MTL'-I 81-50

AD-A188 733TI1C FILE COPJ

HYDROGEN INDUCED CRACKS INARMOR STEEL WELDMENTSFABRICATED WITHAUSTENITIC FILLER ALLOYS

WILLIAM S. RICCIPROCESSING TECHNOLOGY DIVISION

'z.

September 1987

Approved for public relears; distribution unlimited.

DTIC-E•LECTE2 •;•DEC 0•C 07LB

US ARMYLABORATORY COMMANDMATERIALS TECHNOLOGY U.S. ARMY MATERIALS TECHNOLOGY LABORATORYLABORATORY Watertown, Massachusetts 02172.0001

'4Jt I•(

UNCLASSIFIED l 170 "SECURITY CLASSIFICATION OF THIS PAGE (When Date Entered;

REPORT DOCUMENTATION PAGE READ INSTRUCTIONSBEFORE COMPLETING FORM

*. REPORT NUMBER 2. GOVT ACCESSION NO. 3. RECIPIENT'S CATALOG NUMBER

MTL TR 87-50

4. TITLE (and Subtitle) 5. TYPE OF REPORT & PERIOD COVERED

HYDROGEN INDUCED CRACKS IN ARMOR STEEL Final ReportWELDMENTS FABRICATED WITH AUSTENITICFILLER ALLOYS 6. PERFORMING ORG. REPORT NUMBER

7. AUTHOR•() 8. CONTRACT OR GRANT NUMBER(s)

William S. Ricci

9. PERFORMING ORGANIZATION NAME AND ADDRESS 10. PROGRAM ELEMENT, PROJECT, TASKAREA & WORK UNIT NUM6ERS

U.S. Army Materials Technology Laboratory D/A Project: IL263102D077Watertown, Massachusetts 02172-0001 Agency --Accession DA 30 3398.SLCMT-MCD

II. CONTROLLING OFFICE NAAME AND ADDRESS 12. REPORT DATE

Ui.S. Army Laboratory Command September 1987'2800-Powder Mill Road 13. NUMBER OF PAGES

Adelphl, Maryland 20783-1145 i L14.-MON TORINQP AGFENCY NAME &ADORE;SSfi dillweent /trnm Contro~lling Offir') IS. SECURITY CLASS. (of thie -repoto

Unclassified-'S. DECLASSIFICA TION/ODWNOGRA6iXi

SCHEDULE

'-q-O7STRI8-UTIOk STATEMENT lt thlisl Report)

Approved for public release-istr•iInfoI- itcd.

17. DISTAINUTION STATE NT (at the abaltkE Ofltotd In Dhvek 20, it difl,#lefont IMepo@W)

Welding I Cracking 4 '---

Armor lStainless steelIHlydrogen

20. M•SSYNACT (ValiI~w an tov0#00 Sid* 10 tw •011000Y *A idef00titV by bluth• ,MA040q

(SEE REVERSE SIDEl)

DDiA. vi 1413 E.,1,ON OF NO 0o*, ISOBOLE:TE UNCL,'ISSII:IED

UNCLASSIFIEDSECuRiTY CLASSiFtCATION OF THIs PAGE tWhen DAa fnlerd)

BlotNo. 20

ABSTRACT

Hydrogen induced cracks have been identified in the weld metal of castarmor steel weldments fabricated with austenitic electrodes. Cracks propagatedthrough a well defined band of martensite located between the fusion boundaryand the composite region of the weld metal. Martensite band formation wasfound to result from the combined effects of vigorous hydrodynamic forces,especially in large molten weld pools, and the interdiffusion of alloying ele-ments in that system formed between the base metal and the weld metal. Methodsof controlling this form of cracking include limiting base metal dilution,minimizing the concentration of available hydrogen in the welding arc atmos-phere, or the selection of more suitable ferritic filler alloys.

Acoession For

NTIS GRA&IDTIC TABUnannounced "1Justitfoation

.DiStribution/ '

Aval1ability Codes3Ava±1and/orS...:°Dist Special

.

I F I t' A,!

UkL•S IF) l

BACKGROUND-

The work reported here was performed as part of a failure analysis of M60 tanktow lug weldments and was sponsored by the Program Manager-M60 tanks.

INTRODUCTION

Austenitic filler materials are often used to join high hardenability armorsteels. The main reason for their selection over ferritic filler alloys has tradi-tionally been that they provide a deterrent to hydrogen induced cracking (HIC) in rapplications where preheat or post weld heat treatment would be impractical or costprohibitive or when excessive arc atmosphere contamination would be unavoidable.Examples of such applications include the field repair or appurtenance welding ofthick walled ballistic structures. I

Austenitic weld metals have a higher solubility for hydrogen than ferritic weldmetals or heat affected zones (HAZs). Hydrogen in crack sensitive martensitic HAZ8is, therefore, able to escape at a faster rate than it is replenished by hydrogenfrom austenitic weld deposits. This results in a lower hydrogen concentration inmartensitic HAZs, and this reduces the tendency to initiate IIIC.

The lower yield strengths of austenitic filler alloys is also believed tocontribute to their reduced tendency to form hydrogen induced cracks. This ismainly due to the relief of HA jresidual stresses that result from localized plasticdeformation in the weld metal.*

Typically, cracks in armor weldments fabricated with austenitic electrodes arefound in the weld metal oriented transverse to the weld axis. Similarly, hydrogeninduced cracks in armor weldments fabricated with ferritic filler materials aretypically found in the HAZ parallel to the weld axis, Figure 1. Crackq in armorweldments fabricated with austenitic electrodes, located in the weld metal butoriented parallel to the weld axis and away from the weld centerlinep have not,"however, been previously documented.

The occurrence of hydrogen induced cracks in the intermediate zone of armorsteel weldments fabricated with austenitic electrodes will be discussed in thisreport. The term "intermediate zone" refers to parts of the composite region of theweld as well as the partially melted zone, the unmixed zone, and the weld interface,as defined by Baeslack et al.3 (Figure 2).

EXPERIMENTAL

Shielded metal arc welded appurtenances, fabricated with type 307 austentticstainless steal electrodes, were removed from H60 tank hulls. The chemical compost-tions of the weld metal and base metal are shown in Table 1. The composition of thebase metal was consistent with that of cast armor steel (XL&L-A-11356).

1. CASTrM W. R,, tuJ do CAI)ENIKT, Li. k~d~iig 40, k •Jlr uItIJy ,Jt' t Iit fttol• if14t sleeh Camttbtlijge uItveltt Pri, G( ,,i0

11.-1 '44A. .. 1"0 X.W-SOA X /~-e Whwk W~iks.4WSi rd"i 04t

Table 1. CHEMICAL COMPOSITIONS OF WELD METAL AND BASE METAL

C Mn Si Cr Ni Mo S P

Weld Metal 0.1 3.9 0.4 20.0 9.5 0.9 0.03 0.026

Base Metal 0.25 1.16 0.53 1.02 1.01 0.42 0.019 0.016

All ,welds were inspected for cracks by visual and liquid penetrant techniques.Cracks were sectioned and prepared for metallographic examination. A 50% glycerine,30% hydrochloric acid, 20% nitric acid etchant was used to define both weld metaland HAZ microstructures. Hardness traverses were taken from polished and etchedspecimens. Fracture surfaces were examined in an SEM.

RESULTS

Metallography revealed cracks located in the weld metal adjacent to the fusionboundary on both sides of welded joints. Figure 3 shows one such crack. All crackspropagated through a continuous band of martensite at or adjacent to the fusionboundary (Figures 4 and 5). This martensite band ranged in width from 10-140 Um,and in some locations was completely divorced from the fusion boundary by a thinlayer of austenitic weld metal (Figure 5). No cracks were found in areas where thisband was discontinuous or nonexistent (Figures 6 and 7).

A hardness traverse was taken across weld fusion boundaries. Knoop hardnessdata were taken through the weld metal, the fusion boundary, the HAZ, and into theunaffected base metal. This data is presented in Figure 8. The martensitic phasethrough which the cracks propagated had a hardness of approximately 420 KHN, which

-is significantly higher than that of the weld metal (225 KHN), HAZ (320 KilN) andunaffected base metal (280 KIlN).

SEM examination of the fractured surface revealed a mixed mode, intergranular-traiisgranular fracture path. This is characteristic of hydrogen induced cracks thatpropagate at low stress intensity levels.

DISCUSSION

The hydrogen induced cracks observed resulted from the presence of a suscep-tible microstructure (i.e., martensite) in a "hydrogen charged reservoir" (austen-ite). This cracking phenomenon has not been widely documented in the open litera-ture.

Several Invatttgators 4 ' 5 have studied the formation of martenstte at the weldfusion boundary. This martensite formation is believed to be due to the interdif-fusion of alloying elementa in that system between the base metal and the weldmetal. Xinetics alone, however, fall short of explaining martensite formations in

•bands adjacent to, but divorced from, the fusion boundary (Figure 5). Figure 9shows an example of a martensite band extending a relatively large distance into theweld metal along the fusioni boundary of successive weld passes. A more complete

4. O9NA 1$•. •U NiIVi. J,. Wi4'l. I1.1, Andi MINkOIF 1. fi' twu A101#4 1A. w'digg if ILO Agtb Sikiy 101h A Uwair" Aleettidr&~l. W¢ediii humlmtl. v, 0fl. oo(. 11. Ntlivewllsr 1981I

5, LANCASUA4.i1K J.1Q M1J94Nr Of W4•iJq ' Cl)4&a" Mk#1 AIi*•U i lt~til U. I!.4tw:

2

explanation for the observed martensite band formation is that, in addition to*kinetics, hydrodynamic forces play a major role. These forces are sufficient,especially in large molten weld pools, to physically wash away part of the fusionzone and introduce it into the composite region of the weld metal.

The Schaeffler diagram (Figure 10) can provide a first approximation of theeffect of base metal dilution on martensite formation. For the alloy systems ofinterest here, only 17% dilution of the filler alloy with the base metal will resultin the formation of martensite. It should be recognized that 40% dilution is oftenencountered under ordinary welding conditions, especially in the root pass. Onealternative in resolving this problem, therefore, would be the use of higher alloyfiller metals such as 309, 310 or 312 that can tolerate more dilution because oftheir more stable duplex austenite-ferrite structure. For example, 312 weld metalscan tolerate approximately 45% dilution without forming martensite. The use ofthese alloys in other than high dilution passes (such as the root and butteringpasses) is not recommended, however, since high ferrite contents reduce the ductil-ity of the duplex structure.

The elimination of martensite is not the only method of deterring HIC in aus-tenitic weld deposits. It is often more economical to reduce HIC susceptibility byreducing the available hydrogen concentration in the welding arc. This reductioncan usually be attained by using electrodes with low moisture contents, and isgenerally considered to be the preferred alternativo.

CONCLUSIONS

1.. Hydrogen induced cracks have been identified in armor steel weldments fabricatedwith austenitic electrodes.

2. Hydrogen induced cracks may form in a well defined band of martensite locatedS between the fusion boundary and the composite region of the weld metal.

S3*.- Iartensite band formation is-due to the combined effects of vigorous hydro-dynamic forces, especially in large molten weld pools, and the interdiffusion ofalloying, elements in that system formed between the base metal and the weldmetal*.

4# .Cracking can be effectively controlled by limiting base metal dilution andminimtaing the concentration of available hydrogen in the welding are.

• 5, The use of higher alloy filler A l.vs recommended for critical applica-tO4os, especially in the root pass.

* 6. Nore economical ferritic filler alloys should be used whenever low hydrogencondiLtious ca" be properly maintained.

3

METAL

figure 1. Hydrogen Induced crack located In fte HAZ of anarmor steel weldrnant 4150X)

PlART IALLY

F~~~~~~~~ue ~ ~ ~ ~ UE 2.SZL41lut~ntohgth es~so ehwehu El

Fiur Crack In weld mali) adjacent to fusion boundary (37.5X).

Piguee 4. Crack pwp"atflng through a conlinuous band of marttwisitein weld metal along fusion bounidasy -(375X). Ctack deviation was duo:to multipas weldng conditions.

5

II

.. .- -N f. .

" . j. .9z• "

SFigure 5. Cra& in mariensite band sep~rated from fusion boundaryS. .... .. .by a thn laver of austenite (375X).

"t. . ."-

W" .

Figure 6. Crackonmrtinuous bando sprted fomt fusioent boudsybyustion layerof uun (375KM .

- -: .. " " :. • .. -"* t . .• . . ,"-. .- . .-- " " - . . .-'' ..' . • ' ' • "

9... W "' . . . -

- I

.: - . . .. . .. ' ' . .• " ; . . - . . . " . . -- , ' . . . . • - . .

t• 1'

I • " : ~~~a"••••"::' :: "•. •. ":,

i :,fuson boundwv (376K).

• 6

* w.

ISO- = *

*~~ A*~ .*

W" UT-

Nut -4--. ti ~tawos hwl fso bud

fture 9. Manrmnste hand extending along fusion boudauy ofsuccusaiv weld pistm. 3137.X).

44

3 4240

161

U1 2

211

o26

-24

'~20

It, attsustt.'too

* A 6

02 6 at 10 14 1 to 7O? 04 1 is 0 4746Eta1ramium .4uialvaor (%Icr I %n ak IA u N b

""ig=e M0. S" ulcastl d4aga lot "UsstienaOW "e n ctuu e row. "ness stat wow owli.

• • •. " " -.•. .

DISTRIBUTION LIST

No. ofCopies To

1 Office of the Under Secretary of Defense for Research and Engineering,The Pentagon, Washington, PC 20301

Commander, U.S. Amy Laboratory Command, 2800 Powder Mill Road, Adeiphi, M4D20783-1197

1 ATTN: Technical Library

Commander, Defense Technical Information Center, Cameron Stai ion, Building 5,5010 Duke Street, Alexandria, VA 22304-6145

2 ATTN: OTIC-FOAC

Metals and Ceramics Information Center, Battelle Columbus Laboratories,50.5 King Avenue, Columbus, OH 43201

1 ATTN: Mr. Robert J. Fiorentino, Program M~inager

1 Defense Advanced Research Projects Agency, Defense Sciences Office/MW0,1400 Wilson Boulevard, Arlington, VA 22209

Headquarters, Department of the Amy, Washington, DC 203141 ATTN: DAEN-RDM, Mr. J. J. Mealy

Commander, U.S. Air Force Wright Aeronautical Laboratories, Wright-PattersonAir Force Base, 014 45433

1 ATTN: AFWAL/MLCIAFWAL/MLLP, 0. H4. Forney, Jr.1AFWALI/MLIIC, Ur. Stanley Schulman

1 AFWAL/MLLS, Dr. Terence M, F. Ronald1 AFWAL/FIBEC, Dr. Steve Johnson

1 Edward J. Morrissey, AFWAL/MTC, Wright-Patterson Air Force Base, Olt 45433

Cocmander, Army Research Off ice, P.O. Box 12211, Research Triangle Park,NC 277094211

1 AMT: Information Processing Office.IDr. George Mayor

C0twandr, U.S. Army Ptiteriel Coanod, 6001. Eisenhower Avenue, Alex1andria,VA ?2331

1 ATIN-: AMCLO

COw~ander, U.S. Any Anm~went, Ianitlon &nd CheicalI COMuaat,Doviee, NJ 07841

I1 ATIN: W lt Hrry E. I'ibly. Jr.,,PLASIE, 0irectore

Coiwanr, U.S. Any Aviation Syltten t6*Wam , 4300 Good NlIoI DlvI., St.Lul,"D4 03120

1 AY10.1 AiDAV-hS, 14soud L41W

Diretor U.S Ary Balliitic Rowa h Libonatoey, Abou t P1vt~ rsw

I1 ATTN,, AMAisOss (St14r0)

1 41111; SIM.E

6 ~a 40r' U.S. Arty Vrcwign scikeow To lchndlfgy Cnter, No. 7th SotretN t, C lrtw.wiileo, VA I290

A 1 N4Ii: MIlitary tech

Corsu&Isu U.S. A"y Matrici $sItiM 5&lysis Attivity,4&490d~ hrdvlng &bund RI) a100

I. ATTN: Aflt514t', i.C44*0

Cboaeandtr, U.%,. Army Missile Cotwuand, kedstw $C104111( ltafo!Uitioit C tr,14dsti~tAr~n1.AL 35404.5241

I 1 I; ffil: IW-t-/L Owe LitI 411 -9t"1kIAKh#41-ULA. Dr. J"~ . icado

No. ofCopies To

Commander, U.S. Army Belvoir Research, Development and Engineering Center,Fort Belvoir, VA 22060-5606

1 ATTN: STRBE-DI STRBE-G1 STRBE-N1 STRBE-VL

Commander, U.S. Army Aviation Applied Technology Directorate, Aviation Researchand Technology Activity (AVSCOM), Fort Eustis, VA 23604-5577

1 ATTN: SAVRT-TY-ATP, Mr. James Gomez, Aerospace Engineer

Commander, U.S. Army Tank-Automotive Command, Warren, MI 480901 ATTN: AMSTA-RCKM

Director, Benet Weapons Laboratory, LCWSL, USA AMCOM, Watervliet, NY 121891 ATTN: AMSMC-LCB-TLI AMSMC-LCB-PS, Dr. I. Ahmad

David Taylor Naval Ship Research and Development Center, Annapolis, MD 214021 ATTN: Dr. Michael Vassilaros - Code 2814

Office of Naval Technology, 800 N. Quincy Street, Arlington, VA 200171 ATTN: Mr. J. J. Kelly - Code MAT 0715

Naval Research Laboratory, Washington, DC 203751 ATTN: Code 58301 Dr. G. R. Yoder - Code 6384I- Dr. S. C. Sanday - Code 6370

Chief of Naval Research, Arlington, VA, 222171 ATTN: Code 4711 Dr. Steven G. Fishman

Naval Sea Systems Command, Washington, DC 203621 ATTN: Mr. Marlin Kinna - 62R4

Naval Air Development Center, Warminster, PA 189741 ATTN: Dr. E. U. Lee - Code 60632

Naval Surface Weapons Center, White Oak, Silver Spring, MD 209101 ATTN: John V. Foltz - Code R321 Dr. Herbert Newborn - Code R34

National Aeronautics and Space Administration, Washington, DC ?05461 ATTN: Mr. Michael A. Greenfield, Program Manager for Materials, Code RTM-6

Cleveland, OH 44135

1 ATTN: Dr. James A. DiCarlo, Mail Stop 106-1

National Aeronautics and Space Administration, Marslhall Space Flight Center,Huntsville, AL 35812

1 ATTN: R. J. Schwinghammer, EHO0, Dir, M&P Lab1 Mr. W. A. Wilson, EH41, Bldg. 4612

"The Boeing Vertol Company, P.O. Box 16858, Philadelphia, PA 19142I ATTN: Mr. Robert L. Pinckney, Mail Stop P62-061 Mr. Joseph W. Lenski, Jr., Mail Stop P32-09

E. I. DuPont De Nemours and Company, Inc., Textile Fibers Department,Pioneering Research Laboratory, Experimental Station, Wilmington, DE 19898

I ATTN: Blake R. Bichlmeir1 Joyce W. Widrig

1 Mr. Rex C. Claridge, TRW, Incorporated, Manufacturing Division,Mail S~op 01-2210, 1 Space Park, Redondo Beach, CA 90270

I Dr. James A. Cornie, Materials Processing Center, Bldg. 8, Room 237,Massachusetts Institute of Technology, 17 Massachusetts Avenue,Cambridge, MA 01239

No. ofCopies To

1 Dr. Bhagwam K. Das, Engineering Technology Supervisor, The Boeing Company,P.O. Box 3999, Seattle, WA 98124

1 Leroy Davis, NETCO, 2225 East 28th Street, Building 5, Long Beach, CA 90806

I Mr. Joseph F. Dolowy, Jr., President, DWA Composite Specialties, Inc.,21133 Superior Street, Chatsworth, CA 91311

1 Mr. Robert E. Fisher President, AMERCOM, Inc., 8948 Fullbright Avenue,Chatsworth, CA 91311

1 Mr. Louis A. Gonzalez, Kaman Tempo, 816 State Street, Santa Barbara, CA 93101

1 Prof. James G. Goree, Dept. of Mechanical Engineering, Clemson University,Clemson, SC 29631

I William F. Grant, AVCO Specialty Materials Division, 2 Industrial Avenue,Lowell, HA 01851

1 Mr. Jacob Gubbay, Charles Stark Draper Laboratories, 555 Technology Square,Mail Station 27, Cambridge, MA 02139

I Mr. John E. Hack, Southwest Research Institute, 6220 Culebra Road,San Antonio, TX 78284

I Dr. H. A. Katzran, The Aerospace Corporation, P.Q. Box 9?957Los Angeles, CA 90009

Lockheed California Company, Burbank, CA 915201 ATTN: Mr. Rod F. Simenz, Department of Materials and Processes

Lockheed Georgia Company, 86 South Cobb Drive, Marietta, GA 300631 ATTN: Materials and Processes Engineering Department1 Mr. James Carroll

Material Concepts, Inc., 2747 Harrison Road, Columbus, OH 43204I ATTN: Mr. Stan J. Paprocki1 Mr. David Goddard

1 Dr. Mohan S. Misra, Martin Marietta Aerospace, P.O. Box 179, Denver, CO 80201

I Mr. Patrick J. Moore, Staff Enoinoer, Lockheed Missiles and Space Company,Organization 62-60, Building 104, P.O. Box 504, Sunnyvale, CA 94086

I R. Byron Pipes, Profersor & Director, Center for Composite Materials,Untveqrity of Delaware, Newark, DE 19711

1 Dr. Karl M. Prewo, Principal Scientist, United Technologies Research Center,Mail Stop 24, East Hartford, CT 06108

I Or. B. W. Rosen, Materials Sciences Corporation, Gwynedd Plaza 11, OethlehemPike, Spring House, PA 194?7

I Prof. Marc 1. Richman, Division of Engineering, Brown University,Providence, Ri 02912

1 Hr. Ronald P. Tye, Eber y Iateials Testing Laboratory, Biddeford IndustrialPark, Biddeford, ME 04805

1.. Mr. Robert C. Van Siclen, Voughit Corporation, Advanced Technology Center,P.O. Box 226144, Dallas, TX 76266

I1 Prof. Franklin C. Wialwer, Ltpart tint of Mateials Svivence, Sohool of Eqietnerifigand Applied Sciences, University of Virginia, Charlottesville, VA 2290i

I Dr. Cail' Zwebel, Genoral Electric company, Vaalley Forgo Space Centoe/N4O11,P.O.Box 856' Philadelphia, PA 19101

Director, US. Artay Materials, Technology Laboratory. Watertown, KA 021O2-O0I2 ATTN: SLCMT-IML"1 Author •

1-4]oa Z 0 =1 v E

L E 10 dj L. (U

-j -o W L uw- 0 0) =

V 0 . ' V 0.-

-~~~~> 4- 4Jibdi ) U 450

"- 4JCo M Ua 414 C,

M. 0. 0( C4J.36. m

Ii .t o. w 0aZ c 6J -4J O. ' -4 C

.~flu4 41; I mo t*ru

m InW-..4d C: I-

a. O 0414-o 0.2 a wi.0 c .. 0...ýuWcL-001 t C 10. t.i 2, 1- -0 A -riI;a -aju a ;a *oaouS

oj r, cu= *'. 0CC o4 w (c-4 a

0 w00 (n di Q-di 4)

nI CO d 3 41ZfS a Od T3 430 v InnSeN

C2 -j ti ,2C di C44W - 0. -, Ii. di C0i Lt-O 9iW'U. -4 $-

W~~ ~ ~ ~ 10%A9 l.VI'.5m

C.- c~' ap -- oo-o;~ eO x .0 10 01 0- -Al u~ r i0 an-mmrn'ai - "b~~C 714,~ 2 wj tj a- '10-- - .. *

2, WU m.-. -VU L I . L.S I ICU 11>

.0 ,. ;0) u0 "i - .- Ma=

c). 130 =. " .- - 0C - t o~L .JJ0

j-e 5c 1' 3u

14 r 1 -. 41 v 1 wý (0 wlv 0

-- j !M 11) 1

-A W.- 6 r' E

- 9Ct.A 'a- V. 16v ,

'01O-05 0-

t! 1

3 ~1. h

10- 441'i

60 0, -

~ !~ I

1j v 1

~ W I10,4 r-S 14

;2Th~ 3 iiL

C AA ~ 4 týn .A

'A ~A '-A

L -j - -- -