Electro slag welding

http://www.twi.co.uk/j32k/protected/band_3/jk3.html

Description
Electro slag welding is a very efficient, single pass process carried out in the vertical or near vertical position and used for joining steel plates/sections in thicknesses of 25mm and above. It was developed by the aton Institute in the !kraine in the early "#5$s and superseded the very high current submerged arc process for making longitudinal welds in thick%walled pressure vessels. !nlike other high current fusion processes, electro slag welding is not an arc process. &eat re'uired for melting both the welding wire and the plate edges is generated through a molten slag(s resistance to the passage of an electric current. In its original form, plates are held vertically appro)imately *$mm apart with the edges of the plate cut normal to the surface. + bridging run%on piece of the same thickness is attached to the bottom of the plates. ,ater cooled copper shoes are then placed each side of the joint, forming a rectangular cavity open at the top. -iller wire, which is also the current carrier, is then fed into this cavity, initially striking an arc through a small amount of flu). +dditional flu) is added which melts forming a flu) bath which rises and e)tinguishes the arc. .he added wire then melts into this bath sinking to the bottom before solidifying to form the weld. -or thick sections, additional wires may be added and an even distribution of weld metal is achieved by oscillating the wires across the joint. +s welding progresses, both the wire feed mechanism and the copper shoes are moved progressively upwards until the top of the weld is reached. /ee figure 1.

Fig.1. Electro slag welding

.he consumable guide variant of the process uses a much simpler set%up and e'uipment arrangement which does not re'uire the wire feed mechanism to climb. In this case, the wire is delivered to the weld pool down a consumable, thick%walled tube which e)tends from the top of the joint to the weldpool. /upport for the molten bath is provided by two pairs of copper shoes which are moved upwards, leapfrogging each other as welding progresses. .he tubular guides can be further supplemented by additional consumable plates attached to the tube. 0enerally, as the

"

thickness of plate increases, the number of wires/guides increases, appro)imately in the ratio of one wire per 5$mm of thickness, see figure 2.

Fig.2. onsumable guide welding

urrent status
In the fabrication industry, the process continues to be used for thick walled pressure vessels which are post%weld normalised and for structures such as blast furnace shells and steel ladles which are used at above ambient temperatures. .he process is also e)tensively used for the welding of railway points.

!mportant current issues

1onsiderable interest was shown in electro slag welding during the "#2$s when ideas for increasing welding speed were investigated. .his was seen as an important parameter for increasing productivity and as a way of reducing heat input to improve &+3 and weld metal impact properties. &owever, since that time little has been done by way of development. .hose developments that have taken place have been limited to the tuning of parameters and tailoring techni'ues for specific applications.

"ene#its
.he principal benefits of the process are4 speed of joint completion5 typically " hour per metre of seam, irrespective of thickness lack of angular distortion lateral angular distortion limited to *mm per meter of weld high 'uality welds produced simple joint preparation, i.e. flame%cut s'uare edge major repairs can be made simply by cutting out total weld and re%welding

$isks
Electro slag welding is not one of the major welding processes because the high heat input generates large, coarse grained weld metal and &+3s which lead to poor fracture toughness properties in these areas. .oughness improvements can only be achieved by post%weld normalising treatment. +dditionally, the near parallel%sided geometry of the weld, combined with the coarse grains, can make it difficult to identify defects at the fusion boundary by standard ultrasonic 67. techni'ues. .he process has considerable potential for increasing productivity. &owever, its use has been limited because of relatively poor understanding of the process and, for specific applications, the significance of the fracture toughness values. +s a result, use of the process has been restricted to a few niche applications.

%olid wire &!' welding

8etal inert gas 98I0: welding was first patented in the !/+ in "#;# for welding aluminium. .he arc and weld pool formed using a bare wire electrode was protected by helium gas, readily available at that time. -rom about "#52 the process became popular in the !< for welding aluminium using argon as the shielding gas, and for carbon steels using 1= . 1= and argon%1= mi)tures are known as metal active gas 98+0: processes. 8I0 is an attractive alternative to 88+, offering high deposition rates and high productivity.
2 2 2

(rocess characteristics
8I0 is similar to 88+ in that heat for welding is produced by forming an arc between a metal electrode and the workpiece5 the electrode melts to form the weld bead. .he main differences are that the metal electrode is a small diameter wire fed from a spool and an e)ternally supplied shielding gas is necessary. +s the wire is continuously fed, the process is often referred to as semi% automatic welding. &etal trans#er mode .he manner, or mode, in which the metal transfers from the electrode to the weld pool largely determines the operating features of the process. .here are three principal metal transfer modes4 /hort circuiting 7roplet / spray ulsed /hort%circuiting and pulsed metal transfer are used for low current operation while spray metal transfer is only used with high welding currents. In short%circuiting or(dip( transfer, the molten metal forming on the tip of the wire is transferred by the wire dipping into the weld pool. .his is achieved by setting a low voltage5 for a ".2mm diameter wire, arc voltage varies from about "2> 9"$$+: to 22> 92$$+:. 1are in setting the voltage and the inductance in relation to the wire feed speed is

essential to minimise spatter. Inductance is used to control the surge in current which occurs when the wire dips into the weld pool. -or droplet or spray transfer, a much higher voltage is necessary to ensure that the wire does not make contact i.e.short%circuit, with the weld pool5 for a ".2mm diameter wire, the arc voltage varies from appro)imately 22> 925$+: to *5> 9;$$+:. .he molten metal at the tip of the wire transfers to the weld pool in the form of a spray of small droplets 9about the diameter of the wire and smaller:. &owever, there is a minimum current level, threshold, below which droplets are not forcibly projected across the arc. If an open arc techni'ue is attempted much below the threshold current level, the low arc forces would be insufficient to prevent large droplets forming at the tip of the wire. .hese droplets would transfer erratically across the arc under normal gravitational forces. .he pulsed mode was developed as a means of stabilising the open arc at low current levels i.e. below the threshold level, to avoid short%circuiting and spatter. 8etal transfer is achieved by applying pulses of current, each pulse having sufficient force to detach a droplet. /ynergic pulsed 8I0 refers to a special type of controller which enables the power source to be tuned 9pulse parameters: for the wire composition and diameter, and the pulse fre'uency to be set according to the wire feed speed.

%hielding gas
In addition to general shielding of the arc and the weld pool, the shielding gas performs a number of important functions4 forms the arc plasma stabilises the arc roots on the material surface ensures smooth transfer of molten droplets from the wire to the weld pool .hus, the shielding gas will have a substantial effect on the stability of the arc and metal transfer and the behaviour of the weld pool, in particular, its penetration. 0eneral purpose shielding gases for 8I0 welding are mi)tures of argon, o)ygen and 1= , and special gas mi)tures may contain helium. .he gases which are normally used for the various materials are4
2

steels 1= argon ?2 to 5@ o)ygen argon ?5 to 25@ 1= non%ferrous o argon o argon / helium

o o o
2 2 2

+rgon based gases, compared with 1= , are generally more tolerant to parameter settings and generate lower spatter levels with the dip transfer mode. &owever, there is a greater risk of lack of fusion defects because these gases are colder. +s 1= cannot be used in the open arc 9pulsed or spray transfer: modes due to high back%plasma forces, argon based gases containing o)ygen or 1= are normally employed.
2

)pplications
;

8I0 is widely used in most industry sectors and accounts for more than 5$@ of all weld metal deposited. 1ompared to 88+, 8I0 has the advantage in terms of fle)ibility, deposition rates and suitability for mechanisation. &owever, it should be noted that while 8I0 is ideal for (s'uirting( metal, a high degree of manipulative skill is demanded of the welder.

*he &anual &etal )rc process

8anual metal arc welding was first invented in Aussia in "BBB. It involved a bare metal rod with no flu) coating to give a protective gas shield. .he development of coated electrodes did not occur until the early "#$$s when the <jellberg process was invented in /weden and the Cuasi%arc method was introduced in the !<. It is worth noting that coated electrodes were slow to be adopted because of their high cost. &owever, it was inevitable that as the demand for sound welds grew, manual metal arc became synonymous with coated electrodes. ,hen an arc is struck between the metal rod 9electrode: and the workpiece, both the rod and workpiece surface melt to form a weld pool. /imultaneous melting of the flu) coating on the rod will form gas and slag which protects the weld pool from the surrounding atmosphere. .he slag will solidify and cool and must be chipped off the weld bead once the weld run is complete 9or before the ne)t weld pass is deposited:. .he process allows only short lengths of weld to be produced before a new electrode needs to be inserted in the holder. ,eld penetration is low and the 'uality of the weld deposit is highly dependent on the skill of the welder.

*+pes o# #lu,/electrodes
+rc stability, depth of penetration, metal deposition rate and positional capability are greatly influenced by the chemical composition of the flu) coating on the electrode. Electrodes can be divided into three main groups4 1ellulosic Autile Dasic ellulosic electrodes contain a high proportion of cellulose in the coating and are characterised by a deeply penetrating arc and a rapid burn%off rate giving high welding speeds. ,eld deposit can be coarse and with fluid slag, deslagging can be difficult. .hese electrodes are easy to use in any position and are noted for their use in the (stovepipe( welding techni'ue. Features:

deep penetration in all positions suitability for vertical down welding reasonably good mechanical properties high level of hydrogen generated % risk of cracking in the heat affected Eone 9&+3:

$utile electrodes contain a high proportion of titanium o)ide 9rutile: in the coating. .itanium o)ide promotes easy arc ignition, smooth arc operation and low spatter. .hese electrodes are general purpose electrodes with good welding properties. .hey can be used with +1 and 71 power sources and in all positions. .he electrodes are especially suitable for welding fillet joints in the horiEontal/vertical 9&/>: position. Features:

moderate weld metal mechanical properties good bead profile produced through the viscous slag positional welding possible with a fluid slag 9containing fluoride: easily removable slag

"asic electrodes contain a high proportion of calcium carbonate 9limestone: and calcium fluoride 9fluorspar: in the coating. .his makes their slag coating more fluid than rutile coatings % this is also fast%freeEing which assists welding in the vertical and overhead position. .hese electrodes are used for welding medium and heavy section fabrications where higher weld 'uality, good mechanical properties and resistance to cracking 9due to high restraint: are re'uired. Features:

low hydrogen weld metal re'uires high welding currents/speeds poor bead profile 9conve) and coarse surface profile: slag removal difficult

&etal powder electrodes contain an addition of metal powder to the flu) coating to increase the ma)imum permissible welding current level. .hus, for a given electrode siEe, the metal deposition rate and efficiency 9percentage of the metal deposited: are increased compared with an electrode containing no iron powder in the coating. .he slag is normally easily removed. Iron powder electrodes are mainly used in the flat and &/> positions to take advantage of the higher deposition rates. Efficiencies as high as "*$ to ";$@ can be achieved for rutile and basic electrodes without marked deterioration of the arcing characteristics but the arc tends to be less forceful which reduces bead penetration.

(ower source
Electrodes can be operated with +1 and 71 power supplies. 6ot all 71 electrodes can be operated on +1 power sources, however +1 electrodes are normally used on 71.

-elding current
,elding current level is determined by the siEe of electrode % the normal operating range and current are recommended by manufacturers. .ypical operating ranges for a selection of electrode siEes are illustrated in the table. +s a rule of thumb when selecting a suitable current level, an electrode will re'uire about ;$+ per millimeter 9diameter:. .herefore, the preferred current level for a ;mm diameter electrode would be "F$+, but the acceptable operating range is ";$ to "B$+.

-hat.s new
.ransistor 9inverter: technology is now enabling very small and comparatively low weight power sources to be produced. .hese power sources are finding increasing use for site welding where they can be readily transported from job to job. +s they are electronically controlled, add%on units are available for .I0 and 8I0 welding which increase the fle)ibility. Electrodes are now available in hermetically sealed containers. .hese vacuum packs obviate the need for baking the electrodes immediately prior to use. &owever, if a container has been opened or damaged, it is essential that the electrodes are redried according to the manufacturer(s instructions.

&o/ing contact arc welding

b+ -a+ne *homas

Description and e,planation

In moving contact arc welding 981+,:, current is supplied to a shaped metal consumable electrode through a sliding or rolling contact tool. .he lying consumable 9known as a Aidgeback consumable: is flu)%covered and features an e)posed metal ridge that protrudes above the flu). Electrical contact between the tool and the consumable is applied at a relatively short distance from the arc, eliminating the resistive heating path of the electrode.
.8

.he 81+, techni'ue is under development and offers an easy%to%use alternative cladding, welding and repair method. .he process is suitable for either manual or mechanised operation. It can be applied with restricted access and is amenable to remote operation. Fig 1. "asic principle o# & )- using $idgeback consumable
*&

(resent status o# & ).he process is novel in that it differs from the traditional -irecracker 9Elin%&afergut: and other lying consumable welding techni'ues because the contact point is kept as close as possible to the burning arc. .his minimises resistive heating between the contact point and the arc, enabling the

Aidgeback consumable to carry greater currents, or, almost infinite lengths to be used in one operation. Aestricted access and remote operation capability are two of the major advantages uni'ue to the 81+, process.
.8

+lthough the feasibility has been demonstrated, the 81+, process has not yet been used commercially. !nderwater welding under dry conditions in a local habitat is a possible application. .he practicability of restricted access conditions has been demonstrated using a local habitat chamber. In addition, repair welding in haEardous surroundings 9e.g. in the nuclear industry: and welding or repair inside steel tubes and pipes are likely to provide ideal application for the process.

!mportant issues
Ease of manufacture of profiled consumables by hot rolling or e)trusion, etc and application of the flu) covering on consumables in a more cost%effective way are important development issues in successful commercialisation of the process. + further research re'uirement will be to develop the process to become positional, i.e. for vertical and overhead applications. + potential variant of the 81+, techni'ue is sub%arc in which a plain bar stock or powder filled consumable is used. .he sub%arc variant would not need a specially shaped rod, and the consumables need not be flu) covered. .he techni'ue would make use of a hopper feed, granulated flu) and insulated flu) saddles. .he sub%arc variant would eliminate the need for any consumable feed system. Fig 2. )ssessing the #easibilti+ o# mo/ing contact arc welding 0& )-1

"ene#its
otentially, the 81+, process has many beneficial features compared with e)isting processes which utilise consumables. .he process gives greater productivity as the number of start/stops is reduced, and longer and/or larger consumables can be used in one welding operation. /top/starts can be sites for welding defects such as cold laps5 reduction in stop/starts reduces this associated risk. .he use of larger consumables can reduce/eliminate the need for multi%pass welding of thick plate.

81+, uses low%cost, portable e'uipment and is readily automated. It is simple to operate and is essentially a low%skill welding/cladding techni'ue with less associated operator fatigue. It is ideal for applying hardfacing and corrosion overlays and can be designed to operate in restricted access.

$isks
.ypically, the risks associated with traditional arc welding e'uipment will also apply to 81+,.

(lasma -elding
(rocess characteristics
lasma welding is very similar to .I0 as the arc is formed between a pointed tungsten electrode and the workpiece. &owever, by positioning the electrode within the body of the torch, the plasma arc can be separated from the shielding gas envelope. lasma is then forced through a fine%bore copper noEEle which constricts the arc. .hree operating modes can be produced by varying bore diameter and plasma gas flow rate4

&icroplasma: $." to "5+. .he microplasma arc can be operated at very low welding currents. .he columnar arc is stable even when arc length is varied up to 2$mm. &edium current: "5 to 2$$+. +t higher currents, from "5 to 2$$+, the process characteristics of the plasma arc are similar to the .I0 arc, but because the plasma is constricted, the arc is stiffer. +lthough the plasma gas flow rate can be increased to improve weld pool penetration, there is a risk of air and shielding gas entrainment through e)cessive turbulence in the gas shield. 2e+hole plasma: over "$$+. Dy increasing welding current and plasma gas flow, a very powerful plasma beam is created which can achieve full penetration in a material, as in laser or electron beam welding. 7uring welding, the hole progressively cuts through the metal with the molten weld pool flowing behind to form the weld bead under surface tension forces. .his process can be used to weld thicker material 9up to "$mm of stainless steel: in a single pass.

(ower source
.he plasma arc is normally operated with a 71, drooping characteristic power source. Decause its uni'ue operating features are derived from the special torch arrangement and separate plasma and shielding gas flows, a plasma control console can be added on to a conventional .I0 power source. urpose%built plasma systems are also available. .he plasma arc is not readily stabilised with sine wave +1. +rc reignition is difficult when there is a long electrode to workpiece distance and the plasma is constricted, 8oreover, e)cessive heating of the electrode during the positive half%cycle causes balling of the tip which can disturb arc stability.

/pecial%purpose switched 71 power sources are available. Dy imbalancing the waveform to reduce the duration of electrode positive polarity, the electrode is kept sufficiently cool to maintain a pointed tip and achieve arc stability.

)rc starting
+lthough the arc is initiated using &-, it is first formed between the electrode and plasma noEEle. .his (pilot( arc is held within the body of the torch until re'uired for welding then it is transferred to the workpiece. .he pilot arc system ensures reliable arc starting and, as the pilot arc is maintained between welds, it obviates the need for &- which may cause electrical interference.

Electrode
.he electrode used for the plasma process is tungsten%2@thoria and the plasma noEEle is copper. .he electrode tip diameter is not as critical as for .I0 and should be maintained at around *$%F$ degrees. .he plasma noEEle bore diameter is critical and too small a bore diameter for the current level and plasma gas flow rate will lead to e)cessive noEEle erosion or even melting. It is prudent to use the largest bore diameter for the operating current level. 6ote4 too large a bore diameter, may give problems with arc stability and maintaining a keyhole.

(lasma and shielding gases

.he normal combination of gases is argon for the plasma gas, with argon plus 2 to 5@ hydrogen for the shielding gas. &elium can be used for plasma gas but because it is hotter this reduces the current rating of the noEEle. &elium(s lower mass can also make the keyhole mode more difficult.

)pplications
&icroplasma welding 8icroplasma was traditionally used for welding thin sheets 9down to $." mm thickness:, and wire and mesh sections. .he needle%like stiff arc minimises arc wander and distortion. +lthough the e'uivalent .I0 arc is more diffuse, the newer transistorised 9.I0: power sources can produce a very stable arc at low current levels. &edium current welding ,hen used in the melt mode this is an alternative to conventional .I0. .he advantages are deeper penetration 9from higher plasma gas flow:, and greater tolerance to surface contamination including coatings 9the electrode is within the body of the torch:. .he major disadvantage lies in the bulkiness of the torch, making manual welding more difficult. In mechanised welding, greater attention must be paid to maintenance of the torch to ensure consistent performance. 2e+hole welding .his has several advantages which can be e)ploited4 deep penetration and high welding speeds. 1ompared with the .I0 arc, it can penetrate plate thicknesses up to l$mm, but when welding using a single pass techni'ue, it is more usual to limit the thickness to Fmm. .he normal methods is to use the keyhole mode with filler to ensure smooth weld bead profile 9with no undercut:. -or thicknesses up to "5mm, a vee joint preparation is used with a Fmm root face. + two%pass techni'ue is

"$

employed and here, the first pass is autogenous with the second pass being made in melt mode with filler wire addition. +s the welding parameters, plasma gas flow rate and filler wire addition 9into the keyhole: must be carefully balanced to maintain the keyhole and weld pool stability, this techni'ue is only suitable for mechanised welding. +lthough it can be used for positional welding, usually with current pulsing, it is normally applied in high speed welding of thicker sheet material 9over * mm: in the flat position. ,hen pipe welding, the slope%out of current and plasma gas flow must be carefully controlled to close the keyhole without leaving a hole.

%ubmerged3arc -elding

.he first patent on the submerged%arc welding 9/+,: process was taken out in "#*5 and covered an electric arc beneath a bed of granulated flu). 7eveloped by the E = aton Electric ,elding Institute, Aussia, during the /econd ,orld ,ar, /+,(s most famous application was on the .*; tank.

(rocess #eatures
/imilar to 8I0 welding, /+, involves formation of an arc between a continuously%fed bare wire electrode and the workpiece. .he process uses a flu) to generate protective gases and slag, and to add alloying elements to the weld pool. + shielding gas is not re'uired. rior to welding, a thin layer of flu) powder is placed on the workpiece surface. .he arc moves along the joint line and as it does so, e)cess flu) is recycled via a hopper. Aemaining fused slag layers can be easily removed after welding. +s the arc is completely covered by the flu) layer, heat loss is e)tremely low. .his produces a thermal efficiency as high as F$@ 9compared with 25@ for manual metal arc:. .here is no visible arc light, welding is spatter%free and there is no need for fume e)traction.

4perating characteristics

""

/+, is usually operated as a fully%mechanised or automatic process, but it can be semi%automatic. ,elding parameters4 current, arc voltage and travel speed all affect bead shape, depth of penetration and chemical composition of the deposited weld metal. Decause the operator cannot see the weld pool, greater reliance must be placed on parameter settings.

(rocess /ariants
+ccording to material thickness, joint type and siEe of component, varying the following can increase deposition rate and improve bead shape. -ire /+, is normally operated with a single wire on either +1 or 71 current. 1ommon variants are4 twin wire triple wire single wire with hot wire addition metal powder addition +ll contribute to improved productivity through a marked increase in weld metal deposition rates and/or travel speeds. Flu, -lu)es used in /+, are granular fusible minerals containing o)ides of manganese, silicon, titanium, aluminium, calcium, Eirconium, magnesium and other compounds such as calcium fluoride. .he flu) is specially formulated to be compatible with a given electrode wire type so that the combination of flu) and wire yields desired mechanical properties. +ll flu)es react with the weld pool to produce the weld metal chemical composition and mechanical properties. It is common practice to refer to flu)es as (active( if they add manganese and silicon to the weld, the amount of manganese and silicon added is influenced by the arc voltage and the welding current level. .he the main types of flu) for /+, are4 "onded #lu,es % produced by drying the ingredients, then bonding them with a low melting point compound such as a sodium silicate. 8ost bonded flu)es contain metallic deo)idisers which help to prevent weld porosity. .hese flu)es are effective over rust and mill scale. Fused #lu,es % produced by mi)ing the ingredients, then melting them in an electric furnace to form a chemically homogeneous product, cooled and ground to the re'uired particle siEe. /mooth stable arcs, with welding currents up to 2$$$+ and consistent weld metal properties, are the main attraction of these flu)es.

)pplications
"2

/+, is ideally suited for longitudinal and circumferential butt and fillet welds. &owever, because of high fluidity of the weld pool, molten slag and loose flu) layer, welding is generally carried out on butt joints in the flat position and fillet joints in both the flat and horiEontal%vertical positions. -or circumferential joints, the workpiece is rotated under a fi)ed welding head with welding taking place in the flat position. 7epending on material thickness, either single%pass, two%pass or multipass weld procedures can be carried out. .here is virtually no restriction on the material thickness, provided a suitable joint preparation is adopted. 8ost commonly welded materials are carbon% manganese steels, low alloy steels and stainless steels, although the process is capable of welding some non%ferrous materials with judicious choice of electrode filler wire and flu) combinations.

*!' -elding
.ungsten inert gas 9.I0: welding became an overnight success in the "#;$s for joining magnesium and aluminium. !sing an inert gas shield instead of a slag to protect the weldpool, the process was a highly attractive replacement for gas and manual metal arc welding. .I0 has played a major role in the acceptance of aluminium for high 'uality welding and structural applications.

(rocess characteristics
In the .I0 process the arc is formed between a pointed tungsten electrode and the workpiece in an inert atmosphere of argon or helium. .he small intense arc provided by the pointed electrode is ideal for high 'uality and precision welding. Decause the electrode is not consumed during welding, the welder does not have to balance the heat input from the arc as the metal is deposited from the melting electrode. ,hen filler metal is re'uired, it must be added separately to the weldpool.

(ower source
.I0 must be operated with a drooping, constant current power source % either 71 or +1. + constant current power source is essential to avoid e)cessively high currents being drawn when the electrode is short%circuited on to the workpiece surface. .his could happen either deliberately during arc starting or inadvertently during welding. If, as in 8I0 welding, a flat characteristic power source is used, any contact with the workpiece surface would damage the electrode tip or fuse the electrode to the workpiece surface. In 71, because arc heat is distributed appro)imately one%third at the cathode 9negative: and two%thirds at the anode 9positive:, the electrode is always negative polarity to prevent overheating and melting. &owever, the alternative power source connection of 71 electrode positive polarity has the advantage in that when the cathode is on the workpiece, the surface is cleaned of o)ide contamination. -or this reason, +1 is used when welding materials with a tenacious surface o)ide film, such as aluminium.

)rc starting
.he welding arc can be started by scratching the surface, forming a short%circuit. It is only when the short%circuit is broken that the main welding current will flow. &owever, there is a risk that the electrode may stick to the surface and cause a tungsten inclusion in the weld. .his risk can be minimised using the (lift arc( techni'ue where the short%circuit is formed at a very low current level. "*

.he most common way of starting the .I0 arc is to use &- 9&igh -re'uency:. &- consists of high voltage sparks of several thousand volts which last for a few microseconds. .he &- sparks will cause the electrode % workpiece gap to break down or ionise. =nce an electron/ion cloud is formed, current can flow from the power source. Note: As HF generates abnormally high electromagnetic emission (EM), wel ers shoul be aware that its use can cause interference es!ecially in electronic e"ui!ment# As EM emission can be airborne, li$e ra io wa%es, or transmitte along !ower cables, care must be ta$en to a%oi interference with control systems an instruments in the %icinity of wel ing# &- is also important in stabilising the +1 arc5 in +1, electrode polarity is reversed at a fre'uency of about 5$ times per second, causing the arc to be e)tinguished at each polarity change. .o ensure that the arc is reignited at each reversal of polarity, &- sparks are generated across the electrode/workpiece gap to coincide with the beginning of each half%cycle.

Electrodes
Electrodes for 71 welding are normally pure tungsten with " to ;@ thoria to improve arc ignition. +lternative additives are lanthanum o)ide and cerium o)ide which are claimed to give superior performance 9arc starting and lower electrode consumption:. It is important to select the correct electrode diameter and tip angle for the level of welding current. +s a rule, the lower the current the smaller the electrode diameter and tip angle. In +1 welding, as the electrode will be operating at a much higher temperature, tungsten with a Eirconia addition is used to reduce electrode erosion. It should be noted that because of the large amount of heat generated at the electrode, it is difficult to maintain a pointed tip and the end of the electrode assumes a spherical or (ball( profile.

%hielding gas
/hielding gas is selected according to the material being welded. .he following guidelines may help4 )rgon % the most commonly%used shielding gas which can be used for welding a wide range of materials including steels, stainless steel, aluminium and titanium. )rgon 5 2 to 67 82 % the addition of hydrogen to argon will make the gas slightly reducing, assisting the production of cleaner%looking welds without surface o)idation. +s the arc is hotter and more constricted, it permits higher welding speeds. 7isadvantages include risk of hydrogen cracking in carbon steels and weld metal porosity in aluminium alloys. 8elium and helium/argon mi,tures % adding helium to argon will raise the temperature of the arc. .his promotes higher welding speeds and deeper weld penetration. 7isadvantages of using helium or a helium/argon mi)ture is the high cost of gas and difficulty in starting the arc.

)pplications
.I0 is applied in all industrial sectors but is especially suitable for high 'uality welding. In manual welding, the relatively small arc is ideal for thin sheet material or controlled penetration 9in the root run of pipe welds:. Decause deposition rate can be 'uite low 9using a separate filler rod: 88+ or 8I0 may be preferable for thicker material and for fill passes in thick%wall pipe welds. .I0 is also widely applied in mechanised systems either autogenously or with filler wire. &owever, several (off the shelf( systems are available for orbital welding of pipes, used in the manufacture of ";

chemical plant or boilers. .he systems re'uire no manipulative skill, but the operator must be well trained. Decause the welder has less control over arc and weldpool behaviour, careful attention must be paid to edge preparation 9machined rather than hand%prepared:, joint fit%up and control of welding parameters.

*he o,+acet+lene process

(rocess #eatures
=)yacetylene welding, commonly referred to as gas welding, is a process which relies on combustion of o)ygen and acetylene. ,hen mi)ed together in correct proportions within a hand% held torch or blowpipe, a relatively hot flame is produced with a temperature of about *,2$$ deg.1. .he chemical action of the o)yacetylene flame can be adjusted by changing the ratio of the volume of o)ygen to acetylene. .hree distinct flame settings are used, neutral, o)idising and carburising.

6eutral flame

=)idising flame

1arburising flame

,elding is generally carried out using the neutral flame setting which has e'ual 'uantities of o)ygen and acetylene. .he o)idising flame is obtained by increasing just the o)ygen flow rate while the carburising flame is achieved by increasing acetylene flow in relation to o)ygen flow. Decause steel melts at a temperature above ",5$$ deg.1, the mi)ture of o)ygen and acetylene is used as it is the only gas combination with enough heat to weld steel. &owever, other gases such as propane, hydrogen and coal gas can be used for joining lower melting point non%ferrous metals, and for braEing and silver soldering.

E9uipment
=)yacetylene e'uipment is portable and easy to use. It comprises o)ygen and acetylene gases stored under pressure in steel cylinders. .he cylinders are fitted with regulators and fle)ible hoses which lead to the blowpipe. /pecially designed safety devices such as flame traps are fitted between

"5

the hoses and the cylinder regulators. .he flame trap prevents flames generated by a (flashback( from reaching the cylinders5 principal causes of flashbacks are the failure to purge the hoses and overheating of the blowpipe noEEle. ,hen welding, the operator must wear protective clothing and tinted coloured goggles. +s the flame is less intense than an arc and very little !> is emitted, general%purpose tinted goggles provide sufficient protection.

4perating characteristics
.he action of the o)yacetylene flame on the surface of the material to be welded can be adjusted to produce a soft, harsh or violent reaction by varying the gas flows. .here are of course practical limits as to the type of flame which can be used for welding. + harsh forceful flame will cause the molten weld pool to be blown away, while too soft a flame will not be stable near the point of application. .he blowpipe is therefore designed to accommodate different siEes of (swan neck copper noEEle which allows the correct intensity of flame to be used. .he relationship between material thickness, blowpipe noEEle siEe and welding speed, is shown in the chart. ,hen carrying out fusion welding the addition of filler metal in the form of a rod can be made when re'uired. .he principal techni'ues employed in o)yacetylene welding are leftward, rightward and all%positional rightward. .he former is used almost e)clusively and is ideally suited for welding butt, fillet and lap joints in sheet thicknesses up to appro)imately 5mm. .he rightward techni'ue finds application on plate thicknesses above 5mm for welding in the flat and horiEontal%vertical position. .he all% positional rightward method is a modification of the rightward techni'ue and is ideally suited for welding steel plate and in particular pipework where positional welding, 9vertical and overhead: has to be carried out. .he rightward and all% positional rightward techni'ues enable the welder to obtain a uniform penetration bead with added control over the molten weldpool and weld metal. 8oreover, the welder has a clear view of the weldpool and can work in complete freedom of movement. .hese techni'ues are very highly skilled and are less fre'uently used than the conventional leftward techni'ue.

"F

utting with a carbon dio,ide laser

b+ (aul 8ilton

(rinciples o# laser cutting

,hether laser cutting with 1= or 6d4G+0 lasers, the principles employed are basically the same. .he beam from the laser is focused on to the surface of the material being cut by means of a lens..he focused laser beam heats the material surface and a very local melt capillary is 'uickly established throughout the depth of the material. .he diameter of this capillary is usually just slightly greater than the diameter of the focused laser beam. .he great majority of 1= laser cutting is performed using an assist gas. .he significant feature of gas assisted laser cutting is that the molten material is ejected from the base of the capillary by a jet of gas coa)ial with the laser beam. -or some materials this gas can further assist the process by chemical 9e)othermic: reaction as well as physical work. .he cut is generated by either moving the focused laser beam across the surface of the stationary material or by keeping the laser beam stationary and moving the workpiece. &ybrids of these two options are also possible. In this way simple and comple) linear cuts or two dimensional parts can be produced. 8ore comple) systems are re'uired for three dimensional processing.
2 2

8ost laser cutting with 1= lasers is performed in the power range " % ".5k,.
2

%tatus o# the process

.he first ever (gas assisted( laser cuts were made in "#F2 in a joint e)periment between .,I and /EAH 9/ervices Electronics Aesearch Haboratory:. /EAH had, by the end of "#FF, developed a *$$, pulsed slow flow 1= laser 9for military reasons: but were also looking for potential industrial uses. .his development was only two years after the first reporting of lasing action from the 1= molecule, and probably marks the start of the laser materials processing industry as we know it today.
2 2

/heet metal cutting has since become, by far, the dominant industrial use of lasers in materials processing. It is believed that appro)imately "2 $$$ industrial laser cutting systems have been installed world%wide, with a total market value of some ;.5 billion !/ dollars. =ver F$@ of this e'uipment is installed in Iapan. .oday, laser cutting is used e)tensively for producing profiled flat plate and sheet, for diverse applications in the engineering industry sectors. -or three dimensional components, multi%a)is gantry laser beam manipulators have e)tended laser cutting to the automotive sector, this type of e'uipment being used for trimming pre%production body panels at all leading car manufacturers. 8ore recently laser cutting has also found its way, very successfully, into other industry sectors such as shipbuilding, traditionally seen as fairly slow to adopt high technology processes. 8etals, ceramics, polymers and natural materials such as wood and rubber can all be cut using 1= lasers. -or steels the dominant process utilises an o)ygen assist gas, which provides e)othermic energy to the cutting process. +s a result, thick sections 9up to 2$mm: can be cut commercially and the cut 'uality and speed are generally considered high when compared with other thermal cutting processes. Haser cutting is also generally regarded as a (low distortion( process, compared with other
2

"2

thermal cutting options. /tainless steel, aluminium and titanium are also cut using 1= lasers, this time using a high pressure 9up to "5 bars: inert assist gas to aid the process and blow material from the cut kerf. Hess cutting of thermoplastic materials is performed currently because of the nature of the fume generated when some plastics are vaporised.
2

urrent issues
1urrent research in 1= laser cutting involves, on one hand, use of very high beam 'uality lasers 9capable of being focused to small spots of high power density: for high speed cutting of thin metals 9over ";$m/min for $.25mm thick material:. =n the other hand, research is developing techni'ues to increase the thickness of material that can be cut. In this respect steel up to F$mm thick has been cut with moderate 9J5k,: laser power.
2

1ommercial development of 1= laser cutting systems involves use of special focusing lenses, employing twin foci, to improve cut 'uality on non%ferrous metals, and engineering improvements to the machines to improve acceleration and speed, to maintain development of the process itself.
2

+pplication issues involve4 cut 'uality, use of laser cuttable steels, thick section cutting and component nesting technology.

(recautions
.he 1= lasers used for laser cutting are (class ;( and as such, current production level machinery is configured to prevent human access to laser beams during normal operation. .he 1= laser beams can cause serious burns to human tissue and permanent damage to the eyes. 1= lasers also employ high voltage power supplies and there is an associated electrical haEard, with a subse'uent risk of electric shock, during service and maintenance. =ther haEards involve those of moving machinery and fume products from the cutting process.
2 2 2

*hermal 'ouging
.hermal gouging is an essential part of welding fabrication. !sed for rapid removal of unwanted metal, the material is locally heated and molten metal ejected % usually by blowing it away. 6ormal o)yfuel gas or arc processes can be used to produce rapid melting and metal removal. &owever, to produce a groove of specific dimensions, particularly regarding depth and width, the welder must e)ercise careful control of the gouging operation. If this does not happen, an erratic and badly% serrated groove will result. .hermal processes, operations and metals which may be gouged or otherwise shaped4 *hermal process =)yfuel gas flame 8anual metal arc (rocess operations (rimar+ %econdar+ &etals

0rooving How carbon steels, carbon manganese steels 9structural:, 0ouging ,ashing pressure vessel steels 9carbon not over $.*5@:, low alloy steels 1hamfering 9less than 5@1r: cast iron 9if preheated to ;$$%;5$ deg.1: 0ouging 0rooving How carbon steels carbon manganese steels 9structural:, pressure 1hamfering vessel steels, low alloy steels, stainless steels, cast iron, nickel%

"B

based alloys How carbon steels carbon manganese steels 9structural:, pressure +ir carbon 0rooving vessel steels, low and high alloy steels, cast iron, nickel%based 0ouging arc 1hamfering alloys, copper and copper alloys, copper/nickel alloys, aluminium 1hamfering lasma arc 0ouging 0rooving +luminium, stainless steels ,ashing 6ote4 +ll processes are capable of cutting/severing operations. reheat may or may not be re'uired on some metals prior to gouging

%a#et+
It should be emphasised that because gouging relies on molten metal being forcibly ejected, often over 'uite large distances, the welder must take appropriate precautions to protect himself, other workers and his e'uipment. /ensible precautions include protective clothing for the welder, shielding inside a specially%enclosed booth or screens, ade'uate fume e)traction, and removal of all combustible material from the immediate area.

!ndustrial applications
.hermal gouging was developed primarily for removal of metal from the reverse side of welded joints, removal of tack welds, temporary welds, and weld imperfections. -igure " illustrates the value of typical back%gouging applications carried out on arc welded joints., while -ig. 2 shows imperfection removal in preparation for weld repair.

-ig." .ypical back%gouging applications carried out on arc welded joints

-ig. 2 Imperfection removal in preparation for weld repair

.he gouging process has proved to be so successful that it is used for a wide spectrum of applications in engineering industries4

repair and maintenance of structures % bridges, earth%moving e'uipment, mining machinery, railway rolling stock, ships, offshore rigs, piping and storage tanks removal of cracks and imperfections % blow holes and sand traps in both ferrous and non% ferrous forgings and castings preparation of plate edges for welding removal of surplus metal % riser pads and fins on castings, e)cess weld bead profiles, temporary backing strips, rivet washing and shaping operations, demolition of welded and unwelded structures % site work

"#

.hermal gouging is also suitable for efficient removal of temporary welded attachments such as brackets, strongbacks, lifting lugs and redundant tack welds, during various stages of fabrication and construction work.

'ouging processes
0ouging operations can be carried out using the following thermal processes4 o)yfuel gas flame manual metal arc air carbon arc plasma arc

4,+#uel cutting 3 process and #uel gases

(rocess #undamentals
.he o)yfuel process is the most widely applied industrial thermal cutting process because it can cut thicknesses from $.5mm to 2,5$$mm, the e'uipment is low cost and can be used manually or mechanised. .here are several fuel gas and noEEle design options that can significantly enhance performance in terms of cut 'uality and cutting speed..he cutting process is illustrated in Fig# 1. Dasically, a mi)ture of o)ygen and the fuel gas is used to preheat the metal to its (ignition( temperature which, for steel, is 2$$K1 % #$$K1 9bright red heat: but well below its melting point. + jet of pure o)ygen is then directed into the preheated area instigating a vigorous e)othermic chemical reaction between the o)ygen and the metal to form iron o)ide or slag. .he o)ygen jet blows away the slag enabling the jet to pierce through the material and continue to cut through the material. 9Fig.1. Diagram o# o,+acet+lene cutting process1 .here are four basic re'uirements for o)y%fuel cutting4

the ignition temperature of the material must be lower than its melting point otherwise the material would melt and flow away before cutting could take place the o)ide melting point must be lower than that of the surrounding material so that it can be mechanically blown away by the o)ygen jet the o)idation reaction between the o)ygen jet and the metal must be sufficient to maintain the ignition temperature a minimum of gaseous reaction products should be produced so as not to dilute the cutting o)ygen

2$

+s stainless steel, cast iron and non%ferrous metals form refractory o)ides ie the o)ide melting point is higher than the material, powder must be injected into the flame to form a low melting point, fluid slag.

(urit+ o# o,+gen
.he cutting speed and cut edge 'uality are primarily determined by the purity of the o)ygen stream. .hus, noEEle design plays a significant role in protecting the o)ygen stream from air entrainment. .he purity of o)ygen should be at least ##.5@. + decrease in purity of "@ will typically reduce the cutting speed by 25@ and increase the gas consumption by 25@.

hoice o# #uel gas

-uel gas combustion occurs in two distinct Eones. In the inner cone or primary flame, the fuel gas combines with o)ygen to form carbon mono)ide and hydrogen which for acetylene, the reaction is given by 21 & ? 2=
2 2 2

;1= ? 2&

1ombustion also continues in the secondary or outer Eone of the flame with o)ygen being supplied from the air. ;1=?2& ?*=
2 2

;1= ?2& =
2 2

.hus, fuel gases are characterised by their

flame temperature % the hottest part of the flame is at the tip of the primary flame 9inner cone: fuel gas to o)ygen ratio % the amount of fuel gas re'uired for combustion but this will vary according to whether the flame is neutral, o)idising or reducing heat of combustion % heat of combustion is greater in the outer part of the flame

.he five most commonly used fuel gases are acetylene, propane, 8+ 9methylacetylene% propadiene:, propylene and natural gas. .he properties of the gases are given in the .able. .he relative performance of the fuel gases in terms of pierce time, cutting speed and cut edge 'uality, is determined by the flame temperature and heat distribution within the inner and out flame cones.

)cet+lene
+cetylene produces the highest flame temperature of all the fuel gases. .he ma)imum flame temperature for acetylene 9in o)ygen: is appro)imately *,"F$K1 compared with a ma)imum temperature of 2,B"$K1 with propane. .he hotter flame produces more rapid piercing of the materials with the pierce time being typically one third that produced with propane. .he higher flame speed 92.;m/s compared with *.*m/s for propane: and the higher calorific value of the primary flame 9inner cone: 9"B,B#$kI/m compared with "$,;** kI/m for propane: produce a more intense flame at the surface of the metal reducing the width of the &eat +ffected 3one 9&+3: and the degree of distortion.
* *

2"

(ropane
ropane produces a lower flame temperature than acetylene 9the ma)imum flame temperature in o)ygen is 2,B2BK1 compared with *,"F$K1 for acetylene:. It has a greater total heat of combustion than acetylene but the heat is generated mostly in the outer cone 9 see &able:. .he characteristic appearance of the flames for acetylene and propane are shown in Figs#2 an ' where the propane flame appears to be less focused. 1onse'uently, piercing is much slower but as the burning and slag formation are effected by the o)ygen jet, cutting speeds are about the same as for acetylene. ropane has a greater stoichiometric o)ygen re'uirement than acetylene5 for the ma)imum flame temperature in o)ygen, the ratio of the volume of o)ygen to fuel gas are ".2 to " for acetylene and ;.* to " for propane.

Fig.2. 4c+acet+lene gas jet and no::le design

Fig.3. (ropane gas jet and no::le design

&)((
8+ gas is a mi)ture of various hydrocarbons, principally, methylacetylene and propadiene. It produces a relatively hot flame 92,#2FK1: with a high heat release in the primary flame 9inner cone: 9"5,;;5kI/m :, less than for acetylene 9"B,B#$kIm : but much higher than for propane 9"$,;**kIm :. .he secondary flame 9outer cone: also gives off a high heat release, similar to propane and natural gas. .he combination of a lower flame temperature, more distributed heat source and larger gas flows compared with acetylene results in a substantially slower pierce time.
* * *

22

+s 8+ gas can be used at a higher pressure than acetylene, it can be used for underwater cutting in deep water as it is less likely to dissociate into its components of carbon and hydrogen which are e)plosive.

(rop+lene
ropylene is a li'uid petroleum gas 9H 0: product and has a similar flame temperature to 8+ 92B#FK1 compared to 2,#2FK1 for 8+ :5 it is hotter than propane, but not as hot as acetylene. It gives off a high heat release in the outer cone 922,$$$kI/m : but, like propane, it has the disadvantage of having a high stoichiometric fuel gas re'uirement 9o)ygen to o)ygen ratio of appro)imately *.2 to " by volume:.
*

;atural 'as
6atural gas has the lowest flame temperature similar to propane and the lowest total heat value of the commonly used fuel gases, eg for the inner flame ",;#$kI/m compared with "B,B#$kI/m for acetylene. 1onse'uently, natural gas is the slowest for piercing.
* *

*able : Fuel 'as haracteristics Fuel 'as &a,imum Flame 4,+gen to #uel gas *emperature < $atio 0/ol1 8eat distribution k=/m3 rimary /econdary +cetylene ropane 8+ ropylene &ydrogen *,"F$ 2,B"$ 2,#22 2,B22 2,B*; ".24" ;.*4" *.*4" *.24" $.;24" ".B4" "B,B#$ "$,;** "5,;;5 "F,$$$ % ",;#$ *5,BB2 B5,*25 5F,;*" 22,$$$ % *5,22$

6atural 0as 2,22$

(lasma )rc utting

lasma arc cutting has always been seen as an alternative to the o)y%fuel process. &owever, the important difference between the two processes is that while the o)ygen%fuel process o)idises the metal and the heat from the e)othermic reaction melts the metal, the plasma process operates by using the heat from the arc to melt the metal. .he ability to melt the metal without o)idation is essential when cutting metals, such as stainless steel, which form high temperature o)ides. .he plasma process was therefore first introduced for cutting stainless steel and aluminium alloys. .he first plasma torches gave poor 'uality cuts and the process itself suffered from e)cessive noise and fume, especially when cutting thicker material. =ver the last thirty years, the plasma cutting process

2*

has been highly refined and is now capable of producing high 'uality cuts, at increased speeds, in a wide range of material thicknesses.

(rocess /ariants
.he basic plasma torch 9Fig 1a: consists of a central tungsten electrode for forming the arc but, unlike in the conventional .I0 welding process, the arc is constricted by a fine bore copper noEEle. .his has the effect of increasing the temperature and velocity of the plasma emanating from the noEEle. .he temperature of the plasma is in e)cess of 2$,$$$< and its velocity can approach the speed of sound L"M. .he process variants 9Figs 1b to 1g: have been principally designed to improve the 'uality of the cut and the cutting performance, or to reduce operating costs. .he most important process variants are described below. Fig 1. *orch designs 0a1 gas plasma 0b1 dual gas 0c1 water injection 0d1 water shrouded 0e1 air plasma 0#1 contact cutting 0g1 high tolerance

Dual gas: .he process operates basically in the same manner as the conventional system but a secondary gas shield is introduced around the noEEle 9Fig 1b:. .he cutting gas is normally argon, argon/&2 or nitrogen and the secondary gas is selected according to the metal being cut. -or e)ample, when cutting mild steel, air or o)ygen can be used to increase the cutting speed. -ater injection: ,ater can be injected radially into the plasma arc 9Fig 1c: to induce a greater degree of constriction. .he temperature of the plasma is considerably increased 9*$,$$$K1: which facilitates higher cutting speeds and, because of the greater constriction of the arc, much improved cut 'uality. .he presence of an annular film of water around the plasma will protect the noEEle bore, reducing noEEle erosion. -ater shroud: .he plasma arc can also be operated either with a water shroud 9Fig 1 : or even with the workpiece submerged some 5$ to 25 mm below the surface of the water. .he water will act as a barrier in reducing fume and noise levels. In a specific e)ample of noise levels at high current levels in e)cess of ""5 dD, this can be reduced to about #F dD with a water shroud and 52 to B5 dD when cutting underwater L2M. )ir plasma: .he inert or unreactive plasma forming gas 9argon or nitrogen: can be replaced with air but this re'uires a special electrode of hafnium or Eirconium mounted in a copper holder 9Fig 1e:. .he use of compressed air instead of the more e)pensive cylinder gas, makes this variant of the plasma arc process highly competitive with the o)y%fuel process. + variant of the air plasma process is the monogas torch in which air is used for both the plasma and the cooling gas. It is generally considered that air plasma is more widely applied in general engineering and light fabrication industries, eg, in cutting sheet steel within the thickness range " to 2$ mm L*M. .he more popular 2;

materials are 1%8n and stainless steels but the process has also been used for cutting /0 9spheroidal graphite: iron and non%ferrous materials L;M. -or thin section material of a few millimetres, the process is much faster than o)y%fuel, but at thicknesses approaching *$ to ;$ mm, air plasma becomes relatively slow L5M. .he obvious cost advantages of using air in preference to e)pensive gases 9for the plasma and o)y% fuel processes: must be considered when other operating costs have also been taken into account. -or e)ample, the air must be fed at a relatively high pressure 9typically "5$ "/min at 5 bar: and clean. .his will re'uire an ade'uate siEe compressor for a line feed supply with suitable filters for dust particles and oil. +dditionally, special purpose electrodes will be re'uired and the operating life of the electrodes and noEEles can be severely shortened if there are fre'uent stop/starts LFM. How current air plasma torches, typically less than ;$+, are particularly attractive for cutting thin sheet material, in that compressed air is used for both the plasma forming gas and cooling the torch. 8oreover, the torch head can be held in contact with the surface of the metal being cut 9Fig 1f:, without the risk of damaging the noEEle by forming a secondary or series arc between the electrode/noEEle and the workpiece. +s nitrogen and o)ygen suppress the formation of a series arc, compared to argon, contact cutting can be practised with the air plasma system LFM. .he process is becoming more widely used for manual cutting of thin sheet components in both 1%8n and stainless steel, where contact cutting greatly deskills the cutting operation. 8igh tolerance plasma: In an attempt to improve cut 'uality and to compete with the superior cut 'uality of laser systems, a number of plasma systems are available commercially which operate with a highly constricted plasma arc5 the systems under the generic name high%tolerance plasma arc cutting 9&. +1: are manufactured by &ypertherm, <oike +ronson and <omatsu%1ybernation L2M. .he common features of the torches 9Fig 1g: are that the o)ygen plasma jet is forced to swirl as it enters the plasma orifice and a secondary flow of gas is injected down stream of the plasma noEEle. .he <omatsu%1ybernation torch has a separate magnetic field surrounding the arc which stabilises it by maintaining the rotation induced by the swirling gas. It is claimed that the cut 'uality lies between a conventional plasma arc cut and laser beam cutting, but the cutting speed is significantly lower than conventional plasma arc cutting and appro)imately F$ to B$@ the speed of laser cutting L2M. 1utting speeds can be several m/min for materials with thicknesses up to F mm. &owever, for high 'uality cuts, ie, minimum kerf width, the optimum speed is more typically ".5 m/min for " mm, reducing to about 5$$ mm/min for F mm 91%8n: sheet material. +n e)ample of a high definition system, which is mounted on torch manipulation e'uipment with an accuracy of N$." mm is shown in Fig 2.

)pplications
.he plasma process can be used for cutting a wider range of materials than the o)y%fuel process which is largely restricted to 1%8n steels. It is, therefore, not surprising that the plasma arc process is being used increasingly for cutting materials such as stainless steel 9Fig ':, aluminium and coated steels. Even in cutting 1%8n steel, the plasma process can have advantages over the o)y%fuel process such as a higher cutting speed and a narrow heat affected Eone. -or e)ample, there is a substantial speed advantage for cutting thin section materials of less than *$ mm but this disappears very 'uickly as the plate thickness increases. .he gas in the plasma process has a significant effect on performance and it is generally considered that the best 'uality is achieved with o)ygen for mild steel, and nitrogen or argon/ hydrogen for aluminium and stainless steel LBM. &owever, from an evaluation of gases for cutting 1%8n steel, 25

aluminium and stainless steel, the ma)imum cutting speeds were obtained with air and argon/*5@ &2 compared to nitrogen and argon/"5@ &2 L*M. In a comparison with nitrogen, air plasma was faster for 1%8n steel and aluminium, but slightly slower for stainless steel. .he overall conclusion was that in terms of cut 'uality and cutting speeds, air plasma would normally be recommended for cutting 1%8n steels in preference to gas plasma, whilst gas plasma with argon/"5@ &2 is preferred for cutting stainless steel and aluminium. 1urrent industrial practice for mi)ed gas cutting e'uipment is to use argon 25/*5@ &2 for cutting stainless steel and aluminium, which is a compromise between 'uality and cutting speed. &eavy duty cutting systems can be used for cutting stainless steel up to "*$ mm and aluminium up to "5$ mm. +rgon/&2 mi)tures are recommended, but for materials below 25 mm in thickness, nitrogen, air or argon/&2 can be used LBM. *he ad/antages and disad/antages of plasma cutting are summarised below4

)d/antages
1an be used with a wide range of materials, including stainless steel and aluminium &igh 'uality cut edges can be achieved, e.g. the &. +1 process 6arrow &+3 formed How gas consumable 9air: costs Ideal for thin sheet material How fume 9underwater: process

Disad/antages
Himited to 5$mm 9air plasma: thick plate &igh noise especially when cutting thick sections in air &igh fume generation when cutting in air rotection re'uired from the arc glare &igh consumable costs 9electrodes and noEEles: Gou can assess whether plasma cutting is suitable for your own specific application by imputing your data into the utting %elector *oolkit and comparing it with other processes. .he material in this item is based on 2 articles originally published in ,elding and 8etal -abrication 9+pril and 8ay "##*: entitled (1utting processes the right choice % parts " and 2( by , Hucas and / Aennie.

)brasi/e water jet cutting

+brasive water jet 9+,I: cutting is a uni'ue, (cold( cutting process. .he system uses high pressure water, pressurised either by a reciprocating pump or an intensifier pump, which is passed through a fine bore noEEle to form a coherent, high velocity jet. +lthough a water jet itself has sufficient power to cut non%metallic materials, abrasive particles such as flint or garnet, must be entrained in the jet for cutting metals 9 Fig 1:. .he cutting performance of the e'uipment is governed by a number of factors which include the traverse rate, abrasive type O feed rate, stand%off distance, and noEEle geometry 9angle to the surface is usually #$K:. 2F

+,I cutting of the workpiece results from a combination of erosion, shearing failure under rapidly changing localised stress fields and micro%machining effects, depending on the specific properties of the material5 micro%machining refers to the removal of small amounts of the workpiece material by abrasive particles. Fig 1. 4perating principles o# the abrasi/e water jet process. + close up view of the water jet cutting action in producing a bevel edge in *$mm thick plate is shown in Fig 2. .he cutting action results in a striation effect on the surface of the cut edge, and the drag angle of the striations depends on the relationship between traverse speed and cutting power. +s the cutting speed increases for a given cutting power 9water pressure and abrasive feed rate:, the angle of the striations will increase until through thickness cutting is no longer achieved.

Fig 2. lose up /iew o# abrasi/e water jet cutting process in action.

(rocess /ariants
.here are three distinct types of cutting e'uipment available, characterised mainly by the water pressures used and the abrasive entrainment system employed4

/tandard entrainment 9mi)ing water and abrasive in a chamber immediately adjacent to the cutting noEEle: with operating pressures in the region of "$,$$$ psi 92$$ bar: and a typical noEEle diameter of *mm. /tandard entrainment with operating pressures in the range *$,$$$ % F$,$$$ psi 92,$$$ % ;,$$$ bar: and a typical noEEle diameter of "mm. 7irect entrainment 9an abrasive slurrying system with pressurisation of the water and abrasive in a small pressure vessel: with operating pressures of up to "$,$$$ psi and a typical noEEle diameter of *mm.

.he standard entrainment system with high operating pressure 9up to F$,$$$ psi: and small bore noEEle 9"mm: produces an e)tremely powerful jet which is capable of cutting "$$mm thick steel with a kerf width of appro)imately 2mm. .he process is used for precision cutting applications. 7irect entrainment produces higher energy efficiency which can achieve a five%fold increase in the cutting speed compared to a standard entrainment system at the same operating pressure. .he kerf

22

width is somewhat greater at typically *mm because of the larger 9greater than 2mm: bore noEEles employed. E'uipment costs differ widely, largely depending on the operating pressure involved. +t operating pressures up to ";,5$$ psi, standard reciprocating pumps can be used. &owever, to achieve the very high operating pressures of *$,$$$ % F$,$$$ psi, e)pensive intensifier pumps are re'uired at about double the cost of the lower pressure systems.

)pplications
.he standard entrainment systems are generally used in manufacturing environments for precision cutting of almost any material 9metals and /tellite%clad steel: including those materials which are difficult to cut, e.g., composites and some ceramics . +,I cutting e'uipment is often found in jobbing shops, possibly alongside laser and plasma cutting systems, where a wide range of metal and non%metallic sheet and slab products can be accommodated.
L"%;M

.he direct entrainment system is usually used in the contracting industries such as in the demolition of steel and concrete structures. .he systems have also been employed for difficult cutting operations, often in haEardous environments such as in the repair and upgrade of plant in the chemical, oil and gas, and offshore industries and in the cutting of munitions 9 Fig ':. .his is due to the very low fire risk associated with the process 5 due to the inherent safety of the process, a cold work permit is sufficient in sensitive areas. .he common problem of plant which is clad or lined with non%metallic materials, and re'uiring cutting of both the metallic and non%metallic layers, is easily handled by the +,I process.
L5M

Fig 3. )pplication o# )-= process in the cutting o# munitions. (Courtesy of Defence Research Agency).

.he applications of +,I cutting are very wide ranging and include metals, glass, ceramics, and composites 9both plastics and metal based:, in a wide range of thicknesses. .he principal advantages of water jet cutting compared to thermal and mechanical methods include4
L"M

6o thermal or deformation stresses 6o dulling of the tool 8ulti%directional cutting capability 8inimal or no dust, and 6o fire haEard.

1utting speeds are relatively low compared to other processes5 eg, for 1%8n steel, the cutting speed is typically ;$$mm/min for *mm and only "$mm/min for 5$mm thick plate.

2B

.he most important advantage of the process is its ability to cut materials without heat5 thus avoiding the formation of heat affected Eones in metals, which can, in some instances, result in microcracking in higher strength metallic alloys. .his also applies to plastics and composite materials where thermal cutting often degrades or chars the edges of the material being cut. /andwich materials are being used more e)tensively in industry and, with large differences in melting points, a non%thermal cutting process is the only alternative. .he process is particularly suitable for cutting <evlar and similar materials. &igh 'uality cut edges are produced which re'uire no subse'uent edge finishing operations. .he process is also used in glass cutting and, in the automotive industry, for trimming plastic components used for interior fittings such as dashboards, and also for cutting carpet and sound insulation. .he major advantages and disadvantages of +,I are4

)d/antages

6o heat generated 8inimal distortion Effective for a wide range of metallic and non%metallic materials 1an be used safely in haEardous/e)plosive environments

Disad/antages

&igh capital cost /trict handling controls are re'uired for the safe use of the water jet Effective containment of the water used in the process &igh levels of noise 1ut widths are large 1ost of the abrasive can be high and it cannot be recycled

.he utting %elector *oolkit gives you the opportunity to compare abrasive water jet cutting with other processes on both technical and economic grounds. Gou can input your own data to establish if +,I is a suitable process for your particular application.

2#

Electron beam welding 3 e9uipment

Electron Deam ,elding 9ED,: is a fusion process for joining metals which uses a highly focused beam of electrons as a heat source. !sually the electrons are e)tracted from a hot cathode, accelerated by a high potential % typically *$,$$$%2$$,$$$ volts, and magnetically focused into a spot with a power density of the order of *$,$$$ ,/mm . .his causes almost instantaneous local melting and vaporisation of the workpiece material. .he electron beam is thus able to establish a (keyhole( delivering heat, deep into the material being welded. .his produces a characteristically narrow, near parallel, fusion Eone allowing plain abutting edges to be welded in a single pass for material thicknesses ranging from less than $."mm to greater than 2$$mm.
2

.he process was first developed in 0ermany and -rance in the "#5$(s, but is now well established in all industrial countries. It was initially employed in the nuclear industry for fuel element encapsulations but rapidly spread into the aerospace, automobile and electronics industries. 6owadays it is employed over a very wide variety of applications ranging from, for e)ample, high speed welding of band saw blade materials to the joining of thick section marine components.ED, is used for joining numerous metallic materials including steels, aluminium, copper, nickel, titanium and magnesium alloys and refractory metals. .he process produces high integrity welds, with minimal thermal distortion and freedom from component o)idation. In most materials, the mechanical properties of ED welds are better than welds made with conventional fusion processes. 0enerally filler additions and pre%heat are not re'uired. &owever, in some cases, the addition of filler material or post weld heat treatment may be necessary to develop the full mechanical properties. 8aterial cleaning is essential and for ferritic steels and dissimilar metal combinations it is important to avoid residual magnetism and thermo%electric fields to prevent beam misalignment.Electron beams for welding are normally generated in a relatively high vacuum 9better than 5)"$ mbar: but the workpiece can be housed in a chamber maintained at a coarser vacuum level e.g. 5)"$ to "$ mbar. It is also possible to project high power electron beams into the atmosphere and produce single pass welds in steel in thicknesses of more that ;$mm, but the weld width is typically greater than welds made in vacuum.
%5 %*

Electron beam welding e'uipment typically comprises of an electron gun, high voltage power source, vacuum chamber or enclosure, pumping e'uipment and a workpiece or gun manipulator and control system.ED, e'uipment can be 'uite comple) but is very versatile and although relatively e)pensive compared with say arc welding e'uipment, is capable of economic high volume production.1urrently, the main e'uipment research areas are concerned with improvements in power supply design, e'uipment reliability and longevity, beam characterisation, the generation of high intensity beams at low power and the development of systems for use at pressures approaching one atmosphere 9"$$$ mbar:.

Electron beam welding 3 materials aspects

+ll metallic materials can be melted using a focused electron beam and, in conse'uence, most pure metals and alloys can be successfully welded. Indeed, the only pre%re'uisite is that the materials to be welded are electrically conductive and an earth return path for the electrons is provided during welding, otherwise electrostatic charging occurs. In its most simple form, ED welding is carried out by translating the beam, with respect to the parts to be joined, and locally melting the material. 6o filler addition or consumable is necessary and welding is achieved in a single pass almost

*$

irrespective of the material thickness. 1onse'uently the weld 'uality and properties achievable are controlled by the composition of the material alone. -or simplicity metallic materials can divided into discrete groups based on the elements present in most abundance4 %

%teels and iron allo+s

8ost steels that are weldable by conventional fusion welding processes can be successfully joined using the electron beam process. +lso, because of the narrow thermally strained region that results and the hydrogen free welding atmosphere associated with welding in vacuum, many steels which are otherwise considered difficult or impossible to fusion weld can be joined using ED welding without the need for special consumables or preheating. It is important, however, that steels are specified with low levels of impurities such as sulphur and phosphorus to prevent solidification cracking and that materials are sufficiently well de%o)idised, i.e. degassed or aluminium treated, to minimise the risk of gross weld porosity.

3&n and structural steels

1%8n and structural steels can be joined in a single pass in thicknesses ranging from less than "mm to in e)cess of 2$$mm and provided that certain composition controls are recognised, good weld 'uality can be consistently achieved. .he rapid thermal cycle associated with the process invariably results in welds in steels with overmatched tensile strength and hardness. .hus, it is sometimes necessary to add material to modify the weld metal composition or perform a post weld heat treatment operation if high levels of fracture toughness or low hardness are re'uired.

)llo+ steels
In many applications including aero%engine and automobile transmission parts, components are ED welded in high strength alloy steels and are fre'uently used in the as%welded condition. 6i1r8o steels, for e)ample, and high alloy creep resistant steels can be welded in substantial thicknesses, without preheat. +gain low impurity levels are beneficial particularly if toughness properties are important.

%tainless steels
8ost common types of stainless steel are readily weldable using the ED process including austenitic grades, ferritic, duple) and precipitation hardening martensitic stainless steels. .he duple) and austenitic materials are commonly alloyed with nitrogen and thus welding procedures must be developed which minimise the risk of porosity formation due to nitrogen outgassing and which compensate for the detrimental effect of nitrogen loss on phase balance and stability. .he precipitation hardening grades show a slight degradation in tensile strength when ED welded which can be restored, if re'uired, by a post%weld ageing operation.

ast irons
1ast irons are not generally considered to be readily ED weldable predominantly for metallurgical reasons. ,ith the e)ception perhaps of ductile and spheroidal graphite irons, ED welding is not recommended as a joining process for cast irons.

*"

%o#t iron
/oft iron and silicon iron, used in transformer and electric motor manufacture, are ED welded successfully in a variety of industrial applications.

;ickel allo+s
8any of the popular nickel alloys used in welded fabrication can be joined satisfactorily using the ED welding process. ure nickel, nickel/copper alloys and many nickel/iron alloys can be welded without difficulty. .he comple) high temperature alloys, designed to have good creep resistance at high temperature, can be welded using ED, often in preference to arc welding because of the minimal metallurgical disturbance and low thermal strains induced by the ED process. 1are must be taken, however, to prevent &+3 li'uation during welding and to avoid cracking during post weld heat treatment of the more comple) alloys.

)luminium and magnesium allo+s

,elding of the majority of wrought aluminium and magnesium alloys available commercially can be achieved satisfactorily using the ED process. Evaporation of volatile constituents during welding, particularly in the 2$$$ and 5$$$ series +l alloys, can cause difficulties due to gun flash%overs, loss of alloy content and subse'uent degradation of properties. 1leaning prior to welding is especially important and the majority of weld defects that occur are often a conse'uence of poor cleaning practice. 8any of the cast alloys can also be ED welded although the weld 'uality achievable depends heavily on the 'uality of the casting and, in particular, the residual gas content.

opper and its allo+s

!nlike many of the other thermal processes used for joining pure copper, ED welding can be carried out without the need for any pre%heating operation and can join components of in e)cess of "5$mm thickness in a single pass. /o called (pure( copper may contain impurities such as o)ygen, sulphur and carbon which can compromise its weldability and =-&1 copper or phosphorus de%o)idised grades are preferred. .he majority of copper alloys, with the marked e)ception of the brasses, can be welded but again cast materials can be problematic if the parent material 'uality is poor and residual gas content is high. /ome high strength materials, e.g. those alloyed with Eirconium, can suffer from cracking problems if due care is not e)ercised.

$e#ractor+ and reacti/e metals

.he e)treme power density associated with the electron beam and the ability to work in a vacuum environment make it possible to use the process for joining metals which not only have high melting points but also those which are e)tremely reactive when hot or molten. .itanium and many of its alloys can be welded readily using the ED process without the danger of o)idation and subse'uent undetectable degradation of ductility. -or this reason, the process is used widely in the aero engine industry for welding safety%critical titanium alloy parts. /imilarly Eirconium and its alloys, which are also e)tremely reactive, can be welded without difficulty under vacuum. Hikewise tungsten, tantalum, molybdenum, niobium, vanadium and their alloys can be joined successfully using ED, but again impurity levels can profoundly influence the weld 'uality and properties achievable.

*2

Dissimilar metals
=ne of the particular advantages offered by the ED process is that the beam intensity is such that dissimilar metals with vastly different thermal conductivities and melting points can be welded successfully without preferential melting of the lower melting point material. +lthough not all combinations are possible due to metallurgical incompatibility and the formation of undesirable intermetallic compounds, many dissimilar combinations are possible. It should be noted that thermo%electric currents will be generated whilst welding dissimilar metal combinations which can give rise to strong magnetic fields and deflection of the electron beam. .he severity of this phenomenon is very dependent on the material combination, their magnetic properties and the component geometry. ,here the combination of materials gives rise to embrittlement it is often possible to introduce a mutually compatible transition material or to employ an ED braEing/diffusion bonding approach with an appropriate interlayer.

;on3metals
,hilst welding of non%metals using the ED process is generally not possible, drilling, cutting and etching can sometimes be performed.

$educed pressure electron beam welding

(rocess description
Hike conventional electron beam 9ED: welding, reduced pressure ED welding is performed using a beam of electrons from an electron gun having a power up to "$$k,. .he beam is focused onto the joint%line of the two parts to be welded and either the electron gun or the parts are moved so that the beam traverses along the joint%line, penetrating the full depth 9up to 2$$mm: in a single pass. .he difference between reduced pressure 9A : and conventional ED welding is in the design of the electron gun which allows the pressure in the welding enclosure to be several orders of magnitude higher4 "$ % "$ mbar for A compared with "$ % "$ mbar for conventional ED welding.
%" %5 %2

urrent status
Aeduced pressure ED welding is in the process of being adopted for two major applications %

;uclear waste encapsulation

/<D in /weden re'uired a process to seal the lids of 5m long ) "m diameter copper canisters, after filling them with high%level nuclear waste. .he difficulty of vacuum pumping this assembly, due to the long and tortuous evacuation paths, meant that conventional ED welding was not feasible. .he noEEle arrangement, between the electron gun and weld chamber, ensures that no copper particles or vapour reach the electron gun to cause discharging. .he weld is non%penetrating but needs to be about F$mm deep with no root porosity, even in the slope%down region. + beam power of around 2$k, and a welding speed of 2$$mm/min gave a wide 9Bmm: round%bottomed weld profile 9Fig#1:, ideal for avoiding root defects. **

Fig.1. $educed pressure E" weld in copper showing round3 bottomed weld pro#ile

=3la+ pipe girth weld

/aipem /p+. in Italy re'uired a fast welding process to join pipes on a lay%barge using the I%lay system in deep 9P2$$$m: water. It is anticipated that by using two electron guns for the girth weld of a ;"mm thick, 2"$mm diameter pipe 9see Fig#2:, the pump%down time for the inflatable%sealed vacuum chamber could be around *$ seconds and the weld time less than 5 minutes.

Fig.2. *rans/erse weld section #rom 26mm wall thickness 0le#t1 and >1mm wall thickness 0right1 carbon manganese steel pipe

-or both of the above applications, .,I designed and manufactured a (special( reduced pressure ED system 9consisting of an electron gun and transfer column: and integrated it with an intelligent, inverter%driven high voltage power supply and a programmable logic control 9 H1: system. .he prototype pipe welding e'uipment is shown in Fig#'.

urrent issues
.,I operates three, reduced pressure ED systems at its research laboratories. .hese are used for basic research and investigating potential applications for .,I Aesearch 8ember companies. .he effect of the higher pressure on fusion Eone composition, and how this affects weld properties, is an important issue and is under investigation at .,I.

"ene#its
.he benefits and characteristics of reduced pressure ED welding include % weld profiles and properties largely the same as conventional ED welding welds thick%section components in a single pass 9up to 2$$mm in steels: uses a locally sealed chamber to weld components too large to fit in a vacuum chamber welds components which cannot be pumped to low pressure, e.g. complicated structures or dirty/damp components

*;

greatly reduced effect of weld%pool emissions on electron gun stability % virtual elimination of gun discharges reduced capital and running costs of pumping system reduced pump%down times4 up to * times faster than conventional ED systems

(recautions
+part from the usual precautions re'uired for conventional ED welding 9)% rays, high voltage, vacuum entrapment:, the weld metal vapour accumulates as dust in the welding enclosure, rather than being deposited as an adherent layer as in conventional ED welding. .he airborne dust should be e)tracted during the enclosure venting cycle. 7ust which has settled on surfaces can be readily removed by brushing but, again, ade'uate dust e)traction and personal breathing protection ) multi3#aceted protrusion created using %ur#i3%culpt? /iewed using a scanning electron microscope

e'uipment must be employed.

%ur#i3%culpt *& uses electrons to reshape #uture

.,I has announced the development of /urfi%/culpt, a revolution in surface treatment, materials processing and joining technology. /urfi%/culpt uses electron beams to reshape materials precisely and so (grow( protrusions that rise from the surface of the material. -or each protrusion there can be a corresponding hole into the bulk of the material. .he electron beam is controlled by a programmable system comprising hardware and software developed at .,I. ,hile the physical limits of the process are being e)plored, protrusions in the order of 2mm high and 2$$ microns wide have been typical of those created within .,I(s laboratories. +s the electron beam is moved across the surface of a material it creates a pool of molten material in a track, or (swipe(. +s the material vaporises, it creates a pocket of pressurised vapour above the swipe at the point of action of the beam. /urface tension and viscosity of the molten material in the swipe is dependent on its temperature, which varies along the swipe as it cools progressively. .he combined effect of the variance in surface tension along the swipe, and the intensity of the vapour above the swipe, is that material is displaced in the direction opposite to the beam(s travel. /wipes may be repeated or overlapped. ,ith each swipe time more material is displaced from the bulk of the material to a common point. +fter several passes of the electron beam, a protrusion begins to grow and rises out of the swipe path..,I(s principal research metallurgist, Druce 7ance, *5

said4 (-or the first time we have a tool that enables control and modification of materials properties at many different levels. 8echanical, electrical, magnetic, thermal, and other characteristics can be tailored precisely for specific applications, and this creates a multitude of e)citing new possibilities(.

Electron "eam (ercussi/e Forming 0E"(F1

Aecent research at .,I has resulted in the development of novel Electron Deam 9ED: materials processing techni'ue, dubbed (Electron Deam ercussive -orming( 9ED -:. a

In this process, an electron beam of a specific power density distribution is used to treat individual parts of a workpiece surface. .he beam is allowed to interact with each point on the surface for an e)act length of time. .his is usually accomplished using a version of .,I(s high speed ED deflection system to manipulate the beam. In this case, it is not necessary to turn the beam off between treatment positions, as the beam moves so 'uickly that its heat input in transit is negligible. 7uring the treatment, there need not be any deliberate relative movement between the beam and the work, yet an e)citing range of features may be produced, alone or in combination with other treatments. In so many fields of engineering and science, the properties of an object are defined almost e)clusively by its surface. +lthough at an early stage of development, ED - looks set to add another powerful and speedy surface processing tool to those already available.

8ow does E"(F work@

.he process works by utilising some of the same forces that are present in ED welding, ED te)turing and (/urfi%/culpt( technologies, but in a different way. .he schematics below show what happens4

In stage a, the beam has spontaneously melted material, to a depth associated with the electron range into the material, and a diameter related to the siEe of the beam. In stages b and c the molten Eone e)tends. In stage d, the beam is now boiling material in the centre of the interaction Eone. In stages e and f, the molten material is now displaced away from the centre of the point of action of the beam, thus forming a vapour%filled cavity. .his process typically takes just a few microseconds, and is analogous to the initiation of an ED weld (keyhole(.

*F

/e'uence f%m shows what can happen after the beam is turned off or moved elsewhere. /tage (f( is not stable once the beam is no longer incident at this point. .he forces of surface tension will attempt to flatten the melt pool somewhat. If the heat input is low, and the freeEing rate high, then the material may freeEe in stage g, as used in various te)turing processes. &owever, if the heat input is a little higher, the melt pool may flatten appreciably before it freeEes, e.g. stage h. /hould the heat input be high enough, and the viscosity be low enough, the concave (g( melt pool may swing through the flattened (h( stage and become conve) as in (j( and (i(. If this process continues, the melt pool could oscillate several times 9g%m, repeat: before freeEing. &owever, if the heat input is e)actly right, and the freeEing rate is high enough, the melt pool can be made to freeEe in the (j( stage, giving a protruding feature. In some cases there will be an accompanying shallow peripheral depression. In others, this may be more or less absent, even though there is a central protrusion. If the material in and around the melt pool is hot enough, the li'uid metal will more than fill the hole from which it came. .he apparent (e)tra material( comes from a combination of elastic and plastic thermal stresses in the material5 % this is the same effect that gives proud beads in ED welds. Interestingly, the siEe and radius of the protrusion formed in ED - may be just a small fraction of the beam diameter used to make it. .he process appears to work best in materials with a narrow freeEing range, e.g. pure metals. .hese materials freeEe 'uickly, and the li'uid/solid interface is relatively smooth, leading to low viscous forces in the melt pool, and large amplitude oscillations. .he speed of melt pool freeEing is also enhanced if the melting point is high5 radiant heat losses 9which go as .;: help to ensure speedy freeEing.

Dy applying this type of treatment successively, in the same or different form, features of higher aspect ratio may be created and modified. =verlapping treatments may be used to create new types of surface. + typical ED - treated surface. Each feature is formed via a beam interaction lasting typically just a few micro%seconds or so. .his means that several thousand such features may be formed every second in some cases. In this case a beam of Q"$$microns diameter has yielded consistent features protruding several tens of microns from the surface. Each feature has a tip with a radius of appro)imately 5microns% a tiny fraction of the beam diameter used to make them. Dy utilising specially shaped beams, or modulating the treatment via beam focus or 9)/y: deflection during interaction, further control and modification of the protrusion shape may be e)ercised.

*2

Dy adding material before or during the process, a range of alloys and new materials may be created. /ince the material in each feature grows epita)ially from the substrate, substrates with specific grain te)tures can be treated to give features with specific properties dependant on their physical shape and crystallographic orientation. -eatures of a suitable siEe may be made in which there are specific optical interactions, or interactions with other types of wave or physical property. 1ontrol may be e)ercised over the properties of the treated surface, including magnetic, electromagnetic, thermal, mechanical, thermoelectric, ultrasonic, mechanical, emission, electrical properties etc. +lthough some e'uivalent effects may also be produced using laser treatments, the fundamental interaction of the laser beam is different, since volumetric heating of thin surface layer does not occur in the same way. In addition, the speed and precision of laser beam manipulation technology is not sufficient at present to permit areas to be treated in the same way as is possible using an electron beam. ED - treatments, alone or in combination with other treatments are thought to have potential applications in a number of areas, including wettable surfaces 9for printing etc:, altered biocompatibility, the manufacture of cathodes, preparation of surfaces for bonding, surfaces with improved corrosion performance, preparation of surfaces prior to chemical conversion.

*B

arbon dio,ide laser

.he carbon dio)ide 91= : gas laser, is one of the most versatile for materials processing applications, and emits infra red radiation with a wavelength between # and ""Rm, although emission at "$.FRm is the most widely used. =f the several types of 1= laser that are available, the waveguide, the low power sealed tube and the transversely e)cited atmospheric 9.E+: lasers are used for small scale materials processing applications. .he fast a)ial flow 1= laser and the less widely used slow flow laser, are used for thick section cutting "%"5mm and deep penetration welding. ,hile these lasers share the same active medium, they have important functional characteristics, which contribute to the wide range of 1, 9continuous wave: powers, pulse powers and pulse durations available from the 1= laser.
2 2 2 2

.he active medium in a 1= laser is a mi)ture of carbon dio)ide, nitrogen and 9generally: helium. It is the carbon dio)ide which produces the laser light, while the nitrogen molecules help e)cite the 1= molecules and increase the efficiency of the light generation processes. .he helium plays a dual role in assisting heat transfer from the gas caused by the electric discharge used to e)cite the gas, and also helps the 1= molecules to return to the ground state.
2 2 2

%ealed *ube 4 Aasers

2

.hese lasers are operated as conventional gas discharge lasers in the form of long narrow glass tubes, filled with the lasing gas mi)ture. Electrodes at either end of the tube provide the discharge current. + totally reflecting and partially transmitting mirror, usually made from polished metal and coated Einc selenide respectively, form the resonant cavity. .he tube is sealed using Drewster angled windows. Fig#1, shows a schematic drawing of a sealed tube 1= laser. +s the electric discharge in the tube breaks down the 1= , an ordinary gas mi)ture would stop working very 'uickly and so methods are provided to cause the 1= to regenerate, either by addition of hydrogen or water or by the use of catalytic action. /everal thousand hours of operation are possible with sealed tube 1= lasers before the tube has to be cleaned and re%filled or replaced. 71 and sometimes A- discharges are used with these lasers.
2 2 2 2

Fig. 1 %ealed tube 4 laser schematic

2

1, power up to about 2$$, is available from these lasers with good beam 'uality. ulsed power supplies can produce laser pulses lasting $." % "msecs with peak powers 5%"$ times the 1, power level.

-a/eguide 4 Aasers
2

.he waveguide laser is an efficient way to produce a compact 1= laser. It consists of 9see Fig#2:, two transverse A- electrodes separated by insulating sections that form a bore region. .he lateral dimensions of the bore are a few millimetres, which propagates the beam in (waveguide mode(.
2

*#

Fig. 2 -a/eguide 4 laser schematic

2

.he tube is normally sealed with a gas reservoir separate from the tube itself. .he small bore allows high pressure operation and provides rapid heat removal5 both of which lead to high gain and high power output from a compact unit.

*E) 4 Aasers
2

7ischarge instabilities prevent operation of 1, 1= lasers at pressures above about "$$mbar. ulses in the nanosecond to microsecond duration range can be produced by passing a pulsed current transversely through the lasing gas. /uch .E+ 9transversely e)cited atmospheric: lasers operate at gas pressures of one atmosphere and above in order to obtain high energy output per unit volume of gas. + transverse discharge from two long electrodes is employed 9see Fig#':. rior to application of the pulsed discharge, a form of pre%ionisation is used to ionise the space between the electrodes uniformly, thus allowing the discharge to proceed in a uniform fashion over the entire electrode assembly. .he prime attractions of .E+ lasers are their ability to generate short intense pulses and the e)traction of high power per unit volume of laser gas.
2

Fig. 3 *E) 4 laser schematic

2

ulse duration as low as a few tens of nanoseconds up to a few microseconds are possible. ulse energies range from the millijoule region to 5$$Ioules at pulse repetition rates from about *$$&E down to single shot.

4ptics #or 4 Aasers

2

Aeflective mirrors Henses and windows -allplug E##icienc+

% silicon with high reflectivity coatings, gold coated copper. % gallium arsenide and germanium 9not transparent in visible region: and coated Einc selenide 9orange in the visible region:. between 5@ and 2$@

"eam Diameter and Di/ergence

;$

.he shape and length of the laser cavity and nature of the resonator optics determine the beam diameter and divergence of the 1= laser. .ypical ranges are4
2

beam diameter 9mm: beam divergence 9mrads: /ealed tube4 ,aveguide4 .E+4 "%2 "%2 ; % "2 2%F * % "$ $.5 % *

Diode lasers
&igh power diode lasers 9known as & 7H(s: feature a very high electrical to optical power conversion efficiency coupled with a very compact siEe. ,ith suitable (focusing( optics, today(s & 7H(s are suitable for some materials processing applications. .he laser diodes which drive the & 7H(s are also being used to replace flashlamp pumping in solid state lasers. 7iode lasers also e)hibit very high wall plug efficiencies which can be greater than *$@ on commercially available systems. 7iode lasers consist of a p%n junction within a multi%layer semiconductor structure. -or powers greater than about ;,, the only commonly used manufacturing approach produces a diode laser bar about "$mm long, with emission of radiation confined to the narrow junction region 9typically "Rm thick:. +long the "$mm length, many thousands of single emitters, of the order 5Rm wide, produce laser output with, because of diffraction, very large beam divergence. 9/ee Fig#1:. .he resulting beam with its large angular spread is characteristic of semiconductor lasers, and, compared to other types of laser, presents a drawback in terms of focusability. Fig. 1 Aaser diode schematic .he beam divergence is up to #$K perpendicular to the emitting line 9known as the (fast( a)is: and about "$K along the emitting line 9known as the slow a)is:. owers of the order B$, and higher can be achieved from one diode bar. -or high power applications, combining the power from several diode bars is re'uired. -or materials processing applications, the semiconductor material is based on In0a+s on a 0a+s substrate 9#;$nm: or In0a+l+s on a 0a+s substrate 9B$Bnm:. Doth these wavelengths are invisible to the eye. +s a result of the rather unusual beam characteristics of the diode laser and the added complication of increasing power by adding diode bars, several different possibilities e)ist for beam manipulation to achieve the re'uired power densities for material processing applications. It would appear that this is the area in which one (diode laser( supplier may be distinguished from another. + *k, 9highest currently available commercially is Fk,: diode laser 9including beam focusing: is smaller than a shoebo) and its control, power supply and cooling system is the siEe of a two drawer filing cabinet. +s a result, a clear division can be seen between those manufacturers who would

;"

place the laser directly on the arm of a robot say and those who favour fibre optic beam delivery to a focusing head 9the latter very similar to that re'uired for a 6d4G+0 laser:. .he approach to beam shaping and focusing is therefore different for these two cases. .wo of these design configurations, suitable for material processing applications are described below. Henses for beam shaping with diode lasers are usually manufactured from glass or fused silicon.

!ndi/idual "eam %haping 0!"%1 Diode Aaser

Fig. 2a !"% diode laser schematic .his system uses sophisticated optics to combine the beam from three individual diode bars mounted as can be seen in Fig#2a. In addition, special diodes are used where the emitting Eone is confined to 5 areas 5$$Rm wide with a centre to centre spacing of ".5mm. .his design is the basis of improved beam 'uality which permits the generation of focused spots about $.25 ) $.Fmm 2. ,ith its output power of about "5$,, the power density Q"$ 5,/cm 2, is sufficient for conduction welding of metals.

4ptical Fibre Deli/ered Diode Aaser

Fig. 2b Fibre deli/ered diode laser .his laser also uses a comple) optical system designed to minimise the spot siEe from a single bar so that the beam can be launched down a silica optical fibre. +fter fast a)is collimation with a micro lens, the slow a)is is chopped into small beamlets by a special diamond machined mirror. + set of prisms then compresses these beamlets together before collimation with a cylindrical lens and final focusing via spherical optics.

;2

Fig.2b, shows how the combination of diamond machined mirror, prisms and lenses, produces a $.Bmm diameter beam for launching into the fibre. !sing this configuration, a *5, single bar device can produce a power density of about 2 ) "$ * ,/cm 2.

;ew De/elopments
7iode laser technology continues to develop at a fast pace. Its limitations continue, however, to be available spot siEe. 8uch effort has gone into this and most of the higher power 9"k,?: diode lasers now use the techni'ue of wavelength coupling in order to ma)imise power in a small spot.

;d:B)' laser welding

.he 6d4G+0 laser is one of the most versatile laser sources used in materials processing. .he relative robustness and compactness of the laser and the possibility for the ".$F micron light it produces to be transmitted to the workpiece via silica optical fibres, are two features which contribute to its success. 6d4G+0 lasers were first commercialised operating mainly in pulsed mode, where the high peak powers which can be generated were found useful in applications such as drilling, cutting and marking. .hese pulsed lasers can also be utilised for welding a range of materials. 8ore recently, high power 9up to "$k,:, continuous wave 91,: 6d4G+0 lasers have become available. .he 6d4G+0 crystals in these lasers can be pumped either using white light flashlamps or, more efficiently, using laser diodes. .he latter methods are used to produce high 'uality beams, which can be focused to smaller spots 9and therefore produce higher power densities: than the flashlamp pumped lasers. Decause of the possibility of using fibre optic beam delivery, these lasers are often used in conjunction with articulated arm robots, in order to work on components of comple) shape. Decause of the wide range of applied power and power densities available from 6d4G+0 lasers, different welding methods are possible. If the laser is in pulsed mode, and if the surface temperature is below the boiling point, heat transport is predominantly by conduction and a conduction limited weld is produced. If the applied power is higher 9for a given speed:, boiling begins in the weld pool and a deep penetration weld can be formed. +fter the pulse, the material flows back into the cavity and solidifies. Doth these methods can be used to produce spot welds. + seam weld is produced by a se'uence of overlapping deep penetration (spot( welds or by the formation of a continuous molten weld pool. -or the former, once the energy input is sufficient to ensure that the weld does not solidify between pulses, the (keyhole( type weld normally associated with 1= laser welding can be formed. ulsed laser welding is normally used at thicknesses below about *mm. &igher power ;% "$k, 1, 6d4G+0 lasers are capable of keyhole type welding in materials from $.Bmm 9car body steel: to "5mm 9ship steel: thickness.
2

6d4G+0 laser welding is used commercially on a wide range of 1%8n steels, coated steels, stainless steels, aluminium alloys, titanium and molybdenum. .he low heat input welding offered by 6d4G+0 lasers is utilised in the electronics, packaging, domestic goods and automotive sectors, and significant interest has been shown more recently, particularly for the high power 1, lasers, in the shipbuilding, oil and gas, aerospace and yellow goods sectors. Important AO7 issues involve development of high power lasers of better beam 'uality, use of distributed energy in the beam focus, weld 'uality maintenance for both thick and thin sections and weld classification. .he principal risks involved in 6d4G+0 laser welding are4 optical 9the beam can burn the skin or damage the retina if focused by the eye:, electrical, and fume generation. + current application issue is safe use of 6d4G+0 lasers in anything other than a fully opa'ue 9to the 6d4G+0 laser wavelength: enclosure, such as might be found in a shipyard for e)ample. ;*

Cse o# e,cimer lasers #or materials processing

b+ Da/e *a+lor
E)cimer lasers are characterised by short wavelengths, high intensities and short pulse durations. .hese characteristics mean that a single photon is capable of breaking a chemical bond. .he majority of laser materials processing techni'ues are essentially thermal processes in which absorption of a large number of photons heats the material to enable cutting, welding or surface modification operations to be performed. E)cimer lasers were first demonstrated in "#25, some time after many of the other laser sources, and are now fairly well established in their niche applications. .he term (e)cimer( stands for (e)cited dimer(, where (dimer( refers to a diatomic molecule such as =2 or 62. .his is not strictly a correct term, as the two atoms that make up the molecules used in e)cimer lasers can be different. .he most important molecules are rare gas halides such as -2, +r-, <r1l, <r-, Se1l and Se-. .hese do not e)ist in nature but can be produced by passing an electrical discharge through a suitable gas mi)ture. .his means that e)cimer lasers generate ultraviolet energy over a range of wavelengths, depending on the gas mi)ture used 9e.g. "52nm for -2 and *5"nm for Se-:. .ypical average output powers are in the range from less than " watt up to around 2$$ watts. .his is two orders of magnitude less than the more traditional 6d4G+0 or 1=2 lasers which operate in the infrared part of the spectrum. .he high intensity beam of an e)cimer laser is the product of pulse energy 9"$ % "$$$mI:, spot siEe 9governed by focusing optics: and pulse duration 9around "$ns:. +pplications of e)cimer laser are primarily in machining of materials such as plastics, paper, ceramics, glasses, crystals, composites and biological tissue. ,hen illuminated with an e)cimer laser, the relatively weak organic bonds are broken down. E,cimer laser processing using step3and3repeat mask projection .his creates a pressure rise and subse'uent shock wave that removes material, with little heat transfer to the surrounding material, in a process called (ablation(. .his processing is usually most efficient when carried out using a mask with an image of the re'uired feature. E)cimer laser machining is used for its precision, producing features down to appro)imately ;$Rm in resolution, but with virtually no heat affected Eone. /ome research has been carried out in welding and cutting of sheet metals, but showed no significant advantages over 1=2 or 6d4G+0 laser which are available in much higher average powers. E)cimer lasers do also lend themselves to more (niche( applications, such as surface modification of metals and glass for strengthening adhesive bonding and smoothing of machined surfaces to increase wear resistance of components such as camshafts and pistons. +s well as increasing wear resistance, this process is being studied as a means of increasing corrosion resistance through production of a thin, amorphous layer on the surface of the material.

;;

$esistance butt welding

Description
Aesistance butt welding is the simplest form of a group of resistance welding processes that involve the joining of two or more metal parts together in a localised area by the application of heat and pressure. .he heat is generated within the material being joined by resistance to the passage of a high current through the metal parts, which are held under a pre%set pressure. .he process is used predominantly to make butt joints in wires and rods up to about "Fmm diameter, including small diameter chain. .he faces of the pieces to be joined may be flat and parallel or profiled in the case of larger sections. .his reduces the initial contact area and further concentrates the heating at the interface. .he components are clamped in opposing copper dies, with a small amount of stick%out, and abutted under pressure. 1urrent is passed between the dies causing resistance heating of the weld area. .he heat generated during welding depends on the current, the duration of the current, and the resistance. +s the resistance is highest at the joint interface, heating is most intense in this area. ,hen the material softens, it deforms under the applied load, giving a solid phase forge weld. 6o melting occurs. .he current is terminated once a pre%set upset length has occurred, or the duration of the current is pre%set. .he joint is then allowed to cool slightly under pressure, before the clamps are opened to release the welded component. .he weld upset may be left in place or removed, by shearing while still hot or by grinding, depending on the re'uirements.

(resent status
E'uipment is well established for joining steel wire and rod up to about "Fmm diameter, and narrow strip. +utomated dc welding e'uipment is available for joining wider strip, up to about *$$mm wide for automobile road wheel rims, at rates up to about "2 per minute.

!mportant issues
7epending on the application area, issues include control of weld 'uality, production speed, and control of upset shape.

,eld 'uality is normally maintained by good process control coupled with consistent end% preparation of materials. rogrammed force and current are used for the most demanding applications, such as wheel rims In some applications, such as wire frames, a small, smooth upset is preferred, to avoid the need to remove it. 1areful choice of welding parameters allows a balance between ade'uate strength and minimum upset

&ardenable steels can be welded and given a subse'uent in%machine, local heat treatment to restore weld area toughness. +lloy and stainless steels re'uire more care to ensure that tenacious surface o)ides such as chromium o)ide are sufficiently dispersed from the interface during welding.

"ene#its
Aesistance butt welding is a high speed, clean process and is preferred to flash welding for many small components.

;5

$isks
.here are some limitations on component siEe and geometry4 very thin or large sections are unsuitable. .he main haEards are 9i: the risk of crushing fingers or hands and 9ii: burns or eye damage from splash metal.

Flash welding
b+ %te/e -estgate

Description
-lash welding is a development of resistance butt welding and is particularly suitable for butt welding comple) or larger sections. It is used for a wide range of component shapes and siEes from bicycle wheel rims to rails. 8ore efficient energy input, and a more localised and evenly heated Eone can be achieved, compared with resistance butt welding. In flash welding, the components are clamped between dies and brought together slowly with the current switched on, see Fig#1. 1urrent flows through successive points of contact which heat rapidly, melt and blow out of the joint giving the characteristic flashing action. +fter a pre%set material loss has occurred, sufficient to heat the material behind the interface to its plastic state, the components are forged together to e)pel melted material and contaminants. .his completes a solid phase forge weld. .he joint is then allowed to cool slightly under pressure, before the clamps are opened to release the welded component. .he weld upset may be left in place or removed, by shearing while still hot or by grinding, depending on the re'uirements. Fig.1. %chematic representation o# a #lash welding machine ,hen welding larger sections, the parts may be preheated to promote easier flashing. .his is done by advancing and retracting the components to make repeated short circuit contacts under pressure. .he flashing se'uence must be accurately controlled and the forge applied with sufficient force and speed for best results. .he forward motion of the components is normally accelerated as flashing progresses. .he flashing speed should be as fast as practicable for a given flashing voltage, whilst avoiding premature butting or (freeEing( of the parts. .his ensures that the flashing action is continuous and not so coarse that deep craters are blown in the surfaces to be joined, as this ;F

increases the risk of interfacial defects. .hese take the form of a film or planar distribution of o)ide inclusions known as (flat spots(. Insufficient forge may also cause flat spots or result in retained melted or overheated material at the interface.

urrent status

Fig.2. Flash welding o# a 12DDmm diameter aero3engine ring E'uipment is available to join a range of material siEes and types, from thin steel strip around $.Bmm to "5$mm diameter mooring chain. 6ickel alloy and titanium aero%engine rings, see Fig.2, and aluminium alloy sections are also flash welded.

!mportant issues
1oncerns with flash welding are usually associated with weld 'uality.

1areful setting up and control of the welding se'uence, particularly flashing and upset, are re'uired to ensure freedom from weld discontinuities. /ervo controlled machines, which may also have se'uence monitoring, provide adaptive control of the flashing stage. +utomatic monitoring of process parameters, particularly current, force and displacement, can be used to verify conformance with the set se'uence, and provide a guide to weld 'uality. -lash welds in steels have a low impact toughness performance and re'uire post%weld heat treatment for applications where impact toughness is specified. 1ertain weld interface flaws, such as flat spots, are not readily detected by non%destructive testing techni'ues.

"ene#its
-lash welding is ideally suited to producing butt welds in large or comple) sections. ,eld time is relatively short, from a few seconds for the thinnest sections to a few minutes for the largest.

$isks
.he main haEards are4 9i: the risk of crushing fingers or hands5 and 9ii: burns or eye damage from splash metal. ;2

$esistance projection welding

Description
rojection welding is a development of resistance spot welding. In spot welding, the siEe and position of the welds are determined by the siEe of the electrode tip and the contact point on the workpieces, whereas in projection welding the siEe and position of the weld or welds are determined by the design of the component to be welded. .he force and current are concentrated in a small contact area which occurs naturally, as in cross wire welding or is deliberately introduced by machining or forming. +n embossed dimple is used for sheet joining and a (>( projection or angle can be machined in a solid component to achieve an initial line contact with the component to which it is to be welded, see Fig#1. Fig.1. E,ample o# projection welding con#igurations: a1 embossed projectionE b1 stud to plateE c1 annular projection

In sheet joining using embossed projection welds, a melted weld Eone is produced, as in spot welding. &owever, when a solid formed or machined projection is used, a solid phase forge weld is produced without melting. .he plastic deformation of the heated parts in contact produces a strong bond across the weld interface.

(resent status
.he process is well established and is applicable mainly to low carbon or microalloyed steels. .he process is widely used on sheet metal assemblies in automotive and white goods industries for both sheet joining and attaching nuts and studs. ,ire mesh welding is also a large industry.

!mportant issues
rojection welding tends to be more application dependent than spot welding and is less easy to standardise. &owever, similar issues can arise4 ,eldability of coated steels and attachment of fasteners % control of weld 'uality and electrode life ,eld 'uality is controlled mainly by good process control together with periodic testing of samples. ,hile a number of monitors have been developed, there is still a desire to produce a low cost, reliable and robust in%process weld 'uality monitor, in order to reduce or eliminate periodic destructive tests. .he added difficulty with projection welding is that multiple welds are made

;B

Aecent e'uipment developments have included use of capacitor discharge power supplies to make large diameter welds for applications such as gear assemblies

"ene#its
.he advantages of projection welding include its versatility, the speed and ability to automate, the ability to make a number of welds simultaneously and minimisation of marking on one side of joints in sheet materials. 1apacitor discharge supplies used with machined annular projections can compete with power beam welding, as the weld is completed in a single shot within milliseconds.

$isks
.here are some limitations on material weldability but attention to correct setting up and good process control can solve most production problems. .he main safety factors are trapping haEards and splash metal. Hittle fume is produced but may need attention when welding coated steels or when oils or organic materials are present.

$esistance seam welding

Description
.he seam welding process is an adaptation of resistance spot welding and involves making a series of overlapping spot welds by means of rotating copper alloy wheel electrodes to form a continuous leak tight joint. .he electrodes are not opened between spots. .he electrode wheels apply a constant force to the workpieces and rotate at a controlled speed. Fig.1. on/entional wide wheel seam welding .he welding current is normally pulsed to give a series of discrete spots, but may be continuous for certain high speed applications where gaps could otherwise occur between individual spots. /eam welding e'uipment is normally fi)ed and the components being welded are manipulated between the wheels. .he process may be automated5 it is illustrated in Fig.1.

urrent status
.here are a number of process variants for specific applications4

;#

Fig.2a1 -ide wheel seam -ide wheel seam

,heel contact width normally 5 sheet thickness in mm:.

t mm flat 9where t is single

0eneral purpose welding 9domestic radiators up to about F m/min:.

Fig.2b1 ;arrow weld seam 6arrow wheel seam

,heel contact shape typically Fmm radius. <nurl drive on wheel edge with contact surface continuously planished. 1ontrols electrode contamination when welding coated steels, such as for vehicle fuel tanks.

Fig.2c1 onsumable wire seam 1onsumable wire seam welding

/haped, consumable copper wire fed between the wheels and sheets to be joined to give consistent clean contact. !sed for welding coated steels such as tin cans and vehicle fuel tanks.

Fig.2d1 &ash seam welding 8ash seam welding

6arrow overlap of sheet edges, which are partly crushed together during welding. &igh speed welding of tin cans and drums 9$.2mm tinplate up to "$$m/min:. ,ide wheel or consumable wire processes used.

Fig.2e1 Foil butt seam welding -oil butt seam welding

-oil welded on to each side of the butted edges of the sheets to be joined. .ypically ;mm wide stainless steel foil used to preserve corrosion resistance on coated steel.

5$

>irtually flush finish with no crevice and used to produce wide panels.

!mportant !ssues
.he main issues concerning seam welding are in weld 'uality control and welding speeds.
&igh speed welding of coated steels is of particular importance in manufacture of tin cans. /pecialist consumable wire seam welding machines are used. Cuality monitors have been developed for these applications and welding speeds up to about "$$m/min are possible. /teels with alternative coatings to tin, for canning applications, present difficulties due to their high surface resistance. ,eld 'uality is controlled mainly by tight process control together with periodic testing of samples. -actors such as material and pressing 'uality, and alignment of the electrodes to the material, are critical to achieve high speed, 'uality welding, for e)ample in coated steels for fuel tanks.

"ene#its
/eam welding enables high welding speeds to be obtained compared with many other techni'ues, but can be limited by component shape and wheel access.

$isks
.here are some limitations on material weldability but attention to correct setting up and good process control can solve most production problems. .he main haEards are4 9i: the risk of crushing fingers or hands5 and 9ii: burns or eye damage from splash metal. Hittle fume is produced but may need attention when welding coated steels or when oils or organic materials are present.

$esistance spot welding

Description
/pot welding is one of a group of resistance welding processes that involve the joining of two or more metal parts together in a localised area by the application of heat and pressure. .he heat is generated within the material being joined by the resistance to the passage of a high current through the metal parts, which are held under a pre%set pressure. .he process is used for joining sheet materials and uses shaped copper alloy electrodes to apply pressure and convey the electrical current through the workpieces. Fig.1. $esistance spot weld section &eat is developed mainly at the interface between two sheets, eventually causing the material being welded to melt, forming a molten pool, the weld nugget. .he molten pool is contained by the pressure applied by the electrode tip and the surrounding solid metal.

urrent %tatus
5"

/pot welding is one of the oldest welding processes. It can be used on very thin foils or thick sections but is rarely used above about Fmm thickness. It is used in a wide range of industries but notably for the assembly of sheet steel vehicle bodies where more than "$$ million welds are made per day in Europe alone. &igh 'uality welds can also be made in stainless steels, nickel alloys, aluminium alloys and titanium for aerospace application.

!mportant !ssues
.he main areas of research are in materials, 'uality control and e'uipment. &igh strength steels, coated steels and aluminium alloys are of great interest to the automotive industry but can have weldability problems. +lthough the problems are understood, further improvements in weldability and electrode lives are still being sought. .here are also limitations in the ability of spot welding to join sheet to tube, which need to be addressed. ,eld 'uality is controlled mainly by good process control together with periodic testing of samples. ,hile a number of monitors have been developed, there is still a desire to produce a low cost, reliable and robust in%process weld 'uality monitor, in order to reduce or eliminate periodic destructive tests. !ltrasonic non%destructive testing has been proved and widely applied at vehicle manufacturers but still re'uires skill and training to achieve high accuracy. Aecent e'uipment developments have included improved controls, medium fre'uency power supplies, servo operated guns and automatic electrode dressing and changing tools.

"ene#its
/pot welding offers a number of advantages over other techni'ues, including high speed, ease of automation and energy efficiency.

$isks
.here are some limitations on material weldability but attention to correct setting up and good process control can solve most production problems. .he main safety haEards are 9i: the risk of crushing fingers or hands and 9ii: burns or eye damage from splash metal. Hittle fume is produced but may need attention when welding coated steels or when oils or organic materials are present.

old pressure welding

!ntroduction
+ll metals are surrounded by surface layers 9o)ide: which must be disrupted if they are to be welded. 1old pressure welding, carried out at ambient temperature, relies upon the use of high compressive pressures 9";$$%2B$$6/mm for aluminium and at least double that value for copper:. .his provides interfacial deformations of F$@ to B$@ that break the o)ide layers to e)pose fresh, uncontaminated metal that makes contact. In this state, interatomic forces take over to produce the weld.
L"M 2

52

&aterials welded
.his joining method has been used for thousands of years and is applicable to most of the ductile metals and their alloys, such as copper, aluminium, lead, Einc, nickel, silver, platinum, gold, palladium, and cadmium. ublished literature indicates that variations of the methods have been used in production for joining like and dissimilar metal combinations for at least ;$ years.

=oint t+pes
.he process can be applied to both lap and butt configurations . Dutt joints are primarily used for joining wires and rods in diameters from Q$.5mm up to "2mm, depending upon the metal being joined. If the single butt 9pressure: approach is utilised then the surfaces to be joined must be thoroughly cleaned. &owever, greater tolerance to the weld surface condition is possible if a multiple butt techni'ue 9that is, multiple pressure applications: is used.
L2M

,here the application demands the joining of lapped sheets to themselves or bars, a series of small welds can be used. ,hen calculating strength, the designer should consider that at least half the thickness of one of the sheets will be lost due to the applied pressure. It is also possible to make use of drawing processes in which multilayer tubes or solid multilayer 9clad: bars can be cold pressure welded. /imilarly, roll bonding of metals is also possible . /pecialist e'uipment manufacturers supply various forms of cold pressure welding systems.
L2M L*M

)d/antages and limitations

+s the process is performed at ambient temperature, there are no thermal effects on the parts being joined, and the process is fast. It is simple and ine)pensive to operate once dies have been produced. &owever, it is highly specialised with respect to joint design and materials to be welded. +s the welds are made in the (solid state( they are difficult to inspect5 thus reliance must be placed on process control. ,ith the e)ception of butt welds, or welds where the contact surfaces are sheared together, the thickness of the parts is reduced significantly at the weld.

)pplications
erhaps the greatest use of cold pressure welding has been for joining of wire, foil to wire, wire to bi%metals, and sealing of heat sensitive containers such as those containing e)plosives 9detonators for e)ample:. Aod coils are butt welded to permit continuity in post%weld drawing to smaller diameters. In the electronics industry, cold welding processes are used to seal tin plated steel crystal cans and copper packages for heat sensitive semiconductor devices. 0lass packages are also sealed using an indium or tin alloy interlayer. +n interesting application of the process is underground wire servicing where joins need to be made in hostile environments, such as in the presence of e)plosive gases.

Di##usion bonding
!ntroduction
5*

7iffusion bonding is a solid%state joining process capable of joining a wide range of metal and ceramic combinations to produce both small and large components. .he process is dependent on a number of parameters, in particular, time, applied pressure, bonding temperature and method of heat application. =ther e)amples of solid%state joining include cold pressure welding, friction welding, magnetically impelled arc butt 98I+D: welding and e)plosive welding. 7iffusion bonding itself can be categorised into a number of variants, dependent on the form of pressurisation, the use of interlayers and the formation of a transient li'uid phase. Each finds specific application for the range of materials and geometries that need to be joined.

(rocess mechanism
In its simplest form, diffusion bonding involves holding pre%machined components under load at an elevated temperature usually in a protective atmosphere or vacuum. .he loads used are usually below those which would cause macrodeformation of the parent material9s: and temperatures of $.5%$.B.m 9where .m T melting point in <: are employed. .imes at temperature can range from " to F$? minutes, but this depends upon the materials being bonded, the joint properties re'uired and the remaining bonding parameters. +lthough the majority of bonding operations are performed in vacuum or an inert gas atmosphere, certain bonds can be produced in air. +n e)amination of the se'uence of bonding 9Fig#1: emphasises the importance of the original surface finish. .o form a bond, it is necessary for two, clean and flat surfaces to come into atomic contact, with microasperities and surface layer contaminants being removed from the bonding faces during bonding. >arious models have been developed to provide an understanding of the mechanisms involved in forming a bond. -irst they consider that the applied load causes plastic deformation of surface asperities reducing interfacial voids. Dond development then continues by diffusion controlled mechanisms including grain boundary diffusion and power law creep. Fig.1. *he mechanism o# di##usion bonding a1 !nitial .point. contact? showing residual o,ide contaminant la+er b1 Bielding and creep? leading to reduced /oids and thinner contaminant la+er c1 Final +ielding and creep? some /oids remain with /er+ thin contaminant la+er d1 ontinued /acanc+ di##usion? eliminates o,ide la+er? lea/ing #ew small /oids e1 "onding is complete.

a:

b1

c:

d1

5;

e:

%olid phase di##usion bonding

Donding in the solid phase is mainly carried out in vacuum or a protective atmosphere, with heat being applied by radiant, induction, direct or indirect resistance heating. ressure can be applied unia)ially or isostatically. In the former case, a low pressure 9*%"$8 a: is used to prevent macrodeformation of the parts 9i.e. no more than a few percent:. .his form of the process therefore re'uires a good surface finish on the mating surfaces as the contribution to bonding provided by plastic yielding is restricted. .ypically surface finishes of better than $.;Rm A+ are recommended and in addition the surfaces should be as clean as practical to minimise surface contamination. In hot isostatic pressing, much higher pressures are possible 9"$$%2$$8 a: and therefore surface finishes are not so critical, finishes of $.BRm A+ and greater can be used. + further advantage of this process is that the use of uniform gas pressurisation allows comple) geometries to be bonded, as against the generally simple butt or lap joints possible with unia)ial pressurisation. ,here dissimilar materials need to be joined in the solid phase 9and in particular metal to ceramic joints:, it is possible to introduce single or multiple interlayers of other materials to aid the bonding process and to modify post%bond stress distribution.

Ai9uid phase di##usion bonding/di##usion bra:ing

.his techni'ue is applicable only to dissimilar material combinations or to (like( materials where a dissimilar metal insert is used. /olid state diffusional processes lead to a change of composition at the bond interface and the bonding temperature is selected as the temperature at which this phase melts. +lternatively, with the dissimilar metal insert, it melts at a lower temperature than the parent material. .hus a thin layer of li'uid spreads along the interface to form a joint at a lower temperature than the melting point of either of the parent materials. + reduction in bonding temperature leads to solidification of the melt, and this phase can subse'uently be diffused away into the parent materials by holding at temperature, Fig#2.

55

Fig.2 %chematic illustration o# the steps in/ol/ed in making a di##usion3 bra:ed joint

.he techni'ue has been used particularly for the bonding of aluminium alloys where eutectics can be formed with copper, silver or Einc to assist in the break%up of the stable aluminium surface o)ide.

%uperplastic #orming/di##usion bonding

.his techni'ue has been developed specifically within the aerospace industry, and its industrial importance is such that it should be considered separately here. .he process is used commercially for titanium and its alloys, this material being one that e)hibits superplastic properties at elevated temperatures within defined strain rate conditions. .hese conditions of temperature and pressure coincide with the conditions re'uired for bonding, and therefore the two processes have been combined into one manufacturing operation either in se'uence or together. .he process 9known as / -/7D or more correctly 7D// -: is used to produce stiff sandwich structures for airframe parts, or the wide chord, hollow fan blades for aeroengines. Doth these involve skins with internally bonded structures as reinforcing elements.

%ummar+
>ariants of the diffusion bonding techni'ue are available, offering scope for the joining of many new materials and configurations. Decause of the capital e'uipment costs, surface preparation re'uirements and long bonding times, the process is restricted to high value, relatively low production components. &owever, its versatility for unusual material combinations and its relative ease of use for titanium alloys 9including intermetallics: will ensure the process continues to be developed for specific application re'uirements.

5F

E,plosi/e welding
b+ *om 8owes

-hat is it@
.he fearsome destructive power of e)plosives can be harnessed to provide a uni'ue joining method, known as E)plosive ,elding. .o form an e)plosive weld the following conditions need to occur4

.wo surfaces that need to be joined are initially spaced at a small distance 9stand%off distance:. +n e)plosive force brings these two surfaces together progressively at a collision front. .he collision front(s velocity must be lower than the speed of sound in the materials, so that the shock wave precedes the bond being formed. If not, the shockwave would interfere with the contacted surfaces preventing a bond occurring. .he interfacial pressure at the collision front must e)ceed the yield strength of the materials, so that plastic deformation will occur.

+ jet of metal is formed just ahead of the collision front, comprising of the two component surfaces, which is finally ejected from the interface. .he surfaces and any surface contaminants are removed in the jet. Fig.1. D+namic situation at the collision #ront showing the jetting mechanism Dehind the collision front, the now clean surfaces bond, under e)treme pressure, in the solid state. .his dynamic welding situation is shown in Fig#1. In cross section, the materials usually bond together in an undulating wave form and the process can weld a parent plate of thickness $.$25mm to over "m 9the ma)imum flyer plate thickness is one third that of the parent plate:. !p to *$m can be welded in one e)plosion.
2

urrent %tatus
E)plosive welding was first recognised as a solid state process in "#;; when solid state welding had occurred between two metallic discs which had been in contact with a detonator. It was not until the "#F$s that the process was e)ploited commercially throughout the world. .he process was mainly used to clad large areas of one metal with another. .he materials that are commonly clad are shown in Fig#2. 52

Fig.2. ommon clad material combinations

urrent issues
8ore recently the process has moved away from simple cladding and is used to produce corrosion resistant pressure vessels, transition joints for shipbuilding, electrical busbars and heat e)changers for nuclear installations. -uture developments of the techni'ue might create advanced composite structures for the aerospace industry.

"ene#its
.he advantages of e)plosive welding over conventional welding techni'ues are as follows4 !sed to join dissimilar metals +chieves high bond strength 8aintains parent metal 'ualities +chieves welds over large areas Ae'uires low capital outlay roduces minimal distortion of parent metals Enables remote welding to take place Enables welding in hostile environments

$isks
E)plosive welding re'uires a remote location, has the danger of handling and storing e)plosives and therefore needs strict security. .he techni'ue is restricted to a batch process which limits the speed of weld production.

&agnetic pulse welding

Description

5B

8agnetic pulse welding is a high speed single shot welding process for joining tubular components or tube to a solid part in a lap configuration. It is similar in operation to magnetic forming. .he components to be welded are positioned within a coil, through which a heavy current flows from a capacitor discharge supply. Eddy currents are induced in the components and the interaction of the magnetic fields so created cause forces to be developed in the coil and components 9conductors with current flowing in the same direction attract one another and conductors with current flowing in opposite directions repel:. .he forces developed between the components can be sufficiently high that the surfaces impact together at a slight angle to form a jetting action similar to an e)plosive weld. .his produces a solid phase joint with very little heating. Fig.1. %chematic e,ample o# magnetic pulse welding " % lug to tube bond 9taper e)aggerated: 2 % 1apacitor discharge power supply * % ermanent water cooled coil ; % 8agnetic field shaper

urrent status
.he process was developed in the early "#2$s. It has been used for welding caps into aluminium alloy tubes for nuclear application and was demonstrated for a range of automotive applications in steel with an induction preheat included, although recorded applications were rare, possibly because of reliability difficulties at that time. &owever, interest has recently increased for automotive applications in particular and e'uipment suppliers are conducting application studies. .here are also now some production applications for products such as aluminium/steel driveshafts.

!mportant !ssues
+s the process has little published history, there needs to be an understanding of the capabilities of the process and the properties of joints produced. +spects such as joint design guidelines, process tolerance and 'uality control need to be established.

"ene#its
.he techni'ue appears to be capable of joining a range of materials, including dissimilar combinations. .he welding speed is very high as the complete weld is made in milliseconds. .he power demand is low and the process is claimed to have e)cellent reproducibility. In principle, the process is attractive for a range of mass production tubular applications, provided the joint designs allow the welding mechanism to operate effectively without damage to the part.

5#

$isks
,elding performance needs to be established for particular applications but there are few safety issues as the e'uipment can be fully automated. 1oil design is a critical factor to avoid reduced coil life due to fatigue failure.

Cnderwater welding

3 a ke+ technolog+ #or repairing marine structures .,I has been involved in underwater welding research for over 25 years for the marine, offshore and oil and gas industries. 1onsultancy and development work has covered consumables, automated underwater welding, and e'uipment.

Description
!nderwater welding can be divided into three main types4

,et underwater welding, where manual metal arc welding 988+: is the most common process. -lu)%cored arc welding 9-1+,: has been widely used in the former /oviet !nion. -riction welding, which has the advantage of being relatively insensitive to depth, and which lends itself to robotic operation, has the potential for use in deep water repair. 1offer dam welding, which is carried out in the dry, in air, where a rigid steel structure to house the welders is sealed against the side of the structure to be welded, and is open to the atmosphere. &yperbaric welding, in which a chamber is sealed around the structure to be welded, and is filled with a gas 9commonly helium containing $.5 bar of o)ygen: at the prevailing pressure.

urrent status
.he materials which are most commonly welded are microalloyed 1%8n steels of pipelines and offshore structures, and 1%8n steels on ships and harbour works. E)tensive repairs have been carried out to offshore platforms following hurricane or e)plosion damage. 1offer dam welding is most likely to be employed in harbour works or ship repair, although wet underwater welding is also widely used. /emi%automatic -1+, has been used in the former /oviet !nion for repairs to 9shallow: pipelines and to the hulls of sunken ships, prior to refloating.

F$

&yperbaric welding, using 88+ 9/8+:, .I0 90.+: or -1+,, is the preferred process for high integrity welds, particularly for deep water welds, including tie%ins in pipelines and risers in the oil and gas industries.

!mportant current issues

.he technology for hyperbaric welding is well established, and generally well researched. .here is, however, a need to be able to use the process at ever greater depths, with research being carried out in the range 5$$ to ",$$$m. ,et underwater 88+ 9/8+: welding has now been widely used for many years in the repair of offshore platforms, including those which have suffered hurricane or war damage. -uture needs will include the re'uirement to repair offshore structures in deeper waters, fabricated from higher carbon e'uivalent steels. In view of the high weld metal diffusible hydrogen levels generated in wet underwater welding, there is a risk of hydrogen%assisted cracking which increases with increasing carbon e'uivalent. Aevised electrode coating formulations and improved power source technology will be re'uired. ,et underwater -1+ welding may find wider use in the west, if it proves capable of giving acceptable weld 'uality.

"ene#its
!nderwater welding provides a means of assembly or repair underwater. +lternatives, which include clamped and grouted repairs 9which may introduce unacceptably high loading on offshore structures:, and the use of bolted flanges for tie%ins, are not necessarily always satisfactory.

$isks
-or the humans involved, the risks are of three main types. -irstly there is a potential risk to the welder/diver of electric shock. recautions include achieving ade'uate electrical insulation of the welding e'uipment, shutting off the electricity supply immediately the arc is e)tinguished, and limiting the open%circuit voltage of 88+ 9/8+: welding sets. /econdly, hydrogen and o)ygen are produced by the arc in wet welding and cutting. recautions must be taken to avoid the build%up of pockets of gas which are potentially e)plosive. .he other main area of risk is to the life or health of the welder/diver from nitrogen introduced into the bloodstream during e)posure to air at increased pressure. recautions include the provision of an emergency air or gas supply, stand%by divers, and decompression chambers to avoid decompression sickness following rapid surfacing after saturation diving.-or the structures being welded by wet underwater welding, inspection following welding may be more difficult than for welds deposited in air. +ssuring the integrity of such underwater welds may be more difficult, and there is a risk that defects may remain undetected.

E,pertise
.,I(s arc welding engineers and technicians have a wealth of know%how and e)perience gained through many years of industrial problem solving and AO7. 1onsultancy and development work has covered4

88+ welding consumables flu)%cored wire welding consumables O e'uipment remotely operated, mechanised welding e'uipment wet cutting techni'ues

F"

$esource

e'uipment O consumables for wet welding wet welding tank for preliminary trials indoor, heated deepwater tank with secure, 2; hour facilities at .,I .echnology 1entre 96orth East:, 8iddlesbrough advice on process and consumable selection, weld procedures O best practice troubleshooting O feasibility studies process O e'uipment development

Cnderwater #acilities
.,I .echnology 1entre 96orth East:, 8iddlesbrough provides a total support package for the specialist and operational re'uirements of the sub%sea/offshore industry. .his state%of%the%art facility meets the needs of sub%sea/offshore operators, from underwater training and 'ualification of personnel, to AO7, e'uipment testing and development, inspection and calibration.

*he underwater tank is the most modern underwater operations centre in the C2

.he tank is Bm in diameter and F.5m deep, with a volume of *"B cubic metres

!nderwater welding work e)pands at .,I 6orth 9 (onnect, Iuly "##F: +t its 0reat +bington head'uarters, .,I has a smaller wet welding tank, which will accommodate a welder/diver. ,elder/diver training is also carried out in .hailand.

8ow is it that arc welding can be carried out in water@

+rc welding is achieved by arranging an electrical circuit in which current passes from the welding power source through the welding electrode, across an arc to the workpiece, and through a return lead to the power source. -or the safety of the welder/diver during wet underwater welding, it is vital that he or she is completely insulated from the welding circuit. +n insulated electrode holder or welding gun is used, and the welder/diver wears rubber gloves. -or safety, on the verbal directions of the welder/diver, the power supply 9which is controlled from the surface, via a double% pole (knife( switch: is switched off immediately welding stops. =nce the welding circuit is completed by striking an arc, the heat of the arc is sufficient to vaporise the surrounding water. &ence, the arc is surrounded by a vapour shield, similar to the gas shield which surrounds an arc during welding in air. &owever, reaction with the molten metal influences the composition of this vapour, and it comprises typically Q2$@ hydrogen, Q25@ carbon dio)ide and Q5@ carbon mono)ide. .he resulting weld metal diffusible hydrogen level can be e)tremely high 9up to "$$ ml/"$$g of deposited metal for rutile electrodes:, but is much lower for o)idising electrodes 9Q25 ml/"$$g of deposited metal:. .hese levels increase as the welding depth increases, thereby increasing the risk of hydrogen%assisted cracking. /ee Aefs for information on the metallurgy of wet underwater welding.
L",2M

F2

8ost wet underwater welding is carried out using manual metal arc electrodes, which have a waterproof coating over the flu) coating. /emi%automatic welding e'uipment has been developed which uses a solid wire surrounded by a gas shield, retained within a (water curtain(. -ollowing its early use in 0ermany , the process has been revived more recently in Iapan , where it has been used on 8egafloat. + variant of the process uses a wire brush instead of a water curtain to contain the gas shield. ,eld metal diffusible hydrogen levels are comparable with those for welds deposited in the dry. +n alternative semi%automatic process, developed in the former /oviet !nion, uses a flu)%cored wire without any shielding gas . ,et underwater welding has been used for permanent repairs to harbour works, offshore structures, pipelines and ships. /ee Aefs for information on underwater repair.
L*M L;%2M LB,#M L"$%"2M L","$,"","*M

an sound welds be made b+ wet underwater welding@

,et underwater welding re'uires considerable skill and, depending on materials and consumables, carries a high risk of fabrication hydrogen cracking. -or critical applications, it should only be undertaken by e)perienced welder divers using established procedures. +lso, as there are wide differences in the operability and resulting weld 'uality obtained with different commercially% available electrodes, the choice of welding consumables should be made with care. ,hilst satisfactory welds can generally be made in the flat position, and, with the better electrodes, in the vertical position, welding overhead presents a considerable challenge, and weld 'uality in this position is likely to be poorer than in the other welding positions. In spite of this difficulty, it has been recognised for many years that welds of acceptable integrity can be deposited by this process. /atisfactory repairs have been reported for !./.+. 6avy ships and for offshore structures. /uccessful repairs have also been carried out to offshore structures in the 0ulf of 8e)ico, following hurricane damage, and in the 6orth /ea. .he integrity and mechanical properties of wet underwater welds have been evaluated by a number of authors e.g. , and welds that meet +,/ 7*.F class + re'uirements 9that are the same as for welds deposited in air: have been reported . -erritic weld deposits generally show only modest ductility and impact toughness, with 6i%base deposits giving better mechanical properties.
L"%*M L",2M L;,5M LFM L*,2M LBM L#%""M

In nuclear applications, satisfactory repairs to stainless steels have been carried out underwater, using stainless steel manual metal arc electrodes. &owever, it should be noted that the welding of ferritic steel with stainless steel electrodes is generally not successful, as fusion boundary cracking, and sometimes solidification cracking, occur.
L"2,"*M L",";M

.he semi%automatic wet underwater welding processes (water curtain welding( 9with the action of a conical water jet containing a gas shield: and flu)%cored wire welding 9without a gas shield: have been used with some success. .he former is capable of producing high integrity welds. ,hilst success has been claimed for the latter in the repair of ships and pipe%lines, the 'uality of welds has generally not reached that achieved with the other welding processes.
L"5,"FM L"2M L"B,"#M

F*

)rc spra+ing
+rc spraying is the highest productivity thermal spraying process. + 71 electric arc is struck between two continuous consumable wire electrodes that form the spray material. 1ompressed gas 9usually air: atomises the molten spray material into fine droplets and propels them towards the substrate. .he process is simple to operate and can be used either manually or in an automated manner. It is possible to spray a wide range of metals, alloys and metal matri) composites in wire form. In addition, a limited range of cermet coatings 9with tungsten carbide or other hard materials: can also be sprayed in cored wire form, where the hard ceramic phase is packed into a metal sheath as a fine powder. .he table shows the typical performance of arc spraying compared with other thermal spraying processes. omparison o# thermal spra+ing process and coating characteristics (article /elocit+ m.s
31

)dhesion &(a JB "$%*$ 2$%2$ P2$

4,ide content 0in metals1 7 "$%"5 "$%2$ "%* "%2

(orosit+ 7 "$%"5 5%"$ 5%"$ "%2

Deposition rate kg.hr

31

*+pical deposit thickness mm $.2%"$ $.2%"$ $.2%2 $.2%2

%"

-lame +rc

;$ "$$

"%"$ F%F$ "%5 "%5

lasma 2$$%*$$ &>=F$$%"$$$

+ combination of high arc temperature and particle velocities greater than "$$ m.sec gives arc sprayed coatings superior bond strengths and lower porosity levels when compared with flame sprayed coatings. &owever, use of compressed air for droplet atomisation and propulsion gives rise to high coating o)ide content. )rc spra+ing process

amsha#t bearing reco/ered b+ arc spra+ing with 137 r steel

)pplications

F;

+rc spraying has the highest deposition rate of the thermal spraying processes and can be used to spray large areas or large numbers of components on repetitive production line operations. .ypical applications include4 /praying of large structures such as bridges and offshore fabrications with Einc and aluminium to give corrosion protection. Aeclamation of engineering components such as journals, bearings and shafts with steel and bronEe alloys. /praying of electronic component housings with copper, Einc and aluminium to give conductive coatings which provide shielding from electromagnetic interference.

De/elopments
+lthough arc spraying has been available for many years, the technology is still being developed. 6otable improvements are systems which use an inert gas shroud. .his gives a lower o)ide content and improves deposit efficiency, offsetting the higher gas cost. E'uipment suppliers have developed improved power sources that reduce dust and fume levels and give enhanced deposit efficiency and o)ide content.

oating 9ualit+
/urface preparation and the spraying parameters used with the process are key factors affecting coating 'uality. /urface preparation is important for coating adhesion and can affect the corrosion performance of the coating. It is influenced mainly by grit blast profile and surface contamination. /praying parameters are more likely to affect coating microstructure and also influence coating performance. Important parameters include gun to substrate orientation and distance, gas flow rates and powder feed rates. If any of these are changed, they may affect coating 'uality. +dhesion of a thermally sprayed coating is mainly due to mechanical keying. &owever, this does not allow the bond strength to remain independent of the substrate material. .here can be significant differences between the level of adhesion achieved on carbon steel and stainless steel for e)ample. It may be due to the level of surface o)ide or differences in thermal e)pansion. /uch factors should be considered before applying the coating. +ll thermal spray coatings contain a degree of internal stress. .his stress increases as the coating gets thicker. .herefore, there is a limit to the thickness of coating that can be applied. In some cases, a thinner coating will have a higher bond strength.

"ene#its
.here are particular reasons why arc spraying may be selected over other surface engineering techni'ues and they are likely to include a combination of the following4 .he need for lower costs and higher spray rates 1oating of large areas either by manual spraying or using automated e'uipment .he re'uired coating performance can tolerate some o)ide and porosity, yet better adhesion and higher deposition rates are needed than can be achieved by using flame spraying.

$isks and precautions

F5

+s with all the thermal spraying processes there are particular health and safety issues that must be addressed before spraying4 E,traction o# dust and #ume % +ll thermal spraying processes produce dust and fume. .herefore, ade'uate e)traction or ventilation to remove this dust and fume from the particular working environment should be provided. In addition, if the operator is e)posed to this environment, then suitable breathing e'uipment should be considered. )rc e+e % +rc spraying produces large amounts of !> radiation and welding type eye protection is needed. .he correct grade of welding screen must be used. !sers should refer to E6 "F# which specifies a range of permanent filter shades of gradually increasing optical density which limit e)posure to radiation emitted by different processes at different currents. It must be stressed that shade numbers indicated in the standard and the corresponding current ranges are for guidance only. .he operator(s own preference and the application should be taken into account when selecting the shade number for a particular task. /tandard filter glasses are now marked with the 1E mark showing they have been independently tested to meet the full re'uirements of the standard. E6 *2# defines re'uirements for the photosensitive variable density lenses that are now available. .hese can be used with complete confidence, as there are failsafe re'uirements in the standard such that even if the lens does not darken when the arc is struck, daEEle may occur but no permanent eye damage will result. .he overriding benefit of such reactive lenses is the welder(s ability to see and position the electrode correctly before striking the arc. .his can greatly reduce arc initiation defects. 8ore information on arc eye and radiation haEards can be found in4 -+C4 ,hat is arc%eyeU -+C4 !ltraviolet, visible and infrared radiation haEards Fire or e,plosion o# dust and #ume % Duild up of some metals in the form of dust can lead to fire and e)plosions. .his is of concern for powders of aluminium and Einc. .o avoid this, ade'uate provision for e)traction and filtration of the spray dust, along with regular cleaning, should be provided. .he dust should not be allowed to get damp as this can lead to evolution of hydrogen gas. +dditionally, sources of ignition should be avoided such as static charges from rotating e'uipment, for e)ample fans in the ducting.

Cltra/iolet? /isible and in#rared radiation ha:ards

Aadiation is emitted by the welding arc in three principal ranges4% *able 1 *+pe Infrared 9heat: -a/elength 0nanometers1 P2$$

FF

>isible light !ltraviolet radiation

;$$%2$$ J;$$

.hese types of radiation do not cause ionisation of body tissue, but they can still cause damage to body systems and membranes.

Cltra/iolet radiation 0CF1

!> is generated by all arc processes. E)cess e)posure to !> causes skin inflammation, and possibly even skin cancer or permanent eye damage. &owever the main risk amongst welders is for inflammation of the cornea and conjunctiva, commonly known as (arc eye( or (flash(.

)rc e+e
+rc eye is caused by !> radiation. .his damages the outmost protective layer of cells in the cornea. 0radually the damaged cells die and fall off the cornea e)posing highly sensitive nerves in the underlying cornea to the comparatively rough inner part of the eyelid. .his causes intense pain, usually described as (sand in the eye(. .he pain becomes even more acute if the eye is then e)posed to bright light. +rc eye develops some hours after e)posure, which may not even have been noticed. .he (sand in the eye( symptom and pain usually lasts for "2 % 2; hours, but can be longer in more severe cases. -ortunately, arc eye is almost always a temporary condition. In the unlikely event of prolonged and fre'uently repeated e)posures, permanent damage can occur. + person would have to be stubborn and/or stupid to allow himself to be repeatedly e)posed to such arc eye risks without taking some precautions. .reatment of arc eye is simple, rest in a dark room. >arious soothing anaesthetic eye drops can be administered by a 'ualified person, or hospital 1asualty 7epartments. .hese can provide almost instantaneous relief.

Cltra/iolet e##ects upon the skin

.he !> from arc processes does not produce the browning effect of sunburn5 but does cause reddening and irritation caused by changes in the minute surface blood vessels. In e)treme cases, the skin may be severely burned and blisters may form. .he reddened skin may die and flake off in a day or so. ,here there has been intense prolonged or fre'uent e)posure, skin cancers can develop, but there is little evidence of this in welders.

Fisible light
Intense visible light particularly approaching !> or (blue light( wavelengths, passes through the cornea and lens and can daEEle and, in e)treme cases, damage the network of optically sensitive nerves on the retina. ,avelengths of visible light approaching the infrared have slightly different effects but can produce similar symptoms. Effects depend on the duration and intensity and to some e)tent upon the individual(s natural refle) action to close the eye and e)clude the incident light.

F2

6ormally this daEEling does not produce a long%term effect but in welders it is thought to progressively reduce their ability to adapt to e)treme light conditions.

!n#rared radiation
Infrared radiation is of longer wavelength than the visible light fre'uencies, and is perceptible as heat. .he main haEard to the eyes is that prolonged e)posure 9over a matter of years: causes a gradual but irreversible opacity of the lens. -ortunately, the infrared radiation emitted by normal welding arcs causes damage only within a comparatively short distance from the arc. .here is an immediate burning sensation in the skin surrounding the eyes should they be e)posed to arc heat. .he natural human reaction is to move or cover up to prevent the skin heating, which also reduces eye e)posure. .here is very little evidence that welders can be e)posed to the re'uired intensity of radiation long enough for lens cataracts to be formed by infrared radiation. =)y%fuel cutting can also emit high levels of infrared radiation and it is recommended that anti%flash, or impact resistant, eye protection is worn by anyone continuously engaged in heating or thermal cutting processes.

)/oiding the ha:ards

+lthough there are differing effects from !>, visible and infrared radiation, there is one common protection mechanism that is completely effective5 this is to provide a barrier to prevent the radiation reaching sensitive surfaces. .he welder should therefore be e'uipped with protective e'uipment as indicated below.It should not be forgotten that radiation can be reflected off shiny surfaces, and several cases of arc eye attributable to unwanted reflections have been recorded. .he walls, etc, of the work area should have a matt finish.

E+e protection
.he welder protects his eyes by means of a filter glass to absorb the radiation in the dangerous wavelengths, and limit visible light so he can see the progress of the welding process. .here are two basic types4 permanent filters, and photosensitive filters which react rapidly to the incident light from the arc and darken. D/ E6"F# specifies a range of permanent filter shades of gradually increasing optical density which limit e)posure to radiation emitted by different processes at different currents. It must be stressed that shade numbers indicated in the standard and the corresponding current ranges are for guidance only. .he operator(s own preference and the application should be taken into account when selecting the shade number for a particular task. /tandard filter glasses are now marked with the 1E mark showing they have been independently tested to meet the full re'uirements of the standard. D/ E6 *2# defines re'uirements for the photosensitive variable density lenses that are now available. .hese can be used with complete confidence, as there are failsafe re'uirements in the standard such that even if the lens does not darken when the arc is struck, daEEle may occur but no permanent eye damage will result. .he overriding benefit of such reactive lenses is the welder(s ability to see and position the electrode correctly before striking the arc. .his can greatly reduce arc initiation defects.

FB

+lthough arc%eye and other radiation effects appear to be the most significant haEards for welders, more than half all eye injuries are caused by flying particles of slag, grinding, chipping etc. It is therefore strongly recommended that anyone working close to arc welding activities should wear some eye protection even when arcing has stopped.

8ead and #ace protection

-ilter glasses are relatively small and are mounted in a dark, opa'ue shield, either hand%held or pivoted on a head%band so it can be raised or lowered by a movement of the head. .he shield has to be designed to screen the entire face, ears and portions of the neck from the direct radiation from the arc. D/ E6 "25 lays down re'uirements for the basic types.

'lo/es/gauntlets
&ands are usually the closest part of the body to the arc and the work piece. It is therefore important that welder(s gloves provide thermal insulation as well as blocking out !> and visible light fre'uencies. .he gloves should be designed to cover hand and wrist and overlap the sleeves. ,ith manual metal arc and 8I0/8+0 processes, spatter can also be a problem, and therefore gloves need to be able to resist penetration by droplets of molten metal. .he combined effects of !> and oEone can rapidly degrade many glove materials. .he durability of the material has to be taken into account in relation to the process control re'uirements. -or e)ample, tightly woven cotton or supple leather gloves may be ideal for low current .I0 welding where a delicate control of the torch is re'uired, but where little heat, and no spatter is generated. -or most other arc welding processes, which emit high levels of radiation and spatter, much heavier or more substantial gloves are re'uired.

lothing
+lmost any heavy%duty, dark coloured, opa'ue fabric will block !> and infrared radiation. &owever, as with gloves, damage by spatter and the combined effects of !>/oEone may be significant, depending upon the application. .he welder(s clothing must cover all parts of the body, arms, neck and chest that could otherwise be e)posed to direct arc radiation. &eavy%duty cotton overalls are usually the minimum re'uired for protection. 8an%made fibres and plastics are not suitable as they may be melted by spatter or even infrared heat. Dritish /tandard D/ E6 ;2$%" specifies the design features and the spatter resistance for clothing suitable for welders. It is important to prevent the welder from becoming too hot. &e will be close to a source of intense heat and a complete suit of heavy protective clothing might significantly increase his discomfort. Hocal protection in the form of chrome leather aprons, hoods, capes, spats, half jackets or knee%pads is effective. In this way the right degree of protection can be provided where re'uired and the rest of the welder(s body can be protected ade'uately and comparatively ine)pensively, for e)ample, by overalls. -ootwear, not normally subjected to radiation, is also important. It must be able to resist molten spatter falling on it from above, or being trodden on thus melting the sole. +nkle boots with anti% crush toecaps are recommended for all processes e)cept .I0 welding, where shoes 9with protective toecaps: may be ade'uate.

(rotection #or welder.s helpers

F#

+nyone working regularly within 2m of a welding arc needs to be protected against skin and eye e)posure in the same way as the welder. &e should at least have overalls, gloves and a hand%held or head shield if re'uired to look at the arc. +dditionally, he should have anti%flash glasses with side pieces to protect from inadvertent arc eye haEards. .he same recommendation applies to one welder working in comparatively close pro)imity of another. ,elders fre'uently suffer arc eye from inadvertent e)posure not to their own arc, but to that of another welder working one or two metres away from them. .inted anti%flash glasses may be selected from D/ E6 "F#, scale number ".2 % ; for e)ample. If the assistant is to work at a similar distance from the arc as the welder, then the same filter number should be selected for the assistant as for the welder. D/ E6 "25 contains details of the specifications for robustness and impact resistance for the various types of eye/face protection available.

ontact lenses
Aeports that radiation from arc processes can fuse contact lenses to the eye are entirely without foundation. .he !< Employment 8edical +dvisory /ervice and others have investigated the subject and issued statements saying that there is no risk that contact lenses can stick to the cornea due to incident radiation from welding. &owever contact lens wearers can still suffer arc eye, and must remove the lens immediately they feel any discomfort, which could be some hours after the actual e)posure. ,elding also e)poses the eyes to heat and dust so that it may be necessary to remove and wash the lenses more fre'uently than normal.

E+e protection #or others

,here practicable screens, walls or partitions should be provided to stop any arc radiation reaching the eyes of others working or passing through the area. artitions or walls should be painted with matt colours to minimise reflections and glare. /creens or curtains can be either fi)ed permanently or on portable frames where welding can take place at various locations in a shop. -le)ible translucent plastic material is available from welding distributors, which will positively filter out !> glare and the other harmful blue%white radiation. =bviously opa'ue te)tile or plastic materials will also stop radiation and provide complete protection, but they also restrict visibility. 8ost rigid polycarbonate or even ordinary glass windows will also reflect or absorb enough harmful radiation to prevent eye injury to someone observing an arc through the material. &owever obviously the glare will be transmitted and can cause daEEle if the observers and the windows are too close 9i.e. closer than *m to the arc:. If such materials are to be used for prolonged observation of an arc, the supplier should be asked to confirm that the material is suitable. If no screens or windows are possible, then safety must be provided by arranging that unprotected observers should not be allowed to approach closer than "$m to an arc. +t this distance even a wilful observer who insists at staring at the arc for up to say "$ minutes would only be daEEled.

2$

Flame spra+ing
-lame spraying is part of a wider group of coating processes known as thermal spraying. In these processes, a consumable 9usually a powder or a wire: is heated and propelled onto a substrate to form a coating. -lame spraying is the oldest of the thermal spraying processes. + wide variety of materials can be deposited as coatings using this process and the vast majority of components are sprayed manually. -lame spraying has distinct advantages, including ease of application and low cost, compared with the other spraying processes. .hese benefits make it a widely used process. &able 1 highlights the typical performance of flame spraying compared with the other thermal spraying processes. *able 1. omparison o# thermal spra+ing process and coating characteristics (article /elocit+ m.s -lame +rc ;$ "$$
31

4,ide )dhesion content &(a JB "$%*$ 2$%2$ P2$ 7 "$%"5 "$%2$ "%* "%2

Deposition (orosit+ rate 7 "$%"5 5%"$ 5%"$ "%2 kg.hr "%"$ F%F$ "%5 "%5
31

*+pical deposit thickness mm $.2%"$ $.2%"$ $.2%2 $.2%2

lasma 2$$%*$$ &>=F$$%"$$$

-lame spraying uses the heat from the combustion of a fuel gas 9usually acetylene or propane: with o)ygen to melt the coating material, which can be fed into the spraying gun as a powder, wire or rod. .he consumable types give rise to the two process variants4

owder flame spraying ,ire flame spraying

-or the powder flame spraying process, powder is fed directly into the flame by a stream of compressed air or inert gas 9argon or nitrogen:. +lternatively, in some basic systems, powder is drawn into the flame using a venturi effect, which is sustained by the fuel gas flow. It is important that the powder is heated sufficiently as it passes through the flame. .he carrier gas feeds powder into the centre of an annular combustion flame where it is heated. + second outer annular gas noEEle feeds a stream of compressed air around the combustion flame, which accelerates the spray particles towards the substrate and focuses the flame.

%chematic o# the powder #lame spra+ process

2"

%chematic o# the wire #lame spra+ process

In the wire flame spraying process, the wire feed rate and flame settings must be balanced to produce continuous melting of the wire to give a fine particulate spray. .he annular compressed air flow atomises and accelerates the particles towards the substrate.

)pplications
-lame spraying is widely used were lower coating costs are desired and a lower coating 'uality can be tolerated. /ome typical applications include4

1orrosion protection of structures and components 9e.g. bridges, offshore platforms, H 0 bottles: with aluminium or Einc coatings. +luminium is more e)pensive, but has resistance to acidic gaseous atmospheres 9such as those associated with the products of fossil fuel combustion:, as well as neutral solutions, such as salt water. 3inc has resistance to alkaline corrosion. -lame spraying is also used to spray corrosion resistant thermoplastic polymer coatings. Aeclamation of worn shafts, particularly of bearing areas with materials such as stainless steel or bronEe alloys. .he coatings produced are 'uite porous and lubricants can be absorbed into the coating, enhancing the performance of the bearing.

De/elopments
Aecent developments have looked at increasing the particle velocities. /uch processes are known as high velocity wire or flame spraying. .hese processes use internal combustion of the fuel and o)ygen gases to create higher gas velocities and give much higher particle velocities. Improvements in adhesion and o)ide content are claimed, e)ceeding those obtained by conventional flame spraying and arc spraying.

oating 9ualit+
.wo key areas that affect coating 'uality are surface preparation and spraying parameters. /urface preparation is important for coating adhesion and can affect the corrosion performance of the coating. .he main factors are grit blast profile and surface contamination. /praying parameters are more likely to affect the coating microstructure and will also influence coating performance. Important parameters include gun to substrate orientation and distance, gas flow rates and powder feed rates. .he bond of a thermally sprayed coating is mainly mechanical. &owever, this does not allow the bond strength to remain independent of the substrate material. .here are significant differences between spraying onto carbon steel and stainless steel, for e)ample. .his may be due to surface o)ide or thermal e)pansion, and such factors should be considered before applying the coating.

22

+ll thermal spray coatings contain a degree of internal stress. .his stress gets larger as the coating gets thicker. .herefore, there is a limit to how thick a coating can be applied. In some cases a thinner coating will have a higher bond strength.

"ene#its
.here are particular reasons why flame spraying may be selected over other surface engineering techni'ues and they are likely to include a combination of the following4 .he part geometry or the working environment re'uires manual spraying % flame spraying offers the easiest and most tolerable working conditions .here are cost concerns and the area is large and comple) .he re'uired coating performance is reached with flame spraying 7ust and fume levels associated with arc spraying are too high

$isks
+s with all the thermal spraying processes there are particular health and safety issues that should be addresses before spraying % E)traction of ust an fume +ll thermal spraying processes produce dust and fume. .herefore, ade'uate e)traction or ventilation to remove this dust and fume from the working environment should be provided. In addition, if the operator is e)posed to this environment then the provision of suitable breathing e'uipment should be considered. Fire or e)!losion of ust an fume .he build up of some metals in the form of dust can lead to fire and e)plosions. .his especially is of concern for powders of aluminium and Einc. .o avoid this, ade'uate provision for e)traction and filtration of the spray dust, along with regular cleaning of any ducting, should be provided. .he dust should not be allowed to get damp as this can lead to the evolution of hydrogen gas. +dditionally, sources of ignition should be avoided such as static charges from rotating e'uipment, for e)ample fans in the ducting. 7epending on the circumstances, it may be necessary to carry out a (&aEardous +rea 1lassification(.

8igh /elocit+ o,+#uel 08F4F1 spra+ing

.he most recent addition to the thermal spraying family, high velocity o)yfuel 9&>=-: spraying has become established as an alternative to the proprietary, detonation 97%gun : flame spraying and the lower velocity, air plasma spraying processes for depositing wear resistant tungsten carbide% cobalt coatings. &>=- spraying differs from conventional flame spraying in that the combustion process is internal, and the gas flow rates and delivery pressures are much higher than those in the atmospheric burning flame spraying processes.
.8

.,I has three &>=- systems installed at its +bington site near 1ambridge4

7iamond Iet &ybrid I 5$$$ .op0un

2*

8any applications overlap with plasma and 7% gun spraying, particularly for ,1%1o coatings.

.ungsten carbide%cobalt coatings for fretting wear resistance on aeroengine turbine components. ,ear resistant cobalt alloys onto fluid control valve seating areas. >arious coatings for printing rolls, including copper, alumina, chromia. 6i1rD/i coatings 9unfused: for glass plungers. 6i1r coatings for high temperature o)idation/corrosion resistance. +lumina and alumina%titania dielectric coatings.

8F4F spra+ing o# hip prosthesis with h+dro,+apatite

2;