w

NATIONAL RESEARCH COUNCIL

2101 Constitution Avenue
Washington, D. C.

October 23, 1946

\

Chief, Bureau of Ships

Navy Department
Washington
25, D. C..
Dear Sir:
Attached is Report Serial No. SSG5, entitled I/Causes
of Cleavage Fracture in Ship Plate: Hatch Corner Testsll. This
report has been submitted by the contractor as ?-he-freport on the work done on Research Project S%92 under Contract
NObs-31222 between the Bureau of Ships, Navy Department and
the University of California.
B

The report has been reviewed and acceptance recommended

by representatives of the Committee on Ship Construction, Division
of Engineering ad lkdustrial Research, NRC, in accordance with
the terms of the contract between the Bureau of Ships, Navy Department and the Nationsl Academy of Sciences.

Very truly yours,

.\l--*

Chairman, Division of Engineering

and Industrial Research
Enclosure
t

PREFACE
.

..

The Navy Department through the Bureau of Ships isdistributing this

report to those agencies and individuals that were actively associated with tinis
research program. This report represents a part of the research work contracted
for under the section of the Navyfs directive JItoinvestigate the design and
construction of welded steel merchant vesselsrla

The distribution of this report is as follows:

Copy No.
Copy No.

1- Chief, Bureau of Ships, Navy Department

2 - Dr. D. ;~.Bronk, Chairman, National Research Council
Interim Ship Structuxe Sub-Committee

Copy No. 3 - Captain L. V. Honsinger, USN, Bureau of Ships, chai~~an

P. Roop, USN
Copy No. 4 - Captain k?.
Copy No, 5 - Commander R. D. Schmidtman, UXG
Co?y No.
6 - Lt. Comdr. E. U. iiIcCutcheon,USCGR(T)
Copy No. 7 - David P. Brown, American Bureau of Shipping
copy No.
8 - John Vastaj U. S. Maritime Commission
COFJ No. 9 - I. J. Wanless, U. S. Earitime Comniission
Copy No. 10 - J. L. //ilson,
American Bureau of Shipping
Copy No. 11- Finn Jonassen, Liaison Representative, NRC

Members of Advisory Committee of Research Projects

%89,
sR-92, sR-93 and SR-96
Copy
Copy
Copy
Copy
copy

Copy
copy
Copy
Copy
Copy
Copy
COpY
copy

copy
~

copy

No.
No.
No.
No.

12 13 1415 No. 16 No. 17 No. 18No. 19 No. 201!0.21,No. 22NO. 23 No. 8 IJo.10 No.24 -

G. S. Mikhalapov, Chairman
David Arnott
J. L. Bates
H. C. Boardman
paul Ffield
C. H. Herty, Jr.
S. L. Hoyt
J. M. ksSfii-h3
A. Nadai
J. Ormondroyd
H. J. Pierce
d. C. &ith
John Vasta
J. L. Jilson
;i.
k. dilson
Navy Department

Copy
Copy
Copy
COpY
cOpy

No.
No.
No.
NO.
No.

1 25 26 2728COpY No. 29 Copy No. 30 Copy No. 31 -

Vice Admirfl E. L. Cochrane, USN, Bureau of Ships

Capt. R. A. Hinners, USN, David Taylor Model Basin

Comdr. R. H. Lambert, USN, Bureau of Ships
Comdr. R. S. Mandelkorn, USN, Bureau of shi~s
Comdr. J. H. Mc@ilkin, USN, Btieau of Ships
A. G. Bissell, Bureau of Ships
J. il.Jenkins, Bureauof Ships
,...
E. Rassman, Bureauof Ships

,.

Navy llepartment(centd)
cOpY No. 32 Copy No.33 Copy No. 34Copy No. 35 COpy NO. 36Copy No. 37 Copy iio.38 copy No. 39 Copy No. 40 copy No. Q Copies No. 42
Copies No. &

T. L. Soo-Hoo, i3ureauof Ships

B..E. Wiley,Bureau of Ships
K. D. ~Klliams, Bureau of Ships, ~
Noah Kahn, N~w York Naval Shipyard
J. R. Osgood, David Taylor Model Basin
R. M. Robertson, Office ofResearchan@ Inventions
Naval Academy, Post Graduate School
Naval Research Laboratory
Philadelphia NavalShipyard
..
U. S. Naval Engineering Experiment Station
and 43 - Publications Board, Navy Department, via Bureau of Ships,
Code 330c
and 45 --Technical Library,.,Bureauof Ships, Code 337-L

.*
,

U. S. Coast Guard
Copy No. 46- Rear Admiral Ellis Reed-Hill, UWG
Copy No. 47- Captain R. B. kit, Jr., UXG
Copy No. 48 - Captain G. A. Tyler, USCG
U. S. Mcritime Commission
copy No. 49 - Captain T. L. Schumacher, USN
Copy No. 50 - E. E. Martinslq
Representatives of American Iron &nd Steel Institute
Committee on h!alltiactwfi~
?robla.s
Copy No. 51-

hi.Parker, Secretary, General Technical Committee,

Americm Iron and Steei Institute
COPY No. 52 - L. C. ~ibber, Carnegie-Illinois Steel Corporation
Copy No. 17 - C. H. Herty, Jr., Bethlehem S$eel Company
copY No. 23 - E. C. Smith, Republic Steel Corporation
c.

,..

Welding Research Council

copy No. 53 Copy No. 54CopyNo. 55 cop;~ No. 56-

C. A. Adams
Everett Chapman
Lakotte Grover
dilli~l Spraragen
,

Copy No. 57- Dean F. hi.l?eiker,Chairman, Division of Engineering and

Industrial.Research, NRC
Copy No. 58 - Dr. Clyde Jilliams, Chairman, Committee on Engineering Materials
Copy No. 59 - V. H. Schnee, Chairman, Committee on.S@p Construction
~
Copy No. 11 - Finn Jonassen, Research Coordinator, Committee on Ship Construction
cOpy No. 60- M. P. OtBrien, Technical Representative, Research Project SR-92
. .
copy.No. 61 - d. Paul DeGar.mo,Investigator, F@search Project S%92
copy No. 62 - J. L. Meriamj Investigator,Research Project SR-92

CopyNo. 63 - T. R. Cuykendall, Investigator, Research .Roject s&&)

~
t .
H. E. Davis, Investigator, ReseqCh project SR.-92
,
.
COPY NO. 64-

,4~
~,.
.:.
.~

Copy No. 65 Copy No. 66Copy No. 67 Copy No. ~ Copy No, 68 Copies No. 69
Copy
Copy
Copy
Copy

No.
No.
No.
No.

7475 76 -

77 -

M. Gensamer, Investigator, Research Project SR-96

Albert Muller, Investigator, Research Project SR-Z5
E. R. Parker, Investigator, Research Project SR-92
d. M. dilson, Investigator, Research Project SR-93
File Copy, Committee on Ship Construction
thru 73 - Library of Congress via Bureau of Ships,
Code 330c, Navy Department
E. T. Barron, Carnegie-Illinois Steel Corporation
Mark Grossman, Carnegie-Il.l.inois
Steel Corporation
A. B. Kdmzel, Union Carbide & Carbon Research Labs., Inc.
L. P. McAllister, Lukens Steel Company

COpy NO. 78CopyNo. 79 copy No. 80Copy NO. 81Copy No. 82Copy No. 83 copy NO. 84COpy No. 85 Copy No. 86Copy No. $7copy NO. 8$Copy No. 89Copy No. 90Copy IJo.91 copy No. 92 copy No. 93 Copy No. 94Copy No. 95 Copy No. 96Copy No. 97 Copy No. 98Copy No. 99 Copy No.lOO(Copies No. 78 thru 100-

Bureau of Ships)

Total Number of Copies

,::, -,
..:,..,.:..:.:
,..

100

FINAL REPGRT
U.S. Nav-yResearch Project NObe-31222
CAUSES OF CLEAVAGE
FRACTURE IN SHIP PLATE
Hatch Corner Tests
March 1, 1946

to

August 31, 1946

From:
University of California
Department of Engineering
M.P. OIBrien, Technical Representative
Report prepared by:
E. Paul DeGarmo
J. L. Meriam

ABSTRACT
.

This report deals with the testing of seven large welded steel
--

specimens similar in design to a square hatch corner of a ship. These were

the last seven in a series of 26 such specimens which were tested to detenmine the performance of seven types of steel when built into a welded
structure which had severe restraint to plastic flow due to a designed discontinuity. One of the last seven specimens was constructed of a3-1/3 per
cent nickel alloy steel and two frona low carbon steel having 0.82 per cent
manganese.

The others were made of three grades of plain carbon ship

quality steel. Preheating at 400 F was used on three of the specimens.

The effect of preheating was very beneficial. The specimen con-structed fromnickel steel absorbed rather little energy before failure.
This was attributed to the big!.?
yield strength of this material which prevented plastic flow of the plate before failure of the welded jointse
.

Conclusions based upon all of the tests on hatch corner type

specimens are included along with recamnendations for further work~

.:,

FINAL REPORT
U.S. Navy Research Project NObs-31222
CAUSES OF CLEAVAGE
FRAC!I!URE
IN SHIP PLATE

..

Hatoh Corner Tests

~[arch1, 1946 to

All&Jst31, 1946

University ofCalifornia
Department of Engineering
M.-P. OtBrien, Technical Representative
Report pre?ared by:
E. Paul DeGarmo
J. L. Meriam
------

----

Page No.
.!

Introduction . . * .

Procedure .

* .

Results .

11

12

13

.ll+-l+o

Conclusions
Bibliography

Acknowledgment

Recommended Future Work

List of Tables and Illustrations

..,.
.;,;.,~..,
... ..

LIST OF TA,BLHSAND ILLUSTRATIONS

~ge

T&ble

No.

~ - Analysis of Steels . ..o.

. . . . . . . . . . . . . . . .14
.-. .

Table II - Tensile and IIardnessProperties, . . . . . , . . . . . .

Steels for Hatoh CorilerSpecimens

15

Table 111 - Results, Full Scale Hatch Corner Tests for

First 13 Specimens.
..
. .
,
.
Table IV - Results, Full Scale Hatch Corner Tests, .
Specimens 14 through 2.6.
... . . ,,
..
.
,...
.,
Corner
Fig. 1 - Revised D&sign df the Full Scalefk.tch
~~

. ., . . . . . : 16
,,
. . . . . . . .

17.

Model . . . . . 18

Fig..2 - Specimen 20:

Overfill vi.c+w,above

deck

. . , . . . . . . . . 19

Fig. 3 - Specimen 20:

.,
Fig. 4 - Specimen 20:

Overall view, below deck

. . . . . . . . , . . . lg
..

Fig. 5 - SpOcimen 20:

..!.

View of fracture from above deck

. . , . . . . ,20

View of fractures from below deck, . . . . . . 20

outboard, and aft of hatch end beam

-. ,

Fig. 6 - Specimen 20: View of fracture in corner from inside

of hatch

. . . 20

Fig. 7 - Specimen 20:

. . . 21

Fig. 8 - Specimen 20:

, .,

Deck inddoubler fracture patterns,

looking P@d,
..
.
Deck and doublor fracture patterns,,
looking aft

. . . . . 21

Fig. 9 - Spccimcn 21: Overall view, above deck . . . . . . . . . . . .22

,.
.
Zig. 10 - Spccimcn 21: Overall view, below deck. . . . . . . . . . . .22
Fig. 11 - Spezimen 21: Fracture at corner, viewed from above
!.
.,

. . . . .23

Fig. 12 - Specimen 21:

Fracture pattern in doublerJlooking fwd.

.

Fig. 13 - Specimen 21:

Fractures vicwedfromvbelow deck, outboard,

and aft of hzztchend
beam
. .
.,.
.

Fig. 14 - Specimen 21:

Fraottie.in corner viewed from inside of hatch 24

-.
overall view, above deck ~ . . . . . . . . . . 25
.

Fig, 15 - Specimen 22:

.-:,-.,.,:+.,..

:>,<

. . :.
23
, 24
..

Fig. 16 - Specimen 22:

Overall view$ below deck . . . . . . . . .. . . 25

Fig. 17 - Specimen 22:

Fraoture in doubler, viewed from above , . . , 26

LIST OF TABLES AND ILLUSTRATIONS - Cont~d.

Rage No.
.
Fig. 18 - Specimen 22:

Fig. 1.9- Specincn 22:

Fraclmro in deck-hatch cnd beam weld viewed . . 26

from below deck, outboard, and aft of
hatchend beam
Fracturcin corner viewed froi inside of hatch . 27

Fig. 20 - Specimen 22: Fractures viewed from outboard,,below deck,

and aftof hatch end bcxum <
. .
.....
Fig. 21 - Spccimcn 22:
.
.
. ..

Fracture pattern in doubler viewed from

different angles, looking fwd.:

..Fig. 22 ,-Spccirmn 22:. Looking aft at fractufo patterns in deck

and doubl~r
,.
Fig! 23 - Spccimcn 22:
Fig.

24

- Specimen 23:

Fig.26

Fracture viewed fron below dcckD outboard

and fwd, of hatch end bcxan

Fig. 28 - Spccinen 23:

. . . . 28

. . . 28

Overall view, below deck . . . . . . . . . . . . 29

Fracture:vieucd from zbove dock

Fig. 27 - Spccirmm 23:

. . . . . . . . 30
. . . 30

Overall view, abovo

dcok . . . . . . . . . . . . 32
,.

Fig. 30 - Sp~cincn 24:

Overall view, below deck . . . . . . . . . . . . 32

Fig. 31 - $pccimen 24: Fractures viewed from above deck .,. . . . . . . 33

. ,
.
Fig. 32 - Spc6imcn 24: Looking aft at fracture patterns in . . . . . . 33
deck and doubler

Fig. 34- Specimcn 24:

.+

Frackuro in longitudinal coming-hatch aid - . . 30

bean weld viewed frbm below deck, outbo~rd
and aft of lxntchcnd beam
..
,,..
Fracture patterns.indeck, doubler, . . , . . . 33
and 3ftx 3;rbcr, looking f%d.,
.

Fig. 29- Specimen 24:

Fig. 33 - Specimen 24:

..

. . 27

(Werall view, above deck . . . , . . . . . . . . 29

Fig. 25 - S~ci;cn 23:

- Specimen 23:

Fracture in longitudinal coaming-hatch end

. . 34
beamwcld viewed from below deck, outboard,
and aft cf htch end beam
,.

Fractures viewed from below deck, outboard, . . 34

and f%d. of hatch end beam
Fige 35 - Svecimen 25: Iractwe in doubler viewed from above . . . . . 35

LIST 07 TABEM AND ILLUSTRATIONS - Ccmtrd.

Page No*

Fig. 36 - S~ecimen 25:

Deck and doubler fraoture patterns. . . .

looking aft; and section,-lookinginboard

Fig. 37 - Specimen 25: Fracture in corner viewed from inside

of hatch

35

36
..

Fig. 38 - Specimen 25:

Fractures viewed from below deok, . .

outboard, and fWd. of hatch end beam

Fig. 39 - Specimen 26:

Fractures in dock and doubler,

viewed from above

.,.,...;>
::... ( i:. ,

36

. . . .

37

Fig. 40 - Speoimen 26: Fracture viewed fron below dedc, . . . .

outboard, md N/d. of hatch
end beam

37

Fig. 41 - Specimen 26: Fracture patterns in dock and

doubler, looking aft

. . . . .

38

. . . . . . . . . . . . . . . .

39

Fig. 43 - Energy-Temperature Ou.rve . . . . . . . . . . . . . . .

40

Fig. 42 - Load-Elongation Curves

Starting November 1, 1944, a program of research was undertaken

by the University of Californi~.unde~ a contract with the NDRC having as
Fracture of Ship Plate as Influenced by Design and
its title Cleavage.
~[atallurgicalFactors (NS-336).

~{orkunder this projoct continued up

.to,August 31, 1945, and was divided into two parts as follows:
A.

A determination of the influence of metallurgical factors

and temperature on the cleavage fracture of ship plate
containing internnl notches.

B.

The determination of the;effect of variation of material

and temperature on the tendency for cleavage fracture of
welded structural spcci~.enscontaining a discontinuity,
such as,P,a%chcorners.

Part B of this project involved the design znd testing of full

scale ship sections in.order to:
a.

Obtain a spcoimcn approximating an actual section of a ship,

whorcin restraint to plastic fluw is provided by tho inherent
geometry of the structure rather than by artificially induced
notches.

b.

Correlate the effects of temperature, steel, and stress

relief on these specimens with results obtained on flat

plate tests by other investigators,

Since September 1, 1945, this work has been continued by the
University of California under a contract with the United States Navy,
.. .
::%+-.,
,j.
$

.<

Contract I?Obs-31222.

2,

1,2
In previous reports,
published by the Office of Scientific
Research and Development, aocounts were given of the development of a
hatoh corner @pe

specimen containing a corner which had considerable

restraint to plastio flow, and of the testing of thir:teen of these

specimens.

A previous Progress Report of work done under U. S. Navy

Contract NObs-312223 gave the results of tests onsix additional specimens

and some investigations intothe effects of preheating.

This reportcovers the testing of seven additional full scale

hatch corner ~po

isting contraot.

specimens which concluded the work done under the exThis report mkes

use of the data given in the three

reports mentioned above, so that conclusions may be drawn based upon all
the work done to date on the hatoh ookner _&ypespecimens.

,.

1,2
3

See Bibliography
See Bibliography

3.

The design of thu full scale hatch corner type s~pecimenis shown
in

Fig. 1.

reports.2,3

Details of the welding procedure may be found in previous

The analyses andstrength properties of the seven steels

which were used are given in Tables I and 11.

The specimens which wero lmstod arc listed in Tables III and IV.
This report is concorned particularly with specimens 20 to 26, inclusive.
In making specimens 20, 21, and 23 9 pr~heat w2.S ~~cd for all welds ~~ithin
two feet of the corner of the hatch.

~reatingtorches were used to raise

tho temperature of the plates within three inches of the welds to 400 F.
The temperature was not allo~{odto fall below this value until welding
was completed.
100

l?.

In making specimen 22 tho preheat twnperatmre was only

This amount of preheating was used only as a precaution to

avoid cracking since a nickel alloy steel and 25-20 cloctrode wore
It was necessary to conduct two tests on specimen 21.

involved.

On the

first test failure of the aft end connection occurred when the ncninal
stress had reached 33,000 psi.

At this point small cracks had appeared

in the welds acthe corner of the hatch.

the machine and s.new end tab attached,

The specimen was removed from

It was then retested to failure.

Strain gages were attached to the specimens at the locations

shown in the small drawings at the bottoms of Tables III and IV, as had
.

been done in previous tests.

loads of O;

Readings of these gages wore taken at

100,000; 200,000; 300,000;

1,200,000 pounds.

Soo,ooo;

1,000,000; and

Beyond 1,200,000 pounds the readings of four gages

were followed continuously up to failure , or until the gages became

,#::J:
..
:.?:,
;:

inope~ative.
Over-all energy absorption was determined by taking pin-to-pin
2,3
strain measurements, using the method discussed in previous reports.

.40

RESULTS
4; .,

.
.
. .
The results of the tests
specimens are tabul-atd
m
. .of. twenty-sti
.,
..
.
Tables III and IV. It is felt that in order to interpret the ener~ absorption
..
.
.

values shown in these tabies

. . only the values correspondug to failure of the
.,.

should be considered. For specimens where

longitud~.nalto hatch end beam joint
..
cleavage failure occurred this also corresponds to failure of the deck. However,
where she=

fractures occurred if this method of interpretation is not used the

since they depend upon how far the

IIatfailure!!energy values :.
become meaningless

the deck before stopping the test..~is

tear was caused to progress across . . .
distance was not tiiesame in any two cases of shear fracture since in those tests
.

which were conducted at low temperatures,the specinlenwas enclosed ~ a uanvas

,
bag and could not be seen until the longitudinal girder joint had failed and the
.

test was stopped.

.. .
Photographs of the failures in specinens 20 to 26 are shown in Figs.
2 tO 41, inclusive.
. .
,.
Fig. 42.

The load-strain curves for these specimens are shown in

,.

Referrfi.gto spectiens 20 through 24, the beneficial effectsof

409 F preheat are again apparent~., Specimen 20 was essenti~y

the same
The energy

as specimen 8 but reached over 5,000 psi higher mdmuM

.

being about 23 per cent

absorption of tnis specimen was very outstan~g,

greater than that for any other specimen.
.

Sinilarly.in the, case. of

specinen 21 as compared with specimen

., 4, about
stress was

withstood when

preheat was
,,

stress.

7,000

psi

greater

Again
.. in

used.

the

case

.:.
,,.

..,.

.:*:.
+ .:,
,..

.,

,,,.
.
,.

.
. .

of

5*

\
,!

specimens 23 and 24 the use of preheat resulted in about 4,000 psi greater
..
. .
maximum stress even though the non-preheat specimen was a shear fracture.
It should be noted that in the cases of steels C, B, and D,where preheat
and non-preheat specimens were tested at the same temperatures,the in,.

creases due to preheat are from 18 to 36 per cent.

Preheating also appears to produce much greater uniformity of

breaking stresses for specimens made of the various plain oarbon steels.
For specimens made of steels B, C, D, and H, with type E-6020 electrode
, . .

without preheat, the maximum stresses vari6d from 23,200 psi to 31,200 psi,
..
,,,,,,,.
a range of 8$000 psi.
For specimens made from these same steels with .
,,.
...
.
400 preheat tho variation in maxihum stresses was only from 32,400 psi
to 35,400 psi, a range of 3,000 psi.

It will be noted that in the case of specimen 21 the maximum

-.,
..,

,..

stress reached during the first test was slightly higher than for the
..

second test where failure occurred.

ilhe
exact cause for this is not

known but it may be due to strain ago cmbrittlement sinco the retest was
,.
about two weeks after tho first one.
The load strain curve, showm in
.
,, .
Fig. 42, indicates that this specimen was very nearly at its maximum
. ..
possible load when the fs.ilurcof the end tab occurred.
. . .,
. .
While the nominal breaking stress of specimen 22, mude with N
..:
..,,
. ..
.
..
stcol, is considerably greater than was obtained with tho other steels,
tho energy absorption was much lCSS than for sovcwal of the specimens mzndc
.
with the plain carbon stcols.

This situation is not surprising when one

considers that the yield strength of this N steel was around 48,000 psi
and that failure of the spocimcn was brought about by failure of the
wcldod joints.

The yield strength of the plate was so high that the

over-all stress in the spccimon reached a high enough value to oauso

6.

failure of the weld joints befor~ the plate was subjected to a sufficiently
high stress to bring about muoh plastic flow.

Thus in this case the

energy absorbed was the result of a rather high load and very little elongationo
.

This can b6 seen in Fig. 42 which shows the load-strain curve

for this specimen.

The performance of this specimen indicates that if

it is desirable to havo high energy absorption in a welded structure,

the use of alloy steel having rather high yield strength and excellent
impact properties is of little use W.1OSS the performance of the welds
also is improved great~v.
A further indication of the effectiveness of preheating is obtained
...

breaking stress
hatch corner specimen=,.
by computing the ratio nominal...
. of the.-.
.
yield point stress of the material
for specimens made with and without preheat.

Some of these are as follows:

Steel C, no preheat, tested at 68 F, cleavage fracture;

ratio = 0.68

tested c-t70 F, cleevc.gefracture;

ratio = 0.93

Steel C, preheat,

Steel N, no preheat*,tested at 35 F, shear fracture;

ratio = 0,86

Steel H, no preheat, tested at 72 F, shear fracture;

rztio z 0,69

Steel H, preheat,

tested at 31 F, cleavage fracture;

ratio = 0.82

As may be seen in Fig. 28 the cleavage fracture in specimen

23 cx-

tended through the 3 in. x 3 in. bar which was attached to the outboard edge
of the specimen to permit the attachment of transverse restraining beams.2,3
This bar had also fracturedin specimen 2.

Huwever, in the case of specimen

23 this fracture was unique in that the crack

iwnt

outboard in the deck plate

to the outer edge and then inboard through the 3 in. x 3 in. bar.
The deck and doubler plates of specimen 24 were slightly laminated
but it was not felt that this affected the test results.
..,,
.::
,::.
:-.
.
* 25-20 electrode used;

7.3

Fig. 43 shows the energy tornperaturerelationship based upon the

energy values corl*espondingto the failure of the longit~dinaljoint for
This curv~ indicates that the transition
..
temperature for steel C when built into the hatch corner type specimen
specimens 10, 14, 25, and 26.

-is about 85 higher than when determined by Keyhole Char~J tests.

The

fraetureobtained on specimen 26, as shown in Fig. 41, is ofparticular

interest since it started in cleavage, changed to shear, then back to
oleavag~, later going into shear again and then baokinto cleavage with a
small smoun% of shear at the fracture edges.

Although the end of the

fracture was predominantly cleavage it did not progress entirely aoross

the specimen but stopped about 12 inches frcm the outboard edge. At the
-.
point where it stopped there was considerable local plastic flow.

,.

a. .
*

SONCLI?SIONS
. ..

. ....

Thcso;:co~clusi611s
are based upon all of the wori done:&hitch
,. .

..

;.

corrwr type spcciincns&dcr

.<, .,

OSRD Contract OXWsr-1418 and U.S.~avy

..-, !.
...
..

Contract NObs-31222.
..

Fracthcs such as occur in woldod ships oan be rcproducod, both

.
,,
.,
as to frocturetypo and reduction in thid~~.ss~Inlahorator~ type:tests
..
of full scalo hatch eorncr type spccim.cns.
1.

2.

Tho nominalbreaking stress over the load c&ying

area ofthc

spccim&s was as low as 39 per cent of tho nominal

ultimate strength of tho material as detorminod by ordinary tonsilc tckts.
..

3*

,.

The full s~alc models were not as strong in proportion as quarter

scale models (24,000 vs. 36,100 psi nominal breaking stress.)

.4.

Heat tr.catmcntof.a lmtch corner type spocimon for,8 hours at

l?OOO F after welding with type.E-6020 electrode gives ~bout CL25 per cent
incroasc in strength.

such treatment dots not, hovwvcr, .changc.th~tYpC

of fracture and does not give as great m inorcase

in strength as cm be
..,.
. :,
.
.
.
.
.
.
.
.
.
.-.
,L.,
:... .... .:
.,. .
obtained by using preheat at 400 F dtiing welding.
Thti
post heat trcat..... . .::
....
,,...
mmt
dccreasos thohccrdncssof the mid and the he&t affcctcd zone and
.. . . .. .....

alters the miorostructurc~but not as much as preheating at 46.0

F.
. ..

5.
.

,.

c-.

,.

Tho use,of-Z5-20 ~lectrodo in such a structure Zncrc?-s@.s

.th~

bruaking strength by about 15 per cent as compared to E-6020 e,loctro.de.

6.
,., ,

,.:.?;
...
;
~;>;.+
.,

Wh~.nthose
spe-oimen~failed with clc~.vagctype frac$urcs, the
..
.

strengths at.room tompcraturcs were slightly greater than when tostcd at

32 F,

f?

9.

7.

The only steel which produced a shcxm ~pe

built into a hatch oorncw &pe

fracture at 32 F when
-.
specinenwas steel N containing about
;.,.

3-1/3 per cent nickel.

8. :

When fabricated with preheat at 400 F and tested at temperatures

which produced a shear type fracture, steel B was outstanding for its
ability to absorb energy.
,,

It is possibleto obtain good correlation between the transition

9.

temperatures ofsteels in the hatch comer

..
test specimens.

10

type

specimens and in tear

The useof preheating a% 400 F wq.$the most effective procedure

tried, both as to increasing dmmgth

and cner~v absorption, being more

effective than post welding heat treatment at 1000 1?for 8 hours or the
.,,
use of 25-20 electrode.

Maximum ,,
strmgthwas

increased from 18 to 36 per

cent by this procedure.

The performance of the welds v~s,grea.tlyimproved.

Preheating does not appear to influence the type of fracture.

.. . ...,

.,
11.

Keyhole Char~ tests over a range of tetipcnaturesappear to rate

... .

the various steels in the same order relative to transition temperature

..
However, the transition temperas do the full scale hatch corner tests.

atures for tk.esteels when tested as hatch corp-erspecim.e~.s

are considerably higher than for Keyhole Charm

specimens.

For steel llCtt

this

,4

temperature difference is 80 to 90 F.

12

Preheating at 400 F results in a softer weld and heat affected

zono, gives a wider heat affected zone and moduces a different microstructure than is found when welds are net preheated.

..

.
>:.
:::::.
,,
:.:
:..

..
.

23.

When

~. steel

1,00

having a considerably hj.gheryield point (such as

steel llNl~)
than low carbon steel is used M

a rigid and complex structure,

such as the hatch corner typespecimen~ the energy absorption may be less
.

than that obtained from the use of low carbcm steels fabricated with preheat.
This is due to the fact that the welded joints fail before the stress has

become high enough to produce any considerable amount of plastic flow in

the steel. In order to obtain maximum benefit from the use of such higher
strength steels h

welded construction the performance of the welded joints

must be improved.

u.

The substitution of riveting for welding in the hatch corner

type specimens did not give as high nominal breaking stress but did result
in greater energy absorption than was obtained by welding without preheat.
Although cleavage type fractures were obtained in the riveted specimens,

in

no case did they progress farther than the second rivet hole whereas in the
welded specimens cleavage fracture alwys resulted in complete fractlme of
the deck and doubler plates.

15.

The

use of a highly notch resistqnt steel (steel Will)welded

with 25-2-Oelectrode did not result

in

aa high energy absorption as was

obtained with a low carbon, less notch resistant steel welded with type
z6020
,

electrode with preheat at LOO%.

,,

ryarnl
_..A
iI
-------

----

--e

------

-701-

.
.

IL

s.?

r---/--:*,
.,.
.
i

m~

9V

l--f=

-----

....: ..
.-,....
.%-,.
?1

--.?er--j
k

.09

-+

&

11,

BIBLIOGRAPHY

Clewrage Fracture of Ship Plate as Influenced by

Progress Report on l
Design and Metallurgical Factors (11S-336): Hatch Corner Specimen
Testst,OSRD No. 5352, Sorinl No. M-512, July 21, 1945.

Finai Report on Cleavage Fracture of Ship Plate as Influenced by

Design and Wtallurgical Factors (ITS-336): Hatch Corner Specimen
Tests, OSRD No. 6387, Seriai No. M-GOY, December 4, 1945.

Prcgress Report Clcnvr.geFracturo of Ship Plate as Influenced by

Design o.ndMetallurgical Factors: Hatch Corner Specimen Tests,
U.S. Nr.vyResearch Project NObs-31222, SoFtember 1, lS45,to March
1, 1946.

.::.:?,;.
.;...,..:
+s

Serial No. SSC-1, dated July ~,

1946,

u?.

ACKNOWLEDGMENT
.

We wish to express our appreciation of the help given by various

people in oar~ing out the work ofthis project.

Mr. Gee. S. Mikhalapov

and Dr. Finn Jonassen of the ilrar

Metallurgy Committee staff have been
particularly helpful.

Various members of the staff of Richmond Shipyard

Number 3 of the Kaiser Compc.ny,Inc. have given assistance in providing

space and facilitating making

the

specimens.

The staff actively engaged in carrying out the work at the

University of California included the following:

,.

,.......
. ;-:..

Harry L. Aldrich
Winona13ucklin
Earnest Bradford
~~amin G. Dail@Y
E. Paul DeGarmo
Raymond C. Grassi
John T;.Harman
Margaret M. Jordan
James T. Lapsley, Jr.
Douglas M. Nacliillan
James L. Meriam
Clarence Peters
Anne Shultis
Andrew Splinter
Richard Younie

.13

RECOIJMEWDEDI?UTW???$OFW
It appears desirable to test at 32 F a specimen constructedwith

1.

?3ecauseof the outstanding energy

steel Brusing preheat at 400 F.

absorption of specimen 21, Which failed with a shear type fracture, it

would be interesting to know what the energy absorption would be at this
lover temperature even though a cleavage type fracture would probably
result.

2.

Further investigations to determine the effects of preheating are

most desirable.
~Loating

pro~uoes

The results obtained on this project indicate that pregreater

treatment at 1000 F.
.

benefits

than are obtained from post welding heat

In view of the nqy

applications where post welding

heat treatment is impracticable, or very costly, preketitingshould be fully

investigated.

Its effect upon energy absorption is of considerable im-

portance in nany naval applications.

3.

Further studies should be mde

various components of ship cmstruction.

toward improving design details of

The possibility of introducing

more flexibili~ into welded joints and the prevention of severe restraint
should receive careful consideration.

4*
.

In order to determine the manner in which strength varies with

size it wouldbe desirable to construct and test a one-half scale hatch

corner type specimen to tridge the gap in the data now on hand for one\
quzrter and full scale specimens.

5.
..
.-R...
::
,-2
++

It would be desirable to test a hatch corner type specimen the

scum as the existing specimen except having the longitudinal girder continuous instead of the hatch end beam.

14

TABLV I

Analy8is

of.Steels

$s

$ Si.

0.011

0.042

0.02

0.76

0.010

0.030

0.04

0.24

0.49

0.015

(3.033

0.19

0.52

0.01 :.

o. 02,y

0.24

E*

0.23

0.39

0.019

0.032

0.008

H**

0.17

0,82

0.022

0.024

0.15

N*

0,13

0.49

0.018

0.027

0.22

$C

$ Mf2.

A*

0.23

0.47

B*

0.15

C*

Steel

Suppliers ann.lysis

**

Supplied by S. Epstein,

~ Mo.

% Ni.

0.056

0.40

Bethlehem Steel Company

$ Al

3.34

l!ensiie
and Hardness Properties

Steels for Hatch Corner Specimens

.

Plate No.

Yield
M
A-57

Tensile Data (.505 .

Bars)
Break
Uitimate
Elongation
(% in 2)
&
(psi)

Direc.

kmg

35,500

Trans. 38,100
E&l
Long. 35,050
As rolled Trans. 34,000
B-6

(RockwelJ.

Auwd2i)

61,200
60,400

47,400
48,800

39.5
36.2

59.6
56.3

56,900
57,000

38,600
47,5W

40.9
39.6

67,6
58*6

62

59,500

43,400

39*3

43,5m

3$.5

64.0
63.0

64

57,200
68,700
68,000

55,300
57,050

36.0
33.6

59.6
52.5

71

Trans.

35,230
35,750

D-2

Long.
Trans.

37,800
40, 6c0

63,700
63,600

46,900
48,600

37.2
36.6

62.8
59.6

68

E-2

Long. 35,000
Trans. 35,300

58,900
58,200

/!+5,300
46,200

37.2
35.6

59.6
5800

Long.

36,900

Normalized Trans. S6,500

c-l

Hardness

Reduction

Long.

Tensile Data (Full Thickness)

A-57

Long, 35,100
Trans. 34,800

61,4oo
59,800

47,900
49,000

49.2
46.1

58.7

Long .

31,000
31,400

56,5oo
56,400

43,700
45,600

53.2
48.7

66.6
58.4

32,200
32,000

56,900
56,500

Q,1OO
43,400

52.0
51.6

64, o
60.5

c-1

Long. 37,500
Trans. 34,100

66,500
66,200

53,600
56,600

45.5
32.5

56.5
50.4

D-2

Long.
35,900
Trans, 36,100

61,300
60,500

45,8ccI
47,6oo

47.1
46.4

62.3
59.2

E-2

Long.
31,4.00
Trans, 31,000

57j200

4.4.,500
45,600

49.1
45.5

Long.

63,700
63,200

43.0
41.5

68.6
59.0

70

Trans. 33,900
H-2

Long. 37,500
Trans. 34,000

63,900
63,000

~oo
40.5

67.2
60.0

70

N-3

Long.
58,000
Trans. 52,750

82,000
@,300

3s.0
32.5

61.5
54.0

85

B-1
As rolled

Trans

B-6
Long.
Normalized Trans

....5
. .. ....

H-1

37,000

..

56,600

56.3

.. .

L,,!
.: :,.<..

NJ;
.,

RESULTS . FULL
~

..

-r1~/

,*

.-.

LONGITULVAWL

r.

STRAIN CO?U%%VTRATIOIVS
FOR]

S~E

32

32 23,2@- lG- - - - - - - - 6
- -,

I
I

24200

I
I
1
I

HATCH CORNER

TESTS

TOTAL \ STIWIN GMTIOS

.-

SCALE

15 1A #g 2.020 13 -17 !.9

I I I
I

1I----r--I I [..

1321
t

, --- . ,

t-i 77
R.

!715/
,1 15:
-t-t

NOTES
i 2MSE0 ON LOAO CXRRYWG SECTfOIV OF DECK, 0QU6LER.
u
LON61rUDllWlL CL34MING ~LOW OECK.
2 &lSED (WVAVE&lGE
*
LONGITUOlh&lL STRESSES
*
GAGES I-2-3-4
~p
~
~~.
3 ~=
~ AV=T.4=
P LOMSITUOINAL ST/30fS
Kw~JJvs
FOR GAGcS 1-2-3-4
7VP AND 2WTTOM.
A-1
4 C-6W06
E-6020 ELECTR~C
USED #VLCSS ~Tr5D.
5.REAOIN6S @@3TfOA!A13LE.GAGC *II VERY CLOSE 70 00U6LER WELO.

17

-r-h

m
----

.
t

I
1..

1
1-1

.,.

..

.,

%1>

.
L

.*.

:?..

..:

1) v

. .
,

 -s4-

19

.,>.

20

Fig. 4

Fig. 5

Specimen 20:

View of fracture

from

above

deck

Specimen 20: View of fractures from below deck.

outboard, and &ft of hatch end beam

-.

Fig. 6 Specimen 20: View of fracture in corner from inside of hatch

21

Fig. ? Specimen 20: Deok and doubler fracture patterns, looking fwd.

..,
,,,,.;?
.;...

Fig. 8

Specimen 20:

Deok and doubler fracture patterns, looking aft

22

r-l
d

..
A
N

4*
1%

,,..

23

Fig. 11

Specimen 21:

Fractures at aorner viewed from above

..
..

Fig. 12

Specimen 21:

Fracture pattern in doubler,

looking

fwd.

.. ...
d

24

Fig. H

Specimen 21:

Fractures viewed from below deck,

outboard, and aft of hatch end beam

..

Fig.

14

Specimen 21:

Fracture in corner viewed from inside of hatch

A
9+

..

a
N

26

Fig. 17 Speoimen 22: Fracture in doubler viewed from above

Fig. 18

Specimen 22:

in deok-hatoh end beam weld viewed from

below deck, outboard, and aft of hatoh end beam

Fracture

27
,<

Fig. 19

Specimen 22:

Fig. 20

Fracture in corner viewed from inside of hatoh

Specimen 22:

Fractures viewed from outboard,

below deck. and aft of hatch end beam

..4,. .

28

..

Fig. 21

Specimen 22:

Fracture pattern in doubler viewed

from different angles, looking fwd.

.;....

Fig. 22

Specimen 22:

Looking aft at fracture patterns

in deck and doubler

29
q--. ..

----

9-2-+-

.-. ----

.-------

~ :.-

1-

\-

w .

-2

K!

.,

s
G4

30

Fig. 25

Fig. 26

Fig.2?

Specimen 23:

Fracturee viewed from above deok

Specimen23: Fracturesviewed from below deck, outboard,

and fwd. of hatch end beam.

coaming-hatchend
Specimen23: Fracturein longit-1
beam weld viewed from below deek, outboard,
and aft of hatch end beam.

..
, ;.

.,
:

,!

4
-1

Fig. 28

Speoimen 23:

Fraoture patterns in deck, doubler,

and

3 x 3 bar, looking fwd.

Ld

w
t4

_[

Ii
-n%
i)

Fig.

29
,.,,
,:

spcoimcn

24:

Overall tiew, abcwe dook

Fig 30

f@30iEEZi

24:

Overall view, below deok

33

F5g. 31 Speohnen 24:

Fraohwes

viewed from above deok.

,:,.

Fig . 32 Spe d.mem 24:

Looking

aft

at fraoture patterns in deck and doubler

.>,

34

Fig. %5 Specim?m 24:

Fraature in longitudinalaoaming-hatohend beam weld

tiewed from below deok, outboard,and dt

of hatoh
.
end beam.

.,.: .
,.>,

Fig. 34

Speoimen 24:

Fractures viewed from below deok, outboard,and fwd.

of hatch end beam

Fig. 36

Speoimen 25:

Deck and doubler fraoture patterns, looking aft;

and seotion, looking inboard

..

36

-%
-

Fig. 37

Speolmen 25:

Fracture in oorner viewed from inside of hatoh

Fig. 38

Speoimen 25:

Fraotures viewed from below deck,

outboard, and fwd. of hatoh end beam

37

Fig. 39

Spcimen

26:

Fractures
in deok
viewed from abwe

and doubler.

..

Fig. 40 - Specimen 26:

Fracture viewed frcm below deck, outboard,

and fwd. of hatch end beam

38

Overall Mew

of fractures

Closeupat end of fracture

Fig. 41 - Specimen 26:

FYacture patterns in deck and doubler, looking aft

?
t
f?

r 1

VI

I
.

-;,:
,..:,..:,..
.:,.

Id WNLN

-w
1-.

40

.
.

Jooot 000
@oo, 000

800,000

/+

7oo#@oo

600,000

Soo,

000
~

400,000

300,000

200,000

/00,
000
0

20

40

60

80

/00

/@O

/40

/60

,:,,;.,
.,. .
,..