ML20009C146
| ML20009C146 | |
| Person / Time | |
|---|---|
| Site: | Shoreham File:Long Island Lighting Company icon.png |
| Issue date: | 05/12/1981 |
| From: | GENERAL ELECTRIC CO. |
| To: | |
| Shared Package | |
| ML20009C138 | List: |
| References | |
| NUDOCS 8107200323 | |
| Download: ML20009C146 (26) | |
Text
_..
1 1
i i
i GENERAL ELECTRIC FUEL SWELL AND RUPTURE MODEL EXPERIMENTAL DATA REVIEW
~
j AND SENSITIVITY STUDIES i
4 1
i l
May 12, 1983 8107200323 810713 PDR ADOCK 05000322 Q
, _ _ _. - - -. ~. _. _ _.. _ _.,
TABLE OF CONTENTS
1.0 INTRODUCTION
2.0 EXPERIMENTAL DATA 2.1 Ciadding Hoop Stress versus Perforation Temperature 2.2 Circumferential Strain versus Temperature 3.0 SENSITIVITY STUDIES 3.1 Overall Model Sensitivity 3.2 Individual Model Component Sensitivity Studies 3.2.1 Variation of Cladding Strain at Perforation 3.2.2 Variation cf Perforation Stress versus Temperature Curve 3.2.3 Variation of Swelling Initiation Criteria 4.0
SUMMARY
l
[
5.0 CONCLUSION
S i
DKD:csc/170A1
- -.. - ~ _ _. _ - _. -. -.
1.0 INTRODUCTION
This report was prepared in response to the NRC request (Reference 1) to (a) provide supplemental calculations using the NUREG-0630 model and, (b) to revise the cladding models of both CHASTE-05 and CHASTE-06 to confort.: to recent experimental data.
It provides 1) a discussion of the experimental data used to develop the General Electric &nd NUREG-063C cladding swelling and rupture model, 2) the results of sensitivity studies performed using the General Electric heatup model (CHASTE) which show the impact of the NUREG-0630 model on calculated peak cladding temperatures.
It is shown that the NUREG-0630 perforation strain versus temperature curve is not applicable to the BWR due to non prototypicality of the experimental conditions used to generate the curve.
Even so, substitutior.
of a bounding NUREG-0630 curve into the current GE ECCS analysis produces only a negligible effect on the peak clad temperature (PCT).
Therefc e, it is General Electric's position that the current strain model is valid for the BWR and should continue to be used for all ECCS analyses.
This report shows that the GE hoop stress vs. rupture temperature cur e v
is more valid than the corresponding NUREG-0630 curve at temperatures above 1600 F and that the NUREG-0630 curve is morc representative of existing data at temperatures below 1600*F.
A sensitivity study presented using a combination of the two curves (adjusted curve) resulted in a PCT impact of < 10 F.
Even though this PCT impact is small, GE proposes to revise the current model to incorporate the adjusted curve and implement the change at the same time the complete LOCA model improvement package is implemented.
csc:ggo/170C -
2.0 EXPERIEMENTAL DATA 2.1 Cladding Hoop Stress Versus Perforation Temperature The NRC staff has expressed a concern (Reference 2) that the Geners' Electric Hoop Stress Versus Perforation Temperature curve is non-conser sti,e for temperatures above s1000*C (1832 F).
The staff is using as a basis for this concern the data and curves contained in NUREG-063C (Referen:e 3) and are requesting that supplemental calculations be performed with the most conservative curve (O'F/ set) frca NUREG-0630.
General Electric does not believe that this concern is justified and in this section provides the basis for the position that the GE curve should be used for any analysis of GE BWR fuel.
Figure i shows the General Electric perforation curve together witr tre corresponding experimental data base for cladding heatup rates of 11C f /se:.
(10*F/sec is considered a conservative upper bound heatup rate for GE BW:!'s. ) The figure shows that the GE curve is a good representatier. c' the data.
Figure 2 shows the NUREG-0630 correlation for 0 F/se: anc tre GE curve together with NUREG-0630 data for heatup rates of less than 10 F/se:.
l The following points are apparent from Figures 1 and 2:
l (a) The NUREG-0630 curve contains no data for hoop stresses belo.
$3500 psi and temperatures above s 1600 F.
I (b) In the temperature range of concern to the staff (>1832 F), a considerable amount of data typical to the BWR exists to support the GE curve.
Furthermore, General Electric has examined all data in NUREG-0630 (irrespe:-
tive of heatup rate) with perforation temperatures above 1832*F (1000 C).
These data have been plotted in Figure 3.
This figure indicates that the GE curve is conservative with respect to these data.
csc:ggo/170C ;
l In addition, the recent data generated by KfK (Reference 5) have been examined with respect to the General Electric model. These data, which are shown in Figure 4 indicates that the GE model is conservative witn respect to the data for temperatures above s1600 F.
Figures 2 and 4, however, indicate that the GE design curve may be non-censervative wit-respect to the data for temperatures below s1600 F.
General Electric agrees with the NUREG-0630 data presented in this temperature range. However, the overall effect of revising the GE cune in this range is <10 F on the PCT.
Figure 5 shows the data fror Figures 2 and 4 for temperatures below s1600 F together with the GE data base and adjustedcurve.
Figure 6 shows the adjusted GE curve together with data generated by A'..
(Reference 4) which is not included in NUREG-0630.
These data were taken under conditions not prototypical of a BWR (direct heating, u "c r temperature profile, etc.) which are known to produce larger values cf circumferential strain, from a hoop stress versus temperature standp;'.t.
The adjusted curve is an accurate representation of these data alsc.
To summarize, the NUREG-0630 0 F/sec curve is not applicable to the EWR above s1600 F, whereas the General Electric curve is well qualified for l
temperatures in this range.
The General Electric correlation is howe.er, non-conservative with respect to the NUREG-0630 data below 1600 F.
l General Electric believes that the adjusted GE hoop stress versus perfora-f tion temperature curve (Figure 5) is an accurate representation of BWR fuel behavior for LOCA analysis and that the replacement of this curve with the corresponding 0*F/sec NUREG-0630 is not appropriate.
csc:ggo/170C 2.2 Circumferential Strain Versus Temperature The NRC staff it also requesting (Reference 2) that supplemental calcula-tions be performed with a combination of the slow and fast heat up rate circumferential strain versus temperature curves from NUREG-0630. TMs combination of the slow and fast ramps (shown in Figure 7) is biaset t; produce the maximum strain at any given temperature.
General Electric does not believe that the NUREG-0630 strain-temperatu e curves are sufficiently qualified to warrant application to GE.BWR fuel.
Furthermore, the combination of the slow and fast ramp curves represe. s.
a further departure from applicability to BWR fuel. The following paragraphs provide the technical basis for this position.
There is convincing evidence available (including the cata containe: i r.
NUREG-0630) that heat-up rate has a pronounced effect on the value cf cladding circumferential strain at perforation.
Hence the develo pe-t in NUREG-0630 of two separate correlations for strain versus perfora:ic-temperature, one for fast heat-up rates and one for slow heat-up rate:.
is considered appropriate although simplified. The application of data obtained under fast heat-up rates to the BWR (which has a maxim s heatc; rate of <10 F/sec) is therefore not considered technically justified.
Figure 8 shows the General Electric perforation strain-temperature model together with the founding data base for heatup rates of 1 10 F/second.
This figure shows that the GE model conservatively bounds greater than 90% of the data.
Figure 9 shows the GE model compared to 1he slo.
heatup correlation from NUREG-0630 together with the NUREG-0630 data for heatup rates of 110 F/sec.
It is apparent from this figure that the NUREG-0630 correlation is unqualified for temperatures above sI600 f.
Furthermore, General Electric also questions the criteria used to select the data from which the curve has been derived.
Ir: NUREG-0630, it is stated that most of the data falling below this curve were discounted;as they were from tests with features known to reduce perforation strainq.
csc:ggo/170C
i.e., non-uniform temperature profiles, corrosion fission products anc cold shrouds. However. all these features would be present in a BWE during a LOCA.
For example, a large number of BWR rods can freely radiate to a cold surface (fuel channel or water rods) during a '.0CA thus establishing significant temperature variations.
Furthermore, significant axial temperature variations, also known to reduce cladding strain (Reference 7), occur due to the stochastic stacking and tilting of the fuel pellets within the fuel column.
Therefore, as the majcrity of the data used must have been obtained under conditions (heated shroud, uniform temperature) which are not prototypical of the BWR, the applicability of any correlation derived from this data is questionable.
In addition to the NUREG-0630 data, the staff recently supplied GE wit-additional circumferential strain data (Reference 6).
This data is shown in Figure 10.
In this figure, the ORNL data with heated shroc::
displays considerably higher circumferential strain than the data taisr with the cold shroud; i.e., the 5 C/sec hot shroud data at s780 C lies between s95 and 110% strain while the cold shroud data is less than s30%.
It is apparent therefore that to obtain meaningful results, eich are applicable to the BWR, care must be taken to establish test conci;icrs that ensure prototypicality.
Note that the GE perforation strain tempe-ature model is derived from simulated LOCA tests on full scale BWR fuel bundles, thereby imposing the prototypicality criteria. The GE strain temperatcre model is therefore considered more appropriate for the analysis of GE BWR fuel than the NUREG-0630 curves.
To summarize, the General Electric perforation strain curve conservatively bounds the circumferential strain data for slow heat-up rates at temperatures above 925 C (1700'F).
In the alpha phase region, the strain data has been shown to be extremely sensitive to test conditions with the majority of the data contained in NUREG-0630 considered inapplicable to the BWR.
General Electric believes that the GE perforation strain temperature curve is applicable to GE BWR fuel, due to the prototypicality of its founding data base and should be used in GE BWR LOCA analyses for the prediction of perforation strain.
cse:ggo/170C - _,
?.0 SENSITIVITY STUDIES A number of sensitivity studies have been performed to eval; ate the effect of the NUREG-0630 cladding swelling and rupture modtl on CHA5TE heatup calculations.
They show that the peak cladding temperatures (PCTs) calculated with CHASTE have a small sensitivity to various parameters of the cladding swelling and rupture model.
The studies were performed for plants with 7x7 and prepressurized 8x8 -
2 water rod fuel at high exposures to maximize the number of perforations and hence any sensitivity of the calculated PCT.
The plants selectec had long reflooding times and short blowdown periods-This then results in a longer period over which the rods are calculated to be perforated and hence a greater sensitivity to changes in the swelling and rupture model.
Calculations were also performed for plants with shorter reflooding times which are typical of most BWF.5.
Overall model sensitivity studies were performed as follows:
l l
a)
Perforation stress curve for 0 C/second heatup rate belo.
I s1600*F and GE curve above 1600 F (adjusted curve fror Se: tion 2.1);
1 l
b)
Peak strain of 80% below a stress of 1500 psi, peak strain of 90% above 1500 psi; c)
Swelling initiation criteria and strain rate from GE model described in Section I.B.2.5 of Reference B.
l The bounding strain assumption (item b) was made because the CHASTE code does not accept a temperature dependent rupture strain curve.
The GE perforation stress curve was used in the f gh temperature range (T > 1600'f) because it provides a better fit to the available perforation stress data than the NUREG-0630 curve in the high temperature region (see section 2.0 for a detailed discussion on experimental data).
csc:ggo/170C..
Additional calculations were performed with the NUREG-0630 perforation stress curve for O'C/second heatup rate over the entire temperature range.
This curve bounds the 1 to 3'C/second heatup rate typical of BWR fuel during the time when perforations occur.
In addition to the overall model comparisons, sensitivity studies were also perfo.med on individual components of the model.
These studies a e discussed in section 3.2 and include:
1.
Variation of cladding strain at perforation; 2.
Variation of perforation stress versus temperature cur e; 3.
Variation of swelling initiation criteria.
3.1 Overall Model Sensitivity The purpose of this study was to determine the peak cladding tenperature sensitivity to the NUREG-0630 swelling and rupture model. The NUREG-0632 model was approximated in CHASTE using the adjusted stress curve and bounding strain values which were discussed earlier. The GE swelling initiation criteria (swelling begins 200 F before perforation temperature is reached) was used for all cases.
The PCT sensitivity for 7x7 fuel was found to be 5 F for all cases in which perforations occurred.
For 8x8 fuel the PCT sensitivity depends on how fast refloodi'ng of the high power elevation occurs following a LOCA.
If reflooding occurs in s220 seconds or less n: perforations are calculated to occur and the PCT impact is 0 F.
For BWRs which take more than 220 seconds to reflood the PCT impact was found to be small ( 5 F) i for prepressurized 8x8 - 2 water rod fuel.
For 8x8 fuel designs other then prepressurized 8x8 - 2 water rod fuel, no perforations are calculated 1
to occur.
To determine the maximum possible impact, bounding calculations were l
performed using the NUREG-0630 0 C/second heatup rate stress curve for all temperatures.
In most cases the results obtained were the same as esc:ggo/170C described above.
For prepressurized 8x8 - 2 water rod fuel at early exposures tP REG-0630 model can result in a PCT increase of 10 to j
50 F if it re.ults in a large number of perforations at high tempera-tures. However, this result is not con idered meaningful as it is due i
to the
..alously low perforation stress at high temperature predicts:
by the NUREG-0630 model that is not supported by the available expe-i-mental data.
3.2 Individual Model Component Sensitivity Studies 3.2.1 Variation of Cladding Strain at Perforation The purpose of this study was to determine the effect of cladding strain at perforation on calculated PCTs.
The GE and bounding NUREG-0630 rupture strain curves (Figure 11) were used for the comparison.
Tne results show a small (0 to 5 F) PCT decrease with bounding NUREG-053:
strains. This is because even though individual rod temperatures are affected (by as much as 20 F fust after a rod perforates during the transient), the temperature of all the rods in the bundle tends to equalize as a result of redistribution of energy by radiation heat i
transfer.
Consequently, the overall effect on PCT is small. The stud es
[
show that as the strain is increased on an individual rod its temperature l
l decreases, because for larger strains there is a larger area for heat l
transfer.
For smaller strains the temperatures are higher as the area for heat transfer is smaller.
The conclusion from this study is that the cladding temperature of perforated rods is relatively insensitive (<10 F, 15 seconds after perforation) and the PCT is almost completely insensitive to the perforation strains. Hence, continued use of the General Electric strain values is considered appropriate.
i l
l csc:ggo/170C l
3.2.2 Variation of Perforation Stress Versus Temperature Curve The purpose of this study was to determine the effect of changing the perforation stress versus temperature curve only.
Three different stress curves were used:
- 1) the GE curve (Figure 2), 2) the adjustec stress curve (Figure 5), and 3) the NUREG-0630 curve (Figure 2) for all temperatures.
Most cases analyzed had about the same number of perforations for each perforation stress curve and the calculated PCT change was !10 F.
For 7x7 fuel the < CT sensitivity was smaller (15 F).
Calculations using the NUREG-0630 0 C/second heatup curve for all temperatures yielded a higher PCT by about 50 F when a large nu-te< of perforations w rt calculated to occur late in the transient.
As discussed earlier, this sensitivity result is not considered meaning'al as it results from the use of unsupported values of perforation stress.
Cases like this were limited to the early exposure range for prepres-surized 8xE - 2 water rod fuel only.
i One additional study was performed using the 10 C/second :,eatup rate curve from NUREG-0630.
It resulted in a PCT decrease of up to 2*F over i
use of the O C/second curve.
l 3.2.3 Variation of Swelling Initiation Criteria I
CHASTE calculates plastic swelling on rods for all temperatures above a certain temperature. This temperature is nominally set at 200 F below the perforation temperature.
Calculations were done assuming that plastic swelling starts at 0"F, 200 F, and 400*F below the perforation l
temperature.
The results show that for the case of 0 F, the PCT increased by up to 6 F, and for the 400 F case the PCT change was 15*F relative to the 200 F nominal case. The effect on PCT was small (<?O'F), and the l
effect on individual rod temperatures was also small (<
"F), and hence l
it can be concluded that the use of 200 F is still app.
e.
cse:ggo/170C _
4.0
SUMMARY
This report has presented sensitivity studies and a review of the cata used to support the GE and NUREG-0630 cladding swelling and rupture model. These models differ in two areas (perforation stress vs. teg er-ature and perforation strain vs. temperature) which are discussed separately below.
A review of data to support the perforation stress versus temperature curves shows that both models agree well with experimental data in certain temperature ranges. The NUREG-0630 perforation stress curse more closely matches experiemental data at low temperatures (T <1600 F) while the GE curve agrees better with the data at higher temperatures (T >1800*F).
In the intermediate temperature range both perforaticr.
stress curves are similar and provide a good fit to the experimental data.
Sensitivity studies performed with the GE model and with the GE perforation stress curve adjusted in the low temperature range sho a small peak cladding temperature sensitivity (!10 F).
A review of the cirumferential strain data shows that the General Ele:tric design curve conservatively bounds the data for slow heatup rates athe l
925 C (1700 F).
In the alpha phase region, the strain data has bee-shown to be extremely sensitive to test conditions with the majority of the data contained in NUREG-0630 considered inapplicable to the BWR.
l General Electric believes that the GE perforation strain temperature curve is applicable to GE BWR fuel due to the prototypicality of its founding data base.
In addition, sensitivity studies performed using a strain curve which bounds the NUREG-0630 curve show a small (0 to 5"F) l PCT decrease over the current GE model.
l l
l csc:ggo/170C -
5.0 CONCLUSION
S 1)
The GE hoop stress versus rupture temperature curve is more valid than the NUREG-0630 curve for the BWR at temperatures above 1600 F.
Howeier, the HUREG-0630 curve is more representative of existing data for temperatures less than 1600 F.
Sensitivity studies performed using a combination of the two curves (adjusted curve) resulted in a PCT impact of 5 10 F.
Even though this PCT impact is small, GE proposes to revise the current model to incorporate the adjusted hoop stress curve.
Implementation of this curve into the ECCS analysis will be coincident with implementation of the complete LOCA model improvement package.
2)
The NUREG-0630 perforation strain curves are not considered applic-able to GE BWR fuel due to non prototypicality of the experimental conditions.
The small PCT sensitivity shown when a bounding NUREG-0630 burst strain vs. temperature correlation is substituted into the ECCS analysis justifies the continued use of the current GE strair.
curve.
l
[
3)
This report satisfies regulatory position 4(a) of Reference 1 requiring supplemental calculations and should be made available for referencing on individual plant FSAR submittals.
l l
4)
Revisions to the cladding models of both CHASTE-05 and CHASTE-06 (regulatory position 4(b) of Reference 1) are not required, although a revision to the GE burst curve will be made as identified in (1) above.
l
\\
csc:ggo/170C.
REFERENCES 1.
Letter, R. L. Tedesco (NRC) to G. G. Sherwood (GE), "Acceptante fcr Referencing of Topical Report NEDE-20566P, NED0-20566-1 Revision 1 and NEDE-20566-4 Revision 4," dated February 4,1981.
/
2.
Telecon:
D. K. Dennison, D. A. Hamon, A. 5. Rao and R. J. Will ars to R. O. Meyer.(NRC) and D. A. Powers (NRC) on Fuel Swelling and' Rupture Issues, 3/19/81.
3.
" Cladding, Swelling and Rupture Models for LOCA Analysis," NUREG-0E?0.
4.
" Deformation Characteristics of Zircaloy Cladding in Var.aur and Steam Under Transient Heating Conditions," NUREG/CR-034', July 1975.
5.
" Burst Criteria of Zircaloy Fuel Claddings in a LOCA," by F. J. Erbacher, et.al. Fifth International Conference on Zirconiur in the Nuclear Industry, Boston, August 1980.
6.
Telecopy of LOCA data from D. Powers (NRC) to D. K. Dennison (GE),
2/10/81.
7.
Letter, T. F. Kassner (ANL) to K. Kniel (NRC), " Review of CPB Report on ECCS Cladding Models," 1/3/80.
8.
General Electric Company Analytical Model for Loss-of-Coolant Analysis in Accordance with 10CFR50 Appendix K, Volume 1, NEDE-20566P, November 1975.
cse:ggo/170C )
40000 30000 hE DM 202543 T A 8 L E 2.7.1 20000
'O *EATupaATE i'Fi c S a8 gd ME AT-UP R ATE l'f leec
[d ME AT Up A ATE 10*F leec
'O MEATveaATE i'si-c 7a7 - E ME AT-U# R AT E l'F iesc
.5 ME AT UP R ATE 10* F /aec 10000 h OMNL Tu 3676 O cEwa n) 4
( oaNL vu ss7s l
5000 4000 l
3000 l
2000
+
1000 5g
^
G i
e l
S O
My 5, O
- 5. -.
s -
O ADD a
a O
~
i 500 l
0
\\
m 1
O a
l c
400 O
300 203 GENERAL ELECTRIC PROPRIETARY INFORMATION 100 1400 1600 1800 2000 2200 2400 2600 22 :.'.
PERFORATION TEllPEP.ATURE ( F)
HOOP STRESS VERSUS PERFORATION TEf1PERATURE FOR HEAT-UP RATES LESS THAN 10 F/SEC JfMwrR J1.
l 40000_
l 30000-20000=
Symbol NUREG-0630 Data Reference a s E
g V
H 10000.
's 3
0 1
8 5
7
- ',g 5 3 j
5000 -
E kQ
\\
4000-Tf,
\\
j 3000 t
5 O
2000.
s E
s E
W
~
O 2000-i 8
=
l s
4 500 A
s s
400.
's l
's '
300 -
s l
' ~,
1 200
-tJUREG-0630 (OC/SEC) 1 100-i i
i i
+
1400 1600 1800 2000 2200 2400 2600 2800 l
PERFORATION TEldPERATijRE (CF) l DATA FR0fi NUREG-0630 FOR HEAT-UP RATES LESS THAN 10"F/SEC FIGURE 2
l 40000 30000 -
20000-Sy=bol hTREG-0630 Data Reference 10000-g x
3 l
6 K
i l
5000<
l 40CD.
3000<
l 2000-x E
c; s
t.
>U l
D *0
!?
1000-u l
E l
a-8 i
x l
500 400-300 200.
100
+
i e
i i
i i
i l
1400 1600 1800 2000 2200 2400 2600 2803 PERFORATION TEMPERATURE (DF)
NUREG-0630 DATA AB0VE 1000 C FIGURE 3
40000 30000 20000-10000.
6E E
D 5000
!S 4000.
'5 3 G 3000-2000-ai b 100>
0 0
Em g
GE E
500.
400-300.
200 l
100-1400 1600 1800 2000 2200 2400 2600 2800 PERFOR/710N TEMPERATURE (OF)
KfK DATA FOR RAMP RATES BETWEEN 0.8 and 1.6 K/SEC FIGURE 4
40000 30000
~
20000 s*
10000 b,
I E
1 5000 E
t 1
4000 3000
_ 2000
_E r
0 O
b[
O A
a a
O 500 0
0 400 300 200 GENERAL ELECTRIC PROPRIETARY INFORMATION 100 i
1400 1600 1800 2000 2200 2400 2600 2800 PERFORATION TEttPERATURE ('F) iH00P STRESS VERSUS PERFORATION TEllPERATURE E IS'- N! -
5I
-I
/
I 40000 30000 I
i 20000 I.
Sy= bel Ref. 3 0
Table 16:.ber l
o A9 k
O A12
~
10000
+
A13 N
5 A16
~
k X
A19 o
9 o+
0 A20
~
q vo 9
M1 l
- b0 5000 N
E x
4000 l
o 3000 C
C O
E 2000
~
u A
x 0
c.
O
[
1000 L
5 5:0 400 300 400 f
100 1
1 8
i i
1400 1600 1800 2000 220D 2400 2600 2800 l
PERFORATION TEllPERATURE ( F)
, ANL DATA FOR HEAT UP RATES LESS THAN 10'F/SEC
i j
i 4
i 120 4
100 1
n j
80
\\
5 5
l A
S0 C
l 5
1 5
w E
40 u
U i
l j
20 i
4 I
l 600 700 110 0 900 1000 1I00 1200 O
a 1
TEMPERATURE ("C)
FIGURE 7:
4 COMBINATION OF NUREG-0630 SLOW AND FAST llEAT-UP RATE CORRELATIONS
l 1
E 4
3 4
A 34
=
5 0
4 E
4 E
=
C 0
3 4
4 i
C t
C
_=
4 o
u e
h 8
3 44 8
t E
o4 a
5 c
u
~
E O O<
C vi
=
O n
wd a
a qw w
i 3
O C E 8
s ge g
m a
e E
2 e
e3 o
=
m 1
m OQ a
=
=
m
=
l cog g g
e l
e e
5 E
m E
m O
E e
h a
z o
5-5#
e
- e o
s a
- =
w
$, s$ i
.=
3 4 i
w
<tiIIyy q D 5..=
a rL 5
2 i d[{ $.$ '%8 '22 E.E f $
0 w+ $ % I E 2
5 x
I!!!i!!
x a
I"(D 0
E i
I E
TY [cg' l
R$$$$33 $E$3 83 e
m
?
14 y3 y3 IY77777 83 5 s *
- s*
~
9AC3D o o +o e a
v W
e e
i e
I 1
9 I
I I
i 5
R 8
g
.e
.o o~
(O N!YW1S lVIIN38331403HI3
,--.--,-,,.r-, - ~.. - -, - - - -, - -,. _
-,.e.-
,----,,.----...~-e...
=
l l
l CC W
U If Tw
&m g
g, k.
b D o
2 w=a
=m a
o w
C m
r=
6 C
e C
l 3
IS C E w
t, vi m
m f
g 6
g s
Cw c.
D W
s M
C C
C
~
g
=
Q
\\
z 6
W l
>=
2*
l C
i C
%C O
C d
E 3
w W
O Q.
E O
h
=
e6 2
l e
E 1
0 9
(
O o
E 5
m o
~
\\.
s O
I e
I a
9 a
i e
i 1
O i
C O
Q O
O O
m g
N C
CU W
v N
m ee (x) NInis wnN383.iWO3HI3
120 Z
Z X
Z i
x x 0
l HOT SilROUD X
~
6 s C/S RAlf RATE /-
g o
X Z
X
+
00
_77 XO 7
rn m--
~
zg #++ zz Z
1
%o h
- o 6
Z Z
z A
xo
_ _G_E_ _ _ _
a i,, g
,_ _ _ _ _ g. _ _ _ _ _ _ _
m ee i
x
_p4 o
o i
o 5
i g
+
i Z
O e0 g
. r.
l GE t
~
_ Oc eg
_ _S, 2
____yE
^
g
~
14 0 y
~
t' V
6 i
l E
h i
=
g A
a
/
7 Z
\\
COLD. SHROUD 5 C/S RAW RATI 2R Z
Z_
PERf0RMATION TEt0ERATURE ("C) l COMi'ARISON Of RECENT f1RBT AND KfK lifATED SIIR000 l
St.0W RAMP STRAIN DATA WITil NilRf G-0630 MODEL fIGilRE 10
LEGEND FOR FIGURE 10 RA*1P RATE SY"?OL LAE.
(K/SEC.)
SHE E Z
KK 1
HEATED f
X KK 2
HEATED f
C LK 10 HEATED f
4 KK 5
HEATED f
A ORi;L 5
HEATED K
ORNL 1
HEATED C
ORiiL 10 HEALED C
ORi;L 10 COLD 6
Oni;L 5
COLD
+
ORNL 10 HEATED
oo l
,......L..
.. N Eg
.. f' s
,e
.s._
m N
i i
U i
0 a
i
=o"
/
I i
o
\\.
w
=
_,n
=
a o..I I
I a
t F
m
_ o,,
i,
. +. _ _ _ _ _
z g
g e,
O I
e i
o t
t m
eq
=
_2 1 r
i e
w a
k-i I
I e
- ?
t _.t I
O I
Q.
g
=
l zr i
- -,l S -
s o2 g
t
-@ H L;
0 w
s i
$[:f a
u
=
l 2
5 i
{
I 5[
_o $
o l
g--
3 a
@ F-h 1
F 1-o, e
i i
i i
i
~
ii 1
i i,
i a
2 I
i, i
u i-i o e i
_o
.3 l
I 7' %
b
.?.
t:
i l
l p
i L.
l i-o g.
t o
OOl C8 b9
$V OE O
l
("4)
NIVd1S l'dnidn'd
.