ML20206E337

Forwards Detailed Response to Points Made in NRC 870226 Response to 861024 Denial of Violations Noted in Insp Repts 50-327/86-42 & 50-328/86-42.Work Instructions Adequate to Ensure Work Performed Correctly

ML20206E337
Person / Time
Site:	Sequoyah
Issue date:	04/06/1987
From:	Gridley R TENNESSEE VALLEY AUTHORITY
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
NUDOCS 8704130585
Download: ML20206E337 (7)

Text

.

o l TENNESSEE VALLEY AUTHORITY CH ATTANOOGA. TENNESSEE 37401 SN 157B Lookout Place APR 061987 U.S.~ Nuclear Regulatory Commission Attn: Document Control Desk Washington, D.C. 20555 Gentlemen:

In the Matter of ) Docket Nos. 50-327 Tennessee Valley Authority ) 50-328 SEQUOYAH NUCLEAR PLANT (SQN) UNITS 1 AND 2 - NRC-OIE REGION II INSPECTION REPORT NOS. 50-327/86-42 AND 50-328/86 DENIAL OF VIOLATION We have reviewed your February 26, 1987 response to TVA's October 24, 1986 denial of Violation 50-327, -328/86-42-01 (Example 2). Your February 26, 1987 letter concluded the violation occurred as stated in the Notice of Violation.

Upon evaluation of the reasons presented for your determination and our reexamination of the issue. TVA disagrees that the reasons provided in your response enclosure support a determination that environmental qualification of the subject valves was degraded. A detailed response to points made in your response is provided in the enclosure to this letter. Accordingly, we conclude that the work instructions were adequate to ensure the work was performed correctly and therefore no violation of a regulatory requirement occurred.

If you have any questions, please call M. R. Farding at 615/870-6422.

To the best of my knowledge, I declare the statements contained herein are complete and true.

Very truly yours, TENNESSEE VALLEY AUTHORITY l

R. ridley, irector Nuclear Safety and Licensing Enclosure cc: See page 2 B704130585 ADOCK870406 05000327, PDR PDR G

Ol

, l' An Equal Opportunity Employer

U.S. Nuclear Regulatory Commission APR 061987 cc (Enclosure):

Mr. G. G. Zech, Assistant Director Regional Inspections Division of TVA Projects Office of Special Projects U.S. Nuclear Regulatory Commission Region II 101 Marietta Street, NW, Suite 2900 Atlanta, Georgia 30323 Sequoyah Resident Inspector Sequoyah Nuclear Plant 2600 Igou Ferry Road Soddy Daisy, Tennessee 3/379

ENCLOSURE I OIE INSPECTION REPORT 50-327, -328/86-42 l SUPPLEMENTAL RESPONSE l DENIAL OF VIOLATION 50-327, -328/86-42-01 )

Violation 86-42-01 contained two examples involving field installation of environmentally qualified ASCO model NP8316 solenoid valves. The first example involved a field fabricated support which deviated from typical support drawings and a documented variance could not be retrieved. The second example maintained that work instructions were not adequate to ensure the work was performed correctly in that the specific torque values and the tightening pattern provided in the associated vendor bulletin for the valve bonnet end-plate screws were not included in the work instruction. TVA admitted the first example and denied the second example of the violation in our response -

dated October 24, 1986, and NRC responded to the denial (example 2) in a letter from G. G. Zech to S. A. White dated February 26, 1987.

4 The FeLeuary 26, 1987 NRC response enclosure cites the following points as basis for concluding' the violation occurred as stated in the Notice of Violation.

1. It is questionable that work instructions used for installation of the subject ASCO solenoid valves were adequate since the associated vendor bulletin documents specific torque values and tightening pattern for the four valve bonnet (end-plate) screws.

2. Since the vendor was specific with regard to torquing values and

, tightening pattern, these items are considered significant and should be followed. ,

3. Subsequent to identification of this issue, SQN tightened the subject bolts to the vendor values.

4. Since the valve is environmentally qualified as a whole device, TVA's previous argument (that the portion of the valve which would be affected _

by improper installation is isolated from the portion of the valve which  %?l contains the electric coll) is considered not applicable to the situation.

5. TVA stated in response to a deficiency at Watts Bar Nuclear Plant (WBN) that ASCO solenoid valves nre supplied as a complete qualified package with supplier-imposed instructions which must be followed to ensure their ,

application in the plant conforms with qualification tests. This ,; ,

position appears to.be contrary to the TVA SQN position taken in denial of the violation.

NRC then concluded with the determination that environmental qualification of the subject valves was degraded.

The following addresses each of the above points and refutes the basis for l NRC's determination. TVA disagrees with NRC's determination and accordingly denies the violation as stated. The following is not meant to supersede the previous TVA response, only to supplement / augment as necessary to address specific remaining NRC concerns as reflected by the February 26, 1987 NRC response enclosure.

y m_~._ -

. . - ~w , -w ne- . n.2.d$kui

Item 1 Vendor manuals / bulletins are written in various ways, by no consistent standards or criteria, and to serve a variety of purposes. Accordingly, information that is provided, general or specific, cannot immediately be assumed to be critical or noncritical (with respect to qualification or operability) simply by inclusion in the document. Additionally, it cannot be categorically assumed that all critical information is encompassed by the vendor manual. While TVA agrees that vendor manuals / bulletins may contain some instructions which must be followed to maintain operability or qualification. TVA disagrees that simple inclusion of information--general or specific--constitutes requirements for ensuring operability.

In the case of all 10 CFR 50.49 equipment, and specifically for the subject ASCO solenoid valves, information contained in the associated vendor bulletin was thoroughly evaluated to determine those requirements, in addition to those from other sources, which are necessary to maintain qualification of equipment particularly because of its location in a harsh environment. The resultant requirements were translated into the corresponding Environmental Qualification Program Qualification Maintenance Data Sheet (QMDS). Consistent with program development procedures and 10 CFR 50.49, no additional documentation /dispositioning of the information not included in the QMDS was .

required. This process for addressing vendor-supplied requirements and recommendations was questioned by NRC unresolved item (URI) 327, 328/86-42-02

. and again by NRC letter dated July 31, 1986. A special inspection was performed by a joint Office of Inspection and Enforcement / Region II (OIE/RII) team December 8-12, 1986, to specifically address these concerns.

Additionally, the ASCO NP8316 solenoid QMDS was reviewed as part of this inspection, and the torque values for the valve bonnet end-plate screws were specifically evaluated. NRC determined that appropriate vendor requirements, recommendations, and/or suggestions were being properly accounted for in the SQN EQ program. No deficiencies were identified, and open item 86-42-02 is considered closed. These results are documented in Inspection Report Nos.

50-327/86-70 and 50-328/86-70, transmitted by NRC letter dated March 9, 1987, from Ellis W. Merschoff to Steven A. White.

In summary, an NRC-reviewed process for evaluating and dispositioning vendor manual information has been employed at SQN for 50.49 equipment (resultant QMDS), and TVA disagrees that the work instructions for installation of the ASCO solenoid valves were inadequate simply because the vendor manual contained specific information which was not employed in the installation.

. b

- 3 .+;" 17 m sr.yn ,-.,_. . ~ , . sa.

93 = % e ..

m . .

-3 '

Item

• 2 In continuation and reiteration of the previous discussion for item 1, vendors include a varying degree of detail in manuals / bulletins with varying importance, from incidental to critical. The product user should judiciously evaluate the vendor-supplied information both for significance, applicability, and adequacy based upon technical and service requirements of the end-use.

TVA disagrees generically with the concept that if a vendor manual / bulletin is specific in a piece of information then the information is automatically significant. .

For the subject issue, specific torque values and tightening pattern were provided for ASCO model NP8316 solenoid valve bonnet (end-plate) screws in the vendor bulletin. The following summarizes why evaluation of this information determined that the torque values and pattern are not critical /significant to ability of the valve to perform its safety function in either a mild or harsh environment.

o Neither valve nor solenoid housing was disassembled; the lower two of four vendor-supplied end-plate screws were removed to allow field mounting and were then replaced.

o Handtightening of the subject screws and subsequent performance of a leak test (snoop) and functional test per the work instructions provided '

assurance that adequate air pressure would be delivered to and maintained '

on the valve. The specified torque value, 95 t 10 inch-pounds, is easily within the range of torque that could be achieved manually; 1.e., this value is basically "handtight." It may be noted that, in a harsh environment, potential seal leakage should be less severe because of decreased differential pressure across the seal and a tighter seal caused by thermal expansion of materials as a result of high operating temperatures.

A- o IE Notice 84-23 does not allow ASCO test report AQR-67368, Revision 1

\

(August 1, 1983), to be used in its entirety for qualification of NP8316 models because of conflicting test results observed by Franklin tests for high temperatures (higher than used at SQN for NP8316 qualification).

I However, this test report does reflect ASCO's position on leakage at the i valve bonnet end-plate. In addressing an anomaly for leakage on the NP8316 models and other models, ASCO takes a uniform position as f reflected by the following quote, " External leakage, which does not T affect ability to perform a typical safety function, was detected at both

[ Lody/ bonnet joints and at the solenoid base sub-assembly / body joint."

This concludes that some slight external leakage at the valve bonnet end-plates will nec m#fect the valve's ability to perform its safety function in a har.h environment.

9 P 4

,p h 4.,.,,, - "  % . .w- - . L_ . . .. .. u A. Ae., , -sL-

_4_

l 1

o TVA has discussed the torque values with ASCO on four different occasions. The ASCO representative did not identify the torque values as ,

" critical" with regard to ensuring equipment operation in a harsh l environment and agreed qualification had not been violated by the installation process used by SQN in accordance with the associated work instructions.

o ASCO test report No. AQS 21678/TR, Revision A, which was used by TVA to qualify the SQN model NP8316 solenoid valves, did not identify any harsh environment leakage concerns associated with the valve bonnet end-plates.

o All solenoids addressed by the work instructions in question (workplan.

11806) are fail-safe valves. On loss of air or power they fail to their safety position. Any loss of air that could result from potential seal problems because of lack of adequate torquing--which is not expected based on the harsh environment the valves will see at SQN and a review of test results for this environment--should not affect safety function, i.e., if excessive leakage did occur, the leakage should not prevent the valves from performing their safety function, which is fail-safe.

For these reasons, neither functional operability nor environmental qualification for the subject ASCO was degraded by not torquing the end-plate screws to the vendor manual values during valve installation performed pursuant to the cited work instructions.

Item 3 While environmental qualification was not degraded by not using specific vendor-supplied torque values, it would have been clearly acceptable to cite /use the vendor torque values. Using these values, as compared to handtightening in conjunction with subsequent connection leak test (snoop) and functional test, is of no apparent benefit insofar as operability is concerned f as previously discussed. Following identification of the NRC concern (initially an open item), SQN tightened the subject screws on the affected valves to the vendor torque values. This was a conservative action taken in response to NRC's concern. The action does not reflect any change in TVA's k position regarding the operability / qualification of the solenoids before or in f absence of such action and is therefore irrelevant to determining whether a

violation occurred.

[ Item 4 k

TVA agrees that the subject solenoid valve must be evaluated as a whole tested

,-g device, but disagrees that all specifications for the whole tested device must be adhered to unconditionally in order to maintain qualification. The discussion provided in TVA's previous SQN response regarding internal isolation of the affected portion of the valve from the portion containing the electric coil was not intended to indicate the whole qualified device did not have to be considered. The physical composition of the qualified device, test reports, and vendor information must all be considered in determining critical requirements / specifications, etc.; the internal isolation is a physical aspect to consider when evaluating this information relative to equipment performance when exposed to a harsh environment.

w~ , '

fs 3. _

. . . . ~ - ~ . - - - - ~ . v. L .~. .. ~.Y~~ -

. _ . _ _ ._ ~. ~ __ __

I t

Item 5 It appears that the cited statement from the WBN deficiency final report has been viewed by NRC out of context and misinterpreted into a generic policy position which is not consistent with either WBN or SQN EQ program requirements. The deficiency response for WBN addressed a different situation t than that which is under consideration by the subject violation for SQN. The '

valves at WBN were disassembled and reconfigured from the vendor-supplied configuration by a revised mounting method and replacement of the valves'

)

bonnet screws with threaded rods.to facilitate the field mounting method, The WBN response concluded that the field mounting jeopardized the EQ and seismic

qualification, and replacement of the screws with threaded rods to establish a pressure boundary violated the functional qualification. Any technical evaluation of the WBN situation leading to this conclusion should not be confused with the SQN issue and its associated technical evaluation previously discussed in TVA's October 24, 1986 response and in items 1 through 4 of this response.

With regard to the cited WBN statement TVA agrees that the subject ASCO solenoid valves were supplied as a complete qualified package with accompanying vendor instructions. TVA's position regarding consideration of the whole device and evaluation of vendor instructions has been previously discussed under items 1, 2, and 4. The remaining portion of the statement, that the supplier-imposed instructions must be followed to ensure their application in the plant conforms with qualification tests, presents a valid premise but which may need further clarification / delineation. As previously discussed, certain vendor instructions may be determined to be critical (must be followed to ensure component operability when exposed to a harsh environment) based upon technical evaluation of the test reports, vendor i information, physical configuration, and design requirements. The statement applies to critical instructions and should not be interpreted to present a generic TVA position that any deviation from vendor instructions constitutes a violation of qualification. Such an interpretation would have been i inconsistent.with established EQ program requirements both at WBN and SQN.

We understand that the WBN statement, when considered alone, might lead to misinterpretation. However, for the reasons detailed above we conclude that the TVA response for WBN is not germane to determination of whether a violation occurred at SQN and does not alter the technical basis for evaluation of the SQN issue. ,

In consideration of our previous response to this violation and the additional NRC concerns we have addressed in this submittal, TVA continues to maintain that the subject procedures were adequate to ensure the work was performed

. correctly in that neither the environmental qualification nor functional

, operability of the ASCO solenoids was violated as a result of the absence of specific torque values in the associated work instructions. TVA disagrees with NRC's position that the subject deviation from vendor instructions d

degraded the environmental qualification of the valves and accordingly denies that a violation occurred.

l L .

i

-w - -