The fundamental factors influencing the behaviour of welded structures under conditions of multiaxial stress and variations of temperature Part I., The effect of subcritical heat treatment on the transition temperature of a low carbon ship plate steel Par

(1)

9 0(1) In

ò2

z -c U) u.i D U) ( (k 4 P Il U'? 'J s, .i

j

FINAL REPORT (Project SR-99) on

Part J: THE FUNDAMENTAL FACTORS INFLUENCING THE BEHAVIOR OF WELDED STRUCTURES UNDER CONDITIONS OF MULTIAXIAL STRESS AND VARIATIONS OF TEMPERATURE.

Part II: THE EFFECT OF SUBCRITICAL HEAT TREATMENT ON THE TRANSITION TEMPERATURE OF A LOW CARBON SHIP PLATE STEEL.

by

W. M. Baldwin, Jr., and E. B. Evans CASE INSTITUTE OF TECHNOLOGY

Transmitted through NATIONAL RESEARCH COUNCIL'S

COMMITTEE ON SHIP STEEL

Advisory to

SHIP STRUCTURE COMMITTEE

LAFORATORUM VOOR

SCHEEPSCONSTRUCÎES

Division of Engineering and Industrial Research

National Academy of Sciences . National Research Council

Washington, D. C. November 6, 1953 SERIAL NO. SSC-64

LA?.ORitT3RUM \IQOR

,-. .- ,- .'

,HLrLO1'.J

(2)

MEMBER AGENCIES: _{ADDRESS CORRESPONDENCE TO:}

BUREAU OF SHIPS. DEPT. OF NAVY SECRETARY

MILITARY SEA TRANSPORTATION SERVICE, DEPT. OF NAVY SHIP STRUCTURE COMMITTEE

UNITED STATES COAST GUARO. TREA3URY DEPT. U. S. COAST GUARO HEADOUARTERS

MARITIME ADMINISTRATION. DEPT. CF COMMERCE WASHINGTON ss, D. C.

AMERICAN BUREAU OF SNIPPING _November

6, l93

Dear Sir:

As part of its research program related to the improvement of hull structures of ships, the Ship Struc-ture Committee is sconsoring an investigation on the fundamental factors influencing the behavior of welded

structures under conditions of multiaxial stress and variations of temperature at the Case Institute of

Tech-nolocy. Herewith is a copy of the Final Report,

ssc-6L.,

of the investigation, entitled "Part I: The Fundamental Factors Influencing the Behavior of Welded Structures

under Conditions of Nultiaxial Stress. Part II: The

Effect of Subcritical Heat Treatment on the Transition Temperature of a Low Carbon Ship Plate Steel," by W. M.

Baldwin, Jr., and

E.

B. Evans.

The project has been conducted with the _advisory

assistance of the Corrirnittee on Ship Steel of _{the National}

Academy of Sciences-National Research Council.

Any questions, comments, criticism _{or other matters}

pertaining to the Report should be addressed _{to the}

Secre-tary, Ship Structure Committee.

This Report is being distributed to those

indivi-duals and agencies associated with and _{interested in the} work of the Ship Structure Committee.

Yours sincerely,

K. . COWART

Rear Admiral, TI. S. Coast G'uard Chairman, Ship Structure Committee

(3)

Final Report (Project SR-99)

on

Part I: _{The Fundamental Factors Influencing the Behavior} of Welded Structures under Conditions of ìultiaxial Stress and Variations of Temperature.

Part II: The Effect of Subcritical Eeat _{Treatment on the}

Transition Temperature of a Lo Carbon Ship Plate steel.

by

W. I. Baldwin, Jr.

E. B,

Evans

CASE INSTITUTE OF TECIiNOLOGY

under

Department of the Navy

Bureau of Ships NObs

-BuShips Project No. NS-011-073

for

SHIP STRUCTURE COiITTEE

(4)

TABLE OF CONTENTS Page

Introduction . . . , . . . i

PART I: _{The Fundamental Factors Influencing the} Be-havior of Welded Structures.

Selection of Eccentric Notch Tensile Test for

Evaluating the Effects of Welding . . . _. 3

Studies with Eccentric Notch Tensile Specimen . 6

Steel Comparison (A and C Steels) ₆

Preheat and Postheat 11

Specimen Size. . . , . . . , , 11

Inhomogeneity of Plate

₁₃

Selected Locations in Weld Metal ₁₃

Comparison with Concentric Notch and

tfnnotched Tensile Tests

₁₅

Metallurgical Structures ₁₇

Hardness Surveys ₁₉

Temperature Measurements ₂₁

Subcritical Heat Treatment (Preliminary Work) . 21

Part II: The Effect of Subcritical Heat Treatment on

the Transition Temperature of a Low Carbon

Ship Plate Steel.

Isothermal Studies. . . e ₂₅ Air Cooled ₂₅

FurnaceCooled.

.

..

₂₇

WaterQuenched..

...

₂₇

EffectofHeatingMedium....00C

₂₉ Quench-Aging Studies. . . , 29 Microstructures . . o e o o s s . o e e o o 31

CoolingRates

... ...

o. oc.

_.

₃₊

Interpretation of Results: _{Weidments and Subcritically} Heat Treated Plate , e

36

Appendix A . o o o s e o o e e o o e o o o e o

.

Appendix B . _e _e

o s e s e s . s . e e s

Bibliography e . . . . _.

(5)

-ii-LIST 0F FIGURES

No0 _TITLE

L

Notch Properties as a Function of Strength _Level

for Various Steels (2)..

...

. . . 5

Relations between Concentric and Eccentric _Notch

Strength Ratios and Notch Ductility .

5'

Eccentric Notch Strength of the Unaffected Base

Plate as a Function of Testing Temperature. . 8

.

_{Eccentric Notch Strength of the Unaffected Base}

Plate as a Function of Testing Temperature. 8

Distribution of Eccentric Notch Strength at -60°F ₉

Distribution of Eccentric Notch Strength _{at -70°F} ₉

70 Transition Curves of the Region of _Lowest

Ductil-ity for Steels "A" and "C". . . , 10

8.

Variation of Transition Temperature _{with Distance}

from the Weld Centerline . . . 10

90 TransItion Curves of the Unaffected _{Base Plate of}

Steel "C" for Three Welding Conditions. ₀ ₀ ₁₂

100 _{Distribution of Eccentric Notch Strength} _{of "C"}

Steel at -80°F for Three Welding Conditions 12

11 _{Transition Curves of the Region of}

Lowest

Ductil-ity for Three Welding Conditions0 . . 12

l2 _{Effect of Specimen Section Size} _{on the}

Distribu-tion of Eccentric Notch Strength at -70°F .

13. Distribution of Eccentric Notch Strength _{at -80°F} 11+

1 _{Comparison of the Eccentric and Concentric Notch}

Properties as a Function of Testing Temperature

Midthickness Tests of "A" Steel _{Base Plate0} ₁₆

15 Distribution of Concentric and _{Eccentric Notch}

Properties at Various Low Temperatures--Mid-thickness tests of "A" Steel _{Weidments, (100°F}

(6)

16e Regular Tensile Properties of "A" Steel as a

Function of Testing Temperature . . 18

17. Distribution of Hardness across Weld at Various

Positions Throughout the Thickness of the

Plate o e G O O O O O O e e e O G o a 20

l8 Distribution of Microhardness Values at the

Mid-thickness of 3/1+ Plate Weidments of Steels

t(ii

_ja _aLJ.L&,q 9A t

_n

_co _o _o _o o e 0 0 G O O G o e O O

19 Maximum Temperatures Reached during elding for

Two Welding Conditions0 e e e 22

Heating and Cooling Curves in the Region of

Lowest Ductility for Two 6 Pass Weidments . 22

Heating and Cooling Curves in the Region of

Lowest Ductility for Two Welding Conditions 2+

22 Effect ol' Subcritical Heating and Cooling on

the Eccentric Notch Strength of Unaffected

ease Plate. . . O G G O o e a 2+

23 Eccentric Notch Tensile Transition Temperatures

of "C" Steel as a Function of Time at

Various Subcritical Temperatures. Air Cooled. 26

2+ Charpy V-Notch Transition Temperature as Func-tion of Time at Subcritical Temperatures0 Air Cooled0 Plate II o a o o o o o a o e

25' Transition Temperatures of "C" Steel as a

Func-tion of Time at Subcritical Temperatures.

Water Quenched and Aged One Month at Room

Temperature0 Plate II o o o o o

Transition Temperatures of "C" Steel Resulting

from Various Subcritical Heat Treatments in

Nitrate Salt Bath and in Air, Employing an

Air

Cool0

o Q a o o e o o a o e o Q O O Transition Temperatures of "C" Steel Resulting

from Beating at 1100°F for Various Times in

Nitrate Salt Bath and in Air, Employing a

Water Quench0 Aged One Month at Room

Tempera-ture0 . , O O C O Q O t O O O O

26

28

30

(7)

TITLE _PAGE

Effect of Room Temperature Aging Time on the

Transition Temperature and Hardness of "C" Steel after Water Quenching from Temperatures

Indicated. _e o e e o e e O

32

Effect of Various Aging Times and _Temperatures

on the Hardness of "C" Steel after Water

Quenching from

_1300°F

e o o e e e

32 30e

_{Summary Curves Showing Effect of Various Aging} Temperatures and Times on the Transition Temperature and Hardness of "C" Steel.

Specimens Water Quenched from 13000F _before

Aginge o o o o e o e e e o e o e e e e

33

Solution and Room Temperature Aging _{Effects on} the Hardness and Impact Properties of tic,,

Steele o e o e o o e o e e o o o s e

33

Comparison of Cooling Curves in

_{the Regionof}

Lowest Ductility for Two Welding _Conditions

with Those Obtained with Heat _{Treated Test}

Specimens, e s

.

e o e e o e e o e O

35

Al. _{Test Specimens.} . _o _e _o _e _e _o _s _e

s e ₁₊₂

A2,

_{Method of Loading to Obtain 1/1+ Inch Eccentricity.} (Eccentricity and the Position of Fixtures

are Exaggerated0). s s o e s o e o e s 1+3

A3. _{Plate Preparation}

. . e o o o o o o e 1+3

A1+e Welding Procedure o e e e e

.

e e e o s s o o 1+3

Location of Eccentric Notch Specimen at the

SurfaceLevel

e eoo o e. e soe.e

₁₊₁₊

Location of LTnnotched and _{Notched Specimens}

at the Midthickness of Weidment. . . 1+1+

A7e Preparation of Charpy V-Notch _{and Notch Tensile}

(8)

This is the final report _{on a project sponsored by the}

Ship Structure Committee under U. S. Navy Contract NObs-+5*7O

and summarizes the work done in the period from July 1, l97

to December 31, 1952. _{Five Technical Progress}

covered the work completed under this contract.

Three progress _{were issued entitled, "The}

Fundamental Factors Influencing _{the Behavior of Welded}

Struc-tures under Conditions of Multiaxial Stress, and Variations of Temperature, Stress Concentration, and Rates of Strain", The major objectives were to determine the relative notch toughness of various zones in commercially welded ship plate steel _{and, if such zones could be isolated to determine the} dependence of the notch behavior _{upon material, variations in}

the welding process, and heat treatment. Eccentric notch

tensile tests at various low _{temperatures were used to}

eval-uate the ductility of _{a small volume of metal from any}

posi-tion in the weidment. _{A summary of the work toward these} objectives is presented in Part _{I of this report.}

Two progress reports'

_{were submitted under the title,}

s'The Effect of Subcritical Heat Treatment on the Transition

Temperature of a Low Carbon Ship _{Plate Steel",} _{The nrincipal}

objectives were to (1) give an insight into the basic

mech-anism which was responsible _{for the brittle zone found in the}

(9)

-2-max±murn embrittlement possible in base plate by subcritical

heat treatment; and

(3)

suggest possible methods of minimiz

Ing or eliminating this embrittlement in base plate which

may be applicable to weldments. This work is presented in

Part

The important findings of the two different phases of the investigation are integrated to show that the quench-aging mechanism appears to be responsible for the brittle

zone outside the weld area of low carbon ship plate weld

(10)

Selection of Eccentric Notch Tensile Test for

_Evaluatg

Fffects of WeiJ

Numerous investigations have shown that steel _structures

may fall in a brittle manner when subjected to _{certain service}

conditions0 _{The conditions associated with brittle failure} include multiaxial stresses, stress concentration, low

tempera-tures section size, and rate of loading.

A combination of these embrittling _{factors may reduce the}

ductility to a low value0 _{The ductility then dictates a struc}

tures resistance to failure rather than _{the strength, because}

it is known that only a small amount of energy is required to

propogate a crack through _{a region of low ductility0}

In view of the fact that the number of brittle failures in ships had increased with the adoption of welding techniques9 it was felt that the welding process altered the properties of the steel. _{Many investigations employing a number of different} specimens have shown that the ductility _{of a weìdment Is lower}

than the ductility of the steel of _{which the weidment is made.}

In selecting a test for locating _{zones of lowered duc}

tility in weldrnents, _{a specimen was needed which would con}

stitute a very fine probe (these critical zones were expected

to be sniall) _{and still include some of the previously men}

tioned embrittling factors, _{Such a test would serve to lower} the ductility of the entire plate to such a point that the

(11)

critical regions could be located0

The eccentric notch tensile test met these requirements

±n that a very mal1 volume of metal controlled the reaction *

of the specimen the embrittling factors o± eccentric load

ing multiaxial stress, and a stress-raiser were present and

the factor of low temperature could be added0

This specimen was used in previous investigations to

differentiate among low alloy steels (heat treated to the same

strength levels) which were known to have different service

properties67

ifl Fig0

1 the properties of four steels were

compared. at room temperature by means of concentric and ec

centric notch tests0 The data used to draw these comparison

curves were obtained from

Fig0

2, a plot of notch strength

**

ratio as a function of ductility0 The concentric notch

strength seemed to be dependent upon the ductility up to about

two per cent, whereas the eccentric notch strength extended

the dependence of the notch strength upon ductility up to ap

proximately ten per cent by the add±tion of another embrittlng factor, that of eccentric loading.

From

Fig0

1, it can be seen that the eccentrIc notch test

was able to detect differences in the four steels, and rate

them in the same order as the concentric notch test0 The

*The fiber at the notch bottom was subjected to the maximum

tensile stress, and was the fiber in which fracture initiated.

**Notch strength ratio is defined as the ratio between the notch strength and the tensile strength0

(12)

rr--'

_fr_

_&_2

Ar-2.(-'

¿9 3-6' .9 P

I'

'

r r

s 9 87

"y

s 9,- \

.r-r,--' j2--'

'_- 4?, -.J_'fr_

- 1-

t

-

--

----. 2Jy

__37 ,_-_

-'ff'

H t i

/

-j;«''

--'-/

ni. -

--4---\

242/

/

9 s, I

rí< .

-_/.) f7

''2'

(13)

-6-ductility of ship plate steel is too high to show up any regions of lowered ductility at room temperature, but with the added

embrittling factor of low temperature the eccentric notch tensile

test can be

expected to detect differences in the various zones

encountered in a weldment0

*

The specimen design is shown in Fg0 A-1 and the method

of conducting a test illustrated in Fige A20

Studies with EccentrIc Notch Tensile rnn

The ship steels selected for study were A arid C project

steels because they had been shown to have widely different

transition temperatures. although of the same approximate

composition0 The properties reported for these steels are

listed in Table

B-l0

Weidments were made of these steels at

the Battelle Memorial Institute under closely controlled

cori-ditioris0 The details of the plate preparation and welding

procedure are given in

Figs0

A-3 and A+, respectively9 and the welding data in Table B-20 All specimens were taken from

the weidments as shown in Figs0

_Â-5

and A-6 so that the long axis was perpendicular to the rolling direction0 About one

month elasped between time of welding and testing0

The first tests were conducted on specimens from the mid thickness level of weidments of A and C steel made with 100°F

*The letter "A" or "B" preceding the number refers to the corresponding Appendix0

(14)

prehea.t and interpass temperature0 _{The transition temperature}

ranges for the unaffected base plate (2 inches _{or more from the}

weld centerline) are shown in Fig. ₃ _{for C steel, and in Fig.} ₊

for A steel0 _{The superior properties of A steel} _{were evident in} *

a lower transition temperature (_800F), _{as compared to -6°F for}

C steel. _{The same relative rating was found in the}

results of

other investigations.

For both steels the distribution of _{eccentric notch strength}

at selected low temperatures and at various _{distances from the}

weld centerline showed the presence of a minimum at 0.3-0»+ inch

from the weld centerline and a maximum in the vicinity of the

weld junction (0.l--0.2 inch from the weld centerline)0 The mini

mum and maximum positions can be seen in the _{distribution at} _'6O°F

for C steel and at -70°F for A steel in Figs. _{and 6, respectively.}

A comparison of the _{average transition curves determined for the}

position of minimum ductility for _both

steels,

Fig. 7, showed that

the transition temperature for C _{steel (_200F) was about 20°F}

higher than that for A steel (Li0oF),

More complete data on the C steel weldment showed that the weld metal and the location of the maximum (0.l-0.2 inch from the

weld centerline) did not _{go through a transition in the range of}

testing temperatures which _{were used.} _{However, the relation}

be-tween transition temperature and distance from weld centerline can be represented as shown in Fig. 8. _{The zone of minimum ductility}

was definitely defined by its high transition _temperature.

*Traflsitjon temperature is here defined as the temperature at the vertical midpoint of the _{average notch strength curve}

(15)

\

'k

.7

I C-, ,-'&

/60 '---j

H

/0

t L 'k

g.

o O O

Il

o(/9

__c,

g'

¿ o t

₎

.- -2:70'

42 2--JJ'

f1,2

w6>ç-,

-o O

8

-f 4-- dT -'2

-2

-'?

2-'

-7;-

--

1--9

J

-

6-W

-J--

9-' -/00 -60 60

/00

,Jy;*r0

;;

3 .2c,«,9/O

4-62/C-:t9- ¿9_'_

7 e7 /'dY

/9/,4,7 _{4(J' A /*'C>Y&4'}

,J'0

(16)

\ /6

o',

k

o o o o

Poo

o 'J-o

00

o

14

42 0 o o

&

/''

&'f

2';(

-

-t-'

7-'«

4Y79

-6&

9429/

-4?

_{?4LfJ _-4/'_}

-e-o o C

*-&4' &

4-7'

ó>-

,1_-,- -%2

(17)

-lo-.

--

-,;r_f_'4_

,-_o' ?.

-z<P'-w

,_:fç

_'-

_{l.' .}

--62

Z-9e:? >

442 A52e

H- h.

-

r

H

I

97 J742

2'2c5W

-

-4- ¿7_

J>',Z4'

9P

ì7'

1-

'257.W

J12Q

l2

-Ip

A.

ftjf

S -'I

7r

c.c-'

:- -'

(18)

Preheat and Postheat

In order to investigate the possible beneficial effects

of preheating and postheating, weidments _{of C steel were made}

using a +00°F preheat and interpass temperature, _{and a}

post-heat at 1100°F to a weldment made with 1000F prepost-heat0 _A

comparison of the average transition curves for the _unaffected

base plate, Fig0 9, showed that

Data for the 100°F and 1+00°F preheat fitted on the same curve0

The postheated plate had almost the same transition

temperature (-75°F) as the plates without postheat

(-65°F)

A. comparison of the distributions of notch strength across the welds determined at -80°F, Fig. 10, _{showed that the -i-00°F}

preheat brought about adefinite _{improvement, and the 1100°F}

postheat a virtual elimination of the region of minimum duc-tility0 _{The improvement was more clearly revealed in} _the

average transition curves for this region, Fig0 11. _The

tran-sition temperature of -20°F for the 100°F preheat was shifted

to +°F by the +00°F preheat, and to _{-70°F by the postheat}

treatment0

pjnen Size

A smaller specimen than the _{one used might be expected}

to show up greater differences in notch _{strengths, because the}

changes in notch strength were very rapid as the distance from the weld centerline was increased from zero to 0.5 inch.

(19)

k g -« '_*'__ _-__/_4__ e,- -

__,--

__9_r:. 67

-2

-óa - 'a

6'

D

zrro

''A

-- _--_., ,_ --- ----,-_v_,, o O ---1

i

L

I - y-aa

4l/

_4_,a

,

-cj'

_'-ç.--

l---o a 0'r _*Ç_ze- -- J,_/_p /c

_{- r7 _v-}

-'-'

-4-a-'ew --

--

-_3__ -12-k _t )'- _qç4,,-___-ç___4,J: -'29q t

//j'_

--r,-I, "

- _',v ,

59 - ,'tz'-

-çii,P

'- _4_tJ' -vaa -6& -

,'- -'-

_-(rJ e'a 'a da -

5-,

è_ t

-t-Ç9JP

--#cf/

---.

AO'

I

_

-__tç#_'

#/

---1-

-4

---

-___9_fJ' -c //.

lt'&" eJf'Zf

2- - 4'

4-t

-"

--+a

(20)

similar but with the area of the notched section _{one-half that}

of the standard specimen) _{were taken from various locations}

of the A steel weldment and tested at _=70°F0 _{These results}

superimposed on the results from standard specimens, Fig0 12, showed little difference. _{The small size then offered no}

advantage over the standard specimen.

Inhomogene of Plate

To determine any differences due to _{gross inhomogeneity}

of plate, tests were conducted _{on specimens taken from as}

close as possible to the plate surface _{of a C steel weidment}

made with 100°F preheat. _{The transition temperature of the}

unaffected base plate (-60°F) was but 50F higher than that at

the rnidthickness. _{The distribution of notch strength} _{at various}

distances from the weld centerline for _{a testing temperature of}

-80°F,

Fig0

_{13, showed the same general behavior as the mid}

thickness tests, but the minimum was shifted approximately 0.1 inch further from the weld _centerline. _{This shift was to be}

expected due to the geometry of the double-V weld employed.

This same behavior would be expected _{for the A steel at the}

surface level _{based on the similar behavior} _{of A and C steels}

at the midthickness0

Selected Locations in Weld Metal

In order to investigate further the possibility of zones of low ductility in the weld structure, a number of probe tests

(21)

200 160 20 80 40

72;'-4,42

-.1.

- 1

X7

q,i,'o

c_ I*'

ì'zcr r'-'

rj'%7

-r

--#-

&''

r

t

--#

f?4*

-/2

¿7'

t--W

-2---7-4-

31 .-

r;->--

':- -'e' ,4

-72

-I-

icfr2r

;f

r

-WELDMENT-IOO°F PREHEAT ANO

INTER-PASS TEMPERATURE

A

RANGE OF VALUES FOR PREVIOUS

RESULTS AT THE MIOTHICKNESS.

.

s

PLATE SURFACE

0.5 .0 .5 2.0 2.5 3.0

DISTANCE FROM WELD CENTERLINE INCHES

FrG. 13 DISTRIBUTION OF ECCENTRIC NOTCH STRENGTH

T 80° F

'C STEEL

(22)

A0 The coarse structure at the weld junction from Pass _5°

B0 The coarse structure at the weld centerline _{approximately}

003 inch from the plate surface0

C0 _{The coarse structure of Pass 2 at the weld} _centerline0

At these positions only high notch _{strength values were obtained,}

which, in conjunction with the high _{values previously obtained}

at the midthickness and surface levels, seemed _{to preclude the}

existence of low ductility in the weld metal0

parison with Concentric Notch and Unnotched _{Tensile Tests}

To confirm that the eccentric notch strength _{was a measure}

of concentric notch ductility for ship _{plate steel as had been} previously shown for low alloy steels (see

_Figs0

_{i and 2), the} results from a number of eccentric and _{concentric notch tensile}

tests from the midthickness level of the A steel weidment were compared0

In Fig0 l+ the notch properties of the unaffected base plate are shown as a function of the testing temperature0 The

concentric notch strength appeared to be _{dependent upon the}

ductility up to a low value (say, two _{per cent)9 wheras the}

eccentric notch s'trength extended the dependence of the notch strength up to approximately 12 _{per cent0} _{This was in good}

agreement with the previous findings0

In

Fig0 15'

_{the distribution of notch properties at various}

distances from the weld centerline is _{shown as a function of}

testing temperature0 _{The concentric notch ductility} _showed

(23)

z z o H 20 80 40 z 20 w lAi 5 I- Ui IO D4

I

_O _I- _o _z o ₂₀ ₈₀ ECCENTRIC TESTS CONCENTRIC TESTS ONCENTRIC TESTS

/7(

-o 40 -340 -260 -180 -lOO -20 TESTING TEMPERATURE°F FIG. 4

COMPARISON OF THE ECCENTRIC AND NOTCH PROPERTIES AS A FUNCTION TEMPERATURE. MIDTHICKNESS TESTS BASE

PLATE.

w

60

140

CONCENTRIC OF TESTINC OF 'A" STEEL

120 o o o

I

H z W H (n

I

O I- o z 40 20 Io o 20 80 H b' FIG. 5

DISTRIBUTION OF CONCENTRIC AND ECCENTRIC NOTCH PROPERTIES AT VARIOUS LOW TEMPER- ATURES-MIDTHICKNESS TESTS OF

' STEEL

WELDMENTS, (P00°F PREHEAT AND INTERPASS TEMPE RAT U R E S)

- ---e--ECdENTRIC TESTS o -70°F -110°F le CONCENTRIC TESTS I 0-40°F s- 110° F CONCENTRIC 0 -40°F -110°F TESTS 2.5 0 05 .0 1.5 20 DISTANCE

(24)

centerline) in the _{saine manner as the eccentric notch strength,}

whereas the concentric notch strength _{did not define this} _em

brittled zone.

Unnotched tensile tests were also made on the same weld

ment to determine whether _{this test could be made sufficiently}

severe, by the use of very low temperatures, to detect ductility

variations in ship plate _steel. _{For the unaffected base plate,}

the tensile strength did _{not define a transition}

range; the

re-duction in area did define _{this range but only at very low}

temperatures, Fig. 16. _{The tensile properties at -110°F of the} zero and 0.3 _{inch positions were about the same as those for} the unaffected base plate at the same temperature. _{From this}

limited data it appeared that _{the unnotched tensile test was}

not sufficiently _{severe to detect ductility variations in welded}

plate.

Metallur ical Structures

A metallographic study _{of the structural zones at the}

mid-thickness level of the C steel _{weidment made with 100°F}

preheat revealed that the weld _{junction was defined by a sudden large} change in grain size at 0.08 inch from the weld centerline.

With increasing distance _{from the weld centerline,}

the

grain size of this structure _{which was cooled from}

above the upper critical decreased0 _{The structure resulting from} trans-formation from the temperature

range between the up-oer and lower

(25)

160 40 20 00 80 60 UJ -J :;5 40

z

LU H 20 o 70 60 50 40 30 20 Io o

-34°

18-TESTING TEMPERATLIRE-F

FIG. 16:

REGULAR TENSILE PROPERTIES OF

'' STEEL

AS A FUNCTION OF TESTING TEMPERATURE.

-

000 FA

ATURE

STEEL WELDMENT

PREHEAT AND INTERPASS

MIDTH!CKNESS TESTS.

TEMPER-OO,Q

Xt.:

8 UNAFFECTED X 0.0 INCH D 0,35 INCH BASE PLATE FROM WELD FROM WELD CENTERLINE CENTERLINE r-X o

___________

g 40

-20

60 -180 -lOO -260

(26)

03 inch from the weld centerline0 _{There seemed to be no dif}

ference in structure between that _{of the critical zone (0.3-0»+} inch from the weld centerline) and that of the parent plate _when examined at either 100 or 2000 magnifications0 _{The structures}

of the 0.3 inch position _{for the C steel which were pre and}

post-heated, respectively, also _{appeared to be the same as the unaf}

fected base plate0

Hardness Survey

Rockwell B hardness distributions _{were first made at several}

levels of a C steel weldment with 100°F preheat,

_Fig0

_17. _The

maximum hardness at the midthlckness _{level (Rß89) occurred at the}

weld junction and _{was associated with a maximum ductility,}

whereas

a lower hardness (RB83) was found at _{the zone of minimum ductility0}

To obtain the hardness _{distribution at more closely}

spaced intervals, microhardness tests were conducted across all the welds at the center of the plates, Fig. 18. A number of hardness peaks

were found, the highest _{one being at the weld junction.}

The other peaks were due to the composite heat affected zone caused by the six weld passes0 For the C steel weldrnents, _{the overall}

level of the distribution was decreased, in order, with

increas-Ing preheat and with postheat. _{The decrease in hardness}

in the

zone of minimum ductility (0.30.1+ _{inch from the weld}

center-line) appeared to correlate with the decrease in transition temperature in this zone brought about by the pre and postheat treatmentS0

(27)

'9

4 ..-.- -4V

44

,-Ç9-,

J?J

I - ---- -L--- 210

.' .'

t L.

,-- r-i-_z_'

;

/99

L

.

-v !4!.

-__.--_

..'

*tft

? j-24 2

2-2t4-'

-q-lJr

2/O . I

'i--f-

..--/9O

/70 f'-é7 i

-

7-4-'

_*-0'

-f'- ,*Çl,"

'.

/.

_i

1

i c'-I -5-';. 2/O

'-A.4' ---rijtr

L 2/0

-

I A

. .

6r___t

..' .e

-J\t.

OJ5/2lh'c5-w&i' i.4'

-.s.

i

>....1

-:f

-.1 . ' ..7 .7 .7 _P Ç92

lI4

-Çñ .'5'7'.6' A' /7. .û/J 9,6'J?0ìY 07 4'ZfJ ..IC,f'ePJJ 'r7/ .17 1I/06'J A&J/72'0VJ -&/8.

--çr

47 -;5'-,+7'Z'..W/C)4'/Jr 0/ -'--1 J72-W»-.-'ZJ' :7_.___

J>2'j

"

4,12 A

7.9A'0019&6'7 7V.9' z4'/c..rA'76J 07 ìY7 A?A7%.

g . .44'/J .4 -¿tfj /,VW 'À'4'Af

r

. -.o'a.'-.rz.y

.:-.:2$iJt;_1

1 -I -I . . / iC ..4X1J.8 - '9./'9 ¿'7 I 'C4'.#'9N _{..Tó'qjÇ4C.2} _V.CiD'9WT I I i

--. -.

.-1

--

*

. ¡-._../ ₁ I.. .1

._

.-IA/d C "-"9/i9 /,VC,y ¿'.-/7I/2.'IIWT 1 1

-,

\T

,'Ç9ûfl' I -I-I-4--fi

Li'

AXf .D -. ¿'f.9J /XCW 7,9ON 7a7 J9'A'74C.( '97 - - -t---1 t

-H-i

k \

1 -. I . . --4 -t I A .4x/d £ -. ¿'9.f /.VC.4' ,W2n '97

-t-t I --t--4

----t-

If

/

i t I ---I I

(28)

-to that of the C steel with 100°F preheat, but _{the peaks were}

lower and the overall curve was lower0

Temperature Measurements

The results of the temperature measurements _{made during}

the welding of the C steel (at the midthickness level) _are

given in Figs0 19, 20, and 21. _{In Fig0 19, the maximum}

tempera-tures reached during the welding process are shown as a function of the distance from the weld centerline for the weidments with 100°F and +0O°F preheat _{respectively.} _{In the region of low}

ductility (0.30.+ inch from the weld centerline) _{the tempera}

ture evidently never reached the lower critical temperature (Ac1) for either weidment.

In Fig. 20, the complete heating and cooling history is given for both welding conditions. _{A comparison of the}

heat-Ing and cooling cycles for the first pass for the two welding

conditions, Fig. 21, showed that the 100°F _{preheat weldment had}

a faster cooling rate. _{This difference can be shown for all} the weld passes.

Subcritical Heat Treatment (Preliminary _Work)

An attempt was made to duplicate the _{embrittlement in the}

region of low ductility by means of subcritical heat treatment of base plate of C steel. _{Specimen blanks were cooled from}

9500F

to give three different cooling _{rates-air cool, oil} quench, and water quench. _{The notch strengths determined} _at

(29)

&4

.'

ç

-j'

W íZ'

__

f,lJP f

Vó

/?'&

<P

,_- -ç#4,_ ,l_,

ri,'

-, ,1J,J. ' -7.

22-Z'24'- 7? AfeW flÇ''

-i

442 L

-2ó 2

-'»6

-l' ¿--'

__-__'7_

--

;6-6' 6

-lJ'.P

71'-J 4 o o L o

f4rp

1_'_ - --

(30)

-8O0F for all three cooling rates were low, and when compared

to the spread of values for the base plate at _{8O0F, Fige, 22}

it was seen that the material was substantially embrittied by

this heating and cooling0

PART II THE EFFECT OF SUBCRITICAL HEAT TREATMENT ON THE TRANSITION TEMPERATURE OF A LOW CARBON SHIP PLATE

ST EEL0

The detection of a brittle zone in ship plate weldments

in a region which was not heated above the lower critical

temperature, and the preliminary work which showed that base

plate could be embrittled by subcritical heating and cooling

led to an intensive study of the embrittlement of base plate

by subcritical heat treatment0 The effects of the following

factors were studied isothermal solution time and

tempera-ture, cooling rate, and aging time and temperature0

The embrittlement was evaluated by means Of the eccentric

notch tensile test, and also because it has such a wide degree

of acceptance as a standard, by means of the Charpy V-notch

test0 This work was supplemented by ha?dness tests and

microscopic examinations0

The studies were carried out with C steel because it

was felt that its inferior as-received properties portended

a higher degree of embrittlement0 _{The test specimens were}

prepared from the base plate as shown in

Fig0

_A-70 _Unless otherwise noted, all testing was done one month after heat

(31)

'9g

9

-'9

¿2f

c& 5

4g::7

/tf .'2''9

-f-?Jt/'

- .-, -

,Ç/-I.#'

3Pf

'-'5z

,_9cff

- - - -- - -. 4 - A -Ç-& 2/. 414%?

df

7»-¿:7

2df

_-'_7_.

--'9

2%

- - u','2 -' 4J.6

-4,*C2

/ / /

4«

// / /4.f4d /4

J///' J2

40' - ,',y_4_ --- -

- 4?-

_'Wz

--7& 2

-'9 -l'

W-'9.

(32)

treating in order to maintain approximately the _{same aging}

interval as used in the weidment study.

Isothermal Studies

Prior to determining the effects of time and

tempera-ture, a check was made on the within-heat variation _in

transi-tion properties of the two different la?ge plates which supplIed notch tensile specimens0 _{The transition temperature of Plate}

I (-6°F) used in the earlier work on weidments was lower than

that of Plate II (+00F). Due to this appreciable difference,

the results obtained with subcritically heat _{treated plate have}

*

been separated as to plate number

Air Cooled

The relationship between notch tensile transition tempera-ture and time at various temperatempera-tures _{in the 700°--1200°F range}

is plotted in Fig. 23. _{With the properties of the as-received}

plate as a basis of comparison, the magnitude _{of the maximum}

embrittlement appeared to be independent _{of the temperature,}

amounting to a 10°F increase In transition _{temperature for Plate}

II and a 25°F increase for Plate I. _{Time at temperature had}

lIttle effect, other than _{an improvement in properties at very}

long times due to a slightly spheroidized _structure0

In Fig. 21, the transition temperature--isothermal time relationship for impact specimens _{heated at 11000 and l2OO°F}

(33)

u-

I

-50 -70 -Io -301

1

-50 o - Io -30

.

-50 -70 -Io

-30--.

50f

_-70L

o PLATE

I

o PLATE I AS RECEIVED 0.I

-o

800°F 950°F 1100°F 1200° F

.

I0 loo

ISOTHERMAL TIME'-) HOURS

1000

FIG.23: ECCENTRIC NOTCH TENSILE TRANSITION

TEM-PERATURES OF "C" STEEL AS A FUNCTION OF TIME AT VARIOUS SUBCRITICAL TEMTEM-PERATURES. AIR COOLED.

u- o 160_\ 140 Cr

u

criû-

20-u

LUZ

o

1 80

W Z ct % 01200°F .1100°F _RECEIVED ,-AS Ç

--

--4----01 I IO lOO 000

ISOTHERMAL TIME- HOURS

FIG. 24: Ct-IARPY V-NOTCH

TRANSITION TEMPERATURE

AS A FUNCTION OF

TIME

AT SUBCRITICAL

TEMPERATURES. AIR COOLED.

(34)

indicated no embrittlement at the shorter times; however, at * the longer times a slowly increasing ernbrlttlement was evident accompanied by a gradual softening0

Furnace Cooled

Spot checks with impact specimens indicated _{that no sig}

nificant difference in transition properties _{can be expected}

from specimens comparably heat treated and air cooled0 Although

no checks were made with notch tensile specimens, it _{is believed}

that a furnace cool would result in _{no embrittlement because the}

cooling rate would be less than the critical necessary to intro duce quench-aging effects0

Water Quenched

The transition temperature-time _{curves for notch tensile}

specimens quenched from ll00F and 1200°F and aged one month

at room temperature, Fig0 25, were displaced to much higher transi-tion temperature than the comparable _{air cooled curves,}

indicat-ing a severe embrittlement, and with _{the magnitude of the embrit=}

tlement amounted to about _{a 500F increase in transition tempera}

ture above that of the as-received plate; and for the 1200°F

series, an average of about 100°F increase in transition

tempera-ture0

With impact specimens, Fig0 _{25, the transition temperature}

*It should be recalled that although embrittlement was

indicated by three different criteria, _{the upper level of "energy}

absorbed" was raised, indicating _{an improvement in properties at}

(35)

'Io

u-o LiJ

90

Cr

D

70

u

HO-

50

uJ

30 Io cf)

z

F-220 os 200

D

r

180

w

-û-160

I-U..Z

140

LO°

H

(1) 120

z

F--30

---4---

- ±

-AS RECEIVED

-50

L I

o,

o

-28-ECCENTRIC NOTCH TENSILE

2. o CHARPY V-NOTCH i

4-o 1100°F 1100°F -4

0

o AS RECEIVED ç I 0.1 I IO lOO 1000

ISOTHERMAL TIME'-HOURS

FIG.25: TRANSITION TEMPERATURES

OF "C" STEEL AS

A FUNCTION OF TIME AT

SUBCRITICAL

TEMPER-ATURES. WATER QUENCHED

AND AGED ONE

(36)

25°F

and the 1200°F series about

55°F

above their respective air cooled curves.

Effect of Heat in Medium

In the first stages of this investigation, both air and

a nitrate salt bath were used as the heating medium0 It was

noticed that, generally, test specimens heated in salt had higher transition temperatures and hardnesses than specimens

similarly treated in air0 This anomalous behavior was revealed

in both the Charpy V-notch and eccentric notch tensile data0

Both the impact and notch tensile transition temperatures

revealed that the magnitude of embrittlement accompanying an

air cool from the salt heat treatments was increased with

temperature, and time at temperature, as evident in the corn

parison with the results obtained after heat treating in air,

Fig. 26. The impact data also revealed that the embrittlement

increased with an increase in cooling rate, Fig0 27

Metallographic examination9 chemical analyses, and Xrays

showed the embrittling agent to be nitrogen, introduced by dif

fusion into the specimens through a scaling reaction of the

metal with the salt0

flçAin Studies

In order to determine the room temperature aging effects,

impact specimens were tested after various aging times _after

(37)

u) IL o uJ I-200 180 160 40 20 80L4 20 00 80 60 40 20 o -20 -40 -60 -80

/

S RECEIVED VA CHARPY V-NOTCH I,

_/

/

II00E-AIR

iiiiiiui:iiiiii

.700°F-320 280 240 200 60 th -J 120 u- LO 80

-lt

V-NOTCH

/

_I,

J-

I

_/

/

_/

y

J

AIR XAS RECEIVED I J' ECCENTRIC j ¡ NOTCH TENSILE I I

"y

_/

/

_'I

¿I

IA

A..)

4//i

0/

/

#1

'y

_/

.

/

---950°F AIR k..1, "AS RECEIVED 1100°F AIR o 0.1 lO lOO 000

ISOTHERMAL TIME AT IIOO°Fç-HOURS

FIG.27: TRANSITION TEMPERATURES OF "C" STEEL

RESULT-ING FROM HEATRESULT-ING AT 1100°F FOR, VARIOUS TIMES IN NITRATE SALT BATH AND IN

AIR, EMPLOYING

A WATER QUENCH. AGED ONE MONTH AT ROOM TEMPERATURE.

0.1

IO

lOO

000

ISOTHERMAL TIME HOURS

FIG. 26. TRANSITION TEMPERATURES OF

'G' STEEL RESULTING

FROM VARIOUS SUBCRITICAL HEAT TREATMENTS

IN

NITRATE SALT BATH AND IN

AIR,

EMPLOYING AN

(38)

temperature and hardness increased with increasing _solution

temperature and with increasing aging time, reaching _a

maxi-mum level, Fig. 28G

To determine the maximum embrittlernent due to

quench-aging, and how to minimize it, hardness tests _{were first made}

after water quenching from 1300°F and aging _{for various periods}

of time in the 35° to 1100°F _{range, Fig. 29.} _{The peak hardness}

reached and the time to attain this peak decreased with increas-ing agincreas-ing temperature. _{Similar changes were observed with the}

impact transition temperature, Fig. 30, when employing aging

temperatures from room temperature to 600°F.

To obtain a quantitative _{measure of the solid solution}

and aging effects as a function of solution _{temperature from}

room temperature to 1300°F, Fig, 31 was prepared. _Below

solu-tion temperatures of 650°F, the quench-aging _{effects were}

absent, but then increased with increasing _solution

tempera-ture0 _{The maximum embrittlement induced in} _{C steel by the}

quench-aging mechanism amounted to a 90°F increase in impact

transition temperature and 25 points increase in Rockwell B

hardness.

Nicrostructures

To afford a possible explanation of the _{embrittlement}

obtained during subcritical heat treatment, numerous structures were examined at 2000X0 _{All specimens which were air} _{or furnace}

cooled showed no difference in structure _{from that of the} as-re-ceived plate except a slight spheroidization which set in at

(39)

200 90 80 mOE 150 Li 140 130

t

20 U) z OE IO I-lOO 90 80 loo 95 90 85 80 75 70 65 60 AS RECEIVED CHARPY V-NOTCH 1300°F 0°°

/

200°F 105 loo 95 90 65 25°F -_._ 200°F -_ 250°F AS QUENCHED FROM 300°F

--u

- 600°F -900 °F

>( .

m ° 85 -I -J W 80 o 75 > --;---1100°F AS- RECEIVED COOLED FROM 1300° F

)

-: /

It

-' o 35°F 80°F 350°F ALL SPECIMENS AIR COOLED FROM AGING TEMPERATURE

/

1300°F

-1200°F 0°D o <.-__-AS RECEIVED _AA _o 0I Io 00 000 0,000

LOG AGING TIME ---' HOURS

FIG. 28:

EFFECT OF ROOM TEMPERATURE AGING TIME

ON THE TRANSITION TEMPERATURE AND HARDNESS OF C STEEL AFTER WATER QUENCHING FROM TEMPERATURES

INDICATED. 0 0.1 IO lOO 1000 10,000

LOG AGING TIME-"HOURS

FIG. 29: EFFECT OF VARIOUS AGING TIMES AND TEMPERATURES ON THE

(40)

u- Ui

I-.4

W I.

-2

w I-'I) z o I- u, z 4 1°-200 -h CHARPY V-NOTCH 125F----_. 190 - 180 170 60 150 140 130 20 'Io 100 90

OR AIR COOLED FROM 1300°F

80 lOO 95 25°F--350°F AS QUENGED

o

::

r AS RECEIVED 70><.--AIR COOLED FROM 1300°F 65 60 00 80°F _80°F

FIG. 301 SUMMARY CURVES SHOWING EFFECT OF

VARIOUS

AGING TEMPERATURES AND TIMES ON THE

TRANSITION TEMPERATURE AND HARDNESS OF "C' STEEL. SPECIMENS WATER QUENCHED FROM 1300°F BEFORE AGING.

-Jo: loo 95 90 85 80 75 70 65 190 180 u-170 .-150 ¿ 140 130 120 l'o 100 90 80 AS RECEIVED

/H0

AS RECEIVED HARDNESS o AIR COOLED O AS QUENCHED I WATER)

(QUENCHED AND AGED AT ROOM TEMPERATURE FOR ONE MONTH.

CHARPY V-NOTCH AG IN EFFECT0

'

SOLID 1"SOLUTION, ///, EFFECTS .1/

/

'/, /, AGING\ EFFECT

L--0.1 IO lOO 1000 10,000 0 200 400 600 800 1000 1200 LOG AGING TIME -'H0URS SOLUTION TEMPERATURE -'--j FIG. 31: SOLUTION

AND ROOM TEMPERATURE

AGING

EFFECTS

ON

THE

HARDNESS AND IMPACT

PROPERTIES

OF "C" STEEL.

(41)

the longer times at the higher temperatures0 No difference

in microstructure from that of the base plate was detected

In the series water quenched from either 1100° or 1200°F and

aged at room temperature, again with the exception of

spheroi-dization setting In at the longer times0 The first indication

that a precipitation reaction was operative was obtained in _a

series aged at 1+00°F after water quenching from 1200°F, which

showed a general precipitation throughout the ferite grains0 A study of the structures after quenching from 1300°F and aging at various temperatures for various times, indicated

that the precipitation from the supersaturated solution _{was a}

two-stage process0 At low aging temperatures the precipitate

was detected as a mottling of the ferrite grains; at the

higher aging temperatures, where an 'overaged' condition was

rapidly reached, the precipitate had grown so as to be resolvable,

2Q

The results of the temperature measurements made during the cooling of subcritically heated test specimens from 1200°F

are shown in

Fig0

32, in comparison with the cooling curves

associated with the zone of minimum ductility for the first

weld pass for the two welding conditions0 With an air cool the

impact specimens cooled at a slightly slower rate than the notch tensile specimens due to the larger mass of metals but at a rate which was still slower than that obtained in the +00°F preheat

weldment0 Although the 100°F preheat weldment had a faster

(42)

1400 1200 800 600 400 200

\

'D

LEGEND

WELDME NT TEST SPECIMENS

(0.3 INCH FROM WELD )

A.... FURNACE COOLED

B... .AIR COOLED (IMPACT) E.. ..400°F PREHEAT & INTER-C....AIR COOLED (NOTCH TENSILE) PASS TEMP., iST WELD PASS. D....WATER QUENCHED F.... 100°F PREHEAT & INTER- I

PASS TEMP., IST WELD PASS

15 30 45

TIME'- SECONDS

60 75 90

FIG. 32: COMPARISON OF COOLING CURVES IN THE REGION

OF LOWEST DUCTILITY FOR TWO WELDING

CONDI-TIONS WITH THOSE OBTAINED WITH HEAT TREATED

TEST SPECIMENS.

1000

(43)

-36-both these welding conditions were intermediate to the air cooled

and water quenched test specimens0

INTPP1ETATION OF RESULTS

In the weidment s.tudy the variations in ductility across

the weld area were represented by the distribution of notch

tensile transition temperature,

Fig0

8 It was believed that

the maximum ductility observed near the weld junction was associated with a tempered martensitic structure, whose

duc-tility was higher than that of pearlite0

The location of the ductility minimum corresponded to a region which was not heated above the lower critical temperature, which, coupled with a relatively fast cool, pointed to embrit-tiement by the subcritical precipitation of carbide from ferrite

(quench-aging)0 That this mechanism was operative appeared to

be corroborated by the beneficial effects of higher preheat and

of postheat treatments The improvement of the ductility in this

critical region with higher preheat was attributed to a slower cool, allowing less solid solution and subsequent aging pcst heat effects, while the virtual elimination of embrittlemert

by postheat was believed due to

overagigt0

Due to a difference in the general types of microstructure across the weld area, the hardness could not be used as a

meas-ure of ductility0 The peak in hardness was associated with the

maxImum notch strength at the weld junction, but the minimum in notch strength occurred in a subcritically heated region, where

(44)

the hardness was leveling off0 _{However, considering the critical}

region which had the same general type of microstructure, the

hardness appeared to be correlated _{with the ductility, 10e0,} _a

higher hardness signified _{a higher transition temperature0}

From the complexity of the time, _{temperature, cooling rate}

conditions ir a multi-pass weidment, it was impossible to deter=

mine the exact quench-aging cycle _{experienced in the critical}

region0 _{It was believed that each weld pass contributed to the} solid solutIon and aging effects, _{but that the first few weld}

passes controlled the amount of carbon initially _{retained in}

solution, while the following _{passes served mainly as short} accelerated aging treatments0

In the subcritical work, the investigation of the effects of solution time and temperature, _{cooling rate, and aging time}

and temperature revealed _{significant changes in ductility,}

hard-ness, and microstructure0 _{These changes were largely}

reconciled with the quench-aging mechanism0

The isothermal studies showed _{that no appreciable varia}

tion in ductilIty can be expected by varying the time at tem= perature with the exception that _{at long times at the higher}

subcritical temperatures softening _{set in due to slight spheroI}

dization0 This stability with time _{indicated that the critical}

region in weldments _{was not the result of decay of}

an unstable

condition0

Air cooling or furnace _{cooling from subcritical tempera=}

(45)

-38-.

enibrittlement that was obtained by air cooling notch tensile

specimens was about the same as that found in the critical

region of the C steel weidment made with +000F preheat, and a

was shown in FIg0 32, the cooling rates were about the same0

WIth an air or furnace cool, it was reasoned that the degree of supersaturation was low or nIl due to the fact that pre-clpitation occurred largely during cooling; consequently, no apprecIable change in ductility was realized on subsequent

aging0

A consideration of the quench-aging results showed that the degree of embrittlement was influenced by the subcrItical

temperature and the aging time and temperature0 The solubility

of carbon in ferrite increased with increasing solution tempera-turo, resulting in a greater 'as-quenched' hardness, andy ap-parently, a greater initial transition temperature due to the

higher degree of supersaturation0 Upon subsequent room

tempera-ture aging, the magnitude of the peak embrittleinent reached also

increased9 the higher was the solution temperature,

Fig0

28

ThIs WaS in line with other quench-aging systems which showed

that a greater change in properties can be expected with in

creasing solution temperature0

With accelerated aging, the changes in transition

tempera-ture and hardness,

Figs0

29 and 30, revealed characteristic

aging curves, 10e0, the maximum embrittlement attained and the time to reach this maximum decreased with increasing aging

(46)

FTCm trie metalioraphic changes that occurred during aging,

it appeared that a twostage precipitation reaction was opera=

tiVeo

The first stage (evident as a mottling of the ferrite)

was responsible for the greater degree of embrittlement9 while

the second stage in which a resolvable precipitate _{was detected9}

was associated with an improvement in properties corresponding

to a rapid Uoveraged condItion0 No attempt at identification

of the precipitate was made _{however., the sequence of changes}

appeared to be the same as reported in _{a recent paper (8)} _The

first stage was identified as the hexagonal g-iron

_{carbde while}

the second stage overlapped the first and _{was associated with a}

Ttharge in the crystal lattice to cementite (orthorhombic Fe1C)0

insofar as microstructural changes _{were concerned9 the}

brittle zone found in the suberitically heated region in _weld

merits was not identified with a precipitation reaction0 _It

was speculated that because the cooling rate In this region _was

such that the ferrite was riot supersaturated to the maximum pos

sible as in water quenched test specimens9 less carbon was avail

able for th _{subsequent aging reaction and the resulting small} amount of precipitation was not detected0

In regard to the changes in transition temperature _arid hardness the results of the subcritical work _{gave the maximum}

embrittiemerit. possible y quenchaginge Any treatment designed to remoue those effects in base plate should be applicable to the suberitically embrittled region of weldments _OveragIng _cr

postheating treatments at about 6500F would minimize _{the embrittle}

(47)

improvement if a slow cool were employed; however, another quench-aging cycle could be initiated upon fast cooling from

(48)

(49)

A 7/16-20 TMDS

f

/2

II

0.300"

00408

,

_600A. R 0.001

CONCENTRIC NOTCH TENSILE SPECIMEN

l-l/2'

_1/I6'

"40

t-l/2

_-1/16

UNNOTCHED

TENSILE SPECIMEN

I / 6" I-1/2" _-1/16"

ECCENTRIC NOTCH TENSILE SPECIMEN

FIG. AI: TEST SPECIMENS

LII4E

or

TEr4SON _ORCL /4' ECENTRICITY 0.030 60° 5/16,4 d(-.- 3/8" »'--3/ 8 5/ l6 -1/2"

.-l/4--.

4-- I / 2

(50)

ADAPTERS LEFT HAND THREAD FIBER IN MAXIMUM TENSION TENSION TENSION NOTCH SPECIMEN

FIG. A2 METHOD OF LOADING TO OBTAIN

1/4

INCH

ECCENFRICITY.

(ECCENTRICItY AND THE

POSITION OF FIXTURES ARE EXAGGERATED.)

EDGES MACHINE-BEVELED

FIG. A3:

PLATE PREPARATiON

ELECTRODE - 3/16' EGOIO PASSES PASSES

FIG. A4 :

(51)

¶ q

REFERENCE SURFACE

FIG. 1x5

:

LOCATION OF ECCENTRIC NOTCH

SPECIMEN AT THE SURFACE LEVEL REFERENCE

SURFACE

FIG. 1x6

LOCATION OF UNNOTCHED AND NOTCHED SPECIMENS AT THE MIDTHICKNESS OF WELDMENT

ROLLING DIRECTION

HEAT TREAT

FIG 1x7

PREPARATION OF CHARPY V-NOTCH AND NOTCH TENSILE SPECIMENS FROM

hOu STEEL PLATE. CENTERLINE OF HEAT AFFECTED CENTERLINE OF PLATE SPECIMENS ZONE

AND FIBER IN MAXIMUM TENSION ON ECCENTRIC

(52)

(53)

C Steel

A Steel Yield Point P si

39,000

37,950

-+6--TABLE B-1

Properties of

Aand

C

Steel Plate*

Chemica1 Composition .

a

han1cal ProDertles Tensile Strength

P si

67,o0

59,910

Noteg _{Both steels were semi-killed, 3/-f" plate in the as-rolled} condition.

*Technical Progress Report of the

Ship

Structure Committee,

Welding Journal, Vol. 13 (July, 198), _{p. 377s.38+s.}

Elongation er Cent

25.5 (8t1 gage)

33°5

(2

gage)

C Steel 02+

O.8

0.012

0.026 _0.05

0.016

0.02 A Steel 0.26

0»5O

0.012 _0.039

_0.03

_0.012

_0.02

i

C Steel 003

0.03 0Q005

0.003

0.009

0.02

0.01

A Steel 0.03 _0.03 _0.006 _0.003

_o.00+

_0.02 _0.01

(54)

Wat

Harnischfeger

D0 C0 Welder

E1ectrode

3/16

E6010

Reversed Polarity

Current

150 amps

Pass i

160 amps

Passes 2-6

Voltage

25 voits

Passes l-6

W9lding Speed

36 in/min0

Pass i

+8 1r/:iin0

Passes 2-6

Electrode

Btirn Off Rate

85 in/min0

Passes 1-6

(55)

BIBLIOGRAPHY

l _{"The Fundamental Factors Influencing the Behavior of}

Welded Structures under Conditions of Multiaxial Stress, and Variations of Temperature, Stress Concentration, and

Rates of Strain", by G0 Sachs,

L0 J0

Ebert, and A0 W0

Dana, First Progress Report, Ship Structure Committee, Serial Number SSC2+, 10 May l91+9

2 _{"The Fundamental Factors Influencing the Behavior of}

Welded Structures under Conditions of Multiaxial Stress,

and Variations of Temperature9 Stress Concentration,

and Rates of Strain", by

L0 J0

Klingler,

L0 J0

Ebert

and W0 M0 Baldwin

Jr09

Second Progress Report, Ship Structure Committee1 Serial Number SSC-3+, 28 November l9+9

3 "The Fundamental Factors Influencing the Behavior of Welded Structures under Conditions of Multiaxial Stress,

and Variations of Temperaturefl, by E0 B0 Evans and L0

_J0

Iüingler Third Progress Report, Ship Structure Committee, Serial Number SSC-5.+9 1+ October

_l952

"The Effect of Subcritical Heat Treatment on the

Transi-tion Temperature of a Low Carbon Ship Plate Steel", by

L

B0 Evans and

L0 J0

Klingler, Fourth Progress Report, Ship Structure Committee, Serial Number SSC-60,

30 October 193

5 "The Effect of Subcritical Heat Treatment on the

Transi-tion Temperature of a Low Carbon Ship Plate Steel", by

E0 B0

Evans and

D0 J0

Garibotti, Fourth Progress Report, Ship Structure CommIttee, Serial Number SSC-6l,

30 October

_i953

6 "Comparison of Various Structural Alloy Steels _{by Means} of the Static Notch-Bar Tensile Test" _by G0 Sachs, L0 J0 Ebert, and W0 F0 Brown9

Jr0

_{Trans0 Am0 Inst0 Mining and}

Met0

Engr0,

_{Vol0 171 (19-7L Po 6o562l}

70 Notched Bar Tensile Test Characteristics of Heat Treated Low Alloy Steeis" by G0 Sachs9

J0 D0

Lubahn, and L0 J0 Ebert9 Trans0

_{Am0 Soc0}

for Metals,

Vol0

33 (l9+),

o 3+0395°

8 _{"The QuenchAging of Iron", by Anna} _L0 _Tson,

_J0

_Nutting,

and

J0

W0 Menter, Journal of the Iron and Steel _Institute,