ML20198D565
Text
1. Allegation Category: QA/QC 6, QC Inspection
- 2. Allegation Number: AQ-55 and AQ-78
- 3. Characterization: It is alleged that fuel transfer canal liner documen-tation was falsified, that required weld radiography was not completed, and that hold points on inspection travelers for the fuel building were signed off improperly.
- 4. Assessment of Safety Significance: The NRC Technical Review Team (TRT) learned that the liner consists of a series of 3/16-inch-thick stainless steel plates welded together and attached to the inside of a concrete structure within the fuel building. Transfer tubes at either end of the canal connect to similar fuel pools inside the Unit I and 2 containment buildings, which are also referred to as reactor cavities. The method of construction required that all vertical liner plates first be preas-sembled; concrete was then poured around the outside of the plates. The plates also had studs welded to the back and embedded in the concrete to hold the plates in position.
The plates were first joined to a backup bar on the back side with a 1/4-inch spacer temporarily attached between the plates to maintain a weld gap. A channel was then welded over the backup bar to form the leak chase cavity, and concrete was poured around the outside of the liner. Follow-ing concrete backfill, the back side became inaccessible. The final (wa-ter side) seam was machine welded after remov'al of the 1/4-inch spacer.
Plates that did not maintain a minimum gap of 3/16-inch required grinding out to 3/16 inch or hand welding. The final seam weld was made as much as 3 years after the initial back side welds. After completion of welding, the seam w' elds were intended to be examined by dye- penetrant testing, vacuum box testing, and visual examination before the traveler was signed and dated.
The Final Safety Analysis Report (FSAR) for Comanche Peak includes a "Q" list (Table 17A-1), which identifies safety-related structures, systems, and components. This list also iaentifies the items for which a quality assurance program in accordance with 10 CFR Part 50, Appendix B, is re-quired. The stainless steel pool liner is not specifically identified in that list as a "Q" (safety-related) item; however, Gibbs & Hill Specifica-tion 2323-S5-18, Revision 3, dated April 6,1979, pasaraph 11.0, for the fuel pool liner, states that:
The contractor shall establish and implement a Quality Assurance Program which rigidly conforms to the applicable rules and standards as imposed by the Nuclear Regulatory Commission, 10 CFR 50, Appen-dix B, ANSI N45.2-1971, ANSI N45.2.9 (Draft 11, Rev. 0 - January 17, 1973), ANSI N45.2.10 (draft dated 1973), ANSI N45.2.11 (Draft 2, Rev.
2 - May 1973), ANSI N45.2.12 (Draft 3, Rev. 0 - May 2, 1973), ANSI N45.2.13 - NRC Extract (Draft dated May 31,1973) and this speciff-cation. The program shall include pertinent aspects of procurement,
- fabrication, site erection, inspection, testing, handling, shipping, storing and cleaning.
1
' . . _ .__m._...... v',
$ i 863512; )
u9ARDERS . . P D R , J _ ., s 0-219 l
L
i l
l
.ac ;
The allegations refer to a period early in 1983, when the QC inspection responsibility for the liner plate welds was being transferred from ASME, Brown & Root (B&R), to non-ASME, Texas Utilities Electric Company (TUEC),
since the liner welds were not ASME-Code welds. In preparation for that
' transfer, B&R apparently noticed travelers that had missing QC signatures for hold points but that had_been completed.
Part of the allegation was that a B&R QC inspector was coerced into sign-
' ing 112 liner travelers that supposedly were inspected by another inspec- '
t tor; that is, the other inspector had signed the inspection hold point some 5 years previously, filling out a chit documenting completion of the hold point, but not signing the traveler. The B&R QC inspector referenced the original chits as substantiation that the inspections had been com-pleted. According to the alleger, such sign-off constituted falsification of the travelers since the chits did not contain adequate information to ,
. substantiate verification (signing) of traveler inspection hold points.
The TRT reviewed all inspection travelers for liners in the fuel building, i
plus selected travelers for the refueling cavity pool in the Unit 2 con-4 tainment building. The TRT located 47 inspection travelers related to
-the Unit 2 refueling cavity liner, which were signed by two different in-l spectors and .for which hold point No. I had been signed-off out of se-quence (i.e., after later hold points had been signed off). A footnote
~
was placed at the bottom of each traveler, stating: " Late entry per CP-QCI-2.11-11, Para. 5, see Note.
Refer NDE [ nondestructive examination]
chit attached documentation." The footnotes were signed by the inspectors and dated in early March 1983.
The TRT found the " chits" referred to in the traveler footnote attached to the travelers. Most of the chits were dated in late 1978. The CP-QCI-2.11-11, Para. 5, " Note" referred to states: "The QC Inspector shall use information on the NDE Report chits to update Attachment 4-A
[i.e., the inspection traveler] daily and subsequently forward the NDE request chits to the QA vault daily as a status indicator."
The TRT noted that there appeared to be confusion in implementation of procedures related to stainless steel liner inspection, as indicated by the following:
- a. TRT review of CP-QCI-2.11-11, " Welding Inspection and Fit-up of 3
Stainless Steel Liners," showed that the correct number for this pro-cedure was CP-QCI-2.11-1, although pages in the procedure used either number. However, the TRT found that procedure CP-QCI-2.11-1 had been deleted from the QC manual on January 10, 1979, by B&R interoffice memo IM-16606. This procedure was replaced by procedure
- QI-QAP-10.1-4.
The TRT found two procedures numbered QI-QAP-10.1-4: QI-QAP-10.1-4,
" Request for Requalification of Welders by ANI," Revision 0, dated August 24, 1978; and QI-QAP-10.1-4, " Welding Inspection and Fitup of 1 Stainless Steel Liners," Revision 0, dated January 5,1979. The sec-t and procedure was replaced by QI-QAP-11.1-4, " Welding Inspection of Stainless Steel Liners," Revision 0, dated December 26, 1979. This 0-220
1 J
procedure was deleted from the QC manual on January 15, 1982, and was then reinstated without change on January 26, 1982.
The TRT also noted that, beginning with CP-QCI-2.11-1 and continuing through QI-QAP-11.1-4, paragraph 3.1.1, these procedures required fit-up and cleanliness verification of the inside seam welds after con-crete placement as a QC hold point. Similar requirement.s were noted in B&R procedure 35-1195-CCP-38, Revision 3, and in QI-QP-11.14-6, Revision 1. This appears to conflict with testimony prefiled by 'TUEC, which states that there was no procedural requirement for QC to re-
. verify cleanliness (Reference 23).
- b. Texas Utilities Generating Company (TUGCO) procedure QI-QP-11.14-6 references B&R procedure 35-1195-CCP-38 for instructions on issuance and distribution of the inspection traveler and chit. However, CCP-38 dces not contain such instructions; indeed, even the intent of the~
TUGC0 procedure is not clear, since the traveler and chit do not receive " issuance" and " distribution" in the usual sense,
- c. The use of the chit was introduced in B&R specification CCP-38 in Revision 1, dated October 21, 1977, paragraph 3.4.2, which stated:
"The form shown in Figure 3 [i.e., the chit] shall be used in the above noted hold points as notification to QC to perform inspection."
The QC inspector signed the chit after the inspection, providing l craft personnel with a record of completing the hold point and allow-ing them to proceed. The chit was apparently intended to be used as a construction document, not as a QC record to document completed inspections. Procedure CP-QCI-2.11-1, which referenced use of the-chit, was deleted on January 10, 1979. The superseding procedures, QI-QAP-10.1-4 and QI-QAP-11.1-4, contained no reference to the chit, but stated in paragraph 3.1.1:
The QC Inspector shall inspect the following items during fit-up and welding of liner material upon receipt of the Stainless Steel Liner Inspection traveler. [A list of items to be inspected followed.] '
The chit was replaced in CCP-38 on August 14, 1978 by the inspection traveler in Interim Change Notice (ICN) No. 2 for Revision 2, and was reintroduced in CCP-38 by ICN No. I for Revision 3, dated July 26, 1979, and in QI-QP-11.14-6, Revision 1, dated September 8, 1982, both of which indicated that the traveler was to be used to notify QC to perform inspections and that the chit was to be used to record in-spection results for hold points not included on the inspection traveler (e.g., examination of plates after removal of temporary attachments,etc). Apparently, the continued use of chits after January 10, 1979, as a primary means to request and document inspec-tions was not in accordance with the above procedures.
- d. There were frequent changes in the inspection traveler form attached to specification CCP-38. Initially, the-form was issued as part of CCP-38 on August 14, 1978, and showed five weld inspection hold points. The use of five hold points was inadequate for plate-to-plate welds, in that two fitup and cleanliness inspections of the L
0-221 4
same gap were appropriate, but the use of five hold points allowed for only one sign-off. The form was modified to include eight hold i points, one of which was an additional gap fitup and cleanliness l inspection, and was reissued on April 18, 1979. The form was also '
reissued with minor revisions on May 23, 1979, July 26, 1979, June 22, 1982 (decreased to seven hold points),~and September 7, 1982 (in-creased to eight hold points).
The TRT found that welding of stainless steel liners for the refueling cavity in the Unit 2 containment building was also the subject of noncon-formance report (NCR) M83-00795, issued on March 17, 1983, which stated the following nonconforming conditions:
- a. "A random review of stainless steel liner travelers for Reactor II cavity liner welds has found required fit-ups/ cleanliness inspections of inside (water si fe) welds cannot be verified as being performed."
- b. " Quality of welds indeterminate."
- c. "See attached ~ sheet for welds randomly reviewed."
- d. "F'ive hold tags applied."
This NCR was dispositioned and closed on March 23, 1983. However, the NCR was reopened and issued as Revision 1 on March 29, 1983, for the stated purpose of changing the nonconforming condition, i.e., the words " random" and " randomly" were deleted. In the opinion of the TRT, deletion of those words limited the scope of the NCR to the welds listed on the NCR, which was not the meaning of the original NCR. It appeared to the TRT that this revision was issued to avoid the investigation that the original NCR would have required. No explanation was provided on the NCR for revising the description of the nonconforming condition, and there was no indication on the NCR of the inspector's concurrence with the changed description. (The TRT has reviewed TUEC's testimony filed before the Atomic Safety and Licensing Board on November 27, 1984, regarding this subject, and does not accept TUEC's explanation that the revision was simply a matter of word changes.)
The TRT learned that Revision 2 of the NCR was issued on August 13, 1984, l for the stated purpose of deleting weld No. 1225 from the NCR. This was i an additional change in the description of the nonconforming condition, and also was made without explanation and with the signature of someone l other than the original inspectors. This NCR was dispositioned as follows:
l l Subject welds are seam welds utilized to provide leak tightness t
of the liner. Acceptability of welds shall be based on vacuum box and hydrostatic tests.
The TRT also reviewed NRC RIV Inspection Report 79-15, dated June 21, 1979, related to welding of the pool liners, in which the inspector concluded:
0-222
The resident inspector...has become reasonably sure that there were difficulties encountered by the welders with water, moisture, and in some instances with concrete on the weld surfaces, and that in some instances, the welds may not be completely sound inter-nally. These welds, however, serve no strength purpose and need only be smooth and leak free, factors which are established by visual inspection, dye penetrant, and by vacuum box tests of the joint after it is complete. The allegation, while probably true, has no safety significance.
Review of the stainless steel liner by the TRT mechanical and piping group is documented in Mechanical and Piping, Category 43. As indicated in that assessment, the liner is not safety-related, and the liner welds are con-sidered to be acceptable.
The TRT then reviewed that part of the allegation regarding the improper sign-off of inspection travelers, including review of material supplied by the intervenors (References 14 and 16). The TRT found certain recur-ring irregularities in the inspection travelers which were grouped into six categories. The categories listed in References 14 and 16 were not specifically addressed by the TRT, because the TRT reviewed the travelers independently in forming their conclusions. In some instances, the items identified in References 14 and 16 are similar to items noted by the TRT; in other instances, the items did not appear to the TRT to be significant.
The following examples are typical, but not inclusive, of the irregulart-ties noted for liners in the fuel building (unless otherwise noted):
- a. Travelers for Weld Nos. 595, 591, 588, and 589, Dwg. WFB00831,*
were signed by a QC inspector on 12-4-81. However, the " SAT" preceeding the signature does not match the inspector's signa-ture on penetrant testing inspection forms attached to the trav-elers. It appeared to the TRT that someone entered " SAT" and the date on the travelers, then obtained the QC inspector's sig-nature. This inspector confirmed during an interview that this had occurred, and the TRT discovered a number of travelers in which " SAT" was written for Hold Point No. 5 with no inspector's signature (e.g., weld Nos. 6, 15, 19, 34, 36, and 61 for Unit 2).
Also, Step 5 of the traveler for Weld No. 580, Dwg. WFB00831, had a signature and date of 8-5-81; then the date was changed to "7-2-80" and was initialed and dated "7-2-80." The traveler for Weld Nos. 567, 568, 572, and 583 were similarly modified.
The same type of corrections were noted on some NDE examination records (See travelers for weld numbers 60, 70, 104, 114, 126, 129, 144, and 580 for the fuel building). It appears to the TRT that dates were also pre-entered on some NDE examination records; however, the practice of pre-entering the scheduled date was apparently discontinued because of the number of corrections required.
- b. The traveler for Weld No. 564, Dwg. WFB00831, was signed as com-plete on "7-29-81." However, in reviewing the traveler before 0-223 ,
t
. .- . - . = --- . . . ... - .
transmittal to the B&R QA files, the inspector apparently noted that vacuum box testing had not been completed. The vacuum box testing was performed on "8-26-81," then the^signoff_date for weld completion was changed to "8-26-81," and the traveler was receipt-stamped by B&R QA files as "8-27-80 [ sic]." The TRT
' noted similar. occurrences on travelers for Weld Nos. 584 through 586, 558 through 563, and for 565 and 592.
'c . _The procedure numbers for the vacuum box ano leak tests were changed during construction of the liners; however, the numbers on the inspection forms were not changed. Since the form was not changed, the inspector was required to write in the correct pro-cedure numbers, as shown on travelers for Weld Nos. 568 through 571, and for 573 and 575, dated July 2, 1980 and August 10, 1981,.
and signed by another inspector. However, travelers for Weld Nos. 536A .hrough 544A and 546A through 549A, with the same in-spector's signature, were dated in November and December of 1981, and did not have the procedure numbers corrected. The procedure numbers were corrected and dated separately in December 1981 and January 1982, a few days before the travelers were received and date-stamped by the B&R QA files.
- d. . Inspection travelers, 58, 60, 114, 115, 126, 129, and 144 had the signoff dates changed without explanation, including, in some instances, changes to first party inspector dates on the NDE sheet,
- e. In comparing weld rod issue dates with inspection signoff dates for 23 travelers, the TRT located one traveler (No. 62 for Unit 1) in which weld rods were issued on October 3,1978, but the fit-up and cleanliness inspection _ hold point was signed off on December 14, 1978.
- f. The TRT also noted that inspection travelers typically had the welding information completed on the left side of the form to identify the welder, weld filler material log, procedure,'and the hold point against which weld rod had been issued. However, some travelers for welds that had welding performed and inspec-tion hold points signed off did not have information as to weld-er, weld filler material' log, procedure, or hold point (e.g.,
welds 410 through 540 for Unit 2). It is difficult to determine what the requirements were for completing the left side of the traveler form, since the procedures that reference the form identify requirements for QC inspectors to sign off hold points, but do not indicate how the left side of the form is to be used, or if it is required to be used.
- g. In testimony filed on November 27, 1984, p. 21,041, TUEC stated that for the inspector to correctly sign hold point 1 as a late entry, there must be verification that both inspections (inside and outside) have been completed for a plate-to plate weld; i.e.,
two-chits, or one chit plus one personal inspection. This is demonstrated by Unit 2 traveler No. 261, for which hold point 1 1 0-224
was signed on "3-3-83," and which has two chits attached, one dated "9-23-78," the other dated "10-19-81." However, Unit 2
. traveler Nos. 248, 250, 263, 331, 334, 335, 338, 341, and 346, which were also signed on hold point 1 on 3-3-83 as a late entry, have only one chit attached, which substantiated the initial (outside) inspection. The inside weld was completed in 1981 and/
or 1982, as indicated by weld filler material logs; therefore, a second cleanliness inspection before welding could not have beer.
accomplished by the inspector on March 3, 1983. Additionally, these travelers have an NCR (M83-00795, dated March 17, 1983) attached which states that the documentation required to support completion of inspection for the inside weld is not available.
Final sign-off of the traveler was completed after the NCR was dispositionad. Based on the above, the TRT concludes that these travelers were signed off improperly, i.e., without substantia-tion or personal inspection of the inside weld. The TRT does not consider this improfer sign-off to be falsification, as stated by the alleger, because of an apparent absence of an intent to deceive. The note entered by the inspectors on the .
travelers indicated they were signing hold point I with some reservation, and NCR M83-00795 further identified the lack of adequate documentation.
A total of 5,022 inspection travelers related to the liners were issued as of September 14, 1984. Of these, 1,209 were for Unit 1; 2,612 were for the fuel building; and 1,201 were for Unit 2.
Three lift gates in the fuel buildihg fuel transfer canal had gate supports which were welded to the liner. B&R specification CCP-38 required radio-graphy for these Welds (Nos. 1759 through 1772; 1803 thrcugh 1808; and 1823 through 1826). The TRT located the radiograph records. These records showed acceptance of all the above welds. Therefore, the TRT concluded that the required radiography inspection was completed. (This subject is assessed Mechanical and Piping, Category 41, allegation AQW-79.)
The fuel building fuel pool was filled and the liner was satisfactorily leak tested on March 22, 1983 by Test No. XSF-055.
- 5. Conclusion and Staff Positions: The allegation that required rrdiography was not completed is not substantiated, since the TRT found records showing the results of radiography of those welds for which radiography was required.
The primary subject of this allegation was the falsification or improper sign-off of records i.e., inspection travelers. The TRT could not con-clude that the irregularities noted constituted falsification, per se.
Apparently, these irregularities occurred because of poor practices and inadequate inspection forms. Some travelers also appeared to have been signed off improperly.
TUEC representatives indicated that it was common practice for the mill-wright department to write " SAT" and, in some instances, the scheduled
'date for inspection of the completed weld on the traveler, with the inten-tion of obtaining the inspector's signature when the weld was completed and inspected. Welding priorities apparently were then rescheduled and the pre-entered dates were corrected when the traveler was signed.
0-225
b The TRT concludes that there are record anomalies apparent in the liner plate travelers which are not adequately explained on the face of the travelers (e.g., dates changed), which violate procedures (e.g., failure to transfer sign-off from chits to travelers daily), and which employ in-adequate procedures (i.e., confusion over the use of the five-line travele r) .
It appears tr the TRT that the QC documentation relating to the liner plate welds did not meet the standards expected of an effective QA/QC program, or the standards required by Gibbs & Hill specification 2323-55-18, and 10 CFR 50 Appendix B.
The TRT discussed the results of the assessment of allegation AQ-78 in a meeting with the ' leger on December 10, 1984, as documented in the meeting transcript, 'w .ng on page 166. A letter containing these results was also transmis . to the alleger. A meeting with the alleger of AQ-55 was held on March ., 1985, to discuss the results of this assessment. No new issues or concerns were identified, t
e-L
- 7. Attachments: None. a Reference Documents:
- 1. Logbook pages for radiography of Weld Nos. 1759 through 1772, 1803 through 1808, and 1823 through 1826.
- 2. Signoff sheet for Test XSF-055, dated March 22, 1983.
- 3. Fuel Building Spent Fuel Pool Liner Details, 23235-0831 through -0834.
0-226
]
l
- 4. Fuel Pool Liner - Fuel Building Weld Map, WFB-00831.
- 5. B&R Specification 35-1195-CCP-38, " Stainless Steel Liner Erection."
- 6. B&R Procedure CP-QCI-2.11-1, " Welding Inspection and Fit up of Stain-less Steel Liners," with IM-16606, dated January 10, 1979.
- 7. Office Memorandum TUQ-2340, " Stainless Steel Liner Travelers Unit II,"
dated September 11, 1984.
- 8. Interview with alleger A-3, pp. 60 through 89.
- 9. Bostrom-Bergen drawing 2401.
- 10. Testimony of alleger A-4, March 7,1984, pp.18,19, 20.
- 11. Testimony of alleger A-3, August 1,1984, pp. 59,516 through 59,536.
- 12. G&H Specification 2323-SS a, " Stainless Steel Liners," Rev. 3, April 6, 1979.
- 13. Deposition by C. Thomas Brandt, August 16, 1984 (Tr. 45,239-355).
- 14. " CASE's Evidence of a Quality Control Breakdown," dated September 27, 1984.
- 15. Prefiled Testimony of C. Thomas Brandt, October 3, 1934 (Tr. 45,356-480).
- 16. " CASE's Further Evidence of a Quality Control Breakdown in the Con-struction, Installation and Inspection of the Stainless Steel Liner Plate," dated November 15, 1984.
- 17. Testimony of C. Thomas Brandt, Tr. 15,629-697 (9/12/84); 15,978-16,214 (9/13/84); 16,728-777 (9/18/84); 17,264-363 (9/19/84); 20,569-774 (11/26/84); 20,778-21,091 (11/27/84).
- 18. Deposition by alleger A-3, August 1, 1984 (Tr. 59,516-536, 59,640-676) and August 2, 1984 (Tr. 59,773-825).
- 19. Deposition by A11eger A-1, July 31,1984 (Tr. 54,596-617).
- 20. Deposition by Dwight Woodyard, July 24, 1984 (Tr. 56,561-566).
- 21. Deposition by Ted Blixt, July 25, 1984 (Tr. 57,015-036).
- 22. Deposition by Robert Stever, July 25, 1984 (Tr. 58,024-056).
- 23. Prefiled Testimony of C. Thomas Brandt, October 16, 1984 (Tr. 45,373).
- 24. Transcript of TRT Interview, December 10, 1984 (beginning on page 166).
- 25. NRC letter, D. Eisenhut, NRC, to M. D. Spence, TUEC, dated January 8, 1985.
- 26. Liner inspection travelers as noted in text.
t) W. v) %
- 8. This statement prepared by: fcrw bl (, 4 f T'. Curry, TRT /F Date /
Technical Reviewer Reviewed by:
H. Livermore, vtwM/ 6 &[
Date Group Leader Approved by:
V. Noonan, Date Project Director 0-227
1 i
e-(
l l
0-228 <
k n MW .