ML19309H023
| ML19309H023 | |
| Person / Time | |
|---|---|
| Issue date: | 03/10/1980 |
| From: | Sutton J, Whitesell D NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION IV) |
| To: | |
| Shared Package | |
| ML19309H016 | List: |
| References | |
| REF-QA-99900289 99900289-80-1, NUDOCS 8005080170 | |
| Download: ML19309H023 (22) | |
Text
.
80 0508 0l7g U. S. NUCLEAR REGUI.ATORY COMMISSION OFFICE OF INSPECTION AND ENFORCEMENT REGION IV Report No.
99900289/80-01 Program No.
51300 Company:
Nuclear Valve Division Borg-Warner Corporation 7500 Tyrone Avenue Van Nuys, California Inspection Conducted: February 11-15, 1980 Inspectors
[
M [a-8,f[#
4 J. W. Sutton, Coritractor Inspector Date ComponentsSection II Vendor Inspection Branch W As CW/o!X)
D. E. Whitesell, Chief Date ComponentsSection II Vendor Inspection Branch Approved b
/C af -
hjp,a D. E. Whitesell, Chief Date ComponentsSection II Vendor Inspection Branch Summary Inspection on February 11-15, 1980 Areas Inspected:
Implementation of 10 CFR 50, Appendix B and applicable Codes and Standards, including action on previous inspection findings.
Follow-up of reported construction deficiencies to assess cause, corrective act en, and generic impact of the reported problems, nonconformance and corrective action, testing of completed products, authorized nuclear inspection interface and review of Vendors activitie:
- The inspection involved sixty (60) inspector hours on site by tv (; ' NFC inspectors.
2 Results In the six (6) areas inspected, no apparent deviations or unresolved items were identified in five (5) areas, the following was identified in the remaining one (1) area.
Deviations Testing of completed products.
. Measures had not been established to control use of temporary attachments.
(Enclosure, item A)
Unresolved Items None d
i 1
DETAILS SECTION 1 (Prepared by J. W. Sutton)
A.
Persons Contacted:
Nuclear Valve Division-Borg-Warner Corroration (NVD)
N. Bowman, Foreman B. Cost, Lead Man, Testing
- C. A. Dotter, President, General Manager
- R. W. Fritch, Vice President Engineering & QA D. K. Kellough QC Inspector P. L. Milinazzo, Manager, Fabrication Inspection
- N. J. Moore, QA Engineer V/ M. Tenones, Manager QC A. P. Tartaglia, Level III W. R. Wheaton, QA Director Kemper Insurance Companies, Lumberman's Mutual
- M. B. Diana, Authorized Nuclear Inspector
- Attended Exit Interview B.
Action on Previous Inspection Findings 1.
(Closed) Deviation (Report No. 78-02) Calibration inspector not reporting out of calibration equipment as required by QA Procedure 16.9.1.
The inspector verified by review of equipment calibration records a.
that all personally owned tools have been calibrated.
b.
A Personal Tools Discrepant /Out-Of-Calibration list has been generated and is being documented showing the use and disposal of discrepant measuring equipment.
A Training session was held on 1/11/79 for all gage and calibra-c.
tion inspectors.
The training session was documented.
2.
(Closed) Deviation (Report No. 78-02)
Welding procedure specification not at work station, and a.
b.
Electrical characteristics not being maintained.
1
\\
l l
2 3.
The inspector verified that:
a.
The weld procedures are being con: rolled by the weld shop inspector.
Sufficient numbers of weld procedures are being m2 :tained for use at the various weld stations.
b.
The inspector verified that weld procedure WPS 5-1 Revision G, has been requalified for use with 3/32 electrodes. The revised procedure has been approved.
C.
Nonconformance and Corrective Action 1.
Objectives The objectives of this inspection were to verify that:
a.
A system for control of nonconformances and corrective actions has been established and is consistent with NRC regulations, and QA Program requirements; and b.
The system is properly implemented.
2.
Method of Accomplishment The objectives of this area of the inspection were accomplished by:
a.
Review of the QA manual Section 17, "Nonconformities" b.
Review of the corrective action and follow-up logs.
c.
Review of twelve (12) selected Nonconformance reports (NCR).
d.
Review of Material Review Board (MRB) actions.
Inspection of the designated hold and segregation areas.
e.
f.
Discussion with cognizant personnel.
3.
Inspection Findings Within those areas inspected there were no deviations or unresolved items identified.
3 D.
Testing of Completed Products 1.
Objectives The objectives of this area of the inspection were to verify that:
A procedure had been prepared for the hydrotesting of valves, a.
which is consistent with Code, contract and regulatory requirements.
b.
Hydrotesting was performed by trained personnel in accordance with procedure requirements.
2.
Method of Accomplishment The preceding objectives were accomplished by:
Review of Section 12, Paragraph 12.7 of the QA Manual, a.
" Functional Acceptance Testing."
b.
Review of assembly outline and drawing No. 75830, P/R 47689.
c.
Review of Nuclear Test Summary procedure, 75832, Revision A.
for the above.
d.
Review of demineralized water chemistry controls.
Reviewed required controls for use of test equipment.
e.
f.
Reviewed the test summary for valve No. 53082, Part No. 81230.
g.
Reviewed inspection check points listed on Summary Sheets.
h.
Discussion with Test personnel regarding Hydrotest training.
i.
Verification that hydrotesting was a mandatory Authorized Nuclear Inspector hold point for valves with inlet size over four (4) inch nominal pipe size.
J.
Inspection of the following "20-28" Valves, Nos. 80605, 79810, 80429B and 81040, prior to and after testing, to l
determine if a procedure was used.to attach and remove i
temporary attachments.
k.
Discussions with cognizant personnel.
4 4
3.
Findings a.
Deviations See enclosure item A.
b.
Unresolved Item None c.
Comments During the course of the inspectors examination of valves that had been tested or was in the process of being tested, he noted that due to the size of the valves, 20" to 28" lifting lugs had to be attached to the valve bodies to falcilate movement of the valves throughout the fabrication process.
The inspector requested for his review the procedures being used to attach and remove the temporary attachments.
The inspector was informed by management that written procedures did not exist to control this process.
He was also informed, that the welding, grinding and NDE test of the affected valve areas have been controlled by verbal orders and in some cases, had been witnessed and documented by QC personnel and the ANI.
The inspector reviewed the available NDE documentation present-ed to him indicating that the valves cbserved by the inspector had been inspected by QC personnel.
The Shop Traveler does not indicate thet temporary attachments are to be used during fabrication, in aodition, the Shop Traveler does not provide for a documentation sign-off space if temporary attachments are required to be used.
E.
Authorized Nuclear Inspector (ANI) Interface 1.
Objectives The objectives of this area of the inspection were to ascertain whether procedures had been prepared and approved, which describes the system to be implemented for the achievement of interface activities with the ANI, and that the identified activities are consistent with the NRC rules, Code requirements, and the QA Program commitments.
5 2.
Method of Accomplishment The objectives of this area of the inspection were accomplished as follows:
Review of Nuclear Valve Division 1 NVD QA Manual Section a.
20, to ascertain whether the system provides for interface with the ANI and/or the Authorized Inspection Agency (AIA) to review the Design Specification (DS), and provide the inspection services, required by code, on all code items covered by the customer's order and design specification (DS).
b.
Review of nonconforming reports to verify that changes in the customer's design specifications (DS) are reviewed with the ANI to inform him of the status of the inspections and tests of the items when it is removed from the manufactur-ing process.
c.
Review of QA Manual, to verify that measures have been provided to make available Material Certifications and the QC Source and/or Receiving Inspection Reports, for review by the ANI, and that such reviews are documented.
d.
Review of QA Manual, to verify that a system has been provided to maintain the identification of materials, and that the identification is transferred when it becomes necessary to divide the material, also to verify that the ANI is provided the opportunity to verify that the identification of material is properly maintained and documented.
e.
Review of QA Manual; to ascertain whether measures have been established for the ANI to witness any welding procedure, and/or any welder performance qualification tests, and to request the requalification of any procedure or welder, to ensure proper qualification and/or acceptable welding performance.
i f.
Review of the QA Manual, to verify that the program provides for the application of the code stamp only with the authorization of the ANI after acceptable pressure testing, certification of the Manufacturer's Data Report, and stamping only in the pre-sence of the ANI.
g.
The Daily Log Book maintained by the ANI was reviewed, to verify that NVD is making its QA/QC documents available to the ANI for his review, to provide assurance that the code requirements are complied with and that the product quality has been achieved.
l
6 3.
Findings The ANI activities as documented in his log book, and by documents reviewed, supports a finding that the vendor is properly imple-menting its interface responsibilities with the ANI in a manner consistent with the NRC rules, Code requirements, and its QA program commitments. The ANI is a full time Inspector.
F.
Review of Vendor's Activities 1.
Objectives The objectives of this area of the inspection were:
a.
To review the nuclear activity and work load to assess their impact on future NRC inspections.
b.
Evaluate the vendor's fabrication / manufacturing equipment and capabilities.
2.
Method of Accomplishment The foregoing objectives were accomplished by observing manufacturing /
fabrication in progress, and discussions with cognizant vendor personnel.
3.
Findings The Vendor has the capability to design, manufacture and test ASME Code class 1-2 and 3 nuclear valves of the Gate, Globe, Lift Check, and Swing Check type varying from 3/8" to 30" in size. The Vendor also designs and Fabricates Air and Hydraulic Operators.
The Vendor is capable of both manual and automatic machine operation on valves and parts, from plate, castings, forgings, and bar stock.
The Vendor is capable of performing all welding activities including, Hard Facing, Manual, Semi and Automatic operations.
In house heat treating capabilities can accomodate valves up to 24" in size.
The Vendor performs all NDE activities, Radiographic, Ultrasonic, Liquid Penetrant and Magnetic Particle.
All Hydrostatic tests are conducted by the Vendor's personnel.
l
a 7
The Vendor holds valid ASME Certificates of Authorization "N" 1254, Class 1-2-3 Vessels, Valves and Piping systems and NPT-1255 for Class 1-2-3 Vessel and Valve Parts and Appurtenances, Piping Suhassemblies and Penetration Assemblies.
Completed products are shipped by Truck.
The Vendor is currently processing twenty-nine (29) Nuclear contracts for valves of all sizes. This represents approximately 65% of the Vendors total Commercial Nuclear work load for domestic use.
G.
Exit Interview The inspector met with management representative (denoted in paragraph A) at the conclusion of the inspection.
The inspector summarized the scope and findings of the inspection. The management representatives had no comment in response to the items discussed by the inspector.
l l
SECTION II (by D. E. Whitesell)
A.
Persons Contacted E. Frisch, Vice President for Engineering W. Wheaton, Director QA N. J. Moore, Manager QA G. Willis, Engineering Specialist B.
Followup of Reported Construction Deficiency 1.
Block Valve Malfunction a.
Background Information Duke Power Company (DPC) telephoned NRC-R11 on 12-17-79, to report that the results of a special test conducted by them, indicated that the block valve located ahead of the power operated relief valve (PORV), would not fully close under full flow condition at the specified design pressure and temperature.
The block valve was manufactured by the Nuclear Valve Division of Borg-Warner (NVD), in Van Nuys, California, for installation in Catawba Units 1 and 2, as well as McGuire Units 1, and 2.
This valve is an NVD standard 3" 1500 lb, motor operated gate valve. A special test, of this type valve was conducted by DPC at its Marshall Steam Station.
The valve was designed by NVD for a design pressure of 2485 psi at 650 F as specified on the valve list as part of DPC design specification and procurement documencs. The flow rate of 220,000 lbs. of steam per hour, is the flow rate at the design pressure and temperatures which are significantly higher than the normal operating pressure and temperature. DPC was initially of the opinion that the valve failure to fully close, was caused by internal binding, and/or, due to the motor operators being under sized.
During subsequent testing of the valve, with the cooperation of the vendor's engineer, it was determined that the flow rate was of sufficient force to impose excessive deflection on the bottom of the disc, between the guides, to permit the disc to bottom out on top of the seat ring, thus holding the valve in approximately a 10% open position.
l
2 b.
Objectives The objectives of this area of the inspection were to ascertain the following:
(1) What caused the malfunction?
(2) The action proposed or taken to correct the problem; (3) The action prepared or taken to prevent recurrence; (4) The generic impact; and (5) Evaluation of the safety significance and reportability in compliance with 10 CFR 21, 2.
Method of Accomplishment The foregoing objectives were accomplished as follows:
Review of DPC Design Specification No. CNS-1205.00-5, dated a.
2/25, 1974; and addendas No. I through 8, to ascertain what requirements were specified for the design, manufacture, inspections, and tests of the ASME Section III Stainless Steel Gate, Globe and Check Valves; also to ascertain whether any quality items were specified in excess of code requirements.
b.
Review of DPC Design Engineering Data Base, Valve Item List, dated 7/19/79 to verify the quantity, size, design pressure and tempera-ture, pressure / temperature rating, material, operator require-ments and item number for the block valves purchased for McGuire, and Catawba.
c.
Review of NVD drawi gs No. 74380 (P/N), dated 1/21/75, Valve Assembly Gate, 3 insa 1528 lb, with electrical motor operator, to determine whether the customer's quality requirements had been correctly incorporated in the design of the valve.
d.
Review of Design Report No. NSR 7170-1, dated 9/9/76, and the stress analysis for P/N 74380, 3 inch, 1500 lb. block valve, including Rev.
A., dated October 5, 1977 and Rev.
B., dated November 21, 1977, which reconciled the stress report for P/N 74380 with the customer's specified design changes.
It was noted that
(
the design report and stress analysis had been performed for NVD by the Anamet Laboratories, Inc., San Carlos, California and had been reviewed and certified as being correct by a professional engineer.
(PE) Mechanical, registered in the State of California.
3 The required dimensions of analysis report, was compared with the dimensions specified on NVD drawing No. 74380, to verify that the adequacy of the various components agreed with the stress analysis, including the seismic requirements as approp-riate.
Review of the Seismic analysis to verify that the qualification e.
was based on the analytical procedure defined in DPC " Seismic Design Manual" for rigid mechanical systems. Also to verify that the minimum natural frequency of vibration satisfies the specified limit of 33 cycles per second.
f.
Review of NVD PO No. 80948, dated 2/13/76, to Rotork, Inc., for six, No. 72866 Motor with thrust compensator, 14 NA 1, to verify that the PO was in accordance with the material bill, for the actuators.
g.
Rotork, Inc. Certificate of Compliance to NVD PO 809'3, dated 9/2/76, identifying the S/N, description, wiring diagram and tag No. of the actuators furnished, to verify that the items supplied, satisfied the NVD PO requirements.
h.
Review of Certified Material Test Reports, to verify that the material purchased was consistent with the customer specification and the design Material Bill, for the various valve components.
Also to verify whether the chemistry, and mechanical properties were consistent with the ASME Section II requirements for the material, and to ascertain whether the heat number was identified and the identification was maintained during the manufacturing process.
1 4
i.
Review of the NVD LP examination report Nos 005856, dated 9/22/76 for the valve body, 005881, dated 9/23/76 for the body assy; No. 005840 dated 9/20/76, Bonnet; No. 005487 dated 8/27/76, Grate (disc); and 004461 dated 5/27/76, stem to verify that the NDE had beer performed, the procedures used were identified, and the er.terials and batch numbers of the cleaner, penetrate and developer were identified and their halogen contents recorded.
j.
Review of the education, experience, eye examination and qualifica-tion records of 5, NVD examiners identified on the above LP Examination Reports.
l
4 k.
Review of the wall thickness verfication for the valve body and Bonnet, to verify that the actual thickness were within the specified acceptance limits.
1.
Review of the ATP Test Summary for P/N 74300, dated 9/28/76, to verify that the test pressures and holding time met code and contract requirements for the following:
(1) Black seat leakage, dated 9/26/76.
(2) Hydro Shell and Stuffing Box Leakage, dated 9/27/76.
(3) Hydro seat leakage, dated 9/27/76. (Both sides of gate)
(4) Operator Installation, dated 9/27/76.
(5) Torque Switch Setting, dated 9/28/76.
(6) Closing time (10 Sec.), dated 9/27/76.
(7) Final inspection dated 9/30/76.
Review of Engineering Change Notice for Rev. G. dated 1/2/80, m.
to NVD Drawing No. 74380, which modified the gate guide structure; and ECN for Rev.
H., dated 2/12/80, which changed the operator to a larger unit.
It was also verified that drawing No. 74380 had been revised to incorporate these changes, and that the changes had been reconcilled with the stress analysis.
Review of a letter from NVD to NRC-RV, dated February 12, 1980, n.
confirming its telecon of February 7,1980, and formally report-ing the problem to NRC, and identifying the number of similiar valves which NVD had supplied to the nuclear industry.
The report accurately described the problem and identified the nuclear customers receiving similiar valves.
The valve's Part Number and Customers Tag No. and PO were identified in each instance.
Review of the letters from NVD to the nuclear customers, dated o.
February 12, 1980, formally notifying each of them of the potential hardware problem, and identifying each of the suspected valves by the customer's tag number, purchase order number, quantity, and description of the valve, and requesting the customers to determine whether a safety related problem exists in the application, of the identified valves, evaluated on the basis of DPC special test results.
p.
Review of NVD drawings Nos 83767, Guide for 3 inch, 1500 #1b.
gate valve, dated 1/30/80; 83785 dated 1/22/80 for the machining of the gate and Dwg 84-219 dated 2/7/80 for the stem of the 3 inch gate valve; to verify that the valve design had been appropriately revised, reviewed, approved and-released in the prescribed manner, and that the changes had been evaluated against the stress analysis.
1
5 q.
Verified that the revised drawings had been released to production for the manufacture of a protype valve for full flow testing, and the manufacture of the replacement parts required to rework the existing 3 inch,1500 lbs valves to conform to the revised drawings, r.
Inspected a modified valve to verify conformance with the revised
- drawing, s.
Review of NVD Nuclear Practice Bulletin (NPB) number 11-8 dated December 29, 1977, to verify that approved procedures had been developed which prescribed NVD program for reporting defects or noncompliances to the NRC and the effected nuclear customers.
Also that the procedures provided for the evaluation of the safety significance of such defects or noncompliances in a manner consis-tent with the requirements of 10 CFR 21.
t.
Reviewed the documented summary of the special review board committee, which had been convened to evaluate the safety impact caused by the malfunction of the block valve.
Discussions with the NVD cognizant engineers and managers.
u.
3.
Findings The documents reviewed and discussions with the cognizant engineers, and management resulted in the following findings, a.
The subject block valve was tested to full flow condition that would only be experienced at the temperatures and pressures specified by the customer for the design of the valve, which was-verified from the procurement documents as being 2485 psig at 680 F.
The full flow rate at the design pressure and temperature was determined by DPC tests to be 220,000 lbs. steam / hour, which resulted in forces of sufficient magnitude to cause excessive deflection on the bottom of the gate (disc), which permitted the gate to bottom-out on top of the seal ring holdirg the i
valve in a 10% open position.
A review of the customer pro-curement documents provided evidence that flow ra*e of the valve was not part of the design parameters.
6 b.
The test procedures used by DPC in conducting the special tests were not available at the Vendor's facilities for review and evaluation. There was no information concerning the instrumenta-tion for gathering the test data available for review and evalua-1 tion.
c.
It was determined from the objected evidence available for review and evaluation, that the 3 inch, 1500 lb. S.5 gate valves were designed, manufactured, inspected and tested in accordance with the customer's contract requirements, and the ASME code require-ments for a Class 1, Seismic Category 1 valve.
l d.
No deviations from the NVD QA Program were identified.
The vendor's corrective action included the following:
e.
(1) Revised the design drawings and released them to production.
(2) Manufactured a prototype gate and gate guide, and reworked a standard valve to obtain a prototype valve for testing.
(3) Tested the modified valve assembly on a test bed capable of simulating the identified forces which caused the initial manfunction.
(4)
In cooperation with DPC, conducted a series of tests using the pressure, temperature and flow rates, at which the malfunction occured.
The modified valve was successfully tested through twenty-one (21) cycles; and cycled ten (10) more times to accumulate engineering data concerning the opening and closing thrusts, which were needed to properly size the actuator.
(5) Manufactured gates and guides, to be installed in the valves presently installed at McGuire and Catawba.
Those valves not installed, are to be returned to the vendor for modifica-tion.
(6) Prepared welding pre edures and qualified the procedure and-the welder who will participate in the field modification of the valves.
(7) Sent a field crew, including the qualified welder, to mouify on site, those valves which have been installed at the site (s).
i
7 (8) The corrective action of the DPC valves, will be completed by the first of April, 1980. The corrective action at other sites will be based upon the safety evaluation, response, and the needs of those nuclear customers to whom similar valves have been shipped.
f.
The corrective action implemented to prevent recurrence included the following:
(1) Revised the appropriate drawings identified under item (2),
to reflect the required engineerng changes.
(2) All future orders for NVD standard 3 inch, 1500 lb, gate valves will be manufactured in accordance with the revised drawings.
g.
NVD has determined that the problem is generic, and involves thirty-four (34), 3 inch, 1500 lb., gate valves, which were shipped to nuclear customers as follows:
(1) Four (4) to Palo Verde CE Tag No. SI 604 and SI 609, NVP P/N 77910.
(2) Sixteen (16) to Bellefonte; TVA Tag No. 3BWO456-KC-30; NVD P/N 79190.
(3) Two (2) to ANO-2, Bechtel Tag No. 3ACA-GT-M0-2CV4698-1PL and 3ACA-GT-M0-2CV4697-2PL.
NVD P/N 75460.
(4) Six (6) to McGuire, Duke Tag No. 9J-221, NVO P/N 74380; and (5) Six (6) to Catawba, Duke Tag No. 9J-239 NVD P/N 74380.
l All of the nuclear customers other than Duke were formally notified by letter dated 02/12/80, and each customer was requested to evcluate the required design parameters for each valve application, against the pressure, temperature and flow rate used by DPC in its test, and notify NVD if any of the identified valves require modification to sucessfully function in its intended services.
i
8 h.
The review of NVD's procedures for reporting defects and non-compliances to the Nuclear Regulatory Commission, and the affected customer; together with its procedure for evaluating the safety significance of reported defects and nonconfermances demonstrated the following:
(1) NVD was initially notified by DPC concerning the malfunction of the 3 inch, block valve in November of 1979.
NVD sent its field engineer to DPC's Marshall Stem Station to ascertain the cause of the malfunction.,
(2) Several full flow tests had to be performed before the exact cause of the malfunction was determined. The exact cause was determined on approximately January 16, 1980.
In the interim however, DPC had reported the problem to NRC RII as a potential constructon deficiency on 12/13/79.
(3) The purchase orders of other nuclear customers were canvased to determine the generic impact of the problem hardware.
The P0 survey resulted in a finding that thirty-four potentially suspected valves had been shipped to four nuclear customers, including DPC for installation at five (5) sites.
(4) The special review board committee was convened in accordance with procedures, to review the cause 02 the malfunction and evaluate its safety significance in compliance with 10 CFR 21.
(5) The review board committee determined that the hardware problem was a significant safety hazard and was reportable to the NRC in compliance with Section 206 of PL 93-438. The special review board committee's action and findings is documented in an internal NVD memo from the VP for Engineering to the President of NVD.
i.
The NRC Director RV was initially notified of the problem on February 7,1980, followed by a formal notification transmitting NVD's written report on February 12, 1980.
The report to NRC-RV identified the problem, and identified the other nuclear customers and facilities, to which similar valves had been shipped.
- However, NVD had no knowledge concerning the application of the valves and therefore could not evaluate the safety significance of the valves involved.
l
9 j.
In NVD's formal notification of the other nuclear customers NVD requested that the required pressures, temperatures, and flow rates for the intended service of each identified valve, to be evaluated against the pressure, temperature and flow rate, used by DPC in its tests of the block valves.
h.
It was determined that the report to NRC was complete, timely, and accurate.
The DPC procedures developed for the conduct of the special test, and information concerning the instrumentation used to obtain raw data of the test results, were not available at the Vendor's shop for review and evaluation.
C.
Followup of Reported Construction Deficiency 1.
Defective Seat Ring Seal Welds a.
Background Information TVA telephoned RII on 12/13/79 to report that defects had been identified in the seat ring seal weld of a 14 inch gate valve in the decay heat removal (DHR) system in Unit 1 of the Bellefonte facility. The valve had been manufactured by the Nuclear Valve Division (NVD) Borg-Warner, in Van Nuys, California.
TVA is investigating to determine the extent of the problem.
NVD reported that TVA had notified them concerning the problem on 12/27/79.
NVD sent a representative to the Bellefonte site on January 3, 1980, to inspect the weld in question and witness a penetrant examination of the weld. No code rejectable indications were identified and the valve was accepted by TVA. The NVD represent-ative reported that the weld exhibited evidence of poor workmanship which was also observed in other seat seal welds inspected.
The problem appears to be a cosmetic problem rather than defective welds, and TVA has agreed to penetrant inspect any seal weld which exhibits poor workmanship prior to the installation of the valve.
The poor workmanship consisted primarily of mismatch of the Valve Body bore and the seat ring bore which resulted in the seal weld surface to have various slopes in the mismatch areas.
10 b.
Objectives The objectives of this area of the inspection were to ascertain the following:
(1) Why the defective weld was not identified by the NVD QA/QC program; (2) The corrective action proposed or taken by NVD to correct the problem; (3) What steps NVD has or will take to prevent recurrence; (4) The generic impact; (5) Has the safety significance of the problem been evaluated by NVD or the customer in compliance with 10 CFR 21.
c.
Method of Accomplishment The foregoing objectives were accomplished as follows:
(1) Review of the custcmers Design Specification (DS)
No. BNP-DS-1935 2477-R1 dated September 11, 1973 for
" Motor Operated and Manual Valves for Bellefonte Nuclear Plant Units 1 and 2"; and the attached Data Sheet for 4, 14 inch gate valves, EMD, Part Number 80610, Mk3RWO413-ND, to verify the customers requirements for the design, manufacture, inspection and testing the valves.
(2) Review of NVD Dwg. No. 8061, dated 1/31/77, to verify that the quality requirements specified in TVA's purchase documents had been incarporated in the drawing. Also to ascertain whether the seat ring cavity and/or gate guide encroached on the minimum wall thickness of the valve body.
(3) Review of the stress and seismic analysis for P/N 80600 and 80610, 14 inch 1533 lb, Stainless Steel gate valve and electric motor operator to verify that the calcula-tions were based on the customers specified pressure, temperature, and materials.
Also to verify that the dimensions shown on the drawings for the various valve components was equal to or exceeded tle minimum dimensions as calculated in the stress analysis, and reinforced to accomodate the seismic requirements i2 appropriate.
i
1 11 l
(4) Review of NVD LP Report No. 20475, dated 7/11/78, to verify that the seat seal weld of the 14 inch valve body assembly had been examined and no code rejectable indications indentified.
(5) Review of NVD Nuclear Practice Bulletin No. 6-3, Rev. D.
dated November 13, 1978 the vendor's approved written procedures for Visusi Examinations, to verify that it prevides for the qualification of the visual inspector, the visual inspecton method, and acceptance standards, consistent w#+h the ASME code.
(6) Review of NVD internal memo from the Director of QA to the Vice President for Engineering, dated 1/07/80, which documents the chronology of events relating to the seat ring seal weld problem at Bellefonte.
(7) Review of letter from NVD to TVA dated January 10, 1980, transmitting for TVA's approval or comments, NVD's recommended visual inspection acceptance standards for the workmanship and cosmetic appearance of the seat ring seal welds of all NVD valves on site at Bellfonte.
(8) The Director of QA informed the inspector that TVA had made its comments on the above acceptance criteria by telephone, and the procedure had been revised and transmitted to TVA by letter, dated January 16, 1980.
A copy of this letter was reviewed and the letter documented that it was in response to TVA telecon comments. The inspector was informed that NVD has not received TVA's comments, or approval, at the time of inspection.
(9) Reviewed the documented summary of the meeting of the speical review board committee which had been convened by the Director of QA on January 4, 1980 to evaluate the safety significance of the reported defect, in accordance with procedures NPB 11-8.
Based on acceptable l
penetrant examination and leak test results, the committee concluced that the cosmetic appearance of the seat ring weld is not a threat to the safety function of a valve and therefore does not constitute a reportable defect under 10 CFR 21.
i 1
i l
12 d.
Findings From the documents reviewed, determinations were made as follows:
(1) The 14 inch, 1533 lb, stainless steel gate valve was designed, manufactured, inspected and tested in accordance with contract and code requirements, and no breakdown in NVD's QA Program was noted.
(2) The Vendors corrective action will focus on the cosmetic appearance of the seat ring seal well in accordance with written visual acceptance standards which will be approved by both NVD and TVA.
(a) TVA is continuing its LP examination of the seat ring seal welds on all valves purchased from NVD.
(approximately 120-150) and a repair procedure will not be developed for approval until TVA's LP examination is completed, and the results in-dicate repairs are necessary.
(b) A schedule for making required repairs cannot be made until after the repair procedure has been developed, and approved, by both NVD and TVA.
(3) To prevent recurrence, the Vendor proposes to revise NPB No. 6-3, by extending its acceptance criteria to include workmanship.
(a) To minimize mismatch between the seat ring bore and valve body port, both will be counter bored after the seat ring has been installed and prior to seal welding, when a mismatch exists that could affect the appearance of the seal weld, (1) Procedures will be developed or revised as appropriate which established mismatch limits and covers the counter boring operation.
(4) Since the problem was based on the appearance of work-manship, rather than a code rejectable defect, it is concluded that the problem is not a threat to safety and is generic only with customers with visual acceptance standards more rigid, or broader in scope than ASME Code.
l
13 (5)
It was verified that the Vendor had evaluated the safety significance of the reported problem, in a manner con-sistent with 10 CFR 21, and the results of the evaluation was documented that the problem identified by TVA at Bellefonte was not a threat to safety, or would prevent the valve from successfully performing its intended service, and was, therefore, not reportable in compliance with 10 CFR 21.
l l
.