ML20149J266

Safety Evaluation Accepting Changes & Reanalyses in ECCS Evaluation Models & Application of Models for Plant,Unit 2

ML20149J266
Person / Time
Site:	Millstone
Issue date:	07/23/1997
From:	NRC (Affiliation Not Assigned)
To:
Shared Package
ML20149J260	List: ML20149J258 ML20149J266
References
NUDOCS 9707280139
Download: ML20149J266 (4)

Text

--- - - ~ . - - . . . - . - - - - - - - - - . - - . - -.. . - . - - - .

., O Mo u p 4 UNITED STATES 1 NUCLEAR REGULATORY COMMISSION l" wAsHlNGTON, D.C. 2006&4001

%*****/[

SAFETY EVALUATION BY THE OFFICE OF NUCLEAR REACTOR REGULATION j i

EMERGENCY CORE COOLING SYSTEM LOCA ANALYSES l 4

l MILLSTONE NUCLEAR POWER STATION. UNIT 2

}

DOCKET NO. 50-336 .

1.0 INTRODUCTION

[ By letter dated January 29, 1997, Northeast Nuclear Energy Company (NNECO),

the licensee for the Millstone Nuclear Power Station, Unit 2, identified "I i

errors and reported changes to the Emergency Core Cooling System evaluation models and the application of the models-for Millstone, Unit 2, as required by 10 CFR-50.46(a)(3)(ii). The submittal included changes and identified errors in both the large break loss-of-coolant accident (LBLOCA) and small break loss-of-coolant accident (SBLOCA) analyses performed for Millstone, Unit 2, since March 22, 1996.

2.0 BACKGROUND

The Millstone, Unit 2, LOCA analyses uses the Siemens Power Corporation (SPC)

LOCA evaluation models, which include the T00DEE2 computer code. The models were approved by the Nuclear Regulatory Commission (NRC) staff to meet the requirements of 10 CFR 50.46 in a letter dated July 8, 1986, from.

D.M. Crutchfield (NRC) to G. Ward (Exxon Nuclear).

The NRC staff met with SPC, formerly Exxon Nuclear, during August 1995 in regard to the SPC LBLOCA evaluation model. As a result of that meeting, the staff sent a letter to SPC, dated November 13, 1995, which identified problems relating to changes in the T00DEE2 computer code, which, as noted, is part of the evaluation models used by SPC for pressurized water reactors.

By letter dated November 13, 1995, the staff stated that SPC had made c hnges in 1991 to the NRC-approved Fuel Cooling Test Facility (FCTF) reflood heat transfer coefficient correlation used in the T000EE2 code and did not properly assess the significance of the code changes. The following two major changes were made to the FCTF correlation in the T00DEE2 code: (1) nonphysical non-monotonic behavior of the FCTF heat transfer coefficient correlation with respect to the reflood rate was eliminated; and (2) the quench time calculation used in-the model, using data from Cylindrical Core Test Facility (CCTF) experiments, was modified. By letter dated March 13, 1996, the staff requested that-SPC submit to the NRC, for review and approval, all model revisions and corrections in the T00DEE2 code implemented since the staff's approval of the code in July 1986.

On June 2, 1996, SPC submitted Topical Report XN-NF-82-20, "EXEM/PWR Large Break LOCA ECCS T00DEE2 Updates Revision 1, .0.mplement 5," which described 9707280139 970723 #

DR ADOCK 0500 3 6 g

l i

I the updates made to the T000EE2 code between 1986 and 1991. The staff has l completed its review of this report and has concluded that the proposed 1991 j model LBLOCA-ECCS model is not acceptable and the previously approved 1986 i model has an unacceptable error. The staff issued its Safety Evaluation by letter dated November 29, 1996.

! SPC then )roposed new code changes to correct the error in the ?986 model and j to meet tie requirements of 10 CFR 50.46. The corrected model employs linear

. interpolation of reflood heat transfer coefficients used between effective

reflood rates of 1.0 and 1.77 inches /sec to eliminate the nonphysical

! behavior. These changes are an interim step, pending completion of the staff's review of the revised LBLOCA evaluation model submitted in SPC Topical i Report XN-NF-82-20 (P), Revision 1, Supplement 6, "EXEM/PWR Large Break LOCA

{ ECCS Model Updates," December 1996.

l The proposed SPC interim model changes restricts the reflood heat transfer coefficient to increase linearly between the values calculated by the FCTF i

reflood heat transfer correlations for reflood rates of 1.0 in/sec and

1.77 in/sec. The interim model is used for the Millstone Unit 2 reanalysis.

1 By letter dated December 3, 1996, the staff requested that NNEC0 provide i

additional information to verify that Millstone, Unit 2, is in compliance with

' the requirements of 10 CFR 50 46. NNEC0 provided the requested information by letter dated January 29, 1997.

) 3.0 EVALUATION 4

3.1 LBLOCA 4

i NNEC0 provided a LBLCcA reanalysis for Millstone, Unit 2, using the SPC

! interim evaluation model. The interim model uses linear interpolation of heat j transfer coefficients with respect to reflood rate between reflood rates of ,

1.0 and 1.77 in/sec. A break spectrum that ranged from 0.4 to 1.0 for a double-ended cold leg guillotine (DECLG) break was analyzed for Millstone, Unit 2. The limiting break was found to be a 0.6 DECLG break. 'The limiting break calculation included the following information:

1. Maximum linear heat generation rate

2. Reflood rate as a function of time

3. Core collapsed level as a function of time

4. Quench time as a function of core height

5. Core pressure as a function of time

6. Core.subcooling as a function of time

7. Heat-transfer coefficient at the. peak clad temperature (PCT) location as a function of time

8. Clad temperature for PCT location as a function of time The staff has reviewed the information and determined that the SPC interim moc 1 has been shown to be conservative when compared with the SPC FCTF reflood data. Therefore, the LBLOCA reanalysis using the interim SPC evaluation model for Millstone, Unit, 2, is acceptable.

^I l

?

,', i

^

l NNECO also provided the impact of an error in the Z-equivalent model, input 1 for a hydraulic diameter, and nonphysical trend in the reflood heat transfer correlations. The impact of the Z-equivalent error was previously analyzed by

SPC and resulted in a 15-degree F increase in the PCT, which was reported to ,

" the NRC in 1994. The-impact of the error for the limiting PCT case in the reanalysis is a decrease of about 2 degrees F for which NNECO conservatively

took no credit for. The impact of the hydraulic diameter input error is an 4 increase in the PCT of 111 degrees F and the error in the reflood heat i transfer correlations is an increase in the _ PCT of 189 degrees F.

t 2

The revised Millstone, Unit 2, LBLOCA analysis using the SPC interim evaluation model, including the errors, results in a PCT of 2111 degrees F.

This PCT is lower than the 2200_ degrees F acceptance criterion specified in 10 CFR 50.46 and is, therefore, acceptable, f 3.2 SBLOCA i

i NNECO also provided the impact of errors in the previously approved SPC SBLOCA model, ANF-RELAP, T000EE2, and RODEX2. Millstone, Unit 2, SBLOCA was i

reanalyzed with the approved SPC SBLOCA model including the impact of the identified errors. The previous analysis of record resulted in a PCT of 1707 degrees F. The following errors were identified and accounted for in the

( reanalysis:

~

! 1. A steady-state steam generator pressure control system was used instead

of the transient pressure control system.

1

2. An uncertainty in the main steam safety valve set points of 1 percent

, was used in place of the Technical Specification value of 3 percent.

1 3. The time of scram for the decay heat models was incorrect in the T00DEE2

_ heatup calculation.

{

! Millstone, Unit 2, was analyzed with a peak linear heat generation rate (LHGR) j of 15.1 kw/ft. The limiting break size is P.1 square feet. The cortaction r for the errors increased the PCT by 303 degrees F. The reanalyzed PCT for the ,

L limiting SBLOCA, including the error corrections, is 2010 degrees F. This is also lower than the 2200 degrees F acceptance criterion specified in 10 CFR 50.46 and is, therefore, acceptable. 4 4.0. CONCLUSION Based.on the above, the NRC staff has concluded that the reanalysis for both the LSLOCA and SBLOCA are acceptable for the restart of Millstone, Unit 2.

The staff is still in the process of reviewing the revised LBLOCA evaluation model submitted in SPC Topical Report XN-NF-82-20 (P), Revision 1,

. i

1 i

.. l Supplement 6 "EXEM/PWR Large Break LOCA ECCS Model Updates," December 1996.

A plant-specific analysis using the SPC modified final LBLOCA model will be required before startup after the next refueling outage for the unit. ,

l l

Principal Contributors: G. Thomas J. Staudenmeir i

Date: July 23, 1997 '

p 7

1 l(

?

h a

J i

a b

i

\'

.. - , .