ML19257C208

Interim Deficiency Rept Re Rod Drop Analysis.Westinghouse Recommendations Will Be Reviewed & Included in Topical Rept. Final Rept Will Be Forwarded by 800829

ML19257C208
Person / Time
Site:	Seabrook
Issue date:	12/13/1979
From:	Devincentis J PUBLIC SERVICE CO. OF NEW HAMPSHIRE
To:
Shared Package
ML19257C207	List: ML19257C208
References
SBN-109, NUDOCS 8001250241
Download: ML19257C208 (2)

Text

PUE3LIC SERVICE Companyof NewHampshre SEABROOK STATION Engineering OfRce:

20 Turnpike Road weethorough, MA 01ssi December 13, 1979 SEN-109 T.F. Q 2.2.2 U.C. Nuclear Regulatory Commission Region I 631 Park Avenue King of Prussia, Pennsylvania 19406 Attention: Office of Inspection and Enforcement

Reference:

1. Docket No. 50-443 and 50-444

2. Telecen of 11/16/79 between Joseph Mattia NRC and David Maidrand YAEC

3. Westinghouse letter dated 11/15/79 to John G. Davis NRC

4. Westinghouse letter No. MS 'IMA-2167 dated 11/28/79 to A.T. Schwencer NRC

Subject:

10 CFR 50.55 (e) Interim Report on Rod Drop Analysis

Dear Sir:

The following is submitted in accordance with the requirements of 10 CFR 50.55 (e) . This deficiency was reported to the Region I Inspection and Enforcement Office, by telephone, on 11/16/79.

On March 28, 1979 a previous notification under 10 CFR 50.59 was made by Westinghouse which identified the potential for certain single dropped rod eve:cs leading to calculated DNB ratios lower than repcrted to the NRC for certain classes of Westinghouse plants. A comitment was made by Westinghouse at that time to review the SAR dropped rod accident analysis methodology and to revise Technical Specifications as required. A meeting was subsequently held with the NRC on April 12 to discuss the status of evaluations at the time and a further commitment was made to keep the NRC informed as to the results of an ongoing investigacion. As a result of this ongoing review, Westinghouse has detemined that this original concern may involve additional classes of Westinghouse plants.

In addition, a concern relative to multiple dropped rod events has been identified.

This results from thr'ee newly identified potential problem areas.

1808'223 so01350 143

, f if.S.NucledrRegulatoryCommisssion Page 2

. SBN-109 An assumed malfunction in the rod controll'er circuits of two and four loop plants may result in a response similar to that previously identified for three loop plants. The error allowances for the Positive and Negative High Flux Rate Trip protection system as specified in the Technical Specifications may be non-conservative. The power dictributions and reactivity insertion profiles for dropped rod sequences originally used to define the Negative Rate Trip Setpoints may also be non-conservative.

A review of all rod control systems was made to define failure moder with respect to postulating a single failure in the Rod Control System. The four loop plant review shows that a single failure in the nuclear power auctioneering unit of the rod control system may result in more limiting DNB ratios than previously reported. Therefore, the SAR analysis may not represent the most limiting assumption for credible single failure. When this assurytion is made, the four loop plant responses to dropped rods may be similar te the three loop concerns discussed in the March 28 notification.

The second concern involves a potential reduction in conservatism in the error allowances assumed in the analysis for the Power Range Neutron Flux High Negative Rate. The value currently listed in the Technical Specifications is .5% of Rated Thermal Power. This value assumed that the differentiation of the NIS Power signal would eliminate all steady state errors in the circuit before the differentiation therefore, only precision and accuracy errors after the differentiation need be considered. This assumption is non-conservative since errors associated with precision (hysterists, repeatability, etc.) may not be entirely eliminated by differentiator. When these additional errors are factored into, the analysis, a reactor trip may not occur in some cases.

The third concern involves a potential reduction in conservatism in the method utilized to generate power distributions and rod worths for dropped rod events. New analyses were performed using a more conservative reactivity feedback model which indicated that certain multiple rod drops may not trip the reactor.

These concerns are basically a result of application of the single failure criteria in conjunction with conservative methodology. Present experience shows that a dropped rod event has always resulted in a reactor trip for a Westinghouse plant from the present Negative Flux Rate Trip logic and setpoint.

A meeting was held in Bethesda on November 19, 1979 between the NRC and W to discuss these issues.

At the meeting W_ proposed an interim solution for operating plants and committed to review a number of analytic and hardware longer term solutions as an eventual alternative to the interim recommendations. W is preparing a topical report to be submitted to the NRC by mid 1980.

We intend to review the W recommendations which will be included in the topical report and submit a Final Report on the Rod Drop Analysis to you by 8/29/80.

Very truly yours, O

John DeVincentis 1808 224 Project Manager