ML20237E102
| ML20237E102 | |
| Person / Time | |
|---|---|
| Site: | 05000000, Sequoyah |
| Issue date: | 11/14/1985 |
| From: | Myers H HOUSE OF REP. |
| To: | Dircks B, Kammerer C NRC |
| Shared Package | |
| ML082310219 | List: |
| References | |
| FOIA-87-726 NUDOCS 8712280157 | |
| Download: ML20237E102 (33) | |
Text
- _
November 14, 1985 To Bill Dircks, Carl Kammerer From Henry Myers Re: TVA's EQ Documentation To what extent does the NRC's review of TVA's compliance with 10CFR50.49 take account of the following deficiencies specified in TVA documents?
-NSRS Report I-83-13-Nps (June 17, 1983) describes procurement documentation deficiencies at TVA.
"These deficiencies place TVA in violation of 10CFR50, Appendix B, Criterion VII requirements for SQN; and unless changes are made to correct the problems, both Watta Bar Nuclear Plant (WEN) and Bellefonte Nuclear Plant (BLN) will be subject to the same violations." (GNS 84 0301 150, p. 1.]
Among the findings of I-83-13-NFS were:
...the handling of approximately 700 million dollars of vendor QA records is inadequate and will require tens of man-years of effort to collect, verify for completeness and accuracy, and organize and store in an easily retrievable manner records for approximately 1800 completed contract purchases." (Id. p.2.]
S 2 ':UC PR, which was res;- sible for receiving and maintaining all QA records generated during construction and procurement, " reportedly" reviewed films of records.
"However, according to a person in charge of CONST QA records, approximately 80 percent of the CONST-generated QA records for unit 1 are unreadable and the hard copy has been destroyed." (Id.)
Although NUC PR was responsible for maintaining records, the Quality Engineering Branch (QAB) of EN DES, which acquired QA documents in the course of its procurement activities and which was responsible for transferring those documents to NUC PR, was told by NUC PR that "NUC PR did not know what they wanted or needed in the way of QA records and that NUC PR did not want and could not physically store hard copies of QEB's records."
(Id. p. 13.]
.. documentation for like commodities varied from vendor to vendor and may only conclude a vendor certificate of compliance or may include as much as a data package including certification of compliance, mill certifications, test reports, etc." (Id. p.13.]
8712280157 871222 PDR FOIA WANN87-726 PDR
"No record of any transfer of QEB reccrds to WBN could be found.
There was no recorded transfer of QEB records to SQN since 1977; however, bcth QEB and SQN CONST personnel remember that several boxes of contract files were sent by QEB to CONST in the 1979-80 time period and returned because CONST said they did not know what to do with them." [Id. p.14.)
...it took about three hours to track down the pedigree of a component for the polar crane using the combined talent of the person indexing the file, the unit supervisor, and by telephone the TVA resident inspector at the vendor's site." [Id. p.19.]
"The film records received (by NUC PR from CONS) admittedly were not reviewed for accuracy or completeness by any of the QA or engineering groups.
They were reviewed only for readability by document control.
Personnel at SQN NUC PR stated that they did not knew what should be in the QA reccrds they receive. They further stated that they had a difficult time in finding anything in them and generally CONST personnel were requested to go through the film to find something for NUC PR."
[Id. p.23.]
"None of the files in the vurious locations followed a set style.
Some were found in reverse chronological order, some by items produced, some by category of information (correspondence, tests, e::.), some with a cardin?. i:n Of the at:v2 styles, and " e in no logical order." [Id. p. 24.)
"Each file contained what was considered to be superfluous information, such as multiple copies of contracts and attachments and multiple copies of correspondence and irrelevant correspondence.
Microfilmed files contained the same multiplicities and also included machine copies of test patterns, envelopes, and blank pages.
No information was found in any contract file reviewed that the file had ever been reviewed for completeness and no index was found adequate enough to allow someone looking for information to find it without going through the complete file."
[Id. p. 24-25.)
"NUC PR on the receiving end of these records (from MEDS, QEB and CONST) was taking whatever anyone wanted to give them and hoped no one would ask for the records again.
Not knowing what NUC PR wanted, both QEB and CONST took the conservative approach and filmed everything they had in a contract file, including multiple copies of the contract and attachments.
This -- __
was magnified by MEDS filming machine copied test patterns and blank pages because they were in the file.
The results of this mass filming was a considerable amount of duplication between QEB and CONST." [Id. p. 26-27.)
"Since QEB did not review the files for completeness prior to shipment in 1977, no one in QEB knows for sure whether the records were complete." (Id.]
" Assuming the regional inspectors were performing their function, and there was no reason to suspect otherwise, at the time equipment was shipped by the vendor, QA documentation existed.
Therefore, between the records stored by NUC FR SQN, CONST SQN, and QEB records at SQN, Knoxville and the regional offices and records passed by the vendor or subvendors, the QA records should exist.
The data at the vendor's facility becomes important in those cases were TVA does not have all the required QA data.
Of the contracts reviewed 4 out of 16 appeared incomplete, lacking test data and therefore fell into that category." [Id. p.29.) NOTE: What about vendors who go out of business or who do not maintain the records?
j "The overall usability of the QA records found at SQN is highly suspect fer the reasons stated above." [Id. p. 31.)
Ncto: whf ' e t'e :-8 3 ' 3 ~"75 va s c"
,-9 eted in June 1983, it appears not to have been placed in the TVA document control system until March 1984.
When were these deficiencies reported to the NRC?
[ Inspection Report 50-390/84-07 indicates that NRC learned of these l
deficiencies and I-83-13-NPS in the course of an inspection conducted on February 6 - 10, 1984.).What was the NRC's response?
Does the NRC believe that deficiencies described in I-83-13-NPS have been corrected?
What documentation exists to demonstrate that these deficiencies have been l
corrected, and that the corrective action has been reviewed i
and approved by the NRC7 l
-Among the findings of Audit CH-8500-04 were that "..no i
controlled lE list exists."
Deviation Report CH-8500-04-01
)
states: "No overall EQ program procedure existed, however, i
which properly identified all EQ requirements, interface functions and responsibilities, coordination points, j
locations of qualification documents, specifics of a l
qualification data management systems, QMDS and lE list controls, and so forth..... Due to the lack of programmatic procedure (s) (and other information, such as QMSD's and lE lists), an evaluation of each lE device must be performed l'
for each of the nuclear plants."
Deviation Report CH-8500-04-04 states that "The Office of Engineering does not ha.ve quality assurance procedures to provide for development, maintenance, and control of lE lists for NUC PR's nuclear plants as required by 10CFR50, Appendix B, Criterion V."
Note: What reviews have been conducted by NRC to establish the adequacy of TVA's QMDS's and lE lists?
-The report of the Environmental Qualification Management Review Team concluded that: "Although the review team judqcs that certain aspects of the EO program are satisf actory, there are suf ficient other unsatisf actory aspects to lead the team to conclude that documented qualification has not been established for the large majority of equipment reviewed." (Emphasis added. B70 85 0926 007, p. l.]
The report noted that the environmental review was " prematurely suspended" with one result being that the team did not visit the Sequoyah and Watts Bar sites. (Id., p. 4.]
[ Note: In view of the premature suspension of the review which led to cancelling the visit to Sequoyah, what confidence exists that there has been a proper review of the EQ situation at Sequoyah?] Among the team's findings were:
"..without field verification, the cable manuf acturer and type cannot be determined for a given circuit nunber. The team reviewed an " uncontrolled" BFN cable list and found that TVA currently possesses EQ docr entation for less than half of the cables potentially used in Class 1E circuits at BFN (C:ervhtien ll-3rN).
The team believes that cable traceability and cable list completeness problems also exist at SQN and WBN." (Id. p. 13.]
NOTE: In light of the foregoing conclusion, what is the staff's response to TVA's proposal to determine cable type from "QA pull cards" rather than field verification?
(TVA Nuclear Performance Plan, Sequoyah, p.I-1.)
l "In general, the EQ files were found to be incomplete and not readily auditable; where technical information could be found the majority of it was scattered and not easily retrievable.
... The team believes that the identified deficiencies are significant, systematic, and pervasive, in that the same type of deficiency can be expected to be found in other EQ files." (Id. p.13. )
Please provide the following:
Existing NRC staff analyses, memoranda, etc. which indicate NRC's awareness of and response to deficiencies described in the docenents referenced above.
Existing NRC staff analyses, memoranda, etc. indicating that ongoing reviews of Sequoyah environmental qualification activities and documentation for compliance with 10CFR50.49 are being conducted in a manner that assures correction of deficiencies described in documents referenced above.
1 i - _ _ _ _ _ _ _ _ _ _ _
s October 16, 1985 (Rev. 10/21,11/05,11/21/85)
TVA WELD ISSUES ANSI N4 5. 2. 5-197 4, Section 5.5 states: Inspection of structural steel welding shall be performed in accordance with the provisions of AWS D1.1, Section 6.... This inspection shall include visual examination of preparations, welding processes, and post-velding operations.
Prior to welding, verification of welding procedure and welder qualification shall be documented and shall include all essential variables identified in the procedure.
In-process inspections shall include joint fit up prior to start of welding, preheat and interpass temperature requirements, filler metal, control of distortion, post-weld heat treatment, and cleaning requirements." NOTE: FSAR COMMITS TO CONFORMANCE WITH N45.2. 5.
April 16,1980.
Bellefonte NCR 1173 states that a welding inspector had " completed G29C visual inspection records" for a series of welds before the welds had been made.
The NCR said the apparent cause was: Failure of inspector to check weld map versus welds inspected prior to completing inspection records. Action required by original NCR:
" Reinstruct all inspectors on the importance of properly completing QC inspectors records for the inspections that have been made." WRITTEN IN ON 9/22/82, 29 MONTHS AFTER ORIGINAL NCR:
Later investigation revealed this was an isolated case and therefore action required to prever.: recurrence is not applicable.
. m ' 1, 1 ^ U.
w3 NG ?37 5R.
" Welds on the above listed compcnents have been previously accepted and they do not meet the requirements for visual examination.
This was determined f rom a random sample.
70 cable tray supports were inspected and 68 rejected.
40 conduit supports were inspected and 8 were rejected.
22 misc steel items were inspected and 13 were rejected.
Action required to prevent recurrence:
All WEU personnel who inspect fillet welds have been retrained for requirements and have been given mechanical gages to use in weld size determination.
NOTE: DISPOSITION BASED ON RELAXATION OF ACCEPTANCE CRITERIA.
WHAT REINSPECTION WERE DONE TO DETERMINE CONSISTENCY WITH RELAXED CRITERIA?
NOTE: COMPLETE NCR APPARENTLY NOT AVAILABLE. WHERE IS IT?
j July 8,1980.
Cantrell to Wilkins re WB NCR 2375R.
Refers to wilkins memo, 6/23/80 (WBN 800623 006).
Reconrnends each weld examined and not in conformance with the design drawings be identified and made to agree or the location, undersize and length of fillet weld information be sent to EN DES for reevaluation.
WHAT WAS THE RESULT OF EN DES REVIEW 7 Cantrell says new drawings will be issued by ECN 2535 to indicate a reduced minimum length of weld required at cable tray support clip angles.
"This may alleviate part of the clip angle weld deficiency. "
i July 21 - 25, 1980.
Inspection which led to 80-14.
Dahnke and Gilbert contacted.
At exit interview, welding engineering unit supervisor " contended that the visual examination program was basically a good progren and that additional procedural requirements were not needed in the magnetic particle examination procedure because the inspectors were properly trained... This objection by the WEU supervisor precluded discussions of Shen the site would be in compliance with the requirements of ALME B and PV Code and the General Welding Procedure G-29M."
August 22, 1980.
Meds: 800822 006. RWC/ JEW Watts Bar seeks relief from requirements imposed by EN DES for visual inspection of welds on hangers, cable tray supports and clips.
September 3, 1980.
Dielbeler, QA/Const to R.A. Costner, QA, EN DES. BQA 800903 002. Re: Significant Audit Deficiency
-BN-W-80-08.
Found deficiencies at Bellefonte with respect to certain kinds of inspection,s not being perf ormed: verification of welding procedure and compljance thereto; verification of welder qualification; verification of joint fitup; and verification of correct filler metal usage.
Cover letter states that the deficiency is considered sig.11ficant.
Makes findings re kinds of inspections not being perforned.
(See September 12, 1980 R-II to TVA.) A " partial listi:g of the type welding where the required inspections are not being per formed includes Safety related seismic supports, safety related cable trays, safety related misc. exposed steel, and safety related seismic pipe supports. Auditors recommend (Item 5) "Nonconform _all (both ASME and ncn-ASME) safety related welding that has not been assigned
)
the ir < ractic"'/t"r! #'-$ i' ens det a iled in Section 4 of this deficiency."
Revise T/A program to assure that all :equired prior to and during welding inspections are invoked and performed during all safety related welding.
NOTE:
BN-W-80-08 led to a 50.55e report submitted to NRC Region II on October 20, 1980.
Among other things, the 50.55e stated: "Although this deficiency was originally written on Bellefonte Nuclear Plant, it has been determined to be generic to all TVA nuclear plants under construction." (CS File 4.1. )
REFERENCES TO FILLER POVTERIAL: 11/5/80, 6/22/81,10/10/83, 1/12/84, 1/18/84, 1/23/84.
September 4,1980.
TVA EN DES and CONST QA staff issue report
" Watts Bar and later Nuclear Plants."
This report examined TVA's weld program.
The report questioned the adequacy of many significant elements of the welding activity; e.g. qualification and certification of inspectors for visual weld inspection, unavailability of tools necessary to determine weld acceptability, the adequacy of G-29c (finding C-12), audit adequacy, separation of craf t and inspection personnel, adequacy
~of welding engineering personnel and practices, inspector experience, insufficient ratio of inspectors to welders, adequacy of inspector training, inadequate fulfillment of visual inspection requirements. [WAS THIS REPORT PROVIDED TO NRC? WAS IT REPORTED VI A 50. 55e? ) (CS File 0.2) NOTE ALSO THAT REPORT WAS APPARENTLY NEVER PLACED IN THE TVA MEDS SYSTEM.
September 5, 1980.
In course of NRC inspection 80-22, TVA informs NRC of inadequacies in weld inspection procedures. (See entry dated November 3, 1980.]
September 12, 1980. R-II to TVA. Forwards report of Bellefonte inspection 439/80-14 (conducted July 21-25) and NOV based thereon.
Findings were based on NRC inspection involving pipe and structural welds.
"The licensee has identified several hundred piping socket and structural fillet welds which had been accepted by visual examination but did not meet acceptance criteria.
NOV involves inadequate qualification of visual examiners, improper visual examination of pipe weld, f ailure of visually inspected welds to meet acceptance criteria, magr.4Eic particle testing procedural deficiencies, and failure to fulfill 50.55e requirement re deficiencies in safety related welds that had been accepted by QC inspectors.
September 12, 1980.
R-II to TVA.
Attached inspection report (439/80-14) refers to finding from previous inspection, 438/80-07-01, relating to weld material control: "The inspectors reviewed the procedure changes which had been put into operation during the week of July 20.
Inspection of the work areas showed several exampics which indicated that the program was not yet fully implemented in that the licensee's commitment was full i
compliance by August 1, 1980.
l c;;_ -- 10, 1." O.
"7
~ cj ect 'M-::ar te J.C.
Killian, Assistant Manager of Construction.
Requests OEDC assistance in resolving BLN inf ractions specified in 439/80-14-01 and 439/80-14-02. These inspection reports cite BLN CONST for having a visual weld examination program that does not meet ASME(??) Code requirements. BLN's and TVA's quality assurance procedure for NDE certification does not require visual examination as an NDE process.
NRC says that it is and points out that BLN CONST is maiss in not having visual examinee candidates administered an exam, including a practical test, by a Level III NDE examiner. "As resolution of this inf raction could ultimately affect the authenticity of every BLN and TVA Code visual weld examination performed past, present and future, it is requested that responsible NEB Code and OEDC QA personnel formulate the corrective action to be taken along with an engineering evaluation of all code visual examinations performed to date and advise BLN CONST of the results as soon as possible."
NOTE: TVA response to NRC on January 13, 1981 says visual examination is not required.
October 15, 1980.
Cantrell to Wilkins (WBN PM). Discusses attempts to provide relief re problems associated with visual inspection of WBN welds.
(This memo was removed from files per 2/6/81 memo, cantrell to Wilkins.]
l October 20, 1980. 50.55e. Bellefonte. Construction deficiencies; non-ASME welds. Certain categories of safety related welds are not inspected in accord with requirements. States that although j
the deficiency was originally written on Bellefonte,
"..it has been determined to be generic to all TVA nuclear plants under cons t ruc t ion. " (CS File 4.1.] (See 80-22, November 3,1980. )
October 24, 1980.
TVA to Region II.
"TVA does not agree that our program (re visual examination) is in noncompliance with the
{
AsME code." TVA said that although it believed it was not in noncompliance with requirements that it was concerned with deficient conditions and would upgrade its " entire welding and nondestructive examination program to increase its efficiency."
November 5(3), 1980.
80-22 includes following concerning Inspection Report Item 439/80-22-02 re Safety related Non-ASME welding Inspection. On September 5, 1980 the licensee informed R-II that adequate measures did not appear to have been established to assure that certain categories of safety related welding received all required inspection verifications.
As a result, some welds were not receiving inspections required by ANSI N45.2.5 (1974) including verification of: (1) welding procedures; (2) welding procedure compliance; (3) welder qualification; (4) joint fitup; and (5) correct filler metal usage.
The licensee's investigation of this area is continuing.
j
[ NOTE: ANSI N45.2.5 requires that structural steel welding be in l
accord with Aws C1.1. N45.2.5 states that inspections shall include verification of welding procedure and wc; der qualification.
In process inspections shall include joint fitup i
etc.]
NOTE: QTC believes noncompliance with N45.2.5 continues.
December 1,1980.
NCR 2654R reported to NRC/OIE Inspector Cantrell. 245 duct supports inspected prior to 3/27/80, approximately 22% contained unacceptable welds.
Defects included l
undersized welds, incomplete welds, slag inclusions, porosity and overalpa.
Final 50.55e report submitted December 31, l
1980. Revised Final 50.55e submitted January 25, 1982.
(See l
entry of latter date.]
December 3,1980. Cantrell to Wilkins re relaxation of code requirements. Removed from file per 2/6/81 memo, cantrell to Wilkins.
December 5,1980.
Region II to TVA.
R-II does not accept October 24 explanation stating compliance with the ASME code in response to 80-14's finding that there had been a " failure of visual examination program to comply with applicable code requirements.
..In view of the above, the NOV transmitted to you on August 28, 1980 remains unchanged.
Section 2.201 requires you,
L__________-___
to submit to this office a written statement or explanation in reply including.." corrective steps which have been taken, corrective steps to avoid further noncompliance, and the date when full compliance will be achieved." (WHERE IS THE RESPONSE TO THE DECEMBER 5, 1980 LETTER?] (SEE JULY 14, 1981. R-11 TO TVA.)
December 9, 1980.
Cantrell to Wilkins. Apparently re relief from problems associated with visual inspections.
Removed from files per 2/6/81 memo, Cantrell to Wilkins.
January 9,1981.
Barnett visits plant to observe first hand the weld problem.
See 2/5/81, Barnett to Cantrell.
January 13, 1981.
TVA informs R-II that its October 24 response had adequately taken care of problems.
TVA appears to hold to its assertion that its procedures for examination of weld inspectors had always been in compliance with NRC requirements.
TVA states it wanted to meet with NRC on January 22.
It is unclear whether this meeting tock place.
A meeting was held on i
February 2 and follow up telephone conversation occurred on February 4.
(See October 10, 1980 cited above.)
NOTE: If it is true that inspectors had been adequately qualified, why did they continue to accept welds that should have been re1ected?
January 20, 1981.
Barnett to Cantrell re relaxation of weld requirements.
Renmoved from files per Barnett's 2/5/81 memorandum.
February 2, 1981.
Meeting betwcen TVA and R-II.
[WHERE IS THE FICCED7]
February 4, 1981.
Follow up telephone conversation.
(WEAT WAS DISCUSSED 7]
February 5, 1981.
Barnett (Chief, CEB) to Cantrell (SQN and WB Design Projects Manager) re AWS welding requirements.
Supercedes Barnett memo dated 1/20/81, CEB 810120 004. "The superceded memorandum is to be removed from the files.... Following a plant visit on January 9,1981, it became evident problems exist for a large percent of welds where an overstrict adherence to visual inspection requirements of AWS Dl.1 requirements are overly restrictive when literally applied to the type structures involved." Application of less stringent visual acceptance requirements will require review of designs to substantiate a relaxation of D1.~1. requirements.
Revision to the SAR will be required and changes to design criteria is probable. ' Attachment A gives interim acceptance levels, less stringent than D1.1.
NOTE: Where are documents removed from files?
February 6, 1981.
JEW /RWC.
NEB 810206 265.
Transmits ravised visual acceptance criteria for cable tray supports based on a review of the applicable design calculations.
Also provides sampling plans for installed cable tray supports and installed duct supports, requiring 4500 linear inches of welds be inspected under each plan.
This is EN DES response to CONST call for help but cautions CONST that all help offered is based on our best estimate pending (a) revision to SAR, (b) review of applicable calculations, and (c) various sampling programs specified in the memorandum.
None of the relief would be final until completion of changes to design and licensing process. EN DES was confident that the same requirements adopted for cable tray supports could be substantiated for the majority of duct supports, but there was some doubt about using the same reuqirements for pipe hangers.
WHAT CHANGES WERE MADE RE FSAR? WHAT HAPPENED VIS-A-VIS PIPE MANGERS?
February 8,1981.
Cantrell (WB and SQN Design Project Manager) to Wilkins (PM, CONS, WB): Fillet Weld Visual Inspection Requirements. Lists 5 memoranda WBN 800822006, SWP 801017 012, SWP 801203 074, SWP 801210 058, and CEB 810205 001.
"This memorandum supercedes references 2,3, and 4.
These shall be removed from the filing system. Reference 1 emphasized the problems associated with visual inspection of welds on pipe hangers and cable tray supports based on the requirements of AWS D1.1 and the icpact on the WBN cost and schedule.
Subsequent events have indicated that chis problem exists for all welding where strict adherence to visual inspection requirements of AWS D1.1 have been invoked.
It is agreed that in some areas the requirements are overly restrictive when applied to the type structures involved. (NOTE: WHERE ARE REFS 2,3, AND 4?
DID THEY ALLOW RELAEATICS CF D1.1?
February 20, 1981.
TVA sends R-II revised response to 80-14-01.
"The training and certification program was not administered by
..a Level III inspector for the reasons stated in our letter dated January 13, 1981; however, TVA believes our program is equivalent to SNT-TC-1A with the exception that we do not assign Level II or Level III status to individuals."
March 6,1981.
Second interim report re 80-08. See September 3, 1980.
NOTE REFERENCE TO ANSI D1.1.
Indicates nature of ongoing review.
March 10,1981.
Region II to TVA.
R-II acknowledges receipt of TVA's February 20 and other responses to 80-14.
(One issue involved training and certification of weld inspectors. ] "Thank you for... informing us of steps you have taken to correct the item of noncompliance... brought to your attention in our letter of September 12, 1980.
We will examine your corrective actions and plans during subsequent inspections." (WHAT TRANSPIRED? SEE ALSO 50-438/81-16 (JUNE 14, 1981) and JULY 14, 1982 LEWIS TO TVA.]
March 12, 1981.
TL/MNS. NEB 810312 272.
Special Committe.te l
formed to resolve problems associated with welded designs.
EN DES formed a committee to address problems associated with f abrication of welded designs.
They address (a) and (b) per 2/6/81 cited above.
April 3, 1981.
TL/MNS.
NEB 810406 269.
Special TF formed to resolve problems associated with fabrication of welded design.
EN DES welding committee comits to revise G29c and initiate SAR revisions as applicable to reflect TVA commitment to AWS l
l Dl.l.
l May 5, 1981.
Third interim response to 80-08.
June 14, 1981.
NRC Inspection Report 50-438/81-16 closes J
438/80-14-01 re f ailure of visual examination progr a to cocply with applicable code requirements.
NRC inspector accepted "VA explanation that TVA's welding inspectro training and I
certification program (WEU-SOP-715) was " comparable: to the training and qualification requirements " delineated in SNT-TC-1A" except for requalification requirements which TVA commitud to modify. NOTE: 81-16 SPECIFIES SEVERAL OPEN ITEMS REGARDINC VELD l
PROGRAM PROBLEMS.
TO WHAT EXTENT WERE THESE ITEMS ANALYZED BY l
NRC OR TVA TO DETERMINE WHETEER THEY MIGHT BE INDICATORS OF GENERIC PROBLEMS?
June 22, 1981. TVA responds to findings of 80-22 and associated 50.55(e) which followed from construction QA Audit ES-W-80-08. TVA states that it is in full compliance with regard
=, gificnicn cf
- miding y n-Mun ud ccmpid ance therete, verification of welder qualification, and verification of correct-filler metal usage. TVA DID NOT INDICATE WHETHER THERE HAD BEEN NONCOMPLIANCE AND IF So, WHAT HAD BEEN DONE TO CORFECT FOR DEFICIENCIES ARISING THEREFROM. WHAT EVIDENCE EXISTS TO DEMONSTRATE COMPLIANCE IN THESE RESPECTS? TVA'S position on fitup inspection was that the need for same would be specified on a case-by-case basis in engineering drawings.
"However, it has been the usual TVA practice to require more rigorous inspection methods such as radiographic testing, ultrasonic testing, liquid penetrant examination, and dry magnetic particle examination in lieu of specifying joint fitup inspection."
NOTE: THE IMPLICATION IS THAT RADIOGRAPHY ETC. WAS DONE ON STRUCTURAL STEEL l
I WELDING. WAS IT?
TVA said that with respect to fitup, that design engineer's responsibility needed clarification.
WHAT IS l
THE CLARIFICATION AND HOW DID THIS COMPENSATE FOR PROBLEMS THAT MIGHT HAVE ARISEN DURING PERIOD WHEN THINGS WERE NOT SO CLEAR?
THERE ARE ALSO INDICATIONS THAT CLARIFICATION WAS NOT ISSUED.
WAS IT?
NOTE:
NRC accepts foregoing explanation on 8/18/82 per 50-438/82-23, 14 months after receiving it.
Note: SEE 1/25/82 AND 3/1/82 RE RELAXED CRITERIA.
June 29, 1981.
Sprouse to Pierce, Dilworth etc. re " Commitments resulting f rom EN DES Welding Task Force Meeting of June 1, 1981." SAR's to be revised to mention G-29c. EN DES will issue a single source document to cover items that are normally requirement of AWS and to issue implementation instruction l
applicable thereto.
HOW MUCH OF THIS WAS DCNE7 SEE 9/28/81 AND 8/20/82.
~
August 24, 1981.
Cantrell to Raulston. RE: Fillet Weld Sample program per February 6,1981 memo from Cantrell to Wilkins. Lists findings of cable tray inspection.
States that weld samples on cable tray supports requested in accordance with attachment A of 2/6/81 memo (relaxed requirements) have been received and i
evaluated in accord with Attachment A.
Requests Raulston to I
evaluate findings vis-a-vis Attachment A. NCTE:
THE DOCUMENTS INDICATE THAT THE REINSPECTION FOUND DEFICIENCIES VIS-A-VIS ACCEPTANCE CRITERIA SPECIFIED IN THE FEBRUARY 6,1981 CANTRELL TO WILKINS MEMORANDUM; THESE ACCEPTANCE CRITERIA WERE LESS STRINGENT l
THAN THOSE IMPLICIT IN G-29C.
DID NRC REVIEW THE REINSPECTION RESULTS WITH RESPECT TO ADEQUACY OF SAMPLE, ADEQUACY OF LESS STRINGENT ACCEPTANCE CRITERIA, AND COMPLIANCE WITH LESS STRINGENT ACCEPTANCE CRITERIA?
WHAT IS HOW DOES THIS INSPECTION PROGRAM i
RELATE TO INSPECTIONS DESCRIBED IN JANUARY 30, 1984 OQA MEMO WHICH CLOSES CARBO ZINC ISSUE?
September 9,1981.
Banett to Raulston.
Forwards changes from Civil Engineering Branch to Chief Nuclear Engineer re changes to reflect TVA's practice regarding use of AVS for Category 1 structures.
"Please have these changes i.ccrporated in the next s,u s.v s.t s ruwi'. t ed t a the NPC. "
September 9, 1981.
(Same as above as described in 6/18/82 i
Pierce /Jessee chronology. ) JAR / ROB. CEB 810909 007.
SAR l
commitments to AWS.
SAR commitment changes initiated by CEB to l
i NEB and continue by licensing transmittal No. 327 to regulatory staf f in Chattanooga.
It will be submitted to NRC in revision 47 to SAR. Telecon with regulatory staff indicated this will be submitted to the NRC on July 26, 1982.
(Note: AMENDMENT 47 HOT SUBMITTED TO NRC UNTIL JANUARY 4, 1983.]
1 September 16, 1981.
(Referred to in March 26, 1982 memorandum, Austin to all EEU Inspectors, WBN-CONST. ]
Cantrell stated EN DES was able to accept all as-built cable tray support fillet welds made prior to February 6,1982 in all Category I buildings.
WHERE IS THE SEPTEMBER 16 MEMORANDUM?
WEAT DOCUMENTATION UNDERLIES THIS DIRECTIVE 7 (E.G., IS AUGUST 24, 1981 MEMORANDUM J
THE BASIS FOR THIS ITEM 7)
HAS NRC EXAMINED ANY SUCH DOCUMENTATION?
september 28, 1981.
TL/MNS.
NEB 810928 292.
Welding of AISC. G29c revised as a result of the welding task force to L....
deviate from AWS D1.1.
WHO APPROVED THE DEVIATIONS?
ON WHAT DATE?
SEE 6/29/81, 8/20/82.
November 2, 1981. ENDES to CONST (Cantrell to Wilkins): This is J
in response to your verbal request that ENDES consider allowing visual examination of welds in accordance with G-29c af ter coating with carbo zinc. Based on inspection of sample welds and production welds presented for evaluation, this is acceptable provided:
KNOW IT WAS 5 MILS?)
- 2. all work af ter this date is examined prior to prbaing with carbo zinc.
December 17, 1981. Schrandt to QAB Files. Re: WB - Visual Inspection of Welds in Accordance with G-29c - Coated with carbo zinc.
Reference to Cantrell, November 2, 1981, SWP 811102 056.
The basis for accepting a visual examination of welds already coated with carbo Zinc primer was established when a group consisting of Olsen et al. performed a walk-through at WB as well as review of some weld samples in primed and unprimed condition.
The wa;1k-through was for the purpose of selecting at randam a number vf welds previously coated with primer and determining if a competent visual examination could be performed without removing the primer.
Mr. Jessee stated that it became apparent
,th a t _the_pri.ier did not create a cover which_ tended to hide q..
ec.-
.gh _71a r 7 ~-'sp thef sign, but instead,J ade_it_
Inspectior.s were easier to perform the visual inspection.
quite revealing.
It was concluded by all parties that the primed surf aces were probably easier to inspect than the as-welded condition.
It was observed that this was a lacquer primer and did not tend to level and run into crevices like epoxy or enamel.
HOW MANY INSPECTORS WERE IN THE GROUP?
l January 11, 1982. Cantrell to Wilkins (WB CONS Project Manager.) (NOTE: THIS APPEARS TO BE MEMORANDUM LISTED PER JANUARY 14, 1982 IN QTC REPORT.] "This memorandum supersedes my memorandum of November 2, 1981.
... All parties agreed that carbo zinc coating did not tend to hide objectionable weld defects but appeared to enhance their visibility.... Welds inspected for weld quality (defects other than size and locationi as part of an EN DES directed sampling program shall be inspected without primer coating unless exempted by EN DES."
States conditions under which inspection through carbo zinc is permissible; e.g. it is acceptable to visually inspect carbo-zine coated welds provide:
1.
The acceptable criteria for weld defects is in accordance with g-29c.
_9_
i 2.
The carbo zine was sprayed in accordance with the applicable coating application per specification.
3.
The carbo zine thickness is not greater than 5 mils as documented in coating inspection records etc.
"All work preformed af ter this date shall be examined before it i
is primed.
Welds inspected for weld quality (defects other than size and location) as part of an EN DES directed sampling program j
shall be inspected without primer coating unless exempted by EN Des."
NOTE: AUTHORIZATION FOR INSPECTION THROUGH CARBO ZINC DOES NOT MENTION SAMPLING PROGRAMS EXCEPT IN THE ELABORATION; THE WORDING OF THE AUTHORIZATION ALLOWS INSPECTION OF PRODUCTION WELDS AFTER CARBO ZINC COATING.
January 20, 1982.
WATIS BAR NUCLEAR PLANT - NSRS REPORT ON MINIMANAGEMENT REVIEW, R-81-28-WBN. NSRS describes violations resulting from weld inspections being performed by construction organization personnel when the regulations required such inspections to be performed by personnel from the Welding Engineering Unit.
This report also described violations j
resulting from the fact that the Authorized Nuclear Inspector had not approved certain weld inspection procedures.
The overall finding concerning the structural welding was that TVA had not adhered to its commitment to conduct such welding in accord with requirements of American Welding Society ( AWS) Code Dl.1-72. ]
The Jar.uary 20 report was provided to NRC Region II which appears to have taken little or no acticn in response to the vici u':.; which N~os
~raff believad migni'icant.
January 25, 1982.
Mills (TVA) to O'Reilly. Revised Final Report on 50.55e/NCR 2654R.
"In a random inspection of 245 Aux Building duct supports inspected before March 27, 1980, approximately 22%
were found to have welds which are unacceptable.
The defects found include undersized welds, incomplete welds, slag inclusions, porosity, and overlap.
The apparent cause was f ailure to clearly specify acceptance criteria on the applicable drawings.
.. Because the subject duct support and cable tray welds were acceptable as installed, this condition would not have adversely affected the safe operation of the plant. [HOW DO THEY KNOW THE WELDS WOULD NOT HAVE AFFECTED THE SAFE OPERATION OF THE PLANT?
WHAT ABOUT WELDS OTHER THAN DUCT SUPPORTS?]
TVA has reevaluated the subject deficiency.
As a result, an alternate criteria has been established by TVA for visual inspection of fillet welds instead of the stringent requirements of AWS D1.1... "
"This NCR was revised to institute a comprehensive weld sampling program for all previously installed duct supports.
TVA has evaluated the results of this simpling program.
The results obtained from the sampling program were reviewed and it was determined taht the weld quality was acceptable to ensure structural integrity.
Therefore, the welds i
on the subject duct supports are suitable for use as-installed.
The aforementioned corrective actions meet the requirements of the AISC Specification for the design, f abrication, and erection of structural steel for buildings." NOTE: SEE 8/8/83 FOR NRC i
RESPONSE TO RESOLUTION. NOTE ALSO THAT ORIGINAL FINAL REPORT WAS SUBMITTED TO NRC 12/31/80.
4 l
March 1, 1982.
Rankin to Distribution.
Re: 2/16/82 Meeting on
{
NCR 2375R and 3579R.
Discussion re nature of problem and corrective actions.
" Sampled welds should be evaluated to relaxed criteria for fillet welds as defined in memes from Cantrell to Wilkins (2/6/81 and 10/27/81).
Actions: 1001k configuration inspection on all items involved in NCR 3579R; obtain additional information on weld quality. (WHA" WERE THE 1
I RESULTS?]
March 17,1982.
Rankin (Acting WB and SQN design project manager) to Wilkins re Conduit Supports / Filler l
welds. NOTE: INSERT FROM JN IN LIEU OF FOLLOWING. Acceptance of cable tray support fillet welds is not considered acceptance of the conduit support welds for the following reasons, etc.
March 26, 1982.
Austin to EEU inspectors.
(Informal memo, no apparent MEDS number. ] Cites Cantrell memo, 9/16/82 which authorized acceptance of all as-built cable tray support fillet welds known to have been made prior to February 6,1982 without inspections.
"Per Cantrell's memo 9-16-81" shall be written on l
card as justification.
Fillet welds known to have been made i
af ter February 6, 1982, will be inspected and dispcsitioned in am;d with g;;iccio CCP's.
March 29,1982.
Wilkins to Rankin re NCR 3579(?) a.d 2375R re Rankin's March 1 memorandum. Takes exception to Raniin having claimed NCR 2375R was limited to platforms, ladders and stairs or to fillet welds.
Wilkins said it applied to other structural steel items and fillet welds.
If scope is to be narrowed, Wilkins would insist on additional NCRs.
"We believe this will duplicate documentation of a significant condition adverse to quality which had already been reported to NRC."
April 5,1982.
Austin to inspectors cites NCR 2375.
WHERE IS THIS MEMORANDUM 7 May 5, 1982.
Standifer to Wilkins re weld sampling program per NCR's 3579 and 2375R.
Lists attributes for which welds should be examined, including cracks, porosoity lack of fusion, etc. "The results of of this sampling will serve to resolve ICR 2375R."
June 3, 1982.
MAJOR MANAGEMENT REVIEW OF WATIS BAR NUCLEAR PLANT NSRS R-82-02-WBN.
"The overall program established by OEDC to design and construct the Watts Bar Nuclear Plant in accordance with NRC requirements and TVA commitments is considered less than 1 l
i 1
adequate.... The reviews indicated that major deficiencies were present in these areas of sufficent magnitude that managemetn could not have reasonable assurance that activities have been accomplished according to the regulations and commitments."
NSRS states that it had not found "any evidence to indicate" that TVA had sought NRC approval of its having conducted its welding program in a. manner less stringent than American Welding Society (AWS) Code D1.1-72 to which TVA had committed.
NSRS also stated that it believed that TVA's relaxed interpretation of welding QA requirements at Watts Bar "has the potential of becoming a significant problem in the event the NRC does not approve l
relaxation of the AWS Code requirements."
See 11/25/82; 8/8/83.
R-82-02-WBN-24 indicates specific examples of TVA's requirements being less stringent than those implicit in AWS D1.1 regarding structural welding (cable tray supports, conduit supports, instrument tubing supports, piping supports, etc.).
The examples i
included: maximum and minimum weld size; fillet weld configuration; fillet weld undercut; weld splatter; are strikes; acceptability of inspections performed following application of l
j carbo zine primer; and the type of inspection (and related l
documentation) necessary prior to final visual examination; e.g. "..no documented inspections by certified visual examination personnel prior to final visual examination." (The findings with regard to AWS D1.1-72 should not have surprised TVA managers I
since a TVA construction manager in a memorandum (WBN 800822006) dated August 22, 1980 requested an engineering evaluation for relief frem Aws D1.1-72 which appear " overly restrictive" and have had a "gevere immet on Watts Ear _ Nuclear Plant's schedule
- .d
- n. "2 With respect to structural steel welding NSRS recommended that EN DES should provide technical justification for all of the specific AWS-D1.1 code deviations and should obtain written approval from the NRC to allow for these less stringent requirements.
R-82-02 was transmitted with a cover memo f rom H.N.
- Culver, Director, NSRS to G.H. Kimmons, Manager of Engineering Design l
and Construction.
The cover memo stated: "We believe the number and significance of many of the identified deficiencies make it necessary that an in-depth review be performed of one of the safety systems to assure TVA management that Watts Bar has been designed and constructed in accordance with the applicable requirements." WAS THIS IN-DEPTH REVIEW PERFORMED? WHAT WERE THE RESULTS?
DO THEY MEAN B&V? DID B&V LOOK AT WELDS?
June 10,1982.
Wilkins to Standifer re weld sampling program for NCR 2375.
Transmits drawings and inspection sheets for items which have been reworked and/or painted since the initial sampling.
States that certain samples were not returned because structures had been reworked or were no longer accessible.
June 16, 1982. Wadewitz (WP PM/ CONS) to Standifer (SQN and WB Design Projects Manager) re Status of Weld Sampling Program per NCR's 2375 and 3579.
June 18, 1982.
Prepared for presentation to NSRS by Pierce and Jesse: EN DES changes to design process, specifi:ations and licensing commitments made to accommodate f abrication latitude requested by CONST. Contains chronology of events leading to new design requirements and licensing changes now in progress.
WHAT LICENSING CHANGES WERE MADE?
WAS PRESENTATION AT. JALLY MADE?
WHAT WAS RESULT?
June 23, 1982. WATTS BAR NUCLEAR PLANT - INSPEC"'CN PRACTICES OF STRUCTURAL STEEL WELDS - SPECIAL REPORTS NSRS R-82-07-WBN.
"...a small number of welds had been inspected af ter being primed by carbo zinc..and no approved site procedure exis ed for such insepetion.
It was not possible to determine the number of welds that had been inspected in this manner from the records that exist at the plant."
The report states on page 2: Based on l
interviews with QC inspectors, it was concluded that 100 to 150 structural support welds had been inspected thrcugh carbo zine primer without approved procedures."
Compare this with less certain statement on page 4: "Of the 24 inspectcrs interviewed, 9 believed inspection of carbo zine primed welds vas permitted..Of the nine inspectors, four admitted to having inspected the carbo zine primed welds.
While such inspections had been made, the practice apcarently had not been extensive.
Based on these interviews it an:Jj red that only 100 to 150 welds may have been inspe:2d $ n -hi s -' m r.
- hi s could not be substantiated, however, nor was it possible to specifically de. ermine which welds were inspected in this manner. [ Emphasis added. ]
l R-82-07-WBN also found inadequate inspection dccu: mentation, a failure to issue an NCR to document welds for which inspection documentation was not available, and insuf ficie-t documentation to substantiate that a weld sampling program had verified that inspections could be conducted through carbo zf c.
Sometime during 1982, NSRS staff prepared a detailed comparison of the requirements of AWS Dl.1-72 and and G-29:. (See App.
.)
July 16,1982. BN NCR 1888.
Cover memo from Gilbert states that the NCR is considered to be a significant condition adverse to quality. Evidence exists as a result of NCR 1173 investigation that some of the 1/4 inch and larger filet welds required on features in above item description have been fcund to be undersized where visual inspections were made prior to mid-1980.
July 22,1982. OEDC (Kimmons) to NSRS (Culver). Memo concerns 82-07-WBN-02.
OEDC states that inspections through carbo zine were performed only by one engineering unit.
"~nspection records _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ _ - _ _ -
i I
f rom the Record Storage Vault signed by inspectors who stated to f
NSRS that they made inspections through carbo zinc have been examined.
There is no entry on the record that the inspections were made through carbo zinc. All records examined identify the applicable approved procedure as the inspection document.
j n erefore as previously stated we conclude that the inspections were made using an approved procedure."
Note: QTC (Finding 5, p.D-4) states that NSRS did not accept this response.
See ODC, 3
September 2,1982. )
July 23, 1982.
Kimmons to Willio.
Changes were made to ASME Section III which will permit more realistic acceptance criteria j
for welds.
These changes involve sizes and tolerances; they i
I provided a basis for acceptance of similar conditions on non-ASME welding. The latitude provided by changes significantly reduced time required to visually examine welds and allow acceptance of I
minor geometrical surf ace conditions which previously required rework and reexamination.
Code changes will result in "our not spending possible $5 to $30 million at WB.
Additionally a cost s.aving will be realized f rom future construction and m.aintenance. "
WERE THESE CHANGF.S APPROVED BY NRC7 IF SO, BY WHOM AND ON WHAT DATE?
July 29, 1982. Kimmons (OEDC) to Culver (NSRS) Discusses EN DES task force established in March 1981 to address problems a.ssociated with f abrication of welding design.
As a result, "TVA construction specifications have been revised."
SAR revisions initiated by CEB, forwarded to NEB and transmitted to Pover 7egulatory staff in Chattanooga.
"This approach was discussed by
- hone with W.C,,rier te initia, tion of the effert.
It was confirmed that this was not a new approach."
"The SAR revisions will be submitted to NRC in revision 47 expected to be released July 26,1982.
NRC will process revision 47 as they have processed all previous revisions."
NOTE: AMENDMENT 47 SUBMIT:'ED CIN 1/4/83.
WHAT TRANSPIRED IN FHONE CONVERSATION REFERRED TO ABOVE7 WAS NRC INFORMED THAT G29C AS SUBMITTED WOULD ALLOW INSPECTION THROUGH CARBO ZINC?
July 29,1982.
(EDC 82 0729 001.) Kimmons to Culver re NSRS R-82-02-WBN.
Memo responds to Culver's June 3,1982 memo re NSRS R-82-02-WBN.
"It is felt that the identified corrective action taken or to be taken will be adequate to resolve each of your specific concerns... It is our understanding that you will promptly forward to NRC through POWER Nuclear Licensing Staff both the NSRS report and our responses."
August 4, 1982.
Culver to Willis.
States NSRS support for OEDC requested change from commitment to AWS to ASME "since the requirements of the ASME Code are more appropriate to cover the structural welds."
WAS THIS CHANGE IN COMMITMENT SOUGHT FROM NRC AND WAS IT APPROVED?
IF So, WHAT ARE THE RELEVANT DATES?. _ _ _ _ _ _ _
I
~
August 16, 1982.
Culver to Willis.
GNS 820816 050.
States that TVA procedures do not require NSRS review of changes proposed to codes committees.
August 18, 1982.
See August 26, 1982 re closure of BN-W-80-08 etc.
August 20, 1982. (See also 7/29/82 entry above which appears to be same item.]
Attached to 12/22/82, Kimmons to Culver. States that EN DES task force had addressed problems. As a result TVA construction specs have been revised.
SAR revisions were initiated and sent to NEB.
"This approach was discussed by phone with NRC prior to initiation of the effort.
It was confirmed that this was not a new approach."
WHAT DOES THIS MEAN? WHO TALKED TO WHOM, WHEN, AND WHO CONFIRMED WHAT7 "The SAR revisions will be submitted to NRC in revision 47 expected to be released September 10, 1982.
NRC will process revision 47 as they have processed previous revisions." (Note: REVISION 47 SUBMITTED JANUARY 4,1983.. WHERE IS THE TF REPORT 7 ]
August 24, 1982.
Beasley to Sprouse re AWS design and 4
inspection.
Merto states that response to BN-W '80-08 committed to a welding design guide and that the guide had not been issued.
Beaseley said he had been informed by Jesse that guide was no longer necessary due to EN DES EP-4.25 being revised to cover the situation.
Beasley said if it were EN DES position that the design guide was no longer needed, then the commitment to issue it shculd be rescinded and documented.
WAS roCUMENTATICN CCMPILED?
August 26, 1982. Inspection Report 439/82-23 states that item 439/80-22-02 was closed on basis of information provided by TVA in June 22, 1981 letter to R-II.
"The report has been reviewed and determined acceptable.
In their report, the licensee concluded that had the reported deficiency gone uncorrected it would not have adversely af fected the safe operation of the plant.
The inspector discussed the item with responsible licensee representatives and reviewed the documentation to verify that the corrective actions described in the report have been completed. [WHAT DID NRC DO TO VERIFY THAT PROBLEM HAD BEEN ADEQUATELY ADDRESSED, INCLUDING DEFICIENCIES THAT MIGHT HAVE ARISEN WHEN PROCEDURES WERE INADEQUATE?
WHY 14 MONTH INTERVAL BETWEEN RECEIPT OF JUNE 22 LETI'ER AND CLOSING OF ISSUE 7]
Circa August 30, 1982.
Cantrell/Jessee " Report on the Feasibility to Reinspect Steel Structural Welds through Carbo Zinc Primer. ( Attached to September 2,1982 OEDC response to NSRS re 82-07-WBN-02. ]
Refers to walkthrough in " Unit 1 Powerhouse."
Report states " Unacceptable surf ace conditions were more readily apparent on primed surf aces than unprimed surf aces.
Spray priming ehnhanced rather than diminished the visibility of surf ace imperfections... No weld had an unacceptable condition 1
1 i
af ter removal of primer which was not detected by examination before removal of the primer."
Concludes that reinspection but not original inspection could be made through primer. WHA 7 DATA EXISTS TO SUBSTANTIATE FINDINGS?
THIS REPORT WAS PREPARED IN RESPONSE TO 82-07-WBN-02.
NOTE THAT NSRS PER 9/8/82 NSRS TO OEDC FOUND THE RESPONSE ON THIS ITEM TO BE UNACCEPTABLE. HAS NRC EXAMINED THIS DATA?
September 2, 1982.
OEDC to NSRS re R-82-07-WBN-02.
This memo contains detailed responses to 82-07-WBN-02.
DID NRC REVIEW THIS RESPONSE?
DOES NRC BELIEVE THIS RESPONSE TO BE ADEQUATE; E.G INSPECTION / REINSPECTION THRU PAINT, ADEQUACY OF RECORDS, VELDERS' MARKS, ETC. ?
Memo states that "the reported inspection of structural welds involved were confined to the work of one inspection discipline."
Memo appears to modify July 22 j
/
memorandum by stating that acceptance of inspection through carbo zine would be acceptable " assuming approval of the proposed change to the FSAR to commit to AWS as modified by G-29c rather than to AWS without modification.
If the change is not app-oved, the welds will be reinspected."
(Note: Eventual basis for accepting inspections through carbo zine was submittal of Amendment 47 on January 4, 1983, rather than approval by NRC.
See May 19, 1983 memorandum from OQA to OEDC.]
)
September 2, 1982. BN NCR 1968.
Evidence exists as a result of BN NCR 1173 and BN NRC 1888 investigations that some of the welding workmanship, on item description features, does not meet all of the inspection criteria specified in G29c.
Currently, determination of the time f rame in which these nonconforming elds zm Ne k.as not Men established but will be detemined during the course of investigation.
Apparent cause: Inadequate training of weld inspectors with respect to visual examination. [ NOTE: See entry of October 10, 1980 re examination of weld inspectors.]
September 8,1982.
NSRS to OEDC. GNS 820909 051.
NSRS says OEDC response to R-82-07 WBN -06 is not adequate.
NSRS says three viable methods to justify the position that welds could be inspected through carbo zinc: exception to code, demonstrate inspectability through carbo zinc, reinspection per ANSI N45.2 Section 11, Inspection.
NSRS states:
"From the information that has been presented to date to the NSRS, it does not appear that any of the above are being developed to support a technical basis for deviation from Code requirements."
WHAT WAS DONE TO SUPPORT A TECHNICAL BASIS FOR DEVIATION?
September 17, 1982.
OEDC to NSRS transmitting Revision 1 of
" Report on the Feasibility to Reinspect Structural Steel Weld.s Through Carbo Zinc Primer" in response to NSRS R-82-07-WBN-06.
QTC notes (Finding 11, P. D-7) that the " Report" and a subsequent revision were not accepted by NSRS or OQA.
Revision 1 ccoludes that " Visual inspection of welds through spray-applied Carbozine ___
1 1
primer with a maximum thickness of 5 mils will detect all unacepetable weld surface conditions.
Unaccpetable surface conditions were more readily apparent on primed surf aces than on unprimed survaces; therefore, spray primeing enhanced rather tahn diminished the visibility of surf ace imperfections."
IF CARBO ZINC DOES ENHANCE THE DETECTABILITY OF UNACCEPTABLE IMPERFECTIONS, WHY DOES THE CODE NOT PERMIT (OR REQUIRE) COATING WITH CARBO ZINC PRIOR TO INSPECTIONS?
WHY DID TVA ITSELF CONCLUDE THAT REINSPECTION BUT NOT ORIGINAL INSPECTION COULD BE CONDUCTED THROUGH CARBO ZINC 7 September 27, 1982. Culver to Kimmons stating that NSRS will send engineers to OEDC offices and WBN to verify corrective actions re various NSRS reports including R-82-02-WBN and R-82-07-WBN.
WHAT i
HAPPENED IN COURSE OF THESE VERIFICATION ACTIVITIES?
l October 7, 1982. Mcdonald (Quality Improvement Staff to Qual Improvement Files) re October 1 conversation with Dr. Davis, AWS.
Davis does not believe that cracks in the minimum detectable visual range (and unacceptable) would be detected through opaque primer (5 mils) rather than bridged over.
"I told him I wculd call him with feedback if TVA could justify detection.
He said he would be surprised if we could justify detection through carbo zinc."
October 12, 1982. Standifer (SQN/WB Design Projects Manager) to Wadewitz (WB CONS PM) re weld sampling per NCR 2375R.
Presents overall results of sampling program with relatively few details.
I Finds 4.5% of 14,743 inches of inspected weld to be unacceptable. (NOTE: WHAT DOES THIS MEAS?
WHAT WERE THE DETAILS
- F E
- i nC ':E ? 'TW v'fCH tEWooK WAS PEQC RED?
HCW MUCH WAS THE i
NRC INVOLVED IN THIS?)
Standifer says: " An overview of the j
welding program as a whole indicated numerous cases where welders 4
and welding inspectors were either unsure of the drawing size requirements or did not care what they were."
October 20, 1982. OQA to OEDC and NSRS. Supports finding per R-82-07-WBN-06 re insufficient documentation to establish justification for inspection through carbo zinc.
OQA notes that the "G-29c process specification provides for an inspection technique not allowed by Section 3.10 of AWS Dl.1-1972.
OQA said a further clarification was being developed and was expected to be issued by November 2, 1982.
l October 27, 1982.
Revision 2, " Report on the Feasibility to Reinspect Structural Steel Welds through Carbo Zinc Primer."
contains more detailed description of weld evaluation findings.
Revision 2 states that " structures under consideration are not subjected to nor designed for fatigue... Therefore, the ef fects of any defects not detectable by visual examination under carbozine are considered not significant to the serviceability of the subject structures."
Revision 2 sent from Kimmons to Culver on November 10, 1982. WHAT ANALYSIS WAS PERFORMED TO LEAD TO l
4 THIS JUSTIFICATION?
WHAT TVA ORGANIZATIONS REVIEWED "HIS l
ANALYSIS?
DID TVA APPROVE THIS JUSTIFICATION?
November 4, 1982.
Culver to Kimmons re R-82-24-WBN.
This is report on follow-up reviewed conducted at WBN and OEDC in Knoxville re responses to R-80-21-WBN, R-81-11-WBN, R-81-28-WBN, and R-8 2-02-WEN.
With respect to R-82-02-WBN-24, NSRS referred to findings re AWS deficiencies and stated that: EN DES has revised teh TVA Construction Specifications in these areas and has transmitted revision 42 of the WBN FSAR to the NRC for approval.
Additicanlly, EN DES is preparing a report which will include the technical justification for all the AWS D1.1 Code deviations.
This item will remain open until NSRS receives and reviews this report.
NOTE: NSRS SEEMED TO BELIEVE THAT THE FSAR REVISION HAD ALREADY BEEN SUBMI'ITED WHEN IT FACT THIS WCULD NOT OCCUR UNTIL JANUARY 4,1983 PER REVISION 47, NOT REVISICN 42 AS INDICATED BY NSRS. NOTE: REVIEW APPARENTLY DID NOT ENCCF. PASS R-82-07-WBN PER SEPTEMBER 27, 1982 MEMO TO KIMMONS FRCM CULVER.
November 10, 1982. OEDC transmits Revision 2 of Report on Inspection through carbo zine re R-82-07-WBN-06.
QTC says report was not accepted by OQA.
(See March 2, 1983 OQA to Kimmons/ Culver which lays out criticsms of OEDC report. ] [See QTC Findings 14, 17, 20, 22 and 24.]
December 15, 1982. G.H. Kimmons to H.N. Culver: " Technical Justification of Centrasts of AWS D1.1-72 and General Canatruction S. rectification." (EDC 811215 004 sic? 821215 004.)
NCfrE: THIS MEMORANDUM EX?LAINS WHY THERE WERE NO DEVIATIONS FROM CODE. REFERENCE IS MADE TO THIS MEMORANDUM IN CULVER'S AUGUST l
10, 1983 MEMORANDUM TO ANDERSON, MANAGER OF QUALITY ASSURANCE.
contrast 1 states that welds made prior to November 21, 1981 j
may be visually examined without removing primer.
The Technical Justification for "This item is being handled separately.
It is not discussed here per the request of J. A. Crittenden."
(August 1985 QTC report (P.D-9/28) refers to Contrast No.1 re inspection through paint. QTC states that technical justification for " contrast" was not discussed per request of NSRS (sic OQA?).
QTC states "The technical justification (for this deviation, ie. inspecting through p.aint) is not stated in this document or any later TVA document; refer to QTC Finding 18."
SEE ALSO i
ATTACHMENT B, p.14/14.
NOTE: MEMO ADDRESSES WELD CHARACTERISTICS AS OPPOSED TO FILLER MATERI AL, FITUP, ETC.
December 29, 1982. NSRS to OQA.
Tranfers various itens re R-82-02-WBN and R-82-07-WBN to OQA. [82-02-WBN-09 is " Lack of Control of Safety-Related Structures, Systesm, and Cocponents Li s t. "
On January 4, 1983, via FSAR Amendment 47, TVA licensing of ficials revised TVA's commitment to AWS D1.1-72 by stating E
l that TVA's commitment is in compliance with Dl.1-72 as modified by TVA's procedure G-29C. ( Att 7, p.2.; Att. 11, p.2.)
(Note: It appears that the NRC agreed to TVA's belated revision of its commitraent.
It is unclear what review, if any, was performed by NRC prior to its concurrence in the proposed change.
Although NRC may have determined that the change in the commitment vis-a-vis AWS Dl.1-72. was acceptable, NSRS staff did not agree.]
January 12, 1983.
Supports are fabricated and installed using welds which have not been inspected or documented.
Apparent Cause: The site procedure QCP-4.10-2) for inspection of these supports does not require weld inspection or documentation.
NCR form states problem was not generic, that it was unique "f abrication/ installation welds."
WHAT DOES THIS ME.AN?
"This discrepancy is unique to this Watts Bar site procedure."
January 26, 1933. Kimmons (OEDC) to Culver (NSRS) re R-82-24-WBN.
Appears not to address D1.1 issues. With respect to list of equipment important to safety, OEDC states TVA has undertaken a progrm to create a controlled single listing etc.
"We anticipate issuance of a usable quality listing by September 1, 1982(37), with final issue by January 1, 1983.
Upon final issue, all programs and work will be consistent with the controlled single listing.
This schedule will permit extensive trial use prior to fuel loading in August 1983."
January 17, 1983. haes Crittenden (OQA) to Files.
" Notes on Meeting to Discuss Deviations from AWS Code." Differences between D1.1 and G-29c: Undercut, Size of fillet welds, length of fillet welds, inprocess inspections (including fitup), documentation of welder qualification and inprocess inspections, etc., control of weld rods. NOTE: MEETING IS HELD TO DISCUSS DEVIATIONS 18 MONTHS AFTER TVA TOLD NRC ON JUFE 22, 1981 THAT THERE WERE NONE, 5 MONTHS AFTER TVA'S EXPLANATION HAD BEEN ACCEPTED BY NRC ON AUGUST 26,1982, AND 3 WEEKS AFTER TVA HAD REVISED ITS COMMITMENT TO AWS Dl.1.
HOW DID CRITIENDEN ET AL. RESOLVE THE QUESTIONS?
March 2,1983.
OQA to NSPS and OEDC re Finding R-82-07-WBN-06, re the adequacy of documentation to support inspection through carbo zine. OQA does not concur in OEDC report.
OQA specified deficiencies in the OEDC report; e.g. inadequate documentation of observations of carbo zine thickness, inadequate demonstration of sampbles containing a sulficient number of defects, inadequate documentation of weld defects prior to application of primer, f ailure to control prior knowledge and qualifications of test i.nspectors, failure to require retention of test specimens, and f allure to require approved controls and procedures vis-a-vis the test.
... the evaluation of powerhouse welds did not appear to generate pertinent data to support the report's conclusions."
ogA said "The finding remains open.
OQA plans to follow OEDC's _ _ - - - _ -
actions and to close the finding if OEDC satisfactorily conducts an inspection procedure qualification which demonstrates that critical size and type defects are observed."
NOTE: THIS IS TWO MONTHS AFTER SUBMISSION OF AMENDMENT 47 AND OQA CONTINUES TO BELIEVE THAT CARBO ZINC ISSUE HAS NOT BEEN RESOLVED.
April 11,1983. 83-05 refers to verifying fitup etc. of safety related structures and supports outside containment.
States applicable work is done pursuant to ASME B&PV code when in f act structures are via Dl.1.
Furthermore, fitup inspections are not done.
IS THIS CONSISTENT WITH PREVIOUS FINDINGS RE FITUP ETC?
May 6, 1983. Anderson to Kimmons (OQA 830506 0027), apparently accepts December 15, 1982 Technical Justification of Contrasts of AWS Dl.1 and G-29c. NOTE: OQA NOW APPEARS TO ACCEPT CLOSEOUT OF CARBO ZINC NOTWITHSTANDING ITS PREVIOUS POSITION STATED ON MARCH 2,
1983 THAT THE CARBO ZINC JUSTIFICATION REMAINED INADEQUATE.
j
[QTC notes in Finding 18, that the December 15 " Technical Justification" does not contain a justification for the code deviation involving carbo zinc. )
May 19, 1983.
Anderson to Kimmons re R-82-07-WBN.
Accepts closecut of R-82-07-WBN-02.
This item concerned inspection through carbo zinc.
Basis for closing it out was submittal to NRC rather than NRC acceptance.
TO WHAT EXTENT HAS NSRS OR NRC INQUIRED INTO ADEQUACY OF CLOSEOUT OF THESE ITEMS?
June 1, 1983.
Mills (TVA Licensing Manger) to O'Reilly re c a.,x ut of F;R 2F 57 M c her NCR's (e.g. 2111R) including fillet welds on socket weld fittings (e.g. 34,000 inspected, 11,500 repaired). 151 welds were " identified" as inaccessible. These were referred to EN DES for disposition.
[WHAT WAS THE RESULT?)
"A large number of welds on small piping in safety related systems have insuf ficient weld metal buildup which might af fect the strength of the weld.
Therefore, if this condition had gone uncorrected, leaks or small pipe breaks could have occurred (especially during a seismic event) which may have adversely affected the safe operation of the plant." (NOTE: WELDS HAVING PASSED INSPECTION WERE LATER FOUND UNACCEPTABLE AS RESULT IN AMBIGUITIES IN INSPECTION PROCEDURES AND FAILURES TO FOLLOW INSEPCTION PROCEDURES. ]
June 20,1983. WB NCR 4573.
Quality of welding [ structural steel in main stedm valve rooms) is not in strict compliance with drawing and specification requirements.
This structural steel has minor discrepancies which deal with joint and weld configurations.
Welding was previously accepted but not inspected with strict adherence to visual inspection requirements of G-29c. Root cause: Prior to 1-1-80 strict adherence to weld inspection criteria and a lack of knowledge about commitments to drawing configuration yielded a breakdown in quality.
NOTE: WHY WERE WELDS ACCEPTED ON THE BASIS OF MORE STRINGENT CRITERIA THAN
-HC:SE LATER USED AS BASIS FOR REJECTION.
NCR was resolved on basis of criteria used in resolving NCR 2375.
NOTE: IT SEEMEr ONE REASON FOR THE 2375 CRITERIA WAS THAT THE WELDS INVOLVED WERE ON RELATIVELY UNCRITICAL ITEMS.
WHAT ANALYSIS HAS BEEN DONE BY NRC7 June 21, 1983. Kimmons (OEDC) to Anderson (OQA): " Watts Bar Nuclear Plant -
Resolution of NSRS Recommendation R-8 2-07-WBN-06" states : The qualification tests substantiated that weld size etc. but not quality. "OEDC has not interpreted G-29C as permitting acceptance of weld quality based on an inspection through primer."
"The qualification tests did not su.bstantiate a practical method of visual inspection of weld quality through primer in a construction environment." (See Att 7.)
NOTE: CEDC APPEARS TO BELIEVE THAT ABILITY TO INSPECT TEROUGH CARBO ZINC HAS NOT BEEN ESTABLISHED AND THAT SUCH INSPECTIONS ARE NOT PERMITTED BY G-29C.
July 19, 1983. Culver to Willis re NSRS Key Topics Report (July 7,
1983, GNS 830707 055).
"NSRS has been and continues to be concerned about the welding practices at WBN and the managment practices that permit them to exist.
We have identified a number of specific problems that both NSRS and OQA have been working to resolve."
July 27, 1983.
Anderson (QA) to Kimmons (EN DES).
(QTC Ref. 34.]
OQA response to R-82-02-WBN-24; i.e. OQA notes its acceptance of closure of this item, one element of which c m--ed i nspection through carbo rinc. [ Note: Again, reference is made to the December 15 Technical Just.i.icaden maorandum i
wl_ich did not address the carbo zine issue. )
In July 1983, NSRS staf f prepared a draf t memorandum to TVA's manager of Quality Assurance stating that some of the deviations from AWS Dl.1-72 had not been adequately justified by the TVA officials responsible for doing so or for making sure that the appropriate corrective action was taken. One draf t stated: "Our recommendation was (referring to the NSRS review) that these deviations be identified, justified and approved by the NRC.
It appears that ENDES (Department of Engineering Design and Construction) is trying to hide these deviations from the NRC."
August 3,1983.
CEDC to NESB Files regarding R-82-07-WBN-06 stating, in ef fect, that configuration but not quality could be inspected af ter painting. " Aspects of weld qualtiy such as cracks and porosity must be evaluated prior to application of primer."
OEDC was pessimistic as to prospects for establishing inspectability through primer.
August 8,1983.
Verelli (R-II) to TVA. Re NCR 2654R; le.221L of 245 Aux Building duct supports were found to have unacceptable welds. "The defect found include undersized welds, incomplete welds, slag inclusions, porosity, and overlap.
TVA has reevalauted the subject dcficiency.
As a result, an alternate criteria has been established by TVA for the visual inspection of fillet welds instead of the requirements of AVS D1.1.... The aforementioned corrective actions meet the requirements of the AISC Specification for the design, f abrication and erection of structural steel for building.
The Licensee stated that the FSAR had been changed to reflect the change in inspection criteria for HVAC supports."
DID NRC REVIEW THE til'W INSPECTION CRITERIA?
WHERE IS SUCH REVIEW DOCUMENTED?
SEE 1/25/82.
August 10, 1983. This is the final version of the foregoing memorandum (GNS 830811 050).
It addresses carbo zinc, verification of weld filler material, and inspection records.
NSRS says that some of the OEDC technical justifications (cited in the December 15, 1982 memorandum) are not adequate, and deserve further consideration.
NSRS pointed to an inconsistency
{
between the acceptance criteria which required welds to be free j
of cracks and the inspection procedure which allowed inspection through paint.
NSRS stated: "We recognize that the general problem of the application of carbo-zinc prior to weld inspection is being handled by OQA as a separate issue and may have influenced the decision to close this item."
NSRS said the justification for lack of records to trace weld filler material appears to be inadequate. "..a program to determine the chemical and physical properties of a representative sample of the welds may be necessary to establish the acceptability of the filler ma t e rial. "
.. documentation verifying that inspections had been pc. n...ed and fr.2 ;; C ts of the ingection are not available for j
welds prior to July 1982... neither the EN EES reply nor the OQA i
closure file states where the requirements are contained in the TVA QA Program relating to records dealing with fitup, inprocess inspection, or final inspection.
it is net clear even now whether it is intended that the weld inspection program be governed by the weld program through G-29C (peer inspection) or through the quality verifying program (QC inspection).
Regardless of the applicable program, the line organriation was unable to demonstrate that weld insepetion records existed during the NSRS review.
Since documentation verifying that inspections had been performed and the results of the inspectiorn are not avialable for welds performed prior to July 1982, it may be necessary to inspect or reinspect a representative sample of the welds to verify the acceptability of the inspection program and to provide appropriate records of the inspections."
[ NOTE: ASSUMES AWAY PRE-1902 PROBLEMS; NO OBJECTIVE EVIDENCE TO BACKUP THIS ASSUMPTION. NSRS MEMORANDUM NEVER ADEQUATELY ANSWERED. ]
August 24, 1983.
Anderson (OQA) to Kimmons (ENDES) says all required inspections for quality cannot be accomplished via inspections through primer.
The WB site has not identified any welds inspected through paint.
OQA appears to accept closeout of carbo-zine issue.
NOTE: AFTER TWO YEARS OF DISCUSSION AS TO INSPECTABILITY THROUGH PRIMER AND RATIONALIZATIONS FOR ALLOWANCE OF SUCH INSPECTABILITY, OQA ACCEPTS THE POSITION THAT NO SUCH INSPECTIONS OCCURRED BECAUSE THEY HAVE NOT BEEN DOCUMENTED BY OEDC.
WHAT ABOUT 150 WELDS RECALLED BY VARIOUS INSPECTORS? [SEE ENTRY DATED SEPTEMBER 2, 1982.]
WHERE DOES NRC, NSRS, AND OTC STAND ON THIS?
WHAT WAS THE BASIS FOR NSRS FINDINGS THAT " FINAL VISUAL INSPECTION [ OCCURRED) AFTER CARBO ZINC PRIMER HAD BEEN APPLIED 7" October 6, 1983.
Smith to Culver re Differing Opion on the Resolution of the AWS D1.1 Finding.
October 13, 1983. Dilworth (Asst GM) to Anderson (Dir/OQA) says resolution of AWS problem dragging on too long.
[ Note: Dilworth is concerned that AWS problem has not been resolved, when TVA told NRC on June 22, 1981 that it had been resolved; NRC had accepted TVA's resolution on August 26, 1982; and TVA submitted Amendment 47 on January 4,1983.]
October 20, 1983.
TVA OQA Hanager Anderson wrote a memorandum to the Director of the NSRS Culver stating that he considered the carbo zine matter closed via closure of R-82-07-WBN-06.
"OEDC has not documented any inspections through carbozine primer.
The welds which were inspected through carbozine primer were identified by the responsible inspectors and were reinspected and documented af ter the carbotinc primer was removed."
DOES QTC AGREE WITH THIS?
"It is our opinion that the welding program as outlined in G-23c yields a paality pr:0uct if properly implemented;" and that "Our licensing commitment has been properly modified through the FSAR amendment."
Note Item 2C: "The FSAR for Watts Bar was revised by amendment 47 dated January 4,1983 to state that our commitment is [ sic] compliance l
l with AWS D.1.1-72 as modified by G-29C.
To date, the NRC has l
posed no questions in this area.
We believe that questions would I
have been raised by now if there were any to be raised." Iteen 3B: "The records of the final weld inspections are contained in the appropriate inspection procedure for the various components.
CONST at WBN uses one procedure to perform and accept the weld inspection and numerous other procedures to document the results."
[ NOTE: Anderson memorandum ignores NSRS August 10 memorandum pointing to inconsistency between regulations prohibiting cracks and allcuing inspections through paint which would not detect cracks. )
October 20, 1983.
Kimmons to Willis.
Attaches summary of events.
" Note that the ' bottom line' is that NRC and NSRS have closed the structural welding issue (page 4 of summary)."
IS IT A FACT THAT NSRS AND NRC HAD CLOSED THE ISSUE AS OF THIS DATE7 First item re 6/11/80 NCR 2375R and 6/18/82 Pierce / Jesse briefing paper both of which stated reinspection showing that previously inspected welds had not met requirements f or visual examination i._ _ - _ _ -
l 1
was changed in the 10/20/83 chronology to may not meet requirements for visual examination.
With respect to 9/09/81 entry: "Following conversations with cognizant NRC personnel in Atlanta and Bethesda, TVA initiated changes by licensing transmittal 327 to Regulatory staff in Chattanooga.
Chance was submitted to NRC in revision 47 to SAR.
Attachment Sc shows present SAR with Rev. 47. Note: these same changes in commitments have also been made on FBNP and BLNP: and, to date, no questions or comments have been received from NRC." [ Note: Underlined portion presumably refers to contemporaneous (e.g. 10/83) understanding.
On page 4 of summary, re 10/7/83: NRC inspectors performed evaluation of welding within scope of nonconformances l
NCR 2375R and NCR 2111R and NSRS Finding R-82--02-WBN-24 and closed the items on October 7, 1983.
NCR 2375R and NCR 2111R were closed by OEDC on April 22, 1983 and April 18, 1983 respectively.]
October 25, 1983 (circa).
Handwritten notes re inspectors recollections as to inspection of welds through paint. File CS 11.1.
October 28, 1983.
OQA/NSRS meeting re R-82-02-WBN-24 which found that structural welding had not been conducted in accord with all AWS D1.1 requirements.
OQA meeting memo (dated 11/18/83) stated OQA would develop a f actual history of OEDC actions to resolve problems with integrity of welds; OQA will evaluate adequacy of corrective actions. WHAT EAPPENED?
WHERE IS THE OQA FCLLCWUP?
NSRS mceting memo (dated 11/28/83) stated "The g ua y, _ c c f " :
- eting was to discuss the resolution of safety issues relating to the inspection of welds through carbozine and maintenance of weld inspection records.... On the issue of inspection of welds through carbozine primer it was decided that further study was warranted including a determination of whether a sampling program had been conducted and documented." NOTE
- QTC SAYS OQA EAD CLOSED THE ITE!d ON 5/6/83 ; AND NSRS WAS NOT ON THE DISTRIBUTION LIST OF THE OQA MEMORANDUM.
November 28, 1983. Jones memo to files re October 28, 1983 meeting.
December 20, 1983.
QTC (p.D-19/28) states that OQA/SEB memo to NSRS contained a draf t evaluation for review prior to a meeting scheduled for 12/21/83.
Refers to meeting to discuss NSRS agreement and disagreement re resolution of R-82-07-WBN etc. The overall evaluation section of the memo states: "EN DES evaluated the acceptance criteria specified in G-29c and determined that changes were justified based on current industry standards and practices.
The TVA acceptance criteria in G-29c were then revised.
These changes were submitted as changes to TVA's commitments to the NRC."
NOTE: QTC NOTES THAT NONE OF THE ATTACHMENTS TO THE OQA/SEB MEMO, INCLUDING THE EN DES CHANGES IN.
ACCEPTANCE CRITERI A IDDTIFY INSPECTION THROUGH CARBO ZINC AS AN EXCEPTION TO AWS OR CHANGE IN COMMITMENT.
January 12, 1984.
TVA meets with R-II re adequacy of 3-29c. R-II memorandum of meeting [see January 23, 1984 below) states that purpose was to discuss the adequacy of TVA's weld filler material control program. "It is to be noted that ASME Class 1 pipe welding (e.g. Reactor Coolant Piping) has traceability as to heat and lot number on the weld joint control documentation:."
NOTE: TVA STATES THIS NO' WITHSTANDING THAT TRACEABILI':Y WAS NOT MAINTAINED.
At this meeting, NRC " questioned the nature of the j
TVA commitment contained in the FSAR, and informally recommended j
that specific exceptions to AWS Dl.1-72-1972 be docunented."
WAS THIS DONE? WITH THIS QUESTION OUTSTANDING, WHAT IS THE BASIS FOR CCNCLUDING THE NRC ACCEPTED THE TVA EXPLANATION?
January 16, 1984.
John R. Lyons, Systems Engineering Branch (SEB) to SEB Files describing January 12 meeting with NRO R-II.
"The NRC asked several questions regarding the TVA program which we agreed to pursue and to evaluate the need for prog:am improvements.
These questions are contained in Attachment 2."
ATTACHMENT 2 WAS NOT INCLUDED IN PACKET PROVIDC THE NRC AS ATTACHMENTS TO A MARCH 28, 19.85 LE' ITER PREPARED BY TVA FOLLOWING A MARCH 25, 1985 MEETING WITH NRR IN BETHESDA.
ATTAC9ENT 2 CCNTAINS SEVERAL QUESTIONS AS TO BASIS FOR CONCLUDING TEAT WELD ROD NEED NOT BE TRACEA3LE TO SPECIFIC WELDS.
THE RECCRD DOES NOT INDICATE THAT THESE QUESTIONS WERE ANSWERED.
NOTE: WEII E BRIEFING CHARTS PREPARED FOR MEETING INDICATE CARBO Z:NC WOULD BE
-w e ru m, yr"cRANDA D7SrRIBING MEETING DO NOT REFER "O CARBO 5INC AS HAVING 5iEEN' D15CiSS5D CR BAlbG' bid OF THE ISSJES THAT HAD
~
BEEN THE FOCUS OF CONTECVERSY.
QTC BELIEVES IT WAS D*SCUSSED WITH R-II; SEE QTC FINDING 17, P.
D-21.
SEE ALSO ENTRY BELOW, MARCH 25, 19 8 5.
January 17, 1984. OQA to OEDC recommending deleting provision allowing inspection through paint, and modifying FSAR to identify specific exceptions taken to D1.1.
QTC findings state these recommendations "are a direct result of the meeting with USNRC Region II on 1-12-84.."
January 18, 1984.
Briefing of TVA Board.
Among issues raised were Filler Material records, inspection records, inspection through paint, and overall acceptability of AWS welds. "A concern was expressed that, if the NRC-NRR did an in-depth evaluation of the exceptions to this end, they might have significant questions that could seriously affect the licensing of the Watts Bar nuclear plant. To satisfy this concern, a detailed ex:eption will be presented to the NRC-NRR."
At this meeting NSRS agreed with the TVA Of fice of Quality Assurance that the probles had been resolved.
Written notes of the meeting state: "The Board concluded that TVA should present the NRC-NRR (NRC-Of fice of Reactor Regulation) with a detailed description of the (welding) -
progra= changes and should obtain their concurrence." An added note says: "The Scard did not feel it was necessary to obtain this concurrence in writing."
(One neeting participant disagrees with the meeting notes indicating that the Board did not think it necessary to obtain the NRC's concurrence in writing; this person recalls that the Board stated that the detailed description should be presented in writing and that a written concurrence should be obtained from the NRC.]
The meeting notes state that a report was to be prepared for the TVA Board describing the NRC response to TVA's presentation of a detailed description of and justification for its welding program changes. WHERE IS THE WRITTEN REPORT REQUESTED BY THE BOARD?
An attendee at the January 18 meeting who believed the case had not been made that the AWS D1.1-72 problem could be considered resolved later informed investigators that he heard nothing at the meeting that allayed his concern about the unilateral de-concit:nent to AWS D1.1-72.
This fact and the lack of a
documented program verifying work already done at the site was at the heart of his concern.
During the January 18 mee ting, TVA QA officials stated their acceptance of the resolution of the problem involving weld inspections made following application of paint primer. WHAT IS RESPONSE TO OQA/ODC.
WHAT IS ANSWER TO THEIR CON ~ENTION RE PAINT PRIMER?
January 18, 1984.
GM to GM files: describes meeting with Board. The memo indicated that the Board had been told that the
" filler metal records" and " inspection records" issues were presented in terms of "TVA commitments to the NRC", but the "carbe zine issue" was not presented in terms of what the commitment might be. (From QTC p.D-23/28.]
January 19, 1984.
OQA to TVA/GM: OQA and NSRS jointly concluded that the AWS welding program for WBN satisfies regulatory requirements and TVA commitments to the NRC and provides adequate confidence in the integrity of welds made under that program. It is OQA's position that these longstanding concerns have been satisfactorily resolved."
WHAT ABOUT COMMITMENT TO D1.1 VIA FSAR
- 17. 2 COMMITMENT TO N4 5. 2. 5.
January 20, 1984. OQA to SEB files contains statements about the January 18 TVA Board briefing which are different from the GM's memorandum on the meeting. This memorandum said the Board had told the staff to present NRC a description of the weld program change but not a comparison of D1.1 and G-29c, and that it was i
not necessary to obtain NRC's written concurrence.
The GM's memorandu:n had a detailed exception to AWS D1.1 with technical justification, to the NRC-NRR and deterJJine if they have any concerns regarding the TVA program.
January 23,1984. G-29c is changed to delete provisions permitting inspection through carbo zinc. - _ _ _ _ _ - _ _ _ _ _ _ _ _ -
January 23, 1984.
To TVA from R-II "It is our opinion that this (January 12) meeting was beneficial.
It provided a clear i
understanding of the Watts Bar welding filler material
]
program." QTC notes that this letter does not acknowledge any presentation or discussion regarding the carbo sinc issue or inspection of welds through primer as an exception to the AWS Dl.1 code.
January 30, 1984. From TVA OQA to WA/NSRS. Documents closure of NSRS concerns about AWS welding program.
LISTS SEVERAL DOCUMENTS.
DOES NRC HAVE THEM7 NOTE: QTC states: " Review of this memorandum resulted in more unresolved questions than answers, as to the basis for TVA closure of the carbo zine issue." _ Attachments B and C of the QTC carbo zine report are annotated versions of the memo and questions directed thereto.
February 3,1984. TVA/NSRS to TVA/OQA.
NSRS confirms that its AWS concerns have been resolved and closed.
QTC comments q
(p.D-27/28) are to the effect that documents cited in NSRS memo do not support resolution of the NSRS concerns.
February 10, 1984.
TVA's February 17 memorandum stated that TVA provided the background regarding changes made to the AWS code.
According to TVA, NRC indicated it had an understanding of the TVA welding program re G Arements as defined in the General Specs.
TVA said that NFC said the designer is allowed by the Code to interpret requirements, and the NRC " consider (s) that TVA has not modified a commitment but h'. : - rely provided in the G-Spec implementing instructions as permitted by the Code.
Consequently, the NRC has indicated no apparent concern regarding what TVA has done in clarifying requirements for the welding program at WBN.
...NRR representatives indicated.." they had been in discussion with IEE in this matter, and as a result, they had no specific.
concerns regarding the TVA welding program at WBN. They indicated there was no need for TVA to provide additional' information regarding the welding program requirements."
NRC, in its March 1 memorandum describing the meeting, says it was told "there was no technical difference between G-29c and the
... the NRC staff told TVA requirements established by the AISC.
that, after a cursory review of GCS G-29c, and the presentation made by TVA, the staff had no-concerns with regard to TVA's commitment to AWS Dl.1 as it is clarified by G-29c." The NRC said they would provide TVA with a summary of the meeting that will reflect these general conclusions and that the summary will be placed in the public Document Room.
NOTE: TVA DID NOT PROVIDE NRC DETAILED COMPARISONS OF G-29C AND AWS Dl.l.
NOR DID TVA INFORM NRC AS TO EXTENT OF PROBLEMS RAISED BY NSRS.
NRC PREPARED ONLY A BRIEF
SUMMARY
WHICH WAS NOT PLACED IN THE PDR. __
Id. With respect to foregoing meeting, QTC notes (p.D-28/28) that:
(a) GCS-29c was changed on 1-23-84, just prior to meeting to delete references to carbo zine; (b) only a " cursory review" of GCS G-29c was performed.
(c)
No indication that " detailed exception to the AWS Dl.1, with technical justification" was presented to NRC; this as the " Action Item" from the 1-18-84 TVA Board meeting.
March 1, 1984.
NRC summary of foregoing meeting; e.g. memo j
referring to a " cursory review.
August 27,1984. vol 30 - No. 35. p.
A-3.
Commonwealth Edison. S&L commits to D1.1.
D1.1 requires that inspector shall examine work in a way that includes fitups.
Contrary to the above, Napolean Steel Contractors QA program did not require i
fitup inspections for safety related structural steel; quality I
I control personnel did not inspect and assure acceptable gaps for weld joints.
C-E found Napolean did not have documentation for preheat, interpass, temperature, position, etc.
October 2,1984. Investigators were shown notes (File CS A22) where one inspector writes to another: FSAR holds to AWS except as modified by G-29c; G-29c holds to AISC; AISC holds to fitup by inspector. "Now What?"
October 29, 1984.
ENP NCR 3615 states " Reinspection of several structural items in which previous inspections had been finalized, revealed that portions of the installed items do not meet the requirements of BNP-QCP-2.15 and EN DES drawings.
Refer to the attachments for the discrepancies found and drawing numbers.
73% of the total items reinpsected f ailed to meet the requirements.
[ November 27, 1984 version of NCR 3615 states 54.8% of items inspected f ailed to meet requirements. ] (Refer to NCRt 1888 and NCRil968 for welding discrepancies previously identified.) Note: these reinspection are general and not all inclusive. Apparent cause: Without further investigations, the cause cannot be accurately determined, however, it appears to be j
inattention to detail.
November 21, 1984.
First interim report on 50.55e re WB substandard welds on RB conduit supports."There is no reason that these conduit support welds (on 55 supports) had previously been l
inspected or accepted.
Due to the number of conduits on these supports, the support welds are inaccessible to rework.
December 19, 1984. NOV Level V.
Undersize weld on pipe support resulting from f ailure to follow QCP 4.23-4. "This was due to an inspector oversight in assuring that the correct weld size was
' i
This hanger was finalized October 16, 1981.
perf ormed.
Sisce that time hanger QC inspectors have been reinstructed on a recurring basis in the acceptance criteria QCP-4.23-4 and other applicable procedures and specifications."
NOTE: QCP 4.8 DCLWATION WAS DISCARDED.
WHY CONSIDER THIS AN ISOLATED DETICIENCY? WHAT ABOUT INPO FINDINGS?
I March 11, 19 85. 4 3 9 / 85-06. Finds inspections being done to
)
requirements of 0.C.1.l(RO) when requirements mandate inspecti::ns dccie to requirements of 1.C.1.2(R2). Inspectors were " directed' to inspect AWS welds to 0.C.1.1. (RO) which allows substitution of ASME criteria for AISC ( AWS??) welds.
This " direction" to ir:.spect to O.C.1.1(RO) " mandates automatic non-conformance to BN:P-QCP-7.5 (which is the procedure provided to QC inspectors for visual examination of welds. ) This is unresolved Item 438. (WEERE DOES IT STAND?)
Ms_rch 28, 1985. Darling to Denton re AWS Weld Program.
Darling notes that during the March 25, 1985 meeting in Washington that, "You (Denton) raised a concern of whether or not..(NRR).. was aware of the employee concern of this issue when TVA discussed the subject with NRR in 1984."
Darling implies that all issues raised by the employee had been discussed with Region II on January 12, 1984, after which " Region II representatives indicated their full satisf action with TVA's program and that they were satisfied with disposition of the issues as proposed."
IO. NOTE: WHILE R-II BRIEFING CHARTS PREPARED FOR JANUARY 12,
)
1984 MTG WITH R-II INDICATE THAT CARBO ZINC WOULD BE DISCUSSEE, RECORDS OF JANUARY 12 MEETING DO NOT INDICATE THAT CARBO ZINC WAS Di FACT DISCUSSED NOR THE NATURE OF ANY SUCH DISCUSSION.
NOR DC T'EE NRC's RECORDS OF THAT MEETING INDICATE TRAT: " Region II representatives expressed their full satisf ac.an with TVA's program and that they were satisfied with disposition of the issues as proposed."
The NRC said in its Enclosure 1 of its letter to TVA describing the meeting that: "The NRC made several l
comments concerning ways to possibly better control the existing i
program but did not find any regulatory inadequacies in the
(
program as described by TVA.
The NRC did concur that the weld l
filler material program described as being in place at Watts Bar would be acceptable in the AWS Dl.1 Welding program."
The NRC's qualification, "as described by TVA", raises question as to whether NRC would have expressed satisfaction with the program l
and disposition of issues had Region II known the full circumstances of the controversy.
NOTE: DARLING'S MARCH 28, 1985 LETTER INCLUDES AS ENCLOSURE 2 LYONS' JANUARY 16, 1984 MEMO RE JANUARY 12, 1984 MTG WITH R-II; ATTACHMENT 2 TO DARLING'S Mnm WAS A LIST OF NRC QUESTIONS AND OBSERVATIONS CONCERNING, AMONG OTHER THINGS, WELD FILLER MATERI AL. THE FACT THAT R-II EAD S;UCH QUESTIONS SUGGESTS THAT R-II WAS NOT AS FULLY SATISFIED WI""H T'VA' S PROGRAM AS WAS IMPLIED BY DARLING. _ _ _ - _ _ _
a Id. In his March 28 letter, Darling said 1984 meetings with R-II and NRR "were held to ensure that all issues were clearly understood and resolved."
This statement appears in conflict f
with R-II's residual questions re filler material (see Attachment 2 to Lyons' memo) and the failure of the record to show what discussion occurred, if any, re carbo zinc.
March 28, 1985. Novak to Parris re resolution of employee's concerns re G-29c. Novak asks whether employee was satisfied with resolution of issue, whether employee was concerned with adequacy of G-29c as applied to D1.1 or only concerned that staff may not be aware of its existence, and whether TVA was aware of differing opinions today regarding adequacy of G-29c as applied to D1.1.
April 9,1985. TVA (R.H. Shell) letter to NRC (Adensam) responding to March 28, 1985 Novak to Parris re resolution of G-29c issues. TVA states that TVA management had believed employee concerns re G-29c had been satisfied "as a result of TVA actions on the matter and the February 10, 1984 meeting and resultant summary dated March 1,1984.
However, discussions with the employee on April 2 and 3,1985, revealed that certain j
concerns still exist as detailed... below."
NOTE: EMPLOYEE WAS f
NOT INVITED TO FEBRUARY 10 MEETING AND 'WOULD HAVE HAD NO WAY OF f
KNOWING WHAT TRANSPIRED EXCEPT FOR MARCH 1 MEMORANDUM WHICH CONTAINED FEW DETAILS. [SEE MARCH 1, 1984.]
Id.
... all other (i.e. except for concerned employee) TVA personnel involved in the resolution / disposition of the NSRS open items were satisfied with such resoultion before the February 10, 1984 meeting with NRR.... In summary, TVA believes that the employee's concerns regarding the adequacy of the AWS D1.1 weldign program at WBN have been addressed and satisf actorily resolved. To gain further assurance that they have been resolved, these concerns will be reevaluated and discussed with the concerned employee."
May 9,1985. Walker (NRC R-II) to Thompson (NRR) stating that 4
Watts Bar "may be ready for fuel load around the end of May 1985" assuming resolution of readily addressed open items, none of which involved welding.
May 16, 1985. Thompson (NRR) to Parris transmitting a series of questions primarily concerning Watts Bar. Item 7, Enclosure 2 requested that TVA verify its compliance with structural steel welding requirements and that it provide documents " indicating problems with TVA's AWS welding program not previously provided to NRC."
NOTE: SEE JUNE 5, 1985 ENTRY FOR DISCUSSION OF TVA RESPONSE.
June 5, 1985.
TVA response to Thompson (NRR) May 16, 1985 request for information.
"Where provisions of ASME Section III, Division 1, Subsection NF are considered more appropriate they _ _ - _ _ _ _ _. _ _ _ _ _ _
RECEIVED ALLEGATIONS THAT IT CONTAINED NUMEROUS INACCURACIES AND FALSE STATEMENTS.
QTC INORMED W.F. WILLIS 08 OCTOBER 22, 1985 i
THAT IT HAD SUBSTANTIATED THESE ALLEGATIONS PER INVESTIGATIONS PH-85-039-002 AND OG-85-037-004.
August 13, 1985.
OTC submits carbo zine report to W.F. Willis, QTC stated: "In conclusion, it is felt that TVA General Manager.
the closure of the Carbo Zine issue was premature, and that the documentation to support this closure does not provide adequate technical justification and corrective action.
Without the proper closure of the Carbo Zinc issue, structural hardware, as it is related to this concern, is indeterminate."
QTC listed questions raised by existing documentation.
28, 1985.
W.F. Willis to F.W. Whitt requests NSRS August to respond to questions posed in QTC Carbo Zine report.
September 27, 1985.
Whitt to Willis forwarding NSRS respense to QTC Carbo Zine report.
Whitt said: "We believe the responses to l
questions raised in NS-85-001-001 contained in this memorandum
)
resovie any concern of premature closure and inadequate justification for closure of the issue of visual inspction of I
The sampling plan structural welds through carbo zine primer.
being provided by separate correspondence will resolve any reasonable questions concerning the quality of the structural l
Attached report stated on page 5: "In no l
documented case were welds initially inspected and accepted af ter the primer was applied."
"The primer was not removed from some of the welds being reinspected for configuration - an apparent or alleged violation of procedures." NOTE: THE NSRS REPORT IN EFFECT, THAT THE CARBO ZINC DISPUTE INVOLVED 150 TO
- CLAIMED, 200 WILDS THAT HAD BEEN REINSPECTED THROUGH CARBO ZINC AND THAT THERE WAS NO EVIDENCE THAT PRODUCTION WELDS WERE INSPECTED THROUGH PRIMER FOR INITI AL WELD ACCEPTANCE.
DID NSRS SEEK QTC COMMENTS ON NSRS REVIEW OF QTC CARBO ZINC REPORT PRIOR TO FORWARDING NSRS REVIEW TO WILLIS?
DID TVA SEEK QTC COMMENTS ON NSRS REVIEW OF QTC CARBO ZINC REPORT PRIOR TO SENDING THE NSRS REVIEW TO THE NRC7 October 22, 1985.
QTC to Willis forwarding report on inquiry into allegations concerning July 25, 1985 NSRS I-85-14-NPS re i
adequacy of TVA's weld inspection program.
NOTE: SEE ENTRY FOR JULY 25, 1985.
November 1, 1985.
Claude Key to W.F. Willis.
Key says statements in the September 27 NSRS response to the QTC carbo zine report were "not true".
Key said that previous NSRS inquiries into the_ carbo zine issue had revealed that welds other than those subject to reinspection "were inspected through primer for initial weld acceptance.
(NSRS) reviewers also indicated l
that WBN supervisors had informed inspectors in their groups that initial weld acceptance inspections could be performed through 1
_----_ _ _ U
l
.*.a o.
Primer."
NOTE: ON WHAT DATE DID TVA PROVIDE THE NRC KEY'S
. NOVEMBER 1, 1985 LETTER TO WILLIS?-
November 20, 1985.
Knoxville Journal carries story based on the November i Key letter.
TVA "is trying to resovie lingering problems at Watts Bar Nuclear Plant by discrediting some of its own employees, an engineer has charged in a letter to General Manager Bill Willis."
l 1
I J
1 l
l 1 j l
l I