ML20078M872

Forwards Current SBLOCA PCTs & Details on Model Errors & Changes

ML20078M872
Person / Time
Site:	Farley
Issue date:	11/30/1994
From:	Dennis Morey SOUTHERN NUCLEAR OPERATING CO.
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
NUDOCS 9412050183
Download: ML20078M872 (8)

Text

,

Southern Nuclear Operating Company post Omce Box 1295 Birmingham, Alabama 35201 Telephone (205) 86G 5131 k

h

$0uthern Nudear Operating Company Dave Morey ha*riehr'$e*c"l the southem electuc system Novenber 30, 1994 1

Docket Nos. 50-348 50-364 10CFR50.46 U. S. Nuclear Regulatory Commission ATTN: Document Control Desk Washington, D.C. :20555 Joseph M. Farley Nuclear Plant Peak Clad Temocrature (PCT _) Calculation Gentlemen:

Southern Nuclear was recently informed of three new SBLOCA evaluation model errors

_ (in the NOTRUMP code) in the calculation of PCT for Farley Nuclear Plant Units 1 and

2. Additionally, one group of changes to the SBLOCA evaluation model (in the SBLOCTA code) has also been identified. The sum of the absolute magnitude of the above new errors and changes is 53'F for Unit 1 (up-flow configuration) and 93T for Unit 2 (down-flow configuration). The " net effect" of the new errors and changes to the SBLOCA evaluation model PCT is a 417 penalty for Unit 1 and a 577 benefit for Unit 2, resulting in total PCTs of 1976'F for Unit 1 and 18397 for Unit 2.

Accordingly, Southern Nuclear has determined that the above absolute sums constitute significant errors and changes in the SBLOCA analyses for both Farley units and are reportable under 10CFR50.46 to the NRC. provides the current SBLOCA PCTs for Farley Unit I and Unit 2. provides additional details on the model errors and changes.

Southern Nuclear will incorporate model changes at the next licensing action requiring reanalysis of the SBLOCA and for which NOTRUMP would be used as the Evaluation Model. Southern Nuclear proposes not to reanalyze the SBLOCA analysis at this time because:

l (1)

The current magnitude of the PCT for the SBLOCA event maintains considerable margin to the 2200T acceptance criteria.

(2) The overall conservatism in the SBLOCA Evaluation Model has been approved by the NRC.

05009, 9412050103 94t130 PDR ADOCK 05000348 P

PDR g

i y-

~

w

U. S. Nuclear Regulatory Commission Page 2 (3) The NRC is currently reviewing the SBLOCA model changes associated with SI in the broken loop and the improved condensation model (WCAP-10054 forwarded from Westinghouse to the NRC by letter NTD-NRC-94-4278 dated August 31,1994). The NRC review of these model changes may have further impact on the NOTRUMP code.

(4)

The PCT assessment for the recent hot rod model changes was based upon partial reanalysis of the SBLOCA event using the SBLOCTA code (see Attachment 2).

(5)

The revision of the NOTRUMP code to model safety injection to the broken loop and incorporate a new condensation model resulted in a net PCT benefit.

(6)

The assessments performed by Westinghouse to address new issues and phenomena since the previous analyses were conservative.

It has been determined that compliance with the requirements of 10CFR50.46 continues to be maintained when the effects of plant design changes are combined with the effects of the SBLOCA Evaluation Model changes and errors applicable to Farley Units 1 and 2.

Respectfully submitted, SOUTIIERN NUCLEAR OPERATING COMPANY l

lIl]p D. N. Morey DNM:AA/ gps Attachments cc: Mr. S. D. Ebneter Mr. B. L. Siegel Mr. T. M. Ross wd6 gmPCTCAlf

JIA -f 3A-' A e- .=ltW a J..- a 4 - 1 \\ i s 9 l. ATTACHMENT 1 F JOSEPH M. FARLEY NUCLEAR PLANT l UNITS 1 AND 2 PEAK CLAD TEMPERATURE FOR-SMALL-BREAK LOCA ANALYSES I L t i t 1

ATTACHMENT 1 JOSEPH M. FARLEY NUCLEAR PLANT TOTAL RESULTANT SMALL-BREAK LOCA PCT (T) UNIT 1,*F UNIT 2,7 A. ANALYSIS-OF-RECORD (VANTAGE 5)

1. ECCS Analysis

• 1785 1763

2. T, Temperature Uncedainty*

20 20 Total Analysis-of-Record PCT = 1805 1783 B.10CFR50.46 MODEL ASSESSMENTS

1. Prior Reported Assessments 130' 113*

2. Boiling Heat Transfer Correlation Error (1)

-6** -6**

3. Steam Line Isolation Logic Error (1) 18**

18**

4. Axial Nodalization, RIP Model Revision, SBLOCTA Error Corrections Analysis (1) 14**

-49 ' *

5. Change in Burst and Blockage / rime in Life (2) 15**

-20** C. 10CFR50.59 PLANT MODIFICATIONS None 0 0 D. TOTAL RESULTANT SMALL-BREAK LOCA PCT 1976 1839 Previously reported to the NRC under the 10CFR50.46 significant reporting requirements.

• The new changes and errors in the small-break LOCA Evaluation Model for Farley Units 1 and 2 since the last significant changes and errors reported to the NRC. The sum of the absolute magnitude of the new changes and errors for each unit is considered to be a significant error and thus reportable within 30 days under 10CFR50.46 (i.e.,537 for Unit I and 93T for Unit 2).

(i) See Attachment 2 for details. (2) For Burst and Blockage / Time in Life, penalties of 527 for Unit I and 35T for Unit 2 were previously reported to the NRC and were included in B.1 above. Item B.5 reflects changes to the reported values for this issue (see Notes A and B in ).

i 9 e g 9 4 ATTACHMENT 2 DISCUSSION OF SMALL-BREAK LOCA MODEL EROM M GMU t

1 ATTACIBiENT 2 DISCUSSION OF SMALL-BREAK LOCA MODEL ERRORS AND CHANGES i A. NOTRUMP CODE

1. Boiling Heat Transfer Correlation Errors (-6T)

This closely related set of errors deals with how the mixture velocity is defined for use in various boiling heat transfer regime correlations. The previous definition for mixture velocity did not properly account for drift and slip effects that were calculated using NOTRUMP. Particularly affected were the NOTRUMP calculations of heat transfer coeflicient when using the Westinghouse Transition Boiling Correlation and the Dougall-Rohsenow Saturated Film Boiling Correlation. In addition, a typographical error was also corrected in the Westinghouse Transition Boiling Correlation. These errors resulted in a 6T benefit on PCT for both Unit I and Unit 2. See Note A below for impact on the Burst and Blockagefrime in Life penalty.

2. Main Steam Isolation Logic Error (18"F)

This error consists of two portions: (1) a possible plant-specific effect which was the result ofincorrect logic which caused the main steam line isolation to occur on the same signal as main feedwater isolation on S-signal, which is inconsistent with the standard conservative assumption of steam line isolation on Loss of Offsite Power coincident with the earlier reactor trip signal, and (2) a generic effect which was applicable to all previous analyses and was the result of incorrect logic which always led to isolation functions occurring at a slightly later time than when the appropriate signal was actually generated. The plant-specific effect is not applicable to Farley; however, the generic effect resulted in a 18*F penalty on PCT for both Unit I and Unit 2. See Note A below for impact on the Burst and Blockage /fime in Life penalty.

3. Core Node Zirc Oxide Initialization Error (OT)

The NOTRUMP code models two regions for each core node analogous to the two regions (mixture and vapor) in adjoining fluid nodes. During the course of a transient, NOTRUMP tracks region-specific quantities for each core node. Erroneous logic caused incorrect initialization of the region-specific fuel cladding zirc oxide thickness at times prior to the actual creation of the relevant region during the core boil-off transient. However, this error had a OT effect on PCT. See Note A below for impact on the Burst and Blockage /fime in Life penalty.

Note A: Associated with the above NOTRUMP errors is a 10*F increase in the current Burst and Blockage / Time in Life penalty for Farley I and an 8'F increase in the current Burst and Blockage / Time in Life penalty for Farley

2. Note that the Burst and Blockageffime in Life penalty is a function of PCT.

B. SBLOCTA CODE

4. Axial Nodalization, RIP Model Revision, SBLOCTA Error Corrections Analysis (14*F for Unit I and -49'F for Unit 2)

The standard rod model developed by Westinghouse in the 1970's which was used to perform SBLOCTA calculations has 19 axial nodes with a finer distribution in the top elevations. However, sensitivity studies performed by Westinghouse to justify the number and distribution of these nodes could not be documented. Therefore, a series of calculations were performed using increasingly finer axial nodalizations than prescribed for the 19-node model and indicated that the standard SBLOCTA 19-node model was not conservative. Nearly all cases demonstrated a significantly non-conservative behavior with respect to PCT. The penalty is attributed to a net increase in single phase steam enthalpy rise as these nodes uncover sooner and heat up more than the coarser nodes partially covered by the mixture level. It was concluded that a revised model that included a much finer axial nodalization could potentially lead to less favorable results than those predicted in the current analyses. Thus, a revised standard for rod nodalization has been established that ensures an adequate solution to the hot channel calculation by specifying a fine nodalization of 0.25 ft nodes for all elevations that are predicted to uncover during the transient. As a separate, but related issue, Westinghouse implemented a revised model for calculating transient fuel rod internal pressure in the SBLOCTA code. Fuel rod pressure is a governing factor in defining the clad creep, burst and blockage behavior for SBLOCA transients. The NRC has been informed of this modeling change, along with the fact that Westinghouse has validated and instituted the model as a methodology improvement to the SBLOCA model for standard implementation on a forward-fit basis. Since the improved axial nodalization methodology and revised fuel rod internal pressure model can have significant synergistic effects on the predicted PCT, the SBLOCTA calculation from the limiting SBLOCA transient has been rerun with the revised code and methodology in order to obtain an accurate estimation of the net effect of these changes on the analysis-of-record. Several recent code revisions and error corrections oflesser magnitude were also incorporated into the code version used by Westinghouse to conduct this calculation. Normally, these items would have been reported in the 10CFR50.46 year-end reporting

o-3 + s. I summary along with estimates of the effects. As a consequence of using the revised code to obtain the results for this evaluation, these items were implicitly addressed by Westinghouse in the results provided. Since all of the issues relate to portions of the SBLOCTA code and its associated methodology, they are. 1 reported as a single closely-related group of changes under " Axial Nodalization, RIP Model Revision, and SBLOCTA Error Corrections." The reanalysis of the limiting break size using the improved SBLOCTA code and methodology (with I 34 axial nodes) resulted in a 147 penalty for Farley Unit 1 and a 49"F benefit for Farley Unit 2. Note B: As discussed in Note A above, a Burst and Blockage / Time in Life penalty / benefit has also been appropriately assessed (5'F penalty for Farley I Unit I and 28'F benefit for Farley Unit 2) to the SBLOCTA nodalization error. I l _}}