ML18068A470

Annual Rept of Changes in ECCS Models Per 10CFR50.46.

ML18068A470
Person / Time
Site:	Palisades
Issue date:	05/31/1998
From:	CONSUMERS ENERGY CO. (FORMERLY CONSUMERS POWER CO.)
To:
Shared Package
ML18068A469	List: ML18068A468 ML18068A470
References
NUDOCS 9810260194
Download: ML18068A470 (4)

Text

• -

ATTACHMENT CONSUMERS ENERGY COMPANY PALISADES PLANT DOCKET 50-255 ANNUAL REPORT OF CHANGES IN ECCS MODELS PER 10 CFR 50.46 3 Pages qa 10260 1 PDR A00 R

t: iga&Mss \

PDR

\

i_ ___ --- ----

• - ANNUAL REPORT OF CHANGES IN ECCS MODELS PER 10 CFR 50.46 ABB-CE SBLOCA ECCS EVALUATION

SUMMARY

Asea Brown Boveri-Combustion Engineering (ABB-CE) performed a reanalysis of the Small Break Loss of Coolant Accident (SBLOCA) design basis for Palisades using Palisades plant-specific input data in ABB-CE's NRG-approved 1977 ECCS Evaluation Model for fuel Cycle 13. The resultant peak cladding temperature (PCT) from this analysis was predicted to be 1992°F. This result was reported to the NRC in Reference 1.

In January 1997, pursuant to 10 CFR Part 21, "Reporting of Defects and Noncompliance", ABB-CE informed Consumers Energy that a reportable defect existed in the ABB-CE SBLOCA methodology which may have had applicability to the Palisades Plant (Reference 2). The specific concern involved modeling of the piping line losses between the steam generators and the main steam safety valves (MSSVs) while the MSSVs were open. Omission of the line loss from calculations could adversely affect certain LOCA safety analyses where the MSSVs are activated.

An engineering analysis was completed in May 1997 which determined the actual line.

losses between the steam generators and MSSVs for the Palisades configuration. It .

was determined that sufficient MSSV modeling margin existed in the Palisades SBLOCA analysis and that the defect reported by ABB-CE would have no impact on PCT. Corrective action associated with this issue was reviewed and determined to be acceptable by the NRC during the Engineering and Technical Support Inspection which took place at Palisades in July 1998.

• EXEM PWR LARGE BREAK LOCA MODEL FOR PALISADES The EXEM PWR Large Break LOCA Evaluation Model is used to assess large break loss of coolant accidents (LBLOCA) in PWRs. The Evaluation Model consists of four primary computer codes: RELAP4 to compute the system and hot channel response; RFPAC to compute the containment pressures, reflood rates, and axial shape factors; TOODEE2 to calculate the hot rod heatup; and RODEX2 to determine the rod .

conditions at the start of the transient.

Input Preparation Change A change (error correction) was made in the implementation of the EXEM PWR Large Break LOCA Evaluation Model for gadolinia bearing fuel rods. The computer codes were not changed but the manner in which input is prepared was modified (this is *

• considered to be part of the evaluation model).

1

ANNUAL REPORT OF CHANGES IN ECCS MODELS PER 10 CFR 50.46 The radiation model used in LBLOCA analyses (Reference 3, pages 53-73) was originally designed for U02 fuel rods. It was not specifically intended to use the model to analyze gadolinia rods. In applying this model to gadolinia bearing fuel rods, the sink temperature for radiation from the gadolinia rod to surrounding rods was inappropriately determined by using pin-pin power distributions, view factors, and surrounding rod temperatures representative of a U0 2 fuel rod. Therefore, the model described in Reference 3 needed to be used in a slightly different manner when analyzing a "hot" gadolinia bearing fuel rod.

Prior to this change the PWR LBLOCA analysis guideline did not specifically address the analysis of gadolinia bearing fuel rods separately from U0 2 fuel rods. The gadolinia bearing fuel rod calculations were performed with the objective of obtaining a conservative result, and the individual analyst was allowed freedom to define the details of how to achieve this conservative result. In order to analyze the gadolinia bearing fuel .

rods in a uniform manner, and to assure that the results were consistently conservative, a revised method for implementation of the Large Break LOCA Evaluation Model to a gadolinia bearing fuel rod was developed.

Palisades had previously used a very conservative calculation of the cladding temperature for the gadolinia bearing fuel rod which was one degree Fahrenheit (1°F) higher than the .cladding temperature for the U0 2 fuel rod. The change discussed above caused a reduction in the gadolinia bearing fuel rod cladding temperaure such that the U0 2 fuel rod had the limiting peak cladding temperature. Thus the limiting peak cladding temperature decreased 1°F due to the change. The resultant PCT predicted*

for Cycle 13 is 189.1°F.

RELAP4 Excessive Variability Small changes in the input to RELAP4 (part of the EXEM/PWR LBLOCA evaluation model) can result in large changes in the calculated peak cladding temperature during a LBLOCA. These results are considered excessive given the size of the input changes.

Corrections were made to the current Siemens Power Corporation EXEM/PWR LBLOCA evaluation to reduce the RELAP4 excessive variability.

The corrected analysis for Palisades resulted in a new peak cladding temperature approximately 113 ° F less than the previously reported result. The resultant PCT for Cycle 13 is 1778°F. These results were previously reported to the NRC by Reference 4.

2

ANNUAL REPORT OF CHANGES IN ECCS MODELS PER 10 CFR 50.46

REFERENCES:

1. Letter, TCBordine (Consumers Energy) to Document Control Desk (NRC),

"Annual Report of Changes in ECCS Models per 10 CFR 50.46", dated November 26, 1997.

2. Letter, IC Rickard (ABB-CE) to Document Control Desk (NRC), "Report Pursuant to 10 CFR 21 Regarding the Omission of Main Steam Safety Valve Piping Pressure Loss in Safety Analyses", LD-97.-002, dated January 17, 1997.

3. XN-NF-82-20(P)(A), Revision 1, Supplement 3, "Exxon Nuclear Company Evaluation Model EXEM/PWR ECCS Model Updates: Response to NRC Request for Additional Information," dated January 1990.

4. Letter, NLHaskell (Consumers Energy) to Document Control Desk (NRC),

"Evaluation of 10 CFR Part 21 Report Regarding Impact of RELAP4 Excessive Variation on Palisades Large Break LOCA EC~S Results", dated June 1, 1998.

3