...

Cranfield_Report_Mat._No.5.pdf

by user

on
Category: Documents
2

views

Report

Comments

Transcript

Cranfield_Report_Mat._No.5.pdf
CRANFIELD REPORT M A T . N O . 5
nr
'I.
.'4
CRANFIELD
INSTITUTE OF TECHNOLOGY
SIMULATED WELD HEAT AFFECTED ZONE
STRUCTURES AND PROPERTIES OF HY 80 LOW ALLOY STEEL
by
G. T. B. KELLOCK, E . SMITH AND A.
R.
SOLLARS
LR
CRANFIELD REPORT MAT. No. 5
February, 1971.
CRANFIELD INSTITUTE OF TECHNOLOGY
SIMULATED WELD HEAT AFFECTED ZONE
STRUCTURES AND PROPERTIES OF HY 80 LOW ALLOY STEEL
by
G. T. B. Kellock, D. A. E. , A . I . M . . C. Eng., A. F . R.Ae.S.
E. Smith, P h . D . . B. Sc, , A.I.M.
A. R. Sollars, B. Sc.
SUMMARY
Single and double thermal cycle simulation of heat affected zone
(HAZ) structures has been used to study the structural and property changes
produced by submerged arc welding of HY 80 steel. The effectiveness of
the temper bead technique and post-weld heat treatment at 650°C have also
been examined,
Marked degradation of impact properties occurs, particularly in the
grain-coarsened region of the HAZ.
Misalignment of a temper bead may
result in a further impairment of notch-toughness. Post-weld heat treatment is shown to be effective in restoring properties to levels similar to
those of the parent plate.
The significance of the results is discussed in relation to submarine
applications.
CONTENTS
PAGE
1.
INTRODUCTION
1
2.
EXPERIMENTAL
2
2. 1
2.2
2
3
3.
4.
RESULTS
5
3. 1
3.2
3. 3
5
6
7
P a r e n t m a t e r i a l and HAZ
P a r e n t plate banding
Simulated weld HAZ s t r u c t u r e s
DISCUSSION
4. 1
4. 2
4.3
4.4
5.
Materials
Procedure
Simulated weld t h e r m a l c y c l e s
I n t e r p r e t a t i o n of C h a r p y data
Notch toughness r e q u i r e m e n t for HY 80 in s u b m a r i n e s
Significance of the r e s u l t s
9
10
11
12
CONCLUSIONS
15
REFERENCES
17
TABLES
1. M e c h a n i c a l p r o p e r t y data for a s - c y c l e d s i m u l a t e d s p e c i m e n s
and the p a r e n t m a t e r i a l
20
2. M e c h a n i c a l p r o p e r t y data for s i m u l a t e d s p e c i m e n s , post cycle
heat t r e a t e d at 650 C.
21
FIGURES
1. T h e r m a l c y c l e s produced in the p a r e n t plate adjacent to the
weld
2. P a r e n t plate
microstructures
3. Heat affected zone h a r d n e s s
survey
4. The n a t u r e of banding in the p a r e n t p l a t e , after t h e r m a l cycling to
a peak t e m p e r a t u r e of 930°C.
5. Effect of t h e r m a l cycling and subsequent p o s t - c y c l e heat t r e a t m e n t on
C h a r p y V-notch i m p a c t data
6. Effect of subsequent t h e r m a l cycling and p o s t - c y c l e heat t r e a t m e n t on
C h a r p y V - n o t c h i m p a c t data for s p e c i m e n s initially cycled to a peak
t e m p e r a t u r e of 1275°C.
7. Effect of subsequent t h e r m a l cycling and p o s t - c y c l e heat t r e a t m e n t on
C h a r p y V notch i m p a c t data for s p e c i m e n s initially cycled to a peak
t e m p e r a t u r e of 930°C.
8. Effect of subsequent thermal cycling and post-cycle heat treatment on
Charpy V-notch impact data for specimens initially cycled to a peak
temperature of 765°C.
9. Charpy V-notch impact (% crystallinity) data for specimens initially
cycled to 12750C.
10. Charpy V-notch impact (% crystallinity) data for specimens initially
cycled to 930°C and 765°C.
11. Simulated weld HAZ structures produced by single and double cycles,
with initial cycling to 1275°C.
12. Simulated weld HAZ structures produced by single and double cycles,
with initial cycling to 930°C.
13. Simulated weld HAZ structures produced by single and double cycles,
with initial cycling to 765°C.
14. Post-cycle heat treated structures of single and double cycled specimens
with initial cycling to 1275°C.
15. Post-cycle heat treated structures of single and double cycled specimens
with initial cycling to 930°C.
- 1 -
1.
INTRODUCTION
After the end of World War II, much effort has been ex|)ended in the
development of high notch-toughness steels, particularly for naval construction.
Considerable experience had been obtained in the marine industry in using the
quenched and tempered STS steel. Test results on low carbon content heats
of this material prompted modification to a low carbon, nickel, chromium,
molybdenum steel which, when quenched and tempered, developed a yield
strength of 550 N/mm.^ (80,000 Ibf/in^) in thicknesses up to 32 mm, with
notch-toughness values, based on the Charpy V-notch impact test, of
68J (50 ft. Ibf) at -84°C.
With composition variation to ensure good hardenability in different plate thicknesses this steel was referred to as HY 80
and produced in plate form to U. S. military specifications MIL - S - 16216D,
E and F, (Refs. 1-3). The present U.S. military specification, MIL - S 16216G (Ref. 4), quotes only one composition range for all plate thicknesses.
In the application of nuclear power to submarine propulsion the hull
diameter had to be increased over that required for conventional power, in
order to encase the necessary equipment. Due to its extremely good strength :
weight ratio, notch toughness at low temperatures and anti-ballistic properties
when compared with other steels, HY 80 has been extensively used in
fabrication of pressure hulls and other components for the U. S. nuclear
submarine fleet. In the U.K., it is now superseding the British QT 35
quenched and tempered low alloy steel in the production of the Dreadnought
class of nuclear powered submarines. The present Ministry of Defence
specification DG/SHIPS/PS/9027, (Ref. 5) requires plates to the U.S. military
specification MIL - S - 16216G, (Ref. 4).
McKee, (Ref. 6) has discussed the
position of HY 80 steel in relation to submarine design practice and Heller et al,
(Ref. 7) in an excellent review paper, have evaluated its use as a structural
material for nuclear powered submarines.
Since welding is extensively used in submarine fabrication it is
important to be aware of its effect on the structure and properties of the
parent material.
In low alloy steels the fracture toughness of the heat
affected zone (HAZ) may be severely impaired by welding, resulting in
increased susceptibility to brittle fracture and possible cold cracking.
(Refs. 8-10).
Fracture toughness of the weld HAZ is dependent upon the
types of structures produced by the weld thermal cycles and this is controlled
to some extent in HY 80 steel by limiting the heat input to the range 1.2 2. 2. k J / m m (30 - 55kJ/in. ) and using a preheat and interpass temperature in
the range 120°C-150°C, (Ref. 5).
These conditions favour the formation of
structures consisting predominantly of autotempered martensite with perhaps
some lower and upper bainite. In this class of steels this type of structure
has been shown to possess better fracture toughness properties than those
produced by heat inputs above this range, which contain a greater proportion
of upper bainite, (Ref. 11).
However, even when the heat input and preheat
temperature requirements are adhered to, there is still some doubt as to
whether the notch toughness of the weld HAZ provides adequate resistance
to brittle fracture during service. In view of the fact that HY 80 has been
used in marine construction since 1952, it is rather surprising that very
little infornaation regarding its structure as produced by standard heat
treatment procedures, let alone the complexities introduced by the weld
thermal cycle, has been published. Apart from the work of Dolby (Ref. 12)
on weld HAZ structures produced by heat inputs lying outside the specified
range, no electron microscope study of HY 80 steel has been reported
-
2 -
in the literature.
Current U. 5. Navy and Ministry of Defence (Navy) specifications call
for the application of the tempering bead technique for all multi-run welds in
HY 80 steel. Its object is to reduce the high hardness and relatively low
fracture toughness associated with undesirable structures in the HAZ,
particularly in the region of the toe of the weld. The technique is of primary
importance in respect of the final surface runs since sub-surface layers
generally have their HAZ tempered by subsequent layers. The positioning of
the tempering bead must be carefully controlled, since if it is not centered
properly with respect to the two HAZ boundaries in contact with the surface,
a significant tempering effect cannot be obtained in these parts of the HAZ.
In fact, it is possible to envisage a multiplicity of combinations of thermal
cycles in the edge bead HAZ due to the application of a tempering bead. In
an effort to minimise the effect, the welding specification for HY 80 steel
requires that approximately 3. 2mm of the edge beads should be exposed after
deposition of the temper bead. Oldridge (Ref. 13) showed that there are such
immense practical difficulties involved in maintaining accurate positioning of
the tempering bead during deposition that its efficiency in providing a tempering
effect in the HAZ of the last deposited layers, adjacent to the parent material,
may be considered to be open to conjecture.
The Ministry of Defence specification (Ref. 5) considers the post-weld
heat treatment of HY 80 weldments for the purpose of stress relief to be
neither necessary nor desirable. If, for any reason, stress relief is
required it is undertaken at 550°C jl 15°C, this temperature being held for
1 hour per inch of thickness of the thickest member of the weldment. The
potential benefit to be derived from post-weld heat treatment in t e r m s of
restoration of good notch toughness in the HAZ appears to have been
neglected in the past. It is the authors' opinion however that such treatment
could be very beneficial to the weld HAZ and perhaps also to the weld metal,
and thus remove the uncertainty concerning the effectiveness of the tempering
bead technique.
This present research has a three-fold object :(a)
to establish for a given heat input condition and preheat temperature
the nature of the weld HAZ due to a single-run weld.
(b)
by employing double cycle simulation techniques using the same welding
conditions a s in (a), to indicate the general nature of tempering in a
pre-existing HAZ due to the heat flow from a subsequent weld run in
a multi-run weldment and to predict the effectiveness of the temperbead technique.
(c)
to ascertain the effectiveness of a post-weld heat treatment at 650^0
for 1 hour on the structures produced by (a) and (b).
2.
EXPERIMENTAL
2. 1.
. Materials
The 38inm. thick HY 80 steel plate had the following chemical analysis,
- 3 -
which conformed to specification MIL - S-16216G
c
Mn
Si
S
P
Ni
Cr
Mo
V
Al
N
0.16
0.32
0.30
0.017
0, 006
2.54
1.31
0. 28
0.01
0. 015
0.01
T e n s i l e and Charpy V-notch i m p a c t t e s t s conducted on s p e c i m e n s
obtained from the plate m i d - t h i c k n e s s yielded the following r e s u l t s , which
conformed to specification MIL - S - 16216G.
Specimen
Orientation
Charpy Energy*
at -84°C
U.T.S.
Yield Strength
J
ft. Ibf
N/mm
tonf/in
N/mm
Transverse
107
79
576
37.4
719
46,7
Longitudinal
127
93
594
38. 6
735
47.8
tonf/in
* notched in the through t h i c k n e s s d i r e c t i o n
2. 2.
Procedure
2, 2. 1. P r e p a r a t i o n and e x a m i n a t i o n of weld.
A b e a d - o n - p l a t e weld w a s produced by a subnaerged a r c welding unit
using 4. 76mm (0. 1875 in) d i a m e t e r mild s t e e l filler w i r e with the following
conditions:-
Current
(amps)
Voltage
T r a v e l Speed
mm/sec
500
30
7.05
Preheat
Temp.
(°C)
120
Heat Input
kJ/mm
2.13
A t r a n s v e r s e section through the weld HAZ w a s examined using optical
metallography, and h a r d n e s s d e t e r m i n a t i o n s w e r e m a d e at 0. 5mm i n t e r v a l s ,
s t a r t i n g from the fusion boundary, using a Zwick h a r d n e s s t e s t e r and l o a d s of
0 . 5 kg and 5 kg.
2. 2, 2. Simulation of weld HAZ s t r u c t u r e s
Weld t h e r m a l cycle s i m u l a t i o n w a s c a r r i e d out on an a p p a r a t u s
designed and built at Cranfield in which a m a t e r i a l blank of suitable d i m e n s i o n s
i s heated by v i r t u e of i t s own r e s i s t a n c e to the p a s s a g e of an e l e c t r i c c u r r e n t
and cooled by the flow of w a t e r through hollow b r a s s c l a m p i n g b l o c k s .
The
Cranfield s i m u l a t o r h a s been d e s c r i b e d in d e t a i l by Clifton and G e o r g e , (Ref, 14).
The d i m e n s i o n s of the blanks enriployed in t h i s s i m u l a t i o n study w e r e
-
4
-
10. 7mm X 10. 7mm x 10. 7mm x 83mm and were machined from the midthickness of the plate axially transverse to the rolling direction. Coward,
(Ref. 15), showed that the application of restraint by rigid clamping of the
blanks during simulation of a similar steel did not affect the performance
in the Charpy V-notch impact test. In this present study all blanks were
allowed axial freedom of movement during thermal cycling.
Four thermal cycles with peak temperatures of 1275°C, 930°C,
765°C and 650°C, corresponding to those experienced by the grain coarsened,
grain refined, intercritical, and subcritical regions of the weld HAZ
respectively, were used for simulation. The 1275°C and 7650C peak
temperature cycles were measured directly by Smith (Ref. 16) in the HAZ
of a submerged arc bead-on-plate weld using the same set of conditions
described in section 2. 2. 1. The technique has been described by Coward,
(Ref. 15).
The 930°C and 650°C peak temperature cycles were computed
by Kellock (Ref. 17) from a series of programs incorporating the results
of Smith, (Ref. 16).
These programs included functions to account for the
release of latent heat from the solidifying weld pool and the variation of
thermal conductivity with temperature. The computed thermal cycles
correlated well with the measured cycles. The four thermal cycles are
shown in Fig. 1.
The following thermal cycles and combinations of thermal cycles were
used to examine the HAZ structures of single and multipass weldments:-
Peak Temperature, °C
First Cycle
Second Cycle
Peak Temperature, °C
First Cycle
Second Cycle
1275
1275
765
930
1275
650
765
930
765
1275
1275
930
650
1275
930
765
650
2. 2, 3. Post-cycle heat treatment
Half the total number of simulated blanks were given a post-cycle
heat treatment at 650°C for 1 hour, the specimens being placed in the
furnace at 250°C and heated to 650°C in 50 minutes.
After one hour at
temperature the specimens were removed from the furnace and allowed to
cool in still air.
2. 2. 4, Mechanical testing of simulated HAZ structures
Ten standard Charpy V-notch impact specimens were prepared from
the simulated blanks for each condition studied with the notches machined in
1
- 5 -
the control thermocouple position and in the through thickness direction of the
original plate. Impact transition curves were determined by testing between
- 196°C and 40°C. A mixture of iso - pentane and liquid nitrogen was used
for tests below room temperature and hot water for tests above room temperature.
Test temperatures were estimated to be accurate to 't 3°C.
Three Hounsfield tensile test specimens with a gauge length of 7. 6mm
and 4. 5mm diameter were prepared from the simulated blanks for each condition, with the gauge length accurately positioned within the thermally cycled
zone at the centre of each blank. The specimens were tested on a standard
Instron tensile machine using a strain rate of approximately 3 x 10 " sec. " .
The values of 0. 2% proof s t r e s s , U. T. S. , and reduction of area were recorded.
One simulated blank from each condition was sectioned through the
centre of the thermally cycled zone and the hardness determined using the
Zwick hardness tester with a load of 5kg.
2. 2. 5. Metallographic examination of simulated HAZ structures
The sections used for hardness determinations were subsequently
prepared for optical metallographic examination, all structures being
successfully etched in 2% nital. P r i o r to the preparation of carbon extraction
replicas for electron microscopy, the specimens were given a further etch in
2% nital in order to ensure the production of satisfactory replicas.
The
replicas were extracted with a 10% nital solution and examined in a Siemens
Elmiskop lA electron microscope.
2. 2. 6. Examination of banding
Band width measurement and 0. 5kg load hardness surveys were made
on specimens which had been simulated using the 930°C peak temperature
cycle. Specimens in this condition were chosen because they gave the best
inter-band optical contrast after etching. Electron probe microanalysis was
carried out at the Welding Institute to determine the nature of the microsegregation within the HY 80 steel plate.
3,
RESULTS
3, 1.
Parent material and weld HAZ
The parent plate microstructure. Fig. 2, was typical of a quenched
and tempered low alloy steel in that it contained a dispersion of carbides in
a ferrite matrix. Three carbide types are apparent in Fig. 2. There is a
Widmanstatten a r r a y of fine rod-shaped particles situated within ferrite subgrains, having a length of approximately 0. lium. In addition spheroidised
carbides approximately 0. 5^m in diameter and some larger, more elongated
particles about l/.im long are situated mainly at ferrite sub-boundaries.
Dolby (Ref. 18) has indicated that the grain boundary carbides consist of both
M7C3 and Fe3C while the fine intragranular precipitates are invariably Fe C.
Examination of the HAZ shows that it consists of three virtually
-
6 -
distinct regions :(a)
the grain coarsened region extending up to approximately 0. 6mm
from the fusion boundary,
(b)
the grain refined region, between 0. 6mm and 2. 5mm from the
fusion boundary, and
(c)
the intercritical region, between 2.5mm and 3.0mm from the
fusion boundary.
Due to the high temperatures experienced within the grain coarsened
region, rapid austenite grain growth occured during thermal cycling but the
austenite grain size decreased, with decreasing peak temperatures, as the
distance from the fusion boundary increased. Within the prior austenite
grains the structure was acicular but the details could not be resolved
using optical microscopy. The boundary separating the grain coarsened
and grain refined regions is rather diffuse. In the grain refined region
the structure consisted of a fine irresolvable ferrite-carbide aggregate
with a fine, and decreasing, prior austenite grain size. The intercritical
region contained an increasing proportion of untransformed ferrite as the
distance from the fusion boundary increased and the transformed regions
appeared to consist of a fine carbide aggregate, once more not capable
of being resolved using conventional light microscopy.
The results of the 5kg, and 0. 5kg. load hardness traverses are
shown in Fig. 3. The 5kg. results show a continuing increase in hardness
on moving towards the fusion boundary reaching a maximum value of
430 HV5 at the fusion boundary. Large fluctuations superimposed on the
same general trend were obtained with the 0. 5kg. load tests, the fluctuations
being particularly noticeable in the grain refined region. This is thought to
be due to the banded structure of the parent plate resulting from alloying
element segregation.
3. 2.
Parent plate banding
The existence of banding was not very evident in the parent plate or
after cycling to a peak temperature of 1275°C. However, after cycling to
765°C and 930°C, banding was very noticeable at low magnifications.
Banding in a 930°C peak temperature cycled specimen as revealed by
etching in 2% nital is shown in Fig. 4. Measurement of band width and
hardness gave the following r e s u l t s : -
Band type
Band width (mm)
HV 0. 5
Mean
Std. Dev.
Mean
Light
etching
0. 056
0. 031
436
Dark
etching
0. 195
Std. Dev.
HV5
1
Mean
Std. Dev. 1
7
353
0.044
311
11
17
-
7 -
E l e c t r o n probe m i c r o a n a l y s i s of t h r e e typical bands gave the following
results:Weight % in band
Element
Dark etching
Manganese
Nickel
Chromium:
Molybdenum
3. 3.
0.
2.
1.
0.
41
20
70
20
Light etching
Dark etching
0. 59
3. 50
2.40
0. 26
0.39
2.31
1.20
0. 18
Simulated weld HAZ s t r u c t u r e s
The Charpy V-notch i m p a c t t e s t r e s u l t s on single and double cycled
simulated s p e c i m e n s , with and without p o s t - c y c l e heat t r e a t m e n t , a r e given in
Figures 5 - 1 0 .
F i g u r e s 5 - 8 show the e n e r g y absorption - t e m p e r a t u r e
c u r v e s and F i g u r e s 9 and 10 the f r a c t u r e a p p e a r a n c e - t e m p e r a t u r e c u r v e s .
F i g u r e 5 a l s o includes the e n e r g y t r a n s i t i o n c u r v e s for the p a r e n t plate in
both the t r a n s v e r s e and longitudinal d i r e c t i o n s .
The r e s u l t s obtained from t e n s i l e and h a r d n e s s t e s t i n g a r e given in
T a b l e s 1 and 2 for the a s - c y c l e d and p o s t - c y c l e d heat t r e a t e d conditions
respectively.
Data for the p a r e n t plate i s a l s o included in Table 1.
The
s t r u c t u r e s produced by the s i m u l a t e d weld t h e r m a l c y c l e s and p o s t - w e l d heat
t r e a t m e n t a r e shown in F i g u r e s 11 - 15.
3. 3, 1.
Single cycle s i m u l a t i o n
A single cycle to a peak t e m p e r a t u r e of 1275°C produced a s t r u c t u r e
c o n s i s t i n g predominantly of a u t o - t e m p e r e d lath m a r t e n s i t e . F i g . 11a, with
s m a l l a m o u n t s of u p p e r and l o w e r b a i n i t e . F i g . l i b .
There was a considerable
i n c r e a s e in proof s t r e s s , U . T . S . and h a r d n e s s and d e c r e a s e in ductility when
c o m p a r e d with p a r e n t plate (Table 1), while the Charpy v a l u e s w e r e s e v e r e l y
affected (Fig. 6).
After a single cycle to a peak t e m p e r a t u r e of 930 C slightly b e t t e r
e n e r g y a b s o r p t i o n was obtained.
F i g u r e 12a shows a typical s t r u c t u r e
c o n s i s t i n g of a v e r y fine f e r r i t e - g l o b u l a r c a r b i d e a g g r e g a t e and a r e a s of
auto-tempered martensite.
In F i g . 12b the halo affect around s o m e c a r b i d e
p a r t i c l e s s u g g e s t s that not all the c a r b i d e w a s taken into solution during the
thermal cycle.
The m e c h a n i c a l p r o p e r t i e s , Table 1 , show a c o n s i d e r a b l e
i n c r e a s e in proof s t r e s s , U . T . S . and h a r d n e s s and d r o p in ductility c o m p a r e d
with p a r e n t plate.
The C h a r p y p r o p e r t i e s . F i g s . 5 and 10a show a c o n s i d e r able d r o p in e n e r g y absorption c o m p a r e d with p a r e n t p l a t e .
A single cycle of 765 C r e s u l t e d in p a r t i a l a u s t e n i t i s a t i o n .
A high
d e n s i t y of c o a r s e c a r b i d e w a s evident in sonae a r e a s of the s t r u c t u r e . Fig. 13a,
Many of the fine rod-shaped carbides present in the parent plate were taken
into solution and retained there on cooling. The nature of the transformed
a r e a s was difficult to identify but they could have been bainitic or martensitic
formed during the cooling of carbon-enriched austenite. There was a small
increase in proof stress, U . T . S . and hardness (Table 1) although the Charpy
properties were similar to those of the parent plate in the transverse
direction (Fig. 8).
3. 3. 2.
Double cycle simulation
A second cycle to 1275°C after an initial 1275°C cycle produced
only slight changes in structure. The prior austenite grain size was somewhat
larger and more well-defined upper bainite colonies were present. The m.ost
significant change in mechanical properties was the resultant fall in upper
shelf energy from 50J to 37J(37 ft. Ibf. to 27 ft. Ibf. ) as shown in Fig. 6.
A second cycle to 930°C peak temperature after an initial 1275°C
cycle produced a structure composed mainly of auto-tempered martensite
(Fig. l i e ) in which the laths were much finer than the structures produced by
single or double cycling to 1275°C peak temperature. Some upper bainite
was also present. The prior austenite grain size was of duplex type, showing
a larger grain size due to the initial 1275°C cycle within which was a much
finer one due to the lower temperature second cycle. The proof stress,
U . T . S . and hardness were increased slightly by the second cycle (Fig. 6)
probably due to grain refinement.
A noticeable improvement occurred when a second cycle to a peak
tenaperature of 765°C followed the initial 12750C cycle. Fig. l i d shows a
virtually continuous constituent in the prior austenite grain boundaries due to
the initial cycle. As seen in Fig. l i e , this constituent is composed of small
autotempered martensite units. The areas surrounded by this matrix consisted of ferrite and course carbide. There was a marked drop in proof
s t r e s s , U, T, S, and hardness (Table 1) and some improvenaent in Charpy
performance (Figs, 6 and 9b).
Following an initial 1275°C cycle with a cycle to 650°C peak
produced a tempering effect. Carbides precipitated at autotempered
martensite lath boundaries (Fig. 14). Original carbides in the autotempered
martensite show some degree of growth. Precipitation had also occurred
within the martensite laths and in the prior austenite grain boundaries. The
proof s t r e s s , U . T . S . and hardness were lower than after the initial 1275°C
cycle and Charpy performance improved, (Pigs. 6 and 9b). It should be
noted, however, that this second cycle produced slightly higher strengths
than a second cycle to 765°C.
When an initial cycle to 930°C peak was followed by a cycle to
765°C, only partial austenitisation occurred and the resulting structure
(Fig. 12c) consisted of fine-grained ferrite and regions of a fine ferritecarbide dispersion. There was also some evidence of only partial solution
of carbides. The mechanical properties showed a marked drop in proof
s t r e s s , U . T . S . and hardness (Table 1) and a slight improvement in Charpy
performance (Figs. 7 and 10a).
- 9 -
A second cycle to 650 C after a 930°C cycle produced a tempering
effect as shown by an increased density of carbides, although the distribution
was inhomogeneous (Fig. 12d).
The strength properties were higher than for
the parent plate in the transverse direction while the energy absorption
properties, although improved by the second thermal cycle, were still
markedly inferior to the parent plate (Fig. 7).
Specimens initially cycled to 7650C and subsequently cycled to 650°C
showed structures resembling the parent plate (Fig. 13b) probably due to r e precipitation of carbides taken into solution during the initial cycle. The
presence of small rod-shaped particles in some regions (Fig. 13c) indicates
that some auto-tempered martensite was produced by the first cycle. The
mechanical properties were very similar to the transverse properties of the
parent plate.
3. 3. 3.
Post-cycle heat treatment
The post-cycle tempering at 650°C for 1 hour produced structures
consisting essentially of temper carbides in a ferrite matrix. As for the
parent plate, three types of carbides were identifiable viz. fine rod shaped
particles and coarser globular and elongated carbides. The distribution of
these types varied within groups of specimens.
(a) Specimens initially cycled to 1275°C generally contained fine
rod-shaped and coarse elongated carbide particles with only a small
proportion of globular particles. The coarse elongated carbides were sited
mainly at prior austenite grain boundaries and between individual ferrite
laths, as shown in Fig. 14a for a specimen originally given a single cycle
to 1275°C, Fine rod shaped particles occurred in the tempered martensite
laths as shown in Fig. 14b for a specimen given two cycles to 1275°C peak.
The structure produced by tempering after a second cycle to 765°C is
particularly interesting. Figs. 14c and 14d, Martensitic a r e a s delineating
original prior austenite grain boundaries have been tempered to produce a
network of dense, mainly elongated carbides, surrounding a r e a s containing
Utile precipitate. The proof s t r e s s , U . T . S . and hardness of all these
structures were higher than the parent plate (Table 2), and a considerable
improvement in Charpy performance resulted, (Figs, 6 and 9b).
(b) Specimens initially cycled to 930°C contained mainly globular
carbides, with some elongated carbide particles (Figs. 16a and 16b). The
temper carbides were fairly uniformly distributed throughout a ferrite
matrix and tensile, hardness and Charpy properties were close to those of
the parent plate (Table 2, Figs. 7 and 10a).
(c) Specimens initially cycled to 765°C and tempered at 650°C
showed structures very similar to those of the parent plate and the mechanical
properties were also very similar (Figs. 8 and 10b),
4.
DISCUSSION
4. 1.
Simulated weld thermal cycles
The correlation between the measured and computed thernaal cycles
- 10 -
with those produced by the simulation equipment was generally very good.
However, difficulty was experienced in reproducing the sharp peak in the
1275°C cycle, the electrical characteristics of the simulation equipment
producing a broader peak. This could result in the simulated samples
showing a slight increase in the prior austenite grain size and a more
homogeneous austenite compared with the corresponding region in a weld
heat affected zone.
Inflections in the cooling curves were observed for most of the
simulated specimens in the temperature range 550°C - 400°C (Ref. 17) and
have been attributed by Inagaki et al (Ref. 19) to the exothermal transformation of austenite.
4. 2,
Interpretation of Charpy data
The successful performance of a welded steel structure in
avoiding brittle fracture will depend upon the fracture toughness
characteristics of the parent material, the HAZ and the weld metal. Low
fracture toughness in any of these regions may lead to complete failure of
the structure by catastrophic brittle fracture since each of these regions
provide a continuous path along which such fractures may propagate.
The Charpy test has been the most widely used method of
assessing fracture toughness in structural steels. Interpretation of the
results can, however, be misleading unless it is based on a vast amount of
experience or else correlated with some other test procedure which gives a
more realistic appraisal of service performance. This is due partly to
the fact that the energy measured by the test does not permit conditions
for initiation and propagation to be distinguished or the critical s t r e s s and
strain parameters responsible for fracture to be separated. The test also
uses impact loading conditions and thus the results will not be directly
applicable to static loading conditions.
Pellini et al (Refs. 20 - 24), however, have developed a procedure
for the fracture-safe design of steel structures based on a fracture analysis
diagram which takes account of flaw size, s t r e s s level and service temperature.
The procedure defines three critical transition temperatures:
(a) The nil-ductility transition (NDT) temperature below which the
steel loses all ability to deform in the presence of a sharp crack.
(b) The fracture transition elastic (FTE) temperature below which
brittle fracture will run through elastically loaded material.
(c) The fracture transition plastic (FTP) temperature above which
only shear tearing is possible irrespective of severity of plastic loading.
The temperature intervals between these critical transition
temperatures are remarkably similar for a wide range of steels including
mild steels, high strength steels, cast steels, and 12% Cr steels (Ref. 21).
so that it is possible to establish the NDT temperature by the use of the
explosion bulge test. The proposed relationships between the three critical
- 11 -
transition temperatures are as follows:
FTE
= NDT
+
33°C
(+ 5°C)
FTP
= NDT
+
72°C
(t 10°C)
Pellini et at (Refs. 20 - 24) have shown that a correlation can be
established between the NDT temperature measured by the drop weight or
explosion bulge tests and the Charpy energy level at this temperature. It
must be emphasised, however, that this correlation is unique and is only
valid for other steels of similar composition, manufacture, and heat treatment.
When this correlation has been established Charpy data can be used
to predict the three critical transition temperatures and this enables limitations to be placed on the steel which, if adhered to, will ensure that brittle
fracture does not occur.
For welded structures the concept of inherent flaws is considered
to be quite realistic since they may arise from lack of penetration, lack of
fusion, slag inclusion, porosity or cracking. Pellini and Puzak (Ref. 23)
have provided extensive failure and structural test data to support the
validity of the procedure.
4. 3.
Notch toughness requirement for HY80 in submarines
Pellini and Srawley (Ref. 22) have discussed the notch-toughness
requirements of HY80 in submarine structures, two types of structure being
considered.
For non-combatant submarines the lowest service temperature
should be above the FTE temperature, so that fracture a r r e s t protection is
provided even for parts of the structure where tensile s t r e s s e s are close to
yield point levels.
For combatant submarines which are required to withstand explosive loading, resulting in extensive plastic deformation of the hull
structure, the FTP temperature is required to be below the lowest service
temperature.
For HY80 parent material it was established that the FTE and F T P
temperatures are approximately -68°C and-40°C respectively and that the
NDT temperature (-90^0) correlates well with the 40-45 ft. lb Charpy
temperature (Ref. 21).
Winn (Ref. 25) reported a similar correlation for
QT 35 on specimens cut in the rolling direction and this has been confirmed
in recent work at Cranfield (Ref. 26).
Thus, the fracture toughness
requirements for the two types of submarines described above can be stated
as follows, based on safe operation at -10 C which is below the lowest
temperature encountered during submerged operation :(a)
68J (50 ft. lb. ) minimum at -40°C for non-combatant submarines
(b)
68J (50 ft. lb. ) minimum at -68 C for combatant submarines.
A similar attempt to evaluate the behaviour of the HAZ is much
more difficult.
It has been suggested that fractures which initiate in the
HAZ will propagate into either parent material or weld metal since the
welds are normally of V or double V geometry and thus produce a HAZ
slanted 45 - 60 C with respect to the usual s t r e s s vector.
- L
Therefore, protection against fracture initiation only is indicated. However
there is contradictory evidence (Ref. 33) that propagation can occur solely
through the HAZ in such a preparation. Nevertheless, because s t r e s s relieving is not normal practice for submarine structures, high yield
residual s t r e s s e s will be present in the HAZ so that operation above the
NDT temperature is required to protect against fractures initiating from
small flaws for a wide range of nominal s t r e s s e s .
At the present moment it is not possible to state categorically
what the notch toughness requirements are in t e r m s of Charpy data since
the correlations for the types of structure produced in the HAZ have not
been determined. However, Puzak andPellini (Ref. 21) have shown that for
structures of relatively high hardness which have a maximum Charpy energy
of about 41J (30 ft. lb) or less, the NDT temperature is generally related
to the 14 - 20J (10 - 15 ft. lb. ) Charpy temperature. Recent work at
Cranfield (Ref. 26) on plates of QT35 steel quenched to produce structures
similar to those found in the grain coarsened region of the HAZ indicates
that the NDT temperature is equivalent to the 23-31J (17-23 ft. lb.) Charpy
temperature. Thus, it would appear that the fracture toughness requirement
for grain coarsened HAZ structures in HY80 will be in the region of 34J
(25 ft. lb. ) at -10°C.
4.4.
Significance of the results
4. 4. 1,
Parent material
The HY80 used in the present work easily met the 68J (50 ft. lb. )
minimum Charpy energy at 68°C considered necessary for the more severe
conditions required of combant submarines. Fig. 5 shows that at this
temperature the energy levels were approximately 149 and 115J (110 and
85 ft. lb. ) respectively for the longitudinal and transverse directions. It is
of interest to conapare these values with those reported for QT 35 since
HY80 has recently replaced this steel in the U.K. submarine programme.
Smith and Apps (Refs. 26 and 27) have determined equivalent Charpy energy
values at -60°C of 90 and 61J (66 and 45 ft. lb, ) respectively for the
longitudinal and transverse directions of QT35. Thus, the replacement of
QT35 by HY80 in the U.K. would appear to be justified. Dolby (Ref. 12)
has reported a crack opening displacement transition temperature some
20°C lower for HY80 compared with QT35.
The superior notch-toughness
of HY80 has been attributed to the following factors (Ref. 18) :(a)
a smaller grain size giving greater resistance to crack initiation
and propagation
(b)
the higher Ni content of HY80 which improves the basic properties
of the ferrite
(c)
an increased density of carbides and different size distribution
increasing resistance to crack propagation
(d)
A lower dislocation density within the subgrains
The examination of banding described in section 3. 2 is interesting
- 13 -
in that the light etching bands were found to contain a higher concentration
of aUoying elements than the dark etching bands. This is in direct contrast
to another examination of banding in HY80 (Ref. 18), in which the dark etching
bands were found to be richer in alloy content. However, in the present work
banding was examined in material that had been thermally cycled above the
upper critical temperature so that the apparent anomaly can be explained by
the fact that the alloy-enriched bands transformed to martensite during
Cooling and thus appeared lighter than the alloy-denuded bands which transformed to a higher temperature transformation product containing more
carbide precipitates.
4.4.2.
Single pass weld HAZ
The single cycle simulation technique enables a study to be made
of the HAZ in single pass welds which also may be considered to represent
the worst case for multi-pass welding where HAZ hardnesses approaching
those in single pass welds can be found in the final edge bead HAZ.
The results. Fig. 6, show that the transformed HAZ has significantly
lower energy absorption and upper shelf energy compared with the parent plate,
these changes becoming more marked on approaching the fusion boundary. A
similar pattern of results has been reported on other quenched and tempered
low allow steels (Refs. 11, 15, 27-30),
Dolby (Ref, 12) has also shown a
significantly reduced resistance to fracture initiation in the transformed HAZ
of HY80.
This reduction in fracture toughness is accompanied by marked
increases in proof s t r e s s and hardness (Table 4). The high hardness
associated with predominantly martensitic structures, particularly in the
grain coarsened HAZ, indicates that precautions need to be taken to keep
hydrogen to a minimum in order to avoid HAZ cold cracking.
The Charpy data for the grain coarsened HAZ shows an energy
level slightly in excess of 41J (30 ft. lb. ) at -10 C which is above the 34J
(25 ft, lb. ) level considered necessary to indicate an NDT temperature below
-10 C,
Thus, the probability of fractures initiating from small flaws in
the HAZ would appear to be small although more work is necessary to
confirm the correlation between the NDT temperature and the Charpy energy
level at this temperature and to investigate the variation in properties
between different casts of HY80.
A similar investigation into the weld HAZ properties of QT35
steel (Ref. 27) indicated a Charpy energy level of about 30J (22 ft. lb. ) at
-10 C for the same welding conditions used in the present work and would
thus suggest that QT35 is border-line with respect to NDT requirement.
This would appear to be further confirmation of the supjeriority of HY80
over QT35.
Dolby (Ref. 12) also showed a lower resistance to fracture
initiation in the grain coarsened HAZ for QT35 compared with HY80 when
initiation occurred by quasi-cleavage although the position was reversed at
higher temperatures when the initiation mode was by stable tearing. The
superiority of HY80 over QT35 can be attributed to the increased hardenability of HY80 giving predominantly martensitic structures in the t r a n s formed HAZ, the increased Ni content improving the basic properties of
the structure, and the smaller prior austenite grain size caused by the
pinning action of aluminium nitride particles. However, there is evidence
- 14 -
to show that the lower Mn/S ratio and the higher Ni content of HY80 increases
the susceptibility to hot cracking, (Ref. 18).
The improved Charpy performance observed in the grain refined
HAZ compared with the grain coarsened HAZ in HY80 can be attributed to
the much smaller prior austenite grain size and lower hardness observed
in the former.
4. 4. 3.
Multi-pass weld HAZ.
The double cycle simulation technique enables predictions to be
made regarding the HAZ of multi-pass welds where tempering effects will
be expected to improve fracture toughness and reduce hardness by reducing
dislocation densities and coarsening the carbides. Figs. 7 and 8 and
Table 1 show that this is true for regions experiencing a tempering cycle
below the upper critical temperature. These regions easily satisfy the NDT
requirement at -10 C and have much reduced hardness. However, a further
impairment of fracture toughness coupled with hardnesses in excess of
400 HV5 in the grain coarsened HAZ is indicated for regions experiencing a
second high peak temperature cycle, with Charpy energy levels reduced to
about 34J (25 ft. lb. ) at -10 C, which is also the upper shelf energy level.
If such regions exist in a multi-pass weld they would be borderline with
respect to the NDT requirement at the minimum service temperature.
Such a situation may arise either in regions where two fusion boundaries
come into close contact and subsequent weld deposits may then be too far
removed to cause effective tempering or by a temper bead positioned too
close to a final edge bead HAZ. This may be particularly significant in
the final edge bead HAZ which is often a region of appreciable s t r e s s
concentration. Oldridge (Ref, 13) has demonstrated the uncertainties
involved in accurate positioning of the temper bead,
4,4.4.
Post-weld heat treatment
Figs. 7 and 8 show that post-weld heat treatment at 650 C for
1 hour is a much more effective and reliable method of improving Charpy
performance of the weld HAZ in HY80 than relying on tempering effects
occurring in multi-pass welds. Energy absorption is restored quite closely
to parent material levels and additional benefit would be derived from s t r e s s relief. Such structures may be expected to provide complete protection
against brittle fracture for the specific application to submarine structures.
The possibility of delayed HAZ cold cracking would also be eliminated as is
evidenced by the low hardnesses of these structures (Table 2).
There a r e , however, a number of questions which need to be
answered before post-weld heat treatment can be considered as a practical
proposition :1.
Is the strength level of the parent material impaired?
Even
with a localised post-weld heat treatment it is not possible to
avoid reheating parts of the parent material close to the heat
treatment temperature and it is essential that the strength level
of the parent material is maintained, otherwise the load carrying
capacity will be reduced.
- 15 -
2.
Are the weld metal properties impaired?
Then?- is evidence to
suggest that post-weld heat treatment in the temperature range
510 - 650 C may actually impair the notch toughness of the weld
metal (Refs. 18 and 31). The overall effect may be a compromise
between the benefit derived from reduction in residual s t r e s s e s and
the impairment of notch toughness. This will depend on the type
of weld metal (i. e. electrode and flux combination).
3.
Will the weldment suffer from temper brittleness?
There is not
much information on this topic although at least one person has
reported that HY80 is not susceptible to this form of embrittlement,
(Ref. 21). This form of embrittlement is generally associated with
slow cooling or holding in the temperature range 620 - 400 C and
is a function of chemical composition. Increasing Ni, Cr, and Mn
promote susceptibility, while Mo has a counteracting effect.
Until these questions are answered it is not possible to state whether postweld heat treatment will be beneficial to the weldment as a whole. There is
an economic penalty to be paid for carrying out post-weld heat treatment so
that it is necessary to be able to show substantial benefit before the procedure
can be recommended. Heat treatments at other temperatures and times need
also to be examined in order to determine the optimum combination and the
degree of control which would be required. Accumulated experience with
structures and pressure vessels and information from full scale tests have
supported the position that HY80 weldments generally do not require s t r e s s relieving (Ref. 31).
The satisfactory performance of the many welded but
non-stress relieved structures and vessels in HY80, stands as undeniable
evidence that the steel generally does not require post-weld heat treatment
(stress-relieving). The present work, although confined to a single cast of
HY80, has more or less substantiated this view.
5.
CONCLUSIONS
1.
A marked impairment of fracture toughness, as measured by the
Charpy V notch impact test, and hardnesses in excess of 400 HV5 occur in
the grain coarsened and grain refined regions of single-pass weld HAZs in
HY80 using a recommended heat input and preheat temperature of 2.1 kJ/mm
(54 kJ/in) and 120 C respectively.
The structures produced are p r e dominantly martensitic in character and the mechanical properties
deteriorate with decreasing distance from the fusion boundary.
2.
Little impairment of mechanical properties occurs in regions of
the weld HAZ experiencing peak temperatures below the upper critical.
3.
Tempering effects occurring in the HAZ of multi-pass welds
generally improve Charpy performance and reduce hardness. However,
some regions may actually have similar or reduced Charpy performance with
hardnesses still in excess of 400 HV5.
This is particularly true of the
final edge bead HAZ where effective tempering relies on accurate positioning
of the temper-bead.
- 16 -
4.
Post-weld heat treatment at 650 C for 1 hour restores Charpy
performance of all HAZ structures close to parent material levels and
reduces hardness to well below 350 HV5. This is a much more effective
and reliable method of improving HAZ properties than relying on tempering
effects occurring in multipass welds. Other factors, however, need to be
investigated before the treatment can be considered as a practical
proposition.
5.
Tentative correlations with NDT temperatures measured by the
drop weight or explosion bulge tests and applied to the Pellini fracture - safe
design philosophy for submarine structures indicate little risk of brittle
fractures initiating from small flaws in non-tempered weld HAZs in HY80.
More information, however, is needed to confirm this.
6.
HY80 is a suitable replacement for QT35 steel in submarine
structures.
- 17 -
REFERENCES
Steel P l a t e , Alloy, S t r u c t u r a l , High Yield Strength, M i l i t a r y Spec. ,
MIL-S-16216 D (Navy); 12 F e b r u a r y 1959.
Steel P l a t e , Alloy, S t r u c t u r a l , High Yield Strength, M i l i t a r y Spec. ,
MIL-S-16216 E (Navy): 26 June, 1959, amended 3 August 1961.
Steel P l a t e , Alloy, S t r u c t u r a l , High Yield Strength, (HY80 and HYIOO)
M i l i t a r y S p e c , MIL-S-16216 F (Ships), 29 June, 1962.
Steel P l a t e , Alloy, S t r u c t u r a l , High Yield Strength, (HY80 and HYIOO),
M i l i t a r y S p e c , MIL S-16216 G (Ships), 27 F e b r u a r y , 1963.
M i n i s t r y of Defence, Ship Dept. , Welding and F a b r i c a t i o n of HY80
Steel, P r o c e s s Spec. No. DG S h i p s / P S / 9 0 2 7 A , J a n u a r y , 1967.
McKee, A. I.
Recent s u b m a r i n e design p r a c t i c e s and
p r o b l e m s . T r a n s . Soc. Nav. A r c h i t . M a r .
E n g r s . N. Y. 67, p . 632, 1959.
H e l l e r , S. R.
F i o r i t i I. , and
Vasta, J.
An evaluation of HY80 s t e e l a s a
s t r u c t u r a l m a t e r i a l for s u b m a r i n e s ,
Naval Eng. J n l . p. 29, F e b r u a r y , 1965.
p. 193, A p r i l , 1965.
B a k e r , R. G. ,
Wilkinson, F . , and
Newman, R. P .
The m e t a l l u r g i c a l i m p l i c a t i o n s of
welding p r a c t i c e a s r e l a t e d to low alloy
s t e e l s ; Second Commonwealth Welding
Conference, p. 125, 1965.
Boniszewski, T . .
B a k e r , R. G.
Heat affected zone cold c r a c k i n g in low
alloy s t e e l s ; Second Commonwealth
Welding Conference, p. 117, 1965.
and
Watkinson, F . ,
B a k e r , R . G . , and
T r e m l e t t , H. F .
Some thoughts on the p r o b l e m of
welding high t e n s i l e s t e e l s ; Welding
in Shipbuilding Symposium, Institute
of Welding, p. 3 1 , 1962.
Nippes, E . F .
Savage, W. F .
Allio, R . J .
Studies of the weld heat affected zone
of T - 1 Steel, Weld. J n l . . 36 (12),
p. 531-s, 1957.
and
Dolby, R. E .
The effects of welding on the f r a c t u r e
toughness of quenched and t e m p e r e d
low alloy s t r u c t u r a l s t e e l s . Weld. Inst.
Rep. C 2 0 1 / 1 0 / 6 8 , 1968.
Oldridge,
The effect of a t e m p e r i n g bead on the
p r o p e r t i e s and s t r u c t u r e of the weld
heat affected zone in QT35 s t e e l ,
D . A . E . T h e s i s , College of A e r o n a u t i c s ,
Cranfield, 1964.
D.
18
14.
Clifton, T . E . . and
George M. J.
Design and construction of a weld heat
affected zone simulator. Met. Constr. ,
1 (9), p 427, 1969.
15.
Coward, M. D.
The structure and properties of the heat
affected zones of welds in mild and low
alloy steels, M.Phil. Thesis, University
of London, 1967.
16
Smith, E,
Unpublished work, Cranfield Institute of
Technology, Cranfield.
17
Kellock, G. T. B.
A study of simulated weld heat affected
zone structures and properties of HY80
low alloy steel, D.A. E. Thesis, Cranfield
Institute of Technology, Cranfield, 1969.
18.
Dolby, R. E.
Private communication.
19.
Inagaki, M. ,
Uta, M. and
Wada, T.
A new apparatus for determining
SH-CCT diagram for welding and its
application to high strength steel. Trans.
Nat. Res. Inst, for Metals, 6 (6) p. 386,
1964.
20.
Puzak, P. P. ,
Eschbacher, E. W., and
Pellini, W. S.
Initiation and propagation of brittle
fracture in structural steels; Weld.
Jnl., 31, (12), p. 561-2, 1952.
21.
Puzak, P. P. , and
Pellini, W. S.
Evaluation of the significance of Charpy
Tests for quenched and tempered steels,
Weld. Jnl., 35 (6), p. 275-s, 1956.
22.
Pellini, W . S . . and
Srawley, J. E.
Procedure for the evaluation of fracture
toughness of pressure-vessel materials,
N. R.L. Report 5609, Washington D. C.
June, 1961.
23.
Pellini, W. S. and
Puzak, P. P.
Fracture analysis diagram procedures
for the fracture-safe engineering design
of steel structures, N. R. L. Report 5920,
Washington D. C. March, 1963.
24.
Pellini, W . S . ,
Evaluation of engineering principles for
fracture-safe design of steel structures,
N. R . L . Report, 6957, September, 1969.
25.
Winn, W.H.
Welding higher strength steels in warship construction. Second Commonwealth
Welding Conference, p. 209, 1965.
26.
Smith, E.
Unpublished work, Cranfield Institute of
Technology, Cranfield.
- 19 -
Smith, E . , and
Apps, R. L.
Effect of welding and p o s t - w e l d heat
t r e a t m e n t on QT35, Cranfield Report
Mat. No. 2, Cranfield Institute of
Technology, Cranfield, 1970.
G r o t k e , G. E. ,
Weldability and heat affected zone
t o u g h n e s s of HY150 s t e e l . Weld. J n l .
43 (6), p. 2 6 5 - s , 1964.
W e s s e l , E . T. , and
Hays, L. E .
Development of a h i g h - s t r e n g t h , tough,
weldable, s t r u c t u r a l s t e e l . Weld. Jnl. ,
43 (5), p . 2 1 5 - s , 1964.
Savage, W. F . , and
O w c z a r s k i , W. A . ,
The m i c r o s t r u c t u r e and notch impact
behaviour of a welded s t r u c t u r a l s t e e l ,
Weld. J n l . , _45, (2), p. 5 5 - s , 1966.
Doty, W . D .
Welding of quenched and t e m p e r e d s t e e l s ,
Weld. Jnl, 30, (7), p. 2 8 9 - s , 1965.
D o r s c h u , K. E . ,
L e s n e w i c h , A.
Kilpatrick,
I.
and
Development of a filler m e t a l for a high
t o u g h n e s s alloy plate s t e e l with a
m i n i m u m yield s t r e n g t h of 140 K s i .
Weld. J n l . , 43, (12), p. 5 6 4 - s , 1964.
Private communication.
M a t e r i a l Condition
2nd
Cycle
1st
Cycle
1275
Tensile
0. 2% Proof S t r e s s
N/mm2
(tonf/in2 )
Properties
N/mm2
(tonf/in2)
Reduction
of A r e a
%
UTS
Mean H a r d n e s s
HV5
939
(61.0)
1328
(86.3)
51
423
1275
1275
1000
(65.1)
1273,'--
(82.6)
53
412
1275
930
1028
(66.9)
1350
(87.7)
58
43 7
1275
765
793
(51.5)
1023
(66.4)
57
308
1275
650
930
(60.4)
1048
(68.1)
61
352
910
(59.1)
1080
(70.1)
62
353
930
930
765
675
(43.9)
884
(57.4)
62
318
930
650
771
(50.1)
881
(57.2)
66
279
602
(39. 1)
755
(49.0)
72
650
577
(37.5)
720
(46.8)
71
Transverse
576
(37.4)
719
(46.7)
72
234
228
Longitudinal
594
(38.6)
735
(47.8)
70
230
765
1
278
765
Parent
TABLE
1
1
Mechanical property data for as-cycled simulated specimens and the parent material.
M a t e r i a l Condition
1st
Cycle
2nd
Cycle
Te nsile
0. 2% Proof S t r e s s
N/mm2
(tonf/in^)
Properties
UTS
N/mm^
(tonf/in^)
Reduction
of A r e a
%
Mean H a r d n e s s ,
HV5
1275
_
735
(47.8)
837
(54.4)
70
268
1275
1275
751
(48.8)
827
(53.7)
72
272
1275
930
770
(50.0)
825
(53.6)
63
273
1275
765
705
(45.8)
813
(52.8)
70
255
1275
650
743
(48.3)
815
(53.0)
68
273
930
-
608
(39.5)
706
(45. 9)
74
231
930
765
594
(38.6)
698
(45.3)
73
238
930
650
607
(39.4)
705
(45. 8)
72
232
765
-
565
(36. 7)
705
(45.8)
78
230
765
650
556
(36. 1)
680
(44.1)
73
228
TABLE 2
M e c h a n i c a l p r o p e r t y data for simulated s p e c i m e n s , post cycle heat t r e a t e d at 650°C.
1300
(Q),(c}: MEASURED
(b),(d):COMPUTED
THERMAL CYCLES
THERMAL CYCLES
1200
1100
1000
900
800
o
700
UJ
ft:
I
600
UI
O.
500
400
300
200
100 \-
80
70
50
60
40
TIME , SEC.
FIGURE.I. THERMAL
CYCLES PRODUCED IN THE PARENT PLATE
ADJACENT TO THE WELD ,((a) AND (c)) , AND THOSE PRODUCED
BY COMPUTATIONAL TECHNIQUES , ( ( b ) AND (d)).
10
20
30
(a) carbon extraction replica
X4,500
(b) carbon extraction replica
X22,500
FIG. 2
PARENT PLATE MICROSTRUCTURE.
480
• 1 I 1 • 1 1 1 • 1 • 1 • 1 • I
' ' > —1—r
1
FUSION BOUNDARY
—^^
HAZ BOUNDARY
VlWSr
.1
1
f
UO
h
WELD B E N D - X ^ ^ ^ O ^ ,
\ <
400 L
1
\'
* X
w"V
L
]
PARENT PLATE—"'-•^'^-l--^''^
J
UNE OF HARDNESS SURVEY '
'1
r
M^l
ii
^
•'
1
-]
\»
M
(O
UI
360 NI '• A
;•
j
HVO-5
z
a
K
320-
>
4
^
K
UJ
280 >
240
\
Ac
200 1
.
0-4
1
.
•
1
.
1
.
Ac,
08
1-2
1-6
20
2-4
DISTANCE FROM FUSION BOUNDARY, mm
FIGURE 3. HEAT AFFECTED
ZONE HARDNESS
.
l.-L.
SURVEY.
28
J
!_. 1 1
3-2
1
311
HV0,5
436
HV0.5
FIG, 4
THE NATURE OF BANDING IN THE PARENT PLATE AFTER THERMAL
CYCLING TO A PEAK TEMPERATURE OF 930°C. OPTICAL MICROGRAPHS.
ENERGY
X -n
m 5
> c
ABSORBED , f t Ibt.
1
o
•^s
-• V
^
m
>
X
m
z
j
1
m
""
m
o
oi
g^
-• -i
^
X o
oX -n
-• d •
— X
m
2?r-
O ..»
i
9
II
-<
2
O
<1 ^>
z So
-«
o M
o
I
C
o
1
-»
m
<n
-i
s
m
Z o
•fl
5- 1
c
7)
m
o
o
2m
z
^-^
O
^ -D
O
CA
.
t
_
<
O
O
s
8
§
&
ENERGY ABSORBED 7
ENERGY
ABSORBED . FT. L B S .
ENERGY ABSORBED J
&
ENERGY
A B S O R B E D , FT. L B S .
I
1
^H S
Ö
r-
T»
Ï:
5 c
5
„
H r"
a) -J
'
1
T]
s1
\^S^^*fc-
-•
(^ 5* m
r- - I 2
m 2 Tl
3 5"
•Jf
!>
•c
O
X
-t
^
(Fi -* ID
(/)
•c -^ om
•D
c
Ö5m
z
5
c •71
fr! •0
i
. . ^ ^
^
m \
^
^
^
x^^^
~
m
>
c
^
m
m J
m 1
^
^
^
\
^
\
I
\ \
° 2
l
O
V
\
\
-<
o
z
o
ID
W
O
>
z
o
-D
O
cn
\
3!
UI
h
(O
o>
*
o
1
1
1
§
ENERGY
o
-^ m
>
-t
f>
' • <
Q
o
—i
o
-1
10
m
o
-n
cn
UI
o
UI
O)
g
TI o>
O UI
I
o
1
ABSORBEDJ
ABSORBED, f t . lbs.
—T—
;D
m
m P»
5
rri
7-4
> -.-D
>
;».
-i
J
o
—1—
o
\
C0
U> Ci>
J
8
ENERGY
y
^
So "
1
V
so
V.
I-
\
1
<D
5
Z
V\
\
ID
W
\
^
m
-n
i
o
-4
O O
T
-1
(/>
c
—
O CD
-*
in
m
-4
m
z m
n o
I c
> m
z
•30
TJ
-< -^
X
< m
1
Z
-4
z
•D
m
-1
c
o
>
n O
-4
;D
O
o•
I
Z
-n
r-
o
-<
o
f—
z
-» o
rti
o> Ol O) o>
Ol Ol Ol Ol
•
O) o> o»
Ol
o
o o
•
TB
O)
UI
o
T
-t
•D
o
tt f>CTo
X
iÜ zo
i ;^
>
:«
!-)
O
i <
o
z r</i m
o
ENERGY
ABSORBED J .
J
o ^ Z;
I- g m
m y
FRACTURE
FRACTURE APPEARANCE (%CRYSTALLINITY)
o T) I
I o
-r
^
f
o
3
M
to
13
O
O
•j
a>
I
I
o
APPEARANCE (7.CRYSTALLINITY)
&
o
s
•
o
_
L 5
?
PARENT
(LONG)
m c
> a) o
-4 m c
\
5^?
\
\
Mm
— - j co
PARENT
\ (TRANS.)
Z
..* Ol
-O
m
i"5
5 °
c
;o
m
o» > 5
K cn OD
\
\
°. °
*
m c I
o
*
UI
— o :
>
(/>
1
•V
m
o
z
m
z
,
FRACTURE
APPEARANCE (% CRYSTALLINITY)
FRACTURE
o
ë
1
JJ
1
APPEARANCE (•/. CRYSTALLINITY)
ë
S
S S
3
io<oL
Ito u k
o c k
I
1
H>H
[ t I C \\
|o|oi|
k
>
V'\
VA
t"
1
k1
o
"O
VA
\-'\
Ml
1
a> Ol
at Ol
-?ïSS
5i
tt>
Ol Ol
•
Ol
O)
Ol
ü'
1
•
Ol cn UI
• •
—1
to
1275
1275 + 1275
1275*930
"f
t o to
to
Ol
1
-
,ii^i. • !>r-;r ,,-. --..nV
•
u.
^^
is^V
(a) 1275 C carbon extraction replica
X4500
(b) 1275°C carbon extraction replica,
X9000
'4,1 .
f
'i^TX;
•V
(c) 1275/930 C carbon extraction replica (d) 1275/765 C optical micrograph,
X4500
X750
FIG. 11 SIMULATED WELD HAZ STRUCTURES PRODUCED BY SINGLE AND DOUBLE CYCLES
WITH INITIAL CYCLE TO A PEAK TEMPERATURE OF 1275°C
(e) 1275/765°C carbon extraction r e p l i c a , X4500
mmwj:^^m;^r:9ÏÏaT^W^?W^W
if--'-»
\ <
>•[/
•
\.
\'/, .^^
te'
.. ,
•
' •
b:i.
'
*5- , . . •
* ;r^ : •••
.•'•''^W^
• ^ - ^-C^"! -^^'-^^ '4^'-i
(f) 1275/650 C carbon extraction replica, X4500
FIG. 11 (continued)
':-::htvi..; ;
•S' ^::: i>»..K
' "^ •
• ' • • <
• A ' . W ' i/?:^»:.*-^
;
"
• . . '
• ' *'-\ ',.' . * i »
» • ' . ,
'
^ -
^ ^.- .-.
i
.'^-S^^ :^1ir:
,
'<•**'•.
•• >
.;':•<•. • .
'
s
•iü
.'» " , .
^iv-'i
'•r:l^•^ '^^<:M^:i-
•>4-.- 'i'JtCv--
• - • * ' ' 't '
,'' -j;. ^ i
'
^
-
^
,f,^.,
: .^''.,
• > ,-T*
,'/{•
••
•'
: • ; • > •
't .•.-:••»•
••
V ••-
* • • •
-.>T
^
**
» ' ".' "^^ ' "
•* .
••
. > : ^ .
»:
1»
,-;,: -^ iv
' •
•^V4i>^
v?%
. • _
•.' "*-*
••'••'.'.'t • .-Vi;
t-
">•«', r .
•'"^l'
•i"S
::t;,-:^.;:
- •-, -T.^
' » •
;
•• . ' f -
<
•:•••• w
-
. > ,
-•••,
%
fc.^-
>^
••
:.
.*'
•,v 'i' -^.'
3 A. *•*. >. _
_*'4
^^
IK-
t.
(a) 930 C carbon extraction replica,
X4500
(b) 930 C carbon extraction replica
X9000
**-x>-• '.v';-.-'*.
.f
^
••.•.•..,-»
' • • • * . € • '
• •
•.» • *•
•:-;\-'••;•-•
-• ••*••£'» ^-"-'M^i .-'.•>#'-/
'•
.
• *
'.•••*
«•
•. '
•
''*.^-•
•
"AM'
.-•-•-.
,
•
-.•••
•'.^
^m.c
,
. .
• - ,'/.'• f'.
• .•."-•i- :.^-t.--i^-r''* é--i^
•
•
•: '•'
«f • . » _ •
(c) 930/765°C carbon extraction replica (d) 930/650 C carbon extraction
X4500
replica, X4500
FIG. 12 SIMULATED WELD HAZ STRUCTURES PRODUCED BY SINGLE AND DOUBLE CYCLES
WITH INITIAL CYCLE TO A PEAK TEMPERATURE OF 930 C
^'jkyr^
..V .^..
. ^ I » • • . »4 .. ' .'-I
. ' .I I 'I »*'' . ** "" 1
:;*f^
(a) 765 C carbon extraction replica,
X9000
(b) 765/650 C carbon extraction replica,
X4500
(c) 765/650°C carbon extraction replica, X30000
FIG. 13 SIMULATED WELD HAZ STRUCTURES PRODUCED BY SINGLE AND DOUBLE
CYCLES WITH INITIAL CYCLE TO A PEAK TEMPERATURE OF 765°C
,>/
Ï
—^-
'—„•—
^«1
k-^'-'^jV
F- ^
•V-U' "H.'^'^
(a) 1275°C carbon extraction replica
X4500
(c) 1275/765°C optical micrograph X750
FIG. 14
(b) 1275/1275°C carbon extraction replica
X22500
(d) 1275/765°C carbon extraction replica
X4500
POST-CYCLE HEAT TREATED STRUCTURES OF SINGLE AND DOUBLE CYCLED SPECIMENS
WITH INITIAL CYCLE TO A PEAK TEMPERATURE OF 1275°C
(a) 930°C carbon extraction replica
X4500
FIG. 15
(b) 930/765°C carbon extraction replica
X4500
POST-CYCLE HEAT TREATED STRUCTURES OF SINGLE AND DOUBLE CYCLED
SPECIMENS WITH INITIAL CYCLE TO A PEAK TEMPERATURE OF 930°C
Fly UP