SSC-108

NOTCH-TOUGHNESS PROPERTIES OF SHIP-PLATE STEEL AS EVALUATED

BY THE VAN DER VEEN NOTCHED SLOW-BEND TEST

by

E. A, Imbembo

and

F. Ginsberg

SHIP STRUCTURE COMMITTEE

-—.. -.

SHIP STRUCTURE COMMITTEE

MEMBER AG~~=:ADDRESS CORRJ=PONDWCE TO:

SKCRCTARV

BURLAU SF Swl-. D-T. o’ ‘“w

HI LITARV Sti TRANSPORTATION SE RVICC. DI-. Or MAW

SHIP STRUCTURE COMMInC~

UN,tEn STATES COAST GUARD. TREA8Uny DCPT.

U. S. COAST I=uARD H~ABQWAuTms

WASHINaTON 2ti. n. c.

MARITIMI ADMINISTRATION. ~l?FT. or co~McRc~

AMERICAN BuR*AU OF sHl?~lN@

August 31, 1959

Dear Sir:

Herewith is the Second Progress Report, SSC- 108, entitled“Notch-Toughness properties of Ship Plate Steel as Evaluated by thevan der Veen Notched Slow-Bend Test” by E. A. Imbembo and F.Ginsberg, of Project SR-141, “ Semiskilled Steels over One Inch. ”

This portion of the project is being conducted at the MaterialLaboratory of the New York Naval Shipyard under the advisory guidanceof the Committee on Ship Steel of the National Academy of Sciences-National Research Council.

This repoti is being distributed to individuals and groupsassociated with or intere steal in the work of the Ship Structure Com-mittee. Please submit any comments that you may have to the Sec-retary, Ship Structure committee”

Sincerely yours,

E. H. ThieleRear Admiral, U. S. coast GuardChairman, Ship Structure Committee

—

—

—-

-.

,-

_..-

.

Serial No. SSC- 108-.

I ,.’

Second ,Progress. Reportof

Project SR-141

to the‘—

SHIP ‘STRUCTURE C@MMITTEE

on

NOTCH-TOUGHNESS PROPERTIES OF SHIP-PIJYTE STEEL AS EVALUATED BY THEVAN DER VEEN NOTCHED SLOW-BEND TEST

-. .

by

E. A. Imbembo and. l?. Ginsberg,—,

1’

Material LaboratoryNew York Naval Shipyard

,. .

under—

—.

.—

I !.

Department of the NavyIndex No. NS-011-078

transmitted through

Committee on. Ship SteelDivision of Engineering and Industrial Research

National Academy of Sciences-National Research Council,-—

under

—

...

Department of the NavyBureau of Ships. Contract NQbs- 72046

BuShips Index No. NS-731-036

Washington, D. C.National Academy of Sciences-National- Research. Council

Aiigust: 31, .1959

,—

-.

—.

ABSTRWT

This report describes the second stage of research work

on a program designed to study the notch-tcmghness properties

of semiskilled ship-steel plate over one inch in thickness. ThIS

details set forth here are the results of tests ~on the fracture

appearance and duct il ity trans i,t~m.temperatures of sarnp les of

3/4 in., 1 1/4 in. and 1 3/4 in. thick plate,. ~epre sent ing two

heats of a low-carbon, high-manganese, semiskilled experiment-

al steel, as determined by means of the not &hed slow- bend

test devel~ped by J. H. van der Veen.

observations made during this irwest@ation seem to

indicate that the van der Veen tiest provides arL e st hnate ~f the

fracture transition temperature similar to that’ obtained in the

Navy tear test. However, more work remains ~to be done with

regard to obtaining add it icmal. experience with the method and

establishing its correlaticm.with other techniques used f or eval-

—

/—

uatirq notch-sensitivity characteristics of ship~plate steels.

-—

.—

,—

.

l—

—.

TABLE OF CON.TENTS

WE

Introduction .............~ma=o=~ .00~~o

VmderVeenT estMethodoaoo . . . . . . . . . . ...*

ExperimentalE quipment . . . . . . . . . . . . . . . . ..-

Procedure . . . . . . . . . . . . . . . . . . . . .= OOOO

Results . . . . . . . .. O. OO..O.. .o .0000*

Conclusions ..sao.000oo e*o.**.*oa.0*0.

References . . .. OO. .OO. OO. OO... .OOOOQQ

1

2

8

10

20

21

21

,—

.—

-.

—

—.

.

.

---

—

--

INTRODUCTION

In BIovember, 1955, the MateriaI LtioratoW, of the New York Naval SMP-

yard commenced evaluating the notch-toughness., properties of samples of experi-

mental ship-plate steel by means of the 7L7an der ‘Veen notched slow-bend test.

This work. is a phase of a comprehensive program, (W-1419 which was cdqhwilly

set up by the .Shdp Structure Committee to investigate properties of sernikill~ d

steel plate in thicknesses cwer 1 in. that ccm%a,in less carbon and. more rnanya-

,nese than (21ass-B steel as specified in the 1955ABS Rules for plate ranging in

t;hickr~ess from over 1,/2 in. to 1 in. inclusive. It was felt tkt a sernikilkd,

km-t-rdkd. steel conta inhw appmximatejy= O. 1670 mdm and I. 0070 mwame se

mig’ht serve as a possible emerqency substitute for fully kiDed CJass-C steel,

currently specified by ABS Rules for ship plate over 1 in. in thickness. The pro-

gram has been subsequently expanded to include a determination of the ductillby

and fracture- appearance transit ion temperatures of samples of steel plate by

means of the van der Veen notched slow-bend, test--an experimental technique

developed in Europe by J. H. van d~r Veen..1

Its application by the Material

hthrat~~y ~epre ~~~ts the od~ k~.o~~ e~do~atio~ of the method in this couritw O

Under arrangements made by the Ship Structure Committ.am, samples of

plate rolled from. two experimental heats of the proposed steel produced by the

U. S. Steel Corporation were distributed to a, number of Iaboratories for a cmn-

prehensive investigation of the ch.ara cteristics of the material, with part~.cular

emphasis on notch,-toughn,ess properties as evaluated by various experimental

techniques. The se included. Gharpy drop-weight, explosion crack-starter,

R@wtsonj Es so brittle temperature, and Navy tear tests. The groups patiici-

patirq in the program include the American Bureau of Shipping, Naval Research

Laboratory, Admiralty Naval Construction Research Establishment, Esso Research

and .Em@.eering C ompany, Material Laboratory an,d the Unit~d. States Steel C orpo-

ration. The project is of interest

fact Ihat hull plate, 7/8 in. th,ick

cation, MIL-S-1611 3B to be of the

be in the normalized conditiofi.

to the Department of the Navy in view of the

and over, is required under Military Spec ifi-

same type as- ABS Class-G and, in additiori, to

-2-

The purpose of this -report is to describe the techniqup, equipment, and

procedure used by the Material. Laboratory in the van der Veen test. These data, -..

together with those obtained by the various cooperating laboratories ,Qnthe same

steel, will be analyzed and correlated by R. W, Vanderbec k of the ‘United States Stee 1, -.

Corporation; a first progress report on this analysis and correlation appeared in

August, 1956.2 An appraisaJ, of the method as a means of determining the tend-3, 4e.ncy of mild steel to brittle fracture may be found d s.ewhere.

VAN DER mEN TEST METHOD

Description: Specimens of the type shown in Fig. 1 are herd at various

temperatures in a testing machine at a loading rate of 20 mm (O. 8 in. ) per rrdn.

The longitudinal axis of the specimen is taken perpendicular to the direction of

rolling. In the rriiddl,e of one of. the two machined surfaces, a sharp 3-mm deep

notch with au included. angle of 45” is made with a hardened-steel knife, pro-

ducing a radius in the order of 0.004 mm ((0. 00016 in.) at the root af the notch.

After notching, the specimen is brought to the desired test temperature in a

liquid. bath, then removed from the conditioning ba,th and te steal. to failure with-

in, a minute. It is important that the testing be completed within 2 hr. of notcb-4ing to minimize. aging effects.

Heat. exchange with the surroundings is cons i.dered, negligible; however,

the temperature does rise as a result of deformation energy becoming in part

frictional heat, which may influence the transition temperature. It is stated

that the test. thus acquires an adiabatic character. In this conriection,, it has

been stated: “Whether service conditions are approximated more closely by an

adiabatic than by an isothermal test is still an open question. In a,ny case,

the isothermal test is more difficult to perform, nat only because during the

test. the specimen must be present in the cooling bath, but because the test

would have to be carried out very slowly and with interruptions to permit dissi-

pation of the beat locally generated. ” 5

During the test, the changes in load as a function of the deflection of

the specimen at its center are record,ed automatically to produce a complete

—

—

.

-.

—-3-

Lo&d

I—

—w

—

.

.-—

—

.

S~m orFlame Cut Edge

\

SuPPort Roll

?Diameter 40 ~(Approx. 1 5/8”)

(’n-ad-’o”Diameter 60 mm (2 3/8m Approx.

.Machined Edge

Ir

~ Span 200nmzj(7 7/8”)

# 225 mm(Approx. 9“) 4

1

FOriginal?lato Surfam

-J.-

Fig. 1. Van der Veen Notched Slow-Bend Test Specimen.

ryp-Load -De flteiion DiagramFracture App.

Lm .——

a

,,L~ --–

Cw.1

b Fbr.

a

CrJ5L

c

Rbr -

d

a

Iam

Fig. 2. Schematic Representationof Various Shapes of Load–De-flection Diagrams C)btained inthe van der Veen Notched Slow-Bend Test. 3

Fig. 3. Schematic Representation Show-ing Methods of Defining Transition Temp–eratures in the van der Veen NotchedSlow-Bend Test.

hid-deflection

-4-

diagram. As will be. shown later, it is not essential to ?htain

such diagrams since the test result~ may be Qvaluated by other means.

The various forms that may be as surned by the load-deflection curve

are shown sc bematically in Fig. 2. Type “ a“ represents the behavior of. a

specimen failing entirely by shear, the fractme aPPear~nce being tota~~y fi-

brous. Types “b, “ “c, “ “d, “ and “e“ show the behavior of the specimen un-

der the conditions that tend to promote increased brittlene SS. such as would

be obtained by lowering the testing temperature. After the start of cleavage,

a measurable (though small.) amount of work may be -absorbed in. some cases,

despite the fact that the fracture path is almost fully crystalline. In these

cases, the fracture surface usually shows about 1- or Z-mm ‘wide fibrous. edges

bordering on the plate surfaces. h addition? these fibrous edges may occa-

sionally widen and join to form a Go-called thumbnail of parabolic shape. At

time S, this thumbnail condition may lead .to a sequence of alternate fibrous

and crystalline areas in the fracture. In all of these instances, the energy

included under Area IIIis ~greater than zero and t’he load–deflection diagrams

occasionally show a “taiIY “ which is indicated,. by the dashed lines in Fig. 2.

Two transition temperatures are usuaHy determined by testing a numb-

er of specimens at each of a series of test temperatures 10 G (18 F) apart.

The se temperabm% am the fracture-appearance tran~ition temperature and the

ductility transition temperature,

1. The fracture-app—— ea,rance transition temperature is obtained by

mea sur~,n.gthe depth of fibrous area beneath the. notch and plotting the value

for each specimen tested as a function of the test temperature. Since in all

fractures of a mixed type the fibrous portion has a parabolic boundary, meas-

urement of fibrous areas can be avoided by simply taking a scale measurement3

of the distance between the notch root. and the vertex of the parabola. The

curve drawn through the average value of “mm fibrous” at each test temperature

is the transition curve. The fracture-appearance transition temperature is the

temperature at which this average curve intersects the 32-mm line, as iHus-

trated. schematically in graph C of Fig. 3. An equivalent criterion for determin-

TABLE I = IEENTIFICATIQN OF U o So STEEL @J3R:I WORKS] SAMPLE MATERIAL

ii

II

II

!!

II

10

30

40

5.

mat No o

Ladle Analyses, !10

a. E&bDnb. ManganeseC . Phosphorousd o Sumlre o silicon

Plate Thickness

Slab Code

Sample Sizea. _(ino)b. ~~in.)

Direction of Rollingin Sample fora. Tear Testb. W3v Test

ApproximateLocation ofSample withRespect toLength of(kiginal Platea. Tear Testb. VDV Test

2(2 .2850

00161.300.010(?.0.210008

314 in m 1 1/4 in. 1 3/4 in.

B c D

36 X 45 36 X 45 36 X 45:< $ :~

Parallel to 36 in. dimens bnParaIlel to 36 in. dimension

13cmf31n Bottom TopBottom Bottom Top

0.151.040.0130.0220.021

3/4 in o 1 1/4 in. 1 3/4 in. 1 3/4 in.

B G A D

18 x 45 18 X 45 18 x 45 18 X 4518 X 48 18 X 48 18 X 48 18 x48

ParalIel to 18 in. dimensionParallel to 48 in. dimension

Top Top Top BottomCenter Center Center Center

‘~ One sample piece for both tear and van der Veen tests.

-6-

i,nq this temperature uses Lc/L,m = O. 7, where ,Lc represents tlm load at. cleavage

and Lm is the maximum lQad. The fracture--appearance transition through use of

this alternate criterian is obtained from the average curve of Lc/Lm vs. test. tern-—

perature at a point where the curve intersects the O. 7 line of Lc\Lm, as illustrated

schematically in graph D of Fig. 3, The use of “mm fibrous)’ as a. criterion is pre-.

.,..-

.-

fecred by van der VeenL. because of somewhat greater accuracy in its measurement.

T“hi,stemperature is related to the tran sithm from type 11btl to “~” fractures (Fi9. 2)

dnd “therefore m,ay be consicbmd to be a measure of the tendency to start cleavage

from a f ilmous crack propagating in mabxi.al that has been sliq htly deformed in

‘tension. “1

2. The ductility -transition tempera,hme is related to the1-

type “d“ to “ e“ fractures (Fig. 2). Van der Veen has found

fracture, the depth of fibrous area under the notch amounts

tYP e “e, “ the crystalline fracture starts immediately at the

vic imi,ty of the “yield point” of the load,”’”deflection diagram

considerably less than those at which the fibrous incipient12

that

tQ 7

root

at a

—transition from

in, the type “d“

mm average. In

of the notch in the

load and deflection

crack w-odd otherwise

have started. On the basis of previous reports” J and information obtaimed in c or-

respondence with van der Veen, the duct. iJity tram ition Wrnpertature is taken .as the

temperature at which the value of Dun (deflection at maximum load) is about half-

way between the extremes of the average curve of Em vs. test temperature, as..—4

illustrated schematically im graph A of H g. 3. (A recent IIW report takes the duc-

tility transition temperature at a value of Elm +=6 mm). A simi~ar alternative pro-

cedure for determination of the ductility tra,nsitkm temperature is illustrated in

graph B of Fig. 3.

—..

In this case, the transition temperature is defined as that temp-

erature at which the maximum load is about half-way between the etiremes of the

average curve of Lm vs. test temperature. HoweveF, van der Veen 1 considers the—

ductility transition temperature using the criterion of Drn to be more convenient

and reliable.

Ship-Plate Material Under T%st; A defailed description of the ship-plate——

material under investigation is cmata ined in the First Progress. Report on this2

project . In brief, one 30-in. x 64-in. x 89-l/2-in. ingot was cast from each

of two heats. Four slabs, coded A, B, C, and. D, were produced from each ingot.

.-

— -. .-

--

.

—

—.

——

.,..-

-.

—

—

-_

,-

.

.-

-7-

Fig. 4. Jig for Performing vander Veen Notched Slow-Bend Test.

.,:,’’’,’,

Fig. 5. Rear View of Frame of Bend Jig. Fig. 6. Arrangement for CenteringLoading Roll.

— .—

-8-

Slab A was the top slab of the ingot. In the first kat (No, 2G 2850), the top

cut was lost because of pipe and, as a, result., slab A was not obtained. The

slabs were rolled to plate th.icknes ses of ~/’4 k., 1--1/4 in. and. 1- 3/4 in.

A description of the sample material is given in Table 1, which also

shows the size and relative location cd the aampk material subjected to .t&r

tests. It is to be noted that in the case of hQat No. 2G 2850, the tear and. van

der Veen tests were made on the same sample of plate, while in the case of heat

No. 2G 2883, tests were made on samples taken from different locations in the

original plate.

EQ’UIPNIENT

Figure 4 shows the jig that was constructed to perform the notched slow-

bend tests. The ~irnensions of the loadinq and suppmrt. rolls, as well as those

of the span between centers of the support rolls, were made to conform to the

dimensions given in Fig. 1. The jig was desiwned with adjustable stops (,shown

in Fig. 5) for centering specimens of different. plate thicknesses up to 2 in. in

thickness. A disadvad~ge of the test with. respect to the testing of mild steel3

p~a~e iS that it requires a minimum plate thickness of approximately l\Z in.

Thinmer specimens tend .to tilt and. become unstable when loaded, which unduly

affects. the test re s+u%tsar ma~yeven render p~riwnance of the t.e tst impossible.

Figure 6 indicates the .an’a.ngememtfor Iocatirig the l~adi~~.g roll midway

between the support rolls and, ~i~~ita~~~~,dLt?: for ~atera~ centering. The

centering template is also utilized in the inverted position to serve as the

hardened- steel knife holder. Figure 7 shows how th~ hardened-steel knife is

used to cold-pres.,s the 45° V-notch in the test specimen. The knife is made of

52100 steel, hardened to Rc 63/64, ground and .hwnd-honed to a sharp edge.

The edge of the knife W 1ocdted 3 mm above the top surface of the knife holder,

a close-up view of which is shown in F@. 8. tn applying the load for cold-

pressing;, t& loading. roll is removed to wwent deformatim of the sPecimen

edge opposite the notch. ~~~e~oa,d is app~ie~ until the bottom edge of the speci-

men contacts the upper surf ace of the knife holder (as evident ed by sudden in-

— —

. .-

—

.—

—

.—

,.-

-.

---

-9-

Fig. 7. Arrangement for Cold-Pressing 450 V-Notch in TestSpecimen with Hardened Steel Knife.

Fig. 8. Close-up View of Knife Holder and Centering Device.

-.

-1o-

crease in load), thus automatically pressing the notch to the required depth of

3 mm. A dose-up view of a spwimm in the p~mess Of beiw no~~~e~ ~S S~CIWD,,

in Fig. 9.

In order to avoid excessive change in temperature of a specimen d~.ring

the interval between removal .fro.m the conditioning bath. and application of load,

it is considered desirable to ‘have some means of quickly centering the s,pecimen

in the bending jig with subsequent immediate @.Pp~~c~t~Ld of load. Therefore, a

specimen, upon being removed from the cod W.onimg bath, is ba:r],ked aga,inst the-.

spec hen thickness stops [previously described) with the natc h sirm~.lta.neows.ly

centered midway between .t”he snppofi rolls by means of the index point showIm,in—

Fig. 10; the load is then immediately applied.,—.

,-.

PROCEDURE

For each sample plate, three specimens were tested at each of sw~en or

eight temperatures, selected so as to provide the necessary data for develop.merd —.

of the transition curves. The longer dhnension of the specimen was taken perpen-

dicular to the direction of rolling.

Eome preliminary experiments were ~cmducted, to determine whether it was

necessary to machine the lower edges of the. specimen. It kvas found that a care- ---

f-ally made saw-c~~t was adequate. The shorter edges were flame-cut. In th~~

connection, van der Veen recommends that when the longer edges have been flame-

cu.t, the width of the specimen blank should be 100

ante of 15 mm on each side for subsequent removal

dimension of 70 mm.

mm so as. to prcwi.de an aXow-

by machining to the finished .

The rnechmkm of cold-pressing the notch has been previously de scdbed.

NE rwce ssary Io.ads to fully form the notch to the. required depth in the 3\4--in. ,

1-1/4 in. and 1-3/4 in. thick plates were in theranges of 20, 000--25, 000, —

30, ooo-- 35, 000, and 40, 000 to. 50, 000 k, r~spectivel~y. As a precaution against

using a dull knife edge, no more than lo Pre SSiWS we~e made on ‘~”~~e~a~e area of -.

the knife hef ore the knife was band-honed for subsequent pres sMgs. Microscopic

exam, imtion at 500 diam. of several notches showed no measurable radius at the .

-11-

—.

.-

—..

—.

.—

.-

---

—

-—

.-

..-

Fig. 9. Close-up View Showing Cold-Pressing of V-Notch.

Fig. 10. Arrangement for Centering Notched Specimen for Loading.

_12_

—

root. The literature available on the van der Veen test does not coiitain direct

references on the effects of the cold-pressed. notch, cm best results. However>

it imjica,tes two advantages for the cold-pressed notch, namely~ 1)! such a

notch simulates. as. nearly as pos tsible the action of a natural fracture and Z)

the transition, temperature determined by this notch. is relatively irndepen,dent of3

notch root. radius.

Specimens of the 3/4-in. and, 1- l~+in. plates were tested in full plate

thiclmess in a. 120, 00 O-1b capacity hydraulic testing, machine equipped ‘with a

pacing device for controlling the speed of the loading ram. A microformed-type

deflectometer and. automatic recorder were employed .tn chtaln the load-deflection

diagrams o The deflection under load was not mea sured directly beneath the

specimen because of possible. damage to the deflec-knneter, panticularl,y in the

,—

.—

case of brittle failures; rather the deflection was measured as the change in

distance between the laading he ads., which is considered substantially equivalent

to measurement of specimen deflection. This is. especially true in vie-w of the rela-

tively large magnitude of this quantity...

Because ,of insufficient load capacity of

the testing machine described above, it was, necessary to test specimens of the

1-3/4- in,. thick plate in a 600, 00 O-lb hydraulic testing rrm,chlme. Howeverl this

machine is not equipped with apparatus for atitcnnatic recording of the load-

deflection diagram. As..a re suit, the evaluation of the data was ba sed on tlm

criteria of “mm fibrou S, beneath notch” and maximum load. Xn all, cases, the cross-

head speed was, O. 5 in. /mig, since it Twas found difficult to control the % .st at a—.

sped of O. 8 in. /rein a,s used by vam der Veen. ..-For test temperatures below atmospheric temperature, the specimens were

conditioned in an alcohol bat’h contained in an insulated. tank and refrigerated by

direct addition of dry ice. For test temperature’s above atmospheric temperature,

a water bath heated with. an electric i.mmers ion heater was employed. The. 3/4-in.

and.. 1- l\4-in. thick specimens were held at temperature at least 30 min and the

1- 3\4-in. thick specimens 45 rnin before tranfifer to the testing machine. .—

—.

.—

-13-

1 I 1 1 1 ( 11 1 ( I I 1 t

1 1 I 1

1111It;OOOt-

,.lW”+“ m *IulmLnLn111

111111I1

111,m-). nmul,W. .f++

. . .10000

I 1 1 1 I 1 11 1 I 1 1 1 [1 I 1 1 I 1 I

1 1 1 1

111

1

,U)+rld+,%”m-mld OOTF; OOA*; OO 00;0000;1 1 1

mmmmrut-+e.lm oo, m.lo. omm. foo 00-, OOOO. (1 . ...1.... . . . . . . . . .100001000010 --+ 01+-+001-+ --!-+ +-+-+1

1 ( I (

11;mm,om

,mr.lmm

10” 0“ 9’ 01 1 1 I 1 I !1 1 1 1 1 1 (

1 1 1 1

I 1I 11 1.—. !

1

1111 1 1 1 1 1 I

I I I 1 I I 1! I 1 1 1 1 11 I I I 1 1 11 1 1 I 1 1 11 1 1 1 1 1 11 1 1 1 1 I 1

1 (1 11 11 I1 1

, .

11

;Ooooi,0000,

G$:21ImLnm Lnl1 1

—I1

-.— 1 I 1 1I 1 1 11 1 1 1

1 11 11 11 I(Ouloml(. ...1~:mr.l:l1 1

1 1 [ (1 1 I 11 1 I

III 1 1 1lu-lolrlt-bma~to 00o11....1,...1. .1,ro+a+, l-.lm~tilm -+.

I 1 1 1000.0. ;OOOO, ;0 QO+loaool. . . . . . . . .1 . ...1Ot-Qdl Ac-m N,.D: ~:l~q+:laaula(.n. nm.nl 1 1

I ( 1 1I 1 1 11 1 1 1

1 I I II 1 I 1I 1 I 11 1 1

1 11 11 I1 I 1 I 1 I

,oe*t.J:m Pma;t-or--+;1m !-l m m.lm C-JN. Fa,d. r-l N PJl

1 . . . . .1000010” 0” 00100 0011 1 1

1 [ 1 1

?.JA.ca;++am (elmLom:aam-l;I

m -d+m ro, m m m PIlm m w m.,m,rc M co,. . . . . . . . . . .

00001000010 0001000011 1 1 11 1 1 11 I 1 I1 1 1 11 I 1 1

,Q.na. a;1

m-mm1 . ...1100001I I1 11 11 (( 1

1 1 I 11 1 1 11 1 1 I1 ! 1

,0000 ;0000;00 00;!

,0. 0 0 0.,0. 0 0 0.,0 0 0 0.,. . . . . . .

1-4 -- +,+4--,- 444,1 ( 11 I

1 II II 1I 1I 1

1 1 I 11 1 1 1I 1 I II 1 !

( I I II 1 I 1I 1 1 I1 1 1 1( 1 ( 1 1 1 1 I

100001000010 0001,0000,0000,00 00,mmaamm+am moaIhlmmalmm!-mlbd a+lImmln Lrllmmm ml* Lnmml1 I !

0 0 Oio 0010000100001oo~~,goaalo 000,0000,0 Omom-+r-lmol- r-0 ml Ln;ul co.T;:~mLnlmoO of. ru, ml,uulml

I 1 I 11 1 I 11 I I 11 I 1 1

1 I 1

1 II 11 11 11 1

,+—

,-

1 I t 11 I 1 I1 1 11 1 1 11000O1OO.O1OO 001,0000,0000,00 00,mLua.Dmm+Nm Nom

Immcoaltimdmlr- +Q+lIlnulmmlmmm ml+ Ulmml1 I 1 I

0000; 0000;0 0001000010000 ,0000,0 000,0000,mou-lwr~m~o OIflmmwt-dCJm-$mlmtimmlm+ +rob,o owlLnm Lnullmlnmmlm Lnulmlulmmml

! 1

j~

0m I

I

100001,0000,

[%:$:1Imulmml1 I

.—

-14-

1 1 1 I1 1 1 11 ( ( 11 1 1 1

1 1 1 1 I( 1 1 1 I1 1 1 1 11 1 1 1 I

I11

1

1 1 1 11 1 1 11 ( 1 11 I 1 I1 I 1 1

c-aOm~mOmN~*m m+~m+km~-. v. m. T.lm. n. m. W*,W u’ m u’ I.-1 m m LO,. ...1. . .

00001000010 0001000011 1 1 11 1 1 1I 1 1 11 1 1 1I 1 I 1

1 1 1 1 1( I 1 1 11 1 I 1 1I 1 1 1 1

m U.a; oD+d; mrooo ;O +0%”;r.l Om. mJ, ~**w= [l. .DOmlomoco

. . . . . . . . . .0“0010000!0 0-01-0-01

1 1 1 I1 1 I I1 1 I 11 I 1 [

1

I 1 1 ( 1I 1 1 1 (t 1 1 1 1I 1 ( 1

11 1 I 1

0 00100001000 01000010 00,0000,000 0,0000,cJOOOmmot-Om Nr-ml OOr-

+r.JIQ-+o. mlm Qq+ Ida+-]ml W)tilr.1-#mmll+mm r-lm-$m r-1

1 1 1 1

ioooolooooio Oooiooooi,0000 ,0000,0 000,0000,

Lnmu)dt-t-mu.om Mdco.a%nmImaomlwem+$lm AmmkopmcoI+coo-t-l-mu-mlr- 03mmlamm. ol

1 I 1

I(!11I1I1I11111

( 1 ( 1 11 1 1 I 11 1 I I 11,1 1 I 1

1 I 1 I1

1 I 1 11 1 1 11 1 1 11 1 1 I1 1 1 1

1 10000:0000;00 001000010000,0000,00 00(0000,r-mulmo.dmma -+ti-+PlcoomV+r-u-llmmcobl++ +Plld.t.tf.1(mmmmlmmcoml- mwo.1mmmo.1

1 1 1 1

iooooiooooio oooiooooi,0000,0000(0 000,0000,am-+mr-t-mmo mco-ima+m

Io,ad+ol+wm+lm -+o. mbca PmlImmmmlc-mo. mlr-m mmla~malI [ 1 1 1

F1

1 11 11 1

1 1 I I 1 1 I 1( 1 I I 1 1 1 1I 1 1 1 1 1 1 i

1,Om 0. 0, -.:

lr-mmo.!IhDmullnl1 11 11 1I !1,1

1 11 11 1

100001

1 1i 11 11 1( 1

I !I 1 1 1 1 ( 1 11 1 1 I 1 1 1 1z I 1 1 1 1 1 I1 1 ( 1 1 1 I I

;Offlmqo-+om, m*O D,O+ O+JO --;OOOO; 0000 :1 1 1

,mmmlm-)lr-mor-lr - .DO. bmlo. aom. ,yo U.o., ooo. o., o. OOO+l. . . .

10”0“ 0’ 0“10’0’ +’ 0“10”0“ + 01+ 0“~’ 010 ~ o“01+ ~ ~ +1- + + +1

UEAd

1 1 ( 1 I 1 1 11 ( I ! 1 1 I (1 1 1 I I 1 1 11 1 1 1 1 1 1 I1 1 I I 1 1 ( I1 1 1 1 ( 1 I 1

I I

1 11 11 1

I

1 1 1 1~1 ,- ! 1 ,+ 11 1 I 1 1 1 1 1 1 I1 1 1 1 1 I 1 1 1 11 1 1 1 1 1 ( ! 1 1

1 1 1 1 1 1 1 I 1

-15-

0.

$

t-.mw

e-,

+“FJ

0

0

-nu!

r-- II1IIIII

I11I11III

000. . .

r--r-r-QQQ

.-.

000. . .

r’-m t--mma

0.

0In

.U-Im I

IIII

II

I

I

0001000010cOLoalm-F-IUIeJWINl~

000. . .

UJ**ml

0.

mw

000. . .

c-r’- r’-aaa

moo. . .

mt--r-m.a*

L1-)mo. .

*“Am-l+Lnm

000. . .

OAQa%o~

0.

$

w.

mu-’

II

IIIIIIII

0. 0 of. .[

m-d,ddti

11I11IIIIIIII1II

Omot

II

IIII(II

II

0001000010OQO, NAU-)*O+0+1+dd,-l, d

I

I000

. . .0001000010O-+o, caPt- r-lNmmlslAA -,*

t- r.1.ar-

u- NM,+C.le-)mro

1III

1II11I1I!II1III

I0001000010NQOl@lOr.l o++4,-4 PII*AAA, A

IIII

z0

I

I1

oOalr-. .

“’ e“+mo~malv

[II

[1I

I1IIi1I

Oomlln. . -1.

r-~gl+A

111

0001000010Oaol:Co o-tIw-itilti

-—m 0

0m

r--

II1I

01

QPao;II

II

II

I1IIIIIII

III

IIIIIII

Lnolln. .1.

ON NIY+

IIIII1

00010

1III

0001000010o+ml~mow-mold

III

11I

.0

m

o

0

0001000010od*, aat-mooml$I

I .-Id,

I1

I1

II

I1II

000;

1I

0001000010OQNI:madt-mdrn

. . . I I

000’0

III

2IIIIIIII1

I1

IIIIIII

IIIII1III

I

[

E’E:

21

.

-16--—

TYPE

a

b

c

d

e

LOAD-LB$.

I09000

75000

50000

25000

0

I00000

7500C

500DC

25000

e

100000

75000

50090

250013

n

1Ooooc

75000

50000

25DO0

o

100000

75000

!50000

25000

c

LOAD- DE FL ECT; (IN DIAGRAM

TEST TEMP.

+100% \

TEsT TEMP.

+6D° F

M––- ___~c

7’ 5TART OF CLEAVAGE

DIM

TEST TEMP.

+’20 ‘F

/

.-–—~M:~C

START OF CLEAVAGE

DM

TEST TEMP.

Q°F

~–– —LM=LC

START OF CLEAVAGE

J

TEST TEMP,20° F

1 1 1 1 I I I 1 ( I 1

0.2 0.4 M 0.0 1.0 G!

DEFLECTION– INCHES

“RA(jTURE APPEARANCE

Fig. 11. Typical Load-Deflect ion Diagrams and Fractures.(1 1/4 in. Thick Plate, Heat 2G2850, Lab. Code 241C)

.—

,,1,

b i I1,

I

1

1

!

!

Fig. 12. Transition Curves as Determined by van der Veen Notched

-20 0 20 40 60 80 100 120 140

160000

~12 tJ

160000 tB4)

140000 / 4b

4I ●

130W0 r’ ? w

120000 :

110000

1000000

90000Duct llityCriteri*n

70~)2 )3

60

50/ !

40/

2—— — ——

30 — —— — —— —— .—

●

200

10Fraeim

3iF Appearance

0 ~

-20 0 20 40 60 60 100 120 140

s

Fig. 14. Transit ion Curves as Determinedby van der Veen Notched Slow-BendTest. (1 3/4 in, Thick, Heat 2G2850)

Fig. 13. Transition Curves as Determined by van der Veen Notched

Test Temperature ‘F

Slow-Bend Test. (1 1/4 in. Thick Plate, Heat 2G2850)

-18-

.TABLE V - SUMMARY OF TIWNSITION TEMPERATURES ASDETERMINED BY TEAR TESTS AND VDV SLOW-BEND TESTS

Transition Temperatures [” F)Tear Test van der Veen Test

Fracture Transition Fracture !twL!!itiQn Ductility TransitionStart of Middle of 32 mm 0.7 Max.Trans. iilcatter Band —-Fibrous Load D?

Heat No.ZG;28’50;

3/4 in. 70(Slab B)

1 1[4 in. 70(Sk.b C)

Heat No.2G 2883+*

3/4 in. 60(Slab B)

.1 1/4 in. 90(SJab C)

1 3(4 irl. 90(Slab 13)~

60

55

7’5

50

8.5

105

65

57 54

69 67

95 -

42 40

87’ 85

.’?1 -

61 -

5

14

4

~,

9

.-

33 35

17 22

7

—

~ Tear specimens of 1 3K4-in. thick plate reduced in thickness to 1 .1/4 in. fortesting.**samples for tear and .VDV tests were from different portions of plate (see Table I.).

-,.

1 I I I 1 1 L I!_

tl !i: ,,I Ii:! I

0,0 I I I I I

iiI I I I I i I I I I I I I I I I I I 1 I I .J

-*O-8Co ,0 *O 30 .0 m ,0 ,’0m ,0 ,00 -20 -10 0 10 .0 30 m m 60 70 m ,0 100TEST TEhi PER fiTu RE-”F TEST TEMPEROTU3E - “F

—.

Fig. 15. Transition Curves as Determined by van der VeenNotched Slow-Bend Test. (3/4 in. Thick Plate, Heat 2G2883)

Fig. 16. Transition Cumes as Determined by van der VeenNotched Slow-Bend Test. (1 1/4 in. Thick Plate, Heat 2G2883)

1

Test Temp*F*tUrO “F

leooo

130000

120000

110000

100000

90000

80000

70000

70

60

50

40

30

20

10

0

Fig .

-20 0 20 40 60 80 100 120 140

T*8 b T.mperntwe ‘F

17. Transition Curves as Determinedby van der Veen Notched Slow-BendTest. (1 3/4 in, Thick Plate, Heat2G2883)

-20-—

-..

Fig. 18. TransitionCurves as Determined byvan der Veen NotchedSlow-Bend Test.

(1 3/4 in. Thick Plate,Heat 2G2883, Lab. Code41 D.)

I I I II I I I I I I 1 I I I 1 I I I

RESULTS

Results of tests are given in Tables II, III, and IV. Typical load de-

f le ct ion diagrams and fractures are shown in Fig. 11. The transition curves

based on the data of Tables II, III, and IV are presented in Fig. 12 through 18.

The transition temperatures, as previously defined, have been indicated by

heavy crosses on the transition curves. It is to be noted from Fig. 12, 13, 15

and 16 that the alternative methods of determining transition temperature for

either ductility or fracture–appearance criteria yield substant ially the same

results. In determining the ductility trams it ion temperatures, considerable

s caller of the Lm and Dm values was encountered at the lower test temperatures.

In future work, it is planned to conduct a greater number of tests in this region

in order to better define the lower end of the transition curve.

A summary of the VDV and tear-test transition temperatures is given in

Table V. The fracture-appearance transit ion temperatures derived by both

methods show fairly close agreement, particularly if the middle of the scatter

band in the tear test is used as the basis of comparison.

Further analysis of the data will be made by R. W. Vanderbeck of the

United States Steel Corporation who has accepted the task of compiling and

correlating all data obtained on the same steel by the various cooperating lab-

oratoriess.

--

. .

-.

.

On the basis of the limited amnunt af work reported here, the van der Veeri

rmt d-wd slow-bend test appears to provide an estimate ~f the ha cture transit ion

temperature similar t~ that obtained in the Navy tea-r test. A ductility transition

temperature can als D he evaluated hy tk vari tier Veen test, but during the present

study it appeared that more tests should have ‘been conducted at lower temperatures

in cwder RI Penn it nwxe suitable se k cl Ims of the duct 11ity transition temperature.

piak? steels. hn this mnnectkm, the Material ?Mborat,ory is currently applying the

van de.r Veen test to samples Qf N3fl ship plate under inve stigati,on at the National

Bureau d Standards as part K3fShtp Structure Committee project SR- 139$ “Joint

SSG–AISI Study. 0’

40

REF%REN’CES

van der V~en., ]. H. P Devek@rnent of a Test iug-?dethmcl on Brittle Fracture—.. —of Mild steel Platesj Palris: Intern.atic.mml, Inst Itute of We~d !@, 1953..— ——.—.

Vanderbeckj R. w 0$ J&nproved Notch Tcmqhness d Experimental Sernikil~—— .— .. . .@els CXier OnG Inch h ThbdGie ss (Shb Wmctmw Cmmmittee Report Smial— — —.— ——No. SSc–lol)j Washington:N~=&l Academy of SCknee s-}~ational R.e–search council, August 1, 19560

Evaluation of a Method for Determining the Tendency of Mild Steel to Brittle.— —,- -—- —— —< —.-— —.Fracture (Report NO. NW32/lA), l’h~ Hague - The Netherlands: Centrum VOOr——Lastenchniek N.V. L. - T. N. O., April 1955.

“(3mnpari,sm of WDIWNot.ch-l%nci Test and V-Notch Charpy Impact Test”(Intern.a-tionai Inskitute d Wslding Report), & ‘Welding Juwrnal, 34:7,%s~arch Supplement, pp. 338s-346s (July 1955) o

.

—

—J

I i’1,

—

--

—

.

Ii

.—

. .,

1’,,

‘1

,- ,,I

!-

COMMITTEE ON SHIP STEELDivision of Engineering & Industrial Research

National Academy of Sciences-National Research C ouncil

Chairman:John ChipmanProfessor of Metallurgy and Head of

Department of MetallurgyMassachusetts Institute of Technology

Vice Chairman:M. W. LightnerVice President, Research and Technology DivisionUnited States Steel Corporation

Members:C. S. BarrettProfessor, Institute for the Study of MetalsUniversity of Chicago

Paul FfieldAssistant Manager of ResearchBethlehem Steel Company

Maxwell GensamerProfessor of MetallurgyColumbia University

J. R. Low, Jr.Head, Metallurgy Research UnitResearch Laboratory, General Electric Company

T. S. WashburnManager, Quality Control DepartmentInland Steel Company

T. T. WatsonDirector of ResearchLukens Steel Company

i’

/-’/.