https://ntrs.nasa.gov/search.jsp?

R=19870006034 2018-08-15T01:21:09+00:00Z

NASA Tedmical Meummdum 88875

Lubricant Effects on Bearing Life

ErwinV. Zarmky
Lewis Research Center
Cleveland, Ohio

N_7-1E_e7

Uncles
G.3/37 q0238

for the
OEM Design '86 Conference
New York, New York, December 9-11, 1986

A
_4 ¸¸ ,

_f
 LUBRICAN1 EFFECIS ON BEARING LIFE

Erwln V. Zaretsky
Chief Engineer for Structures
National Aeronautics and Space Administration
Lewis Research Center
Cleveland, Ohio 44135

SUMMARY

Lubricant considerations for rolling-element bearings have within the

last two decades taken on added importance in the design and operation of
mechanical systems. The phenomenon which limits the usable life of bearings
"F is rolling, element or surface pitting fatigue. 1he elastohydrodynamtc (EHD)
film thickness which separates the ball or roller surface from those of the
raceways of the bearing directly affects bearing life. Chemical additives
I added to the lubricant can also significantly affect bearings life and reli
ability. The interaction of these physical and chemical effects is important
to the design engineer and user of these systems. Design _-thods and lubricant
selection for rolling.element bearings are presented and discussed.

IN1RODUCI!ON

In rolling, element bearing application, the lubricant can have a marked

effect on bearing life and load capacity, figure 1 illustrates the required
points of |ubricant in a ball bearing, lhese lubrication points are the cage-
ball interface, the cage :ace interface and the ball-race contracts.

Most bearing failures occur because of lubricant starvation at these

points or improper lubricant selection. 1his problem becomes more acute with
high-temperature applications.

A lubrlcant has four major functions:

(1) Provide a separating film between rolling and sliding contacting

surfaces, figure 1, thus preventing wear.

(2) Act as a cooldnt to maintain proper bearing temperature.

(3) Prevent the bearing from being contaminated by dirt and other
contaminates.

(4) Prevent corrosion of bearing surfaces.

Oepending on the type of intervening film and its thickness, a number of

lubrication regimes can be identified. A classical way of depicting some of
these regimes is by use of the we11 known Strtbeck curve (fig. 2). Strlbeck
performed comprehensive experiments on Journal bearings around 1900. He meas
t'-ed the coefficient of friction as a function of load, speed, and temperature.
l.a had difficulty, however, condensing this data into usable form. Some years
later, Hersey performed similar experiments and devised a plotting format based
on a dimensionless parameter, lhe Strlbeck curve, or more appropriately, the

L_
 Stribeck-Hersey curve, takes the form of the coefficient of friction as a func-
tion of the viscosity of the ltquld (Z) velocity (N) and load (P) parameter
ZN/P. ' ' .

At high values of ZN/P which occur at htgh speeds, low loads, and at high
viscosity, the surfaces are completely separated by a thick (>0.25 um)
(>10 -5 In.) lubricant film. Thts area ts that of hydrodynamic lubrication
where frlctlon ts determined by the rheology of the lubricant. For noncon-
formal concentratPd contacts where loads are hlgh enough to cause elastic
deformation of tne surfaces and pressure-viscosity effects on the lubricant,
another fluid film regime, elastohydrodynamlc lubrication (EHO), can be Identi-
fied. In thls regime film thickness (hi may range form 0.025 to 2 5 um (10-6
to 10-4 in.).

As film thickness becomes progressively thinner, surface interactions start

taking place. This regime of increasing friction, which combines asperity
interactions and fluid fllm effects, Is referred to as the mixed lubrication
regime.

Finally, at low values of the ZN/P parameter, one enters the realm of
boundary lubrication. This regime Is characterized by the following:

(1) This regime Is highly complex, Involving metallurgy, surface topo-

graphy, physical and chemtcal adsorption, corrosion, catalysts and reaction
kinetics.

(2) The most Important aspect of this regime Is the formation of protec
tive surface films to minimize wear and surface damage.

(3) The formation of these films Is governed by the chemistry of the

film-forming agent, as well as the surface of the solid and other environmental
factors.

(4) 1he effectiveness of these films tn minimizing wear ts determined by

their physical properties, which include shear strength, thickness, surface
adhesion, film cohesion, melting point or decomposition temperature and
solubility.

Besides the Strtbeck Hersey curve (flg. 2) already described, an Idealized

plot of wear rate as a function of relative load can also delineate the various
lubrication regimes and some wear transitions (fig. 3).

Region OA of figure 3 encompasses the regimes of hydrodynamic and EHO

lubrication, the latter as point A Is approached. Since no surface inter
actions occur In thts region except for startup or shlJtdown, little or no wear
occurs. (1his excludes rolling-element fatigue, which can occur without sur-
face interactions.) Region AX Is the mixed lubrication regime where surface
interactions begin to occur at A and become more prevalent as point X ts
approached. Wear Is low because fluid film effects still exist.

Next there ts region XY In figure 3, which ts the region of boundary

lubrication. The degree of metal to-metal contact and the wear rate increase
_s the load increases. Wear Is mtld and tends to be corrosive to the left of
B and adhesive to the right of B. 1he location of B ts quite variable and
depends on the corroslvlty of the lubricant formulation, for a noncorrosive
lubricant, adhesive wear can occur at X. On the other hand, a corrosive addi-
tive can extend the boundary regime to Z' before boundary film failure occurs.
Region YZ is the regime of severe wear where severe adhesion and scoring occur.
Machinery cannot operate successfully in this region, and, therefore, the loca-
tion of this transition point Is quite important. At point Z total surface
failure occurs, followed by seizure.

In the boundary lubrication regime many properties of t._e liquid lubricant

become important. These include shear strength, ftlm thickness, melting point,
and chemical reactivity with the surface. Operating variables which will affect
lubricant film-performance tnclude load, speed, temperature, and atmo phere, as
already discussed. Additives present in the lubricant to serve specific func-
tlons will also affect behavior. These additives Include antlwear, antlfoam,
antloxtdants, viscosity improvers, and others.

SYMBOLS

a contact ellipse radius transverse to rolling direction, m (in.)

b contact ellipse radius In rolling direction, m (in.)

0 ball or roller diameter, m (in.)

d inner ring track diameter, m (in.)

E modulus of elasticity parameter, N/m2 (psi)

E' material elasticity parameter, N/m2 (psi)

f conformity

6 dimensionless EHO materials parameter

Hc dimensionless central EHD film thickness

Hmin dimensionless minimum EHD film thickness

hc central EHO film thickness, m (in.)

hmln minimum EHO fllm thickness, m (in.)

k contact ellipse ratio, a/b

N speed

P dimensionless pressure

P bearing pitch diameter, p = d , O, m (in.)

p, pressure, N/m2 (psi)

O normal force, N (|b)

3
RA,RB ro111ng radius, m (in.)

Rx equivalent radius In rolling direction, m (in.)

Ry equivalent radius transverse to rolling direction, m (in.)

r radius of body, m (in.)

U average surface velocity in x direction, m/see (ln./sec)

UD dimensionless EHD speed parameter

UA,UB surface velocity of body In x direction, m/sec

kl dimensionless EHD load parameter

x rolling direction coordinate

Y transverse dlrecttorl coordinates

Z viscosity

_p presseire-visco_lty coefficient, m2/N (psi -1 )

13 contact angle, deg

nu absolute viscosity of fluld at ambient prassure N sec/m2

(lb sec/ln.2)

),. ratio of film thickness to composite roughness

Polsson's ratio

P inverse curvature sum, m-1 (In.-1)

o composite surface roughness, m (in.)

OA, oB surface roughness of body, m (in.)

angular rotation velocity, rad/sec

Subscripts:

elastlc bodies

cage

inner race

outer race

reference planes

bodies 1 and 2
 ELASTOHYDROOYNAMIC$ LUBRICATION

Most ro111ng-element bearings have between their rolllng elements and

raceways an elastohydrodynamlc f11m separating the contacting bodies during
motlon. This Is Illustrated fnr the rollers of flgure 4, whose surfaces shown
In the enlarged vlew of the contact are not In direct contact but are separated
by a hlghly compressed, extremely thln lubrlcant f11m. Because of the hlgh
pressures In the contact zone the lubrlcatlon process Is accompanied by some
elastlc deformation of the contact surface shown In figure 5. For thls reason,
this process Is referred to as elastohydrodynamlc (EHD) lubrlcant. Grubin was
among the flrst to Identlfy this phenomenon, which also occurs for other oll-
lubrlcated, rol1_ng machlne elements such as gears. Contour plots for a dlmen-
sionless fllm thickness and pressure are shown in figure 6 for a ball on a flat
plated

Hamrock and Oowson have derived a simplified approach to calculating the

EHD film thickness. They provide dimensionless groupings as follows (refer to
flg. 5):

Dimensionless minimum fllm thickness:

hmtn
Hmln - R (l)
X

which occurs at the tratllng edge of the contact. Dimensionless central film
thickness:

h
H =C
c R (2)
x

which Is the average fllm thickness across the entire contact, lhe eQulvalent
radius In the rolllng direction Rx is given by

1 1 1
- + ---- (3)
Rx rA, x rB,x

and where the equivalent radius transverse to the rolllng direction can be
found from

l l l

Ry - rA,y + rB ,Y (4)

Dimensionless speed parameter:

noU
UD = (s)
X
 where the average _urface speed of bodies A and B is

uA + uB
U

2 (6)

and where the reduced modulus nf e|astlctty ts

E_ = 2
(7)

01menslonless load parameter:

w = --q--
E'R (8)
X

0tmenstonless materials parameter:

G = =pE' (9)

Contact e111pse ratto:

a
k =; (_0)

where a and _ are the radll of the contact ellipse, _lth radius a
oriented perpendicular to the rolllng direction x.

Referring to equations (3) and (4) the equivalent radll for roller and
ball bearings are given In figure 7. Also, from this flgure, the entrainment
velocity or average surface speed U at the inner and outer raceway contact
can be derived for equatton (5) where for a bearing inner race

and at the outer rac=

Uo = p.24p
_ 0 2) (mo - wt) (i,)

Since for most applications the outer raceway does not rotate and wo equal
O, then

U1 = Uo = U (12)

The conventional method of calculating the contact ellipse ratio is to

ftno a solutton to a transcend3ntal equation that relates the ellipse ratto and
the elllpttc Integrals of the first and second kinds to the geometry of the
| contacting soltds. Thts Is usually accomplished by some lterattve numerical

L
_l_lj_w11_." _lF,,._-_'_ -. , _.-. -

procedure. The following stmple expression for k, which eliminates the neces-
sity for that procedure, was derived by Brewe and HaFnrock:

(13)

for 0.01 <_ Ry/R x <_ 100,

The approximate solutton of the ellipse ratto as obtatnzd from equatlon (13)
_s wtthln 3 percent of the exact solutton for k between 0.056 and 18. For
Ry/Rx greater than unity the major diameter of the contact e111pse w111 be
orlented perpendicular to the rolling direction.

The Influence of k and the dimensionless speed Uo, load N, and m_te-
rials 6 parameters on the minimum and central ftlm thicknesses was Investi-
gated theoretically for the viscous-elastic regtme by Hamrock and Oowson. Th_
ellipse ratto was varied from 1 to 8, the dimensionless speed oarameter was
varied over nearly two orders of magnitude, and the dimensionless load param-
eter was varled over one order of magnitude. Conditions corresponding to the
use of solid materials of bronze, steel, and slllcon nltrlde and lubricants of
parafflnlc and naphthentc otls were considered tn obtaining the exponent on
the dimensionless materials parameter. Thirty-four cases were used In obtain-
Ing the following dimensionless minimum-film-thickness formula:

Hml n = 3.63 0 68 G0 "49M-0.073

UO" (1 e -0.68k) (14)

For pure cyllndrlcal roller contact (I.e., llne contact) k = ® and the term
In the parentheses In equatlon (14) equals I.

The procedure used to obtaln the central f11m thlckness Is the same as
that used to obtaln the mlnlmum f11m thlckness and results In the followlng
formula:

0"67GO'53w-O'O67 (1 0.51 e -0.73k) (1S)

Hc = 2.6g U0

lhe measure of the effectlveness of the lubrlcant f11m Is the "lambda" (x)
ratlo (I.e., hc/o, central f11m thlckness dlvlded by the composlte surface
roughness of the rolllng-element surfaces). Usually the root-mean-square (rms)
surface flnlshes of the contactlng bodles oA and oB are used to determlne
the composlte surface roughness as follows:

I/2
2 2
o = (o A + oB) (16)

lhe _ ratlo can be used as an Indlcator of rolllng-element performance

and llfe. For k < l, surface smearlng or deformatlon, accompanled by wear,
will occur on the rolllng surface. For 1 < _ < 1.5, surface dlstress may be
accompanled by superflclal surface plttlng. For 1.5 < X < 3, some surface
glazlng can occur wIth eventual roller fallure caused by classlcal subsurface-
orlglnated rolllng-element fr_tlgue. Figure 8 shows the effect of X on
rolllng-element fatlgue llfe. At k _ 3, mlnlmal wear can be expected wlth
m

| extremely long ltfe; fatlure will eventually be by classical subsurface-

' originated rolling-element fatigue. The most expedient, although not the least
; expensive, way of attaining high x ratio ts to select a high-quality surface
finish.

BOUNDARY LUBRICATION

Extreme-pressure and anttwear additives In the lubricating fluid form a

fllm on the surfaces by a chemical reaction, absorption, or chemtsorptlop.
lhese boundary films can be thinner than 0.025 pm (1 pin.) or several micro-
inches thick. These films are formed from the chemical reaction of sulfur or
from the chemlsorptton of tron stearate. Film thickness varies for various
types of film. The films can separate the metal surfaces when the EHD film
becomes thin enough for the asperities to interact. The boundary film probably
lubricates by mlcroaspertty-elastohydrodynamic lubrication, where the asperi-
ties deform under the load. The boundary fllm prevents contact of the aperl-
ties and at the same time provides low-shear-strength properties that prevent
shearing of the metal and reduce the friction coefficient below that of the
base metal. These boundary films provide lubrication at different temperature
conditions depending on the materials used. For example, some boundary films
wtll melt at a lower temperature than others and will then fail to protect the
surfaces. The "failure temperature" ts the temperature at which the lubricant
film fails. In extreme-pressure lubrication this failure temperature Is the
temperature at which the boundary film melts.

The melttng point or thermal stability of surface ftlms appears to be a

unifying physical property governing failure temperature for a wide range of
materials. It ts based on the observation that only a solid film can properly
interfere wlth potential asperity contacts. For this r_ason many extreme-
pressure lubricants contain more than one chemtcal for protection over a wide
temperature range. For instance, Borsoff found that phosphorous compounds are
superior to chlorine and sulfur at slow speeds, but sulfur ls superior at high
speeds. He explains this as a result of the increased surface temperature at
the higher speeds. (Most extreme-pressure additives are chemically reactive
and increase their chemical actlvlty as temperature is increased.) Horllck
found that some metals such as zinc and copper have to be removed from lubric-
ation systems when uslng certain extreme-pressure additives.

There is strong indication that at k ratios less then 1 1/2, boundary

films can affect the resultant fatigue life of ro111ng surfaces. Rolling-
element fatigue tests were performed by the author and his colleagues in the
NASA five-ball fatigue tester wlth an acid-treated white otl containing either
2.5 percent sulfurtzed terpene, 1 percent dtdoderyl phosphlte or 5 percent
chlorinated wax. Wlth the exception of the chlorinated wax additive, these
additives shewed essentially no statistical differences between the lives using
the base oll with the additive and those without the additive. ]he presence of
the chlorinated wax produced surface distress and a significant reduction In
llfe.

The additives us_ , the base oil did not change the life ranking of the
bearing steels In those .sts where rolling-element fatigue was of subsurface
origin. That is, regardless of the additive content of the lubricant, the
lives of the three bearing steels tested ranked in descending order as follows:
A[S! 52100, A[S[ M-50, AIS! 1018. Recent published work by the authors and hts
 .................... ° ................... _-Z _._LL_.'_.L.. _-

collegues showed that with spur gears, where X Is approximately 1, the

presence of a phosphorous-type load carrying additive In the lubricant produced
a significant improvement tn life over lubricants without this type of addi-
tive. The presence of a sulphus type anttwear additive also showed an improve-
ment In life. However, this improvement was not considered statistically
significant.

The presence of the antlwear and extreme-pressure additive to lubricant

rolling-element bearings ts extremely important especially under heavily
loaded, high-speed, high-temperature conditions. These additives protect the
contacting surface from wear and surface distress even under favorable EHO
lubrication conditions In raceway contacts. In many cases, boundary lubrica-
tion Is dominant between the cage and ball (roller) contact and the cage-race
contact (see fig. 1). Without the additive forming a boundary film, gross
wear can occur.

LUBRICANT SELECTION

lhe useful bulk temperature limits of several classes of fluld lubricants

In an oxidative envlronmenf are given tn table [. Grease lubricants are listed
In Table II. The heat transfer requirements of bearings dictate whether a
grease lubricant can be used. Grease lubrication permits the use of simplified
housing and seals.

lhe most commonly used lubricant Is mineral oil, both tn liquid and grease
form table III. As a liquid, mineral oll usually contains an antlwear or
extreme-pressure (EP) additive, an antlfoam agent, and an oxidation inhibitor.
In grease the antlfoam agent Is not required.

Synthetic lubricants have been developed to overcome some of the harmful

effects of lubricant oxidation. However, synthetic lubricants should not be
selected over readlly available and invariably less-expensive mineral oils If
operating conditions do not require them. It Is usually easier to incorporate
synthetic lubricants In a new design than to convert an existing machine to
thetr use.

1he selection of a lubricant Is not always lndeoendent of the appltcdtlon.

Hence, very ltttle can be done once the machine Is In the fteld. However.
where choice can be made, the designer or the perscn selecting th_ lubricant
should consider the anticipated operating temperature of the bearings. Ix mGs_
cases, the lubricants heat transfer and rheloglcal propert!es wlll affect tPts
temperature. However, a first order estimate can be made and the selection can
be made on the lubricants limiting temperature. Once, the chemical type has
been selected, lubricant viscosity should be a consideration. T_hle III gives
comparative viscosities for various numerical oll classlflc3tlon_. Knowing the
viscosity at the bearing temperature and the pressure viscosity coefficient of
the selected lubricant (_able IV gtve representative values of pressure-
viscosity coefficient _p) the EHD ftlm thickness can be calculated together
with the film parameter X and realtlve bearing life. Where there are too low
values of X or life, consideration may be given to either lowerln_ bearing
temperature, increasing bulk lubricant viscosity and/or changing lubricant type.

I
 LUBRICANT AOOIT[VE SELECTION

Unl!ke the lubricant, It Is almost Impossible to independently select a

lubricant additive package from a given lubricant brand. Whtle the lubricant
chemtcal type Is known, generally the addltl4es contained therein are proprie-
tary to the manufacturer. Unless, there Is economic incentive for the manu-
facturer to blend an additive package for a given use, the machine designer Is
locked Into what ts currently In the market place. Probably no two lubricant
n_lnufacturers supply the same additive package even though supplying the same
base stock. Hence, the major difference between brands x and y wlll be
the additives contained therein.

Some of the extreme-pressure additives commonly used contain one or more

compounds of chlorine, phosphorus, or sulfur or lead soaps. Many chlorine-
containing compounds have been suggested as extreme-pressure additives, but
few have actually been used. Some lubricants are made with chlorine-containing
molecules where the C13°C linkage Is used. For example, either trl (trt-
chioroethyl) or trt (trlchlortert butyl) phosphate additives have shown high
load-carrying capacity. Other chlorine-containing additives are chlr.rlnated
paraffin or petroleum waxes and hexachlorethene.

The phosphorus-contatnlng compounds are perhaps the most commonly used

additives for oils. Some aircraft lubricants have 3 to 5 percent trtcresyl
phosphate or trtbutyl phosphlte as either an extreme-pressure or anttwear agent.
Other phosphorous-containing extreme-pressure agents used tn percentages of 0.1
to 2.0 could be dodecyl dlhydrogen phosphate, delthyl-dlbutyl-, or dtcresyl-
phenyl trtchioroethyl phosphtte and a phosphate ester containing a pentachlorop
phenyl radical. Host of the phosphorous compounds tn olls also have other
active elements.

The sulfur-contalnlpg extreme-pressure additives are belleved to form

Iron sulfide films that prevent wear at high loads and speeds. However, they
give higher friction coefficients and are therefore usually supplemented by
o_iler boundary-film-forming Ingredients that reduce friction. The sulfur com-
pounds should have controlled chemical activity (e.g., otls cont(_lntng dlbenzyl
disulfide of 0.1 or m_re percent). Other sulfur-containing extreme-pressure
additives are dtamyl d_ulflde, dllauryl disulfide, suifurlzed oletc actd and
sperm oll mixtures, and dtbutylxanthtc acid disulfide.

Lead soaps have been used In lubricants for many years. They resist the
slldtng action In bearings and help prevent corrosion of steel In the preseLice
of water. Some of the leaG soaps used tn lubricants ace lead oleate, lead
ftshate, lead-12-hydroxystf:arate, and lead naphthenate. Lead naphthepate ts
used most often used because of Its solubility. Lead soaps are used In concen-
trations of 5 to 30 percent.

Other additive compounds contain combinations of these elements. Host

extreme-pressure lubricants contain more than one extreme-pressure a_dltlva
Needless to say, the selection of a proper extreme-pressure additive ts a com-
pllcated process. The word "susceptibility" Is frequently used with reference
to addltlve_ tn o11s to lndlcaLe the ability of the otl to accept the additive
wtLhout deleterious effects. Such properties as solubility, volatility, sta-
bility, compatibility, load-carrying capacity, and cost must be considered.
Many oll compounds depend on the use of proprietary, or package, _xtreme-
pressure aOdltlves. AS a result, the lubricant manufacturer does not evaluate

10

L
 the additives' effectiveness. Because of this, any selection of extreme-
pressure additives should be supported by an evaluation program to determine
their effectiveness for a given application.

CONCLUSION

lhe selection of a lubricant for a rolling-element bearing application is

as important as the actual bearing design. In fact, good engineering practice
dictates that both the design of the bearing and lubricant selection be inte-
grated. Lubricant type should be selected based upon lubricant temperature
limitations. Mtneral oll type lubricants should be selected over the synthetic
lubricants where temperatures permit. Elastohydrodynamlc (EHD) lubrication
film thickness should be calculated. Effects of the EHD film on bearing llfe
should be determined. Additive selection while not always specifiable, should
be considered and test programs initiated where boundary lubrication is antic-
ipated. Good rolling-element bearing lubrication practice is not "off-the-
shelf" oll can application.

REFERENCES

Anderson, W.J.: Elastohydrodynam!c Lubrication Theory as a Design Param-

eter for Rolling-Element Bearings. ASME Paper 70-DE-19, May 1970.
2,

Anderson, W.J.; and Zaretsky, E.V.: Rolling-Element Bearings. Mach.

Des., vol. 42, no. 15, June 18, 1970, pp. 21-37.

Bamberger, E.N., et. al.: Life AdJubtment Factors for Ball and Roller
Bearings - An Engineering Design Guide. ASNE, 1971.

Beerbower, A.: Boundary Lubrication. GRU.1GBEN.72, Esso Research and

Engineering Co., June 1972.

Borsoff, V.A.: Funaamentals of Gear Lubrication. Annual Rep., Shell

Development Co., June 1955.

Borsoff, V.A.; and Lulwack, R.: F:mdamentals of Gear Lubrication. Shell

Development Co., June 1951.

Bowden, F.P.; and Tabor, O.: lhe Friction and Lubrication c,f Sollds,
Vol. 2, Clarendon Press, Oxford, 1964, p. 365.

B°

Brewe, D.; and Hamrock, B.J.: Simplified Solution for Stresses and
Deformations. J. Lubr. Techno1., vol. lOS, no. 2, Apr. 1983, FP. 171-17l.
,

Buckley, D.H.: Trlbology. Advanced Materlals Technology, NASA CP-225),

C.P. 6}ankenshlp and L.A. Telchman, eds., 1982, pp. 391-409.

10.
Coy, J.J.; Townsend, D.P.; and Zaretsky, E.V.: Gearing. M_chanlcal
Design and Systems Handbook, H.A. Rothbart, ed. McGraw-Hlll NY 1986,
PP. 38.59-38.66. ' ' '

11
11. Felr, R.S.: Chemistry In Concentrated-Conjunction Lubrication. Inter-
disclp;_nary Approach to the Lubrication of Concentrated Contacts, NASA
SP-237, P.M. Ku, ed., 1970, pp. 489-527.

12. Godfrey, O.: Boundary Lubrication. Interdisciplinary Approach to Fric-

tion and Near, NASA SP-181, P.M. Ku, ed., ]968, pp. 335-384.

13. Grubtn, A.N.: Fundamentals of the Hydrodynamic Theory of Lubrication of

Heavily Loaded Cylindrical Surfaces. Investigation of the Contact of
Machine Components, Kh.F. Ketova, Ed., Translation of Russian Book No. 30,
Central Scientlf%c Institute for Technology and Mechanical Engineering,
_oscow, 1949, Chapter 2. (Available from Department of Scientific and
Industrial Research, Great Britain, Transl. CTS-235 and Special Libraries
Association, Transl. R-3554.)

14. Hamrock, 6.3.; and Anderson, N.J.: Rolling-Element Bearings. Mechanical

Destgn and Systems Handbook, H.A. Rothbart, ed., McGraw H111, NY,
pp. 29-5!-29.55.

15. Hamrock, B.J.; and Oowson, O.: Isothermal Elastohydrodynam%c Lubrication

of Point Contacts-Part III-Fully Flooded Results. J. Lubr. Technol.,
vol. 99, no. 2, Apr. 1977, pp. 264-276.

lb. Hersey, M.D.: The 'aws of Lubrication of Horizontal Journal Bearings.

J. Wash. Acad. Sc%., vol. 4, 1914, pp. 542-552.

17. Horlick, E.J.; and Thomas, O.E.: Recent Experiences tn the Lubrication
of Naval Gearing. Gear Lubrlcat%on Symposium, Institute of Petroleum,
1966.

lB. Jones, W.R. Jr., et al.: Pressure-Viscosity Measurements for Several

Lubricants to 5.5x108 Newtons per Square Meter (8x104 psi) ana 149 °C
(300 °F). ASLE lrans., vol. 1B, no. 4, 1975, pp. 249-262.

19. Lowenthal, S.H.; and Zaretsky, E.V.: lractton Drives. Mechanical nes_grl
and Systems Handbook, H.A. Rothbart, ed., McGraw Hill, NY, 1986,
pp. 34.7-34.9.

20. Present, D.L., et al.: Advanced Chem_,.al Characterization and Physical

Properties of Eleven Lubricants. (AFLRL-166, Army Fuels and I_ubr%cants
Research I_ab) NASA CR-1681BT, 1983.

21. Stribeck, R.: Characteristics of Plain and Roller Bearings. VOI Z.,
vol. 46, 1902.

22. lownsend, D.P.; Zaretskyo E.V.; and Sc%bbe, H.W.: Lubricant and Additive
Effects on Spur Gear Fatigue Life. J. lr%bo%oqy, voi. 108, no. 3, July
1986, pp. 468-477.

23. Zaretsky, E.V.; and Anderson, N.J.: How to Use What We Know About EHO
Lubrication. Mach. Oes., vol. 40, no. 26, Nov. ?, 1968, pp. 167-173.

12
 T_RLF I. - L!IRRICANT

SpPcificatinn _II_k fPmD_r_tllr_

TVDp limit in air
i

E _ineral nil _IL-L-6nRI

I qi_stpr m!L-L-7qqq 14q+ i_qo.

Tvp_ If pstpr _IL-L-ORAQq p]o 14oq
T r i_<,tpr _II-L-QORAR

Suporr_fined and c, vnth_t ir

mi n,_ral Oil<,
F lu3rnc,_rb,qn "I 9QR K_q

P,_ ] vph,_n V _ pt h_,r _ %1_ _qn

1_;,_t r#rr_nlmonded.

• AP, r _ Cr)c.',crq

*r

Alrrr_ft '_I!-q-?Q?_n _ 4A t_ +Sqq

/hlqh SDr_rl; hell - _q tn aq_ I ,;,_r q,_ !rhnn

._nl rr_ll#r bP_rinq_

,_Irrr _ft -]qq *n _"qq rhi, k,,nin,: tq,_t,

(SyrlthPt _r _ ,_t r,_m__

pr,_cqur,, qvnfh_tir _ilq, ,_r

_i_t_lrc-q n_ FP

t,b _ : t i v,_

,_lrr ctf t uIl_q__]]64 -]qq tq +'_qq qlmiltr tq v I _ "))_7

f q'tqt h,'* _r ; 7t tn I?] plllq 'Am'\ _
,_v_ I ,,,!_(lPri,yn q 1 fi,l _ f I b_

&lrrr_ft • A_ _n +lqq
_ q,] +,_ 177 qv_/fh,_t jr hv,_r,_r trh,_n

r tnq,, '

_TL q- mr, I_

(,Y,, _ 1 : _t Inq

:_ ] 'VI VtI'v _

!,ltq,_li'/,, _n,1 i_ll

r,, i,t int_

&lr'_ r'tft )_ _ , +_,i"_ ,h_,,_,_,,,,,, I _,!,,,t i,-_

r*,q_%t I_lt L

r_l]} _n, t rqlh,r ',,,irin,)

I
i

t
( ,-,ID,,F 1 t ,irn
.'_ t r._,n., hlqhI 1
 TABLE 3.- MINERAL OIL CLASSIFICATION AND CF)MPARATIVE VISCnSITIES

Cateqorv Am.----T- Approximate viscosity cS

il
3R "C (Inn "F) QQ "C (?In "F_
qear oil

3-5.5

_._-l?
1_-I_

lq-.P4

?(I-35

Extra ---

liqht |OW .... ] 45 4.?

.il 5 --- 61-75

Liqht ?qW .... p 69 5.7

46_ --- gn-lln ___

Medium 30W
......... In
l?_ ---
7nn --- 13n-I _6 ---

Medium 40
......... 11
heavy --- .... 4 lq3 ---

H_avy 5h ......... I?
lq_n l q4-P'_7 1Q

.......... i ---
lSqn --- ?OI-I_R ___
PI_n 417-_)5 ___

.......... _4
14n ......... 25
]l_n ___ f,l_/- 7qq ___

SuDor --- .... h 7

hP_vv .... 4RSq qln-I lPn ---
.... h R ......... -_
?Sq ......... 41
.... hp_ ......... 47
7_qq -- 1 "_7q- ] F,7n ___
.... q .......... 07
.... In : .......... 097

i .... 11 i ......... 4R4

]qr]dp numhpr is PQuiv=iIpnt t_ aw, rtq_ C%vbnlt 'lnlv_rs%l Vi_cnKitv

(S',!S/ _t "r (]qq *gl.
hJqmDqqn,l(_,'t wi th *-art v ni ].

In.74_lq-41
 BALL-RACE ROLLING

,_ NITH SLIDING

r '_ -I
SLIDING .

FIGURE 1.- REQUIRED POINTS OF LUBRICATION IN A BEARING.

 H - 10 .6 - 10-4 IN.

H-IO IN.
_h

H > 165 IN.

j-- I
0.15
I
I
I
COEFFICIENT
OF
I
FRICTION

0.001
-r
I
I
ZN = (VISCOSITY) (VELOCITY)
P LOAD

FIGURE 2.- COEFFICIENT OF FRICTION AS FUNCTION OF VISCOSITY-VELOCITY-LOAD

PARAMETER (STRIBECK-HERSEY CURVE).

SE IZURE I

SEVERE _,_

e_

TRANSITION m

#-
l
f
BOUNDARY-7
i
i
EL AS[OHYDRO-

, / L'L-,./I I
0// A X B y
HY_O-
RELATIVE LOAD
DYNAMIC

FIGURE ).- WEAR RATE AS A FUNCTION OF RELATIVE LOAD

DEPICTING VARIOUS REGIMES C',F LUBRICATION,
 ORIGINAL PA_. _
OF POOR QUALITY

il'-r_ .- HIGH CONTACT

l/lit.i<. ,,_ss°,_
i I

\lllllJ "-PRESSURIZED
LLJ, <o..,c,.,

I1

FIGURE q.- ELASTOHYDRODYNAJ'IICFIL_I BETWEEN TWO BODIES IN ROLLING

CONTACT.

BODY A Q
PLANE X (PLANE
ROLLING)

_RA, y

G PLANE Y

1 _,,._,_s:k ,-_ 1 CONTACT

..1 _k ELLIPSE
ROLL I NG i/_

DIRECIION//,,___ J

O LB Y

FIGURE C_ _ GEOMETRY OF CONTACTING SOLID ELASTIC BODIES.

 D IMENS IONLESS
PRESSURE,
P = P2E'

I .Ix10 -3
1.6
1.5
1.4
1.2
1.0
.7
.3

_"L_.""

(A) PRESSURE.

O H
. . --_ HERTZ IAN DIMENSIONLESS

" CIRCLE FILM THICKNESS,

H = h/R x
/_. x'
A 4.3xI0 -6
B q.G
C 5.0
\ D 5,5
E G.O
F G.G
G 7.4
H 8.2

'\ . /.."

"'..,,.. _;_;'

(B) FILM THICKNESS.

FIGURE G.- CONTOUR PLOTS OF DIMENSIONLESS PRESSURE

FILM THICKNESS. k = 1.25_ U = 0.168x10-11;
W = 0.111x10-G; G = Q522.
 6

o ©
INNER £ACE CONTACI OUTER RACE CONTACT

Rx_ 2(d D+ D) 2 Rx° : o,°,._(_,o(,._)

2(d + D) 2

(A) EQUIVALENT RADII FOR A ROLLER BEARING.

l
D cos I)
?

___L_o
INNER RACE CONTACT OUTER RACE _.ONTACT

D (p _ D cos _) _ D (p + D cos _)
Rx_ 2P RX° 2P

fiD fo D
Ry) ;)El ' Ryo
?fo I

(R) EQUIVA{ENI RADII fOR A BAIL BFARING,

FIGURF /.- G(O_II:IRy AND i:QUIVAIFNT RADII FOR BAli AND

RPA. I [R B(AR INGS.

®
 3.5 --

3.0

c_ 2.5
I-,

IJ.

2.0

U.

..J

; 1.5
,O

l,--

+,,,J

_ 1.0

.5

o ,l,l . L i l i I ,Ill
.G .8 1 2 q G 8 10
RATIO OF MINIMUM FILM THICKNESS TO COMPOSITE
SURFACE ROUGHNESS, X

FIGURE 8.- LUBRICATION-LIFE CORRECTION FACTOR AS FUNC-

T ION OF _.

,)
 T
1. Report No. / 2 . Government Accession No. 3. Reciplent's Catalog No.

NASA TH-88875 1
4. Title and Subtitle 5. Report Date

Lubricant Effects on Bearing Life

6. P_o_lng Organl_tion Code

505-63-11
7. Aut_qs) 8. PedDling Organization Report No.

Erwln V. Zaretsky E-3253

10. Work Unit No.

9. P_orming Or_ni_tion Name a_ Address

11. Contract or Grant No.
National Aeronautics and Space Administration
Lewis Research Center
Cleveland, Ohio 44135 13. Type of Report and Period Covered

12. Sponsoring A_ncy Name a_ A_ress

Technical Remorandum
National Aeronautics and Space Administration
14. Sponsoring Agency Code
Washington, D.C. 20546

15. Supplementa_ Notes

Prepared for the OER Design '86 Conference, New York, New York, December 9-11,
1986.

16. Abstract

Lubricant considerations for rolling-element bearings have wlthtn the last two
decades taken on added importance in the destgn and operation of mechanical sys-
tems. The phenomenon which llmlts the usable life of bearings Is rolling-element
or surface plttlng fatigue. The elastohydrodynamlc (EHO) fllm thickness which
separates the ball or roller surface from those of the raceways of the bearing
dlrectly affects bearlng life. Chemical additives added to the lubricant can
also significantly affect bearlngs life and reliability. The Interaction of
these physical and chemical effects Is Important to the deslgn englneer and user
of these systems. Design methods and lubricant selectlon for rolling-element
bearings are presented and discussed.

17 Key Words iSuggested by Authors)) 18. _lstrlbutlon Statement

Rolling-element bearings; Elastohydro- Unclassified - unlimited

dynamics; Lubrication; Rolling-element STAR Category 37
fatigue

19 Security Clssstf, (of this report) Price"

No, of pages
Unclassified -_20. Unclassified
Security Clssslf. (of this paoe) 2t,

"For sale by the Nahona; Technical Information Service, Springfield, Virginia 22161