Академический Документы
Профессиональный Документы
Культура Документы
ELEVATED-TEMPERATURE PROPERTIES
OF CHROMIUM-MOLYBDENUM STEELS
(An Evaluation of 21/4Cr- IM0 Steel)
05-40006-002
05-40006-002
NOTE
The Society is not responsible, as a body,
for the statements and opinions
advanced in this publication.
Abstract ------------------------------------------------------------------------------------------------------------ 1
Figs. 1-6 ----------------------------------------------------------------------------------------·-.. ·------------- 3
Introduction -------------------------------------------------------------------------------------.. ·---.. ·---·-- 9
Yield Strength and Tensile Strength ------------------------------------------------------------ 9
Creep and Rupture Properties ------------------------------------------------------------------------ 11
Annealed Material - Direct Extrapolation -------------------------------------------------- 11
Annealed Material - Indirect (Parameter) Extrapolation ------------------------ 13
Annealed Material - Comparison of Rupture Strengths
for Different Evaluation Procedures ------------------------------------------------ 13
Normalized-and-Tempered and Quenched-and-Tempered Material ------------ 14
Rupture Ductility _____________________________________________________ .. _____ ..... ________ .. ____ ________ _____ 15
REFERE:OCE: Smith, G. V., "Supplemental Report on The body of the report provides in tables,text,
the Elevated Temperature Properties of Chromium- and figures, details concerning the indentification
Molybdenum Steels (An Evaluation of 2li;Cr-1Mo of the individual lots of material, the evaluation
Steel)," ASTM DATA SERIES, DS 6S2, American Soci- procedures, and the results.Several summary figures
ety for Testing and Materials, 1971. immediately following this abstract, Figs. 1-5,
depict the temperature dependence of important
strength properties for a number of grades of prac-
ABSTRACT: This report offers evaluations of ele- tical interest. Figure 6 provides a comparison of
vated temperature strength properties of 2!i;Cr-1Mo the temperature dependence of yield and tensile
steel in various conditions of heat treatment and strength, expressed in the form of strength ratio.s,
for diverse product forms. The data evaluated in- for several conditions of heat treatnent and pro-
clude test results previously included in ASTM duct form.
Data Series DS 6 (1953) and DS 6Sl (1966), as well
as newly generated data gathered by the Metal Pro-
perties Council. KEY WORDS: elevated temperature, tensile strength,
The data for wrought material are separated yield strength, creep strength, rupture strength,
into a number of strength categories corresponding elongation, reduction of area, alloy steel, mechan-
to current specifications for annealed, normalized ical properties, data evaluation.
and tempered, and quenched and tempered grades,
Data for castings and for weld-metal are evaluated
separately.
Rupture strength and creep strength have been
evaluated at two levels, respectively, as the
stresses to cause rupture in 10,000 or 100,000
hours and as the stresses corresponding to secon-
dary creep rates of 0,1 and 0,01 percent 1000
hours. The temperature dependencies of these pro-
perties as well as of yield strength (0.2 percent
offset) and tensile strength are developed in such
a form as to be useful for establishing allowable
working stresses. Elongation and reduction of area
data, although not of direct usefulness in setting
stresses, are also included,
3
.. - .L
-r
+ 1-
-_:c
i-
,,,.
+t·
+-
:j: -
+
......
Ul
.!<1
Ul 't rlt- ~
Ul '~ f'i\:
~
E-< . +Lt
c -· =h
q_ j· .
T #Ef ih+ . H
11tlHu:_l±ttttl±
t
(/)
L~t·xH+l,_~+. J-l-
4
Figure 3 - Effect of temperature on yield strength, tensile
strength, creep strength (.01 ~per 1000 hours) and rupture
strength (100,000 hours) of quenched and tempered material.
Yield and tensile strengths have been adjusted to 100 and 115
kai at 75 Fr creep and rupture strengths represent averages of
data adjusted to a specified tensile strength of 115 ksi at 75 F
5
Figure 4 - Btfect of temperature on yield strength, tensile
strength. and rupture strength (100,000 hours) of quenched and
tempered material. Yield and tensile strengths have been ad-
justed to 15 and 95 ksi at 15 Pr rupture strength represents
averages of data adjusted to a specified tensile strength of
95 ksi at 15 P.
6
Figure 5 - Effect of temperature on yield strength and tensile
strength of normalized-and-tempered and quenched-and-tempered
cast 2 1/4 Cr - l Mo steel, adjusted to 55 and 85 ksi at 75 F.
Jue~eu ~o f ) ana ~' Ksi at 'l5 ~, rupture streng1;n represents
averages of data adjusted to a specified tensile strength of
95 ksi at 75 F.
7
Figure 6 • Comparison of the temperature dependencies of yield
and tensile strengths, expressed as stre~th ratios, for
annealed (A) normalized and tempered (N'!)• and quenched and
tempered (Qf~ wrought material and normalised and teapePed
and quenched and tempere~ east material (C).
8
INTRODUCTION Yield Strength and Tensile Strength
Since publication in 1966 of ASTM Data Series The original tensile test results, not
Publication DS 6Sl "Supplemental Report on the previously reported in the ASTM Data Series
Elevated-Temperature Properties of Chromium- publications, are tabulated in Table III, sepa-
Molybdenum Steels," not only have additional data rated according to product form. As with previous
on 2-1/4 Cr-1 Mo steel become available, but also, evaluations, the tests are presumed to have been
considerable interest has developed in the use of conducted generally at strain rates within the
this material at the higher strength levels, limits permitted by ASTM recommended practice E 21.
ranging up to tensile strengths of 135 ksi, that The yield strengths are known in nearly all
can be achieved by normalizing or quenching instances to correspond to 0.2% offset, or to the
followed by tempering at relatively low tempera- lower yield point for materials exhibiting a yield
tures. It has therefore seemed desirable for the point. The tabulated elongation values were
Metal Properties Council (MPC) to undertake a re- measured on a 2-inch 2age length and the test
evaluation of the elevated temperature properties specimens were taken from the quarter thickness
of this popular grade of steel and, in particular, position unless otherwise noted.
to assess the effects of heat treatment and product In previous reports covering wrought ( 25 26 )
form upon the basic trends of behavior. austenitic stainless steels and carbon steel, '
The evaluation covers data gathered by MPC it has proved helpful in evaluating yield and
from industrial and other laboratories, and from tensile strengths to normalize the test results by
published literature, as well as the data previous- ratioing the elevated temperature strengths of
ly included in ASTM's Data Series Publica- individual lots to the corresponding strengths of
tion~(l,2,13) A coding key is provided in Table the same lots at room temperature, and then to
to the data from the ASTM Data Series, and the establish by the method of least squares the
newly gathered data are appended to this report. regression curve of best fit -- the trend curve
The data are identified in Tables I and II, as to for the variation of strength ratio with tempera-
specification, heat treatment, product form and ture. This same procedure has been employed in
size, chemical composition, and source. Some of the present evaluations, and its effectiveness in
the data sets from DS6 and DS6-Sl were excluded normalizing the data will again be apparent.
from the evaluation owing to inadequate identifi- In assessing the effect of strength level
cation, and nonconformity with specifications, upon yield and tensile strength of 2-1/4 Cr-1 Mo
including the use of unconventional heat treat- steel at elevated temperatures, it has proved
ments. A distinction has been preserved in the convenient to adopt for the wrought products a
early stages of the evaluation as to product form categorization that reflects current plate steel
although such differences later proved to be specifications:
unimportant in a number of instances. In a few
instances, steel produced to a specific, specifica- Category 1: All annealed material, what-
tion was tested in another form, as when material ever the strength level, and
produced to a pipe specification was actually normalized and tempered
tested in the form of bar. Also, when, in a few material having a tensile
instances, material was identified only as wrought, strength at room temperature
it was arbitrarily categorized as bar. less than 75 ksi.
The properties that have been evaluated in
this report include yield and tensile strengths, Category 2: Normalized and tempered
and creep and rupture strengths. The latter two material having a tensile
properties have been evaluated at two levels each: strength at room temperature
as the stresses causing a secondary creep rate of greater than 75 ksi (ASTM
0.1% or 0.01% per 1000 hours and causing rupture Spec. 387, Grade D) •
in 10,000 or 100,000 hours. These latter properties
are of interest for setting allowable stresses for Category 3: Quenched and tempered material
boiler and pressure vessel systems in the range of having a tensile strength at
temperatures in which creep is important, whereas room temperature less than 115
yield and tensile strengths are employed in ksi (ASTM Spec. 542, Classes
establishing allowable stresses at lower tempera- 1 and 3).
tures. No attempt was made to assess creep strength
in terms of the stress causing a specific creep Category 4: Quenched and tempered material
strain in a given time interval as stipulated in a having a tensile strength at
number of foreign codes, for the reason that the room temperature of 115 ksi or
available data are unsuitable for that purpose. more (ASTM Spec. 542, Class 2).
Although not directly useful to designers,
elongation and reduction of area at fracture in A few data for forgings reported to have been
the tensile and rupture tests are included in this quenched and tempered corresponded better with
report, where available, and some comments are levels to be expected of normalized and tempered
offered concerning the trends of behavior. Since plate material, and were so categorized. After it
the individual strength properties needed for set- was discovered that the strength ratios of quenched
ting allowable stresses are each required over the and tempered material are essentially independent
temperature range in which they may govern, an of strength level (as also appears to be true with-
effort has been made to develop the evaluations in the other categories, as well)C27), a few data
in terms of trend curves defining the variation not meeting the minimum tensile strength of current
of strength with temperature, rather than in terms specifications (95 ksi) were included in category
of strengths at individual temperatures. 3, and a few data exceeding the maximum of current
9
27
specifications (135 ksi) were included in category dependence upon strength at room temperature. c J
4. However, in the latter instance, ductility The scatter in yield strength ratios is
data were not incorporated into the plots. substantially greater than for tensile strength
The few data for quenched and tempered cast- ratios, but whereas some of the scatter may be
ings having a tensile strength greater than 105 associated with the same factors causing scatter
ksi were first considered separately from those of tensile strength ratios, a significant portion
for normalized and tempered castings, but when the is certainly inherent, reflecting testing diffi-
resulting regression curves proved ·to be essenti- culties in the measurement of yield strength, the
ally identical, the two sets of data were combined possible presence of residual straightening
into a common category. The elongation and stresses, and possibly differences in strain rate,
reduction of area results for quenched and tempered factors which are less important in measuring
castings fell within the scatter band of the other tensile strength.
data. A question of interest to some engineers is
Data representing weld metal that had whether there is an effect of section thickness
received a post weld heat treatment were evaluated upon the behavior, particularly in plate, and
as a separate category. Data representing tests accordingly the test results were examined in
of weldments, i.e. of specimens including weld this respect. The available data represented
metal, base metal and heat affected zone, were not sections ranging upwards to more than 7 inches.
evaluated, owing to the basic inhomogeneous The reported observationC27) that strength ratio
character of this type of test specimen, and the at elevated temperature is independent of strength
dependence of the results upon geometrical and level, within a specific heat treatment category,
other factors. indicates that although there might be expected
Finally, a few data became available after to be a dependence of strength, measured at a
the regression analyses had been completed, and specific subsurface location, upon section size,
whereas these data are included in the tabulation owing to the dependence of cooling rate upon
and plots, the regression analyses may not have section size, there should not be any effect upon
been rerun, when inspection indicated that any strength ratio. A comparison of individual strength
change in the regression result would be incon- ratios at 800°F in different section sizes, for
sequential. both normalized and tempered and quenched and
The tensile test results for the different tempered plate failed to reveal any dependence of
categories are plotted as dependent upon tempera- ratio upon section size.
ture in figures 7 through 12. In each figure, Except for weld metal the ratio plots were
part (a) charts yield strength and yield strength evaluated by the method of least squares, and with
ratio, part (b) charts tensile strength and tensile no distinction as to product form for the wrought
strength ratio, and part (c) charts elongation and materials, to establish the best-fit curves
reduction of area. In the plots for the four defining the dependence of strength ratio upon
wrought categories, data for the different product temperature. The resulting "trend curves" have
forms, plate, pipe and tube, bar, and forging have been superimposed upon the plots, and tabulated
been differentiated. However, relatively few in Table V. Because the trend curves for the two
data representing bar or forgings were available, categories of quenched and tempered material were
and most of the data in the normalized and tempered essentially identical, a ratio trend curve was
or quenched and tempered categories represented developed for the combined groups and is given also
plate materials. Inspection of the ratio plots in Table V. A graphical comparison of the ratio
does not suggest the need to distinguish amongst trend curves is given in Figure 6 (following the
the wrought product' forms of a particular strength abstract) for the following categories: annealed,
category, but clearly the sample sizes are limited. normalized and tempered, quenched and tempered,
The ratioing procedure proved generally and castings.
effective in reducing the scatter of results in The trend curves for annealed material are
the strength plots, especially for the tensile strikingly different than those for the other
strength of the wrought products at temperatures categories, particularly for tensile strength.
below about 500°F, At higher temperatures, there A~ indicated previously, the maximum in tensile
is reason to believe that the scatter may reflect strength of annealed material at intermediate
structural changes within the material, sensitive temperatures is characteristic of materials that
to composition and prior treatment, that in turn exhibit dynamic strain aging. The shift in peak
result in differences in strength ratio. The temperature from the range 400 to 500°F observed
rise in tensile strength ratio exhibited by the for carbon steel to 700 to 800°F for 2-1/4 Cr-1
annealed category at intermediate temperatures is Mo steel is to be associated with the presence
characteristic of materials that exhibit dynamic of strong carbide forming elements.(27) The trend
strain aging, and it seems probable that the curves for the remaining categories are essentially
scatter of strength ratios in this range for this similar in form. Differences are somewhat greater
category may therefore reflect differences in for yield strength than for tensile strength. As
susceptibility to such aging. Normalized and a matter of possible interest all of the data
tempered as well as quenched and tempered materials excepting the annealed category have been grouped
do not exhibit dynamic strain aging, at least to together to develop common ratio trend curves;
the extent of a rise in tensile strength at inter- the ratios are tabulated in Table V.
mediate temperatures. However, the leveling of With the dependence of strength upon tempera-
tensile strength of the normalized and tempered ture expressed in terms of strength ratios, it is
material, as well as the increased scatter, possible to compute absolute values of strength,
cited above, may reflect some tendency to dynamic adjusted to any specific level within the limits
aging.(27) In any event, the scatter does not encompassed by the original data. Summary Figures
appear to be related to differences in strength 1-5, referred to in the Abstract of this report,
level, as revealed by plots of strength ratio at include plots showing the variation of yield and
individual elevated temperatures, e.g. 800°F, in tensile strength with temperature, adjusted to the
10
specified minimum levels of several purchase speci- to produce relatively soft material, and it has
fications of general interest. been generally common practice in such evaluations
Although there are too few data available for to ignore a possible dependence of creep and
weld metal, especially at temperatures between 75 rupture strength upon microstructure, which is
and 800°F, to warrant an attempt to develop trend known to have an important influence upon strength
curves, the ratio trend curves developed for cast at room temperature. In the present evaluation,
steel have been superimposed over the ratio plots a possible dependence of creep and rupture strength
for the weld metal to provide a basis for assess- on microstructure has also been ignored, and all
ment of weld metal, Figure 12 a and b. Examina- data for annealed material, including material
tion of these plots suggest that the available normalized and tempered at a relatively high tempera-
data are not inconsistent with the trend curves ture, have been grouped together without distinction
for cast steel. as to strength level. In fact, in the present
Elongation and reduction of area of the instance, as frequently true, room temperature
annealed material fall off slightly at intermediate strength data are so generally lacking as to pre-
temperatures, possibly reflecting the dynamic clude the possibility of assessing the dependence
strain aging, mentioned earlier, and then increase of creep rupture strength on room temperature
progressively to the maximum temperature for which strength. However, it is clear that the creep and
there were data, 1300°F. The normalized and rupture strengths of the higher strength categories
tempered, the quenched and tempered, and the cast of 2-1/4 Cr-1 Mo steel are not independent of
materials also exhibit a slight loss in ductility room-temperature strength, certainly in the range
at intermediate temperatures. The ductility of greatest practical interest, 800 to 900°F,
scatter bands of the various categories overlapped or slightly higher. Consequently, the creep and
extensively especially for reduction of area. How- rupture strengths of the higher strength grades
ever, annealed material did exhibit somewhat have been evaluated in such a way as to recognize
greater elongation at room and slightly higher their dependence upon strength level.
temperatures than did the other categories, and the For extrapolating time-for-rupture data to
quenched and tempered material having a tensile permit assessing the stress to produce rupture in
strength of 115 ksi or greater exhibited elonga- 100,000 hours, there are two broad types of
tion slightly inferior on the whole to the other procedures. In the first, or direct procedure,
categories. the relation between log stress and log time-for-
rupture at constant temperature is extended to
100,000 hours. In the second procedure, extra-
Creep and Rupture Properties
polation is effected indirectly by means of one
The original data not previously reported in or another time-temperature parameter. Both
DS 6 and DS 6Sl (References 1 and 2) are tabulated procedures have been employed in the present
evaluation.
in Table IV, separated according to product form.
The data are plotted in various groupings in
'Figures 13-20. Annealed Material - Direct Extrapolation
The creep and rupture data have been evaluated
with the primary objective of developing best It is possible to effect direct extrapolation
current assessments of these properties in a form either by treating different lots individually, or
directly useful by Code groups for establishing alternatively by defining and then extending the
allowable stresses. In the ASME Boiler and Pressure best fit curve for the scatter band of all data.
Vessel Code, the criteria for setting stresses in The merits of the alternatives have been reviewed
the "creep" range of temperatures include the in the prior evaluations of the current series as
average and minimum stresses to cause rupture in well as in published literature.(25-28) To permit
100,000 hours, reduced by appropriate fractional comparisons, both procedures have been employed
factors, the purpose of which is to insure a in evaluating the rupture strength of annealed
reasonably long safe period of usefulness. The material, but for reasons given in the cited
allowable stress is also limited in the ASME Code references, principal emphasis is placed on
to the average stress to produce a secondary creep individual lot extrapolations. To show both the
rate of 0.01 percent per 1000 hours, which corre- volume and the scatter of the data, all of the
sponds to a rate of 1 percent per 100,000 hours, data, including applicable data from DS 6 and
if the rate were to remain unchanged. The stress DS 6s1(1,2J, are shown in isothermal scatter
for rupture in 100,000 hours can ordinarily be band plots of log stress versus log time for
developed only by extrapolation of the results of rupture (Fig. 13 a,b,c and d); of log stress
tests of shorter duration, since tests lasting versus log of secondary creep rate (Fig. 14 a,b,
100,000 hours are seldom made. Determination of and c); and of percent elongation and reduction
the stress causing a secondary creep rate of 0.01 of area at rupture versus log time for rupture
percent per 1000 hours may require extrapolation, (Fig. 15 a,b,c,d and e). In each figure, data
but may be obtained by interpolation. for different product forms are differentiated,
The data have been evaluated also for the although in the end it does not seem reasonable
stress to cause rupture in 10,000 hours, which to attempt to distinguish amongst product forms.
frequently involved the need for extrapolation, Data for weld metal are considered separately in
and for the stress to cause a secondary creep of a later section.
0.1 percent per 1000 hours. Rupture and creep Because there was no evidence either in the
strengths for still shorter rupture times and scatter bands or in the individual lots that log
faster creep rates have not been explicitly stress and log time-for-rupture were related other
assessed, but can be readily approximated by than linearly, at least at temperatures below
inspection of the scatter band data plots. 1150°F, the time-for-rupture data, whether con-
Past evaluations in this series have been sidered as individual lots or as scatter bands,
concerned with annealed material or material norma- were extended linearly to 100,000 hours. (See
lized and tempered at relatively high temperatures also prior discussion of this question in
11
references 25-28). On the other hand, log extended regression line for 1050°F then inter-
secondary creep rate tended often to vary curvi- sects the 100,000 hour abscissae at 8.87 ksi
1inearly with log stress in the range below 0.1 instead of 7 .67 ksi; the extended regression lines
percent per 1000 hours; the variables were so for 1000 and 1100°F are altered only slightly by
treated, either for interpolation or extrapolation. the exclusion of data less than 500 hours (to
The secondary creep rate data were evaluated only 11.87 and 7.25 ksi respectively). This general
in terms of individual lots not as scatter bands. problem, particularly with respect to the linearity
The individual lot extrapolations for both rupture of the isothermal relations, has been considered
and creep strength were developed visually, giving at some length in another publication(27) and the
weight to the longer-time or slower-rate data. conclusion drawn that the behavior must be accepted
The results of the individual-lot extrapola- at face value, pending the availability of further
tions or interpolations, which were made on test results that would provide a more positive
individual plots too numerous to include here, are basis for disputing the observed behavior.
tabulated in Tables VI-IX, along with similar The perturbation apparent in the family of
results for the high strength categories, to be isothermal regression lines is also evident, though
discussed later. The results have also been less prominently, in the results of the individual-
plotted in Figs. 21 and 22 as dependent upon temper- lot extrapolations, plotted in Figures 21 and 22.
ature. Semilogarithmic coordinates were chosen for Particularly for rupture in 100,000 hours, Figure
the plots of Figs. 21 and 22 because in past 22, the rupture strength values for 1100°F mass
evaluations they have frequently tended to above the line of best fit for the data at 1000,
linearize the relation between the variables. Most 1050 and 1200°F, or considered alternatively, the
of the data plotted in Fig. 21 and 22 represent data at 1050°F lie on the low side. This behavior
pipe and tube. The few data representing bar and is evident also in Figure 21 for rupture in 10,000
castings appear to agree reasonably well with the hours, but less prominently. Inspection of Fig. 22
overall scatter and, in the further evaluation, reveals that the results at 1050°F are weighted
are treated as belonging to the same sample popula- importantly by the three lowest values but no
tion. There were no data for annealed plate. valid reason could be found for rejecting these
The isothermal scatter bands, except those values; attention might also be called to the
at 1150 and 1250°F, for which there were few weighting on the high side at 1050°F, resulting
results, were evaluated by the method of least from three sets of data for bar. Note also the
squares, and the mean curve of best fit extended weighting for rupture in 10,000 hours, Figure 21.
to 100,000 hours. In these evaluations the longer- And, of course, the trend evident in Figure 22
time test results were weighted by arbitrarily receives support from the regression analysis of
excluding rupture times less than 40 hours, and the isothermal scatter bands, Figure 23. The mean
time was taken as the independent variable, for regression lines drawn on the rupture data of
reasons given previously.(28) In the regression Figures 21 and 22 represent best fits for the data
analyses, no significant improvement in variance and correspond to equations of third degree.
was observed in going beyond a linear dependence Minimum curves at a confidence level of 90% have
between log stress and log time-for-rupture, also been developed from the mean curves on the
except at 1200°F. At this temperature, an improve- assumptions that the data exhibit a log normal
ment in variance was realized by assuming a second distribution in strength, that the mean curve is
or even a third order dependency, but the number without error in slope and that the variance is
and character of the data are such as to inhibit independent of temperatures. At a confidence
placing complete confidence in the regression level of 90%, 95% of the individual values should
result. The best fit lines from the least squares lie above the minimum, which is at least approxi-
analyses are superimposed upon the individual iso- mately true. The mean and minimum values defined
thermal scatter bands, and the values of 100,000 by the regression lines in Figures 21 and 22 are
hour rupture strength obtained by extending the included in Tables X and XI. The arithmetic means
regression lines are included in Tables X and XI of the individual values at the different tempera-
for later comparison with strengths developed by tures are also included in Tables X and XI, along
other evaluation procedures. with minimum values at the 90% confidence level,
The family of linear isothermal regression on the assumption of a normal distribution in
lines, shown in Fig. 23, reveal that the slope strength.
for the data at 1050°F is greater than those for The stresses to cause secondary creep rates
1000 and ll00°F, such that the 1000 and 1050°F of 0.1 and 0.01 percent per 1000 hours have also
lines intersect at about 100 hours, and the 1050 been plotted in Figs. 21 and 22. Unfortunately,
and 1100°F curves would intersect shortly beyond the distrihution of the data points suggests that
100,000 hours. It might be argued that the near the relation between log strength and temperature
identity of strengths at 100,000 hours is not may be nonlinear, as was true for rupture strength.
possible. Yet there can be no question that the Thus, in Figure 22, the strengths at 1050°F fall
scatter bands of actual test results at 1000 and within the scatter band of strengths at 1000°F.
1050°F do overlap at 100 hours, and it is not Moreover, inspection of the plots of log stress
unknown for a material to exhibit the same strength versus log secondary creep rate, Figure 14, reveals
at two or even three temperatures (see, for extensive overlapping of the scatter bands. How-
example, Fig. 7b). Alternatively, it might be ever, since there are only about one third as
argued that the data at 1050°F represent a unique many data at 1050 as at 1000 or ll00°F, it seemed
population, but this seems unlikely since relatively possible that the 1050°F data sample might be
few heats were tested only at 1050°F and not at unrepresentative. Examination of the temperature
another temperature. Still another possibility is dependence of the strength of individual heats
that the 1050°F slope breaks upwards, or that the showed that the highest two values at 1050°F,
1100°F slope breaks downward (or both). Some Figure 22, corresponded with the highest individual
support for the view that the 1050°F slope does values at 1000 and 1100°F, respectively, indicating
break upwards is found if data less than 500 hours that the data at 1050°F are, in fact, unrepresenta-
are excluded from the regression analyses. The tive.
12
In view of the character of the data, it has A scatter band plot showing the variation of
not seemed appropriate to develop the trend curve log stress with L-M parameter for the annealed
by regression analysis.* Instead, pending the material is shown in Fig. 24. Values of the
availability of sufficient additional data that parameter corresponding to 100,000 hours at J001l,
would preclude the need for subjective judgment, 1050 and ll00°F are marked on the figure. Th0 dai. a
it has seemed best to define the "trend curve" of plotted in Fig. 24 were evaluated by the method o'.
creep strength, on Figures 21 and 22, by a linear least squares to define the mean curve of-best fit,
line connecting the averages at 1000 and ll00°F, which corresponds to an equation of second order;
and another linear line between the averages at temperature was taken as the independent variah11"
1100 and 1200°F. The creep strengths so defined for the regression analysis. The mean curve is
are tabulated in Table XII. Further, owing to drawn on the plot together with a minimum curve
the character of the data, it has not seemed (90% confidence,level) derived from the mean curve.
appropriate to attempt to develop minimum values. The mean and minimum values for rupture in lG,000
(These are not re~uired for setting stresses in and 100,000 hours are included in Tables X and XI.
the ASME Code.) Approximate minimum values may It is of interest to note in Fig. 24 that the
be estimated by inspection of the plots of Figs. portion of the data l!ncompassing parameter v:1.l •es
21 and 22. that define 100,000 hour rupture strength at
temperatures above 1000°F represents only a small
Annealed Material -- Indirect (Parameter) fraction of the total volume of data, and that the
Extrapolation position of the regression line in this range is
importantly dependent upon the mass of data at
It is commonly agreed that extrapolation by lower parameter values. -
time-temperature parameter should preferably be A scatter band plot for the Manson compromisr-
undertaken on an individual heat basis, but to parameter has not been prepared, but it is pre sum< .1
do this requires more data, particularly at that the plot would resemble Fig. 24. However, tl-.c
different temperatures, than are generally availa- mean and minimum values resulting from regression
ble for the data collected by the Metal Properties analysis of the Manson compromise parameter are
Council. However, it is possible to adopt a plotted in Fig. 25, and 10,000-hour and 100,000-hour
"universalized" approach, which assumes values strengths are included in Tables X and XI.
for the parameter constants, and to "parameterize" A number of features of the parameter scatter
and evaluate all of the data on a scatter band band approach to the evaluation of data for 2-1/4
basis. In spite of inherent reservations concern- Cr-1 Mo steel are considered at length in another
ing such an approach, all of the data, for rupture publication (27), and are not repeated here. How-
times exceeding 5 hours, were evaluated by this ever, the important observation was made that if
procedure employing both the well-known Larson- rupture strength actually varies in the complex
Miller parameter, with an assumed value of 20 for manner, that is depicted in Fig. 22, then the
the constant: universalized parameter procedure, by its inh('rent
nature, is incapable of giving correct valuP.s.
T(20 +log t) = F1 (S) No effort was made to evaluate creep strength
by parameter evaluation procedures.
and the recently proposed Manson compromise**
parameter: (29) Annealed Material -- Comparison of Rupture Strength~
for Different Evaluation Procedures
13
extension of the isothermal scatter band regression data cannot be contemplated.* Furthermore,
line. The most favorable comparisons are afforded parameter methods would seem to be inappropriate,
by the means of the individual extrapolations and even if applied to the data from individual heats,
the values developed by regression of the individual- owing to the increased probability of significant
lot strengths as dependent upon temperature. The microstructural instability, associated with the
latter procedure has the advantage that the data finer microstructure.
can be considered as a whole, thereby also pro- Because scatter band evaluations are inap-
viding a better basis for interpolation, as at propriate, the rupture time and creep rate plots
1150°F. It is of interest to note that both of of Figs. 16 and 17 covering wrought material,
these direct procedures lead to a more reasonable mostly plate, are provided only to convey an
difference between the strengths at 1050 and 1100°F impression of the volume of available data. No
than that indicated by extending the isothermal creep and rupture data were available for quenched
scatter band regression lines. This observation and tempered castings; the few data for weld metal
offers further evidence, as argued in an earlier are considered in a separate section. Rupture
publication,(28) that a scatter band may mask strength and creep strength for individual lots of
individual differences. normalized and tempered or quenched and tempered
The differences between the results of the two material have been derived by individual-lot,
parameter procedures were always less than 10 per direct, visual extrapolation or interpolation on
cent, and neither parameter result differed more log-log plots too numerous to include here, with
than about this same amount from the strength- weighting of the longer time results. The results
temperature regression result. The Larson-Miller are included in Tables 6-9. Although it is
parameter results agreed slightly better with the possible that the plots of log stress vs. log
strength-temperature regression result, except at rupture time may develop curvilinearity at long
1050°F, where the compromise parameter result was times, no clear instances of such behavior were
in exact agreement. All factors considered, it observed, and the rupture extrapolations were
seems best to give greatest weight to the values therefore performed on the assumption of linearity.
developed by the strength-temperature regression, Curvilinearity was evident in the plots of log
and these values have therefore been employed in stress vs. log secondary creep rate.
summary Figure 1. A graphical comparison of the The dependence of rupture and creep strength
results by this procedure with those by the two of these higher strength materials upon tensile
parameter procedures is provided in Figure 26, and strength at room temperature becomes evident when
the similarity of the various results is evident. the results are plotted as in Figure 27. Data for
The parameter procedures do permit an assess- both quenched and tempered material and for norma-
ment of 10,000 hour rupture strength at 1150 and lized and tempered material (relatively few),
1200°F, and hence across-the-board comparisons distinguished by symbol, are included in the plots,
can be made throughout the range 1000 to 1200°F. since they appear to belong to the same family.
Since some of the test results extended to 10,000 Also, all of the data available for quenched and
hours, whereas none had extended to 100,000 hours, tempered material are included in the plots, even
there should be an increased confidence in the though some of the lots have a room temperature
results of the direct extrapolations, relative to tensile strength less than or greater than permitted
the parameter extrapolations. Study of Table X by current ASTM specifications. Examination of
reveals that when compared to the strength-tempera- the plots of Fig. 27 a-d reveals that creep and
ture regression results, the Larson-Miller rupture strengths increase approximately linearly
procedure resulted in higher values at &11 tempera- with 1"ensile strength at 800, 850 and 900°F, the
tures, except at l150°F. Interestingly, the dependence decreasing with increasing test tempera-
Manson compromise parameter, though more conserva- ture; at 1000°F the dependence has essentially dis-
tive than the Larson-Miller parameter, also gave appeared. The data for the individual temperatures
higher values at 1000, 1050 and 1200°F. have been evaluated by the method of least squares,
with tensile strength as the independent variable,
assuming a linear relation between the variables;
Normalized-and-Tempered and Quenched-and-Tempered the resulting lines of best fit have been super-
Material imposed upon the plots, and values corresponding
to specified minimum tensile strengths of a number
Since the high room-temperature strength of of specifications of current interest have been
these materials is achieved by employing tempering included in Tables X-XII.**
temperatures which may be as low as 1150°F for the
normalized material and as low as 1050°F for
quenched material, it is to be expected that the
associated finer microstructure will be inherently *In principle, it would be possible, though hardly
less stable than that of annealed material or of practicable, to develop scatter bands for restricted
material tempered at higher temperature. Moreover, ranges of room temperature strength.
a dependence of rupture and creep strength upon
room-temperature strength, at least in the range **A few apparent inconsistencies are evident (e.g.
of temperatures of practical interest (perhaps the same stress for a secondary creep rate of
800 to 1000°F), has been shown in published 0.1 percent per hour for 95 ksi material at 800
literature, and is corraborated by the present data. and 850°F); these are presumably to be attributed
Because of the dependence of rupture strength to uncertainties in the positions and slopes of
of ~hese higher strength materials upon room tempera- the regression lines.
ture strength, any evaluation procedure, direct
or indirect, that involves the scatter band of all
14
Comparison of the average 100,000 hour at temperatures of 1000°F and higher and mav he
rupture strength for quenched and tempered material compared with the annealed category of wrought
at 1000°F, where it is essentially independent of and cast material.
room temperature strength, Fig. 27c, with the The time-for-rupture, secondary creep rate,
average 100,000 hour rupture strength of annealed and ductility data for the two foregoing categories
material at 1000°F, Fig. 22 reveals an interesting are plotted in Fi?s. 19 and 20, and where possible,
and surprising gap (approximately 21 ksi vs. 12 estimated 10,000 hour and 100,000 hour rupture
ksi). Significant differences may also be noted strengths, and .1 and .01 percent per 1000 hour
in the stresses for rupture in 10,000 hours and creep strengths have been developed and are
for secondary creep rates of 0.1 and 0.01 percent included in Tables VI to IX. Because of the
per 1000 hours. One possible reason for the character of the data, no attempt has been made
differences may be the generally higher carbon to develop trend curves for weld metal. However,
content of the quenched and tempered or normalized it is of interest to compare by inspection the
and tempered material compared to the annealed weld metal results with those developed earlier
material (see Tahles I and II). It is also in this report. Comparing first the strengths for
possible, of course, that the log-log relation at Codes numbered W 15 or lower with the scatter bands
1000°F curves downward beyond the limit of the of Fig. 22, it may be seen that with respect to
availahle data, and, if so, the difference in creep strength, three of the four lots of weld
100,000 hour rupture strength relative to the metal fall within the scatter band, and one lot
annealed material may be less than now indicated. (at 1000°F) falls slightly below. With respect
to 100,000 hour rupture strength, five data for
weld metal fall essentially within the scatter
Rupture Ductility band, and two lie significantly he low.
Comparing the results for lots numbered W 16
The elongation and reduction of area at and higher with the 100,000 hour rupture strength
rupture of the annealed material, Figures 15 a-e, data of Fig. 27c, the four data at 800°F fall
exhibited extensive scatter, especially for tests slightly below the scatter band; four of the data
at ll00°F. At no temperature was there evident at 850°F fall within the scatter band, and the
in the scatter band a well-defined trend with remaining point falls only slightly below the
increasing time-for-rupture, although it is scatter band; at 900°F, two data fall within the
possible to discern in some of the plots, e.g. at scatter band, and the remaining three fall only
1100 and 1200°F, some tendency for reduced slightly below the band. Thus, these limited
ductility at longer times. It is possible that data indicate weld metal rupture strengths trend-
changes in fracture mode from transgranular to ing less than those for base metal. There are no
intergranular with increasing time for rupture, as data for creep strength (0.01 percent per 1000
well as environmental effects, may contribute to hours).
the scatter, but information that would permit con- The rupture ductility of weld metal tended
sideration in such terms is lacking. to fall on the low side of the scatter hand, and
For the normalized-and-tempered and quenched- in some cases below the scatter band, observed for
and-tempered materials, Fig. 18 a-e, rupture corresponding categories considered earlier in
ductility exhibited somewhat less scatter, but the this report.
number of data was substantially less. Elongation
was generally low at 800°F, a reasonable expecta-
tion for high strength material, and exhibited no Acknowled!!lllents
trend with time for rupture, nor any evident
difference amongst the three categorie5 dist ingm ;,hE>d The evaluations nf this report were made tor
in the plots. At 850°F, the few data exhibited a the '1etal Properties Counci 1 under the guidance
suggestion of decreased ductility at longer time. of a subcommittee of which Dr, M. Semchyshen is
At 900°F the average trend of elongation remained chairman. Appreciation is expressed to the memhers
essentially unchanged, but with scatter increasing of that subcommittee for helpful suggestions.
beyond 1000 hours. At 1000°F, a number of low The Council expresses its gratitude to the
elongation values (less than 10 percent) may be many contributors of data.
noted for rupture times beyond 1000 hours, with a
few specimens fracturing at gage punch marks, all
suggesting a tendency for notch sensitivity. The References
low elongation values have their reflection in
reductions of area less than 10%. 1. Report on the Elevated-Temperature Properties
of Chromium-Molybdenum Steels; ASTM DS 6
(formerly STP No. 151) (1953).
Creep and Rupture of Weld Metal
2. Supplemental Report on the Elevated-Temperature
Unfortunately, it has not been possible to Properties of Chromium-Molybdenum Steels; ASTM
identify adequately many of the lots of weld metal DS 6Sl (1966).
for which creep and rupture test results are
available, and the relatively few data are there- 3. R. M. Brown, R. A. Rege and C. E. Spaeder, Jr.:
fore considered separately. As with yield and The Properties of Normalized and Tempered and
tensile strength, data for weldments have not been of Quenched and Tempered 6-1/4-inch Thick Plates
considered. The creep and rupture data for weld of 2-1/4 Cr-1 Mo Steel; Symposium on Heat
metal may be divided into two categories. The data Treated Steels for Elevated Temperature Service,
for lots numbered W 16 and higher were generated ASME, 1966, page 27.
at temperatures of 800 to 900°F, and represent
quenched and tempered or tempered material. Data
for lots having lower code numbers were generated
15
4. L. I. Mandich, E. L. Fogleman and J. A. Gulya: 26. An Evaluation of the Elevated Temperature
Elevated-Temperature Properties of Heavy Gauge Tensile and Creep-Rupture Properties of
Cr-Mo Steel Plates; ibid, page 45. Wrought Carbon Steel; ASTM Data Series
DS llSl, (197.0).
s. o. R. Carpenter and G. N. Ernmanual: The Use
of Quenched and Tempered 2-1/4 Cr-1 Mo Steel 27. G. V. Smith: Strength of 2-1/4 Cr-1 Mo Steel
for the Fabrication of Advanced Design at Elevated Temperatures; presented before
Pressure Vessels; ibid, page 132. the Petroleum -- Pressure Vessels and Piping
Division of ASME, Sept. 1970, Denver,
6. Resume of High Temperature Investigations Colorado.
Conducted-During 1963-65; The Timken Roller
Bearing Co., Steel and Tube Division, 1965. 28. r,, V. Smith: Evaluation of Elevated-Temperature
Strength Data; Jnl. of Materials 4, Nov. 4,
7. C. J. P. Steiner, J. J. de Barbadillo, Dec. 1969, pp. 878-908. -
A. W. Pense and R. D. Stout: The Creep
Rupture Properties of Pressure Vessel Steels - 29. S. S. Manson: Time-Temperature Parameters
Part III; Weld. J. Rsch. Supp. 47, p. 145s- A Re-Evaluation and Some New Approaches;
154s, (1968). - published in "Time-Temperature Parameters
for Creep-Rupture Analysis; Amer. Soc. for
8. A. O. Schaefer: Review of Available Data on Metals Publication No. D8-100 (1968).
Steel in Heavy Sections for Pressure Vessel
Applications; unpublished work, Pressure
Vessel Research Committee, Welding Research
Counci 1, 1964.
16
Table I P
P-lOd II
Ql725, Tll50 II
Tl275
P-12 4-1/8
P-13 4-1/8
P-14a A542 Ql725, Tll50 4
P-14b II
Ql 725 f Tl200 4
P-15a Ql 725 f Tll50 6-1/4
P-15b A542 Ql700, Qll50 6-1/2 Ref. 16
II II
P-15c II
17
Cont. Table I p
II II
P-28 Nl750, Tl300, 6-1/4
Tl230
II II II II II
P-29
II II
P-30a Nl750, Tl325 I 1-1/4
Tl250
II II II
P-30b II
2
II
P-3la II
Nl750, Tl360 6-1/4
P-3lb II
Nl750, Tl420 6-1/4 II
II
P-36b II
4
P-36c 2
P-37 Ql750, TllOO 4-7/16
P-38 A542 Q+T at mill~ 7-1/4 Ref. 9
T975, Tl080
P-39 A542 Ql750, •r1150 I 7-1/4 Ref. 9
T975, Tl085
P-40 A542 Ql 775 I Tl060, 4-5/8 Ref. 10
Tll25 (6h)
II II
P-41 Ql775, Tl060, II
Tll50(10h)
II
P-42 II
Ql775, Tl060, II
Tl275(15h)
P-43 A387 F.G. Ql750, Tl050 7-1/2* Ref. 11
II
P-44 II
Ql750, Tl200
P-45 II II
Ql750, Tl350 II II
P-52c II
Ql725, Tll25,Tl250 II II
18
Table I T
19
Cont. Table I T
T-32 " II
"
T-33 Al550 2-1/2 x Ref. 12
.523
T-34 CD, Al650 5.75x.855 Ref. 19
T-35 "
T-36 II
2.Sx.48 II
T-39 II
"
T-40 "
T-41 "
T-42 II
T-43 "
T-44 Al650 8.x.16 Ref. 20
T-45 Al650 8.x.16 Ref. 20
T-46 Nl600, Tl325 18xl-5/16 Ref. 17
T-47 Nl675, Tl275 II
"
T-48 A213 Nl650, Tl400 13. SOD x "
.358W
T-49 Nl650, Tl400 1. 315x. 787 "
T-50 " Al600* 2.x.388 "
T-51 " II
T-52 " II
"
T-53 " Al600** 2x.36 "
II
T-54 " "
T-55 " " II
"
T-56 " " " "
T-57 " " " "
T-58 " " II
20
Cont. Table I T
T-99 2. 500x.148
T-100 1. 875x. 378
T-101 2.50x.18
T-102 2.50x.203
T-103 2.125x.320
T-104 2.50x.203
II
T-105 1. 875x. 375
T-106 II
II
T-107 2.50x.203
T-108 II
2.50x.148
T-109 A335 Al700 7-8 lOxl. 97W Re . 15
21
Table I - B
Table I - F
F-2 II
Nl640, Tl240 130Dxl5ID x
12 Wide
F-3 Nl700, Tll50 6-7 56Dxl0t
F-4 Al82- Fine- Ql750, Tl275, 7-8 11 x 11 Ref. 22
F22 Grain Tl225, Tl075,
Tl290
F-5 Nl730, Tl225, II
13 x 13
Tl075, Tl250
F-6 II
Nl730, Tl315,
Tl250
F-7 II II
Ql750, Tl275 12 x 12 II
F-8 II
Ql775, Tl225, 6-7
Tl225, Tl075
*Electric furnace, vacuum degassed. Specimens were taken from extra metal
on forgins.
(1) HR-Hot Rolled; T-Tempered; N-Normalized; A-Annealed; CD-Cold Drawn;
Q-Quenched.
(2) Actual grain size except when identified as M for McQuaid-Ehn
(3) Specimens from 1/4 thickness position unless otherwise noted.
(4) DS6-ASTM Spec. Tech. Pub. No. 151 (1953): DS6-Sl-ASTM Data Series
Supplement (1966).
22
Table I - C
23
Table I - w
Identification of 2-1/4 Cr-1 Mo Weld Metal
Code No. Process Preheat Post Heat(!) Electrode Electrode Data Source
Temp. °F Treatment Size Type Source Code No.
24
Table II-P
Code No. c Mn p s Si Cr Mo Ni Cu Al N
25
Table I I - T
Code No. c Mn p s Si Cr Mo Ni Cu Al N
26
Cont. Table I I - T
Code No. c Mn p s Si Cr Mo Ni Cu Al N
27
Table I I - B
Code No. c Mn p s Si Cr Mo Ni Cu Al N
Table I I - F
Code No. c Mn p s Si Cr Mo Ni v Cu
Table I I - c
Chemical Composition of 2-1/4 Cr-1 Mo Cast Steels
Code No. c Mn s p Si Cr Mo Ni Cu Al
Table I I - w
Chemical Composition of 2-1/4 Cr-1 Mo Weld Metal
Code No. c Mn p s Si Cr Mo Ni Cu
W-16 .09 • 55 .008 .018 .22 2.40 .95
W-17 . 07 1. 06 • 017 .022 .33 2.16 1. 00
W-18 .095 . 98 .021 .018 .38 2.23 1. 02
W-19 .10 .94 .019 .020 .34 2.27 1. 05
w-20 .14 .71 .019 .022 .19 2.38 1. 00
W-21 . 073 .90 . 019 . Oll .53 2.20 l . 00 .13 .27
w-22 .11 .96 .016 .020 .41 2.23 1. 00 .08 .33
W-23 .14 .68 . 021 . 017 .20 2.39 1. 02 .14 . 28
W-24 .08 .81 .026 .012 .43 2.16 1. 00 .19 .47
W-25 .15 .74 • 023 .019 .17 2.23 .99 .13 .25
28
Table III - p
Code No. Test Yield Strength Tensile Strength Elong. Red. Area
Temp. °F
29
Cont. Table III - p
Code No. Test Yield Strength Tensile Strength Elong. Red. Area
Temp. °F
30
Cont. Table I I I - p
Code No. Test Yield Strength Tensile Strength Elong. Red. Area
Temp. °F
31
Cont. Table III - p
Code No. Test Yield Strength Tensile Strength Elong. Red. Area
Temp. °F
32
Cont. Table I I I - p
Code No. Test Yield Strength Tensile Strength Elong. Red. Area
Temp. °F
33
Cont. Table I I I - p
Code No. Test Yield Strength Tensile Strength Elong. Red. Area
Temp.°F
34
Cont. Table III - p
Code No. Test Yield Strength Tensile Strength Elong. Red. Area
Temp. °F
35
Cont. Table III ., p
Code No. Test Yield Strength Tensile Strength Elong. Red. Area
Temp. °F
36
Cont. Table III - p
Code No. Test Yield Strength Tensile Strength Elong. Red. Area
Temp. °F
P-52a 75 131.0
P-52b 75 111.0
r'-52c 75 97.0
37
Table III - T
Code No. Test Yield Strength Tensile Strength Elong. Red. Area
Temp. °F
*Data appear unreasonable, and have been excluded from the evaluations.
38
Cont. Table I.II. -· T
Code No. Test Yield Strength Tens.ile Strength Elong. Red. Area
Temp.°F
39
Cont. Table III - T
Code No. Test Yield Strength Tensile Strength Elong. Red. Area
Temp. °F
40
Cont. Table III - T
Code No. Test Yield Strength Tensile Strength Elong. Red. Area
Temp. °F
41
Table III - F
Code No. Test Yield Strength Tensile Strength Elong. Red. Area
Temp. °F
42
Table III - c
Short Time Tensile Properties of 2-1/4 Cr-1 Mo Steel-Castings
Code No. Test Yield Strength Tensile Strength Elong. Red. Area
Temp. °F
43
Cont. Table III - c
Code No. Test Yield Strength Tensile Strength Elong. Red. Area
Temp. °F
44
Table III - w
*Elongation in l"
45
Table IV - P
(1)
Code No. Temp. °F Stress, ksi Test Min. creep At Rupture-%
Duration- Rate-%/hr. Elong. Red. Area
Hours
46
Cont. Table IV - P
Code No. Temp. °F Stress, ksi Test (l) Min. Creep An Rupture-%
Duration- Rate-%/hr. Elong. Red. Area
Hours
47
Cont. Table IV - P
Code No. Temp. op Stress, ksi Test (l) Min. Creep At Rupture-%
Duration- Rate-%/hr. Elong. Red. Area
Hours
48
Cont. Table IV - P
49
Cont. Table .IV - p
50
Table IV - P
51
Cont. Table IV - P
* Test discontinued
52
Table IV - T
Creep and Rupture Data for 2-1/4 Cr-1 Mo Steel - Pipe and Tube
53
Cont. Table IV - T
54
Cont. Table IV - T
55
Cont. Table IV - T
56
Cont. Table IV - T
57
Cont, Table IV - T
58
Cont. Table IV - T
59
Cont. Table IV - T
. '
60
Cont. Table IV - T
Code No. Temp. °F Stress, ksi Test Min. Creep At Rupture
Duration Rate-%/hr. Elong. Red. Area
Hours
61
Min. Creep At Rupture
Du rat ion Rate-%/hr. Elong. Red. Area
Hours
15.5 686.
l3 s 1819.
62
Tnble IV - B
Table IV -c
Creep and Rupture Data for 2-1/4 Cr-1 Mo - Castings
C-14 950 36. 129. .034 40.
28. 1037. .OCJ45 47.
63
Table IV - w
Creep and Rupture Data for 2-1/4 Cr-1 Mo - Weld Metal
64
Cont. Table IV - w
65
Table V
75 1. 00 1. 00 1. 00 1. 00 1. 00 1. 00 1. 00
100 .980 .978 .990 .985 .987 .980 .982
200 .926 .917 .965 .957 • 960 .924 .935
300 .902 .870 .941 .934 .937 .895 . 902
400 .895 .850 • 920 .917 .918 .877 .881
500 .899 .826 .900 .903 .902 .861 .866
600 .905 .810 • 877 .887 .885 .843 .852
700 . 905 .790 .851 .866 .864 .823 . 832
800 • 891 .762 .821 .835 .835 • 796 .803
900 . 856 • 722 .785 .790 .795 .757 .757
1000 .790 .660 .740 .729 .741 • 691 .689
llOO .688 .577 .575 .594
1200 .540 .371 .467
1300 .338
75 1. 00 1. 00 1. 00 1. 00 1. 00 1. 00 1. 00
100 . 980 .980 .985 .988 .986 .980 .985
200 • 920 .921 .949 .958 .953 .941 .935
300 .882 .889 .923 .938 .930 .899 .908
400 . 882 .875 .905 . 925 .912 .877 .896
500 . 915 . 870 • 891 .914 .905 .873 .889
600 . 962 .864 .876 .901 • 892 .873 .881
700 1. 000 .851 .855 .881 .874 .860 .865
800 1. 001 .820 .825 .850 .844 .817 .832
900 . 947 .764 .779 .802 .798 .737 .774
1000 . 828 .673 • 715 .734 .731 .625 .685
llOO .653 .539 . 503 .556
1200 .454 .423 .380
1300 .293
66
Table VI
Temperature OF
Code No. 800 850 900 950 1000 1050 llOO ll50 1200
PART A - Annealed
Pi2e-Tube
67
Table VI - Cont.
Temperature op
Code No. 800 850 900 950 1000 1050 llOO ll50 1200
Bar
Cascings
C-1 9.8
C-2 19.0
C-3 14.0 8.2
C-14 21.0 15.9 9.7 7.3
C-15 20.9 16.7 9.7 8.0
68
Table VII
Temperature op
Code No. 800 850 900 950 1000 1050 llOO ll50 1200
PART A - Annealed
Pipe-Tube
Bar
B-3 12.6 5.6
B-5a 14.4
Castings
C-1 9.0 8.0
C-2 12.0 8.8
C-3 6.1
C-4 14.0 6.3
C-5 9.6 6.0
C-6 9.8
C-7 16.5
69
Table VII - Cont.
Temperature op
Code No. 800 850 900 950 1000 1050 llOO ll50 1200
70
Table VIII
Temperature op
Code No. 800 850 900 950 1000 1050 llOO ll50 1200
Part A - Annealed
Pi12e Tube
T-1 13. 4 6.8 2.8
T-3 12.5
T-7 12.4
T-8 11. 8 8.7 4.9
T-9 11. 0 8.1 6.0
T-10 9.2 7.5
T-ll 13. 9 7.6
T-13a 11.1 8.1
T-13b 8.4
T-13c 9.2
T-14 13.l 8.0 2.1
T-15 11. 0 7.4 2.5
T-16 12.0 6.4 2.7
T-17 13. 3 8.7 2.1
T-18. 10.0 6.5 2.4
T-19 6.0
T-20 7.9
T-21 5.2
T-22a 5.3
T-22b 7.7
T-22c 5.8
T-22d 7.3
T-23 10.l 5.7 1. 9
T-24a 9.6 5.6
T-24b 11. 7 7.1
T-25 12.8 8.2
T-26 12.2
T-33 9.1
T-44 12.0 6.8
T-46 9.1
T-48 10.l 7.2
T-49 7.0 5.8
T-50 8.1 6.1
T-51 7.5 5.6
T-52 7.9 6.0
T-54 5.9 4.8
T-56 6.8 5.9
T-58 5.5 5.7
T-109 16.0 13. 0 10.4
Bar
B-2 31. O* 11. 7 2.1
B-5a 10.0
B-5b 11. 3
B-6 5.3
71
Table VIII - Cont.
Temperature °F
Code No. 800 850 900 950 1000 1050 llOO 1150 1200
Castings
C-1 6.1
C-2 14.1
C-3 9.8 5.6
C-14 16.0
C-15
W-5 3.3
W-12A 12.0
W-12b 11. 4
W-14 4.6
W-15 6.7 4.0 1. 42
W-16 50.0 41. 0
W-21 56.0 28.0
W-22 70.0 45.0 36.0
W-23 66.0 52.0 33.0
W-24 80.0 59.0
W-25 54.0 41. 0
72
Table IX
Temperature °F
Code No. 800 850 900 950 1000 1050 llOO ll50 1200
PART A - Annealed
Bar
B-3
castinss
T-2 3. 8
P-15a 30.0
P-15c 37.5 21.0
P-15d 26.0
P-21 54.0 41. 0
P-23 105.0 68.0
P-24a ll8. 0
P-24b 79.0 44.0
P-40 72. 0 40.0 19.0
P-41 54.0 38.0 18.0
73
Table IX - Cont.
Temperature op
Code No. 800 850 900 950 1000 1050 llOO 1150 1200
74
Table X a
Average
Temp. 0 P Ari th. Mean Linear Ext. Strength- Parameter Scatter Band
of Indiv. of Isotherm. Temp. Regress. Larson- Manson
ExtraE· Regress. Line Individ. Extrap. Miller (C=20) Compromise
Table x b
75
Table XI a
Average
Temp. 0 P Ari th. Mean Linear Ext. Strength- Parameter Scatter Band
of Indiv. of Isotherm. Temp. Regress. Larson- Manson
ExtraE· Resi:ress. Line Individ. Extrap. Miller (C=20) Com12romise
----· ·-··-·-- ..
Minimum
---··--·---··-··--··-
- 90% Confidence
Table XI b
76
Table XII
77
with temperat.ure.
79
Pig. 7b - Variation of tensile strength of annealed
material with temperature.
80
Fig. ?c - Variation of elongation and reduction of area of
annealed material with temperature,
81
_,
;
,, +
;
i- ' : t'
-·jjl-1. ·' t
+4
·;·-r
~-1r;j
·1
~~~ ..
.: .1...,
- >'[
-t-
4 I- J_ff. +·i-·
. ~
]f + t"
1' .. ! . .-H
1'·-
j f -
82
·f:i·
:I
..._
~ ~
{- ~
~ ~i -~ t
'.
i.t
l.
:µ:
83
'f
-~- - +
84
;,- + 1
+'
85
Fig. 9b - Variation of tensile strength of quenched and
tempered material (TS<ll5 ksi) with temperature.
86
,1
'1 +. 1tt l .f:t .. -._·:,jt~-,:~ D~ -: ~H· " ,·
+
.r; 4 .. lil± ~'c,
H
J-
.J .....
r
IF- l:_ ~
, ' 't
+
r. ~1 .•
;- .. ~
',_"+i I
~,_t-L
:' ' ,!: ~fc: ii~ :._ T
f:i ii ~H~ e(i~ tl;
fr·:.i;t,"-'- LI. c
-r ~ 1{j_ f. ':'!]
• l:' l'
tf~ ,:I+ f-1-' !
Ii T
1 '
tH f!i i1f;
+: ... ·• jt~ \·t-t:r ~
1-:--
>:;: 1'1
1:1 ~+'- ' t'
-H
.~ ~;1 - ~~+· ._T.;-l' - _('j' 'if , .. ·:.:t ~~-I~
.u
~ ~~ T±t1 'ft} ~- -~ :i i~ '~ ' .
jl
·I~
·~
1
rll l~1J
·f- ·l~j .!"!:.!"
.+
-11 ~Ii~ : .1 ~ 1-' -I
1
·1-
~ .. :;n j
4 :j_ +-1·
di -.·•
1' i I
' ;·
-+
t: I
'
-i-r - ... 1
1-t t.i ri~
+ Ji.J' l
.Jf.-1.
,--r ' ii,, C
Pl
:l
:;.- \; '•li Le
'
t !i] 11 -· .I.
.:L-1-j ' + ~~: i:u t 1 ,:I ± ·I ±lf i
-'-"
w+ jtl1 "'t' m '·
j' :+
.- l+'r +. :t +
':"+-! l
rt:r - i }4:L~t
+ :t1t-f
Fig. 9c - Variation of elongation and reduction of area of
quenched and tempered material (TS< 115 ksi) with temperature.
87
Fig. lOa - Variation of yield strength of quenched and
tempered material (TS >115 ksi) with temperature.
88
Fig lOb - Variation of tensile strength of quenched and
tempered material (TS>ll5 ksi) with temperature.
89
., .l
...
'
j
"·
:i
f+ -!.
·I'-
90
Fig. lla - Variation of yield strength of normalized-and-tempered
and quenched-and-tempered cast material with temperature.
91
·;. r .; ~ i. ,
! :ti .... ~
:L
•__ 1.
'1
' .
'-±±! ~-.:-
tz.t t·
,. -
+·ti h{~·i t~\ Jt ~1-·
_:t. + -;
tr.-t~ I'Vi,. "'lct-· ,;L1 ::'', 'j
:!± ct ir W , +" , RT
d'. ; ++ .-u -+• ~· .µ 1Jt1i::if!1~ th. fl1 Hh :1 t~lf :tr
~l"tti~ i2i ri , .Ar:::& . '"f"11 ~1 >t ·~ +Hf."- PJ! -tili fi ' , f Hr;: ' . J ~t
:iX I .1--trffiYifJ t ~r+ -!1+~ 'r· fr'+-t1l~· -~.)- ~i+ i·* J1ff ·~~ ".f ij1
, lf;~ J~ ":r- , 1' :tlP , '' ~tt J1, '"1 t .,. , , >1 t t
cj1HJ1f,,,~ uJ .:H~ . , ·~~ ,:t: , ll:, ' WrffV 1- it
~l· :rj. .., + :r~: ,_LJ-1 fit+ ±:i- ::-# ~1 ~~t :w:· .i-1.·~~ t~ J.::~ .1.,t; ,,:+ 1"1-i J~··
r t ~ +. .. -~-:1., ft ~ :Jn . .. ;ii. JJi ~lil tt: ::~~ J-l :ffil
Fig. llb - Variation of tensile strength of normalized-and-tempered
and quenched-and-tempered cast material with temperature.
92
.,
t
-'_t __;J-~.
. r ~-
. it+ h .
-I tt~ -
93
·1 tqLif'iJ lm1, ni il d
µm11 1: j ' , lH• m- tl- J
..
'±i f fi\! r ffii !f,c
,_,1::-tJr:l:t
1,ihH!IH-iltt;t
"[
-:::· .t' t
•+
-·~ :[!j 'J,,,:f,
'. :l-l 1.;r~- ' t i
-·-! .: 4' _,_
; ·....:
.+ I
.. L _;:, ..
• ,...-1-
\. -·
''
94
+
95
Fig. 12c - Variation of elongation and reduction of area
of weld metal with temperature.
96
2 4 5 67891
·.-i
rn
p,
0 5
0
0 4
.-I
3
l.O
-..J
rn 2
rn
Q)
~
.p
(/)
I
9
8
7
6
5
4
101 100
9
8
7
6
5
4
10~
8
7
6
5
4
\0
\0 2 20
I 10
9
8
7
6
5
4
3:
1
10 100 1000 10,000 100,000
TIME FOR RUPTURE - hours
Fig, lJc - Stress vs time for rupture for annealed material.
1001-'I
9
8
7
6
5
4
10~
8
7
6
s:
4
-
0
0
2
11
9
10
8
7
6
5
4
1, l
100,000
TIME FOR RUPTURE - hours
Fig. 13d - Stress vs time for rupture for annealed material.
100
.rn
....;
.!C!
UJ 4
UJ
~ 3
E-<
0 UJ 2
119
B
7
6
5
4
1019 100
•rl 8
Cll 7
.lid 6
5
Ul 4
Ul
~
3
-
0
N
E-t
Ul 2
lt
9
10
8
7
6
5
4
1
• 01 .1 1.0 10
CREEP RATE - PER CENT PER 1000 HOURS
·r-1 3
ti)
p..
2
0
0
0
r-i
lt9 10
8
Ul 7
Ul
6
~ 5
8
Ul 4
-
0
w
3
I 1
9
8
7
6
5
4
.01 .1 1.0 10
CREEP RATE - PER CENT PER 1000 HOURS
Fig. 14c - Stress vs secondary creep rate for annealed material.
oc
Fig, 15a Variation of rupture ductility with time for rupture for annealed material,
00
Fig. 15b - Variation of rupture ductility with time for rupture for annealed material.
-
0
~
Fig. 15c - Variation of rupture ductility with time for rupture for annealed material.
0
-.J
Pig. 15·1 - Variati8n r~!' t'T~t'..ire d•.1ct11.1.: ·1 with time for rupture for annealed material.
0
00
00
Fig. 15e - Variation of rupture ductility with time for rupture for annealed material,
4 5 67!Hll 2 3 4567891 2 3 4567891 2 3 4 567891 2 3 4567891
9
~
7
6
5
4
1001
9
8
7
•.-! 6
ti)
5
.lid
4
-
0
l.O
('J]
('J]
~
E-t
('J]
2
101 100
9
8
7
6
5
4
10~ 100
•r-1 8
l:ll 7
.!ii::
6
5
Ul 4
Ul
~ 3
e-i
.....
..... Ul 2
0
I
9
10
8
7
6
s
4
•r-1 10 I
9
m 8
~
7
6
5
m
m 4
~ 3
~
m
2
I
9
8
7
6
5
4
3
100.9
•.-i 8
C/l 7
..!<: 6
s
(/)
4
(/)
r1l 3
~
-
N
8
(/) 2
1019 100
8
7
6
10
Cl]
~
Cl]
~ ;~
E-i
Cl]
...... ;ii;
w
0
1L
.~
·;r
'~
~~
·~
.~
:~
101
9
•.-4 B
fl) 7
.II: 6
5
t/l 4
t/l
~
3
e-i
.....
..... Cl] 2
.j::>.
100~
B
7
6
5
4
10~
...... B
{I)
7
Ji::
6
5
ti] 4
ti]
....N t1E--t 3
0 ti]
2
10019
B
7
6
5
4
101
1000 10,000 100,000
TIME FOR RUPI'URE - HOURS
Fig. 19a - Stress vs time for rupture for annealed weld metal.
1
1001
9
8
7
6
5
4
101 ·-00
9
8
ori 7
UJ 6
~ 5
4
CIJ
-
3
ell
N ~ 2
8
CIJ
1 I 0
9
8
7
6
5
4
10,000 100,000
TIME FOR RUPI'URE - HOURS
Fig. 19b - Stress vs time for rupture for annealed weld metal.
1 2 3 4 5 67891 2 3 4567891 2 3 4567891 2 3 4567891 2 3 4567891100
101
9
8
7
6
5
4~~
10
.......
!/J
.!<1
5
-
N
N 2
101
9
8
7
6
5
4
I 1 :'";
.01 .1 1.0 10
Fig. 19c - Stress vs secondary creep rate for annealed weld metal.
-
N
w
Fig, 19d - Variation of rupture ductility with time for rupture of annealed weld metal.
-"""
N
Fig. 19e - Variation of rupture ductility with time for rupture of annealed weld metal.
1001~
9
8
7
6
5
4
1019
8
•ri 7
Ul
.i.:: 6
5
4
{/)
{/) 3
-
N
VI
~
~
{/)
2
10019 0
8
7
6
5
4
1019 100
•rl
(!)
e
7
.!<: 6
5
Ul 4
-
Ul
~
3
N 8
0\
Ul 2
iooi 10
e
7
6
5
4
101
.1 1.0 10
CREEP RATE - PER CENT PER 1000 HOURS
Fig. 20b - Stress vs secondary creep rate for quenched-and-
tempered or tempered weld metal
0
00
"'
:;l
+>
p,
J_ Lu_; I LI I j I '' ! I .. _; 1.;.1_1 .i :. Lf_ __ i....! ..J
i -~--~-!-.%_,._~
'' :::s
~
I-<
0
'!--<
'" Ill.
4. ~-"'""""!' µ-·1 .I. ~-L.:. '-~1--.L..L'_'._l. •. ·! .' \ "' l j -~:-. -r_ .+-;t · ;_:~ '.:_.t.Ll_i_ J_.J. .i-LL.-LT'-:f-::E:-{:..::f.."~~-4-:~-l4:.d-it~l~::~f'1: ::r-:y:.J.a w
a
.µ .-1
+>
..<::;
2.. ~:..i:J...ii.J.! __ i ..t :.. : .:._ :).__+..~ -~--+-l_c'..:1 ~--l ..l -'---l-'-'-i. 1 __ ,__;:.L. !. ! .., .i ...;..i .. ; -l. :'-'---i 11-++--"-~:.J-c.i-! :J: .J::).__J.cL ,_ ~-1-1:-l +> QI
..-iIS
ls
'd
>..-i
+> QI
'-9~ -llEt~iL;:~-~-~J;-=
'"
.-1 !I:
El
QI
~
:;l
...
CD
+> H
p, 0
:::s
? .. LLL:LLLL: ;__ L_ l LL.~ -t : !. L e i !J_; l ~ .. .L;_.!_J:..;.::i..c::: c.!-::.. I_;_ : l ..:. _i_;, '.-!.. '· -LLLtt:J::_j_;j_-L:Li.:Ll:.±-l ..LLL:L.l..i ~'t:I
GI
fo-j ~
0 Q)
p.
s:: s
0 CD
'+-1 +"
+> I
Ill -d
..-Is::
~ aS
,_ !_:ELL:EEL: ' :L:c,-:c~::r:r::·. ; :1 L~-J~f-I 1~1- :.: I:~!::.:- _;_j-_J-::..t±.:l..--1..:__l_lLEL~:..:1:.E.:!.: . :..LLJ ~1:=r=r--r-.:1-i-+-FF-FFFF"EF±-l=f±=l Ill I
~·o
G>
I .C:
0
3. - LJ=-lcrd::L-"cl :tc:::tc::r_;__ i:-l•:L'. . .1£Uc:W:=io:;_.L!-~~~"'~c!=l-~Fr"--1"'-f'FPB::rttfi=l"Ft---t++++FH Cl>
OCI>
s::
N
.
tlO~
:::S
D'
'"0
~'H
129
10
9 '"'"
H ~~ -
·- .
E
+----+---!::::
.. - ~~
r-t=
- .
:::
~
~~ 1-1-:::
t-t-~
• - t - ,_,_~
!:::
!:::
!:::
§
t-
' -
"
.... .
,
)
I "" ...
m IB m
~
9
8
7
~ t-
t-
6 ::::
5
~5 0
4
t=
t-
3
t-t-~
ro
,... '
ro
. ..
~
I
9
8
z
"'
I I>.
10
5 '
5
4
2 2
" '-
1 1
130
10
§
~
:=
t:
,_
,.,
I ~
9
8 E
7 E
6
5
~
,_
50
4
E
~
3
I:<
... ~·
; ..
I<
A
)',.
I 10
9
8
7
6
I ~
5
=
= 5
4
- =
"
2
,_ 0
,_
rtl t>11K~ b 11 ~ Ii'
I
131
5 0 7
6
5
4 -.:=..-c
1
9
8
7
6
5
:;E: ~~:::-t-~
~~'cf=!t:tzt-~_fjt-~tl
4
'.;J L.
N 2
__
~~-~~~Ft--:=~
---1 ··:·-·r· -- -
-- -.- - .
~~ -~~=~-~~T----~ -- -
6 ~~~~~~~~~~~~~~~~~~~~1111!111w11~r~~~~~~~~~·ruirr_~~~~-~~~~~s~·-~,=~~~~~~~;~~
2
-.,0::
-'-"'· ·-E.
-F- _cl'=•-
,.c
: J>·· ·
,_ '.
:cLc
1= _. --l.
L:·
.. , _;_
"I; ,[ !- --
-1
~+-·
[,,,
. -I"
- :-r-'-'""
. ~ ·~- -- ~r
i-
:·.r-:-:.;:-:L--L ---f
t:;::"-J
t-
_-
.:+-~ .
-~ L--:t=--=· ~ :j::--:
I
- --r-- -1---
L- - -
1
; -r- r- r- 1---1--1--
I !-,__!-1-r-- l--1-- ..
- ·------ 1--1-
- -- -:---- -- -t-·+-l--
-+·-!-1---
- - ·- -- ---r-r- r- --1-1--
<-+-+-+-l--+-+--+--+-+-+-l--+-+--+--+-+-+-l-+-+-+--+--+-~~~~ ~ J. ,~ I ~~ ·: - 1
H-+++-t-t-t--+-++-+-t-+-+-++-+-t-+-+--1-.l--+--..H- --·- · -~ J;_:~
H-+-+--\--++-t--+-+-+--\--+-t--t--+-+-+--\--+-~--t---t-+-+--t--i-t---t---t--t-- --~r- - . ·f- -- ---r- -- 1--- r-1-1-1-
133
;=-
lo
.... ~~~
~-
HI
m --
L-+=1=1=
§§§
,_
~
,. §§§
- ~~
,,
~
'
,, '~
'~
~
,, ~
; ... ,,
-
~
,...
... . ,,
~
~
,...
I' I'
,, ... .,_
..............
[/) L.; I;' .,
"
L.
~I
8 9 ---
[/)
8
===§
7
5
Iii
~~~
2
,
'~
I
Fi g . 5 -stre·ss
2 Variation of Manson com romise arameter
for rupture of annealed material.
p p w th
134
10
H\
.. .
••
J.
U<U u
' ""
" "
8
II
7
'
3
.
- 0
1
Fig. 26 • Comparison of rupture strength 100,000 hours, trend
c.urv.es developed by several evaluation procedures.
135
Fig. 27a - Variation of rupture strength (10,000 hours) of
normalized-and-tempered and quenched-and-tempered material
with tensile strength at 75 F.
13b
f··-T· 't-;
,j -!--
-··· t
'.1
'
·Ii , f
!Jlc tr t
!111 1. f .-
c1 . -1. ~
f·" :-i • ·1~ i-
i-t-1- ·-t
' ::dl
Fig. 27b - Variation of creep strength (O.l %per 1000 hours)
of normalized-and-tempered and quenched-and-tempered material
with tensile strength at 75 F,
137
Fig. 27e - Variation of rupture strength (lOOrOOO hours) of
normalized-and-tempered and quenched-and-tempered material
with tensile strength at 75 F.
138
.,
f+J
l
.-1-""
:- -1-1
+ ~ i->-
:t '1-
. t Lf-
+ ~
139