ML20099F087
| ML20099F087 | |
| Person / Time | |
|---|---|
| Site: | Grand Gulf  |
| Issue date: | 08/07/1992 |
| From: | Cottle W ENTERGY OPERATIONS, INC. |
| To: | NRC |
| References | |
| GL-89-10, GNRO-92-00106, GNRO-92-106, NUDOCS 9208120114 | |
| Download: ML20099F087 (6) | |
Text
{{#Wiki_filter:n . b ui.,,, ormue,,,, io. c,:
- ENTERGY.
ne s ibi L, z t i,39 EJ t l' gi,Ji.13?[ Q ['-. i W. T. Cof fly m 5-.. 1 August 7, 1992 +- 'o U.S. Nuclent kegulatory Commission 1 Wshirigt on, D.C. 20$55 l (. Attentiont Chief, Rules and Directivo Review Dranch
Subject:
Grand Gulf NucInar Station Unit 1 Docket No. 50-416 1,1conne No. NPF-29 Gomments on. Onneric Lo.tt.or 89-10, Draf t S pplement 5 GNRO-92/00106 s Centlemen; Thn Grand Gulf Nucient Station (GGNS) Staf f has revinwnd the draf t - Suppicment 5 to Generic 1,ett er (GL) 89-10, "Inaccurney of !!otor-Operated Volvo Ding.ostic Enuipment ' Resulting from Yalvn Stem Directional Ef fect s," an noticed in Volume 57 of the redor_allegister, dated July 8, 1992. Comments result.ing from-our review are delinented fii the attachment. Se appreciate thn opportunity to comment on this proposed supplemnnt to GL 89-10 and request. NPC. conalderntion of these comments in formulat ion of the fjnni document. Please contact M. K, Drnndon at-(601) 437-6488 should thorn -be any quantions rogarding our comments. Yo.: I t rul, / ~ / s., - H' CEB/mtc attachment: Coniments on Dra f t Generic Let ter 89-10, Supp' ent. 5 cc: (Sco Next Pngo) 120033 y8 .G9208053/SNLICFLW '1' f 9208120114 920907 l PDR-ADOCK'05000416 .p-PDR
S g_ ^ N.W ^ August.7, 1992 GNRO-92/_00106 Page 2cof 3-ces. Hr. D. C. Illntz (w/a) Mr. J. L. Mathis (w/a) . Mr. R. B. McGohen (w/a) ~ Mr. N.l S. Reynolds (w/a) - 0~ Mr. II. L. Thomas (w/o) i Mr. Stewart D. Ebneter (w/n) Regional Administrator. U.S._ Nuclear Reguintory Commission J Reg on 11 j 101 Harietta St., N.W., Suito 2900 Atlanta, Georgia 30323 Mr. ' Pr W. O'Connor, Project Hansger (w/2) Office.of Nuclear Reactor Regulation U.S.- Nuclear Regulatory Comminsfon Mall 5 top 13113 Washington. D.C. 20555 i 1 l ' d 5 i I 3, j <. t t "b l lI lN lG9208053/SNLICFI.R 2 g J t-i we* v w .,e-w .v m e .
- _., _. _ _ -. _ __.. _.. = -
3 <w mvd'...-- '.~. dew'.,...
1 [i. -{'. h ATTACHMENT GRAhD GULF NUCLEAR STATION COMMENTS ON DRAFT GENERIC LETTER 89-10, SUFPLEMENT 5 -SPECIFIC COMMENTS 1. 57FR30273. Column 2. Last Paraorapjl An MOV calculated to have different than ex)ected margin due to instrument error is not necessarily inoperaale. This paragraph implies that HRC considers any infrit,gement on a conservative margin as justificaticn to declare a valve inoperable. If the NRC is taking the position that thrust settings that infringe on the calculated margins result in an MOV being inoperable, then this is a.new interpretation which requires justi_fication by an appropriate backfit analysis. The still undetermined long term accuracyz and repeatability of ANY diagnostic equipment available today leaves all thrust settings as no more than a ballpark estimate that the original-manufacturers' supplied torque switch , settings are reasonable, and that gross degradation is not present in the valve-actuator assen.bly.
- In assessing the significance of margin reduction, licensees shnuld rely on-the guidance of Generic Letter 91-18 for' dealing with degraded and/or nonconforming conditions.
Suonested Rewg"Jling If a licensee finds-an MOV does not have adequate margin, the -licensee:should take action as designated by their applicable a '89-10 and/or nonconformance programs, as supplemented by the . guidance of Generic Letter 91-18, 2. 57FR30273. Cq1gmD_3..[Lepprtino Reggji.typents (General Copment) 2
- Specific-immediate action over and above the overall action required for GL 89-10 response is not-justified.
The Lncertainty i associated with any vendors' equipment will not_be kn nn until- -years of field data is accumulated and reviewed. This issue could easily be covered and comnitments determined, as necessary, during each plants': follow-up inspections. A special report ~on.just the - use of M0 VATS. equi) ment-without avaluating;other vendor equipment errors seems:somew1at= narrow-fot.ased and premature considering the _ limited amount of data availabic. An industry wide notice, such as a information-only Generic Letter or:an NRC Information Notice, L would seem more appropriate.. This document should provide L' Page 1 of 4 l .,_r ,f - - - 3 W
, e v ^ ATTACHMENT GRAhD GULF NUCLEAR STATION COMMENTS ON DRAFT GENERIC LETTER 89-10, SUPPLEMENT 5 guidelines on the information to be contained in a licensee's final GL 89-10 response ensuring that open vs. close errors have properly been addressed. This would also allow time for further informatien indus+.ry wide to be collected, and give a structured i method to respond to other diagnostic test equiament issues as they arise. It is certain thtt other vendor pro)1 ems and other error mechanisms will develop as more valves are tested and more information is generated. The specific acticn time allotted for GL 89-10 response was done so because the industry and the hRC agreed that major safety concerns did not exist with HOVs, in general. Grand Gulf believes that this remains the case. Additionally, the requested response time is overly restrictive and burdensome on a licensee. In light of the more reasonable time period allowed for actions in response to the original Generic Letter 89-10, the period of time which different test equipment, inaccuracies have been recognized, and the everpresent . remaining uncertainty-In testing mechodologies, a rapid response is not appropriate or beneficial.
- 3..
57fR30273. Column'3 Reoortina__Reauirements. items (c) and (d) 1 The identification and reporting of_ MOVs that are not sized and set to provide sufficient margins are. currently controlled by the existing M0V programs to which each licensee has committed and which the NRC has audited and found-satisfactory for most . utilities.. There are no new findings or problens addressed in tha proposed supplement that would invalidate or require modification to a licensee's current program to determine reportability and acceptability of valve sizing concerns. 1herefore,.a request for specific reporting of individual MOV sizing. outliers and specific corrective actions is-inappropriate and inconsistent with current?y accepted practices. It appears that the actions requested in items (c) and (d) could result in a new and continuous-reporting burden that could be-cycled indefinitely as different findings regarding the assumptiens-and margins associated with valve testing occurs. Items-(c) and (d)_should be deleted as this action is currently addressed by existing programs. Page 2 of 4
i a 4 F.; g., ATTACNMENT 'l l GRAND GULF HVCLEAR STATION COMMENTS ON DRAFT GENERIC LFTTER 89-10, SUPPLEMENT 5 - 4.- 57FR30274 Column 1. Paragraoh 2 To assume that MOVs set below actuator manufacturer's original recommendations have decreased confidence seems misleading and counter productive. LThere have been substantial changas in the way switch settings are derived, such as re-establishing maximum E expected differential pressure in present design vs. original design baris, applying calculated voltage drops rather than nominal casign, and resetting torque switches with diagnostics that include all known errors and allowances. Therefore,. Grand Gulf recommends that sentences 4, 5, & 6 be-deleted or reworded. The. inference that the industry as a whole hac reduced their accident-mitigation capabilities by premature diagnostic testing with unproven technology is inappropriate and unfounded.
- r. #
Spanqated Rewordina for,$salences 5 and 6 The staff recommends evaluation of these MOVs to ensure their confidence to perform has not-been reduced. If performance confidence has been reduced on any MOV, then an appropriate non-conformance evaluation should be implemented, 5.- -57FR3s274. Column 2. Lagi Paraaraoh It is confusing.that NRC has accepted ITI-MOVATS conclusion of Part 21 non-reportability, but:NRC appears to be dogmatic on imposing a restrictive response time criteria (as would be-associated with a safety-significant Part.21 reportable condition) on the licenseet. I f. l Page 3 of 4
i* ~ ATTACHMENT GRAND GULF NUCLEAR STATION COMMENTS ON DRAFT GENERIC LETTER 89-10, SUPPLCHENT 5 GENERAL COMMENIS 1) It appears that the main intent of supplement 5 is to help establish a means for determining the error caused by the differences between the open and closed strokes. If this truly is the case, then it appears that NRC may be premature in endorsing the equation from ITI/MOVATS, In their equation to compute the rate of loading for the close direction a licensee must use the calibration from the open direction. The direct application of this calibration to a close stroke has not been well established. It seems inconsistent to une a open and close error in your calculation, when the point of this whole problem is trying to find out what that error really is. 2) The NRC's comment about lowering torque switch setting when M0V's have only been statically tested, or Differential Pressure testeJ at less than design is unclear. The same values given to the MOV vendors to size the actuator are the same values used to set up an MOV (equipment error is also added during testing). If the NRC feels that setting the torque switch setting to a vendors recommended setting (based on a spring pack curve) is better than setting up 4 MOV to some measured value using diagnostic equipment, then NRC should explicitly endorse this as an acceptable means for establishing valve settings. 3) The ITI/MOVATS equation endorsed by NUMARC and by NRC in the draft supplement assigns pre-selected values to two " variables" in the equations, stem factor of 0.15 and the generic sprin0 pack curve from Limitorque. Not all plants have or use stem factors of 0.15, their actual stem factors may be better or worse than those in the calculation. If a valve's stem factor is better, then it may not need the values that come from the calculation; however, if the stem factor is worse than 0.15, then the valve may not be able to achieve the values of the calculation. The. calculation also relies on the design spring pack curves. This practice may not provide credible results as some of the values-are know to be erroneous, in summary, the validity of this calculational methodology is questionabin' based on the fact that it uses an error to find an error and that it uses two other assumed but urknown values in it's computation. Page 4 of 4 _ _ - _}}