NASA TECHNICAL NASA TM X-52270

MEMORANDUM

(ACCESSION NUMBER) ITHRU)

L
-
A METHOD OF ESTIMATING HIGH TEMPERATURE LOW CYCLE
FATIGUE BEHAVIOR OF MATERIALS
GPO PRICE $

CFSTl PRICE(S) $
by S. S. Manson and Gary Halford
Lewis Research Center
Hard copy (HC) .q-Gd
Cleveland, Ohio
Microfiche (MF) 1 -.
ff 853 July 85

TECHNICAL PAPER proposed for presentation at

Symposium on Thermal and High Strain Fatigue
sponsored by the Institute of Metals
London, England, June 1967
?

NATIONAL AERONAUTICS AND SPACE ADMINISTRATION WASHINGTON, D.C. 1967

A METHOD OF ESTIMATING HIGH TEMPERATURE LOW CYCLE

FATIGUE BEHAVIOR OF MATERIALS

by S. S. Manson and Gary Halford

Lewis Research Center

Cleveland, Ohio

TECHNICAL PAPER proposed for presentation at

Symposium on Thermal and High Strain Fatigue

sponsored by the Institute of Metals
London, England, June 1967

NATIONAL AERONAUTICS AND SPACE ADMINISTRATION

 A METHOD OF ESTIMATING HIGH TEMPERATURE LOW CYCLE FATIGUE

BEHAVIOR OF MATERIALS

by S. S. Manson and Gary Halford

Lewis Research Center

National Aeronautics and Space Administration
Cleveland, Ohio

ABSTRACT

A method i s described whereby s t a t i c t e n s i l e and creep-rupture

p r o p e r t i e s can be used t o estimate lower bound, average, and upper bound

low cycle f a t i g u e behavior i n t h e creep range. The method i s based

primarily on t h e method of universal slopes previously developed f o r

estimating room temperature f a t i g u e behavior, and i n p a r t on a highly

s i m p l i f i e d creep-rupture - f a t i g u e analysis. Reasonable agreement i s ob-

t a i n e d when t h e estimates a r e compared with t o t a l s t r a i n r a n g e - l i f e data

f o r numerous engineering a l l o y s . Included i n t h e study a r e coated and

uncoated nickel-base a l l o y s , a cobalt-base alloy, low and high a l l o y

s t e e l s , and s t a i n l e s s s t e e l s t e s t e d under laboratory conditions over a

wide range of temperatures and cyclic r a t e s .

INTRODUCTION

Low cycle s t r a i n - c o n t r o l l e d f a t i g u e i s now recognized as an important

f a i l u r e mode i n some components of gas turbines, nuclear r e a c t o r s and

other high-temperature s y s t e m . Eie &xcmt of r e ~ d l l j r~ ~ ~ i 1 ~d 5s lt .~SI?

e

l a b o r a t o r y samples i s s t i l l q u i t e limited, and hence, estimates of t h e

performance of s p e c i f i c materials of engineering i n t e r e s t a r e very useful

i n t h e e a r l y s t a g e s of design and materials s e l e c t i o n .

TM X-52270
 Several s t u d i e s have r e c e n t l y been described t h a t provide estimates

of t h e low cycle f a t i g u e c h a r a c t e r i s t i c s of materials a t elevated tempera-

t u r e s (e.g., r e f s . I and 2 ) . The simplified approach of reference 2 i s

based on the method of universal slopes t h a t had previously been developed

f o r estimating room temperature f a t i g u e behavior of small laboratory

samples ( r e f s . 3 and 4 ) . Since t h e universal slopes equation contains

constants obtainable from only simple t e n s i l e t e s t p r o p e r t i e s - elastic

modulus, ultimate t e n s i l e strength, and percent reduction of a r e a - a

f i r s t attempt w a s t o apply t h i s equation using t h e t e n s i l e p r o p e r t i e s a t

t h e high temperature of i n t e r e s t . For t h e few cases i n which a d i r e c t

comparison w a s possible, it w a s found t h a t t h e l i v e s predicted i n t h i s

way were s u b s t a n t i a l l y higher than were r e a l i z e d experimentally. The

discrepancy was e s p e c i a l l y apparent when t h e high temperature f a t i g u e

cracks were i n t e r c r y s t a l l i n e . Several procedures were then examined f o r

modifying t h e equation f o r high temperature use, based l a r g e l y on t h e

concept t h a t i n t e r c r y s t a l l i n e cracks, occurring as a "creep e f f e c t " ,

serve a s f a t i g u e nuclei. Some of t h e procedures involved separating t h e

crack i n i t i a t i o n and propagation s t a g e s of f a t i g u e and applying a d i f f e r e n t

cycle reduction f a c t o r t o each. However, t h e procedure designated a s t h e

"l@ R u l e " avoided t h e d i f f i c u l t i e s involved i n t h e s e p a r a t i o n of t h e

two stages by applying a s i n g l e f a c t o r t o t h e e n t i r e l i f e r a t h e r than

d i f f e r e n t f a c t o r s t o each stage. Only a f e w materials were evaluated,

however, l i m i t i n g t h e g e n e r a l i t y of conclusions drawn i n t h a t r e p o r t

(ref. 2).
 3

This r e p o r t represents a continuation of t h e study t o provide simple

modifications t o t h e universal slopes equation f o r estimating high

temperature s t r a i n - c o n t r o l l e d low cycle f a t i g u e behavior of laboratory

s i z e samples. By examining t h e behavior of numerous high temperature

engineering materials t e s t e d over wide ranges of conditions, more meaning-

f u l conclusions a r e derived regarding t h e v a l i d i t y of t h e universal slopes

equation and modifications t h a t can be applied t o it t o improve i t s

accuracy. It i s believed t h a t t h e conclusions drawn from analyses of

t h e many materials and t e s t variables examined should provide valuable

guidelines f o r t h e estimation of t h e high temperature f a t i g u e behavior

of new m a t e r i a l s as well. The r e s u l t s of our study, and t h e s i m p l i f i e d

r u l e s derived therefrom, are described i n t h e following s e c t i o n s .

METHOD
"he method t h a t has evolved from t h e study i s based on observations

made i n attempting t o apply t h e universal slopes equation, and modifications

t h e r e t o , t o numerous s e t s of high temperature f a t i g u e data. On t h e b a s i s

of t h e s e observations s e v e r a l r u l e s of estimation a r e established t h a t a r e

believed t o be u s e f u l . The method i s as follows:

For s e l e c t e d values of t o t a l s t r a i n range, A€t, determine c y c l i c l i f e

values, Nf, according t o t h e universal slopes equation ( r e f . 3)

 4

where

uu = ultimate t e n s i l e s t r e n g t h

E = modulus of e l a s t i c i t y

D = ductility = In where RA i s t h e percent reduction of

area as measured i n t h e conventional t e n s i l e t e s t

The p r o p e r t i e s au, E, and D a r e determined a t t h e high temperature of

i n t e r e s t a t moderate s t r a i n r a t e s normally employed i n conventional

t e n s i l e t e s t i n g (say, a r b i t r a r i l y , within a f a c t o r of 10 of t h a t s p e c i f i e d

by ASTM Standards). For t h e most p a r t , estimates of c y c l i c l i f e can be

based on c e r t a i n percentages of Nf as determined from equation (1).

However, extreme cases involving low frequencies and r e l a t i v e l y high

temperatures occasionally a r i s e when t h e failures a r e primarily time de-

pendent and it becomes necessary t o make an a d d i t i o n a l estimation of l i f e ,

N;, according t o

Ni
-- Nf
m+O .12

Equation ( 2 ) r e p r e s e n t s a highly simplied attempt t o account f o r t h e

creep-rupture damage a s s o c i a t e d with t h e a p p l i c a t i o n of stress a t high

temperature. It should be considered as an i n t e r i m approach owing t o t h e

assumptions made i n deriving t h e equation. These assumptions and t h e i r

l i m i t a t i o n s a r e discussed i n t h e Appendix. I n t h i s equation

 5

k = e f f e c - i v e f r a c - - m of each cycle f o r which t h e m a t e r i a l may be con-

s i d e r e d t o be subjected t o t h e m a x i m u m s t r e s s . For reasons given

i n t h e Appendix, k i s taken t o be 0.3 i n t h i s r e p o r t .

F = frequency of stress application, cycles per minute (0.0167 Hz) used

i n t h i s report.

A = c o e f f i c i e n t characterising a time i n t e r c e p t of t h e creep-rupture

curve of t h e material a t t e s t temperature. The curve of stress,

ur, a g a i n s t rupture time, trJ i s l i n e a r i z e d on logarithmic

coordinates, and represented by t h e equation

m
crr = 1.7’5cru@) , so that A i s t h e time i n t e r c e p t a t an ex-

t r a p o l a t e d value of or = 1 . 7 5 oU.

m = slope of creep-rupture l i n e (negative value)

Nf = l i f e calculated from equation (1).

Equation ( 2 ) i s applicable only i f Ng < 1 6 Nf and i f Nk < lo5

cycles. Figure 1 provides a simple c r i t e r i o n f o r determining whether t h e

use of equation ( 2 ) i s necessary. If t h e p o i n t representing t h e coordinates

m and AF which apply t o t h e t e s t conditions of t h e material l i e s above

t h e curve and above t h e lowest coordinates shown, i s computed according

t o equation ( 2 ) ; i f below t h e curve, N i need not be computed.

By use of equations (1)and (Z), estimates of elevated temperature

low cycle f a t i g u e behavior can b e made as follows:

1. A s an estimate of t h e lower bound of l i f e use e i t h e r 1% Nf or

N i , whichever i s t h e lower.
 6

2. A s a n estimate of average l i f e use two times t h e lower bound l i f e . I

3. A s a n estimate of upper bound of l i f e use 1 0 times t h e lower

bound l i f e .

The application of t h i s method t o numerous s e t s of f a t i g u e d a t a w i l l

be discussed i n a l a t e r section.

SOURCES OF DATA FOR ANALYSIS

T o evaluate t h e method, high temperature f a t i g u e data have been c u l l e d

from many sources ( r e f s . 5 t o 1 4 ) such a s e x i s t i n g l i t e r a t u r e , t e s t s

conducted i n t h e authors' laboratory and previously unpublished contribu-

t i o n s generously submitted by various i n v e s t i g a t o r s . Forty s e t s of data

a r e included i n t h e analysis comprising 354 d a t a p o i n t s i n t h e l i f e range

from 10 t o lo5 cycles a t temperatures above 1000° F (811' K) . Table I

p r e s e n t s the alloys, t e s t conditions, and t h e p e r t i n e n t t e n s i l e and

creep-rupture properties a t t h e f a t i g u e t e s t temperature. Usually t h e

t e n s i l e properties were c i t e d i n t h e reference source. I n t h e cases where

they were not supplied, handbook or other reference source d a t a were used

as noted i n t h e t a b l e . The t e n s i l e p r o p e r t i e s a r e presumed t o have been

obtained from reasonably standard t e s t s , s i n c e i n some cases t h e authors

d i d not c i t e complete information.

The materials include low and high a l l o y s t e e l s , s t a i n l e s s s t e e l s , a

cobalt-base a l l o y , and various nickel-base a l l o y s both i n t h e coated and

uncoated condition. Ultimate t e n s i l e s t r e n g t h s ranged from 1 6 t o 182 k s i

(l.lX108 t o 12.5X108 N/m2), and reductions of area from 3 t o 94 percent.

A l l t e s t s were conducted i n uncontrolled a i r environments a t homologous

temperatures ( r a t i o of t e s t temperature t o melting p o i n t , both i n degrees

 7

absolute) ranging from 0.48 t o 0.88. T e s t frequencies ranged from 0.017

t o 50 cycles per minute (0.00028 t o 0.83 Hz), with some t e s t s involving

dwell times a t maximum s t r a i n of a s long as 3 hours (10,800 s ) .
I n most cases t h e f a t i g u e t e s t s were conducted i n a x i a l push-pull,

c o n t r o l l i n g e i t h e r t h e l o n g i t u d i n a l or diametral t o t a l s t r a i n range.

These tests a r e thus very similar t o those reported i n reference 3 on

which t h e method of u n i v e r s a l slopes was based. I n only two cases, t h e

Nimonic a l l o y s i n v e s t i g a t e d by Forrest and Armstrong ( r e f . 11) and t h e

Cr-Mo-V s t e e l reported by Coles and Skinner ( r e f . 6 ) , were t h e t e s t s

conducted i n plane bending - although s t i l l s t r a i n - c o n t r o l l e d .

ANALYSIS OF DATA
The l o w cycle f a t i g u e d a t a a r e p l o t t e d i n f i g u r e s 2( a ) t o ( f ) as

t o t a l s t r a i n range versus cycles t o f a i l u r e on logarithmic coordinates.

Symbols representing t h e data a r e l e f t open i f t h e mode of cracking was

not i d e n t i f i e d i n t h e reference source; shaded i f t h e cracking w a s

d e f i n i t e l y i d e n t i f i e d as t r a n s c r y s t a l l i n e ; o r an X i s enclosed i n t h e

d a t a symbols i f t h e mode of cracking w a s d e f i n i t e l y i d e n t i f i e d as i n t e r -

crystalline. Also shown i n each f i g u r e i s a s e t of curves representing

estimates made by t h e proposed method. The lowest curve shown with each

s e t of d a t a i s t h e estimate of t h e lower bound using t h e rule associated

with lC$ Nf or N i ; t h e middle curve i s t h e estimate of t h e average

behavior; and t h e upper curve i s t h e estimate of t h e upper bound. The

dashed s e c t i o n s of t h e curves ( f i g s . 2(d) and ( e ) only) i n d i c a t e t h a t

equation ( 2 ) w a s used i n making t h e estimates, whereas t h e continuous

curves represent estimates based on equation (1).

 8

Examination of f i g u r e s 2 ( a ) t o ( f ) shows t h a t a reasonably good

estimate of t h e experimental r e s u l t s can be achieved by t h e method outlined.

Since t h e r u l e s were, t o some extent, derived from examination of t h e data

i n t h e s e figures, some agreement i s inherent. However, recognizing t h a t

only a l i m i t e d number of e a s i l y determined t e n s i l e and creep-rupture

p r o p e r t i e s a r e required i n making t h e estimates, t h e agreement with experi-

mental data i s g r a t i f y i n g . Hence, a degree of confidence i s established

i n t h e p o s s i b i l i t y of applying t h e s e estimates t o other materials o r t e s t

conditions.

It i s a l s o seen i n f i g u r e s 2 ( a ) t o ( f ) t h a t , f o r a given temperature

and s t r a i n range, l i f e decreases as t h e cycling frequency i s decreased

( f i g s . 2 ( b ) , ( d ) , and ( e ) ) , and t h a t t h e lower frequencies seem t o promote

i n t e r c r y s t a l l i n e cracking. When i n t e r c r y s t a l l i n e cracking occurs, t h e

l i v e s l i e closer t o t h e lower bound than when t h e f a i l u r e s a r e d e f i n i t e l y

transcrystalline. I n t h e r e l a t i v e l y few cases f o r which t h e creep-rupture

c a l c u l a t i o n (eq. ( 2 ) ) w a s applicable, t h e f a i l u r e s were always i n t e r -

c r y s t a l l i n e ( f i g s . 2 ( d ) and ( e ) ) . However, occasional l o w l i v e s were ob-

served where i n t e r c r y s t a l l i n e cracking was dominant, even though t h e

creep-rupture properties d i d not i n d i c a t e a need f o r t h e Ni calculation

( f i g s . 2 ( a ) , ( d ) , and ( f ) ) . Whether t h e s e r e s u l t s a r e due t o inadequate

creep-rupture information (usually, handbook values were used t o estimate

creep-rupture behavior), o r whether they a r e due t o inadequacies of t h e

method requires f u r t h e r study. Some of t h e l i m i t a t i o n s of t h e method a r e

discussed l a t e r i n t h e r e p o r t .
 9

Although f i g u r e s 2 ( a ) t o ( f ) provide a v i s u a l p i c t u r e of t h e r e l a t i v e

v a l i d i t y of t h e method when applied t o individual materials, a more com-

p l e t e o v e r - a l l evaluation of t h e method can be made w i t h t h e a i d of f i g u r e 3 .

This f i g u r e shows a d i r e c t comparison of t h e observed with t h e estimated

f a t i g u e l i v e s f o r a l l of t h e materials c o l l e c t i v e l y . The abscissa i s

t h e estimated average l i f e , determined according t o i t e m ( 2 ) of t h e s e c t i o n

describing t h e METHOD. On t h e ordinate i s p l o t t e d t h e observed l i f e . The

symbols represent t h e experimental data points, using t h e same conventions

as i n f i g u r e s 2 ( a ) t o ( f ) . Perfect agreement between experimental r e s u l t s

and estimates computed from t e n s i l e and creep-rupture p r o p e r t i e s would be

obtained i f a l l t h e data p o i n t s coincided with t h e 4

5' line AB. The
lower l i n e CD represents t h e estimated lower bound on l i f e , according t o

item (l), and t h e upper l i n e EF i s t h e estimated upper bound according

t o i t e m (3).

The t a b l e i n f i g u r e 3 provides an a n a l y s i s of t h e degree t o which t h e

estimated average l i v e s and upper and lower bounds agree with t h e experi-

mental data. For 97% of t h e d a t a points, t h e l i n e EF serves as an upper

bound, and of t h e remaining 3%, a l l but two of t h e points l i e within a

f a c t o r of 1.5 g r e a t e r than t h e upper bound. The l i n e CD serves as a

lower bound f o r 81% of t h e data, and t h e d i s t r i b u t i o n of t h e remaining 19%

i s t a b u l a t e d i n t h e f i g u r e a s a function of t h e l i f e f a c t o r below t h e

lower bound. Since most of t h e points f a l l i n g below CD a r e within a

l i f e f a c t o r of 2 below it, a more conservative estimate of t h e lower bound

i s given by C'D' (5% Nf or 1 / 2 Ni, whichever i s lower). This more

conservative lower bound embraces 95% of t h e data. The t a b l e a l s o l i s t s

 10

the distribution of t h e d a t a p o i n t s from t h e average l i f e l i n e AB by

l i f e f a c t o r s of 2, 5, and 10. Here t h e l i f e f a c t o r r e f e r s t o e i t h e r an

over- or under-estimation of l i f e . This i s t h e same as was done i n

reference 3 f o r evaluating t h e a p p l i c a b i l i t y of t h e method of u n i v e r s a l

slopes t o t h e estimation of t h e average room temperature f a t i g u e behavior

of 29 materials. The t a b l e shows t h a t t h e v a l i d i t y of t h e high tempera-

t u r e method f o r estimating average l i f e behavior i s comparable t o t h a t of

t h e method of u n i v e r s a l slopes when applied t o room temperature r e s u l t s .

DISCUSSION

The method as described f o r estimating elevated temperature low

cycle f a t i g u e behavior i s highly simplified, and does not consider many

of t h e complicating f a c t o r s t h a t influence t h e low cycle f a t i g u e

c h a r a c t e r i s t i c s i n t h e creep range. To o b t a i n g r e a t e r accuracy more

a t t e n t i o n must be paid t o t h e d e t a i l s of t h e f a t i g u e and creep processes

and t h e f a c t o r s t h a t a f f e c t t h e l i f e . The importance of refinements and

t h e b e s t procedures a r e not y e t c l e a r ; however, a few p o s s i b l e approaches

may be outlined a s a guide f o r f u t u r e study.

One approach might be t o r e t a i n t h e s i m p l i c i t y of t h e u n i v e r s a l slopes

method, but t o take cognizance of t h e f a c t t h a t t h e t e n s i l e p r o p e r t i e s

t h a t e n t e r i n t o t h e equation depend not only on temperature b u t on s t r a i n

r a t e , hold-times, and environment as well. Thus, i f t h e appropriate

t e n s i l e properties could be determined, t h e form of t h e equation would

be retained, and only t h e constants would be a l t e r e d . This i s similar

t o t h e approach adopted by Coffin ( r e f . 1) f o r a s t r a i n - a g i n g m a t e r i a l i n

which s t r a i n r a t e was introduced a s a primary consideration. Such an

,
 11

1 . ' approach requires, however, t h e generation of a d d i t i o n a l experimental data.

Alternatively, it may be u s e f u l t o redefine t h e terms t h a t e n t e r i n t o

t h e equations. For example, it may be u s e f u l t o determine d u c t i l i t y i n

terms of t h e s t r a i n which causes surface cracking r a t h e r than rupture, as

suggested by Wells and Sullivan ( r e f . 12), and t o replace t e n s i l e s t r e n g t h

by some creep-rupture s t r e n g t h (e.g., t h e s t r e s s t o cause rupture i n a

s p e c i f i e d time). Such an approach, however, w i l l r e q u i r e extensive and

accurate data before t h e necessary c o r r e l a t i o n s can be made and applied

with confidence t o new materials.

Another p o s s i b i l i t y i s t o take i n t o account some refinements

a s s o c i a t e d with t h e f a t i g u e process which a r e of s p e c i a l importance a t

high temperature. For example, i n reference 2, we considered t h e r e l a t i v e

importance of crack i n i t i a t i o n and propagation on l i f e a t high tempera-

t u r e , and i n p a r t i c u l a r how i n t e r c r y s t a l l i n e cracking might influence

each of t h e s e f a c t o r s . The discussion was highly speculative owing t o t h e

l i m i t e d d a t a on which t o base s u i t a b l e judgement and make reasonable

assumptions. Further p u r s u i t of t h i s approach could be rewarding. Again,

a d d i t i o n a l information w i l l be required before s u i t a b l e refinements of

t h i s type can be applied with confidence.

Finally, procedures f o r incorporating creep-rupture damage together

w i t h f a t i g u e damage as discussed by Taira ( r e f . 15) o r as o u t l i n e d i n t h e

Appendix could be extended f u r t h e r t o make them more q u a n t i t a t i v e . It

would be necessary t o e s t a b l i s h t h e significance of s t r e s s r e v e r s a l , t o

determine more accurately how t o t r e a t t h e continuously changing s t r e s s ,

and t o e s t a b l i s h how t o combine creep and f a t i g u e i n t o an a p p r o p r i a t e ,

cumulative damage law.

 12

U n t i l such time as more accurate and d e t a i l e d procedures become a v a i l -

able, t h e method presented i n t h i s paper o f f e r s an extremely simple way of

estimating elevated temperature low cycle f a t i g u e behavior of small labora-

t o r y samples from e a s i l y a t t a i n a b l e t e n s i l e t e s t and creep-rupture data.

A t very high temperatures, low cycling frequencies, o r when t h e creep-

rupture curve of s t r e s s versus rupture time i s r e l a t i v e l y s t e e p (log-log

s l o p e steeper than -0.12), an a d d i t i o n a l c a l c u l a t i o n should be made t o

obtain a closer estimate of t h e lower bound of f a t i g u e l i f e . This calcu-

l a t i o n represents a highly s i m p l i f i e d computation of t h e damage produced

by creep-rupture e f f e c t s .

Although we have demonstrated t h e degree of confidence t h a t might be

expected i n our estimations of average l i v e s and upper and lower bounds,

care must be exercised i n applying t h e r u l e s given when an a p p l i c a t i o n

involves extreme t e s t conditions well beyond t h e range included i n t h i s

study. For example, t h e proposed method cannot be expected t o apply when

continuous s t r a i n cycling i s i n t e r r u p t e d by excessively long hold t i m e s

spent a t a f i x e d s t r a i n where creep-relaxation can occur. However, by

extending t h e general approach given i n t h e Appendix, it may be p o s s i b l e

t o a s s e s s t h e creep damage induced under such circumstances. In t h e f i n a l

analysis, t h e most appropriate f a t i g u e information i s t h a t generated under

conditions t h a t simulate as c l o s e l y as p o s s i b l e those that w i l l be en-

countered i n a n a c t u a l a p p l i c a t i o n . The i n t e r p r e t a t i o n of even conventional

laboratory f a t i g u e d a t a i n terms of s e r v i c e performance provides many

difficulties. Hence, even more so cam we expect estimates based on t e n s i l e

 13

and creep-rupture p r o p e r t i e s t o have t h e i r l i m i t a t i o n s . If t h e s e l i m i t a -

t i o n s a r e borne i n mind, however, t h e estimates can be very u s e f u l .

CONCLUSIONS

The method of u n i v e r s a l slopes, developed f o r estimating t h e low

cycle f a t i g u e behavior of materials a t room temperature from s t a t i c

t e n s i l e p r o p e r t i e s alone, has been extended f o r a p p l i c a t i o n t o materials

i n t h e creep range, I n most instance6, t h e estimates of l i f e are based

on c e r t a i n percentages of t h e l i f e as computed from t h e method of universal

slopes. As proposed, t h e method permits t h e estimation of a reasonable

lower bound, average value, and upper bound on f a t i g u e l i f e of s t r a i n -

c o n t r o l l e d laboratory s i z e samples. The method has been checked by

comparing t h e estimates with elevated temperature s t r a i n - c o n t r o l l e d Labora-

t o r y f a t i g u e data on numerous materials obtained from various sources.

Without excluding data s c a t t e r arising from t h e diverse t e s t i n g techniques

of d i f f e r e n t l a b o r a t o r i e s , t h e r e s u l t s obtained s t i l l i n d i c a t e t h a t t h e

estimates a t high temperature can be made w i t h about t h e same degree of

confidence as t h a t obtainable a t room temperature with t h e method of

u n i v e r s a l slopes.
 14

APPENDIX - CREEP-RUPTURE EFFECT

Some of t h e damaging e f f e c t of exposure t o high s t r e s s a t elevated

temperature i s inherently accounted f o r by using 1 6 Nf as a lower

bound, Cases a r i s e , however, when t h e creep-rupture curve i s q u i t e

s t e e p (log-log slope steeper than -0.12), t h e temperature i s very high,

and t h e frequency i s very l o w . Under t h e s e circumstances more creep

damage i s done than t h i s r u l e can account f o r . I n such cases it i s

necessary t o examine t h e damaging e f f e c t more closely. Although a r i g -

orous treatment i s beyond t h e scope of t h i s paper, an elementary ap-

proach i s suggested on an i n t e r i m b a s i s u n t i l b e t t e r assumptions can be

established by c r i t i c a l experiments.

An important f a c t o r t o be considered i s t h e magnitude of t h e s t r e s s

l e v e l t h a t develops during s t r a i n cycling. Although estimates must s u f f i c e

i n t h e present a n a l y s i s , d i r e c t measurement of t h i s s t r e s s would, of

course, provide a b e t t e r i n d i c a t i o n of t h e damage done. As a f i r s t ap-

proach, we r e s o r t t o t h e method of u n i v e r s a l slopes f o r an estimate of

t h e stress amplitude. I f t h e l i f e were Nf ( i n t h e absence of a creep-

rupture e f f e c t ) , t h e s t r e s s amplitude, oa, can be estimated ( r e f . 4,

Chapter 4), by

AEeE 3.5 uu -0.12

‘a=-- N;0*12 = 1 . 7 5 uuNf
2 - 2

It has been our experience t h a t t h i s estimate of t h e s t r e s s amplitude

m y possibly be a source of considerable e r r o r (errors of t h e order of

+ 2 6 a r e n o t uncommon). Further complications a r i s e i n t h e considera-

t i o n of how t o apportion t h e damaging e f f e c t s of t e n s i l e and compressive

 15

stresses during s t r a i n cycling. "he influence of compression has

generally been ignored and damage i s considered t o be caused only by

tensile stresses. Whether t h i s concept i s r e a l i s t i c i s unclear, s i n c e

it i s probable t h a t compressive s t r e s s e s may be detrimental (by causing

g r a i n boundary s l i d i n g ) or even b e n e f i c i a l (as a r e s u l t of t h e "healing"

e f f e c t of reversed s t r a i n ) . Definitive experiments a r e necessary t o

c l a r i f y t h e s e opposing e f f e c t s . Another uncertainty a r i s e s due t o t h e

unknown shape of t h e stress-time wave form. Ideally, both t h e shape of

t h e wave form and t h e slope of t h e creep-rupture curve must be known t o

determine t h e e f f e c t i v e f r a c t i o n of each cycle, k, f o r which t h e m a t e r i a l

may be considered t o be subjected t o t h e maximum s t r e s s .

For t h e elementary treatment given here, p a r t i a l allowance i s made

f o r t h e various u n c e r t a i n t i e s described by choosing an appropriate

value of k t h a t r e s u l t s i n f a t i g u e l i f e estimates having q u a n t i t a t i v e

agreement with t h e a v a i l a b l e data. Although t h e b e s t r e s u l t s could be

obtained by s e l e c t i n g d i f f e r e n t values of k f o r each s e t of data, a

s i n g l e r e p r e s e n t a t i v e value of 0.3 was found t o be reasonable.

Equation ( 2 ) can now be derived i n t h e following manner. If Ni

i s t h e a c t u a l llumber of cycles t o f a i l u r e , and t h e frequency i s F

cycles per u n i t time, t h e time t o f a i l u r e i s Ni/F, and t h e e f f e c t i v e

time foi- i.&Lc2L +L- uLAL m nvimJim

,,A-JLA...-.A s t r e s s i n t h e cvcle a c t s i s t' =
IcN;.
- . It
F
i s assumed t h a t t h e creep-rupture damage i s t h e r a t i o of t h e e f f e c t i v e

t i m e a t maximum s t r e s s t o t h e creep-rupture time a t t h i s s t r e s s . "he

next s t e p , therefore, i s t o determine t h e creep-rupture time, t,, under

a s t e a d i l y applied t e n s i l e s t r e s s , or, equal t o t h e f a t i g u e s t r e s s

 16

amplitude. The creep-rupture curve i s represented by a power l a w

relation ur = 1.75 uu , so t h a t t h e s t r e s s - r u p t u r e curve on log-log

coordinates becomes a s t r a i g h t l i n e of slope m and time i n t e r c e p t A

a t an extrapolated value of ur = 1.75 uu ( s e e i n s e t , f i g . 4 ) . Then,

at a stress ur -
- ua = 1 . 7 5 auNi t h e creep-rupture time, t,,

--0.12
becomes A( Nf) . Hence, t h e creep-rupture damage i s given by

The damage due t o pure f a t i g u e i s taken t o be t h e r a t i o of t h e

a c t u a l number of cycles t o f a i l u r e t o t h e cycles t h a t would have been

sustained i n t h e absence of creep damage, t h a t i s Ni/Nf. Assuming a

l i n e a r cumulative damage l a w wherein creep damage i s given by time

r a t i o s , and f a t i g u e damage i s given by cycle r a t i o s , then a t f a i l u r e

- ' + - =t '
N; N1.
-+ 4
Nf tr Nf --0.12
m
ANNf)

or

t Nf
Nf =
m+O .12
1+ k (Nf) m
AF
This approach i s presumed t o have u t i l i t y only when a l a r g e creep-

rupture e f f e c t i s a n t i c i p a t e d and t h e 1 6 Nf rule i s inadequate as a

lower bound; t h a t is, when N;' < l@ Nf. I n t h i s region of i n t e r e s t ,

f a i l u r e i s primarily time dependent and t h e f a t i g u e damage i s nearly

 17

negligible. Since we have r e s t r i c t e d ourselves t o observed l i v e s of l e s s

than lo5 cycles, t h e creep-rupture e f f e c t need be considered only when

N i < 1% Nf C lo5. Hence, i f equation ( 2 ) i n d i c a t e s t h a t Ni > loqb Nf

f o r values of N i > lo5, f u r t h e r use of equation ( 2 ) i s not c a l l e d f o r .

This c r i t e r i o n has been used t o construct t h e curve shown i n f i g u r e 1

for k = 0.3.

An i l l u s t r a t i o n of t h e estimation procedure f o r incorporating t h e

creep-rupture e f f e c t i s shown i n figure 4 f o r Nimonic 9 0 a t 1600' F

(1143' K ) and F = 0.1 cpm (0.0017 Hz) ( r e f . 11). The i n s e t creep-

rupture curve shows t h a t m = -0.24 and A = 2.5 min. (150s); hence, f o r

F = 0.1 cpm (0.0017 Hz) and k = 0.3, t h e point P i s determined as

shown i n f i g u r e 1. Since t h i s point l i e s above t h e curve, t h e computa-

f
t i o n of Nf from equation ( 2 ) i s made. The estimate of t h e lower bound

i s shown i n f i g u r e 4 as CDH. I n t h e region CD, 1@ Nf (where Nf is

determined from curve AB - t h e method of universal slopes, equation (1))

t
gives a lower l i f e than Nf and hence, CD i s used i n s t e a d of GD.

A t high lives, DH provides a lower 'Life than 1% Nf, thus DH i s used

i n s t e a d of DE. Although not shown i n f i g u r e 4, t h e average l i f e estima-

t i o n and t h e upper bound would be determined by multiplying t h e lower

bound l i v e s by f a c t o r s of 2 and 10 respectively. These curves would then

be t h e same as those already shown i n f i g u r e 2 ( d ) f o r t h i s material and

t e s t condition.
 18

REFERENCES

1. L. F. Coffin, Jr., "Proceedings of t h e F i r s t I n t e r n a t i o n a l Conference

on Fracture" (Sendai, 1969, 3, Paper No. D I - 1 1 9 . 1966: Japan

(Japanese Society f o r Strength and Fracture of Materials) .

2. S. S. Manson, I n t . J. Fracture Mechanics, 1966, 2, (l), 327.

-
3. S. S. Manson, Ekp. Mechs., 1965, 5, ( 7 ) , 193.

4. S. S. Manson, "Thermal S t r e s s and Low-Cycle Fatigue". -

1966: New York

(McGraw-Hill) .
5. V. H. Thevenow, Allison Division of General Motors, Materials

Laboratory Report No. MSM 6-13, 1966.

6 , A, Coles and D. Skinner, Roy. Aeronautical SOC. J., 1965, 69, (649), 53.

7. W. F. Anderson and C. R. Waldron, Atomics I n t e r n a t i o n a l , North American

Aviation Report No. NAA-SR-4051, 1959.

8. Anon., General E l e c t r i c Progress Report No. 58, P a r t A, GEMP-58A, 1966.

9. T. Slot, General E l e c t r i c Report No. GE-TM66-6-11, 1966.

10. E. E. Baldwin, G. J. Sokol, and L. F. Coffin, Jr., h o c . h e r . SOC. Test.

Mat., 1957, 57-, 567.

11. P. G. F o r r e s t and K. B. Armstrong, " J o i n t I n t . Conf. on Creep" (New York-

London, 1963), Paper No. 1. 1963: - London ( I n s t i t u t i o n of Mechanical

Engineers). See a l s o : P. G. F o r r e s t and K. B. Armstrong, J. I n s t .

Metals, 1966, 94, ( 6 ) , 204.

1 2 . C. H. Wells and C. P. Sullivan, Trans. h e r . SOC. Metals, 1965, 58, (31,

391.
 19

13. R. W. Swindeman, "Joint I n t . Conf. on Creep" (New York-London, 1963),

-
Paper No. 32. 1963: London ( I n s t i t u t i o n of Mechanical Engineers).

14, Anon., Data courtesy of P r a t t and Whitney A i r c r a f t , Division of

United A i r c r a f t Corporation, East Hartford, Connecticut.

15. S. Taira, "Creep i n Structures", p. 96, 1962: - New York (Academic

Press).
 TABLE I. - ALLOYS, TEST CONBITTIONS, AM) m T I " PROPERTIFS

7 G - ptemperature
designation
Z- Test frequency, F RA,
percent
Tensile s t r e n g t h , uu m a s t i c modulus, E I Creep-rupture

I ksi 100 N/mZ 1 0 3 bi Id-1N/m2 Slope I n t e r c e p t , fi

I OF
9(

-
26.8
-m
<0.12
ain

011 31. 182. 12.5 1.L5

923
-
.083 - 0.83 30. 167. ll.5 23.1 1.59 <. 1 2
923 18. 115. 7.93 23.0 1.59 CO.12

A-206 1033 57.6 75. 5.2 10.0 1.31 0.20

Inco 901 1 1400

I
1033 16.1 1M. 7.17 20.0 1.43 0.135 0.17

956 42. 179. 12.3 22.0 1.52 <0.12 ------ *

030 1, 1 0 0.017, 0.17 70. 69.5 4.79 24.** 1.65 <0.12
I -
L-605
I 1000
1200
811
923 .083 - 0.83
48.2
47.1
106.
95.4
7.31
6.58
27.
25.
1.9
1.7
<0.12
C.12

304 S t a i n l e s s
steel I 1000
1200
1500
811
923
1009
.05
.05
-
-
0.30
0.45
-~
67.3
50.w
42. **
55.5
41.H
20.**
3.03
2.0
1.4
23.**
22.**
19.**
1.6
1.5
1.3
<0.12
.18
.18

347 S t a i n l e s s 1110 073 67.7 55. ** 3.8 22 .** 1.5 <o. 12

steel

Nimonic 75 1200 923 10 0.17 29.5 81.5 5.62

1300 1023 10 .17 42.0 53.1 3.66
1600 ll43 10 .17 62.5 28.0 1.94 12.2**
1000 1253 10 .17 66.0 16.2 1.12 9.7** ------
Nimonic 9 0 1200 923 10 0.17 24.5 144. 9.93 26.5** 1.03 CO.12
1300 1023 0.1, 10 .0017, 0.17 12.0 119. 8.20 25.** 1.7 .135
1600 1143 0.1, 10 ,0017, 0.17 14.0 76. 5.2 23.** 1.6
1800 1253 0.1, 10 .0017, 0.17 94.0 16. 1.1 el.** 1.45 .25

Nimonic 1 C 6 1200 923 10 0.17 13.7 142. 9.79 25.""

1380 1023 10 .17 18.0 151. 10.4
1600 1143 10 .17 32.5 99.5 6.06 22.**
68.0 39.2 2.70 1 0 .**
1000
-
1253 10 .17

Udimet 700 1400 1033 1-2 I 0.017 - 0.033 31. 155. 10.7 <o .12
Astroloy 1400 1033 20 0.30 29.6 141. 9.72 1.6 (5)

Inconel 1500 1009 3.017, 0.1 0.00028, 0.0017 63.3 25. 1.7 1.2 0.15 1.0 60.

I N 100 1000 011 6 0.10 9.3 121. 8.34 27.1

(PWA 47 Coat) 1700 12oc 6 .10 11. 96. 6.6 22.7
2 000 1367 6 .10 10.3 26.8 1.05 19.3
-
MAR M zoo 1300 97e 6 0.10 5. 120. 8.27
(€VA 47 Coat) 1700 12oc 6 .10 4. 77. 5.3
-
B 1900 1300 978 6 0.10 5. 145. 10.0 <0.12
(PWA 47 Coat) 1700 l2Oc 6 .10 3. 95. 6.5 23.0 1.59
2000 1361 6 .10 9. 35. 2.4 20.5 1.41 .26

*
NASA data - E. R. Roberts.
**
Approximate handbook v a l u e s .
 F

:i
.36
Note: If point i s above curve, N' should be
calculated from Equation 12). If pint
i s below curve, or belpv the lowest
coordinates s h w n , Nf need not be
calculated.
g

.24 -
8 P (Nimonic 90 1600" F (1143"K) -
n 0.1 cpm (0.0017 Hz))
%
-
V
L

/
0

ri - .18
rl v1
r-
M Equation (2) for k = 0.3 and N; = IO5 = 10 percent Nf
I
w I I I I I I I

10-1

10-2
A-286 CR 34 percent

10-1

10-2

10-1

10-2

10-3
101 Id 101 Id
Cycles to failure
(a) Iron-nickel-chromium-molybdenum alloys.
Open symbols - Unidentified cracking
Shaded symbols - Transcrystalline cracking
X in symbol - Intercrystalline cracking
f 7
Upper bound - - -
Average life -- -
Lower bound - - -

Figure 2. -Comparison of estimated and observed fatigue behavior.

 10-1
+ s iI 0 10 corn (0.17Hz)
1 cpm (0.017Hz)
A 10 m i n (600s) dwell
10-2 30 m i n (1800s) dwell
ir
-=I B 100 m i n (6000s) dwell
d
m Cr-Mo-V steel D 180 m i n (10800 5) dwell
c
1050' F (838"K)
.- 10-1
m
L
c
Y)

-
(D
c
I-0

10-2

lo-:
101 Id 101 103 105
Cycles to failure
(b) Cr-Mo-V steel and cobalt-base alloy.
Figure 2. -Continued.

10-1 \ I 10 cpm io. 17 Hz) \ 3-18cpm (0.05k.30 Hz) I

Q
a
3
CI,
%rLW-4
10-2 -Anderson and

E
.-c 10-1
E
c
v)

IU -101 102
~ ~~

101 Id Id
Cycles to f a i l u r e
(c) Stainless steels.
Figure 2. - Continued.
 10-1

10-2
4-
pi
c
m
e
.-= 10-1
c
E
VI
-
3 10-2

10-3
101 Id 101 Id
Cycles to failure
(d) Nickel-base alloys - Nimonic series, Forrest and Armstrong (11).
Figure 2. - Continued.

10-1

10-2

10-2
d
c
0
E I I I I Plane bendinq I 1
.-c -
-
E
VI
IU

10-2

10-1

10-2

10-3
101 Id 101
Cycles to failure
Id Id
(d) Continued. Nickel-base alloys - Nimonic series, Forrest and Armstrong
(11).
Figure 2. - Continued.
 Cycles to failure
(d) Concluded. Nickel base alloys - Nimonic series, Forrest and Armstrong
(11).
Figure 2. - Continued.

Cycles to failure
(e) Nickel-base alloys.
Figure 2. - Continued.
10-1

10-2

6 cpm (0.10 Hz)

10-1
\

10-1

10-2

10-3
101 101
Cycles to failure
(f) Coated (PWA 47) nickel-base alloys, Pratt & Whitney (14).
Figure 2. - Concluded.
 2

105

lo4
5

-01 2
-
K
103
-d
.-
c

F 5
e
*
0)
n
0 2

102

101

Estimated average life, cycles

Figure 3. -Comparison of observed fatigue life with high temperature method of estimation.

10-1

10-2

10-3
lo0 101 102 Id 104
Cycles to failure
Figure 4. - Creep-rupture effect calculation used to establish lower bound
for Nimonlc 90 at 1600" F 11143" K) and 0.1 cpm 10.0017 Hz).