NA SA TECHNI CA L NOTE

m
h
04
N
I
n
z
c
4
m
4
z
NASA TN D-2975
4 . /
-
-4
- 0
"
OPTIMIZATION OF TIME-TEMPERATURE
PARAMETERS FOR CREEP AND
STRESS RUPTURE, WITH APPLICATION
TO DATA FROM GERMAN COOPERATIVE
LONG-TIME CREEP PROGRAM
by A Zexunder M endelson, Ernest Roberts, Jr.?
QndS. S. Munson
Lewis Reseurch Center
CZeueZund, Ohio
NATI ONAL AERONAUTI CS A ND SPACE A DMI NI STRA TI ON WASHI NGTON, D. C. 0 AUGUST 1965
k
OPTIMIZATION OF TIME-TEMPERATURE PARAMETERS FOR CREEP
AND STRESS RUPTURE, WITH APPLICATION TO DATA FROM
GERMAN COOPERATIVE LONG-TIME CREEP PROGRAM
By Alexander Mendelson, Ernest Roberts, Jr.,
and S. S. Manson
Lewis Research Center
Cleveland, Ohio
NATIONAL AERONAUTICS AND SPACE ADMINISTRATION
For sal e by t he Cl ear i nghouse for Feder al Sci ent i f i c and Techni cal Inf or mat i on
Spr i ngf i el d, Vi r gi ni a 22151 - Pr i c e $2.00
OPTIMIZATION OF TIME-TEMPERATURE PARAJMETERS FOR CREEP AND STRESS RUPTURE, WITH
APPLICATION TO DATA FROM GEFWAPT COOPERATIVE LONGTIME CREEP PROGRAM
by Alexander Mendelson, Ernest Roberts, J r., and S. S. Manson
Lewis Research Center
SUMMARY
By the use of orthogonal polynomials developed f or di screte sets of data,
the l east-squares equations f or determining the optimized stress-rupture para-
metri c constants are obtained i n nearl y uncoupled form; thus the use of high-
degree polynominals i s permitted without the l oss of si gni fi cant fi gures.
Optimum values of the constants can thereby be accuratel y obtained. The method
i s appl i ed to the data obtained from the German cooperative long-time creep
program by using a general parameter of which the Manson-Haferd and Larson-
Mi l l er parameters are speci al cases. Good correl ati on was obtained. An analy-
si s i s al so made of creep data obtained f or columbium al l oy FS-85 with good
resul ts. A complete Fortran I V computer program i s included to ai d those wish-
i ng t o use the method.
I NTRODUCTI ON
One method of extrapol ati ng short-ti me creep-rupture data to predi ct long-
time l i f e involves the use of a time-temperature parameter. This concept i s
based on the assumption that all creep-rupture data f or a given materi al can be
correl ated to produce a si ngl e "master curve" wherein the stress (or l og stress)
i s pl otted agai nst a parameter involving a combination of time and temperature.
Extrapol ati on to l ong times can then be obtained from thi s master curve, which
can presumably be constructed by using only short-ti me data. Three well-known
parametric methods are the Larson-Miller, Manson-Haferd, and Dorn parameters
(refs. 1 to 3). These parametric methods have the great advantage, at l east i n
theory, of requi ri ng only a rel ati vel y s mal l amount of data to establ i sh the
requi red master curve.
More recentl y a general creep-rupture parameter was introduced by one of
the authors (ref. 4) that i ncl udes most of the currentl y used parameters as
speci al cases. The anal ysi s i n the present paper i s therefore based on this
general parameter.
A si gni f i cant advance i n t he pr act i cal appl i cat i on of t he par amet r i c
met hods was t he devel opment of an obj ect i ve l east - squar es met hod f or det er mi n-
i ng t he opt i mumval ues of t he par amet r i c const ant s wi t hout pl ot t i ng and cr oss-
pl ot t i ng t he dat a and wi t hout t he use of j udgment on t he par t of t he anal yst
(ref . 5). Thi s l east - squar es met hod i nvol ves, however , sever al pr act i cal di f -
f i cul t i es t hat ar i se f r omt he f act t hat i n f i t t i ng t he mast er cur ve by a pol y-
nomi al , t he set of l i near al gebr ai c equat i ons f or t he coef f i ci ent s ( t he nor mal
equat i ons) ar e ver y i l l - condi t i oned.
shown t o be r el at ed t o t he Kl ber t det er mi nant ( ref . 6), whi ch r api dl y ap-
pr oaches zer o as i t s or der i ncr eases. Thus f or pol ynomi al s above t he second
degr ee, i t i s necessar y t o use doubl e- pr eci si on ar i t hmet i c (16 si gni f i cant
di gi t s or mor e) on t he comput er , and f or t he f i f t h degr ee and above t he r esul t s
become uncer t ai n even wi t h doubl e- pr eci si on ar i t hmet i c. Thi s di f f i cul t y i s
i nher ent i n t he nor mal l east - squar es equat i ons and i s not l i mi t ed onl y t o t he
st r ess - r upt ur e probl em.
The det er mi nant of t hese equat i ons can be
The pr esent r epor t pr esent s a met hod f or avoi di ng t he above di f f i cul t y by
usi ng or t hogonal pol ynomi al s i n t he r epr esent at i on of t he mast er cur ve ( appen-
di x A). The use of or t hogonal pol ynomi al s f or r epr esent i ng di scr et e set s of
unequal l y spaced dat a i s descr i bed i n r ef er ence 6 and i n mor e det ai l i n r ef er -
ence 7. A f ur t her i mpr ovement can be obt ai ned by per f or mi ng a l i near t r ans-
f or mat i on on t he st r esses ( or t he l ogs of t he st r esses) so t hat al l t he val ues
of st r ess ( or l og st r ess) l i e bet ween 2 and - 2, as r ecommended i n r ef er ence 7.
As a r esul t of t hese i nnovat i ons, i t became possi bl e t o per f or mal l t he compu-
t at i ons i n si ngl e- pr eci si on ar i t hmet i c ( ei ght si gni f i cant di gi t s) up t o 18t h
degr ee pol ynomi al s wi t hout appr eci abl e r ound- of f error.
I n addi t i on, t hi s r epor t cont ai ns a compl et e anal ysi s, i n whi ch t he gen-
er al par amet er was used, of al l t he dat a f or t hr ee st eel s t hat wer e obt ai ned by
NASA t hr ough t he cooper at i on of Dr. K. Ri char d of Faber wer ke Hoechst i n
Fr ankf ur t and t hat wer e i nvest i gat ed i n a l ong- t i me cooper at i ve cr eep pr ogr am
i n Germany. Some of t he dat a f r omt he l at t er i nvest i gat i on ar e i ncl uded i n t hi s
paper.
Fi nal l y i t i s shown by means of a concr et e exampl e how t he par amet er t ech-
ni ques can be appl i ed t o cr eep dat a t o pr edi ct l ong- t i me creep. For t hi s pur -
pose t he dat a f or col umbi umal l oy FS- 85, as r epor t ed i n r ef er ence 8, ar e used.
A compl et e For t r an I V pr ogr am, as used on t he I BM7094 comput er i n maki ng
t he cal cul at i ons, i s pr esent ed i n appendi x B. Thi s pr ogr amcan be used f or t he
obj ect i ve anal ysi s of any set of cr eep- r upt ur e dat a by t he Lar son- Mi l l er ,
Manson- Haf er d, or t he mor e gener al par amet er of r ef er ence 4.
SYMBOLS
A, B l i near t r ansf or mat i on coef f i ci ent s
a, b, c el ement s of coef f i ci ent mat r i x
D st andar d devi at i on
2
K
m
n
p( 0 )
Q
(2
r
S
T
Ta
t
ta
U
X
X
Y
Y a
a, P
(5
z
degree of freedom
degree of polynomial
number of data poi nts
creep-rupture parameter
polynomial
stress exponent
temperature exponent
sum of squares of resi dual s
temperature
temperature i ntercept
time to rupture
time i ntercept
coeffi ci ent of polynomial functi on
scal ed l og stress
l og stress
l og time
l og time i ntercept
constants from recurrence rel ati on
stress
o ' ( T - Ta)r
Subscri pts :
max maxi mum
min minimum
PROCEDURE
GeneraJ . Parameter
The general creep-rupture parameter introduced i n reference 4 has the f ol -
3
lowing form
l og t
- - lnp. t
a
0
--
I
where Ta, l og ta, q, and r are materi al constants to be determined from the
avai l abl e experimental data. The parameter P(o) i s a functi on of the stress
and, when pl otted agai nst stress, i s referred t o as a master curve (fi g.
1,
p. 9). If q = 0 and r =1, the Manson-Haferd parameter i s obtained. I f
q = 0, r =-1, and
q =1 and r =1, the stress-modified parameter suggested i n reference 9 i s ob-
tai ned. Fi nal l y, i f q = 0, equation (1) reduces to the parameter proposed by
Manson and Brown (ref . 10).
Ta = -460' F, the Larson-Miller parameter resul ts. If
The obj ect i s t o fi nd the best values of the constants q,, l og ta, Ta,
and r so that the master curve best f i ts the data. To f i nd these values, the
method of l east squares i s used whereby the master curve i s represented by a
polynomial i n the logarithm of the stress, and the best f i t i s obtained by mini-
mizing the sumof the squares of the devi ati ons (the resi dual s) of the data
from the curve. The cal cul ati on procedure w i l l now be described. The detai l s
of the deri vati on are given i n appendix A, and a Fortran I V computer program
using thi s method i s given i n appendix B.
Cal cul ati on Procedure
To si mpl i fy the notati on, the following symbols are introduced:
( 3 )
I
T E o ~( T - Ta)r
y =l og t
x =l og 0
ya E l og ta
Then from equation (1) it follows that
Y = o%a +TQ(X)
where i n reference 5, Q(x) was represented by a simple polynomial of the form
Q(X) =
+alx +a2x2 +. . . +a,mxm
(4)
The l east-squares equations obtained sometimes l ed to di f f i cul ti es as i ndi cated
i n the INTRODUCTION. These di f f i cul ti es can be avoided, however, by rewri ti ng
equation (4) i n terms of polynomials that are orthogonal over the set of data,
as defined i n appendix A. Thus assume
4
where uj i s an unknown constant, m i s the degree of the hi ghest degree
polynomial, and Q. (x) i s a polynomial of degree j - 1 that sati sf i es the
orthogonal i ty condi ti ons descri bed i n appendix A. The use of orthogonal poly-
nomials permits the sol uti on of the l east-squares equations di rectl y i n closed
form, thus the l oss of a l arge number of si gni fi cant di gi ts i s avoided.
method of cal cul ati ng Qj w i l l be discussed i n appendix A.
m +5
unknown constants resul ts f or the case of the general parameter. For the case
of the l i near parameter there are m+3 constants, and f or the Larson-Miller
parameter there are m +2. It i s necessary that the number of data poi nts n
always equals or exceeds the nurhber of unknown constants.
J
The
If equation (5) i s substi tuted i nto equation (3), an equation with
The constants are determined so that equation (3) fi ts the data best i n
the l east-squares sense. To accomplish thi s, the sumof the squares of the
devi ati ons i s minimized; that i s,
n
i s made a minimum. Because the equations are nonl i near i n some of the unknown
constants a tri al and error procedure must be used. A set of values i s assumed
f or q, r, and Tat and the corresponding best values of ya and uj are de-
termined. A di fferent set of values f or q, r, and Ta i s then chosen, and
agai n the best values of ya and uj are cal cul ated. Several sets of values
of q, r, and Ta are tri ed, and the values corresponding to the overal l best
f i t are determined. For the case of the l i near parameter, only the value of
Ta i s vari ed (q i s always equal t o zero, and r i s always equal to 1). For
the Larson-Miller parameter, Ta i s equal to -460' F, and no tri al and error
procedure i s needed.
As a measure of the f i t, the standard devi ati on D, defi ned by
i s used, where K equals
m + 5 general parameter
m + 3 l i near parameter
m+2 Larson-Miller parameter
The smal l est val ue of D w i l l correspond t o the best fi t.
5
1 . . . . .. .. .. ._ . . . . ._ - . .. .. . . .. _. - - . . - . . . . - .. .- . . . . ._. . . ..._... .... .-....----.-
To determine the best values of ya and U j f or a given set of values of
T,, q, and r, the following cal cul ati ons are made. Fi rst, the logarithms of
the stresses are scal ed so that they l i e i n the range -2 to 2, as suggested i n
reference 7. The reason f or thi s i s discussed i n appendix A. Thus defi ne a
vari abl e X by
X =A x +B ( 94
4
%ax - %in
A =
%ax +xmin
%ax - %in
B = -2
The polynomials Q j ( q ) are now cal cul ated f or each of the data poi nts by
usi ng the following formulas:
-
Pj -
n
i=l
j n
a =
i=l
n
i=l
m
i=l
Q1
where n i s the number of data poi nts, Y i s the scal ed value of l og f or the
ith data poi nt, and ~i i s equal to &(Ti - Ta)r f or the ith data poi nt f or
the chosen values of Ta, q, and r.
It i s to be noted that the degree of the polynomial Q(x) of equation (5)
can be i ncreased by merely computing the next polynomial i n the seri es
without having to recompute any of the previous ones. This i s one of the ad-
vantages of using orthogonal polynomials.
%+2
Once the values of Qj have been computed f or each of the data poi nts,
ya and uj can be cal cul ated as follows:
6
I
Let
n
i=l
n
co = c a;yi
i=l
where j =1, 2 . . . m +1.
Then
j=1
J
Note that i f q =0, a. equals the number of data poi nts n. Thus by means of
equations (9) t o (12), the best values of Y a and U j t o f i t the data are
found f or a given choice of Ta, q, and r. The Fortran I V program described
i n appendix B automati cal l y scans all the desi red values of Ta, q, and r and
chooses the best set from all the submitted values as determined by the smal -
l est value of the standard devi ati on D, as defined by equation (7). The
method can be i l l ustrated by a simple example: consider a set of theoreti cal
data, which f i t the following equation exactl y
7
9'5 - log =
7.02 +0.467 x +0. 061 x2 +0.00928 x3)
T - 600
Eight data poi nts sati sf yi ng thi s equation are given i n columns 2 to 6 of
tabl e I. For this data Ta = 600' F and l og ta =Y a =9.5. Suppose, however,
that these ei ght data poi nts were obtained experimentally and that the val ues
of Ta and l og ta were not known. The problem then i s to f i nd the best
val ues of Ta and l og ta to f i t the data by the l i near parameter. These
values can readi l y be found by using the equati ons of the previous secti on.
Fi rst, from column 6 of tabl e I
(l og O)max = 4.75051
(l og a)min =1.81954
Theref ore from equati ons (9b)
A =1.36474
B = -4.48319
and by means of equati on (sa) the were computed and are given i n column 8.
For i l l ustrati ve purposes three val ues of Ta were chosen, 500, 600,
and 700 F.
Qj ( X i ) were computed by means of equations (Z), (l o), and ( l Oa3, and the val ues
of aj, bj , and cj were computed by equations (11). The resul ts are tabu-
l ated f or
thi rd degree polynomial.
For each of these values of Ta, val ues of
Ti , CL., pj , and
Ta = 600 i n columns 9 t o 12 of tabl e I and i n tabl e I 1 up to a
The values of Y a and uj were then computed by using equations (12) f or
each of these three val ues of Ta by fi rst assuming m= 2, then m =3, and
f i nal l y m =4, corresponding t o polynomials of second, thi rd, and fourth de-
grees, respecti vel y. For each of these cases the standard devi ati on D was
computed from equation (7) with S being given by equation (6) and Q by
equation (5). The resul ts are summarized i n tabl e 111. The l east value of D,
si gni fyi ng the best fit, i s obtained f or m =3 and Ta = 600 F. The cor-
responding value of . ya i s 9.5. These val ues, of course, correspond t o equa-
ti on (13), from which the data were generated.
Application t o Dat a from German Cooperative Long-Time Creep Program
As part of the German cooperative long-time creep program, a suf f i ci ent
amount of material of each of three steel s was suppl i ed to NASA to permit the
running of short-ti me tests necessary to predi ct the resul ts at long t i mes ob-
tai ned i n the German test program.
tabl e I V.
The composition of these steel s i s shown i n
The resul ts of the NASA tests, which wer e used i n the subsequent anal ysi s,
8
f
h
Parameter, P =(T - xw))/(log t - 16.54)
Figure 1. - Master curve for steel K (27b KK), calculated
from NASA data between 10 and 3700 hours.
are shown i n tabl e V. Table V I shows the
results of the long-time German test pro-
gram. The three steels w i l l be designated
bri ef l y as steel K, steel Cy and steel P.
With the use of the test data shown
i n tabl e V a complete anal ysi s was made
by the previously descri bed method. The
general parameter di scussed i n the
INTRODUCTION was used, and the best values
were obtained f or the parametric constants
f or each of the three steel s.
Al l the data obtained f or these
steels are shown i n tabl es V and VI .
Many of the data poi nts were obtained f or
purposes other than the appl i cati on t o
time-temperature parameters, as described
i n thi s report. As al ready discussed i n
references 4 and 11, a much smal l er
amount of data i s needed when an accel -
erated program i s desi red; however, si nce
these data were al ready avai l abl e, all the
data i ndi cated i n tabl es V and V I were
used to obtai n the best possi bl e para-
metri c constants.
For al l three steels the anal ysi s showed the stress exponent q t o be
zero, but the temperature exponent r to be di fferent f or each of the three
materials. For steel K the best value of r was 1, which i ndi cated that the
best f i t i s obtained by the l i near parameter. For steel P a value of r of
-1 was obtained, which i ndi cated a parameter s i mi l ar t o the Larson-Miller param-
eter; however, the corresponding value of Ta was 200' F rather than -460 F
used i n the Larson-Miller parameter. For steel C the value of R was 2.5.
Figure 1 shows the results f or steel K. Here the master curve consi sts of
a pl ot of stress agai nst the optimized parameter (T - 300)/(l og t - 16.54).
Figure 2 shows the isothermals computed by using the optimized parameters,
as shown on each of tne fi gures. The range of the NASA data used to obtai n
these parameters i s al so shown on each of the fi gures. The data poi nts shown
are the German results obtained t o date. Tne predi cti ons up t o 100 000 hours
from the NASA data based on the optimized parameters agree wel l with the
German data, if scatter and di fferences i n testi ng technique between the t wo
organi zati ons are considered.
Fi gure 3 shows a comparison f or each of the three steels between the best
l i near parameter, the best Larson-Miller parameter, and the best general param-
eter. Although f or some of the steels fai r agreement can be obtained with one
or the other of these parameters, it i s cl ear that the general parameter i s
superi or when dl the materials are considered j oi ntl y. If any one of the
speci al cases of this parameter i s t o be chosen f or al l materials, the l i near
9
\
\
Predicted -
German data above 5 hr
100 102 104 11
0
(a) Steel K (27b KK); parameter,
P =(log t - 16.54)/(T - 300).
100 102 1
\
100 102 1
i T(No fail,
?6 4F--
failure
106
~.
Time, hr
(b) Steel C (23b CK); param ter, (c) Steel P (14a PA); parameter,
P =(log t - 8.87)/(T + P =(T - 2OO)(log t +11.13).
Figure 2. - Analysis of German steel data by generalized parameter wi th opti muT constants (where T is temperature, and t is ti me to rupture).
LV
(a) Steel K (27b KK).
40
20
0
100 102 104 106
Time, hr
(b) Steel C (23b CK).
Temperature for
German data
above 5 hr,
"F
0 932
0 1022
0 1112
A 1202
0 1292
(c) Steel P (14a PA).
Figure 3. - Analysis of German steel data by several parameters (where T i s temperature, and t i s time to rupture).
10
I
I
10
- 1
l l
Strain,
percent
.-
v) CL
0-
10
m-
v)
E
c
m
lfl
0
-
A
1
2
5
\
h
J ~~ _ _
2200 3000 100 150 200 250 300 350 400
Temperature, T, OF
T - 1400
Parameter, P =- log - ~ t - 5.76
(a) 5-Percent strain. (b) Master curves obtained for I-, 2-, and 5-percent strain.
Figure 4. -Analysis of creep data for columbium alloy FS-85 by linear parameter.
parameter would appear to be the best choice.
Application to Creep Data
Although there i s no fundamental reason why the same parameter i s capable
of representi ng both creep and rupture data, it has neverthel ess been found
empi ri cal l y (refs. 1 and 2) that the dual rol e of the same parameter l eads to
reasonable resul ts. Experimental data f or creep are much more l i mi ted, however,
than that f or rupture, and such data tend to contai n more scatter; hence,
anal ysi s of creep data by the parametric approach has been l i mi ted i n the past.
The method of the present report can be appl i ed di rectl y to creep data
without any change. Al l that i s necessary i s to redefi ne t as the time to
attai n a speci fi ed amount of creep rather than as the rupture time.
i s assumed that f or a given amount of creep, say 1 percent, a pl ot of l og CT
agai nst a parameter, such as that given by equation (1) , w i l l produce a si ngl e
master curve. For a di f f erent amount of creep, say 5 percent, a di f f erent
master curve can be obtained, but it i s assumed that the parametric constants,
such as l og ta
tai ned from rupture data.
Thus, it
and Ta, remain the same and that they equal the values ob-
Cal cul ati ons of thi s type were performed f or columbium al l oy FS-85.
The
creep tests were l i mi ted to runs of approximately 1000 hours; the data are
11
given i n tabl e ' m1, as taken from reference 8. Figure 4(a) shows the data f or
5-percent creep strai n, and f i gure 4(b) shows the master curves obtaFned f or
1-, 2-> and 5-percent strai n as wel l as the parametric constants obtained by
the method of thi s report. While scatter i n the creep data i s high, the cor-
rel ati on must be regarded as good. I n general , the poi nts agree well with the
master curve.
Although these resul ts are encouraging, much more work i s necessary before
it can be concluded that the parametric approach i s completely val i d f or creep
data. If it i s eventual l y concluded that the parametri c approach i s val i d f or
creep data and i n parti cul ar that the parametric constants are the same f or
both the creep and rupture processes, it i s obvious that a great saving i n test
f aci l i ti es and test program planning w i l l resul t. It therefore seems very
worthwhile i n f uture studi es to gi ve more attenti on to the correl ati on and
extrapol ati on of creep data by the parametric method.
Lewis Research Center,
National Aeronautics and Space Administration,
Cleveland, Ohio, May 3, 1935.
12
APPENDIX A
ORTHOGONAL POLYNOMIALS AND LEAST-SQUARES DEFERNDI "I 0N
OF PAl3AMETFiIC CONSTANTS
A set of polynomials Qj(x) are sai d to be orthogonal over an i nterval
wi th respect to the weighting functi on
rel ati on
z(X) i f they sati sf y the following
Si mi l arl y a set of polynomials can be defi ned to be orthogonal over a set of n
di screte poi nts xi by the fol l owi ng rel ati on
It can be shown (ref. 6) , that all orthogonal polynomials sati sf y a three-term
recurrence rel ati on of the form
Thus by starti ng with Ql = 1 and p1 =0 an i nf i ni te set of orthogonal poly-
nomials can be generated by means of equation (A3) i f val ues f or % and Pk
are known.
(Al) or (AZ)). Fr omthe rel ati on (A2) i t f ol l ows that
These can be determined f r omthe orthogonal i ty condi ti ons (eqs.
i=l
and
When the recurrence rel ati on ( A3) i s used to el i mi nate Qk+l, there i s obtai ned
n
i=l
13
When the orthogonal i ty condi ti on (A2) i s used, equations (&a) and (A5b) reduce
t o
i=l
Solving equations (A6) f or and pk gi ves
i=l
n
ak =
i=l
n
XiT?QkQk-l
id
' k = n
i=l
Thus a set of orthogonal polynomials can be generated that are orthogonal over
a f i ni te set of di screte values of the variable x. Note that these val ues
need not be equal l y spaced, a condi ti on that i s obviously necessary f or stress-
rupture data.
Scal i ng of Polynomial Argument
From the recurrence rel ati on ( A3) wi th
Q1 = 1, it follows that the l ead-
i ng term of Qk+l(xi) i s xi. k Therefore, depending on the values of xi, the
values of This procedure can
l ead t o a l oss of si gni fi cant fi gures i n performing the cal cul ati ons. I t i s
shown i n reference 7, by comparison wi th the Chebyshov polynomials, that if
x i s scal ed so that al l the values of l i e between 2 and -2, the polynomial
Qk+l(xi) can become very l arge or very smal l .
14
val ues
scal i ng, $et xmax be the maxi mum value of l og a
value of l og a; then l et
Q . ( q ) w i l l al l be of approximately uniform si ze. To perform thi s
and %in be the minimum
-2 =kn +B
and sol vi ng f or A and B resul ts i n equations (9b).
It has been found i n practi ce that scal i ng the values of x as i ndi cated
does indeed preserve the si gni fi cance of the cal cul ati ons.
Least-Squares Procedure
I n terms of the orthogonal polynomials, equation (3) can be wri tten
m+l
Y = "qYa +7
ujQj(X)
j=1
To f i nd the best values of ya and uj that f i t the data, the sumof the
squares of the resi dual s i s minimized. Thus l et
Then i n order t o fi nd the val ues of ya and uj that wi l l make S a minimum,
S i s di fferenti ated i n turn wi th respect to ya and each u and the resul ti ng
equations are set equal t o zero. When thi s i s done, the following set of equa-
ti ons i s obtained:
j y
7
%Y a +al U 1 +a2U2 +* * - +%+l Um+l =CO
?ya +blul +0 +. . . +0 = c1
c2
a2ya +0 +b2u2 +. . . +0 =
where
15
n
= cr%.Q.(Xi) j =1 , 2 . . . m + l
j I - l J
i=l
n
b. J = T B Q ~ ( ~ ) j =1 , 2 . . . m + l
i=l
i=l
n
c = ziyiQj(Xi) j =1 , 2 . . . m + l
j
i=l
It i s t o be noted that the only nonzero elements i n the coeffi ci ent matri x of
equations (A12) are the diagonal elements and the elements of the fi rst row and
f i rst column. All the other elements are zero because of the orthogonal i ty
properti es of the polynomials used. This i s one of the major advantages i n
usi ng orthogonal polynomials.
ments of the fi rst r ow and fi rst column, except f or the fi rst element, would
al so be zero; and the equati ons would be completely uncoupled, each
being computed completely independent of the others, without the necessi ty of
sol vi ng any sets of equati ons with the resul tant l oss of si gni fi cant fi gures.
I n thi s parti cul ar case because of the added constant Y a, the equations are
not completely uncoupled, but they are very nearl y uncoupled and can readi l y
be solved.
I n the usual case of data f i tti ng, al l the el e-
uj
Thus f or any equation after the fi rst
Substi tuti ng i nto the fi rst equation and sol vi ng f or Y a gi ve immediately
16
APPENDIX B
FORTRAN N PROGRAM
B I D Y A G 1202 E g N t S T KOL3CRTS9 J K. - 140 M- S - P A X 6 1 5 2
B L I B S I O CON I I N U C
b I EJ Ot l S UURCE
d I6 F T C P R M T R 1 L 1 S T 9 K E F t O t C K
C
C
C
c.
C
C
C
C
c
C
C
C
c
C
C
C
c
c
C
C
C
c
L
C
C
C
C
C
c
c
C
C
i
C
c
C
C
C
C
C
C
C
L
c
C
C
C
C
C
C
C
C
C
C
C
c
C
C
L UE E P / >T R E S S - K UP T UR E P AR AME T E K P KOGKAM P R K l
P K MT
P K P T
DO S T ANDAR D L , VI AT I I - l N P U k T
K K DE L R E F O F F R E E OUY P RMT
K M NUMt3ER OF VAL UE S OF M REAL, P K P T
KU NUMBE R OF VAL UE S OF I) R t A O P R C T
KK FuUPlHCK UF VAL UE S OF H R E n U P K P T
K T A NUMBE R OF V A L U t S OF T T A K L A O P R HT
M UE ti R E E P U L Y N U P I A L P K P l
IV NUMBE R ( I F D A l A P O I N T 5 P KMT
PP P AKAME T E Y P R P T
d S T R E S S E X P U N t N T PKMT
Od P O L Y Y U V I A L P K K T
n T t MP E R A T U K F t X P O N E N T P R NT
R A T I U A B b ( Y - Y Y ) / D U P K K T
5 I LIMA 5 T R t S S P K NT
5I ti d S l u P A * * O PI<MT
T 1 I M t P KMT
TA T I M E I N T t f l C E P T P K P T
T A U S I b M A * * O * I T T - T T A ) * * K P KMT
T AUS dR T A U **2 PRCIT
T I M k C AL C i J L AT E d T ( l O . * * Y Y ) P K MT
T TA P RMT
X LUb S I G N A P KMT
Y Lob T P U k T
Y A L U b TA PKMT
Y Y C A L C U L A T E 3 L U G T P R MT
P R K I
A L L L U A & T I T I E h I h CUMMUQ WI T 1 4 THI5 PKUGRAlV, A"\L) T l l l S P A P t l i P H Ml
A ? : P t P R k S F N T t O BY T : i E SAME 5YMt3C)L, WI T H K F P t A T t t ~ P RMT
L t T T t R h I N DI L A T I ' \ I G lrll- U P P t K C A S t ANU I ; %ttK L t l I L H 5 4 E l V U \ P L L L L ' l - P A i <T
OUT. P RMT
P K K T
P K k T
P K MT
S E L E C T > P AR AME T E R P R DDU C I N L S MAL L E S T R t S I O U A L A h d GUTPUT:, A P R Ml
C OMP L E T E T ABL E . K F S U L T S OF A L L U T H t R V A L U E S AR F S U MK A R I ZE D I'd P KF i T
A S HOATE K TAtrLE . P K MT
P KMT
* ~ + * ~ + + ~ * + a + ~ ~ ~ ~ ~ f f * n C n r c a + , r + a P RMT
P RMT
T l T L t L A K D ~ VIOL) CARD, AND F I V E ( 5 ) S E T S OF DATl i . AT Ti l e [NU UF P R K T
E A C H S t T OF uATA MJ L T dE A C AR U WI TH T t i t WURO ' t U U ' I N THE F I R S T P R P T
MUS T HAVE tl L ANK S I N THE F I R S T T H R E E COLUMNS . L U L U h, \ h 7 3 - 8 0 ARE P RMT
1 NLOAEI ). PRK T
P R M l
T I T L E - ANY A L P HA ME R I C I i XF !3KMATI UN- - HE AD5 t A C t i P AGE O F U U I P U T P R MT
P K k T
NOME NC L AT UR E I S A S F DL L Oh' S P u K T
Tr T E MP E R AT UK E P n N i
T EfiP E R A T UR E I ii T t R C t I' T
VU OGKAM E X T R AP OL A T E S CR E E P / S T K t S S - A UP 1 UR L .)A I A U 5 1 ' UC A
P P = ( Y / S 1 GMA * ' 0- Y A / ( T T- 1 T A 1 * *K ,
GE I L E R AL I L E U P A K A M t T t K P r<r I
IHqtE COLUMN>. AL L DA T A C A ~ D S (EXCF P TI Y L TITLE ANI) vni)E C AR DS ) P R V T
MODE C AR D - bNE D F I t i R t i k WORDS I N C UL UMN5 1 - 6 9 ' L A K S J N ' , ' L I N A R ' p P R M T
OR 'GENRAL'. Tt i I s C AR D DEFIYES ' K K ' , T ~I E DEGREL OF P K w i
E UE E DOM, USED I N C A L C U L A T I N G GUODNESS UF F I T . P f i MT
P R K I
U A T A Sk T 1- - VAL UE S U F T r A T I 1 t3E I N V t S T I G A T k D - - D N F P t K C A 4D P R K T
17
L
c
C
C
c
C
C
L
C
C
C
c
c
C
C
c
C
C
C
C
C
C
c
c
i;
C
c
c
C
C
C
C
C
C
C
C
C
c
C
C
c
C
C
C
c
C
C
C
C
C
c
C
C
C
c
1
F O 2 M h l ( 3X, F l O. O1- - 50 VAL I J E S MAXI MI J M P R R T > Y
P RMT 60
J A T A S t T Z- - VAL UE S UF T t MP F E A T U K E E XP O. \ l tQT, 2, TJ J E I X V E b T I C A T F O P K MT ti
UNE P E A C AK I I - - F OK MAI ( 3X 9F l O. O1- - 2C V A L U E S HA X I MU M P R MT 62
P R K T 6 3
ONE P E R CAKD- - F ORMAT ( 3X 9F l O. O1- - 2C V A L U E S f"lAXI P I UM P R P T 6 5
P K MT 66
QAT A S t T 3- - V A L J E S UF S T K F S S CXP UNE NT. ' i , T O HE 1' 4V E bT I G A T E D P R P T b 4
OA T A b t T 4- - U E G K E t 5 O F P U L Y N U H I A L , M, T O 13E I N V E S I I Z A T t d P K MT b 7
UNE P E ; ~ CARD- - F URMAT (~X,I Z)--MAXI MUY VALUE ~ O I ro P R MT t~
E x C E t u L o - - L t R u MA Y .mr ~ J E USED. P RMT 5 9
P K MT 70
U A r A j t T 5- - L)ATA P O l U T 5 I N THE OK DE K T t M P t K A T U K t t S T I I E S S , A N 3 P R P T 7 1
1I ME - - - c) 1ut > r P L R CAKD- - F URMAT 1 3 X t 3 F l G . 3 ) P RMT i 2
T H t VAL OE UG S T Y t S S I S A V T O MA T I C A L L Y D I V I D E 0 r3Y 103C P R MT 73
F OR A L L C A L C d L A T I U N S E X C t P l F I N U I N G THE L OG STRL j S. P RMT 14
L O O 5E T a MAXI MI J M. P K MT 75
P R NT 76
n * n a r n + a a a t r & t ~ F A T a x + a a c c + + c a r ? &MT 77
P RMT 7d
L h C H UF T H t F I V E S t l S U F iDATA MUS T trE F UL L UWE D 6 Y A C AR U H A V I N G P ANT 79
T I i t dud( ; E kD I Y Tt4E F I R b T T hK E E C OL J MNZ. P K MT PO
ALL ~ A T A cAg,J s ( E x c t P r I x b r I T L L AND MOO^ C A ~D S ) Y U ~ T HAVE i r l ~ P R MT 01
P K MT 8 3
' d I TI 1 I N E A C h b E T , I I AT A MAY -)E I N ANY bR. I Et7. 1 1 HI L L 3E P XUC L S S C O 2K C T 6 4
I:\ T l i E TJ KDEg P KE S E I L TE O Ti) THE MAC HI NE . I' t <E" T b 5
P R K T L O
T H t L A L C U L A T I O Y S A t l t P t R F O R Mt O 1U F UUR ( 4 ) L OOP b. P KC. 1 87
GUI NL ; F R OM I i U Y F S MOS I T[ J LI UTF KMLI ST, Lit L i J A N T I T I t S A&E V A ? I E 0 P R MT t H
It\ THt F U L L I J r l I NG U R J E R PREAT b 9
L J E G K E ~ P C ) L Y NUI Y I AL t M P R MT 4(1
V A L U E LJ F T T A P RMT 3i
I E MP L R A T U K E E xP U U t8r l T , c( P K MT 9 2
>T dE S S tXP ONtmVT t LI P R MT ' i 3
P RMT 74
T HE U U r P U T T A B L E S U T I L I L t L F S S T!4A8\l 120 C 3L UMNS 3'4 T r i E P K I k T E R P K MT 9 >
Aha E X P E C T :.ILI C A 3 R I A b E C UNT R OL S OTHEK rHAl V 1, 0 , + AVO dL ANK . PRMT 96
A L I f \ l E CUJ I VTtR 15 I I VCORI 'URATEU TU L I M I T I J L I T P dT 1 3 69 L h t S P t R P K V l 37
P AL E . F OK E A L H U F d P AC E I HE T I T L E AND A P P K U P R I A T E C G L U Y Q HE A' I I UGS P K MT 3 r :
PHMT 101;
P A G t CuUNTI Nb ANL) E : tKOK T Y AP S MUS T i l E P ROV1T)EI ) t l Y T t l E O P E K A l I N b P RET 1 L l
P KMT l b 3
PKlJ G17AoY N I T t i I d S Y S AYD I OC>Pl dI L L K UN U N A 1 6 K MP C I I I U E P KMT 1 3 4
P R MT I O 5
P RMT 1Vb
L OG I C A L T R bGi C l , T 9GGA2, I R ZG 3 3 P RMT 1 u 7
P K MT lC'8
d I V t 1 \ b l C & T I l L E ( 1 Z ) , T A ~L t I S , 1 1 O ) ~I T t r L l 6 ~1 1 0 ~ P HMT I33
P R M T I I C
E U U I V A L E N C t ~ l A ~ L ( l t 1 ) , I T ~ L t ( l ~ 1 ) ) ?REIT 1 1 L
P KMT 112
COMMON / D A T A / S I G MA l L O 1 I , T I 201 ) , T T ( 201 1 P K M l 113
I / T R Y S / E A l 2 1 ~ ~ ~ ~ 5 1 ~ ~ K ~ ~ l ~ ~ T T A l 5 I I PK PT 1 1 4
3 / C A L C / P P I ~0 0 ) , 9 A T 1 0 ( 2 0 0 ~~T I M t ~2 ~~) t Y Y ( Z C O ~ P Y MT 116
4
P K P T 117
5 / P L Y h P 1 L / L l T , l t N l ( 4 7 2 1 1 tY A , I I T tl E R Z( 6 3 ) P RMT l i b
PRPIT 1 I - J
P R MT 121..
I N P U T P R MT 121
P K MT 1.2~
d. r l <I Tt (6,99941 P R Ml L L 3
XEAi ) ( 5 9 Y O C l ) ( T I T L t I K ) t K =l t l 2 1 P RMT 1 2 4
F I dST THRE E Ci j LU'I NS dLAi \i K. P n f i r 8 c
A & t P R I NT E I ) . I 'KUGKAM V I I S r l I 1 1 l A T R A L S F t Y TU T h L I Y I T I A L d t A 3 . P R MT
i Y i T E M . P K M l 122
2 / F DA T A / S I G: 3( LOO , T A U ( 200 1 , T A U bO d( 2OC I , X( L O P I t X X I 2.20 1 t Y ( Z O C ) PRM1 115
/ t N U / L Q*l/ Y / N / UI)/ d 3 / IO GK t t / D E Gd E C
--
18
10
15
20
25
30
100
c
C
c
110
c
C
c
C
c
C
c
C
112
K E A U (5, 9001) DE G R E t
K = O
K = K +L
K t A U (>,90021 C H t C K , T T A ( K )
I F ( C HE C K . Y f . tND) GO TO 10
K I A = K - 1
K = O
r ( = K +L
R E h U ( 5 , 9 0 0 2 ) C H E L K , K ( K )
1 F ( C HtC K . NF . tND) GO 10 15
K K = K - 1
K = O
K = K +L
R E A D (5, 90021 C H E C K , d ( K )
I F ( CHL CK. NE . L ND1 GCI TL) 2 0
K U -L. I(-1
K = O
K = K + l
A E A U (5, 90031 C I I F C K , M( K I
i F ( C H L C K . N F . ~N J ) GO T I 1 25
K M = K - 1
U = O
K = U + l
R t A 0 ( 5 , 9 0 0 4 ) C t I F C K , T T ( K ) , S I L M A ( K l r T ( K 1
I F ( L Ht C K . Nt . E NU ) L O TO 39
N- K - 1
t N 0 UF I NI ' U1
t l N D L b G S T K t b 5 A"4U L OG T I M t
dU 100 K =l , N
X ( K I =A L U L 1 0 ( z I G MA ( K ) 1 + 3 .
Y ( K l =A L 0 6 1 0 ( 1 ( K ) )
L U NT INclE.
I N I T I A L I ZE CO, \ I 5TANl S
J Ul =l . t 5
L I \ t S =' j l
I I < ti tili 3 =. F A L 5 t .
IN T i l Y =0
hC A L E L OGS OF S T K k S S
C A L L S L A L E
t-I ,\d H I G H E S T d t G N t E P U L Y k O MI A L
MAX = U
UO 110 K =l , K M
MAX = MA X O ( MA X , M( K l )
L ONT I UUE
MAJ OK L UUP - C A L C U L A T E S A L L Y ( A ) b ANU Y t b I L ) I J A L i
wR I T t S SUMMARY T A D L t
i-I Nl)S S MA L L E S T R E b I i I U A L
UO 500 K S =l t r ( i )
C A L C U L b T F SILMA**O
19
1,
C
c
c
115
118
119
120
C
c
C
c
c
c
d 0 30U K 3 =l r K T A
L A L C U L A T E T A U ANLJ TAU**, ?
iJ0 120 r<=l , i \ l
ILJIFF=aBS(TT(Kl-TTAiK3)1
T A U ( K I =O.
b0 TCI 119
T A U ( K I =SI GO( K l *TI ) lFF*ad( K 4 1
T a U S P H i K l = T A U ( K l * * 2
I F ( T ! J I F F l l l b r 115,118
L ONT I l vuE
t V A L U A T E P 1.1L Y d 0 M I A L S
C A L L P d L Y ( P A X I
u0 200 K 2=l . r ( M
L A L C U L A T E T i i t U K E T I C A L L OG TI r4l th A X 0 T I ME S
L
C L C M P b T t K E S I O u A L
L
C A L L K t S I D ( M I K 2 1 1
c
L MAKE Us\ t Q l K Y I N SUtbllVARY T A B L t
L
C
L
i
C
L
c
170
175
180
N T d Y = N I K Y +1
T A d L E ( L 1 N T R Y ) =<i K 4 I
Tal l Lt( 4, NT K Y I =T T A ( K 3 1
I A d L t i 5 , N T K Y ) =Y n
T A I ~L ~( ~, N T K Y ) =J I J
I RGL KL =NTRY . E O. 2*L LdES
r A t l L E i L, NT K Y I =L A( K 5 I
1 T tdLb ( 3 9 NT K Y =M i K 2 )
I F ( l K b G K 2 1 bd TLI 170
bG T U 199
d L T P U T h 0 U t PA GE 13F SUMMA<Y T A d L E
u b T P b T T I T L E A ND HE A QI Y G S F OR S UMVARY T A L L E
P R P l 1%1
P RMT 1 j L
P 2kT L 94
P K K T 195
P Y MT llt
P RMT 1 3 7
P R MT 178
P K HT 199
PRMT 2 0 5
P R MT 2 b i
P RMT 2 L Z
P R MT 223
P K MT 2 6 4
P K MT 2U5
P RMT 226
P K P T 2 0 7
P RMT 2C8
P R M I 2 0 0
P RMT 210
P KMT 211
P K NT Z L L
P KP T 214
P K MT 215
P dMT 216
P R K T 217
P R K T I l k
P K P T 219
PP.MT 2 ~ 0
P RMT 221
P K K I Z C L
P R HT 2 d 3
P RMT 2 ~ 4
P KMT 225
P K P T 2 2 6
P A C T 2 ~ 7
P gP T 2 ~ 8
P R P T Ld. 7
P K P T 2 3 2
P KMT 231
PRF.!T 2 j L
P 4k T 2 3 3
P KY , T 2 3 4
P HMT 2 3 >
P R P T 2 3 b
fXlJ T 1 93
P w r z r 3
P R P T 237
P H#T 23c
PRPT 23Y
t<P,l 2 4 5
P K h T ?41
P RMT 242
P K MT 2 4 3
P R M1 244
P R NT 2 4 5
P RMT 246
P R k T 2 4 7
P KMT 2 4 6
P RMT c 4 9
P R P T 2 7 0
PY >l l 271
P RlVT 2 2 3
P kMT i 5 4
P RMT 255
P K P T 2 5 6
P RMT j b i
20
I
C
C
c
190
200
500
400
500
C
C
L
C
C
L OO0
I F ( T R b C 4 3 ) GU TO 1000
S AVE V A L U t S P R ODUC I NG S MAL L E S T R t S I d U A L
1F ( D D i . L t . 0 0 ) GU T U 200
M l = M ( K 2 )
T T A l =T T A ( K 3 )
R 1 = K ( K 4 )
U A = U I K 5 1
r A A = YA
uo1 =DO
i C N T I k U E
L ONT I h d t
LUNT I NuE
C OhT I N b E
I F ( NTKY . NE . 01 GU TU 170
E ND MAJ OR L OOP
LIIITPUT OP T I MU M VAL UE S ANU H L A U l N G F OK F U L L T A d L E
TRGGR3=. TRUt.
L O ~ T I N U t
C
C
C
L
C L A L C U L A T E T H t J K E I I C A L T I K E S , R A T I O 5 UF D I ~F L ~E U C F J
i T O R k S I D U A L , A Ql I V A L U L S UF T H E P AK UP l tT E K , F UX THF
C P A K A M t l E R P K U d U C I N G T Ht MI NI i Y UM R t 5 l U U A L
c
UC 1035 K =l t i \ i
T O I F F =n B ~( T T ( K ) - T T A l )
b I L U ( K ) =S I b M A ( K ) **ti1 .
I F ( T U I F F ) 1 0 3 2 , 1 0 3 1 , 1 0 3 2
1 0 3 1 l A b ( K ) =O .
1 0 3 2
1 0 3 4 T A U L L K ( K ) = I A U ( K l * * 2
b0 T O 1 0 3 4
T A U ( K ) =5 I G &( K ) *TJ I F F +*K 1
1 0 3 5 LONT IKJF
uu=c31
C A L L P u L Y ( P A )
L A L L Y b U b A ( M 1 )
C A L L Y l h ( M I )
C A L L R A T I O 1
C A L L P AR AM
C
C UUT P UT FULL T A B L E
C
K = O
A R I T E (6,9012) T T ( K ) , S I G M A ( K ) , X ( K ) , T ( K I , T l ~I ~( K ) , Y ( K ) ~Y Y ( ~) ~
L I N E S =L I k ES t 1
1040 K = K + l
1 R A T I U( K ) , P P ( K )
1F (K. Ei 2. V) L U T O I
I F ( L I i kF S . L T . 601 GU T U 1 0 4 0
w K I T k (6,9005) ( T I T L E ( K K K ) , K K K =L , l Z ) , D ~L ~~t
H R I T E (6,90111
L I NE S =e,
b C T O 1040
t k D UF P R OL R AV
P RMT 2 >7
P KMT 258
PKk8T 0 9
P K K 1 7 (, 0
P kMT 201
P K k l Z C J ~
P dMT 2 6 5
P R k T 2b4
P R k T 265
P RMT 166
P R K I 267
P KP T 268
PRET io')
P k V l 2 1 0
P P MT Z I l
P RMT 2 I L
P RMT ? 1 >
P RMT 2 7 4
P K hl - 2 1 5
P KP T 27. 5
P KWT 277
P K P T 2 7 h
P K PI T 2 I ' I
P P NT i t 0
PHMJ Z t , A
P K P l Z U L
P K bT ? b 3
P P R T 21-14
P R h I 285
P K MT l t ' b
I P RVT 2 b l
PKI.:T St35
P R P T Zbir
P RMT Z, >O
P RMT 291
PRPrT 2YLI
P RE 1 2' 13
P R K T 2 4 4
P K h T T35
P R K T r"i6
PRPcl 2'1 I
P KMT Z'f P
P R bT c 9' 1
? KHT 39L
P R M T I i 1
PRP!T 3 i L
P K NT 3 5 3
P R N T 3 ~ 4
P R K T !u5
P' KMT jL'6
P R r T 3L7
P R K I j u8
P RMT 13Y
P R M I '5AG
P R K T 3 1 1
PHP. 1 3 l L
P RMT >A3
P RMT 3 1 4
PK PT 315
? R k T 3 1 6
P RWT 3 1 I
P R K T 318
P R K T 3 1 ' 9
P R M l 3 L O
P R h T 3 2 1
P K K I -czz
21
c P K k T 3 ~ 3
C F ORMAT S T A T t M E N T 5 F OR P ROGRAM P R K T 3 r 4
c P RF ; l 3 2 5
c F OR MAT b F OR i i \rP UT P R MT 3 ~ 6
C P R M l 3 ~ 7
9001 F ORMAT ( 1 2 A 6 J P R K T 3i8
9002 F UAMAT ( A3, F l O. O) P R MT 329
9003 F ORMAT ( A 3 9 1 2 ) P K MT 3 3 Q
9004 F OR K AT ( A3~0P F 10. 0~3P F 10. 0, OP F 10. 0) PRMl 331
L P R k T 3 3 1
C F OK MAT b F O i i GUT P UT P RMT 333
C PK K T 334
c T I T L E (S KI P . , TU N t W P A G E ) P KWT 3 3 5
C P RMT 336
c P R MT 3 3 6
C P RMT 3 4 0
9005 F O K M k T ( l H 1 ~2 U X ~l Z A 6 / 1 H , 30X , A b, l OH P A R A C E T E K / l H 1 PKFIT 3 3 1
C AUMMAUY Ol- I N P U T P K Y T 33y
9006 F OK MAT ( 1 H , l OX , 45t i L R E t P / R U P T U K E P A R k M E l F K b AK F I S I V t S T I b A T t U F dK/ P K r T 3 4 1
11H , I 2, 18H V U L U E ( S ) OF T ( A ) , , I 3 , 2 5 H T E MP trR AT i J R t t X P t l U t N T ( S ) , t I 3 , PUP1 3 4 2
ZL4H b T R k S S E X P O U E N T ( S ) , A NO, I 3, 14H P O L Y N J M I A L ( S ) / l I l tl :!)i tStI dSI \i G~ P KY NT 3 4 5
314,12H D A l A P O I ' v T S / l I 1 ) P K Y l 3 4 4
C P RMT ? 4 5
C H E A D I N ~S FGA L dMMAR Y T A B L t , U N t LI h E OF S UY P ARY T A G L t P K MT 3 4 5
c PHWT 3 4 7
9007 F OK MAT 9 2 ( 2 X , 1 I-1G t 7 X 9 1 HR 9 6 X 9 1 HM 7 5 X 9 4 H T ( A ) p 5 X 9 4 H Y ( A 1 9 4 X 9 PRF: 1- 3 4 8
1 8HSTO. DEV. , I O X ) / l H 1 P K K T 3 4 9
9008 F OdP A T ( 1H ~0P F 5. 2~~8. 2, 15, F 9. 01F 10. 2, 1P E 11. 2l P l l Z) P R P T 35b
9009 F U d k A T (lH+r~RX,O?F5.2,F~.21I51F9.O,F1C.ZllPEll-2) P KMT 3 5 1
C P RMT 3 5 Z
C U P T I MU N V A L d t 2 P K MT 353
c P R K T 3 5 4
9010 F U K M A T L l H 10X 44H V A L U E S P R ODUC I NG S MAL L E S T STAI LL7LXD i )tVI UTI LJ [', /3HI )u'=P KMT 3 5 5
l F S . Z, 4H, R =F 5. 2, 4Hr M=I 2,7ti , T ( A ) =F 6. 0, 7H, Y ( A ) =F Y . 3, 11i l , S TO. J F V. P RMT 356
2 =1 P t 9 . L / l H O ) P KMT 3 3 7
C P R K T 5 5 6
C H E A D I N ~S F OX F U L L T AHL E , ONE L I N t OF F U L L T A b L E P R P T 359
C P KWT 363
9011 F OR K AT ( 5H T t H P 1 4 X , O H S T ~t S j , 3 X , 3 H L ~~, 6 X , 4 H r I M E , ~~X , 6 ~i L A L ~l ~, j X , P R Y T . 361
13HL OG, 3X, 8HC AL C L O G , Z X ~~~~E V / S L J , 3 x 9 9t I P A R A ME T I H/ 1 t i T bX, t; i i ( at- 3) t 2 X t P R P 1 3 h L
Z 6 H ~ T ~ E j S , 1 4 X , 4 H T I M t ~ 5 X ~ 4 H T I ~ E ~ 4 X ~ 4 ~ T l M ~ / l ~ ) P K M l 3 6 5
( 1H
9012 F ORMAT ( 1H ~ ~ P F 5 . 0 ~ ~ 8 . 1 ~ F A . 3 ~ 2 F l C ~ l ~ 3 F Y . 3 ~ l F l Z ) P RCT 3 6 4
c ? RMT $ 6 5
9999 F OKMAT ( 1H1) P R C T 3 b b
c P R Y 1 357
t ~i i P UMT 3b &
22
S I B F I C P R MdL K L I S T ~R E F I D E C K
C S E T S F I R S T P U L Y N O MI A L TO U N I T Y A T A L L S T A T I O N S AVD S T OR E S
C A L P HA ME R I C CUDt MOROS
C
BL OC K UAT A
COMMON / P L Y N M L / Q Q ~ 2 1 ~ 2 0 0 l ~ 0 T H E R S ~ 8 5 ~ / E N O / E N D / N A M E S / ~ A M E S ~ 2 ~
DA T A (~Q(l,K)~K=lr200)/20O*l./~END/3HEND/~
1 (NAMES(K),K=lr2)/12HLARSONLINEAR/
E ND
B I B F T C P AR AM L I S T I R E F I DE C K
C S U t l R OU T I NE F OR E V A L U A T I N G T HE P AR AME T E R A T E A C H P O I N T
L
S U B R OU T I NE P AR AM
COMMON / F D A T A / S I G O ( Z O O ) ~ T A U ( 2 O O ~ ~ O T H l i R S ( 6 O O ~ ~ Y ( 2 O O l
1 /CALC/PP(200)~0THERl(6OO)/N/N
2 / P L Y N ML / O T H E K 2 ( 4 2 2 1 ) r Y A r O T H E R 3 ( 6 3 1
C
UO 10 K =l r N
P P ( K ) = ( Y ( K ) - I G Q ( K ) * Y A ) / T A U ( K )
10 C O N T I N U E
R E T UR N
E ND
C
C
C
C
10
20
30
P RMB
P RMB
P RMB
P RMB
P R MB
P R MB
P RMU
P RMB
P AR M
P AR M
P AR M
P AK M
P AR M
P AR M
P AR M
P AR M
P AR M
P AK M
P AR M
P AR M
P AR M
dI t 3F T C Y T H L I S T I R ~F I DE C K
S U U R OU T I NE F UR C A L C U L A T I N G T I ME S AND L OG T I ME S F ROM T HE P AR AME T E R Y T H
S UBROUT I N E Y T H( M I
CUMMON /CALC/OTHERS(400),TIME(200)~YY(200)
1 / F O A T A / I G Q ( L O O ) ~ T A U ( 2 0 O ~ ~ O T H E R l ~ 8 0 0 )
2 / P L Y NML / U0( 21, 200) rU( Zl ) , Y A, UT HtR 2( 63)
3 / N/ N
U O 10 K =l , N
Y Y ( K ) = 0.
M l = M+1
L ONT I N u E
00 30 K =l , N
00 20 J =l , M l
Y Y ( K ) = Y Y ( K ) +Q O ( J , K ) * U ( J )
C ONT I NU E
Y Y ( K ) = T A U ( K ) , Y Y ( K ) +S I G O ( K ) ~Y A
T I M F ( K ) = 1 0 . * * Y Y ( K )
C ONT I NU E
R E T UR N
t N D
Y T H
Y T H
Y T H
Y T H
Y T H
Y TH
Y T H
Y T H
Y T H
Y T H
Y T H
Y T H
Y T H
Y T H
Y T H
Y T H
Y T H
Y T H
Y TH
Y T H
Y T H
1
2
3
4
5
6
7
8
1
2
3
4
5
6
7
8
9
10
11
12
13
1
2
3
4
5
b
7
ti
(3
10
11
12
13
14
1 5
16
1 7
18
19
20
21
22
23
p
6 I B F T C R A T I O 1 L I S T t R E F t D E C K
C S U B R OU T I NE F UR C A L C U L A T I N G R A T I O S
C OF I N D I V I D U A L R E S I D U A L S TO ROOT- ME AN- S QUARE R E S I DU A L
C
S U B R OU T I NE R A T I O 1
r
b
COMMON /FDATA/OTHERS(1000~tY(200)
1 / C A L C / O T H E R 1 ( 2 0 0 ) ~ R A T I O ~ 2 O O ~ ~ O T H E R 2 ( 2 O O ~ ~ Y Y ~ Z O O ~
2 / N/ N/ U D/ OD
C
UO 10 K =l r N
R A T I O t K ) = A B S L Y ( K ) - Y Y ( K ) ) / O D
10 ZONT I N U E
R E T UR N
NU
R A T O
R AT O
R A T O
R A T O
R A T U
R AT O
R A T 0
R A T O
R A T O
R A T O
R A T 0
R A T O
R AT O
R A T O
B I B F T C R t S I D L I S T t R E F t O E C K
C S UaR L l UT I NE FUR C A L C U L A T I N G R E S I D U A L KESU
C K t S U
C T H t R E b I D U A L I S dA S E D ON T HE L O L OF T H t T I ME . R t S U
C I r I S D E F I &E u A S THE S QUARE ROUT OF T H t SUM OF THE S QUARE S OF R E S U
C T HE I N D I V I D d A L R E S I U U A L S D I V I D E D B Y T H t D I F F E R L Y C t B E T WE t N T t i t ' 4UMR tS U
C F RE E DOM, K K , DE P E NDS ON THE P AR AME T R ( S t E MA I R BOOY OF K E P OK T ) . R t S D
L K K =2 F UR L A R S U N- MI L L E R P A K A Mt T E R R t S U
C K K =3 F UR L I N E A R P AR AME T E R R E S U
C K K =5 F UR G E N t R A L P AR AME T F K R E S D
t R t S U
C BE R OF DAT A P O I N T S AND T HE DE GR E E S OF kRE CDOM. THE DE GR E E L OF R E S O
C
C
C
C
10
20
30
40
DO = S i J R T ( ( Y - Y Y l * * 2 / ( N - M- K K ) l
S UBR OUT I NE K t S I D ( M )
COMMON /FDATA/OTHEKS(lOOO)rY(2OOJ
1 /CALC/OTHER1(600)rYY(200)
2 / OD/ Du/ N/ N/ OtGR E E / DE GR E E / NAME S / F AME S ( 2)
1 F ( DE GR E E . E Q. F AME S ( 2) ) GO T U 20
I F ( DL R E E . E Q . F A ME S ( l ) ) GO T U 10
ti0 TO 30
L O TU 30
K K = 5
K K = 2
K K = 3
OD = 0.
D = N-M-KK
UO 40 K =l t N
OD = D D +( Y ( K ) - Y Y ( K ) J * * Z
COh T I N U E
DO = S hl R T ( DD/ D)
RETURi'i
t N D
R t S U
R E S U
R t S D
K E S D
R E S U
R t S U
K E S U
R t S O
R E S U
R E S O
R t S U
R t S D
R t S O
R E S O
R t S O
R E S U
R t S D
R t S U
R k S U
R tS L )
R t S U
R t S D
nESu
1
2
3
4
5
6
7
8
9
10
11
12
13
14
1
2
3
4
5
b
7
G
Y
16
1 1
1 2
13
14
15
Ab
1 7
18
19
iL
21
22
2 3
14
2 5
L b
17
LE;
ZY
3C
31
3 2
3 3
34
24
E I B F T C Y bUdA L I S T , K t F , U t : C K
c SUt3RbJ I I NE FUR E V A L d A T l N b Y ( A 1 Y3UI3 I
YSUtl L
Y 5Utj 3
YSUR 4
Y SUt, 5
Y aUti 6
Y SUH 7
Y S U J h
Y SUU Y
Y S Ud I C
Y SUO L i
YSUd 1 2
YSU6 13
Y S Ud 14
Y bUtl L',
Y S Ud I b
Y 3U d 1 7
Y S Ud 18
Y SUd 1 7
Y SUY L >
Y bUd i l
Y S Utl L L
Y bU d L 3
YSUd c 4
Y S Ud / >
Y bU 6 L:,
Y S Ud 27
Y S U8 cti
YSUb L 9
Y S Ui l 3 0
Y S J J 3 1
Y S Ud 3.2
Y S Ud 3 3
YSUO 3 4
Y3UB 55
Y S Ud 36
Y SUB > I
Y S UB 3 8
Y S Ud 3 9
Y S U8 4.1
Y SUB 4 1
Y S dl 3 4 L
Y hU d 4 3
25
51b FTC P U L Y L I S T , K E F , O t ZK
C
C
C
C
C
C
C
C
c
c
10
20
30
40
50
S US R C I dT I NE F UR k V A L u A T I Y G OK T HObUhGL P O L Y N O MI A L S
A L L P O L Y h O M l n L S UP TO MAXI MUM U t S l r ( t D U E G R E t A R t t V A L U A T t U
A T E A C h DA T A P O I N T
TH5 F I A S T P OL Y NOMI A L I S I D E N T I C A L L Y t B U A L TU UP ;I TY
T H t S t V A L U E S AR E S T OR E D B Y A BL OC K DAT A S UBRUUTI ' i E
bU t J R U uT I NE P O L Y ( M1
COMMON / F DA T A / OT I i E R A ( 400) pT AUS OR I 200) , UT HE R Z I 2 0 0 1 9 X X I 2 0 0 )
1 U T H E R 3 ( 2 0 0 )
2 / P L Y NML / ~Ql 2l r 200) , 0T HE K S ( 45) 1A L P HA ( 2O) , ~T ~( 20)
3 / N/ h
P O L Y A
P O L Y i
P OL Y 3
P O L Y 4
P O L Y 5
P O L Y t
P OL Y 7
P OL Y 8
P cj L Y 4
P O L Y 1c.
P O L Y A i
P O L Y 12
P OL Y 13
P OL Y 1 4
P b L Y 15
P U L Y Ab
P U L Y 17
P OL Y i E
P U L Y i s
P OL Y L i '
P U L Y 2 1
P OL Y 2 2
P U L Y L 3
P OL Y 2 4
P OL Y 2 5
P U L Y L U
P U L Y 27
P OL Y i b
P U L Y 2Y
P OL Y 3r,
P OL Y 3 i
P t i L Y 32
P OL Y 3 3
P U L Y 34
P O L Y 35
P U L Y 36
P OL Y 5 7
P b L Y 3 8
P LI LY 3 Y
P U L Y 4 L
P i l L Y 4 1
P OL Y 4 2
P U L Y 4 3
P OL Y 44
P cl L Y 4 5
P OL Y 46
P OL Y 47
26
S I B F T C S C A L L I S T t K E F , O E C K
C
C
C
C
C
C
10
20
S U dK OU T I NE FUR S C A L I N G L OGS OF ST RESb
i H t 5 C A L t O V A L U E S L I E I N T HE K t G I U N - 2 T O 2
5 U d K U U T I h E S L AL E
COMMON
ellti = 0.
S MA L L = L e t 5
/ k D A T A / O T l i t K L ( 60O) , X ( 2 0 0 ) ~X X ( 2 0 0 ) , OT tl tR 2( 20' 3) / N/ N
UO 10 K =l , N
ti I G = A M A X l ( b I G , X ( K ) )
J MAL L = A M I N L ( $ M A L L , X ( K ) 1
L ONT I NU
A = 4. / ( bI G - >MA L L )
U =2 . * ( ~I G +S M ~L L ) / ( ~I G - S M A L L )
00 20 K =l , N
X X ( K 1 = A e x ( ~1 - B
CONT I :QUE
RETURN
t i l 0
S C AL 1
S C AL i:
S C A L 3
S C AL 4
S C AL 5
S C A L t
S C A L 7
S C AL i:
S C A L 9
S CAL L O
S C AL 11
S C A L 1.2
S C A L 13
S C A L 14
S C AL 15
5 C A L 1 7
S C AL 18
5C A L 19
$C AL LC,
S C A L 2 1
S C AL It,
27
IllllllllIlIIllll IIIIIIIIII I I 1 I 1 Ill1 I II lIIm~11ll1l1l 1l1ll1l II Ill
REF' EI I ENCES
1. Lar son, F. R. ; and Mi l l er , J ames: A Ti me- Temper at ur e Rel at i onshi p f or
Rupt ur e and Cr eep St ress. Trans. ASME, vol . 74, no. 5, J ul y 1952, pp.
765- 771; Di scussi on, pp. 171- 175.
2. Manson, S. S. ; and Haf er d, Angel a M. : A Li near Ti me- Temper at ur e Rel at i on
NACA TN 2890, 1953. f or Ext r apol at i on of Cr eep and St r ess- Rupt ur e Dat a.
3. Orr, Ramond L. j Sher by, Ol eg D. ; and Dor n, J ohn E. : Cor r el at i ons of
Rupt ur e Dat a f or Met al s at El evat ed Temper at ur es. Trans. ASM, vol . 46,
1954, pp. 113- 128.
4. Manson, S. S. : Desi gn Consi der at i ons f or Long Li f e at El evat ed Temper a-
t ures. NASA TP 1- 63, 1963.
5. Manson, S. S. ; and Mendel son, A. : Opt i mi zat i on of Par amet r i c Const ant s f or
Cr eep- Rupt ur e Dat a by Means of Least Squares. NASA MEMO 3- 10- 593, 1959.
6. Hammi ng, R. W.: Numer i cal Met hods f or Sci ent i st s and Engi neers. McGr aw-
Hi l l Book Co. , I nc. , 1962, pp. 223- 246.
7. For syt he, G. E. : Gener at i on and Use of Or t hogonal Pol ynomi al s f or Dat a-
Fi t t i ng wi t h a Di gi t al Comput er. J. SOC. I ndust . Appl . Mat h. , vol. 5,
no. 2, J une 1957, pp. 74- 88.
8. Ti t r an, Rober t H. ; and Hal l , Rober t W. : Hi gh- Temper at ur e Cr eep Behavi or
of a Col umbi umAl l oy, FS- 85, NASA TN D- 2885, 1965.
9. Mur r y, G. : Ext r apol at i on of t he Resul t s of Cr eep Test s by Means of Par a-
met r i c For mul ae. Vol . 1 of Proc. J oi nt I nt . Conf . on Cr eep, I nst . Mech.
Eng. , 1963, pp. 6- 87 - 6-100.
10. Manson, S. S. j and Br own, W. F. , J r. : Ti me- Temper at ur e- St r ess Rel at i ons
f or t he Cor r el at i on of Ext r apol at i on of St r ess- Rupt ur e Dat a. Proc. ASTM,
vola 53, 1953, pp. 693- 719.
11. Manson, S. S. j Succop, G. j and Br own, W. F. , J r. : The Appl i cat i on of Ti me-
Temper at ur e Par amet er s t o Accel er at ed Cr eep- Rupt ur e Test i ng.
Trans.
ASM, V O ~. 51, 1959, pp. 911- 934.
28
1
Index,
i
1
2
3
4
5
6
7
8
Time,
t,
hr
4954.68
L1365.9
2908.
1340.
625.342
117.371
34.4856
995.25
TABLE I. - CALCULATION OF POLYNOMIALS FOR THEORETICAL DA!l?A FOR IMLRD DEGREE WLYNOMlY&
[Temperature i ntercept, Ta, 600' F. 1
Stress,
0,
psi
56 300
19 800
30 300
5 080
12 900
778
4 190
66
2
Tempera-
T,
OF
ture,
1100
1100
1200
1200
1300
1300
1400
1400
0
1
2
3
4
3 1 4
- - - - - - -
0.27092
-.57813
.28432
.41260
5
l og t
3.69501
4.05560
2.79612
3.46359
2.06956
3.12710
1.53764
2.99793
6
log a
4.75051
4.29666
4.48144
3.70586
4.11059
2.89098
3.62221
1.81954
7
q(T-Ta)l
500
500
600
600
700
700
800
800
8
Scaled
1% 0,
X
2. 0
1.3806
1.6328
.57433
1.1267
-.53777
.46017
-2.0
I
9 1 10
Polynomial
Ql
1
1
1
1
1
1
1
1
TABLE 11. - I NTERMEDI ATE CALCULA'LIONSFOR THEORETICAL
DATA FOR THIRD DEGREE POLYNOMIAL
[Temperature i ntercept, Ta, 600' F. 1
Q2
1.3619
.30341
.e5576
-. 80869
.le925
-2.2709
2.8030
.51879
l l 2
Index, I a
J I
3
P
-------
0.
1.6548
1. 3315
.a0618
4
a
8.0
786.18
465.68
286.01
5200.
5
b
- - - - - - - -
3.48 Xl Ot
5.7589
7.6678
6.1816
6
C
23.743
14897.
2616.
3796.9
2694.6
Q3
-0.19594
-1.6875
-1.4583
1.5741
2.5067
-. 90880
-. 78251
.015445
-.57205
1.7195
-1.0745
.92564
-. 77354
-. 98133
29
TABLE 111. - FI T FOR SEXEFLAL VALUES OF LINEAR
PARAMETER FOR THEORETICAL DATA
Degree of
polynomial
Temperature,
Ta
500
500
500
600
600
600
700
700
700
Vari abl e,
ya
10. 54
10. 54
10. 55
9. 49
9. 50
9. 50
a. 44
8. 46
8. 45
Devi ati on
0. 008049
. 010660
. 004859
. 000002
. 000003
. 015412
. 013937
. 014469
. 0097a6
TABLE I V. - COMPOSITION OF STEELS RECEIVED
FROM GERMAN COOPERATIVE LONG-
TIME CREEP PROGRAM
[&-recei ved, 20- mm- di am. bar stock. 1
Element
Carbon
Si l i con
Manganese
Chromium
Molybdenum
Columbium and
tantal um
Ni ckel
Ti tani um
Vanadium
Tungsten
I
Composition, percent
I
C
( 23b CK)
0. 065
. 47
. 60
17. 24
2. 08
. 02
11. 90
. 39
. 10
Less than
0. 005
Steel
P
( 14a PA)
0. 270
. 26
. 60
2. 62
. 27
Trace
. 14
Trace
. 26
Trace
~
K
(27b KK)
0. 068
. 45
. 73
16. 14
2. 10 I
*44 I
13. 12
Trace
. 05
Trace
30
TABLE V. - NASA f i - DATA
(a) Steel K (27b KK)
Stress,
0,
psi
20 000.000
20 000.000
20 000.000
20 000.000
20 000.000
20 000.000
eo 000.000
20 000.000
20 000.000
40 000.000
40 000.000
40 000.000
40 000.000
40 000.000
40 000.000
40 000.000
40 000.000
60 000.000
60 000.000
60 090. 000
60 000.000
60 000.000
20 ooo.ooo
Temperature,
T,
OF
1022.00
a1022. 00
alo22. 00
a1022.00
a1022.00
a1022.00
"1022.00
a1022. 00
a1022.00
a1022.00
aloaz. 00
a1022.00
1112. 00
1112.00
1112. 00
"1112.00
a1112.00
"1112. 00
a1112.00
a1 112. 00
a1112.00
a1112. 00
1112.00
Time,
hr
t ,
0.400
1.900
4.450
23.700
25.500
38.
136. eo0
394.800
704.600
1212.
2.700
7.500
15.203
44.400
377,
1417.
2110.
5367.
. 610
1. 250
4.400
4.500
10. : >OO
lemperature
T,
??
a1600. 00
a1620. 00
a1660. 00
"1680.00
a1700. 00
"1740.00
a17P0. 00
a1425. 00
a1450. 00
"1480. 00
Stress,
0,
psi
77 000.000
72 500.000
72 000.000
70 000.000
68 000.000
66 000.000
66 000.000
65 030.000
62 500.000
60 000.000
60 000.000
55 000.000
68 000.000
65 000.000
62 500.000
60 000.000
57 500.000
55 000.000
52 500.000
50 000.000
45 000.000
43 000.000
37 000.000
a1500. 00
a1520. 00
"1560.00
T i me,
hr
t,
1.500
13.800
10.
36. 700
60.400
73.300
107.600
201.300
250.400
990.
817.500
.750
2. 250
4.300
13.900
22.700
51.500
147.500
2R3.
1 020.
1 579.
13 140.
3 680.
"1570.00
a1600. 00
1650.00
1700.00
a1260. 00
a1202.00
a1290.00
"1292.00
a1292.00
a1320. 00
a1360. 00
a1400. 00
1440.00
i 4eo.m
a1112.00
"1160. 00
allso. 00
a1202.00
?tress,
0,
psi
5 000.000
5 000.000
5 000.000
5 000.000
5 000.000
5 000.000
5 000.000
1 000.000
1 000.000
1 000.000
1 000.000
1 000.000
1 000.000
1 030.000
1 000.000
1 000.000
1 000.000
1 000.000
1 000.000
1 000.000
1 000.000
1 000.000
1 000.000
1 000.000
1 000.000
1 000.000
1 000.000
) 000.000
1 000.000
) 000.000
) 000.000
Time,
t,
hr
570.200
156.800
91.600
62.700
40.500
10.600
1690.
550. 300
270.
170.
120.500
40.
;71.500
15. ROO
5. 250
1.750
3307.
667.400
255.
347.100
363.
180.400
28.900
ie6.600
82.
9.
2.500
4258.
1110.
696. 300
350.
Temperature,
T,
OF
1600.00
1560. 00
1520.00
"1480.00
?460.00
y440.00
?400.00
a1360. 00
"1340.00
"1320. 00
1320.00
1290.00
&IZSO. 00
a1230. 00
a1170. 00
"1140.00
"1125. 00
112. 00
1200. 00
1170.00
1150.00
1140.00
al l 20.00
(b) Steel C (23b CK)
Temperature,
T,
OF
a1230. 00
a1250. 00
a12eo. 00
1292.00
a1310. 00
a1112. 00
a1120.00
a1202.00
a1202.00
a1210.00
a1220.00
a1150.00
a1170. 00
a1240. 00
1270.00
12bO. 00
1292. 00
1300.00
a1112.00
a1112.00
~1l 12. 00
a1112. 00
a1112. 00
"1112. 00
a1112.00
a1112.00
"1112. 00
a1202.00
a1202.00
a1202.00
Stress,
0,
psi
30 000.000
30 000.000
30 000.000
30 000.000
30 000.000
40 000.000
40 000.000
40 000.000
40 000.000
40 000.000
40 000.000
40 000.000
40 000.000
40 000.000
40 000.000
40 000.000
40 000.000
40 000.000
34 000.000
43 000.000
46 000.000
46 000.000
50 000.000
52 000.000
54 000.000
57 000.000
25 000.000
34 000.000
35 000.000
48 ooo.ooo
.-
Time,
hr
175. 700
103.500
t,
58.100
!1 300.
22.500
667.900
785.400
266.700
127. 900
44.100
74.
40.500
37. 800
17. 200
4.500
1. 200
1.300
.e00
363.100
233.900
261. 400
1P3.100
65.600
39.300
23.300
199.400
124.300
2 274.
84.500
1074.
Temperature,
T,
OF
a1112.00
a1110.00
a1080.00
aioeo. 00
a1050. 00
a1030. 00
a1022.00
a1020.00
1040.00
1022. 00
a1000.00
a9s0.00
a960. 03
a940. 00
a920.00
a1120.00
"1200. 00
al28O. 00
"1340.00
"1500.00
a15G0. 00
"l5nO. 00
"1540. 00
Temperature
T,
OF
a1202. 00
a1202. 00
"1202.00
a1202.00
a1202.00
1202.00
1202.00
1202.00
1202.00
a12.a 00
al 2!~2.00
a1232.00
al2:~2.00
?2.'2.00
a12.'2.00
11292. 00
"1292.00
1292.00
12!'2.00
1292.00
a1060. 00
a1300. 00
a1360. 00
a1430. 00
ai4eo. 00
a1570. 00
a1630. 00
a1140. 00
a1320. 00
a1480. 00
a1540.00
Stress,
0,
psi
$0 000.000
;o 000.000
50 000.000
$0 000.000
;o 000.000
50 000.000
$0 000.000
j0 000.000
15 000.000
75 000.000
75 000.000
75 000.000
75 000.000
'5 000.000
75 000.000
50 000.000
LO 000.000
50 000.000
?5 000.000
.5 000.000
12 000.000
!O 000.000
.o 000.000
Stress,
0,
psi
36 000.000
12 000.000
14 000.000
65 000.000
4G 000.000
18 000.000
19 000.000
50 000.000
10 000.000
23 000.000
25 000.000
2Q 000.000
29 000.000
32 000.000
33 000.000
34 000.000
36 000.000
37 000.000
38 ooo.ooo
38 ooo. ooo
60 000.000
25 000.000
19 000.000
15 000.000
12 000.000
8 000.000
6 000.000
34 000.000
15 000.000
8 000.000
6 000.000
Ti me,
hr
t,
12.900
34.
52.200
37.400
239.
445.
989.900
817.500
.330
5.850
15.600
46.500
138.
542.
579.600
186.100
130.203
132.700
125.800
51.300
41.700
32.400
148. 200
-
~-
Time,
t,
hr
68.601
59.30(
24.401
14.5M
22.901
~
7.
2. R51
2.551
1.47(
t'5!1. 701
1!>4. 601
75.
34. GO1
31.
13. 30
19. ROI
10. 401
2.751
7.60'
1. 65'
42.50
89.60
35.
71.40
147.90
104.
140. 90
1077.
1505.
2237.
1258.
-
lata poi nt i n parametri c anal ysi s.
31
TABLE V. - Concluded. NASA FUPITRE DATA
(c) Steel P (14 PA)
Cemperature
3
932.00
&932.00
932.00
932.00
932.00
932.00
932.00
932.00
932.00
1022.00
1022.00
1022.00
1022.00
1022.00
1022.00
1022.00
1022.00
9022.00
a1022.00
1415.00
1340.00
1315.00
129o:oo
Stress,
psi
65 000.000
60.000.000
60 000.000
51 500.000
55 000.000
52 500.000
40 000.000
30 000.000
27 000.000
58 000.000
55 000.000
50 000.000
47 000.000
47 000.030
45 000.000
42 500.000
40 000.000
37 500.000
25 000.000
10 000.000
10 000.000
10 000.000
LO 000.000
0,
-. -. ~-
Time,
t,
hr
-
3.800
14.150
14.400
10.
18.900
51.
623.
7 592.
11410.
.580
.717
1.280
2.450
6.200
3.500
6.300
22.500
12.
382.200
. 170
1.500
3.700
6.100
Temperature
T,
OF
1250.00
1220.00
U80. 00
1140.00
1090.00
1100.00
1080.00
1060.00
1050.00
1040.00
p o . 00
lO20.00
9010.00
1000.00
990.00
980.00
960.00
940. 00
a930. 00
%oo. 00
932.00
as7.00
a860. 00
Stress,
=,
psi
10 000.00c
10 0oo.ooC
10 00o.ooc
10 000.00c
10 000.00c
10 000.000
40 000.000
40 000.000
40 000.000
40 000.000
40 000.000
40 000.000
40 000.000
40 000.000
40 000.000
40 000.000
40 000.000
40 000.000
40 000.000
40 000.000
70 000.000
70 000.000
70 000.000
-
Time,
hr
19.20c
42.
167.
203.4oC
608.
2639.
t,
1.30C
2.20c
4.30C
6.800
7.400
22.500
20.100
63.300
51.200
80.603
192.100
427.900
623.
372.
1.400
5.800
31.200
~. -
Temperature,
T,
OF
740.00
&785. 00
880. 00
a932. 00
~20. 00
1022. 00
1050.00
1090.00
1090.00
1120.00
1160. 00
&1230.00
1290.00
740.00
780.00
a830. 00
880. 00
932.00
a980. 00
1000.00
1030.00
1070.00
1150. 00
1220.00
-.
-. .~
Stress,
0,
psi
90 000.00(
70 ooO.OO(
60 000.00(
50 000.00C
30 000. OOC
25 0OO.OOC
20 0oo.ooc
20 000.00c
16 000.OOC
13 0OO.OOC
8 000. ooc
5 000.00c
30 000.000
70 000. OOC
50 000.000
50.000.000
35 000.000
30 000.000
35 000.000
!O 000.000
L6 000.000
8 000.000
5 000.000
ao ooo.oo(
Time,
t,
hi
57.100
195.800
120.
103.500
186.700
123.500
79.500
112.400
183.500
100.300
97.900
139.700
996.600
84.
1122.
948.800
599.
1902.
754.800
970.700
804.800
948.500
960.
1084.
Data point used i n parametric analysis.
32
TABU VI . - GERMAN RUPTURF: DATA
remperature, Stress, I Time,
-
T
4
1022.00
1022.00
1022.00
1112.00
1112.00
1112.00
1112.00
1112.00
1112.00
1112.00
1112.00
1112.00
1112.00
1112.00
1112.00
1112.00
1112.00
1202.00
1202.00
1202.00
1202.00
1202.00
1202.00
1202.00
1202.00
1292.00
1292.00
1292.00
t,
psi I hr
Steel K (27b KK)
76 899.999 0.100
66 899.999
72 500.000
64 000.000
55 500.000
Steel C (23b CK)
~
14 200.000
17 800.000
28 400.000
28 400.000
28 400.000
35 600.000
44 100.000
51 200.000
59 800.000
11400.000
14 200.000
17 800.000
22 800.000
28 400.000
35 600.000
42 700.000
52 600.000
l l 400.000
11400.000
13 900.000
60 000.
30 000.
3 500.
3 000.
2 200.
1 200.
520.
150.
82 790.
15 000.
6 500.
1 800.
550.
124.
5.
30 000.
20 000.
4 500.
. l oo
. 100
Temperature,
1292.00
1292.00
1292.00
1292.00
1292.00
1292.00
-
932.00
932.00
932.00
932.00
932.00
932.00
932.00
932.00
932.00
1022.00
1022.00
1022.00
1022.00
1022.00
1022.00
1022.00
1022.00
1022.00
1022.00
1022.00
1022.00
1022.00
Stress,
0,
psi
Time,
t,
hr
ateel C (23b CK)
17 21 800 4 0 0 : O O O ~ 000 1 100.
21 400.000
27 000.000 180.
27 000.000 140.
47 000.000 . l oo
;eel P ( 14a PA)
84 000.000
75 500.000
78 399.999
55 500.000
44 100.000
34 200.000
27 000.000
22 800.000
17 100.000
72 599.999
69 699.999
65 500.000
59 800.000
35 600.000
27 000.000
22 800.000
22 800.000
17 100.000
13 900.000
13 900.000
11100.000
8 830.000
~-
0. 100
. l oo
2.
150.
1 700.
2 600.
16 000.
22 000.
100 000.
. l oo
. l oo
1.200
1.500
150.
300.
400.
900.
2 100.
6 500.
8 000.
10 000.
68 000.
33
i
I i
I lemperature;
~
T,
OF
2005
1900
1790
2175
2400
2300
2200
2100
2100
2000
2000
2000
25 75
2200
2800
2620
2200
2 900
3000
2450
TABU3 VII. - CRFEP DATA FOR COLUMBIUM ALLOY FS-85
~
Stress,
0,
psi
25 000
25 000
25 000
10 000
10 000
10 000
10 000
10 000
10 000
10 000
8 500
6 000
6 000
4 000
4 000
4 000
3 000
2 000
2 000
~
18 ooo
1-Percent creex
0.6
26.
210.
4.9
3.4
25.4
54.
355.
380
775.
900.
2480.
425.
5.6
3.4
14.4
2. 6
4. 6
1140.
------
Time, t, hr
2-Percent creep
3.0
33.
257.
7. a
5.7
41.
84.
500.
570.
1325.
1420.
10.
710.
------
6. 4
5-Percent creep
6.1
45.
332.
13.
10. 8
68.
133.
765.
875.
2175.
------
22.2
13.5
1370.
56.
13. 8
33. 2
------
950.
34
NASA-Langley, 1965 E- 2781
The aeronautical and space activities of the United States shall be
conducted so as t o contribute . . . t o the expansion of human knowl-
edge of phenomena in the atmosphere and space. The Administration
shall provide for the widest practicable and appropriate dissemination
of information concerning its activities and the results thereof.
-NATIONAL AERONAUTICS AND SPACE ACT OF 1958
NASA SCIENTIFIC AND TECHNICAL PUBLICATIONS
TECHNICAL REPORTS:
important, complete, and a lasting contribution to existing knowledge.
TECHNICAL NOTES:
of importance as a contribution to existing knowledge.
TECHNICAL MEMORANDUMS: Information receiving limited distri-
bution because of preliminary data, security classification, or other reasons.
CONTRACTOR REPORTS: Technical information generated in con-
nection with a NASA contract or grant and released under NASA auspices.
TECHNICAL TRANSLATION Information published in a foreign
Scientific and technical information considered
Information less broad in scope but nevertheless
language considered to merit NA % A distribution in English.
TECHNICAL REPRINTS: Information derived from NASA activities
and initially published in the form of journal articles.
SPECIAL PUBLICATIONS: Information derived from or of value to
NASA activities but not necessarily reporting the results .of individual
NASA-programmed scientific efforts. Publications include conference
proceedings, monographs, data compilations, handbooks, sourcebooks,
and special bibliographies.
Det ai l s on t he avai l abi l i t y o f these publ i cati ons may be obt ai ned from:
SCIENTIFIC AND TECHNICAL INFORMATION DIVISION
N AT I 0 N A L A E R 0 N A U T I CS A N D SPACE A DM I N I STRATI 0 N
Washington, D.C. PO546