ML20202E466
| ML20202E466 | |
| Person / Time | |
|---|---|
| Issue date: | 01/29/1999 |
| From: | Birmingham J NRC (Affiliation Not Assigned) |
| To: | Mcphatter F FRAMATOME |
| References | |
| PROJECT-693 NUDOCS 9902030031 | |
| Download: ML20202E466 (6) | |
Text
____
January 29, 1999 Mr. Frank McPhatter, Manager Framatome Cogema Fuels 3315 Old Forest Road P.O. Box 10935 Lynchburg, VA 24506-0935
SUBJECT:
REQUEST FOR ADDITIONAL INFORMATION FOR REPORT BAW-10227P,
" EVALUATION OF ADVANCED CLADDING AND STRUCTURAL MATERIAL (MS)
IN PWR REACTOR FUEL,"
Dear Mr. McPhatter:
By letter dated September 30,1997, Framatome Cogema Fuels (Framatome) submitted Report BAW-10227P," Evaluation of Advanced Cladding and Structural Material (M5)in PWR Reactor Fuel," for NRC review. On October 26,1998 the staff sent a request for additional information to i Framatome. Since then, the staff and its contractor have continued their review and have '
additional questions regarding Appendices C,D,E, and G of the report. The information being requested was discussed with you in a phone call on January 28,1999 and the questions are l
enclosed. To assist us in completing our review, please provide your response by the end of i February 1999. !
If you have any questions regarding this matter please contact me by phone 301/415-2829 or email jlb4@nrc. gov.
Sincerely, Original Signed By:
Joseph L. Birmingham, Project Mgr.
Generic issues and Environmental Projects Branch Office of Nuclear Reactor Regulation
Enclosure:
As stated Project No. 653 cc w/ encl: Mr. M. Shoppman, Manager ff Rockville Licensing Operations /
Framatome Technologies, Inc.
1700 Rockville Pike, Suite 525 Rockville, MD 20852-1631 Mr O b DISTRIBUTION:
Hardcopy Email Docket File ' PUBLIC JBirmingham PGEB r/f SRXB r/f FAkstulewicz j TEssig OGC SLWu ACRS JLyons DOCUMENT NAME: G:\JLB\BAW10227.WPD OFFICE PGEBv/ b PGEB c/
NAME JBirmingham:sw FAkstulevvicz l DATE 01/r: /99 ' / W /99 OFFICIAL RECORD COPY THL/
b g }, ),
?
(10!'#p'qC 1
9902030031 990129 PDR C
TOPRP EMVBW PDR hh
^
p s* "%q _
f' UNITED STATES
~'$
j ,y
'I NUCLEAR REGULATORY COMMISSION WASHINGTON, D.C. 20555 4 001 O
i 49
SUBJECT:
REQUEST FOR ADDITIONAL INFORMATION FOR REPORT BAW-10227P,
- EVALUATION OF ADVANCED CLADDING AND STRUCTURAL MATERIAL (MS)
IN PWR REACTOR FUEL,"
Dear Mr. McPhatter:
By letter dated September 30,1997, Framatome Cogema Fuels (Framatome) submitted Report BAW-10227P," Evaluation of Advanced Cladding and Structural Material (MS)in PWR Reactor Fuel," for NRC review. On October 26,1998 the staff sent a request for additionalinformation to Framatome. Since then, the staff and its contractor have continued their review and have additional questions regarding Appendices C,D,E, and G of the report. The information being requested was discussed with you in a phone call on January 28,1999 and the questions are enclosed. To assist us in completing our review, please provide your respon'>e by the end of February 1999.
If you have any questions regarding this matter please contact me by phone 301/415-2829 or email jlb4@nrc. gov.
Sincerely, l ? bt m. g k/
Joseph L. Birmingham, Project Mgr.
Generic issues and Environmental Projects Branch Office of Nuclear Reactor Regulation l
Enclosure:
As stated Project No. 693 cc w/ encl: Mr. M. Shoppman, Manager f Rockville Licensing Operations Framatome Technologies, Inc.
1700 Rockville Pike, Suite 525 Rockville, MD 20852-1631 l
l
l l'
i Questions for Appendices C, D, E, and G of BAW-10227 General Comments on Annendix C The premise for the development of the Framatome Technologies Incorporated (FTI) high-temperature swelling and rupture model for M5 cladding is that it should provide reasonable predictions of the Chapman bundle tests because these bundles were used in the development of the NUREG-0630 model. However, as noted in NUREG-0630, the NUREG-0630 model overpredicted
- the flow blockage from the Chapman bundles because it was concluded that these tests were ;
l somewhat non-conservative compared to in-reactor bundle conditions. The reason for the non-conservatism of the Chapman tests was believed to be due to the extra heat loss through the shrouds around these test bundles. The degree of flow blockage for the NUREG-0630 model was not only I based on the Chapman tests, but was also based on the more conservative results from the Treat l FRF-1 test bundle and the JAERI bundle tests that resulted in the greatest flow blockage. The NUREG-0630 predictions of flow blockage provided better predictions of these test bundles than the overprediction of the Chapman test data.
l The subject topical report suggests that the FTI model is consistent with the NUREG-0630 model;, l however, as noted in Question 16 below, the FTl model predicts less flow blockage than the NUREG-0630 model for standard PWR bundles in the a phase accounting for the difference in a i phase transformation temperatures between M5 and Zircaloy-4 (Zr-4).
It is also noted that no bundle flow blockage data is provided for the M5 cladding. This is acceptable because as noted in the subject topical report (page C-6), the n ain difference between the behavior of M5 versus Zr-4 for LOCA analyses is that the a to a+p phase transformation temperature is lowered by 70 to 100 C. This results in a shin in the rupture strain versus rupture temperature curves for M5 and Zr-4 in the a and a+ regions by a similar temperature shin for both slow and fast temperature ramp rates. However, the overall magnitude and shape of the rupture data for M5 cladding remains well within the uncertainty of those data for Zr-4. Therefore, the behavior of the M5 cladding should be similar to Zr-4 cladding at high temperatures (including flow blockage) as long as the shin in phase temperature is taken into account. This conclusion has previously been observed based on proprietary data at high teinperatures from other cladding types similar to M5 available to the NRC.
Ouestions
- 16. In order to compare the flow blockages of the FTI and NUREG-0630 models for licensing applications of full size bundles for fast and slow temperature ramp rates, Figures C-34 and C-l 35 from the subject topical report were compared to equivalent flow blockage curves in Figure 16 of NUREG-0630 taking into account the shiR in phase transformation temperatures for M5 j and Zr-4. This comparison demonstrated that the FTI model provides an overprediction of flow blockage in the p phase but an underprediction in the a phase of 5% (relative) for slow ramp rates and underprediction in the a phase of 18% (relative) for fast ramp rates relative to those 1-l Enclosure
e l
l for NUREG-0630. Please explain why this underprediction of flow blockage by the FTI model I in the a phase in relation to NUREG-0630 is acceptable.
- 17. How was the magnitude of the PDF function (Figure C-16 in topical report) determined? Were the Chapman test bundle data used?
- 18. The FTI high temperature swelling and rupture model are considerably different from the NUREG-0630 model. One of the differences is the FTI pre-strain sub-model. There are some assumptions used in the development of the pre-strain sub-model that appear to be inconsistent with the data or with the general behavior of M5 and Zr-4. Some of the inconsistencies are:
a) It is assumed that pre-strain is a function of rupture strain and a significant contributor to flow blockage. However, examination of the Chapmam bundle pre-strain data show that the pre-strain data on average for each bundle do not appear to change much between those bundles with lower rupture strains (bundles B-1 and B-2 with less flow blockage) and the bundle with greater rupture strains (bundle B-3 with greater flow blockage) in the a phase.
This would suggest that: 1) the amount of pre-strain is not a function of rupture strain as assumed by the FTI model, and 2) if there is an effect of pre-strain on flow blockage, it is rather constant for rupture strains greater than 40% and that significant flow blockage is primarily a function of rupture strain in the a phase.
b) The pre-strain values from the MS Edgar tests are significantly lower than those obsen ed from the Chapman bundle tests. It would be expected that M5 and Zr-4 should give similar i I
pre-rupture strains because both have similar rupture strain behavior at the high temperatures typical of a LOCA, as noted above in the general comment. This would suggest that there may be some problems with the single rod Edgar tests, e.g., axial temperature gradients, that make the pre-strains non-prototypical of those in a bundle during a LOCA where cladding temperatures are more uniform. This would further suggest that the FTI pre-strain model significantly underpredicts pre-strains in the a phase resulting in an under prediction of flow blockage in th; a phase for PWR bundles during a LOCA.
c) The pre-strain FTI modeling in Figures C-8 and C-9 for slow and fast ramps, respectively, assume the same pre-strain values in the a and a+p regions. Examination of the Edgar slow ramp data in Figure C-8 demonstrates that the a phase has higher pre-strains than the a+p phase on average. This would be expected because it is known that the latter has less strain l
capability than the former, and this is further demonstrated in the rupture strain data. This would suggest that the FTl pre-strain model underpredicts pre-strains of the Edgar tests in the a phase.
a Please comment on the above inconsistencies in the FTl modeling assumptions of M5 pre-strains and what is the impact on the calculation of flow blockage in M5 bundles (Mark B and BW designs) during a LOCA, if the assumptions regarding the FTI pre-strain model are incorrect.
l l
l 19. The shape of the FTI pre-stram curve m the p phase (between 1000 to 1200 C) does not appear l to bejustified based on the data presented (3 data points for slow ramp rates and 2 data points l for fast ramp rates). In addition, the p phase transformation is complete by 1000 C, which l would suggest that perhaps the peak of the pre-strain should be at 1000 C. There is also one !
pre-strain data point near 1000 C for slow ramp rates that would suggest that the peak is near
, this temperature, but that is ignored (sig-ificantly underpredicted) by the FTI pre-strain curve. ;
! Are there additional pre-strain data to substantiate the FTI pre-strain curves?
If there are no additional data that are applicable, there are two alternatives: 1) Ignore the single rod pre-strain data altogether and adopt the NUREG-0630 methodology for determining flow ,
l blockage (based on rupture strains only -- thejustification for this modeling change would be '
l the observation in item 18-b above), or 2) assume that the pre-strain begins to peak near 1000 C and still fits the few data points that exist in the p phase. Please discuss the impact of tne above two attematives to modeling flow blockage in the p phase on M5 bundles (Mark B and BW designs) for LOCA analyses.
. 20. Has the composition or specifications for the fabrication of the M5 cladding changed from that l used to develop the data in the subject topical report or that u;ed in LTA irradiations? If so, please provide the differences and identify the data impacted.
- 21. Please provide new M5 test and LTA data that have become available since the publication of the subject topical report.
- 22. Please provide cycle lengths in full power days for each of the plants listed in Table E-1. It appears that the LTA data taken to date are based on power operation that does not appear to be particularly aggressive, e.g., six cycles to achieve a burnup of 63 GWd/MTU and data from LTAs with more aggassive operation will not be obtained until calendar years 2000 to 2001.
l Is this observation correct?
- 23. Please provide the standard deviation and also the maximum and minimum values of the l azimuthal average thicknesses quoted on page G-6 for the oxide, oc, and p phase thicknesses from M5 and Zr-4 samples. Also, please discuss the differences in the alpha-incursion behavior l between Zr-4 and M5 samples at 1200 to 1300' C. Were equivalent cladding reacted (ECR) measurements performed on the other failed and un-fai':d specimens from the table on page G-5, other than those provided for the un-failed speciniens with maximum oxidation? If so, l please provide these values and how they were determined.
- 24. The peak local oxidation values provided in Table F-3 (LOCA calculation at 40 GWd/MTU bumup) do not appear to include the cladding oxidation from normal operation. If so, please l justify why this initial oxidation from normal operation is not included in the total amount of j t
oxidation for LOCA to assess whether the 17% oxidation limit is exceeded. Also, for this same calculation (Table F-3), please provide the burnup level at which the gap closed for the Zr-4 j clad fuel rods and for the M5 clad fuel rods. i c l l
. t l
- 25. Please provide arrhenius plots of the Zr-4 and M5 high temperature oxidation data. Please provide a discussion of the uncertainties and potential biases in the optical pyrometer temperature measurement Appendices D and G. Were independent temperature measurements performed on oxidized M5 material to confirm uncertainties and lack of bias in the optical pyrometer measurement?
r P
i t
l I
l I
i
_