ML20234D036
| ML20234D036 | |
| Person / Time | |
|---|---|
| Site: | Braidwood  |
| Issue date: | 07/01/1987 |
| From: | Rosalyn Jones BUSINESS & PROFESSIONAL PEOPLE FOR THE PUBLIC INTERES, ROREM, B. |
| To: | |
| References | |
| CON-#387-3935 OL, NUDOCS 8707070077 | |
| Download: ML20234D036 (77) | |
Text
-
r o' Y -- p D X7 E ~ r
UL'i; C UNITED STATES OF AMERICA NUCLEAR REGULATORY COMMISSION I d P40 BEFORETHEATOMICSAFETYANDLICENSINGAPPEAL[OAda
)
In The Matter Of: ) Docket Nos. 50-456
) 50-457 COMMONWEALTH EDISON COMPANY
)
(Braidwood Station, Units 1 and 2) )
OPENING BRIEF OF INTERVENORS-APPELLANTS BRIDGET LITTLE ROREM, ET AL.
Douglass W. Cassel, Jr.
Robert Guild Robert L. Jones, Jr.
Business and Professional People for the Public Interest 109 North Dearborn Street Suite 1300 Chicago, Illinois 60602 (312) 641-5570 f
8707070077 870701 PDR ADOCK 05000456 PDR G
/[O3
]
t
/
TABLE OF COMTENTS Page TABLE OF CONTENTS . . . . . . . . . . . . . . . . . . . . . i TABLE OF AUTHORITIES . . . . . . . . . . . . . . . . . . . iii QUESTIONS PRESENTED . . . . . . . . . . . . . . . . . . . . 1 INTRODUCTION AND BRIEF PROCEDURAL HISTORY . . . . . . , . . 2
- 1. THE EXTENSIVE RECORD EVIDENCE OF THE PERCEPTION OF AND EXISTENCE OF IMPROPER PRODUCTION PRESSURE, COUPLED WITH THE LACK OF CREDIBLE EVIDENCE OF SATISFACTORY PERFORMANCE BY COMSTOCK QC INSPECTORS, COMPELS REVER3AL OF THE BOARD'S FINDING THAT EDISON HAS MADE THE " REASONABLE ASSURANCE" SHOWING REQUIRED BY 10 C.F.R. 550.57(a) . . . . . . . . . . . . . . . . . . 6 A. " Reasonable Assurance" Can Be Established Only Through A Quality Assurance Program That Complies With Critierion I of Appendix B, 10 C.F.R. Part 50 . . . . . . . . . . . 6 B. There Is Extensive Evidence, Much Of It Uncontradicted, That The Comstack Quality Assurance Organization Failed To Meet Criterion I Requirements . . . . . . . . . . . . . 7
- 1. Production Pressure Imposed On Comstock By Edison . . . . . . . . . . . . . . 8
- 2. Pressure Imposed By Comstock QA Managers . . . 8
- 3. Perceived Incidents of Harassment and ,
Retaliation . . . . . . . . . . . . . . . . . . 9 {
i C. There Is No Credible Evidence In the Record To i Support The Board Majority's Finding That Comstock QC Inspectors Performed Satisfactorily In Spite of the Pervasive Atmosphere of Production Pressure . . . . . . . . . . . . . . . . 11
- 1. Reinspection Evidence . . . . . . . . . . . . . 12 I 2..' Laney Testimony . . . . . . . . . . . . . . . . 14
- 3. Individual Inspector Denials . . . . . . . . . 14 ;
l 1
/
Page II. THE RECORD EVIDENCE CONCERNING DEFICIENCIES IN COMSTOCK'S PRE-1983 GRID AREA WELD INSPECTIONS COMPELS REVERSAL OF THE BOARD'S FINDING THAT THOSE INSPECTIONS PROVIDE "NO CAUSE TO BE CONCERNED" . . . . . . . . . . . . . 16 III. THE BOARD MAJORITY IMPROPERLY PLACED THE '
BURDEN OF PROOF ON INTERVENORS . . . . . . . . . . . 18 CONCLUSION . . . . . . . . . . . . . . . . . . . . . . . . 20 i
l l
i l
l l
l 11 a
/
TABLE OF AUTHORITIES j Page(s)
Louisiana Power and Light Company (Waterford Steam Electric Station, Unit 3), ALAB-732, 17 NRC 1076 (1983) . . . . . . . . . . . . . . . . . 18 Niagara Mohawk Power Corporation (Nine Mile Point Nuclear Station, Unit 2), ALAB-264, 1 NRC 347 (1975) . . . . . . . . . . . . . . . . . . 6 Northern States Power Company (Monticello Nuclear Generating Plant, Unit 1),
ALAB-611, 12 NRC 301 (1980) . . . . . . . . . . . . 6 l
Union Electric Company (Callaway Plant, Unit 2), ALAB-740, 18 NRC 343 (1983) . . . . . . . . 11 l
10 C.F.R. S2.732 . . . . . . . . . . . . . . . . . . 18 10 C.F.R. Part 50, Appendix B, Criterion I . . . . . 6, 7, 10 10 C.F.R. S50.7 . . . . . . . . . . . . . . . . . . 9, 11 10 C.F.R. S50.57(a) . . . . . . . . . . . . . . . . 6, 15, 18 l
l l
l iii l
e~-
/
UNITED STATES OF AMERICA NUCLEAR REGULATORY COMMISSION BEFORE THE ATOMIC SAFETY AND LICENSING APPEAL BOARD In The Matter Of: )
) Docket Nos. 50-456 COMMONWEALTH EDISON COMPANY ) 50-457
)
(Braidwood Station, Units 1 and 2) )
OPENING BRIEF OF INTERVENORS -
APPELLANTS BRIDGET LITTLE ROREM, ET AL.
QUESTIONS PRESENTED
- 1. Does the extensive record evidence of a pervasive atmosphere of harassment and production pressure in the L.K.
Comstock Quality Control organization compel reversal of the Atomic Safety and Licensing Board majority's finding that Applicant Commonwealth Edison Company has carried its burden of demonstrating licensability of the Braidwood plants?
- 2. Does the record evidence of deficient grid area weld inspections prior to October 1983 compel reversal of the Licensing Board majority's finding, over the dissent of its Chairman, that those inspections provide "no cause to be concerned" over the safety of the plant?
- 3. Did the Atomic Safety and Licensing Board majority improperly place the burden of proof in the licensing proceedings on Intervenors?
1
_ = - m-
.c
/ . ,
1 i
INTRODUCTION AND BRIEF PROCEDURAL HISTORY This is an appeal from the Atomic Safety and Licensing Board's (" Board") May 19, 1987 Concluding Partial Initial Decision */ concerning Quality Assurance Contention 2.C. filed by ;
Intervenors-Appellants Bridget Little Rorem, et al., in which the Board ruled by a 2-1 vote, over the dissent of its Chairman, that there is reasonable assurance that the Braidwood Units can be operated without endangering the health and safety of the public.
QA Contention 2.C., which was first filed on March 8, 1985, j i
charged that Applicant Commonwealth Edison Company (" Edison") and ]
its electrical contractor, L.K. Comstock Engineering Company 3 4
("Comstock"), had violated 10 C.F.R. Part 50, Appendix B, ]
Criterion I and 10 C.F.R. 50.7 by subjecting Comstock Quality Control inspectors to systematic pressure, including harassment, f intimidation, retaliation and other discrimination that discouraged the identification and correction of safety-related !
problems at Braidwood.
I QA Contention 2.C. was initially filed as part of a much l broader late-filed contention alleging deficiencies in the Braidwood QA program. On April 17, 1985, the Board rejected the i
I contention (LBP-85-ll, 21 NRC 609) but granted Intervenors leave to file an amended version, which they did on May 24, 1985. The 1
- / Although the Licensing Board titled its May 19 decision its
" concluding" Partial Initial Decision, it is still considering Intervenors' May 6, 1987 " Motion To Admit Late-Filed Contention On Financial Qualifications." In its June 15, 1987 Memorandum on Licensing Board jurisdiction, the Board expressly retained jurisdiction over that motion "until further action c/: the Licensing Board." .In addition, on July 1, 1987, Intervenors filed a further Motion To j Reopen The Record To Admit Late-Filed FQ Centention.
]
2
)
f
/
Board rejected two parts of the amended contention outright, l
reserved ruling on Part 2.C., and otherwise accepted the revised I
( contention. (LBP-85-20, 21 NRC 1732 (1985). On July 28, 1985, the Board admitted Part 2.C. pursuant to a stipulation by the parties.
On April 24, 1986, the Commission reversed the Board, l l
dismissing all of Intervenors' QA contentions except for Part l l
2.C. and remanding Part 2.C. to the Board for a determination whether it met the late-filing requirements of 10 C.F.R.
S2.714(a)(1). (CLI-86-6, 23 NRC 241 (1986)). On May 2, 1986, the Board found Contention 2.C. admissible.
Evidentiary hearings on QA Contention 2.C. occupied nearly 100 hearing days between May 6 and December 17, 1986. The l
record includes approximately 18,000 pages of testimony from 60 witnesses and 500 exhibits.
During the hearings, Intervenors presented extensive evidence in support of their allegations of improper production pressure directed to Comstock QC inspectors, including evidence to support their allegations of specific incidents of harassment and retaliation for raising safety concerns. 1 l
Edison did not deny the existence of an atmosphere of j l
improper production pressure at Comstock; indeed, Edison conceded in its Proposed Partial Initial Decision (" Proposed P.I.D.") "a widespread perception of harassment and pressure to compromise quality" among Comstock QC inspectors. Proposed P.I.D. at 49.
Rather, Edison argued that any such perception was the result of misunderstandings and was not attributable to any wrongdoing by 3
. - , _ _ y I
' /
1 Edison. Edison also presented evidence that in its view demonstrated that the QC inspectors performed satisfactorily in spite of any pressure that may have been imposed on them.
On May 19, 1987 the Board issued its Concluding Partial Initial Degision disposing of QC Contention 2.C. The Board found that although some actions taken by Comstock management in j dealing with QC inspectors " crossed the line of acceptable behavior," those actions were not "of sufficient severity to warrant the precipitous action of license denial." Majority Op. t l
74 */. Moreover, the Board majority found " considerable )
\
evidence" that Comstock QC inspectors had performed well and with integrity. Id. at 9.
In dissent, Chairman Grossman found considerable evidence of l 1
improper production pressure imposed by Edison on Comstock, and by Comstock on its QC inspectors; indeed, Chairman Grossman found incidents of harassment and retaliation so severe as to warrant a substantial civil penalty against Edison. Minority Op. 1-2.
i Chairman Grossman agreed wtth the majority that, notwithstanding j the improper pressure, the QC inspectors had performed satisfactorily and that thstre was reasonable assurance of plant safety in spite of any harassment. Id. at 2. Nevertheless, because he found that Comstock's grid system of weld inspections i
- / Citations to the record and to the Licensing Board's opinions and findings will be in the following form:
Licensing Board Majority Opinion - " Majority Op. "; ,
Licensing Board Minority Opinion -- " Minority Op. "; I I
Licensing Board Minority Findings of Fact - " Minority FF
"; Witness Testimony - "(Name of Witness), Tr. ";
Exhibits - " Edison Ex. " or "Intervenors' Ex. __".
l 4
)
/
prior to October 1983 " totally lacks credibility," Chairman Grossman would have denied an operating license absent further evidence. Id.
In this appeal, Intervenors challenge both the evidentiary and the legal basis for the Board majority's conclusion that notwithstanding the atmosphere of improper production pressure at Comstock, Edison has met the " reasonable assurance" requirement of 10 C.P.R. 550.57(a). The Board majority's decision turns on l its finding that Comstock QC inspectors performed satisfactorily in spite of any production pressure. See Majority Op. 9-10,74-l
- 75. Partibelowarguesthat the record evidence compels 1
reversal of that finding. l l
j Part II below argues that the record evidence also compels l reversal of the Board majority's finding that Jeficient grid area weld inspections pose "no cause to be concerned." Majority Op.
71.
l Part III below argues that the Board majority's opinion not only runs counter to the manifest weight of the evidence, but is further flawed by an incorrect allocation of the burden of proof. Although the Board acknowledged that Edison bore the burden of proof of " reasonable assurance" in this proceeding (Majority Op. 77), the Board majority clearly -- and improperly --
placed the burden of proof on intervenors.
The regulations and the record compel reversal of the l
Licensing Board's ruling.
5
/
a I. THE EXTENSIVE RECORD EVIDENCE OF THE PERCEPTION OF AND EXISTENCE OF IMPROPER PRODUCTION FRESSURE, COUPLED WITH THE LACK OF CREDIBLE EVIDENCE OF SATISFACTORY PERFORMANCE BY COMSTOCK QC INSPECTORS, COMPELS REVERSAL OF THE BOARD'S FINDING THAT EDISON HAS MADE THE " REASONABLE ASSURANCE" SHOWING REQUIRED BY 10 C.F.R. S50.57(a).
This Appeal Board may reject Licensing Board findings when I it is " convinced that the record compels a different result."
Northern States Power Company (Monticello Nuclear Generating Plant, Unit 1) , ALAB-611, 12 NRC 301, 304 (1980) (quoting Niagara Mohawk Power Corporation (Nine Mile Point Nuclear Station, Unit 2), ALAB-264, 1 NRC 347, 357 (1975)). The record in this case compels the conclusion that Edison has not carried its burden of demonstrating with " reasonable assurance" that the Braidwood plants may be operated safely. 10 C.F.R. S50.57(a). j A. " Reasonable Assurance" Can Be Established Only Through A !
l Quality Assurance Program That Complies With Criterion I of Appendix B, 10 C.F.R. Part 50 The Commission's regulatory scheme for assuring the protection of the public's health and safety in construction of nuclear po*+!er plants such as Braidwood requires effective implementation by licensees such as Edison of the Commission's Quality Assurance Criteria. 10 CFR Part 50, Appendix B, " Quality Assurance Criteria for Nuclear Power Plants and Fuel Reprocessing Plants." The " reasonable assurance" standard for licensing is met through effective adoption and implementation of a quality assurance program that complies with Commission regulations.
Criterion I of Appendix B in the licensing regulations requires the establishment of a Quality Assurance organization with " sufficient authority and organizational freedom," including
" sufficient independence from cost and schedule when opposed to F
6
- 2 safety considerations," to " identify quality problems; to initiate, recommend or provide solutions; and to verify I implementation of solutions." 10 C.F.R. Part 50, Appendix B, Criterion I. The responsibility to establish and execute such a program is entirely on the Applicant's shoulders. Id.-
B. There Is Extensive Evidence,14uch Of It Uncontradicted, i That The Comstock Quality Assurance Organization Failed To Meet Criterion I Requirements. l i
The most important fact in this case is also undisputed:
that "a widespread perception of harassment and pressure to compromise quality" pervaded the Comstock QC department after i
August 1983. Edison's Proposed P.I.D. at 49. That fact was established by_ extensive record evidence, but it was brought home most clearly by accounts of the unprecedented March 29, 1984' visit by 24 Comstock QC' inspectors to the NRC resident i
inspector's trailer at the Braidwood site to complain of l
harassment and production pressure at Comstock. (That visit is recounted in Minority FF 202-211.) */ During the meeting, the 24 inspectors expressed their unanimous opinion that Comstock Quality Control management was emphasizing quantity over quality.
i McGregor Tr. 17534-35; Minority FF 210, The QC inspectors' perceptions resulted from their i
observations and from shoptalk concerning production pressures imposed bya. Edison on Comstock, the words and behavior of Comstock
/ Frequent reference is made in this brief to the Minority Findings of Fact. EacE of the Minority Findings of Fact to which reference is made is attached hereto as Appendix A. '
Intervenors incorporate those attaching findings, with their record citations, as part of this brief.
7 !
- /
managers, and numerous incidents which the inspectors perceived as retaliation by Comstock against QC inspectors for raising safety concerns:
- 1. Production Pressure Imposed on Comstock by Edison.
As the Board majority acknowledges, Edison Construction Superintendent Dan Shamblim began heavily pressuring Comstock QC Manager Irv DeWald in mid-1984 to eliminate a large inspection backlog at Comstock: "There is no doubt that considerable pressure was put on DeWald by Shamblin, including the possibility of a work shutdown if progress on reducing the backlog was not made." Majority Op. 15. Shamblin informed DeWald by memorandum that " reduction of this backlog must be the first priority of LKC Production, Engineering and Quality Control personnel" and that
" positive results must be seen very shortly" (emphasis in original). Intervenors' Ex. 8 at 1; Minority FF 25. Shamblin ordered DeWald to report to him every Monday on Comstock's progress. Intervenors Ex. 8 at 2. The record is devoid of any concerns expressed by Shamblin about the quality of Comstock's work. Edison's pressure on Comstock QC for faster production and its threat to cancel Comstock's contract were matters of considerable shoptalk among Comstock QC inspectors. E.g.,
Bossong Tr. 9857; Gorman Tr. 5840-41, 5871, 5884-85; Seeders, Tr.
7568; See Minority FF 39.
- 2. Pressure Imposed by Comstock QC Managers.
Comstock management, primarily through QC Manager Irv DeWald and QC Supervisor Rick Saklak, created and reinforced in a variety of ways the perception that Comstock was much more 8
T r L ;
concerned with quantity of work than quality. After Shamblin began pressuring'DeWald, DeWald. instituted weekly meetings.with O<
the QC inspectors at which he rel4yed-Edison's. threats and exhorted them to increase the speq9 of their work. Minority FF 38, 39. There.is no mention in the record that DeWald ever made similar exhortations at those meeting with respect to the quality of-work. DeWald's statements were understood by inspectors as a 2
direction to emphasize quantity over quality. Id. That impression was. strengthened by DeWald's introduction of a " status tracking system" which projected the number of inspections an inspector should perform in a day and compared them*to actual results. Minority FF 40.
- 3. Perceived Incidents of Harassment and Retaliation.
A great deal'of the evidence in the record concerns incidents in'which' inspectors believe they were harassed or subjected to retaliation by DeWald and Saklak for raising safety concerns. The Board majority, Chairman-Grossman, and the parties all have different views about which of those incidents actually constitute actionable retaliation under 10 CFR S 50.7. But there
.is no dispute that there was considerable shoptalk about such incidents and that inspectors widely. viewed the actions of DeWald and Saklak ao retaliatory and as part of an effort to promote quantity over quality. See, e.g. Holley, Tr. 5210, 5266-67; Gorman, Tr. 5900-03; Hulin, Tr. 18128; Peterson, Tr. 5965-66; Rolan, Tr. 4914; Hunter,'Tr. 8646, 8650; Martin, Tr. 9233; Bowman, Tr. 6955, 6950-52; Wicks Tr. 7133; Gorman Tr. 5886; Perryman Tr. 9636; Int. Ex. 195 at 15-17, 49; Int. Ex. 196.
I J
w
=
l Those perceptions are not without considerable foundation in the record. 'Even the Board majority concedes that the behavior -
of Saklak and DeWald toward individual inspectors left much to be !
desired, and that "some actions taken in dealing with QC inspectors crossed the line of acceptable behavior even for a l large construction site." Majority Op. 74. For example, the Board found Saklak unsuited for his position (Id. at 41) and l
acknowledged his "scurrilousness" (Id. at 40), his " outbursts of I i
temper" (Id. at 52) and his practice of threatening inspectors j l
with discharge even though he was powerless to follow through. j l
(Id.) j
. The Board majority does not dispute the existence of a I
pervasive atmosphere of pressure at Comstock QC to emphasize quanity over quality. But neither does the Board even acknowledge that fact, much less attribute to it the decisive weight it demands. The Board's position seems to be that as long as inspectors continue to bring some safety concerns to management, the system must be working properly. Majority Op. 30.
But Criterion I demands much more: that OC inspectors feel completely free to bring all safety concerns to management's attention. That freedom does not exist in an organization such l as Comstock's in which inspectors commonly believe (whether correctly or incorrectly) that management prefers speed over careful work and even retaliates against inspectors who raise safety concerns. The Commission cannot possibly condone a QC 10
-, s mg
- e organization in which inspectors believe they may face a choice between doing their jobs properly or being punished. That is the very reason for the existence of 10 CFR S50.7. When such perceptions are pervasive, as at Comstock, among a large group of inspectors over a period of years, "there has been a breakdown in quality assurance procedures of sufficient dimensions to raise legitimate doubt as to the overall integrity of the facility and its safety-related structures and components." Union Electric Company (Callaway Plant, Unit 1) ALAB-740, 18 NRC 343, 346 (1983).
C. There Is No Credible Evidence In The Record To Support The Board Majority's Finding That Comstock QC Inspectors Performed Satisfactorily In Spite Of The Pervasive Atmosphere of Production Pressure.
The pivotal finding in the Board majority's opinion is that "the QC inspectors, in spite of management harassment and schedule pressure, performed their inspection duties in a professional manner and the fruit of their labors was not poisoned by management's actions." Majority Op. 74-75. It is perhaps because of that finding that the Board failed to explore in more depth the fact of and consequences of the pervasive atmosphere of production pressure at Comstock. Apparently, in the Board's view, the poisoned atmosphere at Comstock was inconsequential as long as the QC inspectors performed well. */
- / That view is strongly at odds with the NRC licensing regulations. The heart of the NRC licensing requirements is that an effective and independent QA organization must identify, evaluate and initiate corrective action where called for on all conditions that fall short of acceptance criteria. The licensing scheme equates safety with an (footnote continues on following page) 11
In fact, there is no credible evidence in the record that the QC inspectors actually performed satisfactorily. Edison failed by a wide margin to produce evidence that could rebut the
" legitimate doubt" about the integrity of Comstock quality control created by the evidence of harassment and production pressure at Comstock.
The Board majority cites three categories of evidence that in its view demonstrate effective QC performance: the expert testimony of Mr. Laney concerning the integrity of QC inspectors generally (Majority Op. 75), the testimony of " cote than a dozen" Comstock QC inspectors that they had never personally succumbed to pressure to suppress safety concerns (or witnessed others do so) (Id. at 9, 75), and evidence from two reinspection programs (Id. at 10, 76). For the reasons explained below and at greater length in the Minority Findings of Fact, each of those categories of evidence is woefully deficient.
- 1. Reinspection Evidence.
Two separate sample reinspection programs (the BCAP and PTL programs) were conducted at Braidwood for purposes unrelated to measuring the effectiveness of Comstock QC inspector performance.
Nevertheless, Edison attempted to use the results of those (footnote continued from previous page) effective QA program: if " reasonable assurance" could be established through after-the-fact sampling, then the QA program requirements would be superfluous. The very existence of the requirements for an independently structured QA program, comprehensive documentation, and prompt and comprehensive identification of discrepancies indicate that minimal after-the-fact sampling is insufficient to meet licensing criteria.
i 12 i
l
programs as evidence of effective Comstock inspections. In fact, for reasons explained cogently in Minority Findings of Fact 330-373, the reinspection evidence has no probative value whatever on the issue of OC inspector performance. As Chairman Grossman found, "Neither of these programs was designed to measure Quality {
Control effectiveness and neither program, as presented by Applicant, was able to offer any assurance that Comstock's Quality Control program was effective or that the electrical system was properly installed by Comstock." Minority Op. 39.
Rather than repeat the arguments in the Minority Findings at length here, we merely list and summarize the principal defects in the BCAP and PTL data:
BCAP Defects (a) Time Periods. The BCAP program examined only those items that were QC inspected and accepted before June 30, 1984; only 24% of the electrical construction items were therefore included. Much of the record evidence of harassment and production pressure concerns incidents and j time periods after June 30, 1984. See Minority FF 343. The BCAP statistics can therefore address at most a small portion of the production pressure evidence.
(b) Types of Statistics Generated. The " agreement rate" and l
" design significance" statistics generated by the BCAP I I
program can indicate at most the incidence and severity of defects remaining in the sample portions of the plant. */
l
- / Although the design significance statistics are in principle j a means of assessing the hardware quality at Draidwood, they (footnote continues on next page) 13
- l
Abseng information that Edison.does not possess, such as the rate of. errors committed by craft workers, those statistics provide no measure at all of inspector effectiveness. See Minority FF' 356-360.
PTL Progam The PTL program. suffers from the same basic infirmities as
.the BCAP program, buL also suffers from the additional severe infirmity that its sample was not chosen in a statistically random manner that permits generalization beyond the sample itself.' Minority FF 370.
- 2. Laney Testimony.
The. Board majority apparently gave considerable weight to the testimony of Mr. Laney that, in his experience, QC inspectors
-are generally people of high personal integrity who would resist pressure to compromise themselves. (Majority Op. 75). Such general testimony hardly establishes that individual Comstock inspectors were unaffected by harassment and production pressure; indeed, it is so general and speculative that one may question whether it has any relevance at all.
- 3. Individual Inspector Denials.
The Board majority also gave heavy weight to the testimony of-a dozen or so inspectors that they personally had not allowed (footnote continued from previous page) are unreliable even for that purpose, because they reflect considerable subjective judgment by Sargent & Lundy, the Braidwood architect / engineer, which had a substantial stake in the results of those judgments. See Minority FF 362-369.
J 14
)
a ,
harassment or production pressure to compromise the integrity of their inspection work (Majority Op. 75). Even if the testimony of that small fraction of Comstock inspectors was credible, it certainly does not establish that Comstock's many other QC inspectors were not affected. Indeed, as Chairman Grossman points out, the testifying inspectors would be those most likely to resist production pressure, since they were the inspectors who had the courage and integrity to complain to the NRC in the first place. Minority Ultimate Finding of Fact and Conclusion of Law 14 (p. 281).
More importantly, little or no weight can be given to denials even of the testifying inspectors because of the severe consequences of an admission of wrongdoing. Even Mr. Laney testified that he would give little or no weight to ',
uncorroborated denials of wrongdoing by an inspector, becaut+
such deniahs are self-serving. Laney Tr. 17300-301.
In sum, the evidence cited by the Board majority falls far short of establishing acceptable work performance by Comstock QC inspectors. The lack of evidence of effective QC inspections, coupled with the extensive evidence of the poisoned atmosphere at Comstock, compels reversal of the Board majority's finding that Edison has met the " reasonable assurance" requirements of 10 C.F.R. S50.57(a). */
- / In his minority findings, Chairman Grossman took pains to indicate he gave no weight to the evidence relied upon by the Board majority in their finding that Comstock QC inspectors had performed adequately. Minority Ultimate FF & CL 14 and (footnote continues on following page) 15
a o II. THE RECORD EVIDENCE CONCERNING DEFICIENCIES IN COMSTOCK'S PRE-1983 GRID AREA WELD INSPECTIONS COMPELS REVERSAL OF THE BOARD'S FINDING THAT THOSE INSPECTIONS PROVIDE "NO CAUSE TO BE CONCERNED."
A substantial amount of evidence was presented establishing that prior to October 1983, Comstock QC employed a system of grid area weld inspections under which the inspection of as many as 1000 or more welds was documented on a single inspection checklist. See Minority FF 315-329. Those checklists, which covered as many as three or four days of inspection activities, regularly indicated that no discrepant conditions had been discovered. See Minority FF 318, 320, 323. The grid area inspection procedure was not permitted by any approved Comstock quality procedure. (Martin Tr. 8358).
Chairman Grossman found that the grid area weld inspections are " totally lacking in credibility" in light of the record evidence that the inspection standards of a significant portion of the weld inspectors were substandard, that the inspectors failed to observe significant numbers of discrepancies, and that the weld inspectors failed to document discrepant conditions as required by Appendix B . . .
Minority Op. 37.
(footnote continued from previous page) 15 (pp. 280-81). Nevertheless, Chairman Grossman found, on the basis of unspecified "other evidence," that the Comstock inspectors had performed satisfactorily in spite of the harassment and production pressure at Comstock. There is no record citation or indication of any kind in either the Minority Opinion or Minority Findings of Fact to what "other evidence" Chairman Grossman refers. Intervenors know of no such evidence that would support that finding, which is strongly at odds with both the letter and spirit of the balance of the Chairman's Findings and Opinion.
16
The BCAP reinspectors found defects in 16% of the welds that had .
already been QC inspected and accepted. Minority FF 352.
Considering that fact, and the fact that the defect rate is almost certainly higher among welds not already QC inspected and accepted, Chairman Grossman found it " inconceivable" that the grid area reports indicating no defects could accurately have reflected the original condition of the welds. Minority Op. 37.
The Board majority acknowledged the problems with grid area inspections, but found "no cause to be concerned" about them because some welds inspected by the grid area method were reinspected in the BCAP program, which uncovered no so-called
" designed-significant" defects. Majority Op. 70-71. */ But, as explained above, the BCAP results provided no measure at all of the effectiveness of QC inspector performance. See Minority FF 356-357. Thus, BCAP provides no reassurance at all that grid area inspections were adequate.
In sum, the record evidence concerning grid area inspections all tends to raise serious concerns about their effectiveness; there is no substantial evidence to the contrary. Edison's failure to carry its burden of probf on that issue ca.mpels reversal of the Board majority's opinion.
4
- / The Board also defended the prc-1983 practice of inspecting j only 35% of welds. Majority Op. lu. But the issue raised I by Chairman Grossman is not whether the sample was large enough, but rather whether the inspection of the sample was performed properly. If the inspections were unreliable, they would not be acgeptable even if 100% of weldn had been inspected.
1 17
. . . . . . . . _ . . . . . _ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . - . . . . . . . . . . . . . .. _.__.__...__.._.J
III. THE BOARD MAJORITY IMPROPERLY PLACED THE BURDEN OF PROOF ON INTERVENORS.
The Board majority acknowledged at the end of its opinion that Edison, as required by 10 C.F.R. S2.732, bore the burden of proof in this case. Majority Op. 77. There is no question that Intervenors met any burden of going forward with evidence of a quality assurance breakdown that may have been necessary to trigger Edison's burden of persuasion on quality assurance issues. See Louisiana Power and Light Company (Waterford Steam Electric Station, Unit, 3), ALAB -7 3 2, 17 NRC 1076, 1091 (1983).
Nevertheless, and despite the Board's suggestion to the contrary, its opinion makes clear that it placed the ultimate burden of proof in this case on Intervenors. The very framing of the issues at the outset of the majority's opinion is the first indication that the burden was misplaced. Majority Op. 9-10.
The question before the Board, under 10 C.F.R. SS2.732 and j 50.57(a), was whether Edison had proven, in the face of considerable evidence of a defective Comstock QC organization, that Comstock QC had performed effectively, in compliance with l l
Commission regulations, and that there is " reasonable assurance" that the plant may be operated safely. Instead of thus focusing on Edison's burden, the Board posed -- and answered -- its questions in terms of the quantum of evidence that Intervenors had produced. The ultimate question in the Board's mind was not whether Edison had proven effective QC performance, but whether Intervenors had proven a "sufficiently large breakdown in quality assurance procedures that there is no ' reasonable assurance' provided that the safety systems at Braidwood will l
l 18
a: ,
. perform their functions and the public health and safety will be protected." Majority Op. 9.
1 The clearest' indication of the Board's misallocation of the burden-of proof is the statement at page 74 of its opinion that it does not consider Edisen's and Comstock's " indiscretions of sufficient' severity to warrant the precipitous action of license denial." (emphasic added). Implicit in that remarkable phraseology is a strong presumption in favor of granting a license, a3d a correspondingly heavy burden on Intervenors to produce compelling and persuasive evidence that a license should not be granted. The Board has thus turned the regulatory scheme upside down.
A further indication of the Board majority's distortion of the regulations occurs at the bottom of page 75, where the majority makes much of the fact that Intervenors failed to produce evidence of significant hardware defects that are as yet unidentified and uncorrected by Edison. Whether or not that i assertion is true,*/, the Board cannot thrust upon Intervenors the impossible burden of proving unidentified defective hardware; once Intervenors have gone forward with evidence of serious quality control deficiencies, it the the Applicant who must provide proof of acceptable QC performance and hardware quality.
- / A strong inference can be drawn from the BCAP data (which show thousands of defects in the BCAP sample that were overlooked by QC inspectors) that the electrical construction items not included in the BCAP sample are 3 also riddled with defects not found by the original QC inspectors.
19 i
l The Board majority has turned the regulatory scheme upside l down. It has created a presumption in favor of granting a license and imposed a heavy burden of proof on Intervenors. Its decision must be reversed.
CONCLUSION The Licensing Board majority committed errors of fact and law that compel reversal of its May 19, 1987 Concluding Partial Initial Decision.
Respectfully s b itted i riks
(- /~
One of the Agtorneys for Bridget Little Rorem, et al.
DATED: July 1, 1987 1
l
~
Douglass W. Cassel, Jr.
Robert Guild Robert L. Jones, Jr.
Business and Professional People for the Public Interest <
109 North Dearborn Street Suite 1300 Chicago, Illinois 60602 (312) 641-5570 20 m
- APPENDIX A 51
- 11. MINORITY FINDINGS OF FACTS _
A. Orcanization in toto, as Findings I accept Staff's Prcposed Findings 1-15, 1-15.
(" Applicant" or Commonwealth Edison Company
- 1. [1.]
located in
" CECO")
is the owner of the Braidwood Nuclear Station, ultimately is Braceville, Illinois. As the owner, Applicant responsible for the design, construction, and operation of the facility. Applicant engaged Sargent & Lundy ("S&L") to design the Varicus contractors were retained by Applicant to Braidwood Station.
construct the facility. Only one of which -- Comsteck, the electrical In particular, our contractor -- is pertinent to these findings.
contracter focus is on certain activities and events involving that durina the period August 3934 through March 1985.
l
\
1
- 2. [2.] Comstock was awarded the electric 61 contract 4r )
Tes tin.ony Braidwood Units 1 and 2 by Applicant on Feoruary 5,1979. !
Prior te of Ecbby Treece, A.13 at 6. ff. Tr. 12881 [Treece Test.). ,
the electrical work had been performed by E. C.
LKC's involvement, !
Ernst Company. .: d_.
I
~ --- - - - - - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
, i 62 Mr. DeWald's plan, 20 LKC Quality Control inspectors were dedicated to f
perfonning backlog inspections and 34 Quality Control inspectors were 1 l
assigned to " currenti .nspections." Int. Ex. 12 at 4-5.
l
- 25. [25.] Mr. Shamblin was not entirely satisfied with Mr.
DeWald's target completion date of September 1984 for the junction box In a and small equipment backlog. See Shamblin Test., A.13 at 11.
letter dated June 9,1984, Mr. Shambiin informed Mr. Rolan and Mr.
DeWald that according to CECO's records, the amount of all backlogged Int. Ex. 8 at 1. Mr. Shamblin inspections exceeded 6,000 in number.
observed to Mr. DeWald and Fr. Rolan that CECO was "very concerned about the large existing backlog" and stated that "[r] eduction of this backlog must be the first priority of LKC Production, Engineering and l
[ Emphasis in original .] Mr.
Quality Control personnel." M.
Shamblin also took note of the fact that LKC's request for cdditional time to reduce the inspecticn backlog had been granted, indicating that " positive results (i .e. significant current inspection backlog
[ Emphasis in original.]
reductions) must be seen very shortly." M.
If such results were not soon forthcoming, Mr. Shamblin was prepared Shamblin Test., A.18 at 15. Finally, to suspend LKC's cperations.
Mr. Shamblin directed Mr. Rolan and Mr. DeWald to report to him every Monday on the progress in eliminating the backlog that had been made I
the previous week. Int. Ex. 8 at 2.
p
.o 6 70 In addition 1 add the following five findings (37-41) as follows:
newly promoted Quality Control In July, 1982, 37.
h d with the mission. by Comstock Supervisor Richard Saklak was c arge construction of trying to bring the quality control department under control and to organize a production system for responding to the Tr. 8014-15. At installation reports from the production department.
24 years of age, the young Mr. Saklak had previously been employed as a cost and scheduling engineer at Edison's LaSalle station and <
l reassignment had been a inrnediately prior to his Quality Control Tr. 7992.
planning and scheduling engineer for Comstock production.
Very quickly after He had no prior Quality Control work experience.
Mr. DeWald's appearance as Quality Control manager in August of 1983, he evaluated Mr. Saklak as a "very aggressive individual" who had taken on added responsibilities under him, duties that would have been performed by an Assistant Quality Manager, "with great enthusiasm zest." Mr. DeWald concluded that " Rick is a real asset to the Int. Ex. 52. When Mr. Saklak became a Braidwood QC department."
Quality Control supervisor in July of 1982, he shared his supervisory At about the time Mr. DeWald became 90sition with another individual. of Mr. Saklak became the sole supervisor
' Quality Control Manager Quality Control inspectors. Tr. 8000.
38.
LKC replaced its prior Quality Control Manager Thomas 1983, because Mr. Corcoran had Corcoran with Mr. DeWald in August,
v n
> :s e p 71 been too quality conscious and not sufficiently construction oriented.
L Mr. DeWald relied upon the Friday meetings, referred to Tr. 1220-27.
Quality Control above,: as a primary means for communication with Tr. 1786. Management usually described those areas of inspectors.
inspection that were behind and those areas which needed more manpower l
' allocated to them. Tr. 4241. Assistant Quality Control Manager Larry Seese would read the status reports which detailed the progress being.
made on projects to eliminate inspection backlogs and the projected Tr. 4243, 6871-73, 9663. John dates of completion of those projects.
d about
. Seeders . testified that at these meetings Mr. DeWald commente Tr. 7567. From these being under schedule pressure from Edison.
meetings, Seeders understood that the quantity of inspections was emphasized over inspection quality because the weld inspectors would DeWald pushed "DeWald wants numbers again" when Mr.
comment that Tr. 7566. Quality Control inspectors for greater productivity.
Inspector Terry Gormar also interpreted these weekly meetings as L
reflecting management's emphasis of quantity over quality in urging Tr. 5798. Mr. Gorman inspectors to perfonn more inspections.
f recalled Mr. DeWald's comp sints that not enough work was being accomplished because too many people were sitting around the office when they should have been out in the field performing more Tr. 5776-77. Quality Control Inspector Robert Wicks inspections.
l testified that he believed quantity was emphasized over quality trying to meet Edison-impcsed because Comstock management was It was shoptalk among Quality Control 4 deadlines. Tr. 7077-78. }
)
o ;
72 Tr. 7087.
inspectors that Comstock stressed quantity over quality.
Several inspectors remembered Mr. CeWald talking about a minimum required number of inspections to be performed as an attempt to Tr. 6866-67, 9240-41.
eliminate inspection backlog.
39.
Six inspectors testified that Comstock Quality Control management was pressuring inspectors for production under an Edison threat to cancel the Comstock contract if the inspection backlog was Gorman, Tr. 5840-41; Holly, Tr.
not eliminated by certain dates.
8499-8500, 8744-47; Peterson.
5151-52; Bossong, Tr. 9857; Hunter Tr.
Tr. 5950-51; Seeders, Tr. 7567-69. . Three inspectors acknowledged that the threatened loss of Comstock's contract was shoptalk among the 5840-41, Bassong, Tr. 9857; Gorman, Tr.
Quality Control inspectors.
5871, 5884-85; Seeders , Tr. 7568.
Mr. Seeders testified ihn such its shoptalk was f airly common when Comstock was not meeting Tr. 7568. Inspector Danny Holley recalled a meeting in deadlines.
the summer of 1984 at which QA manager Robert Seltmann indicated th f
if the backlog of inspections was not eliminated, it could mean that Tr. 5151-52. f l
the livelihood of Comstock at Braidwood wculd the beweekly lost.
Inspector R. D. Hunter testified that more than once at meetings during 1984, Mr. DeWald had stated that Comstock was in danger of losing its contract if it failed to satisfy certain promised Inspector Dean Tr. 8499-8500, 8655, 8744, 8747.
completion dates.
Peterson recalled a special meeting where assistant Quality Control Manager Larry Seese indicated that things were looking very criti
)
. 6 73 CORRECTED ,
l l
Tr. 7087.
inspectors that Comstock stresseo quantity over quality.
< Several inspectors rernembered Mr. DeWald's talking about a minimum required number of inspections to be perfonned as an attempt to eliminate inspection backlog. Tr. 6866-67, 9240-41. f Si inspectors testified that Comstock Quality Centrol
- 39. )
management was pressuring inspectors for production uncer an Edison l threat to cancel the Ccmstock contract if the inspection backlog was not eliminated by certain dates. German, Tr. 5840-41; Holly, Tr.
5151-52; Bossong, Tr. 9857; Hunter, Tr. 8499-8500, 8744-47; Peters on, l
Tr. 5950-51; Seeders, Tr. 7567-69. Three inspectors acknowledged that the threatened loss of Comstock's contract was shoptalk emong the Quality Control inspectors. Bossong, Tr. 9857; Gorman, Tr. 5840-41, 5871, 5884-85; Seeders, Tr. 7558. Mr. Seeders testified that such its shoptalk was fairly conmon when Comstock was not mcating deadlines. Tr. 7563. Ir.spector Danny Holley recalled a meeting in the suntner of 1984 at which QA manager Robert Seltmann indicated that if the backlog of inspections were not eliminated, it could mean that Tr. 5151-52.
the livelihood of Comstock at Braidwcod would be lost.
Inspector F., D. hunter testified that more than once at the weekly meetings during 1984, Mr. DeWald had stated that Cemstock was in danger of losing its contract if it failed to satisfy certain promised Tr. 8499-8500, 8655, 8744, 8747. Inspector Dean completion dates.
Peterscn recElled a special meeting where assistant Cuclity Centrol Manager Lnrry Seese indicated that things were looking very critical
l 4 $
74 ,
CORRECTED for Comstock and that everyone's help was needed to eliminate the backlog. Tr. 5950-51. Mr. DeWald acknowledged such a rumor that Comstock was in jeopardy of losing its electrical contract. However, he recalled the rumor circulating in January, 1985. Tr. 1345-47.
Ultimately, Comstock did lose its contract for a portion of the electrical work on Unit 2. The Gus K. Neuberg Company has replaced Comstock for a portion of the Unit 2 electrical installation and inspection work. Tr. 1349,
- 40. In order to monitor inspector productivity and manage the inspection backlog elimination program as well as the performance of inspections on current installations, Comstock's Quality Control I management developed a status tracking system. Under this system, the scheduled' completion of various inspection tasks, including the inspection backlogs which existed in the spring of 1984, vias projected on the basis of the number of average inspections an individual inspector was expected to perform in a day, e.g., an average expected level of performance, goal or quota. Int. Ex. 23; Seese Pref., ff. Tr. 2320 at 3-10; Seese, Tr. 2350-51; Saklak, Tr.
8116-18. For example, Mr. DeWald's early June 1984 backleg completica schedule was based on the average of five welding, equipment ana configuration inspections per day; six termination inspecticns per day; and seven conduit inspections per day on average. Int. Ex. 12.
The status report figures showing the number of inspections actually performed ' were compiled from incividual inspectors' daily reports,
1 1
4 J 75 CORRECTED then passed through the inspectors' leads, who summarized and routed I
them to the status department. Comstock management posted the l l
periodic status reports for Quality Control inspectors' information. <
j Seese, Tr. 2498-99; DeWald , Tr. 1576-78: Comstock management acknowledged utilizing the status reports and tracking system to regulate inspector overtime assignments, and to transfer inspectors .
from one inspection area to another. Seese Pref., ff. Tr. 2320, at 9; Seese, Tr. 2350.
1
- 41. It is against this background that I consider alleged harassment, intimidation, and discrimination cited by Intervenors in its inspector harassment contention. First, I take up the matter of Worley O. Puckett, a Level III Weld Inspector, who, according to Intervenor, was fired by LKC because he "make numercus complaints about safety and quality deficiencies which he identified in the ;
course of his duties at Braidwood." Second, I consider the case of i John Seeders, an LKC inspector who allegcdly was transferred out of LKC Quality Control Department to a clerk position in LKC's Engineering Department "in retaliation fcr his expression of quality concerns." Contention 2.C. Third, I discuss the complaints of harassment and intimidation mace to the fiRC by 24 LKC inspectors in March of 1985. Finally, I discuss additional instances of alleged harassnent and inti:aidation not cited specifically in Intervenors' Cantention.
i
179 discussed the incident with some of his colleagues who agreed to accompany him to the offices of Mr. McGregor and Mr. Schulz, the Senior Resident inspectors at Braidwood. Tr. 4205-06. Based on his previous experience, Mr. Snyder considered Mr. McGregor .to be a fair and concerned individual. Tr. 4593.
202. [197.] At approximately 8:15 a.m., the next morning, Friday, March 29, 1985, Mr. Synder, accompanied by five other Quality i Control inspectors " walked into the NRC Braidwood office with numerous allegations which 'effect' the' quality of work being accomplished by I the electrical contractor," LKC. Int. Ex. 42 at 1; Tr. 4210, 11567. .
Mr. McGregor advised the inspectors of their right to remain anonymous
' but informed them that the NRC would like to know their identify in order to obtain further information from them if necessary and to I advise them of the results of the meeting. Int. Ex. 42 at 1; Tr. j
)
1 11567. )
203. [198.] During this meeting, Mr. Snyder informed the NRC Inspectors of the threat made against him the previous day by Mr.
Saklak. Tr. 4211, 11569. Other inspectors complained to the NRC about Mr. Saklak's conduct as well. App. Ex. 11, Tr. 11569, 11736.
In addition, the Quality Control inspectors raised a number of other f I
complaints against LKC Quality Control management in this meeting.
App. Ex. 109. Among these complaints were that CECO's " Quality First" 1 Program was not effective; that unqualified persons were awarded lead
180 inspector positions; that certain of LKC's Quality Control management team harassed and intimidated Quality Control inspectors; and that management was more concerned with the quantity rather than the quality of the inspectors' inspections. Ld.;Tr.11569.
204. [199.] Af ter the meeting adjourned, Mr. McGregor and Mr. Schulz contacted their superiors in .the regional office to bring Weil to their attention the events that transpired that morning.
Test. , A.63 at 16; Tr. 11569-70. Participating in that conference Weil Test., A.16 at call were Mr. Warnick, Mr. Weil, and Mr. Forney.
/
- 16. The NRC inspectors informed the region that six LKC' Quality Control inspectors had complained to them about harassment and l intimidation from Mr. Saklak and an over-emphasis on quantity at the Mr. McGregor expense of quality of LKC's Quality Control management.- ,
the LKC- quality and Mr. Schulz also reported to the region "that control inspectors were threatening a walkout the following Monday."
j Weil Test., ff. Tr. 11948, A.16 at 16. I
(
l 205. [200.] Mr. McGregor and Mr. Schulz also recommended to the region that someone from the regional office be sent to Braidwood imediately to take sworn statements from the Quality Control 11582.
Mr. McGregor and Mr. Schulz also inspectors. S,ee Tr.
the Region consider issuing an order stopping LKC recommended that from performing further work pending an inspection of the quality of l l
work already performed. App. Ex. 109. The inspectors believed these l l
- . s 181 actions appropriate because they had previously notified CECO of discontent in LKC's Quality Control Department and Ceco apparently had failed to take sufficient action to address the problem. Tr.
11740-56.
206. [201.] After speaking with Mr. McGregor and Mr.
Schulz, Mr. Warnick, Mr. Weil, and Mr. Forney discussed the matter !
among themselves and decided that CECO should be notified of the substance of the Quality. Control inspectors' allegations. Weil Test., .
i ff. Tr.11948, A.65 at 16. [This-eevese-ef-ett4en-wes-< ens 4+ tent-*4th the-pelfey-ef-the-NRG-wh4eh ureeegn4mes-that-an-applicant-has-a-streng interest-4e--lcrnbg- of- irnd-taking-appropr4ete-+c44en- to eeeeeet-any problems- wMoh -may--+f-feet-the-etW-4-meh--f aeMp 2 id.r t-See- Apps-4*.--149 -*t -1. ] The region determined that it'would be appropriate to notify CECO of the substance of the allegations that had been made "because of the allegations involved CECO personnel and the information to be provided Applicant did not appear to be of such character as to enable Applicant to compromise a subsequent NRC inspection or investigation." Weil Test., id_., A.66 at 17; see App.
Ex. 119. Mr. Weil was therefore asked "to advise the six LKC quality l control inspectors of the NRC's proposed course of action and ascertain whether any of them desired to remain anonymous." Ld.,A.65 at 17.
i 182 207. [202.] Mr. Weil then called Mr. McGregor to ask him to arrange a telephone conference with the six Quality Control inspectors. Tr. 11570. McGregor in turn contacted some of the inspectors and asked them to attend a meeting in his office during their lunch break. Tr. 4265. Mr. McGregor indicated that any other inspectors who wanted to attend should feel free to do so. Tr. 4265, 11571.
208. [203.] At approximately 12:00 p.m., the conference call began. Tr. 11571-74; Weil Test., A.67 at 18. Mr. Weil was informed at that time by Mr. McGregor that 18 Quality Control inspectors, in addition to the original 6, were present in the NRC office. Weil Test., ff. Tr. 11948, A.67 at 18; see Tr. 11573.
209. [204.] As stated above, the purpose of the telephone conference was to advise the six Quality Control inspectors of the action Region III proposed to take and determir.e whether any of them remain anonymous. I d_. , A.65 at 17; Tr. 11971-72.
wished to Accordingly, Mr. Weil spoke with each of the original six Quality Control inspectors and asked whether there was any objection to the NRC notifying CECO of the substance of the allegations. l_d., A.68 at 18; Tr. 11,972. None of these Quality Control inspectors expressed Id.
any disagreement or objection with this proposal to Mr. Weil.
Mr. Weil also asked each of these inspectors whether they wished to remain anonymous and was informed by each that confidentiality was not
e r 183 desired._ I d_.
Mr. Weil then afforded the other Quality Control inspectors in attendance an opportunity to speak; ten of those Quality Control inspectors took advantage of this opportunity and made statements. Id.
210. Senior Resident Inspector McGregor testified
{newl that at some point during the meeting a request was made for a show of _
hands to determine how marty Quality Control inspectors agreed that Comstock Quality Control management was emphasizing quantity over quality. Mr. McGregor recalled that the 24 inspectors' agreement with the statement was unanimous, without abstentions or denials, and that he~ or Mr. Schulz relayed that agreement to the Region during the conference call. Tr. 17534-35.
211. [205.] The telephone conference lasted between 30 and 40 minutes. Tr. 4269. Mr. Weil then notified 01 Director Pawlik of the allegations received from the Quality Control inspectors and was informed by Mr. Pawlik that an " investigation by 01:RIII was not warranted" based on the information then available. Staff Ex. 23.
f At approximately 1:15 p.m. that afternoon, 212. [206.]
another telephone conference was held, this time between officials of Region III and Commonwealth Edison Company (Ceco). Tr. 11579.
Present on behalf of the region were Mr. Forney, Mr. Williams, Mr.
Weil Test., ff. Tr.
Weil, and Rogelio Mendez, an NRC inspector.
l 1
i
. 6-
,f 237 o
I-D. ' Grid Area Weld Inspections 315. 'In 1981, Quality Control Manager Irving DeWald had worked at Braidwood as a Level II QC Inspector for L. K. Comstock. He-and a few other weld inspectors including Richard Martin perfonned so-called " grid area basis" inspections, documenting large numbers of welds on single inspection reports. DeWald Pref. Test. , ff. Tr.1700, at A.19. -There was a general and consistent belief among. Quality Control inspectors that Mr. DeWald had signed a checklist that documented his inspection of a thousand or more welds in a single day.
Eight of the Quality Cor trol inspectors who testified had heard through general talk among the inspectors of a Mr. DeWald.1,000-plus checklist, but had not seen it themselves. Hunter, Tr. 8495-98; Martin, Tr. 8294; Mustered, Tr. 5061-62,- 5086; Rolan, . Tr. 4762-63, 4769-71; Stout Dep., Tr. 144-145; Klachko Dep. , Tr. 192, 265-66; Hii, Pref. ~est., ff. Tr. 16608, at 3; Gorman,'Tr. 5817-18, 5828. Six of the inspectors that testified claimed to have actually seen one or more Mr. DeWald 1,000-plus weld checklist. Bossong, Tr. 9848-50; Bowman, Tr. 6890-91; Holley, Tr. 5154-56; Perryman, Tr. 9652-57; Peterson, Tr. 5933-35; Wicks, Tr. 7151-54.
316. Mr. DeWald testified that he is " pretty certain" that he had never documented a thousand or more welds on a single inspection checklist. Tr. 4092. He doubted that it was possible that he ever did so. At - the direction of his Supervisor, Tony Simile,
1 o 4 ._
238 Quality Control Inspector Bowman undertook a search for this checklist during the course of the proceeding, but was unable to locate the document.-. Tr. 6894; Bowman Pref. Test., ff.- Tr. 16000 at A.13-A.15.
317. On deposition, Mr. DeWald had been asked the maximum number of welds he had ' documented on a single checklist and could not remember,. until found in a casual search through .his old weld
' inspection checklist, documenting as many 'as 551 welds on a single inspection checklist. Tr. 15000; Int. Ex. 19, 318. Although the inspectors originally believed that those inspection. checklists . represented inspections done over the period of one day, they were subsequently informed that the ' checklist could have Mr..
been the result of several days' work. See , e.o. , Tr. 6892.
DeWald testified that, while the inspected welds may have been covered by a single checklist, "it may been a day, two days, three days it Tr. 1482. With took me to complete all the particular inspections."
regard to the particular checklist covering 551 welds (Int. Ex.19),
Mr. DeWald believed that it took two, three or four days to complete He had found all the welds acceptable.
his inspections. Tr. 1490.
Tr. 1491.
319. Quality Control Inspector Richard Martin bad observed a checklist covering 60 cable tray hangers and associated aux steel that Mr. Martin, who had could have covered up to 2,500 to 3,000 welds.
239 inspected on the grid basis alongside of Mr. DeWald, recalled actually Tr. 8376.
inspecting on the order of 300 to 350 welds in one day.
Earlier, on disposition, he even recalled inspecting 500 welds on one Mr. Martin had also seen a checklisc filled out by day, Tr. 8377-78.
a'. thousand welds.
Tr.
Quality Co'. ,'ol Inspector Thomas witn over 8294.
320.
In~ December of 1984, inspectors John Walters, Mike Blake_ and Dan _ Asmussen reviewed a 1979 checklist by Quality Control Inspector Richard Yankeitis, documenting, on a single sheet, the inspection and acceptance ofJ 1,166 welds.
In a letter of concern to management, Mr. Asmussen stated, "I can not accept a 0% reject rate for that many welds inspected." One of the 77 hangers listed _on the 1984 grid inspection cover sheet was later ~ the subject of a reinspection which identified extensive welding defects not identified Mr. Asmussen, speaking for himself in the original grid inspection.
and the other inspectors', indicated that they recognized their responsibility to bring their concerns to management's attention and '
"immediate management this situation deserved felt that investigation." Int. Ex. 18 at 5-6.
321.
Mr. DeWald, as Quality Control management, agreed that '
the total number of welds (1,166) being inspected in a single day did appear to be a considerable number for one individual to accomplish.
However, he dismissed that problem on the basis that the inspections I
< a 240 written on. the checklist could possibly have represented a ' total of l
several days' work. He dismissed the other concern, regarding the 1 acceptance of all.the welds, as follows:
The other questionable item brot.ght out by Mr. Asmussen is zero ' (0) rejects. To him, this is questionable, although.
it is felt the individual was a competent inspector.
Mr. DeWald concluded that if Mr. Asmussen had any question concerning 1
the validity of the inspection, he could reinspect the items himself "to ensure a good valid inspection." M.at1.
the grid basis weld 322. As Mr. DeWald described l
inspections, there was only a small number of inspectors, they would complete an area, document it on the PTL coversheet, and fill out the i
inspection report on various days.
They did not complete their inspection reports on each and every day that they had done inspections. Tr. 1479. The reason that they didn't fill out inspection reports as they completed each component was because there Tr.
were only three or four inspectors covering a hundred welders.
1483.
323. Mr. Martin, who had served as a weld inspector along with Mr. DeWald, described the weld inspection documentation practices Tr. 8343-78, 9384-97. As a rule, no official in more detail. !
documentation of rejectable conditions was ever made unless the craf t I
l t
~ ,
241 couldn't fix the. defect promptly. Only then would an Inspection j Correction Report be issued writing up the defect. Tr. 8349. Mr.
Martin would simply note rejectable conditions in his personal notebook without indicating the identity of the welder or the Only . acceptable . items particular weld found defective. - Tr. 8351.
were documented.on the official weld inspection checklist. Tr. 8352.
This system, employed by Mr. Martin and the others for performing and documenting weld inspections, was not provided for by any Comstock quality procedure. Tr. 8358. I 324. It was not until October,1983, after a Commonwealth L
audit, that the practice of documenting : weld Edison Company
. inspections on personal notebooks and completing checklists later in 4 the office was uncovered and brought to an end. Tr. 9570-77.
4 325.- In 1984, Mr. DeWald took newly hired Level III welding 3 I Mr.
l inspector Worley Puckett on a tour of the Braidwood facility.
DeWald pointed out welds to Mr. Puckett that he (DeWald) had inspccted l
when he previously worked as a Level II weld inspector.- The welds were on a large hanger. Mr. Puckett testified that although he just l
glanced at the welds, he saw welds that he (Puckett) would not have accepted. The welds he had observed had undercut, excessive spatter, slag, overlap, and excessive craters. Mr. Puckett indicated that he would not have had inspectors working for him that would have accepted some of thuse welds. Tr. 6215-17.
-. 4 I'
242 326. Robert D. Hunter joined the L. K. Comstock Company at Braidwood in October of 1983. Within 30 days thereafter, he became a welding inspector. He had had plenty experience as a welder and Tr. 8471-81. When he first began inspecting at welding inspector.
Braidwood, Mr. Hunter was asked by Quality Control Manager Irving DeWald to review some of Richard Martin's welds. Mr. Martin had been one of the few inspectors inspecting welds under the grid system, and had been trained by Mr. DeWald. Mr. Hunter reported to Mr. DeWald 1 that Mr. Martin's work was lacking in certain areas. Mr. Martin would mi::s things such as undercut, cold-lap, and other things of that nature. Subsequently, in early 1984, Mr. Hunter accompanied Mr.
Martin to the field, reviewed Mr. Martin's work, and discussed Mr.
Martin's prior training with him. According to Mr. Hunter, Mr. Martin didn't know some simple things about welds.. For example, he didn't know what rod-craters were, and what the face or toe of a weld should Tr. 8484-92. Quality Control Inspector Thermond Bowman look like.
also testified with regard to reinspecting Mr. Martin's early welding inspections. Although he testified reluctantly on this matter, Mr.
Bowman indicated that he had found one-third of the welds inspected by Mr. Martin to have been rejectable. Tr. 6888.
and 1985, the $"Braidwood Construction 327. In 1984 Assessment Program" was instituted which reinspected samples of prior Of over 13,000 welds !
Quality Control accepted construction work.
reinspected, approximately 16% were fcund to be deficient in one or
243
. more respects that might possibly have an affect on their safety
-The function. Other, lesser - types of deficiencies were ignored.
sampling was done on a statistically random basis and, presumably, If the approximate i should have represented the population at large.
16% figure for discrepant welds represents discrepancies after_ at least one original Quality Control inspection, it is inconceivable that any large numbers of uninspected welds would be free of discrepancies. If the percentages arrived under the BCAP hold true, in. an inspection of 500 welds, one might expect 80 welds to be.
discrepant (500 x 16%), even after the welds were inspected at least once by Quality Control. Assuming at least a 50% Quality Control effectiveness on the welds examined under the BCAP Program 160 welds out of 500 would have been discrepant originally (i .e ._, before inspection). In the case of Mr. Yankeitis's 1,166 welds examined .by Mr. Asmussen, one might similarly expect at least 340 welds to be discrepant. Not only is it inconceivable that the weld inspection reports indicating acceptances of multi-hundred welds could have reflected the original condition of the welds, but it is also inconceivable that such large numbers of discrepancies could have been reworked or repaired during the one, two, three or four days between the beginning of the inspection and the signing of the inspection report. Neither time nor space would be adequate for such operations even if craft were not otherwise occupied in its further construction
'1 activities. .
e- .
244 i
[ 328.
Moreover, the ' failure to record' discrepant conditions, j
which surely must have existed in the multi-hundred weld inspections under the grid' system, if observed, would violate Criterion XVII ~ of j
10 C.F.R. Part _50, Appendix- B, which requires , as a minimum, a record f of any deficiencies noted.
329.
On the . basis of the evidence. adduced, which indicates that the inspection standards of a significant portion of the weld inspectors was substandard, thdt the inspectors . failed to observe significant numbers of discrepancies, and -that the weld inspectors failed to document discrepant conditions as regttired by Appendix B, the weld inspections performed under tne grid system, in effect until October of 1983, lack credibility.
f i
l l
1 l
l l
o a 245 E. Applicant's Sampling Reinspection Programs -
t 1
i 330. In an attempt to prove the effectiveness of the 'l Comstock Quality Assurance Program, Applicant presented- the results of.
two large' sampic . reinspection programs. The first program was the Construction Sample Reinspection (CSR) conducted as part of the BraidwoodConstructionAssessmentProgram(BCAP). This data spans the i
time period from the start of construction until June 30, 1984. The l second set of data results from the routine overinspection of Comstock
- Quality Control accepted work by Pittsburgh. Testing Laboratory (PTL).
for Applicant's Quality Assurance Department for the period July '1, l 982 to June 30, 1986. These reinspection programs were conceived, designed and carried out independently of each other. DelGeo'rge, Reb.
Prep. Test. at 6, 9, ff. Tr.16740; Kaushal, Reb. Prep. Test, at 7-8, i ff. Tr. 15568.
'331. The Braidwood Construction Assessment Program (BCAP) was a program of reinspections and reviews carried out by Edison in 1984 and 1985 covering safety-related construction activities at-Braidwood. The BCAP was comprised of three principal program elements. These were (1) the Construction Sample Reinspection (CSR),
(2) the Reverification of Procedures to Specification Requirements (RPSR), and (3) Significant Corrective Action Program (RSCAP).
Kaushal,- Pref. ff. Tr.13068 at 4. Only the CSR program element was presented at hearing.
1
a e i 246 COPPECTED 332. The CSR consisted of a visual reinspection cf a sarrpie June 30, of on-site, safety-related ccnstruction wcrk which, as of Tne sample l 1984, had been completed and (uolity Coritrol inspected.
was selected based in part on engineering judgment and in part en :ne use of statistical concepts. The reinspecticns were carried out frem
~
October, 1984 through July, 1985.
Kaushal, Pref. ff. Tr. 13069 at ;,
13-16.
I 333. The CSR (and otNr elements of ECAP) were carried cut by the SCAP Task Nrce. Tnc BCA? Task ~crce Director was Edison Kaushal reported airectly to the employee Dr. Narindar Kaushal.
Braidwood . project manager, Mike Wallace, who had principal production responsibilitics at Braidwood. Kaushal Pref. f f. Tr. 13C.:9 at 9-10.
The BCAP GA group, a par: Of Edison's QA department, uride r the direction of an Ediscn employee, Neil Smith, oversaw the ECAP Task Force activities, id. it 10.
by BCAF CSR 1:spectors were 334. ciscrepancies r eur.d evaluated foe design sigr.ifi ance by Sargent &
andj f L , wnich as responsible for develop nc tne d . acn drawinc :;ecification: fcr Braidwood.
The ai:tivities of Sarcent & Lundj, the ECAP Tap Fcrte, and BCA/ QA were reviewed oy an inderercent E ger t Cverview Srcap t! reu ., the i. A nt. ion - e s e c ec 'i Lerpor ticr.
(IEOG) cstablit.hed .
Kuushal, Pr"#. ff. Tr. 00" at 10 !!
y 3.--- i 247- I 335. The . NRC Staff assigned inspector Ronald Gardner to monitor on-site the implementation of the BCAP program. Gardner was on-site from August 20, 1984 until June,1985, during which time he i engaged.in daily oversight of BCAP activities. Gardner Pref. ff. Tr.
17606 at 3,'7..
336. BCAP Director Kaushal was assigned to BCAP .in March, 1984, after BCAP was conceived but before it was implemented. Tr. l 13098. NRC Inspector Gardner was assigned to BCAP in August, 1984.
Gardner Pref. ff. Tr. 17606 at 3, Tr. 17569.
337. BCAP was not designed to look at suspected problems or to respond to the possible effects of harassment and production Nor was BCAP pressure on Quality Control inspector work performance.
Rather, designed to look for isolated design significant defects.
BCAP was designed as a quality " confirmation" program; the program design assumed that construction quality was- good and relied on a size that would reveal only recurring, programmatic sample construction problems. Kaushal, Pref. ff. Tr. 13068 at 3-6, 16-17; Tr. 13326-28.
338. The record contains evidence that NRC officials had misgivings about the sufficiency of the BCAP design, but no evidence that the NRC actually approved that design. Edison forwarded the BCAP program document to NRC Inspector Keppler and his Staff for comments
i 0 CORRECTED 248 Keppler's response to Edison, Intervenor's Eihibit in Jur.e, 1984 I 140, made 27 specific recommendations for changes in the procram design.
With only one exception, Edison resocnded to each of Keppler's comments that " Edison believes that no change to tne Kaushal, Tr. 13114-17.
existing BCAP document is warranted." .
ta 339. The Board ruled tnat Mr. Garaner was not comcetent j vouch on bePalf of tne NRC fcr the adequacy cf the BC,AP program cesig !
j cr tne S&L design significance evaluations because he hac no role in t
/
either aspect anc that if Edicon or the Staff wished to establish thet design or design significance f ha:i approvec the BCAP the NRC Tr.
they would have to present other witnesses.
evaluations ,
No such witnesses were ever prodJCed. i 17666-6C5.
Jor purposes cf taking 340. The CSR was a san.ple program.
samples, the entire pcpulaticn of on-site contractors' safety-related construccion work was divided into 30 " construction categories," wnich 9
were definea as groues of hardware constructed usicig sir.fiar processes i Six af those znstructior er containing similar types of cc:::penents.
cabies; cable pans, caole pan categories centair.ed electriai work:
electrical equipment hangeri, and hangers, conduit, concuit The l Kaushai, Reu. Prep. Test.
at 11-13, ff. Tr. 13068.
installation.
consisted of three parts.
total se:cple f or each construction category a racnr.er as te I I
" random" pc rtion was cr.osen in s ac r.
The first er confidence tupport fortal statis ti ca l conclusions witt. et Mast Si !
I
g m.
9' D 249 95% reliability concerning each sampled construction and at least category.
Kaushal, Reb. Prep. Test. at 13-16, ff. Tr.13068; Frankel, For the second portion of Reb. Prep. Test at 9-11, ff. Tr. 17082.
the sample, engineering judgment was used to determine sample size and This portion emphasized types of items which are to select items.
It was emergency core cooling systems.
part of safe-shutdown or initially intended that the " engineering judgment" portion of the CSR sample vould also emphasize the types of items which had previously However, for each of exhibited deficiencies at Byron and Braidwood.
the electrical construction categories it was determined that none of the previously identified deficiencies could be limited to a subset of Therefore, additional items were chosen the construction category.
Kaushal, Reb. Prep. Test. at 14-15, ff. Tr.
using random methods.
13068.
Although both the " random" and " engineering judgment" portions of the CSR sample already included more highly-stressed items, in the cable pan hanger category,10 additional more highly-stressed items Kaushal, Reb. Prep.
I were added as the third part of the CSR sample.
Under the Test. at 15-16, ff. Tr. 13068; Kostal, Tr. 15074-75.
provisions of the CSR, if any design significant discrepancies had been found in the initial CSR sample, the sample size would have been These sample expansion provisions could have led to a 100%
increased.
reir.spection. However, since no design significant discrepancies were Kaushal, Reb. Prep. Test, at identified the sample was not expanded.
6-7, ff. Tr.13068; Tr.13756-57,14148-49.
9 o.
250 341.
. The CSR inspection checklists and instructions were developed by the BCAP Task Force engineers based on relevant design The attributes selected for reinspection infonnation provided by S&L.
were .those that (1) are required by applicable codes and standards, (2) potentially have an 'effect on the item's ability to perfonn its safety-related design function, and (3)are currently observable.
Kaushal, Reb. Prep. Test, at 18-19, ff. Tr. 13068.
In the electrical
' construction categories, the CSR checklists and instructions were not based on the original Comstock inspection checklists, and did not include attributes with no potential for design significance.
Kaushal, Tr. 13180-86, 13375, 13385.
342. The CSR electrical sample was chosen from items that 30, 1984. f had been Quality Centrol inspected and accepted as of June Edison estimated that only 24% of the total research construction items in the plant were " valid" and had been Quality Control inspected
]
and accepted as of June 30, 1984; only those items were therefore App. Ex. 133, Int. Ex. {
eligible to be included in the CSR sample.
159. Thus only 17% of conduit hangers. 29% of electrical equipment installations, 39% of cable pans, 42% of conduits, and 59% of cable App. Ex.
pan hangers were eligible to be included in the CSR samples.
133; Int. Ex. 159. The remaining 76% of the electrical construction items in the plant (some 72,216 items out of total of 94,947 electrical items.in the plant) were ineligible for the CSR samples and were thus not covered by the CSR program at all.
J
251 The fact that the CSR program covered only 24% of the 343.
total electrical construction population at Braidwood limits the program with
.overall conclusions that can be drawn from the BCAP The CSR cutoff date respect to Quality Control inspector performance.
of June 30, 1984 bears no relation to Intervenors' contention Many of the incidents :
concerning harassment and production pressure.
exhibiting harassment of production pressure that have been developed in this record occurred after June 30, 1984:
the backlog was
- The Comstock campaign. to eliminate During reaching its most intense period in June 1984.
that month, DeWald received a memorandum from Shamblin ,
emphasizing the urgency of eliminating the backlog and announcing weekly meetings for progress reports.
- The termination of inspector Puckett, arguably the most egregious incident of harassment in this extensive record, occurred in August 1984.
Comstock Quality Control
- On March 29, 1985,- 24 inspectors went to the NRC to complain about problems in at Comstock, including production pressures that, their view, placed an emphasis on quantity over quality in the Comstock Quality Assurance organization.
- Allegations of harassment and as production pressure exemplified by the continued well into 1986 retaliatory incidents involving Richard Partin and Gregory Archambeault.
Early in the CSR program the NRC Construction Assessment Team identified deficiencies on three of six pipe supports / restraints which The BCAP the BCAP Task Force inspectors had previously reinspected. ,
In Task Force reinspectors had not identified these deficiencies.
252 addition, IEOG overinspections identified deficiencies' associated with a concrete placement which had not been identified during the BCAP CSR In reinspections. Gardner, Reb. Prep.- Test. at 8-9, ff. Tr.17606.
response to these findings and following a meeting with Mr. Gardner on January 23, 1985, Mr. Kaushal temporarily suspended CSR reinspections.
Gardner, Reb. Prep. Test. at 8-9, ff. Tr.17605; Kaushal, Reb. Prep.
Test. at 21-22, ff. Tr. 13068; Int. Ex. 148. Corrective actions were taken to address the identified CSR reinspection discrepancies and to ensure that future.CSR reinspections were performed in an acceptable manner.: These actions included the partial repeat reinspection of previously reinspected mechanical pipe supports, the review of electrical conduit support packages, _ and partial repeat reinspection of such supports, where necessary, the implementation of additional training for BCAP inspectors, the revision and clarification of BCAP checklists and instructions, and the initiation of. the BCAP Quality Control overview of BCAP Task Force inspections. Gardner, Reb. Prep.
' Test, at 9-10, ff. Tr. 17606; Staff Ex. 25 at 5; App. Ex. 135; Wozniak, Reb. Prep. Test, at 5-7, ff. Tr.13068; Smith, Reb. Prep.
Test, at 7-14, ff. Tr.13068.
344. Kaushal believed that the root cause of the CSR reinspection errors identified by the CAT and the IE00 prior to January 23, 1985 was a misunderstanding by the BCAP Task Force Tr. 13941-42.
inspectors of certain attributes on their checklists.
Mr. Gardner, on the other hand, concluded that the root cause of these
COMi.O !.
d F53 gro at whicF LCAP Tutt CSR reintrection deficiencies wc; the tast Gardncr, Reb. Prep, Test, tt 10, ff.
Force inspectors were working.
Tr, 130f8, Although Mr. Gardner did not discast this curcern with Kaushal in their meeting on January 23, 1965, or document it in nis Inspection reports, he continued to monitor BCAP inspectors' 6ttitudes Subse6uently the CSR reinspectors were and instructicns. Tr.18369.
cs instructed to disregard tny race ccr.cerns anc take as mu:h tir;e Tr. 17023-24 necessary to perfonn their inspecticns. j 315.
Three types of d;ta were produ;ec as a result of the BCAP CSR program.
The first, the raw cata frcm the CSR reinspectiors sere tabulated w terms of the number cf the ciscrepancies and tne nuce- of acceptcble concitiens identifica by the CRS overinspectcrs.
Lecond, thase
- r. umbers were used t'a c mpute :o-called "eareerrent Thirc the discrepancies were analyced to deter:rine wretner r& tes .
tne;y were cesign significant.
546.
All CSR reir: spec +. ion ottervations reportac by the ECAP inspectors were reviewed Dj thcir l e:.d disci plir.e hss force Vaushal, Reo.
intrectors for c' c ri ty , ccmpleter.ess and .iccurMy.
at 22-25, ff. Tr. 130CF. If suiteble for furth<:*
Prep. Test.
prccessing, tr e rtservn'.icts were wa it.a tea icr vn' i di ty Lj L 3 f'
, :. ^ ,. > , C5F OhSertbt1Dn5
( ne r BCAP ;.rDc9dures ( ! !:'.
c091neers.
i t t. c aritra c t ,.ir',
v.! i cit hcJ pr:.vicatly tier ider;t i t .e6 b, , pplie; Pot c<
- ,tn4:r controll+.d system wers.
tn en e/,11 tir g encen'omco ,r r t-pc-
.o .
254 considered to be invalid. Conditions that were in accordance with current design documents or design documents current at the time of the original Comstock inspection were also not valid. Kaushal, Reb.
Prep. Test. at 23. ff. Tr.13068; Int. Exs. 143, 154; Tr. 13588-603. ?
Observations that related to items not within the CSR sample or attributes not on the CSR checklists were declared "out of scope." In l
addition, because the objective of.the CSR was to look for previously unidentified and unaddressed construction problems, observations which pertained to known conditions addressed prior to the CSR through existing procedures or other documented plans for future construction completion activities (for example, all cable pan hanger configuration observations) were also declared "out of scope." Kaushal, Reb. Prep.
Test, at 22-24, 26-27, ff. Tr. 13068; Tr. 13535-38, 13799-802; Int.
Ex. 143. The remaining (valid, in-scope) observations were termed
" discrepancies" and were transmitted to Sargent & Lundy for evaluation of design significance. Kaushal, Reb. Prep. Test. at 2.5-26, ff. Tr.
13068.
347. Early in the CSR program, Sargent & Lundy engineers reviewed each discrepancy sent to them by BCAP for validity as well as I for design significance. However, in March 1985, NRC Inspector Gardner assessed an item of noncompliance against BCAP for invalidating 37 so-called " red-line" observations all relating to a Phillips-Getschow documentation practice, based on an inadequate rationale provided by S&L. Although the focus of the NRC Staff's
I I
i 3
255 .{
concern was the invalidation itself, rather than S&L's role, after this time BCAP-06 was modified to emphasize that S&L could only recommend invalidation and only the BCAP Task Force could invalidate BCAP observations. Thereafter, S&L played little or no role in the invalidation process. Gardner, Reb. Prep. . Test. at 11-12, ff. Tr.
17606; Gardner, Tr. 17764-67, 18328-34; Kaushal, Reb. Prep. Test. . at 25, ff. Tr. 13068, 13489-503, 13828-34, 14343-45, 14476-77.
348. Sargent & Lundy categorized 'all discrepancies sent to it for evaluation of design significance as either: " insignificant,"
" notable." or " design significant," depending on its severity. App.
Ex. 179 at 15; Thorsell, Reb. Prep. Test. at 9-10, ff. Tr. 14270.
Discrepancies which reduced an item's capacity by less than 10% but did not impair its ability to perform its . safety-related design function were termed " insignificant." Discrepancies which reduced.an item's capacity by 10% or more but did not impair its ability to perform its safety-related design function were termed " notable." Any discrepancy which would impair the item's ability to ' perform its Mr. Gardner safety-related design function within code allowable stresses was called " design significant." Thorsell, Reb. Prep. Test.
at 9-10, ff. Tr.14270; Kostal, Reb. Prep. Test, at 16-17, App. Ex.
179; Kostal. Reb. Prep. Test. at 28, ff. Tr. 13068. Sargent & Lundy's evaluatica of discrepancies for each of the six electrical construction categories concluded that there were no design significant discrepancies.
256 349. The Board - heard substantial testimony .regarding S&L design significance evaluations for CSR discrepancies. ' CSR sample items cable pan hanger ("CPH") 104, and Cable- ("CBL") 130 were vehicles for a comprehensive evidentiary review of S&L's approach and methodology; see, generally, Kostal, Tr. 14641-86, 14755-805, 15517, 16675-76; Thorsell, Tr. 14453-60, 14477-90, 14565-66; Int. Ex. 155, ISSA, 155B , . App. Exs. 159, 173, 180. Sargent & Lundy, initially calculated the design margin for- CPH - 104, taking into account CSR identified weld discrepancies, to be 1.03. Tr. 14781-83; . Int. Ex.
155B, pp. 14-15. Any v'alue equal to or greater than 1.0 is not design significant and therefore acceptable. Tr. 14781; Int. Ex. 161.
Subsequently, a revised calculation was performed using the actual cable tray weights that existed in ,the pan, rather than the conservatively estimated load use in the initial calculation. Tr.
14756, 14784-85, 15181-82; App. Ex. 159. That calculation resulted in a design margin of 1.89, but an improper shortcut was taken in the second calculation. Correcting for the shortcut, the design margin was calculated at 1.28. Tr. 14781-84. With respect to Cable 130, Sargent & Lundy erred in closing out a minimum bend radius violation observation on the basis of technical acceptance criteria contained in a letter from the cable manufacturer, Okonite Company, without firs t pointing that cable cut to the manufacturer's specifically representative, or providing a written description of the bend radius violation to the manufacturer. Thersell, Tr. 14482-83.
e .
257 COPPECTED
. 350. The criteria in the letter 'cr approval of the bend radius did not apply to Cable 120 without a further determinatiori by the manufacturer of the condition of the cable, and a dif ferent cable was examined by the manufacturer's representctive, than assumed by Tr. 14456-62, 14482-89, 14565-67. The errors in Sargent & Lundy.
both the CPii 10't and Cable 130 design significant evaluations were not ciscovered and corrected until the SSL experts were cross-examined by Intervenors' ccunsel at hearing.
351. The q"ality control inspection of cn item such as a cable or a cable pan hanger requires the inspector to verify that the item conforms to cesign requirements fcr each ettreute on his checklist. Verification of each such. attribute may require one or Kaushal Ret. Frep. Test, at 19-20, ff.
more inspection judgrrents.
Tr. '.3068; Tr. 137Cl-62. Moreover, the items incluced in the CSR sample varied greetly in their complexity and thus in the number of fer init iai Oualfty Control inspe: tor judcmenti recuirr.c tne T r . 13758-59, 1>,166- 73 ; App .
insr.ection ano f or tr>e C3R reinspection.
En . 143, 14a. Tc pennit meaningful juegment -f inspec:or performance 2nd treaningf e; comparisen of inspector perfccmarce with respe:.- ::
items of Mering complexity, the ECAP Tcsk Force tcgethe with LCAP uncy deveioped te L concept of Sudlity A::urance aric Sargent &
~' U I k - 5 ':: , L: 7 /fJ ,
"in;pOctior foints" and Ni E repan',j PDDM.
facn in5pectOr ( NC k to Cetsrnin? 'h> P CCei ti:D !Ii b
. ;, / . ' ; , ]4]73 ~9.
c.r . ejectabl iity o' en item or ar, Ottekte ws identi fied arc termec
i i
.c 153 CORRECTi.D en "inspectico point." Each inspection point wrich resultef in a ZP discrepancy was t'rmed a " discrepancy point." Kausnal, Rcb. P r e ;, .
Test. at 19-20; ff. Tr. 13068; Kcstal, Reb. Prep. Test, at 13-14, ff.
Tr. 14270; Tr. 13760-6'. On this basis, over 9FA of the inspecticr points were found to ce ccerect (nondtscrepant) and re thun two-thirds of the discrepancy points were insignificant. App. E:< . 1 ~9 at 16; Thorsell, Reb. Prep. Test, at 11, f f. Tr. 142,7,; Lcstal , Rea, Prep. Test, at 22, ff. Tr.14270.
l i
352. Applicant alsc presented tna CSP. results fcr the electrical construction categcries on a per veld basis. About 84% of
+ne welos had nc uiscrepancies. Acp. Ex. 192.; Reb. Prep. Test, et 33, ff. Tr. 16740. The caraparable figure for the PTL overinspa:tica data f or the period July 1,19F2 to June 30, 1965 is 33%. Fcr the perix in which the two cata bases %erlap (July 1, bE2 to June 10. 1966),
the agreemert rates are 89% and 90%, respectively. DelGecr;e, Reb.
Orep. Test. at 37-38, ff. Tr. 1674 , 1660',-C2.
OF3. A tnird say of locking et the CSR rescit: was supported by intervencrs in this proceeding. A n '7 item witr one or more discrepancies wcuid be terr.ec a "aiscrepant iter, ir e TEC hac originally required that any cenc;usions cc exNnding the C3? s r*
size be based on tne percentge of acaptctle 1te:6. irres,;cc ti ve M the number of .tr i bt: M i e ' r.gettic r. peint; . Irt. E.. 10 ct E .' AF Co=1:en ts !I-4; Tr. .71J-l'. 4plican*. c ra i t t ed itsei' t; this
1 259 I
l requirement. App. Ex. 128 at Atachment A, p. 3 of 7. Although Applicant's statistician did some early analyses based on an item, rather than inspection point, basis, Applicant inexplicably breached ,
its commitment to the NRC and abandoned that basis. Tr. 17141-42, {
17631, 17710-18. On an item basis, 60.0% of the cables, 64.4% of the cable pans, 59.0% of the conduit, 56.4% of the conduit hangers, 86.2% j of the cable pan hangers and 72.5% of the electrical equipment installation would . be deemed " discrepant i tems .'? App. Ex. 181.
Applicant's witnesses did not view this as a reasonable or fair measure of Comstock Quality Control inspector performance _-- both because it masks the actual number of inspector errors on each item and because it equates very dissimilar reinspection outcomes. For example, a huge cable pan hanger with hundreds of welds, one of which might be discrepant due to an arc strike, would count the same as a conduit wall strap support which was totally missing. Kaushal, Tr.
13758-59; Shevlin, Tr.13770; Kaushal, Shevlin, Woznial ad Smith, Tr.
14173, 14179.
354. NRC inspector Gardner agreed that in grading inspector J
performance he would not equate such dissimilar " discrepant items."
Mr. Gardner did not believe that Applicant Exhibit 181, standing along, presents a balanced portrayal of the CSR reinspection results.
Nonetheless he recommended that the Licensing Board should consider
~
all the data available to it, including the data presented on an item basis. Mr. Gardner stated that his own personal standards were high,
260 l
and he would expect a good inspection program would have resulted in lower rates of discrepant items than as .hown s in App. Ex. 181. l However, he conceded that he had never developed acceptance criteria ;
for differentiating good from average or poor inspection programs using data presented in Intervenors' suggested " item basis" fonnat.
1 Tr. 17633-45, 17807-11, 18347-49. In Mr. Gardner's view, the Comstock Quality Control inspectors were not effective in the " classical" ;
\
10 C.F.R. Appendix B sense of identifying all defects, but they were j effective and adequate in the sense that they did not miss any design
'significant' defects. Tr. 17807-09, 17813-15. {
l 355. There is merit in both Applicant's and Intervenors '
positions. Clearly, if we are dealing with a complex component containing a number of welds, each of which are evaluated on the basis
)
l of,17 design significant attributes, it would be unrealistic to judge the original Comstock inspection as a failure if one attribute on the j
cibponent were discrepant, as Intervenors suggest. On the other hand, l
jut}ging the quality 'of the original inspection en the percentagp of l
agributes that were discrepant, as Applicant proposes, is simil!)rly undealistic. As an example, welds were divided into 17 inspection poi!)ts (or attributes). It seems unlikely that any weld that had more I
thanI two or three discrepant inspection points (f.e., attributes) l l P wculd have become the subject of an original inspection by an L. K.
r Ccmstock Quality Contre 1spector. If a craf tsman were to weld a f ,,
~
- weldment with more than two or three f aulty attributes, such as being
4 s 261 undersized or cracked, lacking fusion, etc., it is likely that he would redo that weld himself without waiting for quality control to reject it. On a practical level then, the original Quality Control inspector is inspecting welds that might have, at most, one, two, or three defective attributes (although any of those, such as a crack, But, even if we might render the weldment totally nonfunctional).
were to assume that the Quality Control inspector inspected and passed only discrepant welds (those with one, two, or three defective attributes), his percentage of acceptable calls (i.e., his " agreement On its face, an rate" under BCAP) would range between 82% and 94%.
82% to 94% rate does not seem egregious, even though it should because, in our example, the Quality Control inspector missed every single discrepant weld that the craftsmen would not have redone of their own volition.
356. There are infirmities in the BCAP CSR reinspection beyond the question of whether components, program that go sub-components (such as welds), or inspection points should be tallied I l
to determine the percentage of discrepancy. Even if we were to choose )
one of these, we would still lack the perspective to judge the quality ]
of the original Quality Control inspection. The main element lacking '
in the evaluation would be the number of the discrepant items (components, sub-components, or attributes) that the original Quality Control inspector reported, as opposed to those that he missed, only the latter being disclosed under the BCAP program.
i
4 .
262 357. As an example, let us use welds as the unit of measurement and 15% of the welds as being found discrepant under the BCAP reinspection program. (Applicant's Exhibit 181 indicates that approximately 16% of the welds examined by the BCAP inspectors were found to be discrepant.) If we assume that the craftsmen had welded 45% of their welds discrepantly, the Comstock Quality Control inspector would have had to miss one-third of those discrepant welds (1/3 x 45%) to have been found 15% discrepant under BCAP. If, on the other hand, the craf tsmen had welded 20% of the welds discrepantly,-
the Comstock Quality Control- inspector would have had to miss
=
three-quarters of the discrepant welds (3/4 x 20% 15%).
Consequently, unless we know either explicitly or deductively (or indeductively, as the case may be) how many discrepancies were reported by the original QC inspectors, we do not know whether the Comstock QC inspectors were 67% effective, 25% effective, or any other percentage.SI SI - The calculation made in this example is somewhat simplified.
Since I do not distinguish between weld discrepancies missed by an original Comstock QC inspector and any Comstock reinspector, but use only the final products of their cumulative inspections, it is unnecessary to adjust the calculation for welds that were reinspected, as in the examples presented at hearing where the examples began with a hypothetical 100 welds to be initially inspected.
I also do not take into account in this simplified calculation the possibility of the BCAP reinspectors' not being 100%
accurate. I recognize that they could be expected to have missed
[ FOOTNOTE CONTINUED]
e 263 q 358. There would seem to be no reason why the discrepancies uncovered by the BCAP reinspectors could not be compared to the ,
}
discrepancies originally reported by the Comstock inspectors , as l
- 1 contained in the inspection packages for the sampled components. 1 d
Under the requirements of Part 50, Appendix B, Criterion XVII, the J original inspection records should be retrievable. Criterion XVII states, inter alia:
i Criterion XVII. Quality Assurance Records:
... inspection and test records shall, as minimum, identify the inspector or data recorder, the type of observation, the results, the. acceptability, and the action taken in connection with any deficiencies noted. Records shall be '
identifiable and retrievable.
4 359. It would appear that even at this point a comparison can be made between the discrepancies found by the BCAP inspectors and those found by the original Quality Control inspectors. We need only
[F0OTNOTECONTINUED]
discrepant welds, as offered by Intervenors and Applicant. I do not also accept the proposition offered by Applicant that these i BCAP reinspectors would have erroneously reported nonexisting discrepancies under a rate comparable to, or greater than, that of missed discrepancies. I do not believe that one could expect inspectors to find things that weren't there, except in unusual circumstances. That does not go to say that there might not have been differences in judgment between the BCAP reinspector and the ,
original Comstock QC inspector, but I would expect that any i errors in judgment on the part of the BCAP inspector would have l been weeded out in the BCAP review that he was subject to which, in all probability, also weeded out any marginal calls he made, f even if correct. ]
\
l
264 examine the original sampling packages, with no need for any further sampling, if we wish to measure the effectiveness of the original Quality Control inspectors. Whether any such comparison was ever made has not been disclosed and is not a part of the record. In the absence of such comparison the BCAP program cannot be accepted as any measure of the effectiveness of the original Quality Control inspector. It might also be noted at this juncture that if a comparison had been made, the entire controversy over which units (i.e. components, sub-components, inspection points) should be measured would have been obviated. Had Applicant compared only those attributes examined by the BCAP reinspector with the comparable ,
attributes originally inspected to by the Comstock Quality Control inspector, Intervenors would have no basis for challenging the results. Of course, even if Applicant had measured apples against apples and oranges against oranges, it would only have arrived at a percentage of the effectiveness of the original Quality Control inspector. Unless those results were determinative on their face (i.e., either an extremely high rate of Quality Control inspector effectiveness or an extremely low rate), the results would still have i
to be evaluated by the experts and the Board.
360. Without any measure of effectiveness of the Quality Control inspector, and with only a measure of the absolute numbers of ,
discrepancies missed, a meaningful comparison cannot be made between different periods of inspection activity. Moreover, any BCAP sampling i
i
,a. ..
265 comparison'between the pre-DeWald (as Quality Control Manager) era and the period. in which' the contention alleges .that management harassed and intimidated inspectors, is parti::ularly inappropriate. Mr. DeWald became Quality Control Manager in August of 1983, shortly before the grid area basis for weld inspectors was discontinued in October of 1983. The grid system was not a proper or effective method of inspection (seeMin.FF 315-329, supra) and, consequently, neither the grid system period nor the DeWald-Saklak period represents a standard against which any other period can be judged.
361. In the absence of any measure of Quality Control l
effectiveness based upon a comparison between discrepancies missed and discrepancies reported, the BCAP evaluations of. " design significance" .
were presented as.a measure.of Quality Control effectiveness. But the question of whether a discrepancy is " design significant," is totally irrelevant to the function of a Quality Assurance inspector. He is not charged with seeking out design significant discrepancies or even with determining whether any putative discrepancies are significant from a safety standpoint. His obligation is to report all discrepancies. Any attempt by him to ignore those discrepancies that p
he might consider insignificant would interfere with this obligation.
l The question of whether a discrepancy is design-significant is uniquely in the presence of an engineer to evaluate based in part on the inspector's findings but also based on a variety of other data and expertise that as not imediately known to a quality control
266 inspector. The measure of the qualification of a quality control inspector is whether he can inspect to established acceptance )
criteria. Tr. 16775-76.
362. The only value, therefore, that BCAP could have for ub considering the way it was programmed, is with regard to the constructed hardware, rather than with regard to the effectiveness of the Quality Control Inspection Program. However, even there little weight can be given to the results. The main problem here is with the the party selected-to make the determination of. design significance, Sargent & Lundy.
363.- The BCAP program document recognized the need for independence of- the Independent Expert Overview Group reviewing the program. The document provided that the IE0G members "will be free of any significant contacts with Commonwealth Edison Company" and "will not have participated in the design, construction, or quality assurance activities related to the Braidwood Station or with Braidwood side contractors within the last five years." App. Ex. 137 at V-2. The IE0G was not shown at hearing as being any more than a token oversight group. However, the BCAP Director, Mr. Kaushal, was an Edison employee. He and BCAP were answerable directly to Edison management in the person of Mike Wallace, the Braidwood project manager who was responsible for ccst and scheduling considerations at Braidwood. Kaushal Tr. 13716. More importantly, Sargent & Lurdy, f f
I
< i
267 design-significant evaluations, the only which performed the evaluationsofanyimportanceEl in the BCAP program, did not meet the ,
independence criteria. Sargent & Lundy failed the independence ,
criteria on almost all grounds. As architect / engineer, it designed As Braidwood and was intimately involved with its construction.
f consulting engineer, it advised on construction and dispositioned NCRs and ICRs that documented discrepant construction. activities. It was .
in day-to-day contact with Applicant, the Commonwealth Edison Company.
Because of its intimate involvement in the construction activities, whether 'or not it was the case, Sargent & L' undy sppeared to the Quality Control Manager of L. K. Comstock to be the prime electrical Mr. DeWald, Tr.
contractor and Comstock only the subcontractor.
1805-06. Were the Braidwood facility to fail to meet its licensing requirement or were its construction to prove deficient in some degree, it is likely that Sargent & Lundy's liability exposure would surpass that of even Applicant, the Commonwealth Edison Company.
for design 364. An example of Sil's direct liability significant defects is Cable .130, which violated the manufacturer's l
El It would not have mattered cne iota whether there had been half as many.or twice as many discrepancies or notable discrepancies found by the BCAP inspectors. The entire conclusion as to whether the facility passed the BCAP test was founded on S&L's design determining whether any of the discrepancies was significant. ,
268 bend radii limitations and which S&L evaluated (incorrectly, at first) as having no design significance. The bend radii's violation was attributable to the cable being placed in a junction box that was too small to permit the cable to be bent properly. S&L had designed and j ordered the equipment and installation, and had failed to require an adequate junction box. Tr. 14923-25.
365. Not only would Sargent & Lundy have failed the independence test in the BCAP program document, but other reasons exist for questioning the objectivity of its evaluations on the BCAP.
Just prior to the critical period in issue in this proceeding (beginning with the summer of 1984), Sargent & Lundy had complained that the Comstock Quality Control inspectors were being "over-critical and were marking discontinuities which S&L felt were acceptable."
App. Ex. 1 at 1. Sargent & Lundy had complained in particular about the Quality Control inspectors' interpretation with regard to overlap, undercut, arc strikes, lack of fusion, tack welds base metal reduction and spatter. Sargent & Lundy was concerned that these interpretations constituted "overinspection." Ibid.
a .
269 i
i 366. Consistent with the evidence of Sargent & Lundy's concern about inspections being too critical, was the Ouality Control inspectors' recurrent testimony during the course of the hearing that they were becoming somewhat demoralized because of Sargent & Lundy's practice, as evaluating engineer, of dispositioning QC-determined discrepancies on a "use as is" basis. Tr. 8162-64, 10485, 10576, 12320, 12632, 17363; Test. of Mendez and Neisler, ff. Tr. 10490 at 30.
Many of the Quality Control inspectors believed these dispositions to be unjustified and had voiced complaints to the NRC. Ibid. Some had successfully challenged the S&L engineers' "use as is" dispositions. -
Tr. 8162-64.
367. Further disquieting about the role of Sargent & Lundy as an objective evaluator' under BCAP were its participation in the improper termination of Mr. Puckett and its testimony in defense of that termination, its errors and evaluations of the two sample BCAF packages randomly selected by Intervenors for examination at hearing, its improper invalidation of 37 red-line drawings under BCAP, its complaints to Comstock's management concerning Quality Control inspectors who sought engineering advice from them for not having gone through channels (i.e._, through their Quality Control supervisors with whom they disagreed), its unilateral departure from FSAR standards (e.g., with regard to response spectra (see Tr. 15176-79,15197-201))
in evaluating BCAP design significance, and the seeming inability of
I 270:
Sargent & Lundy witnesses to ' answer Intervenors' questions directly with regard-to the BCAP program.
368. As 'a general matter, by virtue of its direct involvement in the design and construction of the Braidwood plant and its potential liability to Commonwealth Edison for any construction or licensing problems,' Sargent & Lundy is too comitted to the licensing of the plant to be considered an objective . evaluator. While it is certainly entitled to evaluate the plant's construction under BCAP or any other program for. its own purposes to determine, for itself whethe,r the plant is properly constructed, its comitment to the licensing of the plant is too strong for acceptance of its opinions as impartial.
Furthermore, its past actions and testimony at trial confirm its partisanship in that regard. Its attitude in general appeared to be that it had designed the plant with so much safety margin that no deficiencies in construction and inspection in the electrical area could impair the ability of the electrical equipment to function safely. While that might be the case, that opinion should be expressed by someone other than the designer of the plant to be afforded much weight.
369. Further questions exist with regard to whether the by Sargent & Lundy are design significant evaluations made satisfactory samples for statistical application. In the statistical process, one can select sufficient items on a random basis to project
271 to the population at large. The population being sampled, however, must have a degree of homogeneity in order for that statistical projection to be valid. But in this case, the calculations and for design significance appear to be unique evaluations made calculations suitable only for the particular items selected. While evaluations with standard design Sargent & Lundy began its calculations, it departed from these standards through a series of so-called " refinement" when the design margins became minimal.
Sargent & Lundy's design significance calculations were carried out by using successive levels of " refinement." When one set of calculations l produced results that indicated a concern about design significance, Sargent & Lundy turned to a "more refined calculations" which, by eliminating purported conservatisms in the first set of calculations, enhanced the acceptability of an item. Sargent & Lundy employed multiple layers of refir.ement in order to arrive at its conclusions Tr. 15076, 14083-85.
that no discrepancies were design significant.
These refinements took many forms, including examinations of the "as-buil t" configuration of the sample item and its neighbors to determine if additional safety margins exist, departures from the FSAR specifications to those based upon its own engineering judgment to see if further safety margins exist, and departures from the equipment manufacturers' specifications and requirements on the basis of its own engineering judgment to detentine if further safety margins exist.
Many of these departures from the original design specifications adopted in the FSAR were based on a_d hoc exercises in engineering .
a w 272 judgment and all were in the direction of finding additional safety Absent were any suggestions of margins in the as-built construction.
refinements . in the direction of reduced _ safety margins because of as-built conditions that might have included observations of less than item, or in a
' satisfactory workmanship or materials in the sampled Given l neighboring item that might adversely affect the sampled item.
the predisposition of Sergent & Lundy to validate the construction of.
Braidwood as satisfactory, which was the expressed purpose of the BCAP program, and considering Sargent & Lundy's resourcefulness and their predisposition for searching only for matters that would show an increased safety margin, it is difficult to see how they would ever
' find a discrepancy of design significance.
370. For the same principal reasons that the CSR agreement rates are not indicative of the efficacy of the original Comstock Quality Control inspector, because there is no. comparison between the discrepancies he missed and those that he found, the PTL results are similarly unilluminating. Furthermore, in addition to the sampling's not being done on a statistically random basis, there is further doubt I The PTL inspectors were !
as to how representative the sampling was. )'
Although the PTL panel pennitted to overinspect welds through paint.
experts claimed that this amounted to only 7% of the inspected welds, this testimony was questionable.
=.. ,
i 273 371. They arr'ived at the 7% figure by reviewing PTL's.
overinspection records for July .1982 through June 1986 and detennined how many - of the welds were noted in the remarks section of the ;
inspection reports as having been inspected through paint. This i amounted to 7% of the total welds that were.overinspected. The PTL witnesses testified that it was PTL's practice, although not a written l
. procedural requirement, to indicate in the remarks section of PTL's inspection reports which welds were overinspected through paint. They f i
believed that the PTL inspectors follwed this practice whenever they l l
inspected through paint. Tr. 15749-54.
l 372. That testimony is not acceptable. On its face, the 7% !
figure seems very low considering the practice of Comstock of coating welds with Galvanox after the initial installation and Quality Control 4 8541. Galvanox was a heavy, thick, inspection. Tr .. 8533, gray-colored paint' used a protection coating to prevent welds from !
rusting. Tr. 8531, 8540. It would be less surprising if the. figure l given for welds covered by Galvanox by the time of the PTL overinspection were 70%, rather than 7%. Furthermore, not only did l PTL's written procedures not require noting the welds overinspected through paint, but neither did the checklist given to the PTL overinspectors. Tr. 15780-81. Nor was there any other written direction to note those inspections through paint. Tr. 15782. But PTL, no less than the original Quality Control inspection group, is ;
l i
l
,u : 2, 274 required by Part 50, Appendix B, to document in writing its procedures and instructions. Criterion V, states, in pertinent part, as follows:
Criterion V. Instructions, Procedures and Drawings:
Activities affecting quality shall be prescribed by documented instructions ...
Criterion XVIII states, in pertinent part:
Criterion XVIII. Audi ts :
The audits shall be performed in accordance with the written procedures or check lists by appropriately trained personnel ..."
373. If PTL, which had been overinspecting visual
'insr ections of welds since 1977, had not memorialized any directions to its overinspectors to note welds inspected through paint (as it had documented its other requirements) by the period for which it offers its conclusions, 1982-1986, we cannot accept the testimony that these instructions existed and were uniformly applied. And, if in fact a large number of welds were inspected through paint, PTL's high agreement rate with Comstock Quality Control inspectors me6ns very little because many discrepant attributes would have been obscured by the paint. The Galvanox coating could obscure cracks, undercut, cold lap, porosity, and other attributes. Tr. 8531-32. Moreover, one the sampling by PTL, which was not cculd not be sure that e , j
Is UNITED STATES OF AMERICA +
NUCLEAR REGULATORY COMMISSION BEFORE THE ATOMIC SAFETY AND LICENSING AEOEAU'BdkRb In the Matter Of: ) b0t
) Docket Nos. 50-456 COMMONWEALTH EDISON COMPANY ) 50-457
)
(Braidwood Station, Units 1 and 2) )
CERTIFICATE OF SERVICE I hereby certify that I caused copies of the foregoing OPENING BRIEF OF INTURVl!NORS-APPELLANTS BRIDGET LITTLE ROREM, ET AL. to be served on ,ill parties on the attached service list by depositing copies in the U.S. mail, first class postage prepaid this 1st day of July 1987, except that counsel for Applicant were sened by personal delivery and counsel for Staff by Federal E)' press on thi s same date.
4 Robert , / Jo es , Jr.
Douglass W. Cassel, Jr.
Robert Guild Robert L. Jones, Jr.
Business and Professional People for the Public Interest 109 North Dearborn Street Suite 1300 Chicago, Illinois 60602 1
~
< , r.
1 BRAIDWOCD SEPVICE LIST Gary J..Edles, Chairman Michael I. Miller, Eso.
arJ Administrative Judge Peter Thornton, Esq.
Atomic Safety and Licensing Ishem, Lincoln & Beale Appeal Board Three First National Plaza U.S. Nuclear Fegulatory Commission Chicago, Illinois 60602 Washington D.C. 20555 Docketina & Service Section Christine N. Kohl, Esq. Office of the Secretary Administrative Judge U.S. Nuclear Regulatory Atomic Safety and Licensing Commission Washington D.C. 20555 Appeal Board U.S. Nuclear Pegulatory Commission 20555 C. Allen Bock, Esq.
Washington D.C.
P.O. Box 342 Urbana, Illinois 61801 Dr. Reed W. Johnson Administrative Judge Atomic Safety and Licensing Bridget Little Rorem Appeal Board 117 North Linden Street U.S.. Nuclear Regulatory Commission Essex, Illinois 60935 Washington D.C. 20555 Thomas J. Gordon, Esq.
Waller, Evans & Gordon Ivan. Smith, Esq.
Chairman and Administrative Judge 2503 South Neil Atomic Safety and Licensing Board Champaign, Illinois 61820 U.S. Nuclear Regulatory Commission Washington D.C. 20555 Lorraine Creek Poute 1, Box 182 Manteno, Illinois 60950 Richard F. Cole Administrative Judge Atomic Safety and Licensing Board Region III Office of Inspection &
U.S. Nuclear Pegulatory Commission Washington D.C. 20555 Enforcement U.S. Nuclear Pegulatory A. Dixon Callihan Commission Administrative Judge 799 Roosevelt Road Glen Ellyn, Illinois 60137 102 Oak Lane Oak Ridge, Tennessee 37830 Atomic Safety and Licensing Stuart Treby, Esq. Board Panel l U.S. Nuclear Regulatory NRC Staff Counsel Commission U.S. Nuclear Pegulatory Commission 20555 7335 Old Georgetown Road Washington D.C.
Bethesda, Maryland 20014 Atomic Safety and Licensing Joseph Gallo, Esq. Appeal Board Tcham, Lincoln & Beale U.S. Nuclear Regulatory 1150 Connecticut Avenue N.W. Commission '
Washington D.C. 20555 Suite 1100 Washington D.C. 20036
- s
\
BRAIDWOOD SERVICE LIST Robert Guild 4 413 Pall Mall l Columbia, S. Carolina 29201 William Little, Director Braidwood Project Region III United States Nuclear Regulatory Commission 799 Roosevelt Road Glen Ellyn, Illinois 60137
' Charles Jones, Director Illinois Emergency Services and Disaster Agency 110 East Adams i Springfield, Illinois 62705 Ms. Janice A. Stevens l
United States Nuclear Regulatory Ccmmission 7920 Norfolk Avenue Bethesda, Maryland 20014 George L. Edgar, Esq. '
Thomas A. Schmutz, Esq.
1615 "L" Street, N.W.
Washington, D.C. 20036 1
l
- - - - - _ ___ - ______ - -___ - __ - _ _ __-___ - -_ - _ __ _ _ _ __ _ - _.