraulics are better described as deck machinery.
They include anchor winches and capstans. Automatic mooring winches normally employ
radial hydraulic motors and can be rendered in almost any size and capacity in neat and
compact form. On larger vessels the pump is usually engine-driven (and thus located in
the engine room), although an electrically driven pumping unit may be preferred in some
installations. An electrically driven pumping unit has advantages for working deck
machinery and similar services which may be required when the vessel is stationary since
the main engines do not have to be run to provide hydraulic power.
Hydraulic windlasses for handling anchors are made in all sizes from gear designed as
small yacht equipment up to the largest sizes required for ocean going vessels. They are
produced both as independent units with their own hydraulic pump (electric motor-driven
or engine-driven) and as pure hydraulic units for coupling to the main hydraulic system.
In the latter respect it should be noted that in a 'complete' hydraulics system the demand
for many of the services will seldom, if ever, be simultaneous. Thus the system power level
can be proportioned to meet the maximum simultaneous demand likely to be encountered.
In many cases it may be possible to reduce this demand to that of the single system with
the greatest demand (for example, certain services like power-operated steering will only
be required when the vessel is under way; and other services, like windlasses, when the
vessel is stationary).
In the 'all-hydraulic system', or where a number of individual hydraulic services are
operated from a common source of hydraulic power, an accumulator can materially reduce
the size of power plant required and also provide 'reserve' power to meet peak demands,
such as an unusual set of circumstances calling for a simultaneous demand greater than the
design maximum operating level. This can apply particularly when a variety of deck
machinery may be hydraulically operated off a common system.
A typical scheme for an hydraulic drive for deck machinery on a medium size fishing
vessel comprises a double pump driven directly by the main engine and feeding four
separate hydraulic motors distributed over the deck. Each service is independently
selected and controlled and can operate simultaneously, if necessary. Additional services
can be added, or other combinations used, within the power-output capabilities of the
pump.
A series system is shown in Figure 8.1 1.4 where a single pump is used to operate all the
deck machinery, each machine being fitted with the same size of hydraulic motor. A
second pump is incorporated as a standby or can be run simultaneously unloaded and
brought in only when peak power demands have to be met. Advantage can be taken of the
normal 'non-simultaneous' service requirements to keep the size of the pump and prime
mover down to a minimum.
All the units in this circuit are connected in series via a ring main and each motor is
controlled by a four-way combined input bypass valve. When this valve is operated the
required amount of fluid can be diverted through the motor before the main flow is passed
on to the next unit in the series connection. Each motor can be operated independently of
the others and the pressure drop across each is proportional to the load on it. Normal
maximum 'simultaneous' demand would be from the three winches, representing the peak
system demand, and the pump size would be selected accordingly. A smaller pump may
be specified on the basis that only two winches are likely to be used simultaneously (or any
other two services together), when the standby pump could be run off-load and brought
in should circumstances call for the operation of three services simultaneously.
Hatch covers
Many cargo ship designs employ hydraulically operated hatch covers, the specific
advantage being the very considerable saving in time compared with manual labour to do
the same job. System design is specific to the size and type of hatch and the loads involved,
but can be met with the simplest mechanical linkage connected to hydraulic jacks of
suitable proportions.
Power demands for the actual operation may be relatively high, but since this service
is only required at infrequent intervals accumulators are widely favoured as the power
source. Accumulator size can be selected to cover a complete cycle of operation with
sufficient reserve for emergencies, and then re-charged during the long idle periods of the
system by a very small pump and motor.
Stabilisers
Stabilising fins are now an accepted feature of ship designs for roll damping and are
employed on naval as well as commercial passenger-carrying vessels. Stabilisers differ
from bilge keels or similar fixed devices in that they are rotated to produce positive or
negative 'lifting' forces, with their motion governed by a master roll controller (usually
a gyroscopic roll detector). The necessary motions of the stabilising fins are almost
invariably produced by hydraulic power. Pumps are used which give both variable and
reversible delivery and these are controlled through a powerful servo-system capable of
giving full effect to control impulses in a fraction of a second.
The advantage for passenger vessels is a much smoother ride, and obviously on all
vessels crew prefer a steadier environment. The reduced roll also reduces fuel consump-
tion due to improved propeller efficiency.
An example at the larger end of the scale is the 2000 passenger ship 'Oriana' (the P &
o flagship), which uses two stabilisers with an areaof21.5 m2 These giant fins (supplied
by Brown Brothers & Co of Edinburgh), which can be extended from the ship's hull when
required, continually move to reduce the ship's roll even in severe conditions and thus
allow the vessel to maintain its cruise speed in all conditions. The stabilisers are controlled
by an on-board computer connected to a full proportional hydraulic closed loop system
involving Denison fixed displacement vane pumps and proportional valves.
A similar system is available for smaller vessels and this is shown in detail in Figure
8.11.5. It was produced with the co-operation of Saunders Roe (Anglesey) Ltd, William
Denny & Brothers Ltd, Brown Brothers & Co Ltd, Muirhead & Co Ltd, and Automotive
Products Co Ltd, who are responsible for the electrical and hydraulic systems.
As will be seen from the drawing there are two fins which are extended when in
operation and retracted when not in use. The main fin is oscillated by power through an
angle of 40 and the tail fin automatically travels through a total of 1000 through a simple
shackle device. Any tendency of the ship to roll is resisted by the forward motion of the
ship acting on the fins, which are turned in such a way as to correct it. A roll to starboard
wo,uld be corrected by turning the starboard fin anti-clockwise (looking from the outside)
and the port fin clockwise.
A watertight box A is attached to the hull structure and to a hydraulic cylinder H which
is pivoted at its upper end. The fin unit consists of the fin F, a crosshead X which runs in
slots in the box A to transmit the turning motion and a double-acting piston operating a
cylinder H. When extended, the crosshead makes ajoint with the seating at the bottom of
the box A, and by so doing all fin loads are transferred direct to the ship structure.
The fin assembly, including cylinder and water-box, is oscillated by the hydraulic jack,
which is connected to the lever L. At the upper end the control valve spool is attached to
a floating lever, one end of which is connected to an arm on the upper end of the cylinder
H and the other to the gyroscopic control.
Sailing ships
Even sailing yachts and ships can make excellent use of hydraulics. One example is
steering by means of an autopilot unit, shown in Figure 8.11.6.
Another concerns the reefing and furling of mainsails on large yachts which is such a
complex and heavy operation when undertaken manually; one patented method involves
the use of an hydraulic system which draws the sails automatically into the boom. In this
example two lines of rubber rollers, hydraulically clamped, rotate together as they grip the
sail- to draw in or release against a vertical tension. Figure 8.11.7 shows the basic idea.
Two hydraulic power packs provide the rotation and tensioning as required.
And, of course, when the wind stops, instead of the cradle breaking, an auxiliary motor
powered by hydraulics can drive even a 500 year old (replica) sailing ship - John Cabot's
Matthew (Figure 8.11.8). In this case a Caterpillar diesel engine drives tandem Linde
pumps, which in turn drive two propellers (out of sight!). The thrust from each propeller
is 3.5 kN at a maximum speed of 750 rev/min. which produces 6 knots on a calm day -
similar to a 22 knot wind and all the sails out.
FIGURE 8.11.8 - The sailing ship Matthew powered by hydraulics [Linde]

689
OFFSHORE AND SUB-SEA
OFFSHORE AND sub-sea operations experience extreme and hazardous environments.

Not only is there the corrosive nature of sea water, there is also the loading from storm
forces and the pressure developed many kilometres below sea level (1 km"" 10 MPa (100
bar)). Temperatures, too, may vary considerably in offshore work.
Hydraulics, suitably designed or adapted, are involved in a variety of operations. These
include
Oil and gas rig platform build
Drilling
Remotely operated vehicles (ROY) used underwater
Offshore and underwater cranes
Sub-sea trenching and pipe laying
'Roughnecks' (hydraulically powered torque and spin for pipes)
Deck and rig machinery
Water operated hand tools
Fluids
There are two major problems as regards the fluids used in offshore work. One is the low
temperatures experienced and the other the possibility of environmental pollution.
Obviously the optimum fluid (if above freezing), and considerably cheaper to use, would
be the surrounding sea water; with suitable de-salination this may well take on in the 21st
century; its problems currently are
Low viscosity - giving leakage and erosion
Poor lubricity - wear
Saline medium - corrosive
Highly contaminated - silt, solids, micro organisms
Electrically conductive - possible short circuits
Some progress is being made with seawater, and various research projects are in progress
to produce pumps (possibly with separate mineral oil lubrication) and valves to operate
at up to 25 MPa (250 bar) at depths of 6 km (60 MPa (600 bar) ambient pressure). One
special design of a sub-sea val ve actuator is being considered at well heads of offshore oil
and gas wells. The advantage here is enormous because of the extremely high cost of a
supply line to carry hydraulic fluid, perhaps over several kilometres (costs over several
1000' s per metre have been quoted). In this particular example a multi-stage centrifugal
pump, operating at the well head, supplies unfiltered seawater at 1.7 MPa (17 bar) to a
single acting spring return cylinder (Figure 8.12.1).
Mineral oils are currently more often used, but they are not environmentally friendly.
That leaves, as a reasonable compromise, water based fluids with a suitable anti-freeze
content such as glycol. Indeed, water glycol with no other oil content, is the preferred
current hydraulic oil.
Components
All components used in seawater need to be protected from the corrosive effects of the
water. Plastic coatings can protect the external surfaces but the internal features need the
addition of specially hardened and corrosi ve resistant surfaces such as provided by plasma
spraying, sputtering or other weld deposition techniques (e.g. TIG or MIG).
Seals, too, need to be compatible with the seawater and with sunlight and possible
weathering effects including ozone and temperature. Sub-sea use requires an ability to
withstand high ambient pressures.
Platform build
More correctly this is the exploration of suitable positions, piling operations and the
putting together of the various components of the platform in situ. This tends to require
a semi-submersible construction plant able to winch, clamp and jack-up the items
presented to it, as well as perform numerous other small functions. One example is the
OFFSHORE AND SUB-SEA 691
MSV Tharos (Figure 8.12.2) which is powered by a pair of Volvo Penta 7 L turbocharged
diesel engines each developing 139 kW at 2200 rev/min. Each engine drives a Poc1ain
pump with six separate banks, and a Hall twin segment pump.
Drilling
The drilling and insertion of pipes in the bore holes require as steady a situation as possible.
However, North Sea weather conditions are not often calm and hence some means of
stabilising the drilling block by means of compensators is necessary. Indeed all hung or
hook supported equipment needs to have a compensated link between the platform (which
may be moving) and the item suspended (which should be stationary). A pair of single
acting cylinders, mounted at 90 in a Vee arrangement, act against a spring force of high
pressure air held in air-pressure vessels. Any movement of the platform-connected
hydraulic cylinder forces the air in the pressure vessels to alternately expand and
compress, thus enabling the support (hook) to maintain a relatively stationary position.
The cylinders used in rig compensation can provide strokes up to around 8 m with a force
of almost 3 MN, which enables drilling to continue in all but the most severe weather
conditions. (The pressures in the system at 20 MPa (200 bar) are usually adequate although
they are the upper limit where water glycol is used. Higher pressures need mineral oils.)
One of the earliest uses of hydraulics was in the preparation of the pipes, with hydraulic
pipe spinners and drill collar handling.
A range of downhole pipe handling and drilling features are combined in the top derrick
drilling machine shown in Figure 8.12.3. Controlled rotation is combined with pipe
handling to perform all tripping, running, casing and drilling functions. While, for pulling
out of a downhole or going into it, a swivel coupling enables the pipe handler to be freely
rotated while air and oil lines remain linked up.
Roughnecks
The 'roughneck' is an hydraulically powered machine designed to torque up and spin out
the drill pipe and drill collars. This enables the total drilling operation of make-up and
break-out sequences for the drill pipe to be performed automatically. The torque wrench
is in two parts, one above the other, rotated through 30 by two hydraulic cy linders working
in contradirection.
Pipe laying and moving

Pipe laying requires very accurate pipe tensioning, and special pipe-laying vessels are
used with hydraulic tensioners.
Remotely operated vehicle (ROV)

Much of the under-sea operation is unsuitable for human presence due to the depths
involved. For this work specially designed ROV's (sometimes termed remote guidance
vehicles, RGV' s) are used, supported by appropriate surface vessels. Equipment has to be
able to withstand the seawater and the external pressures involved and, in particular, shaft
sealing may have to be up-graded for depths over 1 km.
A major feature of the ROV is that of being able to position itself accurately; two Linde
56 kW hydraulic thrusters have been found to be able to give 300 mm horizontal
tolerance and 5 heading at a depth of 150 m. (See Figure 8.12.4.)
Another example is shown in Figure 8.12.5. In this case similar Linde pumps driven by
37.5 kW electric motors are used at depths of lOOOm as part of an Offshore Systems
Engineering Ltd electric hydraulic unit which is used totally unprotected from the sea. The
pump powers the thrusters and robots of the remotely controlled hydraulically operated
subsea work vessel. The vessel manoeuvres at 3 knots using seven hydraulically powered
thruster propellers, each thruster controlled by a system of electro-hydraulic servo valves.
Deck machinery
Deck machinery includes winches and cranes as well as other handling facilities, all of
which can be operated hydraulically. Hoist, slew and luff motions may be powered by
axial-piston pumps driving through radial or axial piston motors within a closed loop.
695
MINING AND INTRINSIC SAFETY
IT HAS been estimated that about a quarter of all UK production of fluid power equipment
has been used in mining in the past. However, the decline in coal mining and the advance
in other industries has seen this proportion dropping in recent years. Nevertheless, mining
still is a major user of hydraulics because of the higher output powers and forces available
from the more compact hydraulic equipment.
Underground mining has a number of distinctive requirements not seen in other fields.
For instance, the difficulty of assembly and maintenance oflarge machinery. There is also
the dirt laden atmosphere which is very conducive to contamination of hydraulic systems.
The most important requirement, though, is that of the need for intrinsic safety, entailing
special electrical circuitry and fire resistant fluids.
Mining in this chapter refers not only to coal mining, or exploration for other minerals
inland, it also relates to drilling and tunnelling below ground, for whatever reason.
Fluids
Hydraulic fluids in most mining applications require to be non-inflammable. There are a
number of types available (see chapter on Types of fluid) but one of the major fluids used
in underground coal mining, for powered roof supports where some leakage is expected,
is the HFA dilute emulsion (95% water, 5% solcenic fluid). It is much cheaper than other
fire resistant fluids, but it does require special valves and pumps.
For mobile machines and rotary transmissions, the HFB invert emulsion (40% water,
60% mineral oil) may be employed, permitting the use of most standard industrial
hydraulic components, although pumps and motors must only be of types able to tolerate
the low lubricating properties.
Component and circuit design

Reservoirs
Reservoirs in mines should be fully enclosed, with a filtered breather, and preferably the
refilling faces on the sides so that fluid must be pumped in under pressure, since experience
has shown that dirt is often introduced during open pouring. Another complication is that
mixtures are supplied in their two parts, and it can be quicker to pass 100% water into the
tank than the correct 95% mix unless a premix arrangement is included. Reservoirs should
696 THE HYDRAULICHANDBOOK
contain sludge weirs and magnetic filters. Underground reservoirs should also be fitted
with safety devices such as low level and high temperature cut outs. Care should be
exercised on the type of painting or treatment used in reservoirs to avoid any possibility
of flakes of 'paint' falling into the fluid after long use.
Pumps and other moving components

The pump suction is provided with a relatively coarse strainer, often ofthe rotary cleanable
type. Because of the need for reservoirs to be alongside the pump, the inlet pressures at the
pump may sometimes be below atmospheric causing cavitation unless great care is taken
to ensure the reservoirs are kept close to the maximum level, or boost pumps are fitted.
Exposed mechanical sliding surfaces are very liable to damage and corrosion. Where
possible, rams are designed so that the rod is retracted in the rest position. Plating or hard
chrome is employed on rods. Gaiters are sometimes used but are rather subject to
mechanical damage. On surfaces, such as ram barrels, which are wholly immersed in the
fluid, surprisingly little corrosion occurs even with water-based fluids, and bare steel is
satisfactory. More critical components such as manifolds and valve bodies are often
phosphated, whilst their moving parts are usually of stainless steel.
Filters
The finest filter in the system, down to I5 11m, is usually a replaceable element unit placed
in the pressure line immediately following the pump where the pressure loss it causes will
not be important. Local filters protecting sensitive components such as servo valves are
also used. Because of the difficulty in replacing filters, large capacity filters should be used
and a regular maintenance programme enforced.
Circuits
A typical underground circuit showing the mixing of the fluid, power pack, and supply to
a coal-face powered support installation is shown in Figure 8.13.1.
MINING AND INTRINSIC SAFETY 697
Underground drilling equipment

The use of hydraulics allows a greater advance in mechanisation. Booms have been
designed where both the boom and the feed can be swung quite freely. All the movements
of the boom and feed are actuated by hydraulic cylinders. Unlike air, hydraulic oil is
relatively incompressible, so that it is possible to position the feed or boom to drill a hole
at a pre-set point and in a pre-set direction. The hydraulic fluid remains trapped between
the valves when the position is fixed.
Hydraulic cylinders have the advantage over such mechanical alternatives as long
screw feeds, in that these are exposed to wear and disruptions. The hydraulic systems of
booms are normally powered by a compressed air motor, although the rig may also feature
an auxiliary diesel or electrically driven hydraulic pump, so that the booms can be swung
when the rig has to negotiate curves.
An important advance over simple positioning by means of hydraulics is 'parallel
holding', whereby a feed can be kept parallel to its original position when moved vertically
and/or horizontally. Here hydraulic oil is transferred from one cylinder to another to
compensate for the upwards, downwards, or sideways movement of the boom and keep
the feed at the same angle to the tunnel face. On most tunnelling and drifting rigs the feeds
are also hydraulically powered.
In a single pass, drilling can cut through many different materials ranging from soft sand
to hard granite, and at each new level a different torque is required on the main hydraulic
motor. These changes from one level to another generate shock loads and backlash which,
in deep passes, can cause a string of drill pipes to twist, especially when the bit is cutting
through tough materials. Then, as the bit hits a softer strata, the pipe suddenly untwists,
sending huge pressure surges back into the rig's hydraulic system. Pressure spikes to over
80 MPa (800 bar) can be experienced which can cause major failure in the rig. Denison
Hydraulics hydrostatic pumps monitor the hydraulic system and rapidly de-stroke the
pump (in 50 ms) preventing pressure overshoot more than 10 %, and the performance is
unaffected by the inherent pressure spikes.
Hydraulic pit props

A typical prop is shown in section in Figure 8.13.2. Basically, it embodies two steel
cylinders fitted one inside the other, which can be extended by hydraulic pressure derived
from a pump incorporated in the assembly. Provision is also made for quick release of
pressure, whilst a relief valve in the hydraulic circuit also ensures that when the hydraulic
pressure in the prop exceeds a specified figure the prop yields.
As the diagram shows, the main or pressure cylinder is enclosed by a guard tube. The
inner cylinder, which slides in the pressure cylinder, forms a fluid reservoir as well as
containing the pump and valve mechanisms. The pressure cylinder and inner cylinder
operate together as a hydraulic ram, with a main bearing located at the open end of the
pressure cy linder. This bearing also forms part of an outstop should the prop reach the limit
of its travel. Included in the bearing assembly are a metal scraper ring and wiper ring to
prevent dirt and water from entering the pressure cylinder.
A piston head is welded to the bottom of the inner tube and carries the pump cylinder,
the main release valves and the relief valve capsule plus a gland ring which forms a high
pressure seal, an anti-extrusion ring and a piston ring. Valves and rings are retained by a
detachable plate.
A control shaft is mounted transversely in a housing at the top of the inner tube, linked
via a crank to a connecting rod to the pump and a push rod to the valve assembly. An
oscillatory motion imparted to the control shaft via a suitable handle operates the pump
for setting the prop in position. Lifting the handle right up past the normal suction stroke
position first compresses the spring in the valve assembly and then operates the relief valve
mechanism for withdrawing the prop.
The pump is a single acting type with the pump cylinder projecting from the top face
of the main piston head. The piston is pin-jointed to the connecting rod and a non-return
valve forms part of the piston itself. Fluid is admitted to the pump cylinder via this non-
return valve and then forced into the pressure cylinder by the pumping action. The prop
thus extends a specific distance for each stroke of the pump, the setting up load being
established by continued operation of the pump handle. This setting up load is normally
from one quarter to one third of the maximum load for the prop. Thus once set the prop
will continue to accept increasing roof loads up to the maximum figure, at which the relief
valve is automatically operated and pressure is reduced by fluid passing from the pressure
cylinder back to the reservoir. Immediately the load on the prop drops below a slightly
lower figure, the relief valve re-seats
In principle, a powered support unit consists of several props mounted on a common
base which also contains a horizontal ram for advancing the whole assembly. A typical UK
support unit, termed a 'chock', is shown in Figure 8.13.3, connected as in Figure 8.13.1.
As with individual props, it is the trapped fluid within the legs which supports the roof, but
powering permits the legs to be made heavier, and the rated thrust per leg is from 30 to 100
tonnes in current designs.
Supply pressures up to 21 MPa (210 bar) feed the face through as much as 1.5 km of
pipe. Since the fluid is dilute emulsion, the only suitable pumps are slow speed 3- or 5-
horizontal piston types with oil-lubricated crankcase. These are fixed delivery pumps so
that an unloader valve must be provided to control the output. Usually at least two pumps
are provided to ensure reliability, the total output being about 120 Umin, depending on
the size of the chocks and the number of them to be moved per minute to keep pace with
the other mining operations.
Underground loaders
Increasing demands from all phases of mining and civil engineering have meant that
loaders are being asked both to travel longer distances than before and to carry heavier
loads. Haulage distances of 100 - 300 m make compressed air lines impracticable. Air
pressure is lost, the hose is exposed to wear and is generally difficult to handle. The
necessary increase in hose diameter aggravates these problems.
Equally, the air cylinder, used in the past to raise the bin and load, took up most of the
space between the wheels of the unit. Additional disadvantages were lack of speed and
strength in certain circumstances. Since the hydraulic system, which replaced pneumatics,
works at a much higher pressure, e.g. 15 -25 MPa (150-250 bar), more speed and power
became available and from much more compact equipment.
Once hydraulic power has been introduced for these machines, it can also be used close
to each wheel, to provide four-wheel drive as well as good manoeuvrability. The diesel
engine powers a series of hydraulic pumps; one for each pair of wheels and one for the
bucket and bin. The number of hydraulic cylinders required can be kept low by the bucket
and bin. A small extra cylinder can be added to increase the digging power of the bucket
at the critical moment.
For cutting coal from the seam and loading it out, the main drive is normally electric,
the inertia of the motors proving an advantage in coping with intermittent peak loads.
Hydraulics are used, however, for all ancillary operations such as boom movements and
for traction. Extreme compactness is required, so that proprietary pumps, motors and
valves are seldom suitable and machine builders must develop their own. These compo-
nents are often located within gearboxes so that mineral oil rather than fire resistant fluid
is used, but this is tolerable since there is not much chance of an external high-pressure
leak.
The most common long wall power loader is the shearer, of which an example is shown
in Figure 8.13.4. The 200 kW water-cooled motor drives the two cutting drums through
gear trains. A drive shaft passes through the haulage section of the machine where there
is also a power take-off to the hydrostatic transmission. This comprises a variable-deli very
swashplate pump delivering up to 1.8 Lis at 20 MPa (200 bar) coupled in a closed circuit
to the haulage motor, which in this model is of the radial piston type with a multi-lobe cam.
This drives the haulage sprocket through 3.6: I gearing. This propels the machine at up to
10 mlmin (30 ftlmin) by pulling on a fixed chain stretched along the face. Rack systems
are also in use as an alternative to chain.
The pump swash angle may be varied to control the speed and direction of travel. There
is also a torque motor sensing the load on the main electric motor so as to reduce haulage
speed in case of overloading. The haulage section casting itself forms the reservoir and it
also contains a priming pump, filter and oil cooler.
There are further pumps of 6 L/min capacity within the gearbox at each end of the
machine for powering the ancillaries - boom lift, cowl rotation and machine tilting jacks.
Other longwall machines, such as the trepanner, cut and load the coal in a different
manner but the haulage and other hydraulic aspects are basically similar.
Tunnelling
Perhaps the most captivating recent tunnelling experience has been that of the Channel
Tunnel (UK/ France) covering a total tunnel distance of 50 km (sub-sea for 38 km). In this
example special tunnel boring machines (TBM's) were manufactured just for the one
application - and then buried! They were more than 200 m long, weighed 1350 tonnes and
had an 8.36 m bore. The five Vickers hydraulic power packs, covering 650 kW power,
were
two for steering and propelling the cutting head - 3000 L,
two for placing the tunnel sections in position after boring - 1500 L,
one for positioning shuttering for grouting operations - 380 L.
Tunnelling in the channel tunnel mainly involved chalk, however, in general the
methods employed vary widely in accordance with the hardness of the rock. In the
tunnelling associated with longwall coal mining, only rocks of medium hardness are
encountered, but the need to keep pace with the relatively slow advance of the face renders
the most sophisticated machines uneconomic, and also special problems are posed by the
congestion of machinery.
One of the obstacles in exploiting hydraulic oil as a power medium in rock drilling lies
in its major difference from air - its incompressibility. Hydraulic oil can be compressed
about 0.5% under a pressure of 100 MPa (1000 bar). If a piston mechanism is employed,
the high shock pressures generated in the hydraulic fluid could break parts of the rock drill.
In this case the inclusion of an accumulator can solve the problem.
Working pressures in hydraulic systems for rock drills are about 20 MPa (200 bar) as
compared to 0.5 - 0.8 MPa (5 - 8 bar), or in certain cases up to 1.5 MPa (15 bar), in a
pneumatic system. This means that the size and weight of the rock drill can be reduced for
the same amount of impact force, or the force can be increased without excessive demands
for room and bearing power. At the same time the energy requirement for the same drilling
capacity is only about one third as much. In addition, the higher impact force is achieved
without increasing the stresses in the drill steel, which are a major factor in drill steel life.
Hydraulic pump units are smaller and can be located on the rig, facilitating changes in
pressure to match particular drilling circumstances.
With the closed power transmission system of the hydraulic rock drill, there is no noise
due to escaping 'power' and the noise level can be 10 -15 dB(A) lower thaR for an air-
powered drill. This is particularly important since the exhaust noise of a pneumatic drill
is, to a great extent, low frequency sound, which is the most difficult to muffle. Finally,
the absence of cold exhaust precludes the precipitation of humidity as mist and fog.
Freezing problems are also completely eliminated.
Ripping
In coal mining, ripping is the operation of cutting down stone to enlarge a tunnel after the
coal in the cross section has been removed. One of the types of machine used is that shown
in Figure 8.13.5. This has direct electric drive to the rotary cutting head and conveyor, but
all other motions, including the caterpillar tracks, are hydraulic. There is a triple-section
gear pump supplying a maximum of 422 L/min at 13.6 MPa (136 bar) maximum pressure.
A total cutting area of 6 m by 4 m is achievable. The machine is capable of working on
invert emulsion fluid.
'Hydraulic mining'
To avoid confusion, it should be explained that hydraulic mining means washing out a
mineral by powerful water jets. It is used for surface working of sedimentary ores and
occasionally for soft coal underground. It is also possible, with a water jet, to cut quite hard
rock such as would be encountered when tunnelling; exceptionally high pressures, e.g. 700
MPa (7000 bar), are necessary, obtained by a hydraulic or pneumatic intensifier.
Underground conveying and winching

Wherever possible the mineral product is brought out by conveyor, either to the shaft
bottom or all the way to the surface up an inclined drift. Some loading machines
incorporate a small conveyor of the scraper-chain type. These are frequently jammed by
large rocks and hydraulic drive is thus very suitable. Low speed motors requiring no
gearbox are generally chosen. 'Armoured' face conveyors, also ofthe chain type, may be
up to 300 m long with a drive of up to 180 kW at each end. Radial hydraulic motors have
been used for this purpose, offering the useful safety feature of rapid stopping due to their
low inertia, but the substantial power packs needed and the problem of synchronising the
drives at the two ends lead to this form of drive not being preferred to the standard one in
which electric motors drive through traction-type fluid couplings and gearboxes.
Hydraulics play a vital part in the discharge arrangements for underground storage
bunkers. In some, hopper doors release the mineral, and these are operated by rams fed
from a stationary power pack In other cases a very robust chain conveyor forms the whole
bottom of the bunker, and discharge is effected by dri ving the chains at slow speed. This
may be done by automatically-reciprocating rams driving through pawls or there may be
a chain sprocket fitted with a pair of hydraulic motors.
Much of the transporting of men and materials is by means of 'rope' or steel cord
haulage (belts). A stationary haulage engine usually has electric drive, either direct or
through a fluid coupling, but hydrostatic drives are being introduced for haulage where
control of speed and torque is critical, for example, for winches intended for dragging
coalface machinery into position. In order to provide long reliable life to belts it is
important not to exceed permissible belt tension, thus a suitably rapid response is required
on the hydraulic tension system - 3% accuracy of tension can be achieved by a Denison
Gold Cup pump with its 50 ms response.
A further use of hydraulics is in over-riding train brakes. The latest designs, for a fail-
safe action, are spring-applied and taken off by hydraulic pressure. A centrifugal device
is also fitted, to release the pressure in case of over-speeding.
For loading, tipping or docking, mine cars are being manipulated increasingly by
hydraulic means. Several different methods are employed for this purpose, from small
isolated units to comprehensive automatically controlled schemes.
To conserve power and prevent excessive oil heating when the cars are held stationary
by the retarder, an unloading valve is provided which cuts out the main pump when the
pressure rises to about 5 MPa (50 bar) leaving a small subsidiary pump in circuit which
blows off at 7 MPa (70 bar) through a relief valve. As soon as the retarder is released the
main pumps immediately pick up the load at the selected speed. The retarder grips the
wheel flanges between spring loaded bars to apply the braking force and is released
hydraulically.
Shaft winding
The main winding engines of a mine may be of 2000 kW or more and the hydraulic
application here is that of the braking system, which must obviously be capable of delicate
control and must be as safe as is humanly possible. Brakes are usually spring-applied and
released by a single-acting ram which overcomes the spring force. Unlike a weight loaded
brake, the system has practically no inertia and it is a simple matter to incorporate a time
lag which reduces the initial rate of retardation, although normally this need not exceed
a fraction of a second with this type of brake. In the event of pressure failure the brakes
are applied automatically.
The usual arrangement (Figure 8.13.6) has a high pressure twin-delivery pump, the
main delivery being unloaded when the 'brakes off' pressure is reached, leaving the minor
supply to maintain pressure.
Because so much depends on local conditions, hydraulic systems often differ in detail.
On sinking winding, for example, the brake must be able to discriminate between an
unbalanced descending load and the same load ascending, and also compensate for the
weight of rope paid out. These conditions will vary as sinking proceeds.
Exact positioning of the cage can be achieved by solenoid valves actuated by the
mechanical parts. As the end of the wind approaches, the first solenoid valve is released,
thus connecting the brake cylinder to a relief valve, and the pressure falls slowly to allow
the brakes to contact the drum lightly. Just before the end of the wind the cage releases a
second solenoid valve, introducing another relief valve set at a lower pressure than the first
- the position when this occurs can be timed to suit the cage speed. If necessary a third
solenoid valve applies full holding force when the cage comes to rest.
Centrifugal governors can be integrated with the hydraulic braking system and if
desired, the speed controlled over the whole cycle including whilst stopping.
Intrinsic safety
All sub-surface mining requires a measure of intrinsic safety to prevent the occurrence of
explosion. The idea is that oflimiting the amount of electrical energy that might be present
in the atmosphere, should an electrical fault develop. In practice this means the apparatus
is designed so that any electrical spark that may occur is incapable of causing ignition of
the flammable gas or vapour present in the prescribed area.
There are basically two approaches. One is to seal all hazardous electrical components
in a totally isolated container, the other, described further below, is to limit the energy
available by suitable electronic control interfaces.
The interface may either be a shunt-diode safety barrier, or galvanic isolators. These
interfaces have the ability of being able to divert any fault energy away from the potentially
hazardous atmosphere.
Shunt-diode safety barriers

These are the simplest of the two types of interface, providing a return path for the
excessi ve current. Because of their simpler construction their lives are considerably longer
than the isolators, and, if allowable, they are to be preferred where maintenance is difficult,
such as offshore. A fuse protects the Zener diodes which restrict the voltage, and a current
limiting resistor restricts the current that can be delivered when a fault occurs.
Barriers are passive devices and hence are more accurate and consume less power than
isolators; they are also generally less expensive unless additional circuitry is required in
the total circuit.
Galvanic isolators
The galvanic isolator provides an insulating barrier between the safe and hazardous areas,
and hence cut off any connection between the two. The power transfer may use a
transformer, and the return signal an opto-coupler, transformer or relay. The final power
limitation comes from the use of a diode/resistor network similar to the shunt -diode safety
barrier.
Isolators are generally safer, but such increased insurance may not be necessary. They
do, however, have greater immunity to electrical surges than barriers, and they offer better
protection to instrumentation
Code of practice
Although international standardisation is gradually occurring, Europe is at the forefront
in developing a uniform standard throughout the European Community with the conform-
ity CE mark in addition to the community mark. A number of standards for electrical
equipment are being put into statute books of the member countries and are expected to
be fully law by the year 2003. These are basically CENELEC standards and, in particular,
two are relevant here:
EN 50020 - intrinsic safety
EN 60079-10- area classification.
The mining area, with potentially explosive gases present, is currently considered the
most hazardous, and is designated Zone 0 or Category 1. But pending the finalisation of
the standards there is some confusion with Germany forbidding the use of shunt-diode
safety barriers, whereas they are permitted in the UK where a choice is made depending
on the type of project.
707
HYDRAULICS IN MEDICINE
IT IS often assumed that because electrics and pneumatics are 'clean' (and require no
reservoirs) that hydraulics has no place in medical applications. But that is far from the real
picture. True, there must be no leakage at all and obnoxious smell is anathema, but even
so, hydraulics with its inimitable power/weight ratio possibilities is ideal for heavy
applications in confined spaces. Trolleys and lifts and general movement of personnel are
ideal applications even in homes and hospitals.
Hydraulic power sources

The source of power is either a manual pump or a low voltage electric motor. The hand
operated pump is much simpler and lighter and can be operated without any connection
to an electrical supply. It is very suitable for homes and nursing homes where carers are
involved as well as those who are being cared for.
Electrically operated and controlled power packs have the distinct advantage that a
patient with the minimum of muscular movement can operate the unit themselves; thus a
domestic lift required to raise and lower a person into a shower or bath complex can be
easily operated by the person in question with no outside help. Another value of the
electrically powered unit is where very precise movements are necessary, as with an
operating table.
Figure 8.14.1 shows a miniature power pack, weighing under 4 kg, able to develop 24
MPa (240 bar) from a 12 V or 24 V source.
FIGURE 8.14.1- The Series 108 miniature power pack [Oildyne].

Lifting and moving equipment

A hand operated pump application is shown in Figure 8.14.2. Here a single stage direct
acting ram (it could also be two-stage) allows a gradual raising of a slung person, perhaps
just a few millimetres a stroke, and the fixed support of them at the required level. A
variable position unloading valve, gradually unscrewed, permits a very gentle lowering
of the hoist. Obviously the heavier the load the greater the force required on the jacking
lever; because ofthis the heavier personnel (e.g. over 140 kg) are normally only jacked
by electrical motor activated pumps.
HYDRAULICS IN MEDICINE 709
The electric activated hoist may use a self-contained linear electro hydraulic actuator
such as shown in Figure 8.14.3. This particular unit has a stroke of 350 mm and can apply
a direct force exceeding 11 kN. The performance curves for the unit are shown in Figure
8.14.4 for a 24 vDC input. The lowering speed would not exceed 12 mm/s (i.e. the
maximum lifting speed at no load). Fluid flow is up to around 4 Llmin at 7 MPa (70 bar)
and a person of 200 kg can be quietly and safely raised in 20 s. The Smiths' Autopack,
operating at 150 W, has a reversal feature meaning that the occupant is both raised and
lowered under power, for increased safety.
FIGURE 8.14.4 - The performance curves for the above actuator [SI Hydraulics Company].
Other medical items which can be controlled and manipulated by this type of actuator
are
Hospital beds
Step lifts
Chairs
Baths and bath seats
Stand aids
A major disadvantage of many lifting and stop/start functions, where invalids are
concerned, is the jolt which can be experienced. A soft function valve has, therefore, been
developed which reduces typical acceleration/deceleration values from 0.5 g to around
0.002 g (at 6 L/min, an actuator speed of70 mrn/s and at a pressure of2.5 MPa (25 bar.
The valve is an electronically controlled poppet valve providing proportional control
through electronic pulsing. The valve and a typical circuit is shown in Figure 8.14.5.
HYDRAULICS IN MEDICINE 711
Operating tables
The operating table needs to be accurately positioned, able to be inclined yet extremely
stable and firm at the set position. There is a possibility of hydraulic power here. By using
a reversible power pack the number of control valves can be reduced by half. Figure 8.14.6
shows a typical circuit diagram and the quality of hydraulics required.
Artificial limbs
One valuable contribution here has been the fitting of an Enidine shock absorber to the
artificial leg of a skier. The person involved was already a competent ski instructor before
losing one leg in an accident. An ADA adjustable rate absorber has been fitted in a
modified knee joint with a tiny motor installed to change the damping rate. The motor is
activated by a button on the ski pole handle (Figure 8.14.7).
FIGURE 8.14.7 - A hydraulic shock absorber fitted to an artificial limb [Enidine].
Exercise equipment
Hydraulics may be used as the restriction to any physical action, but it is particularly suited
to water activities such as rowing. A rowing machine has conventionally incorporated an
air dampened flywheel, a friction device or electromagnets, but by using a paddle wheel
in water, with a Sprague clutch, an almost totally realistic resistance is achieved. Rowing
is seen as one of the best forms of aerobic exercise available, because of the smooth coiling
and uncoiling of limbs and muscles with low impact forces on the joints.
Waste disposal
The transport and disposal of hazardous waste is a major headache for hospitals. Health
and Safety guidelines recommend that waste be containerised, but the containers are not
always of suitable size and, if using 1100 L wheeled bins, may be too wide for standard
doorways. Instead of using these large wheeled bins, smaller 540 L containers are taken
by trolley to a pick-up point where one operator can use an hydraulically operated
telescopic arm, with four diagonally connected suction cups, to lift the boxes and put them
on an adapted 8 m lorry. The telescopic arm attached to the lorry is run from the PTO and
all motions are hydraulically operated by a four section Danfoss proportional motion
control valve.
ADVERTISER'S
INDEX,
BUYERS GUIDE
AND
TRADE NAMES INDEX
713
ADVERTISERS INDEX
Applications Engineering Ltd, Unit 7, Ridgewood Industrial Estate, Facing p. 465

Uckfield, East Sussex, TN22 5SX, United Kingdom
Tel: +44 (0)1825 764737 Fax: +44 (0)1825768330
B.E.P. (Portsmouth) Ltd, Amside Road, Waterlooville, Portsmouth, Page 156

Hants, P07 7UP, United Kingdom
Tel: +44 (0)1705 268733 Fax: +44 (0)1705 251104
Cosappo Sri, 85034 Francaulilla SS, F2 Zona Industriale, Italy Page 124
Tel: +39973 577024/577355 Fax: +39973 577355
Denley Engineering Co. Ltd., Spen Vale Street, Heckmonduke, Page 384
West Yorkshire, WF16 ONQ, United Kingdom
Tel: +44 (0)1924 413400 Fax: +44 (0)1924410109
Fairey ArIon Ltd, Unit 29, Fareham Industrial Park, Fareham, Facing p. 341
Hampshire, P016 8XG, United Kingdom
Tel: +44 (0)1329826161 Fax: +44 (0)1329825758
Flotron Ltd, 445 Perth Avenue Trading Estate, Slough, Berks, Page 446
SLl 4TS, United Kingdom
Tel: +44 (0)1753865137 Fax: +44 (0)1753 865137
Forward Industries Ltd, PO Box 8, South Marston Park, Swindon, Facing p. 95

Wiltshire, SN3 4RA, United Kingdom
Tel: +44 (0)1793823241 Fax: +44 (0)1793828474
Guyson International Ltd, Hansen Coupling Division, PO Box 18, Facing p. 296
Otley, West Yorkshire, LS211RD, United Kingdom
Tel: +44 (0)1943 464027 Fax: +44 (0)1943850042
Hagg Lunds Drives Ltd, Foxbridge Way, Normanton, Wakefield, Page ix

West Yorkshire, WF6 ITN, United Kingdom
Tel: +44 (0)1924 220100 Fax: +44 (0)1924890111
Houghton Vaughan Pic, Legge Street, Birmingham, B4 7EU, Facing p. 340

United Kingdom
Tel: +44 (0)121 3596100 Fax: +44 (0)121 3592458
D.G. Howell (Hydraulic Engineers) Ltd, 78-84 Commercial Street, Facing p. 431
Risca, Gwent, NPI 6BA, United Kingdom
Tel: +44 (0)1633612044 Fax: +44 (0)1633601581
Hydraulic System Products Ltd, Monckton Road, Wakefield, Page 480

West Yorkshire, WF2 7AL, United Kingdom
Tel: +44 (0)1924 364748 Fax: +44 (0)1924 290450
Hydraulik Ring Ltd, Meir Road, Park Farm North, GB-Redditch, Facing p. 139
Worcestershire, B98 7SY, United Kingdom
Tel: +44 (0)1527 516666 Fax: +44 (0)1527 516777
IAHR, P.O. Box 177,2600 MH Delft, The Netherlands Facing p. 1

Tel: +31 152858585 Fax: +31 152858582
e-mail: iahr@wldelft.nl internet: http://www.iahr.nl
Institute of Filtration and Techniques of Separation (IFTS), Facing p. 634

Rue Marcel Pagnol, Foulaxronnes, France
Tel: +33 05 5395 8394 Fax: +3305 5395 6695
Kawasaki Precision Machinery (UK) Ltd, Ernesettle Lane, Facing p. 579

Ernesettle, Plymouth, Devon, PL5 2SA, United Kingdom
Tel: +44 (0)1752 364394 Fax: +44 (0)1752 364816
Linde Hydraulics Ltd, 7 Nuffield Way, Abingdon, Oxfordshire, Facing p. 578

OX14 lRJ, United Kingdom
Tel: +44 (0)1235522828
Manuli Hydraulics Ltd, Unit 26, Piccadilly Trading Estate, Facing p. 297
Adair Street, Manchester, Ml 2NP, United Kingdom
Tel: +44 (0)161 2737383 Fax: +44 (0)1612737564
Mikrospin Ltd, Mikrospin House, William Street, Littleborough, Facing p. 465

Rochdale, OU5 8JP, United Kingdom
Tel: +44 (0)1706 378625 Fax: +44 (0)1706 373988
ADVERTISERS INDEX 715
Moss Plastic Parts Ltd, Langford Lane, Kidlington, Oxon, Page 323
OX5 IHX, United Kingdom
Tel: +44 (0)1865841100 Fax: +44 (0)1865 370135
MP Filtri UK Ltd, Bourton Industrial Park, Bourton-on-the-Water, Facing p. 341

Gloucestershire, GL54 2HQ, United Kingdom
Tel: +44 (0)1451 822522 Fax: +44 (0)1451 822282
Parker Hannifin Pic, Parker House, 55 Maylands Avenue, Page vi

Hemel Hempstead, Herts, HP2 4SJ, United Kingdom
Tel: +44 (0)1442 358401 Fax: +44 (0)1442 238112
Parker Hydraulics & Pneumatics Ltd, Unit 5, Ashbourne Estate, Page 247
Mile Cross Lane, Norwich, NR6 6RY, United Kingdom
Tel: +44 (0)1603 403190 Fax: +44 (0)1603 406190
Pressure Hoses Ltd, Nasmyth Business Centre, Green Lane, Facing p. 275
Paticroft, Manchester, M30 OSN, United Kingdom
Tel: +44 (0)161 7878085 Fax: +44 (0)161 787 8085
Rota Engineering Ltd, Wellington Street, Bury, Lancashire, Facing p. 431

BL8 2BD, United Kingdom
Tel: +44 (0)161 7640424 Fax: +44 (0)161 7629729
TG Scott Subscriber Services, 6 Bourne Green, Wrotham Road, Facing p. 274

Borough Green, Kent, TN15 8DG, United Kingdom
Tel: +44 (0)1732 884023 Fax: +44 (0)1732 884034
Sterling Hydraulics Ltd, Sterling House, Crewkerne, Somerset, Facing p. 138

TA18 8LL, United Kingdom
Tel: +44 (0)1460 72222 Fax: +44 (0)1460 72334
UCC International Ltd, PO Box 3, Thetford, Norfolk, IP24 3RT, Facing p. 578
United Kingdom
Tel: +44 (0)1842 754251 Fax: +44 (0)1842 753702
e-mail: fluidpower@ucc.co.uk
World Dredging, Mining & Construction, 17951-C Skypark Circle, Facing p. 635
Irvine, CA 92614, USA
Tel: +1 7145530836 Fax: +1 7148639261
717
TRADE NAMES INDEX
100, 200, 300 Series - High pressure filters - Fairey ArIon Ltd
900, 800 Series - Old High pressure filters - Fairey ArIon Ltd
AQUALINK - Sub sea control fluids - Houghton Vaughan PIc
ATOS - Hydraulic valves - Parker Hydraulics & Pneumatics Ltd
ATOS - Hydraulic & Electronic Technology - D.G. Howell (Hydraulic Engineers)
Ltd
BETAFLOW - High pressure filter range - DCC International Ltd
CM20 - Portable particle counter range - DCC International Ltd
COMPACT - Hydraulic Motor Shaft Driving - Hagglunds Drives Ltd
DYNAP AD - Axial piston pumps - Forward Industries Ltd
ELETTROTEC - Pressure Switches - Applications Engineering Ltd
EO - Hydraulic tube fittings - Parker Hannifin PIc
ERMETO - Hydraulic tube fittings - Parker Hannifin PIc
FESTO - Pneumatics - Parker Hydraulics & Pneumatics Ltd
FINENEX - Level Switches - Applications Engineering Ltd
FLUID CARE - System and fluid management service - Houghton Vaughan PIc
GATES HYDRAULICS - Hydraulic hose & couplings - Parker Hydraulics &
Pneumatics Ltd
HANNIFIN - Cylinders - Parker Hannifin PIc
HANSEN - Quick-release and self-sealing hose couplings and fittings - Guyson
International Ltd
HELAC - Rotary actuators - Rota Engineering Ltd
HELLER HYDRAULIK - Hydraulic valves/pumps/motors - Hydraulik Ring Ltd
HOUGHTOLUBRIC - Synthetic hydraulic fluids - Houghton Vaughan PIc
HOUGHTO-SAFE - Fire resistant hydraulic fluids - Houghton Vaughan PIc
HPFE - High performance filter element - Fairey ArIon Ltd
HYACS - Hydraulic cylinders - Forward Industries Ltd
HYDRALUBRIC - Fire resistant hydraulic fluids - Houghton Vaughan PIc
HYDRAULIK RING - Hydraulic valves/pumps/motors - Parker Hannifin PIc
HYDRA-TEST - Hydraulic System test equipment - Hydraulic System Products Ltd
HYDRA-VIS - Fire resistant hydraulic fluids - Houghton Vaughan PIc
HYDRO-DRIVE - Oil-based hydraulic fluids - Houghton Vaughan PIc

HYDROFIT - Hydraulic hose fittings - Manuli Hydraulics Ltd
HYDROLUX - Logic elements and D.C. Valves - Forward Industries Ltd
HYDROPA/HYDROSTAR - Pressure switches - Rota Engineering Ltd
MANATROL - Hydraulic valves - Parker Hannifin PIc
MANULI - High pressure hydraulic hose - Manuli Hydraulics Ltd
MARATHON - Hydraulic Motor Shaft Driving - Hagglunds Drives Ltd
MUL TICLAMP - Pipe clamping system - DCC International Ltd
MULTIFLOW - Tank mounted return hose filter range - DCC International Ltd
NORSEMAN - Wheel Hub Motor - Hagglunds Drives Ltd
NO-SKIVE - Rubber hose/fittings - Parker Hannifin PIc
P ARFLEX - Thermoplastic hose - Parker Hannifin PIc
P ARKRIMP - Rubber hose/fittings - Parker Hannifin PIc
PDI - Pressure Devices inc. Pressure Switches - Applications Engineering Ltd
POWERLINE - Hydraulic cylinders - Forward Industries Ltd
PRADIF A - Seals - Parker Hannifin PIc
PTI TECHNOLOGIES - Quality cost effective filtration products - D.G. Howell
(Hydraulic Engineers) Ltd
PURE TECHNOLOGY - Design concept - Fairey ArIon Ltd
RACINE - Pumps & valves - Forward Industries Ltd
RY-CONTROL - Hand controlled hydraulic valves - British Engineering Produc-
tions Ltd
SERVOSA VER - Servo Filter - Fairey Arion Ltd
SENSOCONTROL - Measurement system - Parker Hannifin PIc
SLENDERLINE - Hydraulic cylinders - Forward Instruments Ltd
STACK-MAGIC - Blowout preventor fluids - Houghton Vaughan PIc
SUN HYDRAULICS - Cartridge valves - D.G. Howell (Hydraulic Engineers) Ltd
SYSTEM 20 - Condition monitoring range - DCC International Ltd
SYSTEM MATCHED FILTRATION - Filtration concept - Fairey ArIon Ltd
TANKTOPPER - Tank return filters - Fairey ArIon Ltd
TELL TALE - Parker Hannifin PIc
THERMAL TRANSFER - Oil coolers - Forward Industries Ltd
TITAN - Hydraulic cylinders - D.G. Howell (Hydraulic Engineers) Ltd
TORQUEMASTER - Semi-rotary actuators - Forward Industries Ltd
TRIPLELOX - Hydraulic tube fittings - Parker Hannifin PIc
UCC - Filters - Parker Hydraulics & Pneumatics Ltd
VICKERS - Pumps, valves, spares & repairs - D.G. Howell (Hydraulic Engineers)
Ltd
VIKING - Hydraulic Motor Drum Driving (Rotating Case) - Hagglunds Drives Ltd
VOAC - Hydraulic valves/pumps/motors - Parker Hannifin PIc
VSG - Pumps & Motors - Forward Industries Ltd
WILFIT - Replacement element - Fairey ArIon Ltd
719
BUYER'S GUIDE
PUMPS Rota Engineering Ltd
Piston - radial Mikrospin Ltd
D.G. Howell (Hydraulic Engineers) Ltd Parker Hydraulics & Pneumatics Ltd
Mikrospin Ltd
Parker Hannifin PIc Pressure Intensifiers
Hydraulik Ring Ltd
Piston - axial
D.G. Howell (Hydraulic Engineers) Ltd PUMP DRIVERS
Forward Industries Ltd Air
Hagglunds Drives Ltd Parker Hydraulics & Pneumatics Ltd
Hydraulik Ring Ltd
Kawasaki Precision Machinery (UK) Ltd Couplings
Linde Hydraulics Ltd D.G. Howell (Hydraulic Engineers) Ltd
Parker Hannifin PIc Mikrospin Ltd
Parker Hydraulics & Pneumatics Ltd
Vane
D.G. Howell (Hydraulic Engineers) Ltd MOTORS
Forward Industries Ltd Piston - radial
Hydraulik Ring Ltd Forward Industries Ltd
Parker Hannifin PIc Hagglunds Drives Ltd
Parker Hydraulics & Pneumatics Ltd Kawasaki Precision Machinery (UK) Ltd
Mikrospin Ltd
Gear - external Parker Hannifin PIc
D.G. Howell (Hydraulic Engineers) Ltd
Hydraulik Ring Ltd Piston - axial
Mikrospin Ltd D.G. Howell (Hydraulic Engineers) Ltd
Parker Hannifin PIc Forward Industries Ltd
Kawasaki Precision Machinery (UK) Ltd
Gear - internal Linde Hydraulics Ltd
Hydraulik Ring Ltd Parker Hannifin PIc
Parker Hannifin PIc
Parker Hydraulics & Pneumatics Ltd Vane
Hand pumps Parker Hydraulics & Pneumatics Ltd
Gear Ilotating Hydraulic Couplings

D.G. Howell (Hydraulic Engineers) Ltd Cosappo Srl
Hydraulik Ring Ltd
Mikrospin Ltd Linear Transducers (for inside cylinders)
Parker Hydraulics & Pneumatics Ltd Rota Engineering Ltd
Gerotor Motors 1l0TAIlY ACTUATOIlS

Parker Hannifin PIc Vane type
Hydraulik Ring Ltd
CYLINDERS
Single acting Piston type
Cosappo Srl D.G. Howell (Hydraulic Engineers) Ltd
Denley Engineering Co. Ltd Forward Industries Ltd
D.G. Howell (Hydraulic Engineers) Ltd Rota Engineering Ltd
Forward Industries Ltd Parker Hannifin PIc
Hydraulik Ring
Parker Hannifin PIc Valve Actuators
Parker Hydraulics & Pneumatics Ltd Rota Engineering Ltd
Double acting VALVES

Cosappo Srl Directional control
Denley Engineering Co. Ltd D.G. Howell (Hydraulic Engineers) Ltd
D.G. Howell (Hydraulic Engineers) Ltd Forward Industries Ltd
Forward Industries Ltd Hydraulik Ring Ltd
Hydraulik Ring Ltd Mikrospin Ltd
Parker Hannifin PIc Parker Hannifin PIc
Parker Hydraulics & Pneumatics Ltd Parker Hydraulics & Pneuamtics Ltd
Sterling Hydraulics Ltd
Hydrostatic bearings
Denley Engineering Co. Ltd Flow control
Parker Hannifin PIc D.G. Howell (Hydraulic Engineers) Ltd
Forward Industries Ltd
Cylinder mounts Hydraulik Ring Ltd
Cosappo Srl Mikrospin Ltd
Denley Engineering Co. Ltd Parker Hannifin PIc
D.G. Howell (Hydraulic Engineers) Ltd Parker Hydraulics & Pneumatics Ltd
Parker Hannifin PIc Sterling Hydraulics Ltd
Pressure control
Ilotaryffelescopic D.G. Howell (Hydraulic Engineers) Ltd
Denley Engineering Co. Ltd Forward Industries Ltd
Hydraulik Ring Ltd
Telescopic Cylinders Hydraulic System Products Ltd
Cosappo Srl Mikrospin Ltd
Parker Hannifin PIc
Top Link Cylinders Parker Hydraulics & Pneumatics Ltd
Cosappo Srl Sterling Hydraulics Ltd
BUYERS GUIDE 721
Proportional Pipes
D.G. Howell (Hydraulic Engineers) Ltd D.G. Howell (Hydraulic Engineers) Ltd
Hydraulik Ring Ltd Parker Hydraulics & Pneumatics Ltd
Mikrospin Ltd Pressure Hoses Ltd
Parker Hannifin PIc
Parker Hydraulics & Pneumatics Ltd Pipe couplings
Sterling Hydraulics Ltd D.G. Howell (Hydraulic Engineers) Ltd
Guyson International Ltd
Servo Parker Hannifin PIc
D.G. Howell (Hydraulic Engineers) Ltd Parker Hydrauluics & Pneumatics Ltd
Hydraulik Ring Ltd Pressure Hoses Ltd
Parker Hannifin PIc
Parker Hydraulics & Pneumatics Ltd Hoses
Relief
Hydraulic System Products Ltd
Manuli Hydraulics Ltd
Hydraulik Ring Ltd
Parker Hannifin PIc
Parker Hydraulics & Pneuamtics Ltd
Mikrospin Ltd
Pressure Hoses Ltd
Parker Hannifin PIc
Hose couplings
Load Control Hydraulic System Products Ltd
Sterling Hydraulics Ltd Manuli Hydraulics Ltd
Parker Hannifin PIc
FITTINGS & ACCESSORIES Pressure Hoses Ltd
Accumulators
D.G. Howell (Hydraulic Engineers) Ltd Seals
Parker Hannifin PIc
Parker Hydraulics & Pneumatics Ltd Hose/Pipe Cleaning System
Applications Engineering Ltd
Shock absorbers Moss Plastic Parts Ltd
Parker Hydraulics & Pneumatics Ltd Parker Hydraulics & Pneumatics Ltd
Coolers Plugs
Forward Industries Ltd Moss Plastic Parts Ltd
Mikrospin Ltd
Parker Hydraulics & Pneumatics Ltd FLUIDS
Mineral oils
Reservoirs and tanks Houghton Vaughan PIc
Houghton Vaughan PIc Parker Hydraulics & Pneumatics Ltd
Hydraulik Ring Ltd
Mikrospin Ltd Water
Parker Hannifin PIc Houghton Vaughan PIc
Emulsions Mikrospin Ltd

Houghton Vaughan PIc MP Filtri UK Ltd
Parker Hannifin Ltd
Water-based glycols Parker Hydraulics & Pneumatics Ltd
Houghton Vaughan PIc uee International Ltd
Tank
Synthetic
Houghton Vaughan PIc
Fairey ArIon Ltd
Ecologically acceptable
Mikrospin Ltd
MP Filtri UK Ltd
Analysis
D.G. Howell (Hydraulic Engineers) Ltd Parker Hannifin PIc
IFTS
uee International Ltd
Parker Hannifin PIc OfT-line
Additives Fairey ArIon Ltd
Houghton Vaughan PIc Forward Industries Ltd
Hydraulik Ring Ltd
FILTERS Mikrospin Ltd
Media MP Filtri UK Ltd
D.G. Howell (Hydraulic Engineers) Ltd Parker Hannifin PIc
Fairey ArIon Ltd Parker Hydraulics & Pneumatics Ltd
Mikrospin Ltd uee International Ltd
MP Filtri UK Ltd
Parker Hannifin PIc Indicators
Parker Hydraulics & Pneumatics Ltd D.G. Howell (Hydraulic Engineers) Ltd
uee International Ltd Fairey ArIon Ltd
High pressure Hydraulic System Products Ltd
D.G. Howell (Hydraulic Engineers) Ltd Mikrospin Ltd
Fairey ArIon Ltd MP Filtri UK Ltd
Hydraulik Ring Ltd Parker Hydraulics & Pneumatics Ltd
Mikrospin Ltd uee International Ltd
MP Filtri UK Ltd
Parker Hannifin PIc Breathers
Parker Hydraulics & Pneumatics Ltd D.G. Howell (Hydraulic Engineers) Ltd
uee International Ltd Fairey ArIon Ltd
Low pressure and return Hydraulik Ring Ltd
D.G. Howell (Hydraulic Engineers) Ltd Mikrospin Ltd
Fairey Arion Ltd MP Filtri UK Ltd
Hydraulic System Products Ltd Parker Hydraulics & Pneumatics Ltd
Hydraulik Ring Ltd uee International Ltd
BUYERS GUIDE 723
Filter testing Parker Hydraulics & Pneumatics Ltd

D.G. Howell (Hydraulic Engineers) Ltd Sterling Hydraulics Ltd
Fairey Arlon Ltd
Hydraulik Ring Ltd System checking
IFfS Denley Engineering Co. Ltd
MP Filtri UK Ltd D.G. Howell (Hydraulic Engineers) Ltd
Parker Hannifin PIc Houghton Vaughan PIc
Parker Hydraulics & Pneumatics Ltd Hydraulik Ring Ltd
Linde Hydraulics Ltd
Stainless Filters Mikrospin Ltd
MP Filtri UK Ltd Parker Hydraulics & Pneumatics Ltd
SYSTEMS SYSTEM CONTROL

Hydrostatic transmissions Servo-systems
Denley Engineering Co. Ltd Denley Engineering Co. Ltd
D.G. Howell (Hydrualic Engineers) Ltd Flotron Ltd
Forward Industries Ltd Forward Industries Ltd
Hagglunds Drives Ltd Hydraulik Ring Ltd
Linde Hydraulics Ltd Parker Hannifin PIc
Sterling Hydraulics Ltd Electro-modulated hydraulics
Denley Engineering Co. Ltd
System design D.G Howell (Hydraulic Engineers) Ltd
Denley Engineering Co. Ltd Flotron Ltd
D.G. Howell (Hydraulic Engineers) Ltd Forward Industries Ltd
Forward Industries Ltd Hydraulik Ring Ltd
Houghton Vaughan PIc Linde Hydraulics Ltd
Linde Hydraulics Ltd Rota Engineering Ltd
Mikrospin Ltd
Parker Hannifin PIc Computer control
Parker Hydraulics & Pneumatics Ltd Flotron Ltd
Sterling Hydraulics Ltd Hydraulik Ring Ltd
Parker Hannifin PIc
Computer simulation Rota Engineering Ltd
Flotron Ltd Linear Transducers
Hydraulik Ring Ltd Rota Engineering Ltd
System build
Denley Engineering Co. Ltd INSTRUMENTATION
D.G. Howell (Hydraulic Engineers) Ltd Leakage
Forward Industries Ltd Parker Hydraulics & Pneumatics Ltd
Houghton Vaughan PIc UCC International Ltd
Hydraulik Ring Ltd
Mikrospin Ltd Pressure
Parker Hannifin PIc Applications Engineering Ltd
Hydraulic System Products Ltd Filter Tests

MP Filtri UK Ltd IFTS
Parker Hannifin PIc
Parker Hydraulics & Pneumatics Ltd Contamination Monitoring
Rota Engineering Ltd uee International Ltd
uee International Ltd
SPECIAL HYDRAULICS
Flow High temperature hydraulics
Applications Engineering Ltd Houghton Vaughan PIc
Parker Hannifin PIc Water hydraulics
Parker Hydraulics & Pneumatics Ltd Denley Engineering eo. Ltd
uee International Ltd Fairey Arlon Ltd
Temperature Parker Hydraulics & Pneumatics Ltd
Applications Engineering Ltd
Hydraulic System Products Ltd Screw Type Cartridge Valves
uee International Ltd TRAINING
eourses
Condition monitoring - containment Fairey ArIon Ltd
MP Filtri UK Ltd Hydraulik Ring Ltd
Parker Hannifin PIc IFTS
Parker Hydraulics & Pneumatics Ltd Linde Hydraulics Ltd
Parker Hannifin PIc
Condition monitoring - other Parker Hydraulics & Pneumatics Ltd
Parker Hannifin PIc
uee International Ltd Publications
Flotron Ltd
Test systems Houghton Vaughan PIc
Hydraulic System Products Ltd Hydraulik Ring Ltd
Hydraulik Ring Ltd IFTS
IFTS Parker Hannifin PIc
uee International Ltd Equipment
Hydraulik Ring Ltd
Sampling points IFTS
Hydraulic System Products Ltd Parker Hannifin PIc
IFTS
Parker Hannifin PIc Standards
uee International Ltd IFTS
Calibration APPLICATIONS
Hydraulic System Products Ltd Mechanical handling
IFTS Denley Engineering eo. Ltd
BUYERS GUIDE 725
D.G. Howell (Hydraulic Engineers) Ltd Injection moulding machines

Fairey Arlon Ltd Denley Engineering Co. Ltd
Forward Industries Ltd D.G. Howell (Hydraulic Engineers) Ltd
Hagglunds Drives Ltd Fairey Arlon Ltd
Houghton Vaughan PIc Forward Industries Ltd
Hydraulik Ring Ltd Hagglunds Drives Ltd
IFI'S Houghton Vaughan PIc
Parker Hannifin PIc IFI'S
Parker Hydraulics & Pneumatics Ltd MP Filtri UK Ltd
Rota Engineering Ltd Parker Hannifin PIc
Sterling Hydraulics Ltd Parker Hydraulics & Pneumatics Ltd
Rota Engineering Ltd
Industrial robots Sterling Hydraulics Ltd
D.G. Howell (Hydraulic Engineers) Ltd Presses
Fairey Arlon Ltd Denley Engineering Co. Ltd
Houghton Vaughan PIc D.G. Howell (Hydraulic Engineers) Ltd
Hydraulik Ring Ltd Fairey Arlon Ltd
IFI'S Forward Industries Ltd
MP Filtri UK Ltd Houghton Vaughan PIc
Parker Hannifin PIc Hydraulik Ring Ltd
Rota Engineering Ltd IFI'S
Sterling Hydraulics Ltd MP Filtri UK Ltd
Parker Hannifin PIc
Machine tools and automation
Fairey Arlon Ltd
Road and rail vehicles
Hydraulik Ring Ltd
IFI'S
Fairey Arlon Ltd
MP Filtri UK Ltd
Hydraulik Ring Ltd
Parker Hannifin PIc
IFI'S
MP Filtri UK Ltd
Parker Hannifin PIc
Workshop tools Rota Engineering Ltd
Denley Engineering Co. Ltd Sterling Hydraulics Ltd
Hydraulik Ring Ltd Mobile
IFI'S D.G. Howell (Hydraulic Engineers) Ltd
MP Filtri UK Ltd Fairey ArIon Ltd
Parker Hannifin PIc Hagglunds Drives Ltd
Rota Engineering Ltd IFI'S
MP Filtri UK Ltd Rota Engineering Ltd

Rota Engineering Ltd Mining
Agriculture D.G. Howell (Hydraulic Engineers) Ltd
Fairey Arlon Ltd Forward Industries Ltd
Hydraulik Ring Ltd Hagglunds Drives Ltd
IFrS Houghton Vaughan PIc
Parker Hannifin PIc IFfS
Parker Hydraulics & Pneumatics Ltd MP Filtri UK Ltd
Rota Engineering Ltd Parker Hannifin PIc
Sterling Hydraulics Ltd Rota Engineering Ltd
Aircraft
Denley Engineering Co. Ltd Medical
IFfS Denley Engineering Co. Ltd
Parker Hannifin PIc Hydraulik Ring Ltd
Rota Engineering Ltd IFfS
Parker Hannifin PIc
Marine
Denley Engineering Co. Ltd Aircraft SullPort
D.G. Howell (Hydraulic Engineers) Ltd Fairey Arlon Ltd
Fairey Arlon Ltd
Forward Industries Ltd Water
Hagglunds Drives Ltd IFfS
Hydraulik Ring Ltd Steel Industry
IFfS Houghton Vaughan PIc
MP Filtri UK Ltd
Parker Hannifin PIc Military
Parker Hydraulics & Pneumatics Ltd Sterling Hydraulics Ltd
Sterling Hydraulics Ltd MISCELLANEOUS
Power packs and ring-mains
Off-shore and sub-sea Denley Engineering Co. Ltd
Denley Engineering Co. Ltd D.G. Howell (Hydraulic Engineers) Ltd
Forward Industries Ltd Houghton Vaughan PIc
Hagglunds Drives Ltd Hydraulik Ring Ltd
Houghton Vaughan PIc Mikrospin Ltd
IFfS Parker Hydraulics & Pneumatics Ltd
MP Filtri UK Ltd
Parker Hannifin PIc Consultants
Parker Hydraulics & Pneumatics Ltd Denley Engineering Co. Ltd
BUYERS GUIDE 727
D.G. Howell (Hydraulic Engineers) Ltd MilitarylNaval

IFfS Manifolds
British Engineering Productions Ltd
Custom Manifold Blocks
EDITORIAL
INDEX
729
EDITORIAL INDEX
air-hydraulic cylinders, 553
A Air Release Value, 24
accumulator (s), 213 aircraft, 13
hydraulics, 249, 669
control, 232
reservoirs, 257
duties, 230
emergency power, or energy storage, 232 systems, 232
holding devices, 232 aliasing, 461
all-metal hose, 293
sizing, 226
altemative pilot operation, 196
standards, 560, 568
type devices, 233 aluminium tubes, 264
amplifier valves, 633
acoustic,
amplitude ratio, 437
damping, 515
aniline point, 15, 23
filters, 520
annular,
treatment, 515
actuator (s), 644, 680 orifices, 38
dimensions, basic equations, 98 piston accumulator, 218
double-acting, double-ended, 96 anti-foam agents, 328
double-acting, single ended, 96 anti-wear additives, 329
performance, 385 articulated,
semi-rotary, 395 booms, 583
mechanical systems, 593
single-acting, 96
artificial limbs, 711
types, 96
asynchronous electric motor, 81
vibration and noise, 517
attachments, 608
additives,
auto ignition, 15
anti-wear, 329
Auto Ignition Temperature (AIT), 24
for mineral oils, 326
auto-frettage, 542
aerospace, 13
automatic transmissions, 662
valves, 205
automation, 597, 603
agricultural,
machine tools and, 597
hydraulics, 643
automobile hydraulic brakes, 655
machines, 326
auxiliary services, 660
applications, 336
axial piston,
air,
pump, 52-4, 75
blast coolers, 243
motor characteristics, 86
contamination, 364
B C
bacterial contamination, 333 cartesian, 593
baffles, 256 mechanical systems, 593
baling presses, 621 cartridge valves, 154
ball valves, 147 cast iron pipes, 260
barrel, 506 central by-pass valves, 170
base units, 2 changing cross section, flow through, 35
basic, charge pump, 407
circuits, 41 check valves, 161
computer interfacing, 459 chevron seal, 313
requirements of all slide valves, 142 circuit, 696
bearings, 506 design, 458, 550, 648
benders, 608 diagrams, 613
bent-axis piston pumps, 55 clamps, 608
Bernoulli, cleanable filter elements, 350
effects, 150 cleanliness,
forces, 150 coding systems, 367
bite coupling, 276 levels, 360
bladder type accumulator, closed loop,
open-bag / open-top, 221 basics, 431
closed bag, 219 control, 462
board presses, 621 system,
body panel presses, 620 advantages of, 432
bonded washer, 309 closed-centre valve, 160
booms, articulated, 583 closed-circuit hydrostatic transmission, 403
boost supply, 407 cloud point, 23
brake, temperature, 15
lining presses, 621 code of practice, 705
systems (standards), 560 coding systems, cleanliness, 367
brazed, coiled tubing, 294
connection, 275 coining presses, 629
coupling, 282 cold climates, 249
breathers, 254 Cold Finished Seamless (CFS), 263
bridge operation, 580 cold-forging presses, 620
British Fluid Power Association guidelines combination ring seals, 314
and data, 572 combined bite ring with elastomeric
British Standards, seal, 278
BS4231, 17 common conversion factors, 3
BS 5000, 78 compatibility of fluids, 337
BS 5242, 107 compensation, 440
BS 7201, 233 component (s), 690
buckling length, 121 circuit design, 695
burst disc, 185 composite sealing rings, 314
by-pass, compressibility, 15
compensated proportional valve 199 'delay', 393
valves, 182 effects, 394
of fluids, 20
computer (s),
EDITORIAL INDEX 731
aids and simulation, 423 critical rod lengths, 119

generic methods, 427 cryogenic coupling, 280
programme, current drive amplifier, 448
component selection, 424 cushioning, 101
categories and requirements, 423 cylinder (s), 550, 601, 607
for control, 459 buckling, 119
concentric cylinders, 39 construction, 111
conductivity, 24 cushioning, 102
connections, 598 double acting, 374
standards, 569 end cover, 110
connector, 275 equations, 98
contaminant (s), 341 friction, 123
particles, size of, 342 mounting, 116
contamination, position feedback in, 442
air, 364 re-assembly, 506
control, 357,361 rod, 109
design, 361 selection and performance, 122
ingression, 361 single acting, 373
steps in, 361 speed controls, 375
effects in servo spools, 203 standards, 561, 568
insensitivity, 362 cylindrical-type accumulator (open-bag /
life, 361 open-top), 221
measurement of, 357
reliability, 361 D
water, 342, 363 damping, 213
continuity of flow, 30 of pump,
control (s), 594 pressure surges, 230
circuit design, 603 pulsations, 220
developments, 590 ratio, 438
pressure, 380 of vehicles, 213
valves, 640 dash,
and selectors, 647 number, 259
controller structures, 440 sizes, 289
controlling, data, analysis of, 497
flow, units for, 611 DC or AC electric drives, 593
pressure, units for, 611 DCV centre position alternatives, 160
conventional hydraulic fluids, 531 deadband compensator, 451
coolers, 239 deceleration valve, 168
cooling, 240, 533 deck machinery, 683, 693
copper, 264 density, 15, 19
corrosion, derived units, 1
inhibitors, 328 design,
monitoring, 495 considerations, 419
counterbalance valve, 182 and pipework installations, 272
coupling (s), 77-8, 83, 275,466 procedure, 419
standards, 569 diaphragm accumulator, 221
cracking, 467 diesel,
cranes, mobile, 585 drivers, 514
engine (s) 83-4 electrohydraulic

digital control, 441 control, 71
digitisation, 460 servo-valves,
direct acting electrohydraulic servo- direct acting, 203
valves, 203 pilot stages for, 205
direct acting relief valve, 176 electromagnetic compatibility, 453
directional, electrostatic filters, 350
control, EMC Directive, 523
of movement, 373 emulsions, 330
and flow control valves, 157 EN 50020, 705
disposable filter elements, 350 EN 60079-10, 705
dither signal, 449 end covers, 110
double acting cylinders, 374 energy equations, 33
drawing, engine,
presses, 622-4 starters, 661
tools, triple-action, 623 powered lift trucks, 582
drilling, 691 Ethylene Propylene (EP), 286
drive (s), 588,593 excessive,
circuits, system pressure, 508
common features of, 447 temperature, 468
couplings, 644 exercise equipment, 711
DC or AC electrics, 593 external gear pump, 59-60
hydraulic, 593 extrusion, 467
pneumatic, 593 presses, 629
dry ice press, 621
dual-pressure circuits, 232 F
dump valve, 163 faults,
duplex filters, 355 causing,
duty cycle, 421 complete stoppage of the system, 465
dynamic, excessive temperature, 465
(absolute) viscosity, 16 leakage, 465
analysis, 415 malfunction of control or opera-
response, 413 tion, 465
filling, 254
E filters, 341, 343, 550, 696
eccentric cylinders, 40 construction, 348
ecologically acceptable fluids, 336 duplex, 355
eddy current device, 444 electrostatic, 350
elastomer, elements,
compound compatibility, 307 cleanable, 350
properties, 306 disposable, 350
electric (s), . integrity of, 347
fork lift trucks, 581 location, 351
motors, 78, 514 magnetic, 354
efficiencies of, 81 maintenance of, 502
versus hydraulics, 9, 10, 12 media,
versus pneumatics, 457 types of, 348
electro-modulated hydraulics, 447 rating (s), 343
EDITORIAL INDEX
733
choice of 353 cell press, 624

size, compatibility with, 347
choice of, 353 compatibility of, 337
standards, 562 conditioning, 408
strength of, 347 ecologically acceptable, 336
suction, 353 hydraulic, 325
testing, 355 level indicator, 254
filtering, maintenance of, 502
pressure-line, 351 properties of, 15
return-line, 352 silicone, 335
filtration, 341-343 standards, 563, 569
off-line, 352 synthetic, 334
fire, types, changing, 338
point temperature, 23 types of, 325
resistance, 337 ultra high pressure, 539
fitting connection to hose, 297 fluorocarbon rubber (FKM & FPM), 286
fittings, 275, 680 flush flange fitting, 114
with formed-tube, 280 food applications, 336
standards, 563 force,
five-way valves 159 analysis, linear actuator, 387
fixed, controlled solenoid, 189
delivery vane pumps, 58 motor operated, 204
sequence robots, 591 multiplication, 28
flanged, forging presses 625
and-bolted covers, 112 fork lift trucks, electric 581
couplings, 283 formed,
flared coupling, 280 flange end coupling, 281
flash point, 23 nipple type coupling, 281
temperature, 15 tube, 275
flat face coupling, 302 couplings 281
flexible hose, handling of, 507 four-way valves, 158
float centre valve, 160 frequency response, significance of, 437
float-type accumulators, 223 friction in cylinders, 123
flow, 476
capacity, 345 G
control valves, 164, 190 galvanic isolators, 705
summary of, 165 gantry mechanical systems, 593
dividers, 168 gap-type presses, 620
forces, 150 gas,
limited, 392 flexible-separator-type accumulators, 219
measuring methods, 476 loaded,
units for controlling, 611 accumulators with auxiliary bottle, 229
velocity, spring, 231
considerations, 265 rigid-separator-type accumulators, 215
in tube, 265 gaskets, 309
fluid (s), 548, 689, 695 gear,
application suitability, 337 motor, 90-1
basic properties of, 15 pump (s), 59,61-3
generic computer methods, 427 maintenance of, 505

gland end cap, 505 power output of, 101
glycols, water-based, 333 drive (s), 12
goods lifts, 580 applications of, 92
gravimetric level, 358 filter media,
grinding wheel presses, 621 types of, 348
ground support, 13 fluid shock,
equipment, 674 absorption of, 235
fluids, 325
H fuses, 164
hatch covers, 685 hose, 285
head, 396 construction of, 285
heaters, 249 cover variations for, 289
heating, maintenance of, 507
sources of 239 reinforcement of, 287
heavy gauge tubing, 260 storage of, 507
high ambient temperature, 296 lock, 149
high temperature, mining, 702
hydraulics, 531 motors, 85, 578, 649
seals, 320 operations,
testing rigs, 535 examples of, 12
high-technology robots, 591 pit props, 697
hobbing presses, 621 power packs, 637, 638
hose (es), 550 selection of, 640
abrasion, 296 power sources, 707
construction, 285 presses, 617
correct line of, 295 small, 617
couplings and fittings, 297 pressure, 640
fitting termination, 299 pump selection, 72
fittings, choice of type, 299 ring mains, 637
installation, 294 suspensions, 662
pressure rating, 289 systems,
problems, 466 complete stoppage of, 469
selection, 294 efficiency of, 34
skiving, 298 mobile, 581
standards, 289, 564, 569 static, 579
and tube standards, 564 valve,
vibration and noise, 516 classification, 133
human monitoring, 490 requirements (special), 632
hybrid pumps, 64 workholding, 600
hydraulic (see also hydraulics), hydraulically stressed bolts, 545
advantages, 9 hydraulics (see also hydraulic),
applications, 11 in aerospace, 669
brakes, 232 classification of, 85
components, characteristics of, 438 in medicine, 707
computer programs, 427 hydro-pneumatics, 553
cylinders, hydrodynamics, moving fluids, 30
classification and definition, 95 hydrostatic,
EDITORIAL INDEX 735
bearing, 114-5 L
principle, 116 laminar flow, 30, 39
extrusion, 631 laminating hot-plate presses, 621
principle, 41 land-based transport, 655
transmission (s), 399, 664 landing gear, 94, 671
circuit, 45 large presses, relief valves for, 632
dynamics, 413 leakage, 213, 465, 495
stationary fluids, 27 compensation, 213
unit selection, 408 legislation, 521
lift trucks,
I electricfork, 581
in-line piston pumps, 50 engine-powered, 582
industrial hydraulic shock absorbers, 237 lifting equipment, 708
industrial robots, 591 lifts,
ingression, 361 passenger or goods, 580
injection, 612 scissor, 579
moulding machines, 611 line,
inner tube, 286 connections, 104, 255
instability, 433 mounting, 153
Institute of Petroleum Standards, 571 linear,
instrumentation, 254, 471 actuator, 98
integrated digital sensing, 445 basic speed relationship, 385
intensifiers, 554, 702 force analysis, 387
internal, circuit (s), 41,373
combustion engines, 82 cylinders, 601
internal gear pump, 62 inductive position sensor (LIPS), 444
design, 63 resolver, 445
intrinsic safety, 695, 704 spool, 134
International Standards Organisation valve orifice and flow path, 135
(ISO), 5 valves, 139
ISO 1219, 5 liquid springs, 544
ISO 3320, 110 liquid-metal fluids, 535
ISO 3448, 17 load independent proportional flow con-
ISO 4395, 118 trol, 198
ISO 4406, 367 lobe-rotor motors, 91
ISO 6020, 116 logic,
ISO 6022, 116 availability of, 455
standards, 559 circuit devices, 456
viscosity classification, 17 elements, 155
isostatic pressing, 542 loop gain, 435
J M
jet pipe valve, 207 machine tool (s), 597
hydraulic system for, 597
K magnetic filters, 354
K values, 36 magnetostrictive, 444
kinematic viscosity, 16 maintenance,
aspects of pipelines, 270
ease of, 362 mould actuation, 612

programmes, 501 movement, directional control of, 373
malfunction, moving equipment, 708
of control and operation, 469 moving fluids, hydrodynamics, 30
manifold blocks, 550
manipulators, manual, 591 N
marine, National Aerospace Standard (NAS), 371
hydraulics, 679 NAS 1638, 371
system pressures, 679 natural frequency, 438
measurement of ultra-high pressure, 538 NC robots, 591
mechanical, needle valves, 146
efficiency, 34 negative force port, 151
handling, 577 neutralisation value, 24
systems, Newtonian fluid, 16
articulated, 593 NFPA Standards, 567
cartesian, 593 Nitrile-butadiene rubber (NBR), 286
gantry, 593 noise, 511
polar, 593 sources of, 512
rectilinear, 593 nominal size, (or NG), 137
mechanical wear, 468 non-gas accumulators, 223
media, advantages and disadvantages of non-Newtonian fluid, 16
different, 10 non-SI Units, 2
membrane, notation, 5
optical density, 358 nozzle flapper, 206
piston accumulator, 218
metallic flexible hose, 293 o
mineral oils, 326 a-rings, 308
additives for, 326 off-line filtration, 352
miniature hydraulics, 525 offshore, 689
minimum bend radii, 295 oil,
mining, 695 analysis, 493
mobile, capacity, 640
cranes, 585 flow, 640
hydraulics, 643 heat exchangers, 250
hydraulic systems, 484,581 rigs, 223
machinery, 249 on/off valve designs, 161
momentum forces, 150 open-centre valve, 160
monitor, open circuit hydrostatic transmission, 400
choice of, 498 operating tables, 711
monitoring, 489 optical density, 358
techniques, 489 orifices, flow through, 35
Moody diagram, 32 outer cover, 286
mount, overall efficiency, 34
choice of, 118 oxidation inhibitors, 326, 328
motion control, 598
motors, 396, 548 P
hydraulic, 578, 649 parallel circuit,
standards, 569 by-pass valve, 172
EDITORIAL INDEX 737
configuration, 648 Polytetra-fluoroethylene (PTFE), 287

particle (s), Polyurethane (PU), 287
concentration, assessment of, 359 poppet valve,
count of, 359 orifice and flow path, 134
retention, 345 guided, 144
size filtration, unguided, 144
efficiency of, 343 positive displacement pump, 69
Pascal's Hypothesis, 27 potential hazards, 521
passenger lifts, 580 pour point, 23
peak pressures, 242 temperature, 15
performance monitoring, 497 powder compacting presses, 621
permanent tube grip fittings, 279 power, 78, 239
pick-and-place robots, 591 curve for a positive displacement
PID controller, 441 pump, 69
pilot operated, in circuits, 46
reducing valve, 181 packs, 637, 638
relief valve, 177 applications of, 641
pipes, 550, 680 regulation, 68
bending, 271 source (s), 606, 640, 643
fittings and seals 540 steering, 589, 659
laying and moving, 692 prefill valves, 163, 635
vibration and noise, 516 preheaters, 249
pipeline installation, 270 presses, 231
pipework, baling, 621
calculations, 259 board, 621
couplings and fittings, 275 body panel, 620
piston (l08), brake lining, 621
downstroke, 43 coining, 629
motors, 86 cold-forging, 620
pumps, 50, 52, 56 drawing, 622
rotary actuators, 128 dry ice, 621
seal, 108 extrusion, 629
type accumulator, 215 fluid cell, 624
upstroke, 42 forging, 625
plain cover styles, 321 gap-type, 620
plastic moulding presses, 621 general-purpose, 618
plasticising screw, rotation of, 613 grinding wheel, 621
platform build, 690 hobbing, 621
play-back robots, 591 hydraulic, 617
pneumatic (s), laminating hot-plate, 621
logic controls, 455 large, 620
versus electronics, 457 medium-sized, 618
versus hydraulics, 10 plastic moulding, 621
polar, 593 powder compacting, 621
mechanical systems, 593 rubber die-forming, 620
Polychloroprene (CR), 286 wheel rim, 620
Polyester elastomers (YPBO), 287 pressure, 396, 471
polymer compounds, 286 compensated,
flow control valve 167 energy efficient, 382

valves 166 cost-effectiveness, 73
control, 380, 440 delivery, 73
valves, 175, 188 design, 514
drop, 345 drivers, 77
at bends, 268 vibration and noise, 514
calculations, 267 drive speed, 73
in tube, 267 efficiency, 67
gauges, 471 fluid, 74
holding and leakage compensation, 231 in hydrostatic systems, 29
limited, 392 maintenance of, 504
line filtering, 351 and motors (standards), 565
losses at fittings, 268 performance, 65
measuring methods, 472 power:weight ratio, 70
override, 176 ripple, 213, 230
rating, 66, 289 standards, 569
reducing valves, 179 vibration and noise, 75,514
relief valves, 175 purge circuit, 407
strength of tube, 268
surges, 230 Q
switch, 642 qualifications, 576
transmission, velocity of, 396
units for controlling, 611 R
variations in the system, 508 rack and pinion rotary actuators, 128
printed gaskets, 309 radial piston,
priority valves, 170 pumps, 51,71
process plant, hydraulic drive in, 12 motors, 87-8
profile cutting ring, 278 rail hydraulics, 667
tube, 278 Ram Air Turbine, 83
progressive two edge cutting ring, 277 ramp control, 452
proportional, rapid-motion valves, 377
2-way flow control valve, 191 reciprocating dynamic seals, 311
4-way flow control valve, 192 rectilinear, 593
controller, 432 mechanical systems, 593
flow control valve, 190 reducing valve, 175
pressure valves, 188 reinforcement, 286
solenoid (s), 187 relief valve (s), 175
force controlled valve, 192 large presses, 632
power limit, 193 remotely operated vehicle (ROV), 692
stroke controlled valve, 193 reservoir (s), 519,550,695
valve circuits, 450 basic configuration and design 251
valves, 187 cooling, 240
protective covers or caps, 321 design, 252
pulsation dampers, 520 filling, 254
Pulse Width Modulation (PWM), 450 flexible separators, 364
pump (s), 548, 643, 644, 670, 679, 696 instrumentation, 254
bulk and weight, 74 standards, 570
circuits, and tanks, 251
EDITORIAL INDEX 739
resistivity. 24 damage, 466

retum-1iDefiltering. 352 failure. 467
rig testing. 481 friction, 317
rigid tube materials. 260 materials. 305
ring mains, 637 standards, 566, 570
ripping. 702 wear, 317
road vehicles. 232 sealed reservoirs, 256
robot (s). sealing ring couplings, 279
ftxed sequence. 591 secant bulk modulus. 20
high-technology. 591 second generation bite type connectors, 277
industrial. 591 secondary,
NC. 591 control, 416
other uses for. 596 systems, 660
pick-and-place. 591 selectors, 647
play-back. 591 self-sealing couplings. 301
size. semi-axial, 56
effect of. 595 semi-rotary actuators, 395
variable-sequence 591 sequence,
rod (s). 109,506 control, 461
end ftttings. 118. 505 valve, 183
seal, 505 sequencing, 381
rotary spool valves, 143 series circuit,
rotating. by-pass valve. 173
block pumps. 55 conftguration, 648
dynamic seals. 310 series compensated,
roughnecks. 692 proportional valve, 200
roughness. valve, 167
for tubes and pipes, 33 series-parallel conftguration, 649
rubber die-forming presses. 620 servo assistance, 431
rubbers. hardness of, 306 servo-operated brakes, 658
servo-valve,
S drive circuits, 449
safety. 521 spool, 201
design codes, 233 servosystems, 431
requirements, 524 servovalves, 201
sailing ships. 687 shaft winding, 703
sampling points, 362 sharp-edged orifices, 37
sandwich mounting. 153 shock,
scissor lifts, 579 absorbers, (see also vibration) 235
scraper, 316 preventers, 235, 520
screw thread, short courses, 575
piston actuators, 130 shunt-diode safety barriers, 705
rotary actuators. 130 shuttle valves, 163
screwed, SI prefixes, 3
construction. 112, 113 side-loading trucks, 584
unions. 284 Silicone (MQ), 287
-on ends, 111 fluids, 335
seals. 305, 534, 550 single acting cylinders, 373
single vane, surface tension, 15, 22

actuator, 127 surge,
rotary actuator, 126 collectors, 235
six-way valves, 159 damping accumulator, 233
skip loaders, 585 swage coupling, 279
skiving, 298 swing clamps, 601
sliding plate valves, 142 symbols, 5
solenoid operated, 204 synthetic fluids, 334
specialised cylinders, 601 system (s), 670
specific, maintenance, 507
gravity, 19 pressure, 507
heat, 15, 19 malfunction seen at the actuator, 509
speed controls, cylinders, 375 pressure, 72
spiral twisting, 467
spontaneous ignition, 15 T
Spontaneous Ignition Temperature (SIT), 24 tailored fluids, 533
spool valve assembly, 141 tandem,
spring-loaded accumulator, 223 centre valve, 160
stabilisers, 685 circuit,
stability, 437 by-pass valve, 173
stainless steel pipes, 264 configuration, 649
standards, 559 pumps, 64
static, tangent bulk modulus, 20
applications, 12 tank, (see also reservoir), 251
hydmulic systems, 579 sizing, 252
O-ring seals, 505 target cleanliness levels, 360-1
power installation, 485 temperature, 249, 479
seals, 308, 468 terminology, 135
stationary fluids, hydrostatics, 27 testing,
steady state, complete hydraulic systems, 484
analysis, 497 components, 481
errors, 433 cylinders, 483
steel, filters, 355
mill rolls, 231 pipes and tubes, 484
pipes, 263 pumps, 482
tubes, 263 systems, 481
steering, 94, 681 valves, 482
systems, 232 tests,
stepper motor valve drive, 196 properties of fluids, 25
straddle-carriers, 584 thermal,
stretching machines, 631 conductivity, 15, 22
stroke controlled solenoid, 189 expansion, 15, 22
sub-plate mounting, 153 thermography, 494
sub-sea, 689 thermoplastic hose, 292
suction filters, 353 threaded cover styles, 322
Superlip-System, 63, 64 three stage valves, 211
supporting clamps, 271 three-way,
surface damage, 467 regulator, 167
EDITORIAL INDEX 741
valves. 158 V
throttle valves.. 164 V-ring seals, 313
tie-rod, III vacuum-hydraulic brakes, 657
construction. 112 valves, 549
tilted-body pump, 55 amplifier, 633
titanium tubing, 265 as a hole, 133
tools, workshop, 605 closed-centre, 160
torque motor, 205 connection and mounting, 152
Total Acid Number - TAN, 24 construction and connection, 139
Total Base Number - TBN, 24 contamination, 148
total pressure drop, 268 control elements, 140
training, 575 drive, 447
transduction methods, 442 float centre, 160
transfer, functional classification of, 135
function, 433 hydraulic forces in, 149
paths, 512 open-centre, 160
valves, 163 operating forces in, 148
transient positions, 161 prefill, 635
transmission configurations, 404 rapid-motion, 377
triple-action drawing tools, 623 relief, 632
trucks, side-loading, 584 standards, 566, 570
tube, tandem centre, 160
calculations, 259 terminology of, 135
gripping, 275 vibration and noise, 517
fittings, 276 ways, 135
materials, 107, 260, 286 working elements, 139
size calculations, 266 Valvistor, 197
standards, 569 vane,
tungum tubes, 264 actuators, 125
turbulent flow, 30 construction, 127
friction factors for, 32 motors, 89
two stage valves, 208 pumps, 57
two-way valves, 157 vapour pressure, 15, 22,532
variable,
delivery vane pumps, 58
U resistive vector transducer (VRVT), 444
D-ring seals, 312 unit control, 409
ultra-high pressure, sequence-robots, 591
generation of, 537 vehicle,
hydraulics, 537 configurations, 412
measurement of, 538 transmissions, 412
ultrasonics monitoring, 492 vena contracta, 37
underground, venting, 254
conveying, 702 vibration, 511
drilling equipment 697 monitoring, 491
loaders, 699 reduction in, 513
winching, 702 sources of, 512
unloading valve, 184 viscosity, 15, 532
effects of, 249

index, 19
improvers, 329
volume compensation, 231
volumetric efficiency, 34
W
warehousing, 580
waste disposal, 711
water, 330
based glycols, 333
cooled heat exchangers, 240
contamination, 342, 363
hammer, 518
hydraulics, 547
applications of, 551
separability, 24
wear debris analysis, 492
weight-loaded accumulators, 224
welded,
connection, 275
construction, 112
coupling, 282
wheel rim presses, 620
wiper (s), 316
seal, 505
housing, 505
and scrapers, 316
work-holding control, 600
workshop tools, 605
Z
zinc dialkyl dithiophosphate (ZDDP), 329

Hydraulic Handbook PDF

Caricato da

Informazioni sul documento

Titolo originale

Copyright

Formati disponibili

Condividi questo documento

Condividi o incorpora il documento

Opzioni di condivisione

Hai trovato utile questo documento?

Questo contenuto è inappropriato?

Copyright:

Formati disponibili

Hydraulic Handbook PDF

Caricato da

Copyright:

Formati disponibili

THE HYDRAULIC

Copyright 1996 ELSEVIER SCIENCE LTD

All rights reserved

Other engineering publications by Elsevier Advanced Technology include:

Printed in Great Britain by BPC Wheatons Ltd, Exeter

For more information contact:

There are also two Supplementary Units relating to angle

Force = Mass x Acceleration,

(Note: 1 cP = 1 mPa g 1 cSt = 1 mm2 g-l )

and the commonly used units of time

are acceptable: but not acceptable:

NOTATION AND SYMBOLS

THE HYDRAULIC ADVANTAGES

Electrics - by means of electrons at a potential

Is the environment 'unfriendly'?

Figure 1.3.1 is a suggestion of the most commonly acceptable regions of applicability.

Typical hydraulic applications

Examples of hydraulic operations

Example of a hydraulically operated aircraft gangway

Example of a drive for a travelling saw

AN HYDRAULIC fluid has several functions to perform. For instance it must

air release value

Other practical viscosity values

ISO viscosity classification

where Vo = volume at atmospheric pressure

where v = viscosity (cSt).

Alternatively, 'compressibility' may be expressed in terms of relative density, namely.

Spontaneous Ignition Temperature - SIT

Air Release Value

BASIC HYDRAULIC THEORY

HYDRAULIC SYSTEMS are a combination of hydrostatics and hydrodynamics. Hydro-

Hydrostatics - stationary fluids

where g = acceleration due to gravity,

where A = base area.

with a force multiplication of A2 / Ai .

These equations define the performance in a basic hydrostatic system of force

Pumps in hydrostatic systems

Hydrodynamics - moving fluids

where p = fluid density.

Laminar or turbulent flow

where v = mean fluid velocity (m S-I)

Flow through changing cross section and orifices

THE HYDROSTATIC PRINCIPLE

where Np = the pump speed

'The subscript R is used to denote the "retract" stroke of the piston.

In-line piston pumps

Radial piston pumps

Axial piston pumps

In order to increase the performance of axial piston pumps, there is a tendency to

Bent-axis piston pumps

FIGURE 2A.1.7 - Bent-axis piston pump [Denison Hydraulics].

Fixed delivery vane pumps

Variable delivery vane pumps

position. The main requirement is to establish a suitable delivery/eccentricity gradient so

FIGURE 2A.l.11- The Powermaster patented interlocking body [Ultra Hydraulics].

Hydraulic pump selection

Bulk and Weight

Noise and Vibrations

50 MPa (500 bar) - the new benchmark?

PUMP DRIVERS AND COUPLINGS

FIGURE 2A.2.2 - An integrated

Asynchronous electric motor

Internal combustion engines