57

CIGHIEENTH CONFERENCE

1951

THE LOW HYDROGEN (LIME-FERRITIG) ELECTRODE tZy l'. II. Z,ANGX>ON

The tleveloprr~entof the low hydrogen or lime-ferritic electrode as it is sometimes termed, is possibly the greatest achievement to date in n1etallit arc weldi~ig with flus coverecl electrodes. This electrode produces welds which are more resistant 1:o cracking than any other .fen-itic structural grade electrode, ancl its application has widened the range in romposition of steels weldable without resorting to some form of heat treatment in corijunction with the welding operation. As the name implies, the electrode employs a ferritic or special mild steel core wire with a flus covering which consists essentially of a lime \nse material. The principal feature of Ihe electrode is its virtual elirnmation of the h!-clro~en which is normally evolved in the arc atmosphere during weltlmg with ordinary structural grade electrodes.

This electrode was evolved as a result of research in America on steel electrodes the weldability of steel armour with austenitic stai~~less Just prior to and during World War II. I t was shown that the lime type stainless steel electrode was superior to the titania or rut& type :;tainless electrodes for we1tlin.g armour and that the lime type stainless electrode acl-lievecl its superiority due t.o the small amount of hydrogen cvolx-cd during welding. TABLE I.

I

Ordinary structural grade

Electrode Yield point tons per sq. in. ensile strength tons per sq. in. e r cent. a n 3.5411). uction of area per cent. impact ft. lb. . .

.

.

33-35

.. ..

.. ..

23-26 29--33 28-32

28-32

..

..

40-60 $5 (min.)

55-70 65 (min.)

".p .-

38-42

P

-Uthough the lime-ferritic electrode is a ntiltl steel type, it usually cleposits welt1 metal of higher tensile strength than that deposited by ordinary mild steel electrocIes. 1x1 addition, the weld metal possesses exceptional toughness and good ductility. In Ta.ble I typical mechanical ~xopertiesof lime-ferritic and ordinary structural grade electrodes are given. These physical prc3perties alone would render lime-ferrjtic electrodes u~orthyof use on many difficull apphcations, particularly those in which the weid is under a high degree of restraint. The primary advantage of

EIGHTEEN i H CONFERENCE

these electrodes, .however, is much more fundamental, and before considering this factor it will be necessary t o digress a little in order to discuss the reasons for difficulty in welding certain steels. Anyone associated with welding will agree that metallic arc welding of the liigher carbon ancl low alloy high tensile steels has always presented difficulties lxcause of the tendency of these steels to produce defective welds ant1 heat affected areas. Unless special precautions are taken the welding of these materials with ordinary mild steel electrodes generally results in the loss of cluctility of the welds and adjacent weld inctal and the frequent occurrence of crrtclring. In the past mild steel electrode coatirzgs have fallen into three more or. less well clefined categories. Firstly, the cellulose type whic1-r contains a large percentage of cellulosic material, anti relatively small amounts of mineral constituents, secondly the oxide type wllich consists basically of iron and manganese ooxicles and certain siliceous materials, and thirdly the titania type, of which titanium tlioxide is an essential constituc:nt. The last class includcs the majority of present day geIiera.l purpose smooth running clectrocles. T n the case of lime-ferritic electmdes, calcium carbonate andior calciuxn oxide forms tlzc basis of the coating. of a. steel is largely a function of its hardenability, The vc~eldabilit~. i.c.., the rnore deep hardening is, the rnoro difficult it. is to welcl. .I his is true with few exceptions.a. steel In order to unclerstand the I-ner1ianisn.r of hardening in steel, it will be necessary to discuss briefly the fundnmentals of ferrous metallurgy.

.

Fig. l-Showing

pearlitc in a n annealed carbon steel containing 0.8 per cent. carbon ( X 1000).

Metallurgy of' Steel. The reactioh upon wlrich ferrous nlet.;ilhirgy is based is the allotropic chrtnge which occurs in pure iron a.t $)OSo C. This change i s one of crystal structure, and profoundly alters the properties of the iron, particularly in its ability to dissolve carbon. Iron at room temperat~~rc is known .as "alpha-iron" or ferrite, aztl in this form will dissolve only some 0.007 per cent. carbon, which rnay be neglected for all prartical purposes. At temperatures above 903" C., iron exists in its second forin

i

1951

LIGHTEEN7"H CONFERENCE

59

and is known ,as "gamma-iron" and the solid solution of carbo~rin gamma-iron is referred to as austenite. 'The maximum solubility of c.\arbon in garnrna-iron occurs at 1100" C., where 1.7 per cent. carbon may be held in solution. This variation in the solubility of carbon is of the utmost importance, as will be shown later. At room tenlperatures carbon exists in steel as iron carbide and in plain carbon steels 'the a.rr~ountand distributiorr of this compound deter-mines the physical properties of the steel as a whole. When a steel is slowly cooled from the austenite region, iron carbide separates owing to the vcry low solul-~ilityof ca.rhun In ferrite. In separating, the iror~ carbide associates itself with a certain proportion of ferrit.e, to form a constituent known as pearlite which actually consists of alternate plates of ferrite and iron carbide, a^., showrn in Fig. I. The thicliness of the plates is clependent upon the rate of cooling, becoming increasingly fine as the cooling rate increaqes. 'The hardr~essand te~lsilestrengtl-1 of tile sl&l increase as the pearlite becomes finer, while the c1uctilii.y and toughness decrease. 'The arnount of pearlite present depends on the carbon coxlterlt and "Le cooling rate. I n very slowly coolect (annealed) steels, the rrsicrostructure corshists entirely of pearlite at 0.84 per cent. carbon. iiielom this carbon colltcr~tthe stru(:t~lrei:; pcarlite plrss excess fen-itc, while carbon cwntents above 0.81 per cent. carbon yield a structurr consisting of pearlite $ 7 ~ 1excess ~ iron carbide. By increasing the coolirig rate the amount of pearlite for a given carboil contc$nt i s inc:reasecl. Thus small air-cooled samples of 0.6 per cent. carbon steel mal- have a. structure- c-onsisting entirely of fine pe;rrlite.

Fig. 2---Martensitie of a 0.5 par cent. carbon steel (X10 0 0 ) .

Fig. 3-The coarse carbide dispersion resuftingl from heating at 7 0 0 " C. the steel shown in Fig. 2.

If the cooling rate is increased beyund a certain value the lamellar type of structure is no longer formed, and the amstenite transforms t o a new acicular or needle-like structure called martensite. I t is the hardest known constituent of steel, hut it is also extremely brittlti. The appearance of martensite is illustrated in Fig. 2. If martensite is re-heatecl to sub-critical temperatures (below 725" C.), its hardness dcrreases as the temperature is ra.ised, ~rntilat 700" C. it is cornparable

60

EIGHTEENTH CONFERENCE

195 1

with an annealed steel. This softening effect is brought about by a coalescence of the very fine dispersion of iron carbide which exists in martensite. As the temperature increases, the size of the carbide particle.; increases until at 700" C. a relatively coarse carbide dispersion i~ ohtainecl as shown in Fig. 3. Although in hardness it is similar to an annealed steel, the iron carbide exists in the form of roughly spherical particles, as opposctd to the lamellar structure of annealed steel. This "spheroidised" structure may also be obtained by heating the lamellar pearlite at 700: (:. for prolonged periods. The cooling rate at which the formation of pet~rlite is just supprcssed, and martensite forms, is called the "criticai cooling rate," and its actual value gives a measure of the bardenability of the steel. Hardenability refers to the ease with which full hardening may be obtained, and not the maximum hardness which may be obtrzined. It is possible to have a steel of low hardenability but with estrernely high martensitic hardness, or, conversely oi high hardenability and low niartensitic hardness. The plain carbon tool steel5 are examplc5 of the former type. The maximum hardness which is obtainable in :i givcn steel is a function of the carbon content. Tt should be noted at this stage that the I-lardness obtairled on quenching is dependent to some extent upon the time for which the steel has been held in .the austenitic region. The reamn for this is tlrat the change to homogeneous austenite is not instantaneous, a definite time being required for the carbide particles to dissolve in the garnmrt-iron. Thus a steel which has only been held for a short time in the austenitic region will not develop full hardness, since the carbon in undissolved carbides does not assist in developing martensitic hardness. In the case of alloy steels the rate of solution of the complex carbides formed is even slower, requiring longer tirnes than plain carbon steels. For this reason hardenability and weldability are not entirely analagous, since liardenability is based upon the reaction of a steel to quenching after it has been held long enough in the austenitic range to obtain a homogeneous austenite with all carbides in solution. \Veldability, on the other hand, is concerned only with the reaction of the steel to welding, where the heating and cooling cycle is rapid, and c-oniplete uniformity of structme is not obtained.

Alloy Elements: -The foregoing discussion has included only piain carbon steels, and before continuing, it is necessary to describe briefly the effect of alloying elements. The most common alloying elements ill steel are nickel,, chromi~im,molybdenum, and manganese, and the discussion will be confined to these. Alloying elements may exist in steel. either dissolved in the pure iron (ferrite), or combined with iron t-arhidc, forming complex carbides. Ilsually the particular element exists partly in each forrn, the a.monrit in each forrn depending upon the relationship between its solubility in ferrite, its affinity for carbon, and the cornposition of the steel as a whole. For example, in a high carbon low chromium steel, the chromium would be almost entirely coml~inedwit11 the iron carbide ; on the other hand, in a low carbon, high chronliu~n steel, most of the chromium would be uncombined. All alloying elements, with the sole exception of cobalt, increase tlle hardenability of a steel. This means that the tendency to harden is

1951

EIGHTEENTH CONFERENCE

.-

61 p -

greater in alloy steels than in plain carbon steels of similar carbon content, ancl consequently greater care is required in .welding these alloy steels. ,. llle effectiveness of the various alloying elements in increasing harden-. ability is cliffererlt ; carhon itself being the most effective, followed by I-nol?-htlenurn,chrornit-lrn,manganese and nickel iri that order. By apply--in{; a suitable empirical factor, the effect of each alloying element may 1)e expressed in terms of carbon, and the sum of these incliviciual val~lcs. givcs a figure lillowxl as the "'c~arboneqnivalent." The formula for the . calculation of the carbon equivalent i s : -hIn , per cent. C'r per cent. N-I per rcilt Xlo ('.E per cent (' 4 per cent. - -- - --4 - 15 B 5 4 170~-teisl containing 0.25 per c.c~~-t. C, 0.9 per cent. Xln, 0.73 per cent. C r , 3 0 IWI- cent. XI, and 0.2 per tent. MO, the carbon ecluix.alent woul(i be 0.8 ~ C rI e r ~ t . :

?'Ill.: :~llo>-steel yill &us react t.o wclclin:: in a niarliler si?nila.r to a pl;rili carbon steel coirtainirlg 0.8 per cenl. rarbolr. I t is irnportnnt t o realise that this formula is empirical and is, therefore, IJO! iilfallil-)l?,h u t it does give a guide, and is useful wherr alloj- steels are met. I-lo\vctvcr, this formiila. i5 ilpplic.a.ble 0111k- 10 tllc low n.lloy steels. ('oizditio!lr di$viil,q kl'i.ldiitc:. -1Iilrirr~~veldirl.:, a teinperaturc gradivnt 1111 in thc bar. heial, from t11e melting point at tile junction of t1:e (i(il)(l.?it and the Inse rnetal, to snbstantiall>-cold 11le'.;11 some clistancc frorn tl-e uic;~lautclc.r:r.~.k- ;lrt. 5hn1r.n in 1;iy. t i . The cracks in thr illristrntinn r m ~ I I P lrft II;II I, ~ ~ ~ n c t r ; t tnlnlrbqt rrl to t h r rclnts o I the rcrlds.

.\11tr1-cratkingi.; ustr:i11!. :ittnl~.itedto low strength arid rlur-tility :t t his!\ tc:nl~ratl~rr.;, t l ~ ewrld havini: a lower st rcnytI1 rltan t Ele I,a~elnr.t:~l, 5nIr-c thc lat trr 1s rel:~tivcl\.rn~~clr rr~oIcr. 'rlli? crnnrliticm ir ayarnv:~tctl r v l ~ miniltE qtccl clcrtmdcs arc uscd on t i i ~ hrarhon or alloy strrl, C)H'~IIS tlr t l ~ cinl:rrmtly Iliqlkcr strrncth of t h r 1;tttcr m-ttcrinls. I t will hr a ~ p r r c ~ a t cthat r l a s t h e lveltl ancl arljaccnt fircnq cnnl c-oi~tractlr~n take< pl;rr*r ant1 intrrtlal strrs~cl;xsr wt 1111 in the wrld arca. In nsscrnhlirq rv11c:r thr parts arc free to rnovc t h c TC~IIIU:LI ~ stresses ;trc T ~ : ! I I C P ~ ~I)!' mnl-cmcnt nf thc plates anrl any rr:nainin.: s t r ~ s sis arrornm~~latcrl),v t l ~ rrlrirtility nf t l i ~\vrlrl. I t ia irnportarlt to realise t h a t , ;~lthnzrgl~ rcsirlli,~lstress- are reduced by i~llorrrinyfret rnnvcrncnt of the plntrs, 11lc-l.;ire not r*ornplrtcIy rliminatrrl and x.itr3-cm-kin: mat- occur rrncn in unrrstsainrd walds in t11;~teriaInf Inw ~vvFtlahrlitv. The reason is t h ; ~ t tht. prk-ul' of rar1,r)n nr allcl!. from the lxnc metal i+c:lucrs the ductilif!. of thc rvclrl nlctal, wl~iclimay not he ahlc to acrommnrlatc the contraction ytrwsr?;. rtncks ilr the Rnse llelal.-One important aqpect of the weldinc of steels !raving low weldability, and one which i s Ily no means full?. appreciated, is that the a h e n r e of cracking in the ~velrl metal is not nwessanll; an indication of a sound titeltl. %me t y ~ osf qterl, particolarlv certain of the low alloy tvpes. are very prone to crark in the ],;\SF metal unless certain precautions are taken to obviate this. Rnsc

1951

EIGHTEENTH CONFERENCE

65

metal cracking may be of the hot or cold type, and usually is confined t o the heat affected zone. A particular form of cracking resulting from a condition known as caustic embrittlement sometimes occurs in boilers, solution vats, etc., containing an alkaline solution. This cracking, which is intercrystalline and irregular in nature, seems to originate in highly stressed zones and often penetrates into both the weld and the base. nletal proper. Since most base metal cracks develop in the heat affected zone, il is i:nportant to consider- the nature of thi.; zone. It. consists prim~riip of two parts :- (a) An Inner zcne, next to the ~ e l d ,which is heated durrng

uelding to a sufficiently high temperature to permit at least a partial change to austenite. Dependent on the cooling rate, this zone is transformed into a much harder and stronger product than the base metal, because if the cooling rate is rapid enough, the austenite phase transforms to the hard brittle micro-constituent martensite. Martensite is always assoc~atedwith cold cracks, hence the term "hart1 cracks. *' (h) An outer zone, which is not heatecl to the tra.nsformation temperature, but which is usually heated sufficiently to reduce the hardness and strength of the base metal, j~rovided the base metal is not in the fully annealed state.

There is usually a steep gradient in structure of the two zones, as characterised by the metallic arc process. The hot type of haze r.c:tal cracking is ra.re, but n x y he enco~ii~tcrecl in welding S.A.1:'. 1-130 aircraft steel. This is a nickel--clrroim.molybdenum steel containing 0.30 per cent. carbon, ancl it exllibits a. terldency to crack at the junction of the weld and base metal. This hot cracking tendency of steels depends not only upon the chel-nical coirrposition, but also upon the rnicro-structure of the particular stecl being welded. The llardness of clrlenched steel is clepcnclent to a c.ertain extent upoir the time of heating in the a.ustr:nitic region. Since the weldting cycle results in fairly rapid heating and cooling, the smaller the carbide particles, the grca.ter will I:e the dcgree of solution. Thus for a gi-:en stecl, such as S.A.E. 4130, the tendency for cracking will he greater when theedispersion of carbide particles is fine than \vlren it is coarse. Hot cracks are intercrystalline in na.ture ancl are associated with weaknesses between the grains or dendrites.

""Cold" cracking of the base metal takes place in the hardest zone of the heat affected area after the weld and base metal are substa!l?.ially cold, and it may not take place until some hours or ever1 days after welding. This l.ype of crack, which is transcrystalli~iv in naiurc, is a1waj.s associated with the formation of martensite in the heal: affected zone, ancl is frequently cauzed by internal stresses set up during cooling. Tke most common type of cold crack is the "underbead" crack, so ca.lled l;ecause it occurs just bnzieath the weld, and may remain hiclden, leaclin.= to failure in service. Cold cracks may also occur in the base meta.1 at the surface of the plate, just adjarent to the weld junction, ancl may extcnd from the ].eat affected zone ixto the unaffect.ed platc. 'This

66

tlGHTEENTH CONFERENCE

195 1

type of rrat- kin^ may be referred to as "toe" crackinq, but is still rindrr!>cat1 crackine. This type nf cracking and a micmn"maplr nf a crack are slinwn in I-ir. 7.

Fig. f-lllustrnfing

a typical "toe"

crack and a microgmph of this crock,

C A ! r-rar.kii1.q is thc most' scl-iom ant1 common drfcct in the heat illtc~-tetl ;ronr, and, inrlced, in ;I w~lrlecl conlllonpnt. I t is as~or'iatrrl witll crrtlin;ls! hcaiI, IiIIct and butt wcltls. 11 15 apparcilt that rvcn thr I>rittIrn1artrn3itr will not rmt-k, t~nlcwit is sr~ficirntly~ t r e s ~ tn ~ cesreecl l it< q t r ~ n11.~ t I-nlurti~nnteIv,such stresspi arc tnn f r e q t ~ ~ n t lprrs~nt y , nr icinc from thr followinq pnssilde sotlrres :5trrs~r.;rl~lct o i h c vo1111ncchanqc that take.; plnre on transfnma'tir,n. t
195 1

67

ElGkITEENT'H CONFERENCE

of the a.rc atmospheres of various types of electrode in an endeavour to determine the reason for the superiority of lime type coatings over other types. These analyses showed that much less hydrogen was evolved by lime coatings than, any other type, and experiments were then carried out t o obtain quantitative data on the effect of this element. In orcler to saturate the a.rc atmosphere with hydrogen, welds were rnade using hollow electrodes, through which a. corltinuous stream of hytlrogen flowed during xvelding. I t was found that welds rnade with an arc saturated with hydrogen showed a much greater incidence of underbead r:racl-to cracking lulider a given set of conditions.

EIGHTEENTH CONFERENCE

68

1951

The effect of making welds in atnlospheres containing various amounts of hydrogen is shown in Fig. 8. The increase in the incidence of underbead cracking as the proportion of hydrogen increases is obvious from the graph, and this confirmed the theory that hydrogen played an important part in causing unclerlsead cracking. Tests were subsequently carried out using various types of electrode, in order to determine the effect of the electroie type on underbead cracking. These tests curlfirintd t' e rcsults obta~nedin previous tests, ancl provecl that the advantage of the lime type coating lay primarily in the low hydrogen content of the arc atmospi:erc. TABLE 11. *Analysis of arc atmosphere Electrode

No.

--Lime type coating

CO per cent.

I

Water peZ?:nt.

Volume of gas produced cc. per inch of electrode

I

-2

----Or dir,ar y mildsteel t Y Pes

--- I

---

72 --

'I'ls analy-cs of the arc atmospkcl cs obtained from varioos electrodes are given in Table 11. I n these tests underbead cracking was experiencetl with all electrodes not having a llme type coating.

Cold cracking is conbidered to be caused by the resultant of the following factors, of which the trigger effect of hydrogen play- a rnajor role. Chemical composition and structure of the base metal. Cooling rate after welding. Stresses in the heat-affected zone. Hydrogen z~bsorptionand precipitation. The embrittling effect of hydr0gen.i~by no nleans confined to tbe heat affected zone, altholigh its effect is probably most noticeable in this region. Tbe deposited metal itself is adversely- affected by excessive hydrogen. Tke only visua,l eCfect of hydrogen is porosity, and even this may not show at the surface, but during a tensile test these gas holes open up and become very obvious, as inay be seen in Fig. 9. Here also is illustrated a cross-section of a broken tensile specimen. The effect of hydrogen is obvious, and areas of contamination are known as " fisheyes" and "flakes."

195T

69

El GHTEENTH CONFERENCE

In Table IT1 are given the physical properties of all-weId metal trst bars, welded with experim~ntaleIectrndcs, and these give some idea of the effect of hydrogen. The residrlal h y d r o ~ e ncontent is pro~mrtinnalto t h e cooling rate, and thus the air coalc,l sp:imcns represent t hc I~ighest hydrogcn content, while thr funl-ire cnn1e:l s p e ~ i m e n ?:irr t h e Ioiifest.

Fig. 9-lllustroting ITeft) the parosity o f a tensile specimen due t o hydrogen crnbr~ttlcmrntand Irighk), the "firheyes" noticed o t end o f fractured oll weld metal kensile bet.

Limitation of the Low Hydrogen Electrod?. Carhot? and Lozt' ,~illml N i ~ hT c l ~ s i ! ~Slre1s.-'The , low h!~drt~gcn elcctrude, which was not markctrrl until after the war, is primarilv used for wrltlinc rarhnn nnrl low alFov Ilich tcnqilc strels. The ranqe of thcsr. rnatrrials ~vcldal~lc wi tIlnu t p r e - l i c a t ~ n ~ rtc., , 1r1ll of conrse bc d e p n d e n t TABLE Ill. Yield point tons per sq. in,

Ultimate tensile strengrh

tons per

Air cooled-normal runs Air cooled-large runs C w l e d in ashes Furnace cooled

..

. ..

5q.

29.7 25.9 25.6

38.0 33.5 3 L .8

23.8

29.9

in.

Reduction in area per cent,

31.4 34.5 48.6 46-1

Elonga-

tton per cent.

15.0 19.7 19.0

30.0

,111wn the particular lime-territ ic elcctmde employed. Tn general, however, it may Iw said that for t tle alwve steeIs, the range in composition which is weldable tvitl~outsome form of heat treatment k i n g used in c n n j ~ ~ n c t i nwith n weldin?, is approsirnatelv rloubk that capahIe of being

70

EIGHTEENTH CONFERENCE

195 1

satisfactorily welded with ord inary mild steel electrodes. This limitation [A-CA9 'Ii'elding Code, wh~chrestricts the c a r l ~ ~ n is based an the S/: ., tIor . . . content of carhon stccl smmctural purpuses to 0.25 per cent. maximum. Thus with the low hyrlrogen ekctrode structural welds can be d e p s i t ~ d on normal plain carbon steeIs containing up to 0.5 per ccnt, carlxln withotit heat treatment. Similarly. strength uPeldsand weldments can he made without heat treatment, on low ;rlIoy hieh tensile steels with a carhon equivalent up to 0.8 per cent. J t is emphasized that tl~esespecifications are bayed on analysis only, and for any particular application consideration must he given to other nperatlng factors, such as thc size of the section involved, the design of the joint, spnticc! requirements. etc. These factors really resolre into standard practice and welding procedure and are therefore in t h e hands of the opcratnrs. : l good cxnmplc oi the use of thc lime-ierritic electrode nsetl for making a s t r c n ~ t hjoint on high carhon steeI is in the welding of rail qtee1. Jn Fig. l 0 is il111stratecla welded ra~lsection (70 11.1, rail, containing 0.65 per cent. c a r l ~ l nand 0.0 per ccnt. rnanpncsc) before and aftcr destructive testing.

Fig. I M h o w i n g o butt welded 70 lb. roil redion befare ond offer de&rucrion.

In Tables IV and V arc set out the results of hubt wrlrlcd tcnsik tests and hardness survevs on T 611et welds made with a law hyrlrogen clectrocle on 1045 and IOFO steels. I t will be noted that the 10M i n 10Kn and 1045 to lnfjO combinations, although sat isfactor?.. exceed the allove recommendation. I t is considered rlcsirable to adhere to thc above limits which, as mentinned previously, are based on analysis only, and thus ensure a greater factor of safety. Revond these limits, pre-heating on]?? is usually required. For highly stressed parts it may be desirable to pre-heat and then stress relieve after welding. Heavql Seclimts.-The use of the IOWhydrogen electrode for welding particularlv heavy sections is also recommended. Sometimes cracking . occtirs in the foundat ion n3n of filIet welds made on heavy sections, sucl~ as machine frames, etc., due t o the eftect of the combined strcss whirh cscceds the capacity of the deposited metal. The use to the lime-femtic electrotlcs on heavy section work is illustrated in Fig. 11. f f a ' ~ J 1 Tensile Steel.-Weld deposits from lime-ferritic electmdes generally have high yield point and masimum tensile stress, typical figures h i n g :-

1951

EIGHTEENT'k1 CONFERENCE

71

Yield point ... ... ... 32-25 tons per sq. in. Ultimate tensile stress ... 38-42 ,, ,, ,, These features in adtlition to excellent ductility and toughness, and of course low hydrogen content, make the electrode ideally suited for high tensile requirements. The successflrl welding of .motor car axles and springs represents two applications. TABLE I?!.-.-Butt

Piate

Area of section sq. in.

Tensile Tests (& in.

X

4

in. material).

Yield point per t o n per sq. in.

Remarks

24.3 23.8

Specimen broke in plate. Fracture through weld-caused by slag inclusions. Fracture through weld and heat affected zone adjacent t o weld.

-

The percentages of the alloying elsrnents in the t w o steels were :C. Mn. S. P. Si. Cr.

1045 1060

0.44 0.59

0.54 0.54

0.032 0.049

TABLE V.-Vickers Location of test

0.030 0.027

0.15 0.18

0.14 -

Hardness Test.* Range i n Hardness

Average hardness

198 290 244 210

Heat affected zone

*" T " fillet weld in

202-228 245-294 208-258 1 % in.

X

4 in.

212 268 222 material.

Cast 1ro.n.-The lime ferritic electrode may be used in the welding of both grey and rrialleable cast iron for work of secondary importance. Welds can, if necessary, be carried out without pre-heat, but owing to the tendency of cast irons to harden in the heat affected zone, a pre-heat to 200 250" G. is recommended, particularly where the maxrmum strength is required. Welds deposited on these materials by lime ferritic electrodes are normally not machinable, but may be made machinable by stress relieving at 650-700" G. after welding. .

In Fig. 12 and 13 are illustrated applications of the low hydrogen electrode, for welding high tensile steel and grey cast iron.

71

EIGHTEENTH CONFERENCE

195 1

S?rb-normal Tc~nfi~rafztrcs.-Of special consideration is the maintenance nf cluctility and toughness under conditions of extreme cold. This property gives the lime-ferritic ~Icctrodc special application for equipment requlred to operate in low temperature zoncs, h great deal of work remains to IH: done in this field, hut the lime-fenitic electrode is, at present, the best choicc for,machinerbl operating under sub-normal temperatures.

Fig. I l--Showmg

a heavy section welded with the low hydrogen electrode.

Cnsf Steels.--Tbc lime-ferritic electrode is also the first prcfcrence for wclrling cast steels. which usually have from 0.25 pcr ccnt. to 0.5 per cent, carhtln, hut cast steels with a carbon content of u p to 0.9 matf 11e welded. The same limitations apply with cast stcel as for plain carbon or low alloy steels. I [ .thc composition is Jouht ful, and particularly where a 11cavy mass of metal is concerned. a slight pre-heat to about 150" C. is advisable. .+Inapplication involving the use of the lime-ferri t i c * clcctrode on cast stccl is shown in Fig. 14. M'ro~rght and Cast Slepls r o n l a i n i n ~High S?~lphztrand Phos$hor~rs highly l~asictype slag produced hy this electrmle is ideally suited to thc removal of sulphur and phosphorus by s l a q i n ~ action, and since practically no hydrogen is evolved in the area the tendency to produce sulpE~urousgases, whiclr may dissolve in the weld metal, is eliminated, In addition, the penetration of the low hydrogen electrode is adequate but not excessive, and consequently the parent metal is not greatly affected. il'hilst the Iime-ferritic electrode can he Confenfs.-The

1951

EIGHTEENTH CONFERENCE

73

tiscd succ~ssfdlvfor welding sulphur bearing mild steel, its use i 3 not part icrllarly rccnmrnendcrl in preference to othrr clcctrorles dcvelnped l o t this work hrcaitsc tlle latter usually hxve slightly srnootl~crweldinq properties, nrr cheqwr, and are cntirelv satisfactory. A s kir a5 ~ ~ e l t l i n c Iiigll plicrsnhnnis strrl is concerned, exact Eimit?; have not y t bee11 rlrtcrminrd, but it will suffice tn state that a limp-lezritic ctcrtrcrrlc h 2 s hecn s~tcrrssfulll;tle;etl for welding a largc drii*inq shaft cnntaininq $:.b[

prr rrnt. phosphonls. cightrrr%months.

The shaft has no\\,h e n in service Ior \wll over

Fig. 72 (lefr) and Fig. 13 (right b 4 h w i n g respectively a truck axle ready for welding mnd a heavy cast iron frame repoired by welding with low hydrmen

electtode.

EnnnreIlin,n Steels.-One of the principal t m n l ~ l r scnwuntercrl when wr.ldinq stecl which is to 'tw suhsequentlv vitrrn~is~namellcdis that gas d t f f i w ~from ~ , the H'PICIS on aycinq and creates hlistrrs at i n t ~ r f a cof~ the is t h r prinripaI mctal and enamrl. It is grn~rall; acccptrrl that F~~~rlroqm gas rrvlvrd ancl t h p Eirnr-frrritic elcctrotlc has b r . 1 i~srdwith sttr'rrss for this claw of rvnrk. Hiclt Sz'licn~r Trransf~rmer Iron.-The effcrt of siliron on crack susceptibility appears to Ix more pmnouncerl w h ~ nthe rnangnnrsc cnntrnt of w~Ftlmetal is low. The lime-fcrritic electrode generally ha5 a higher manyanese contcnt than ordinar!? 11.5. tvpes and ha4 proved satisfactory for rveldin~ this material. Transformer iron has Iligh m a ~ n e t i cpermcabilitv and electrical resistance, and while variolzs gradcs arr availahlt, :L tvpical analvsis woirl(1 he :-4:arhon, 0.07 per rcnt.; !4n, 0.(1!)per rent.; Si, 4.3 pcr cent. The elwtrmb also has low smrrntiI7ilitl; to cratkinq with hip11 s~llphurand phnsphonls stcds, and thr-rforc the comhinctl cffcct of silicon and thew elements have less c h ~ n c eof promoting cracking with a lime-fewitic c!crtmdr than with other t y s .

The l i m e - f e q ~ t i electmde c achieves its spcial puqmsc Iwcause practicallv no hydrogen is evolved during welding. and it has csrrlrent mechanical properties, narnrly, a high yielt! point with high masimum

EIGHTEENTH CONFERENCE

74

1951

tensile strength; and extrerne toughness with good ductility. This conhination of properties rnakes the electrode the best s t r ~ ~ c t u rtype al available as regards resistance to \veld cracking. I t also has adequate but not excessive perletration and, besides, the highly basic lime type slag has the facilities for the removal of detrinlental impurities.

Fig. 14---Showing

a large cast steel turntable base fabricated with the low hydrogen electrode.

This type of electrode is not a "cure-all,? but, nevertlieless, by an ~mderstandingof the limitations of its functions, a much wider range of materials may be successfully welded without special treatment, anti in Inany other cases a modified heat treatment procedure Inay be adopted. The lime-ferritic electrode is not recommencled, nor is it likely to be used, for general structura.1 mild steel welding. Present-day types of smooth rlxnninq and general purpose electrodes, besides being very satisfactory for this mrorli, hare certain advantages over the lime-ferritic electrode, s~anlel!~,ease of running and manipulation, weld a.ppearance, and of course, cost. The lime-ferritic electrode is manufactureil by special processesancl u~itllspecial material, consequently the cost of procluction is higher than that of ordinary mild steel types.

E.,U.F. Electric Co., Melbozlnza.