ML20207M194
| ML20207M194 | |
| Person / Time | |
|---|---|
| Site: | Braidwood  |
| Issue date: | 01/05/1987 |
| From: | COMMONWEALTH EDISON CO. |
| To: | |
| Shared Package | |
| ML20207M066 | List: |
| References | |
| OL, NUDOCS 8701130114 | |
| Download: ML20207M194 (180) | |
Text
r i
January 5(KEILi 1987 00 US14RC UNITED STATES OF AMERICA NUCLEAR REGULATORY COMMISSION 87 JAN -8 P4 :52 BEFORE THE ATOMIC SAFETY AND LICENSING BOARD OFFICt G
~~A:
00CnEfIN;a x fWCf.
EF M:3 In the Matter of
)
)
COMMONWEALTH EDISON COMPANY
)
Docket Nos.30-456
)
50-457 (Braidwood Station Units 1 and 2
)
FINDINGS OF FACT AND CONCLUSIONS OF LAW 1.
Commonwealth Edison Company (" Applicant" or
" Edison") has not performed its own construction services for Braidwood Units 1 and 2; rather, it has issued contracts to various firms to perform the construction of Braidwood Station.
Applicant oversees and has ultimate responsibility for the work of these contractors.
One such contractor is L. K. Comstock &
Co.,
Inc., a contractor who is performing electrical installa-tion and construction work at Braidwood.
(Shamblin, Prep.
Test. at 3-4 ff. Tr. 16274.)
2.
The quality assurance and quality control func-tions are performed by Comstock Engineering, Inc. ("Comstock").
In order to assure organizational freedom and independence from the cost and schedule concerns, Comstock is organized separate-ly from and independently of the construction activities conducted by the Production Department of L. K. Comstock & Co.,
Inc.
(" Production Department").
Both L. K. Comstock & Co. and 8701130114 B70105 ~
DR ADOCK 05000456 PDR
- Comstock Engineering, Inc. are members of the Comstock Group.
(Shamblin, Prep. Test. at 6 ff. Tr. 16274.)
3.
Comstock's QA and QC managers at Braidwood report to Comstock's Regional Manager, QA/QC Services, who is located in Chicago.
The Regional Manager reports to Comstock's Cor-porate Manager of QA/QC Services, who in turn reports to the head of Comstock Engineering, Inc.
(Seltmann, Prep. Test. at 4 ff. Tr. 1960; DeWald, Prep. Test. at 3 ff. Tr. 1700; Shamblin, Prep. Test. at 6 ff. Tr. 16274; Int. Ex. 4.)
The head QA and/or QC person on site does not report to the head on-site production person, Franco Rolan.
These two positions and their departments are parallel and independent of each other.
(Shamblin, Tr. 16276; Int. Ex. 4.)
4.
The independence from the Production Department permits Comstock's QC Department to fulfill more effectively its responsibility to inspect all safety-related installations.
(DeWald, Prep. Test. at 2 ff. Tr. 1700; Shewski, Tr. 10120.)
Similarly, independence from the Production Department permits Comstock's QA Department to oversee and measure more effective-ly the effectiveness of the QC and Production Departments through regular audits and surveillances.
(Seltmann, Tr. 1969; Shewski, Tr. 10120.)
Edison's site and Corporate QA Depart-ments conducted regular audits and surveillances of Comstock's QC Department.
(Shamblin Prep. Test, at 6 ff. Tr. 16274.)
The separate organizational and reporting structure of the Produc-tion and QA/QC Departments are intended to ensure that contractor quality remains independent of engineering and pro-duction concerns.
(Id. at 4-6.)
5.
Daniel Shamblin, Applicant's Construction Super-intendent, is the official at Braidwood who administers the Braidwood on-site construction contracts, including the con-tract for the electrical work. performed by L. K. Comstock &-
Co.,
Inc.
(Shamblin, Prep. Test. at 3 ff. Tr. 16274.)
'Shamblin has held the position of Braidwood Construction Superintendent since May 1984.
(Id. at 1.)
Shamblin is responsible for implementing the Applicant's Quality Assurance Manual procedures, and monitoring on-site contractor quality control procedures, engineering, production, and work perfor-mance to ensure the quality of construction.
(Id. at 3-4.)
Shamblin, as overseer, was responsible for building quality into the construction.
(Id. at 4-5.)
6.
Shamblin's stated philosophy is to build quality into the construction by ensuring that skilled contractor engi-neers and craftsmen are used; that these personnel receive the appropriate training for their jobs; and that they perform their work to approved installation procedures.
He verifies the quality of completed construction by ensuring that trained and certified quality control inspectors using approved inspec-t tion procedures inspect the construction work, and by ensuring that the contractor's quality organizations are independent of contractor production organizations.
Shamblin also assures that the contractors' QA organizations conduct audits and i
l l
surveillances of their engineering, production and QA areas.
1 (Id. at 3-4.)
7.
While quality must function independently of f
engineering and production, all three areas should ultimately work as a team to achieve successful completion of a project.
That is, engineering personnel and craftsmen must be indoctri-nated with quality principles, so that the work is correctly performed in the first instance; quality personnel should be indoctrinated with_those same quality principles and advise the craft of the results of QC inspections in timely fashion, so that adverse trends in craft work can be identified and cor-rected; and by overseeing all three areas, Shamblin could ensure that the proper checks and balances necessary for each contractor to produce a quality product would be maintained.
(Id. at 4-5.)
8.
Irving DeWald became QC Manager for Comstock at Braidwood in August 1983.
(DeWald, Prep. Test. at 1 ff. Tr.
1700.)
DeWald had previously served at Braidwood as a Level II welding inspector for Comstock from November 1980 to October 1981.
(DeWald, Tr. 1258-59.)
9.
By August 1983, Comstock had developed what DeWald initially calculated to be a backlog of over 14,000 quality control inspections, largely because of increased inspection requirements and an insufficient number of QC inspectors.
In November 1982, Applicant had begun requiring Comstock to perform 100% quality control inspection rather than a 35% sample of installed conduits, junction boxes, safety-related cable pans, and conduit and cable pan hangers.
At approximately the same time, the NRC requested that inspections of current work be performed within 30 days from the time installation was completed.
(DeWald, Prep. Test. at 7 ff. Tr.
1700: DeWald, Tr. 1228; Seese, Tr. 2694; Saklak, Tr. 8040; O'Connor, Tr. 10045-46.)
10.
DeWald also calculated that a backlog of over 50,000 quality control documents required review to satisfy a prior correction action commitment.
(DeWald, Tr. 1228-30, 1249; DeWald, Prep. Test. at 7 ff. Tr. 1700; Seese, Tr. 2694; Saklak, Tr. 8040; Int. Ex. 215 at 1-3.)
11.
In August 1983, Comstock employed only 36 QC inspectors at Braidwood to inspect the work of more than 400 craft workers.
This complement of inspectors was insufficient to perform the work necessary to reduce the inspection backlog and keep current with ongoing installations.
(DeWald, Prep.
Test. at 7 ff. Tr. 1700; DeWald, Tr. 1230, 1242.)
12.
By early fall of 1983, DeWald requested and received permission to hire additional QC personnel, and i
Comstock began intensive recruiting efforts to obtain qualified and experienced inspectors.
(DeWald, Prep. Test. at 7 ff. Tr.
1700; DeWald, Tr. 1246; Seese, Tr. 2549.)
Constock's QC workforce grew slowly; however, its manpower needs were not met immediately merely through hiring of new personnel.
Accord-ingly, the Comstock QC department received authorization in October 1983 to work 10-hour weekdays and eight-hour Saturdays.
(DeWald, Prep. Test. at 8 ff. Tr. 1700; Seese, Tr.
2534-35; Int. Ex. 33 at 100; Saklak, Tr. 8122; Int. Exs.
8, 9.)
13.
In order to address Comstock's inspection backlog more efficiently and to attempt to keep up with current inspec-tion work, DeWald began using schedules and establishing target goals.
(Seese, Prep. Test. at 7-8 ff. Tr. 2330; Seese, Tr.
2545; DeWald, Prep. Test. at 21 ff. Tr. 1700; Int. Ex. 247.)
Comstock established desired completion dates for various critical project goals,- based on conservative estimates of the average number of weekly inspections which could be expected and the number of inspectors per discipline which would enable them to achieve those goals.
(DeWald, Tr. 1851-54, 1356-61; Seese, Prep. Test. at 8 ff. Tr. 2330; Ex. 12; Saklak, Tr.
8117-18; Seese, Tr. 2749-50, 2363, 2346.)
DeWald implemented these management tools to permit better utilization of his increasing workforce without sacrificing quality.
(DeWald, Prep. Test. at 21 ff. 1700; DeWald, Tr. 1397-98.)
14.
Assistant QC Manager, Larry Seese developed a i
status tracking system to keep management informed of QC l
l progress toward those goals at Braidwood.
(Seese, Prep. Test.
at 8 ff. Tr. 2330.)
The first of Seese's status reports was issued on May 5, 1984.
(Id.)
These reports reflected the status of work performed in each inspection area (e.g., welding termination, configurations, cable pulling).
(Id.; Seese, Tr.
2500; DeWald, Tr. 1420-21.)
15.
Comstock's status tracking system allowed manage-ment to monitor progress of backlog reduction and to assess the y
n,- - -,--,.
L
~..- -
. status of QC activities and workload.
Both Comstock and Edison i
believed that effective supervision was particularly important in light of the existing inspection backlog.
(Seese Prep.
Test. at 9-10 ff. Tr. 2330; Int. Ex. 8; DeWald, Tr. 1330-31, 1816-17; Seese, Tr. 2939-42, 2750; Saklak, Tr. 8116-17; Mendez, Prep. Test. at 13-15 ff. Tr. 10490; Shamblin, Prep. Test. at 14 ff. Tr. 16274.)
When informed by the tracking system that actual progress fell behind desired goals, Comstock management was able to take actions such as increasing inspector overtime hours, hiring additional inspectors, and transferring inspec-tors from one inspection area to another.
(Seese, Prep. Test.
at 8-10 ff. Tr. 2330; Seese, Tr. 2350-51; Saklak, Tr. 8116-18.)
16.
A backlog of approximately 6,000 inspections remained by the spring of 1984.
(Shamblin, Prep. Test., Attch.
2 ff. Tr. 16274; Int. hex. 8.)
When Daniel Shamblin became Edison's Braidwood Construction Superintendent in May of 1984, he found that the Comstock QC Department had a number of problems relating to that backlog:
(1) it was having diffi-culty coping with the inspection backlogs; (2) it did not fully understand the dimensions of work that it was facing, and therefore did not understand its manpower requirements; (3) the craft was proceeding with current installations far ahead of the related QC inspections; (4) the total number of open ICRs and NCRs were increasing; (5) the department's ongoing inspec-
)
tion programs were not progressing because of a lack of inspec-i tion support; and (6) the department needed to be reorganized to enalle it to accomplish its task of eliminating the backlog.
(Shamblin, Prep. Test. at 8-9 ff. Tr. 16274; Shamblin, Tr.
16293-98.)
Shamblin also believed that some tasks should more appropriately be performed by other organizations or depart-ments.
-(Tr. 16293.)
17.
The existence of the backlog and its related problems concerned Shamblin because unless inspections were performed in a timely fashion, a volume of installed work of indeterminate quality existed, and Edison would be unable to determine whether there were any adverse quality trends to be l
detected and then corrected.
(Shamblin, Prep. Test. at 8-9 ff.
j Tr. 16274.)
Because of the backlog, Comstock was unable to l
-perform trend studies of current inspection results or to perform in-process inspection activities.
(DeWald, Tr. 1815-16; Seese, Tr. 2535-36; Saklak, Tr. 8137.)
18.
Shamblin took a number of steps to assist the L
Comstock QC workforce in eliminating the backlog.
One of Shamblin's first steps was to develop in cooperation with DeWald a list of Comstock's priorities, based on material provided by DeWald, including a matrix prepared by QC Supervi-sor Richard Saklak reflecting the converging and interrelated demands on Comstock's QC Department.
By setting these priori-ties, Shamblin enabled Comstock to address more effectively the various tasks it faced.
(Int. Exs. 8, 12; Shamblin, Prep.
Test., Attch. 2 ff. Tr. 16274; DeWald, Tr. 1352-55, 1298-99, 1306-07, 1854; Saklak, Tr. 8125-26, 8227-28, 8249-50; Shamblin, Tr. 16309-14.)
The top three priorities, in order of impor-tance, were to finish the first line incomplete field
. inspections, to perform inspections on new work so as to prevent a second backlog, and to complete the field inspections necessary to close ICRs and NCRs.
(Shamblin, Prep. Test. at 11-12, Attch. 2 ff. Tr. 16274.)
19.
In addition to setting these priorities, Shamblin sought to alleviate the workload on Comstock so that they could proceed with inspections.
In addition to assigning four Sargent & Lundy " loan servants" to Comstock to perform QC inspections (Id., Attch. 1), by early June he had developed nearly two dozen other action items intended to reorder, reallocate and reduce the workload facing Comstock QC.
(Id.,
Attch. 2.)
20.
Shamblin monitored Comstock's progress on its reduced workload through periodic meetings with, and weekly status reports from, Comstock QC management.
Those reports provided both Edison and Comstock with data to judge Comstock's progress and to target areas for improvement.
While he was prepared at all times to suspend work, his evaluation of
+
Comstock's QC Department led him to conclude that Comstock could cope successfully with both the backlog and the inspec-tion work necessary to support ongoing installation efforts.
l l
(Id. at 9-10, 14.)
21.
Shamblin made it clear to Comstock personnel in these discussions and in memoranda that maintaining the quality of the work was just as important as getting the job done.
One of the ways he reinforced Edison's commitment to quality was to obtain the participation of Edison QA personnel in the meetings
..-.,-..n_
..-n n--,n-.
,-----m he held with LKC.
(Id. at 14-16; Attchs. 3, 4; Tr. 16304, 13608-09.)
22.
As Shamblin continued observing Comstock's performance, it became apparent that Comstock would not elimi-nate its inspection backlog by the original July 1984 target date.
Shamblin observed that Comstock was making reasonable progress without jeopardizing the quality of its inspection work, and therefore agreed to schedule extensions.
(Id. at 17-18.)
23.
Edison Site QA and Edison Corporate QA indepen-dently monitored Comstock so that Edison would be alerted to any possible deviations.
Throughout the summer of 1984, as the Comstock QC inspectors moved through the backlog, Shamblin personally reviewed the QA audits and surveillances in detail to ensure that there was no effect on quality.
Comstock's General Office QA also performed a thorough audit of Comstock's implementation of its quality assurance program in September 1984.
Shamblin also reviewed Pittsburgh Testing Laboratory
("PTL") unit conception inspection reports throughout the summer of 1984.
Finally, Shamblin conducted personal field walks, during which he made personal observations of Comstock QC-accepted work.
All of these review e fforts showed no deviations in quality to Shamblin as tha Comstock QC inspectors worked to eliminate the backlog.
(Id. at 6, 19-21; Shamblin, Tr. 16300-01, 16380-84.)
24.
The backlog was ultimately eliminated in Septem-ber 1984.
(Id. at 16-17.)
25.
In July 1984, Saklak verbally abused QC Inspector Mike Mustered when Mustered had refused to retract three ICRs which he had initiated on three conduit hangers, the installa-tions for which differed from their design drawings.
The drawings noted a requirement that the base plates of the hangers be painted; and Mustered. issued the ICRs because he found the plates to be unpainted.
(Mustered, Tr. 4970-72.)
26.
Saklak had recently attended a meeting with Sargent & Lundy and learned that the requirement for painted Saklak base plates would be deleted from the design drawings.
advised Mustered that the drawings were to be revised soon.
(Mustered, Tr. 4972-73.)
27.
Mustered did not withdraw the ICRs because the drawing revisions which would moot them had not yet been made.
Mustered told Saklak that he would continue issuing ICRs for unpainted base plates on conduit hangers until the drawings were in fact revised.
(Mustered, Tr. 4972-73.)
Saklak became abusive, but did not threaten Mustered.
Mustered did not feel intimidated.
(Mustered, Tr. 4973-74.)
28.
Mustered did not report the incident to either Edison or Comstock management; nor did he report it to the NRC l
j i
at that time.
He considered the confrontation as one that occurs from time to time at a construction site.
(Mustered, Tr. 4974-75.)
29.
From early 1983 until September 28, 1984, John D.
l Seeders was employed by Comstock as a Level II QC inspector.
During that period, Seeders was the only inspector assigned to
_ perform calibration inspections, although he was not the only inspector certified in that area.
(Seeders, Tr. 7431, 7538-41, 7665-66.)
30.
Seeders was required to turn in a daily status C
report regarding the previous day's work.
He understood the reports were due every morning by 9:00 AM.
(Seeders, Tr.
7366.)
31.
Seeders failed to hand in his status report for August 10, 1984.
(Seeders, Tr. 7357.)
Seeders handed in his status report late on August 14, 1984.
He missed the 9:00 AM deadline because he lost track of the time of day.
(Seese, Prep. Test., Attch. 3 ff. Tr. 2330.)
32.
Seeders missed the deadline for his status report on August 17, 1984.
Upon inquiry, Seeders told Saklak that the report would be in soon.
(Seese, Prep. Test., Attch. 3 ff. Tr.
2330.)
33.
During their conversation about Seeders' August j
17 status report, Seeders also requested overtime work for Saturday, August 18, 1984.
Saklak told him it would be neces-sary to provide a detailed explanation justifying why Seeders needed to work on Saturday.
Seeders then made loud, negative comments t: Saklak in the presence of other inspectors.
Saklak lost his temper and a screaming session ensued.
(Seese, Prep.
Test., Attch. 3 ff. 2330; Seeders, Tr. 7367, 7370-72, 7438.)
i 34.
Seese issued a written warning to Seeders on August 17 which, among other things, reflected that Seeders failed to hand in his status report for August 10, and he had handed in his status reports late on August 14 and 17.
(Seese Prep. Test., Attch. 3 ff. Tr. 2330.)
35.
On the evening of August 17, 1984, Seeders wrote a letter to DeWald with copies to Edison and the NRC's Robert Schulz, in which he complained, among other things, of harass-ment and intimidation.
(Int. Ex. 23; Seeders, Tr. 7425-26, 7445-47.)
Seeders testified that Saklak harassed him by screaming at him on the morning of August 17 about turning in his status reports late.
(Tr. 7438.)
36.
DeWald interviewed six inspectors who had witnessed the encounter with Saklak.
(App. Ex. 4; DeWald, Prep. Test, at 36-37 and Attch. 5 (Revised) ff. Tr. 1700.)
In general, these witnesses, who were Rick Snyder, Worley Puckett, Don Coss, Bob Wicks, Jan Lobue and Myra Sproull, attributed the incident to a personality clash between Saklak and Seeders (App. Ex. 4; DeWald, Prep. Test. Attch. 5 (Revised) ff.
Tr. 1700).
37.
Larry Phillips, Seeders' lead inspector, normally conducted material receipt inspections at Comstock.
(Int.
Ex. 23; Seeders, Tr. 7397.)
Phillips took funeral leave as of August 8 and then some vacation time.
(Seeders, Tr. 7397.)
38.
Seeders was assigned the additional responsi-bility for conducting material receipt inspections during Phillips' absence.
(Seeders, Tr. 7397.)
Phillips was gone for no more than one or two weeks.
(Seeders, Tr. 7454.)
Seeders was relieved of the assignment upon Phillips' return to work.
(Id.)
39.
Thirteen material receipt inspections were performed by Seeders during the period August 8 through August 17; that was the extent of such inspections for Comstock for that period.
(App. Ex. 90; Hii, Prep. Test. at 2-3 and Attch. 1 ff. Tr. 16608; Hii, Tr. 16612.)
At least as to one of the thirteen, which appears to us to be fairly typical of the rest, seeders testified that it would take him perhaps a half hour to perform the inspection.
(Tr. 7396; See Tr. 7412-16.)
40.
Certain aspects of material receipt inspections involving clerical tasks could be performed by individuals who were not certified in that discipline.
(Seeders, Tr. 7814.)
During the week or two that Seeders conducted these inspections while Phillips was away, Seeders performed all aspects of the inspections, including those tasks not requiring certification in material receipt inspections.
(Id.)
41.
Seeders stated both in his letter and while testifying that despite his complaints, he maintained the highest level of professionalism in his work to the best of his ability.
(Int. Ex. 23; Tr. 7429.)
42.
On November 5, 1984, inspector Francisco Rolan wrote an NCR that identified a nonconforming condition in an i
installation drawing.
The drawing required revision to pro-perly disposition the NCR.
(Rolan, Tr. 4665-68, 4688-89, l
4691.)
43.
Rolan had not followed the Comstock practice of identifying the electrician who had installed the work because Rolan believed that such identification would unjustly blame l
l the electrician.
(Rolan, Tr. 4694.)
Rolan's lead had directed him to handle the inspection in the manner he had.
(Rolan, Tr. 4665-68.)
44.
Saklak insisted that the electrician be identi-fied on the NCR.
(Rolan, Tr. 4668-73.)
Saklak wanted the electrician added so the NCR could be tracked properly.
(Saklak, Tr. 8050-52.)
45.
When Rolan refused to add the electrician's name to the NCR, Saklak threatened Rolan with words to the effect that when the NRC was through with him, Rolan would not have any certifications left.
(Rolan, Tr. 4674.)
Rolan understood Saklak as threatening to fire him (Id.)
Rolan reluctantly added the electrician's name to the NCR.
(Rolan, Tr. 4673.)
46.
Saklak was issued a written warning from DeWald after Rolan complained to DeWald about it.
(Int. Ex. 43, 44; Saklak, Tr. 8055.)
47.
Rolan testified that he believed Saklak had no authority to fire QC inspectors.
(Tr. 4676-77.)
48.
Sometime aetween August 17, 1984 and March 28, 1985, Saklak apparently reprimanded Mike Lechner for joking around and laughing loudly with fellow QC inspectors in Saklak's vicinity while Saklak was speaking with an NRC inspec-tor.
Saklak yelled at Lechner later for interrupting him with laughter.
This was a trivial incident involving no quality concerns.
(Snyder, Tr. 4229-82; Simile, Tr. 3396-98.)
49.
In September of 1982, Saklak approached QC inspector Rick Martin and asked if he had a tape measure.
\\ Martin responded that he had lost his tape measure that day, and Saklak cursed him.
Martin asked Saklak if he had a problem.
Saklak then blew up, cursed Martin again, and ordered him into the field.
Martin told him he would not tolerate Saklak speaking to him in that way and refused to go to the field.
Saklak told him that he was fired, and began preparing separation paperwork on Martin.
(Martin, Tr. 8439-43, 8445.)
50.
After completing the paperwork, Saklak met with Tom Corcoran, at that time the QC Manager.
Martin was even-tually called into Corcoran's office, where he told Corcoran that the lost tape measure was not Comstock's, but his own.
Martin testified that he told Corcoran his side of the story in such a way as to appease Saklak.
Saklak then recommended to Corcoran that he " smoke him."
Martin was not fired.
(Martin, Tr. 8443-46; 8451; App. Ex. 104.)
51.
Martin regarded Saklak's behavior in this inci-dent as intimidating, but stated that it had no relation to quality matters.
He regarded it as poor management.
(Martin, Tr. 9421-22.)
52.
In early 1983, inspector Danny Holley was threatened by Saklak.
Holley had earlier gone to Edison for the purpose of inquiring about job opportunities elsewhere.
Saklak found out that Holley had gone to Edison's offices and told Holley that he would fire him on the spot if Holley went to Edison again without permission.
(Holley, Tr. 5100.)
53.
Holley testified that he knew Saklak was without authority to execute such a threat.
(Holley, Tr. 5101-02.)
. - - _,..,., -, - _ _., - - -,,... - - ~. - -. - - -
54.
Sometime between October 1983 and March 28, 1985, Saklak and an S&L engineer met with a team of QC inspectors to assess whether drawing revisions maintained at print stations in the field were current or voided.
(Martin, Tr. 9440-41.)
There was no procedure for this endeavor, and Saklak apparently intended to adapt a checklist ordinarily used for Comstock 4
document reviews for this effort.
(Martin, Tr. 9440.)
55.
Martin objected to Saklak's proposed use of the document review procedure checklist because he thought it was a violation of the document review procedure.
(Martin, Tr. 9439, 9442, 9448.)
When Martin raised his objection to Saklak, Saklak belittled him in front of the others.
(Martin, Tr. 9439.)
In particular, Saklak repeatedly asked Martin how he would document this inspection. (Martin, Tr. 9440.)
56.
Martin then went to Seese to suggest that Saklak's use of the document review checklist for this purpose should somehow be documented.
(Martin, Tr. 9445-46.)
Seese agreed and memorialized Saklak's special use of the checklist.
(Martin, Tr. 9446-47.)
According to Martin, Seese also com-mended Martin for his judgment. (Martin, Tr. 9446.)
57.
Although Martin testified that he felt degraded by this incident, he did not regard it as something intended to coerce him to compromise the quality of his work.
(Tr. 9449-50.)
58.
Sometime between October 1983 and March 28, 1985, Martin and QC Inspector Dunbar bypassed their lead, their supervisor, and QC manage:aent, and went directly to a top-level l
Sargent & Lundy person whom Martin characterized as the " head man" to ask a question about a detail on a drawing.
(Martin, Tr. 9410-12.)
Martin had not expected to get an adequate answer to his. question from Coms'tock.
(Tr. 9411-15.)
d 59.
Sargent & Lundy apparently reported this to Comstock, and Saklak, togethe: with DeWald, reprimanded Martin and his fellow inspector.
(Martin, Tr. 9412-13, 9416, 9423.)
Although he felt the reprimand was an example of poor manage-ment practice, Martin did not regard it as designed to coerce him to compromise the quality of his work.
(Tr. 9422-26.)
60.
Also sometime between October 1983 and March 28, 1985, Saklak assigned Martin to inspect cables pans for trash, dirt and debris.
Saklak gave Martin a form to fill out, and directed him to write "N/A" for all other information asked for on the form.
Saklak then demonstrated the assignment to Martin by crumpling a piece of paper, throwing it on Martin's desk and t
telling Martin that that was was what he ought to inspect for.
Martin testified that he told Saklak he would like to review the procedure first, and that Saklak agreed.
(Tr. 9417-19.)
After doing so, Martin concluded that the applicable procedures called for duties for which he was not qualified, but rather than returning to Saklak, consulted Seese about it.
(Martin, Tr. 9419-20.)
Saklak then withdrew his request that Martin conduct the debris inspection.
(Martin, Tr. 9420.)
/*
61.
Martin testified that this incident and the preceding two involved only poor management practice on Saklak's part, not harassment or intimidation, since they did not related to quality work (Tr. 9420-26.)
62.
Sometime in 1982 or 1983, Saklak lost his temper at Joseph Hii in the vault area.
Hii could not recall many details about it, only that that Saklak and he were standing in the vault area, and that Saklak was yelling at Hii because Hii had not finished a project.
Hii could not recall the nature of the project.
(Hii, Reb. Prep. Test. at 4 ff. Tr. 16608.)
63.
Hii was not intimidated by the incident.
(Hii, Reb. Prep. Test. at 5 ff. Tr. 16608.)
64.
The next day Hii pulled Saklak aside and talked to him about the incident and the problems he was having with the project.
They discussed both the incident and the project and resolved them both during that discussion.
(Hii, Reb.
Prep. Test, at 4-5 ff. Tr. 16608.)
65.
QC Inspector Richard Snyder had replaced John Seeders as the calibrations inspector.
(DeWald, Tr. 1617.)
Among his responsibilities was to issue ICRs on weld machines he found to be out of calibration tolerances, under Comstock Procedure 4.9.1, Revision C.
(Snyder, Tr. 4182, App. Ex. 24.)
Sometime before March.28, 1985, engineering personnel had accepted "as is" welds made with out-of-calibration machines on which ICRs had been issued.
(Snyder, Tr. 4187-90, 4193-94.)
66.
When the factu in the preceding Finding came to Snyder's attention, he asked both Saklak and simile about it.
They explained that because all welds made with the machines were themselves inspected, any discrepancy caused by improper amperage would be caught.
(Snyder, Tr. 4193-94.)
67.
The result of these discussions was that, as of March 28, 1985, LKC procedure 4.9.1, Revisicn C was being revised to eliminate the requirement that weld machines be calibrated.
(Snyder, Tr. 4182.)
This procedure revision was approved in early April 1984.
(App. Ex. 38 (Procedure 4.9.1, Revision D).)
68.
Snyder testified that since the new procedure revision had not been finalized by March 28, 1985, he wrote another ICR on a weld machine that was out of calibration on that date.
(Tr. 4181-82.)
Given his past experience with ICRs on weld machines, Snyder testified that he expected the engi-neers to accept and disposition the ICR on the basis of the weld inspections.
(Tr. 4187.)
on the afternoon of March 28, Saklak told Snyder and Nemeth to close the ICR without sending it to engineering because they knew what the disposition would be.
(Snyder, Tr. 4182-83, 4185, 4195.)
69.
Snyder and Nemeth tried to explain to Saklak that Comstock procedures still required an engineering evaluation of the ICR.
(Tr. 4185-86.)
Snyder checked with Seltmann, the QA manager, who assured Snyder that he was correct and Saklak was wrong.
(Snyder, Tr. 4186; Seltmann, Prep. Test, at 20 ff.
Tr. 1960.)
Snyder then asked both gentlemen if they could bring Saklak into the office and straighten him out.
This request was declined because Saklak was "too hot." (Snyder, Tr. 4186-87.)
. 70.
Snyder then returned to his desk with Nemeth, Saklak came up and stood four or five feet in front of them with his legs spread apart and said:
" Rick, sometimes -- we are going to go round and round and you make me so pissed off that if beating was legal, you would be dead."
Snyder replied:
" Rick, whatever you think you have to do."
(Snyder, Tr. 4196.)
71.
Saklak walked away.
After Saklak walked away, Snyder testified that he was shocked and in a state of dis-belief, and discussed with Nemeth whether Saklak had meant was he said.
(Tr. 4198.)
Nothing like that had ever happened to Snyder in his line of work.
(Tr. 4198-99.)
72.
Snyder wrcte down what had happened because he knew his " rights as an inspector", he "did not have to put up with that."
(Snyder, Tr. 4196-97.)
Snyder believed that Saklak was upset because Snyder had bypassed Saklak by going to Seltmann.
(Snyder, Tr. 4467.)
Seltmann agreed with Snyder's assessment.
(Seltmann, Tr. 2097.)
73.
On the morning of March 29, 1985, Nemeth told Snyder that Saklak wanted to apologize; Snyder told Nemeth that it was too late, that he had thought about it all night and f
that he was going to the NRC.
(Snyder, Tr. 4200-01.)
Saklak nevertheless apologized to Snyder and told him that he had had a bad day.
Snyder responded, "Well, you know, we can all have bad days, but you have got to learn to control it."
(Snyder, t
Tr. 4201.)
74.
Saklak's apology did not cause Snyder to change i
his mind about going to the NRC.
(Snyder, Tr. 4201.)
He and
___ _. other inspectors complained to the NRC during two meetings on March 29, 1985.
The first meeting, at which Snyder was ac-companied by five other inspectors, was sometime before 9:00 a.m.
(Snyder, Tr. 4201, 4204.)
Snyder returned at the NRC's request at noon, and was accompanied by 23 Comstock QC inspectors.
(Weil, Prep. Test, at 17-18 ff. Tr. 11948; McGregor, Tr. 11570-71; Snyder, Tr. 4265.)
75.
As a result of the incpector's complaints to the NRC, Saklak was first suspended for a day and then suspended pending completion of an investigation.
Thereafter, Saklak was laid off by the Corporate QA Manager.
Saklak never returned to the site (DeWald, Prep. Test, at 31-32 ff. Tr. 1700; Saklak, Tr. 8033; Snyder, Tr. 4270-71.)
76.
Snyder and a number of other inspectors testified that Saklak's removal from the site satisfactorily resolved the 4
problem that had motivated them to complain to the NRC on Mar. 29.
(Snyder, Tr. 4462-63; Stout, Dep. Tr. 152; Mustered, Tr. 4990-91, 5073; Rolan, Tr. 4747.)
77.
Edison held a meeting on Monday, April 1, 1985 i
with the Comstock inspectors.
They were told that Edison expected all work to be performed in a quality manner and that CECO would accept nothing less.
(Maiman, Prep. Test. at 7 ff, i
Tr. 3806.)
Edison barred Saklak from all safety-related work at any of its nuclear sites.
(Id. at 8.)
Edison also took several longer term actions.
Among them was to provide inter-l i
personal skills training to Comstock management and supervisory i
_ personnel.
(Id. at 8-9; DeWald, Prep. Test. at 33-34 ff.
Tr. 1700.)
78.
Although Constock's site management had some concern over Saklak's temperament, they all felt that he was a conscientious, quality-minded and knowledgeable supervisor.
(DeWald, Prep. Test, at 26-29 ff. Tr. 1700; DeWald, Tr. 1287-88, 1760, 1850, 1910; Seltmann; Prep. Test. at 22-23 ff. Tr. 1960; Seltmann, Tr. 2060, 2063-64; Seese, Prep. Test.
at 22 ff. Tr. 2330; Int. Ex. 44.)
79.
This high opinion was reflected in Saklak's regular performance reviews.
(DeWald, Prep.
Test. at 27-29 ff. Tr. 1700; Int. Exs. 52-57; Saklak, Tr. 8001-27.)
In a December 22, 1983 review, Saklak was described as "an exceptional supervisor" who was strong in identifying problem areas concerning construction practices, and who "wants the job done effectively and to the detail."
(DeWald, Prep. Test. at 27 ff. Tr. 1700; Int. Ex. 53.)
Saklak's January 5, 1985 annual review found him to be "an aggressive, energetic and quality minded individual continually striving for a quality product."
i He was described as " exceptionally knowledgeable of all areas in which he supervises, although he is not certified in every discipline" and as " extremely beneficial" to Comstock's QC Department.
(Int. Ex. 56.)
80.
Several inspectors apparently agreed with Comstock management that Saklak was a quality-minded and knowledgeable supervisor.
For example, Bowman testified that Saklak was knowledgeable in discharging his responsibility to interpret Comstock procedures (Tr. 6784); Hii "always found (Saklak) to be quality-conscious and dedicated"-(Hii, Prep.
Test, at 5 ff. Tr. 16608); and, Peterson found Saklak helpful in answering his questions calmly, capably and competently.
(Tr. 6091-92.)
81.
From the time Saklak started work as a QC supervisor in 1982, he began making suggestions to-improve Comstock's procedures and practices.
Saklak was initially assigned to organize cable pan installment reports and cable pan weld inspection reports.
He promptly recommended that inspectors limit cable pan weld inspection reports to single trays, so that those inspection reports would no longer be done on a grid basis.
This suggestion greatly simplified the tracking and filing of those reports.
(Saklak, Tr. 8165-67, 8078-79.)
82.
Saklak also took the initiative to recommend revisions to several loosely written or vague inspection procedures on which he had received complaints from inspectors.
(Saklak, Tr. 8009, 8080-82.)
In addition, frustrated by the lack of cross-references in Comstock's filing system, Saklak established a new filing system to improve Comstock's ability to track inspection reports and respond to audit findings and surveillances.
(Saklak, Tr. 8009, 8083-84.)
3 83.
Saklak also began keeping a personal file, which he called his " Pearl Harbor file."
That file, which eventually i
j grew to fill roughly half a file cabinet drawer, contained documents which Saklak categorized by inspection area.
(Saklak, Tr. 8064-67, 8240, 8182; Hii, Prep. Test. at 7-8 ff.
Tr. 16508.)
Saklak kept these documents so that he would be able to document, if necessary, various concerns raised by or addressed to comstock QC and document the manner in which they were resolved.
(Tr. 8064-66.)
84.
Saklak used documents compiled in his Pearl Harbor file to persuade Rolan, the Manager of the Production Department, that QC was receiving duplicate installation reports from the Production Department.
Those duplicate reports caused needless reinspection of various hangers at a time when Comstock was addressing a large inspection backlog.
(Saklak, Tr. 8183-85; See also Int. Ex. 231.)
After Saklak proved the validity of his concern by presenting copies of numerous duplicate installation reports, Rolan took steps to reduce the number of duplications.
(Saklak, Tr. 8185.)
85.
Saklak also took the initiative in addressing several quality-related matters.
When informed by inspectors Sproull and Hall in the spring of 1983 that scrap cable was stored in an area where it might be used for safety-related pulls, Saklak investigated the concern and confronted Franco Rolan, Comstock's Production Manager, with the problem.
Saklak's action led the Production Department to take measures to further ensure that scrap cable was used only for non-safety related work.
(Saklak, Tr. 8240-41, 8253-54; Int. Exa. 212, I
233.)
86.
In June 1984 Saklak took the initiative to approach DeWald with the concern that Comstock QC could not
_ effectively perform the various tasks before it unless an order of priority was assigned to those jobs.
To help in establishing priorities, Saklak created a matrix showing the various areas for which Comstock QC was responsible and the manner in which those areas were interrelated.
(Saklak, Tr. 8046-48, 8227-28; DeWald, Tr. 1349-50; Int. Ex. 12.)
87.
In addition, Saklak frequently raised in Comstock staff meetings his concerns about the quality of the inspect-ors' reports, about paper flow and other problems which hin-dered the inspectors' work and about poor construction prac-tices.
(Seese, Tr. 2528-32, 2564-66, 2584-86, 2591-92; Int.
Ex. 33 at 98, 114, 117, 118, 140; Int. Exs. 272, 282.)
88.
From early 1983 until September 28, 1984, John D.
Seeders was employed by Comstock as a Level II QC inspector.
During that period, Seeders was the only inspector assigned to perform calibration inspections, although he was not the only inspector certified in that area.
(Seeders, Tr. 7431, 7490, 7536-41, 7665-66; App. Ex. 2.)
89.
Seeders was responsible for ensuring that tools and equipment were calibrated before being issued for use.
The tools would then be returned to QC after a prescribed period and he would check their calibration again.
If a tool's calibration was acceptable upon its return, all work performed between calibrations was considered acceptable.
(DeWald, Prep.
Test, at 35 ff. Tr. 1700; Seese, Prep. Test. at 16-17 ff.
Tr. 2330.)
If Seeders found a tool out of calibration or l
returned to QC with an expired calibration date,
- Paragraph 3.5.7 of Procedure 4.9.1, the calibrations procedure, required him to prepare an ICR.
(App. Ex. 24.)
90.
Seeders testified that when he found a tool out of calibration, he typically placed a hold tag on it at;once.
He would then record the deviation on the appropriate forms and prepare an ICR.
(Seeders, Tr. 7316-18.)
These ICRs served to document both the discrepancies in the tools to be resolved before the tools could be used again, and any indeterminate quality conditions in the field which might have resulted from using out of calibration tools.
(DeWald, Prep. Test. at 35 ff.
Tr. 1700.)
91.
Edison Audit Report QA-20-84-528, which was issued May 29, 1984, found that Comstock's calibration Procedure 4.9.1 was not being followed, in that ICRs were not h
being issued to evaluate the validity of previous use of
~
(
instruments and gauges found to be out of calibration; thermometers were not calibrated as frequently as required; and calibration records for certain equipment Lent to an outside agency had not been reviewed for acceptablility.
(DeWald, at11hf.
l Prep. Test. at 35 ff. Tr. 1700; Gieseker, Prep. Test.
Tr. 2771; Seltmann, Prep. Test. at 11-12 f f. Tr. 1960.)
92.
Comstock forwarded to Edison its proposed re-sponse to the audit findings.
This consisted of issuing an NCR and two ICRs to document particular deficiencies, as well as providing additional training for Seeders.
(Gieseker, Prep.
Test. at 12 ff. Tr. 2771.)
Edison requested Comstock to assure that no additional deficiencies existed and to take measures to
(
prevent reoccurrences.
(DeWald, Prep. Test. at 36 ff.
Tr. 1700; Gieseker, Prep. Test, at 12 ff. Tr. 2771.)
Seltmann was responsible for this further response, which Edison had requested by July 20, 1984.
(Seltmann, Prep. Test. at 12 ff.
Tr. 1960.)
93.
Sometime between late May and early July Seeders was directed to perform a 100% review of the Comstock cali-bration records to identify whether in fact ICRs or other documentation had been issued for work performed with tools found to be out of calibration.
(DeWald, Prep. Test. at 35 ff.
Tr. 1700.)
The witnesses would not recall the date when this assignment was made, but DeWald testified that by July 12, Seeders had received the assignment and it should have been in progress.
(Id. at 36 and Attch. 6 thereto.)
94.
On Friday, July 20, 1984, the day the response was due, Seltmann asked Seeders the status of his review.
Seeders said that his review was incomplete, but that he and Snyder intended to work on it the following day.
(Seltmann, Tr. 2104.)
Seltmann extended the deadline for completing the review and made it clear to Seeders that he should perform a 100% review of his records, rather that merely a sample review.
(Seltmann, Prep. Test, at 14 ff. Tr. 1960.)
95.
On Monday, July 23, 1984, Seltmann contacted Seeders and was informed that the review was still incomplete.
Seltmann gave Seeders another time extension for completing the review.
He pointed out that Comstock's response to Edison's audit was already late because of Seeders' failure to complete
. _ _ _ _ __ the review, and he told Seeders that he would send Edison a memorandum stating that a complete review wr.s underway and that an updated response would be submitted by August 13, 1984.
(Seltmann, Prep. Test. at 13 ff. Tr. 1960.)
Seltmann issued the memorandum to Edison.
(App. Ex. 85.)
96.
Seltmann periodically checked with Seeders regarding the status of his review.
(Seeders, Trs. 7673.)
On August 14 Saklak showed Seltmann several handwritten papers which purported to constitute Seeders' calibration documenta-tion review.
(Seltmann, Prep. Test. at 13 ff. Tr. 1960; App.
Ex. 26.)
The sixth page of that document read "No further action need be taken.
All research completed as per 4.9.1.C.
Please see following attachment research", followed by Seeders' signature.
(App. Ex. 26, Seeders, Tr. 7339.)
Seeders testified that this notation meant only that this portion of his review was completed and no discrepancies had been dis-covered.
(Tr. 7340.)
97.
Seltmann was dissatisfied with the documentation shown him and he and Saklak met with Seeders about the review later that day.
(Seltmann, Prep. Test, at 13-14 ff. Tr. 1960.)
I Seeders testified that Seltmann threw the papers back at him.
l (Tr. 7675-77.)
Seltmann testified that Seeders was very l
arrogant at that meeting, and at one point accused DeWald and Seltmann of being responsible for the deficiencies cited in the Edison audit.
Seltmann became angry and told Seeders to leave l
and get back to work.
(Seltmann, Prep. Test. at 14 ff.
Tr. 1960.)
98.
Seltmann then wrote another memorandum to Edison stating that Seeders' review of the calibration records was still only partially completed and that.a final report was expected by August 24, 1984.
(Seltmann, Prep. Test, at 13-14 ff. Tr. 1960.)
99.
Seeders was relieved of his assigt. ment to com-plete his review of the calibration records, although he could not recall when this occurred.
(Seeders, Tr. 7427, 7809-11.)
Sometime in August 1984, QC Inspectors Richard Snyder, Don Coss and Myra Sproull were assigned the responsibility of performing a review of all electrical work calibration records in Comstock's custody, without regard to whether they were Comstock's or the prior electrical contractor's.
(Synder, Tr. 4054-56; 4331-34.)
100.
On August 17, 1984, Seeders wrote the complaining letter we have referred to earlier.
The letter complained that Seeders had been required to perform a 100% review of cali-bration records by a certain date or be subject to disciplinary action.
When he explained to Seltmann that a proper review could not be completed in that time, he was " harassed and intimidaoed into being done with my review on the given date no matter how I did it."
Seeders took this as a direction to falsify documentation, which he stated he would never do.
(Int. Ex. 23; Seeders at Tr. 7425-26.)
101.
Seeders testified that Seltmann never suggested to him to falsify documentation to get the job done, and that no one else told Seeders to do so.
(Tr. 7426-27).
He also testified that Seltmann had not given him a completion date.
Rather, Seltmann had approached him a number of times to ask i
him how the review was progressing, and told him to get it done as soon as possible so that Seltmann "would have something to submit to CECO."
(Tr. 7424-26.)
Although his letter said that he had been required to complete the review "by a certain date," he agreed on the stand that it would have been more accurate to say "as soon as possible."
(Int. Ex. 23; Tr. 7423-26.)
It was Seeders' opinion that because Seltmann pressured him to complete the work, he was being harassed to make him ignore quality.
(Seeders, Tr. 7428-30.)
102.
Seeders sent a copy of his August 17 letter to Cosaro, Edison's former Construction Superintendent.
Dan Shamblin, the present Superintendent, received it.
(Shamblin, Reb. Prep. Test. at 18 ff. Tr. 16374; Seeders, Tr. 7445.)
In response, Mike Wallace, the Braidwood Project Manager, Shamblin and Jim Gieseker, an Edison engineer assigned to work with Comstock, met with DeWald to discuss the letter.
Shamblin then l
directed Gieseker to meet with Seeders to understand his concerns better.
(Gieseker, Prep. Test. at 6 ff. Tr. 2771.)
Shamblin also asked DeWald to investigate the matters relating to Seeders' work situation and the harassment he claimed to have experienced.
(Shamblin, Reb. Prep. Test. at 22 ff.
Tr. 16274.)
103.
Gieseker met with Seeders on August 21, 1984 and again on August 23.
(Seeders, Tr. 7466-69; Gieseker, Prep.
l Test. at 6-8 ff. Tr. 2771; Seeders, Tr. 7468-7469.)
At the
4 second meeting he told Seeders that Edison understood Comstock QC was undergoing a stressful period because of ongoing reinspection programs and other problems, including the cramped working conditions resulting from a fire in the QC office and problems scheduling training because the training coordinator had been in an accident.
(Seeders, Tr. 7471; Gieseker, Prep.
Test. at 9 ff. Tr. 2771.)
104.
Gieseker then discussed the steps that Edison had taken to improve-matters.
He told Seeders that Edison PCD had added personnel, including himself and Larry Tapella, assigned to the electrical area so as to provide further support to Comstock's activities.
He also told Seeders that he was working on revisions of Comstock's qualification / certification procedure to facilitate the certification of additional in-spectors hired by Comstock.
Finally, Gieseker informed Seeders that DeWald was then in the process of interviewing other QC inspectors to hear their views and suggestions.
(Gieseker, Prep. Test. at 9 ff. Tr. 2771; App. Ex. 93.)
105.
Tapella, present at the second meeting, told Seeders that Edison was most concerned about having quality t
work performed and that the quality of work should not suffer in order to meet schedules.
(Seeders, Tr. 7472.)
- Finally, Seeders was told that he should continue trying to resolve his concerns with Comstock management, but that he was always free to contract Shamblin, Gieseker and Tapella.
(Seeders, Tr. 7472; Gieseker, Prep. Test. at 10 ff. Tr. 2771.)
Gieseker said that he also told Seeders that his employment would not be jeopardized because of his letter.
(Appl. Ex. 78.)
Seeders failed to recall being told that.
(Seeders, Tr. 7473.)
Seeders testified that he felt better after discussing his concerns, but he did not contact Gieseker or Tapella again.
(Seeders, Tr. 7473-7474, 7491) 106.
DeWald documented the findings of his investigation into the issues raised'in Seeders' letter in a memorandum dated September 25, 1984.
(App. Ex. 4.)
DeWald concluded that many of Seeders' allegations and statements in his letter were either inaccurate or had been resolved.
107.
For example, Seeders alleged that 30 inspectors left the jcb site after a $12 per hour minimum wage for Level II inspectors was established -- in fact only 6 in-spectors left.
(App. Ex. 4 at p.1 of 11.)
Seeders alleged that there was constant pressure to train new inspectors -- in fact then currently certified inspectors had to train new inspectors in order for them to get their initial certification so that the new inspectors could perform the jobs and give I
support to the other inspectors.
(App. Ex. 4 at 2 of 11.)
Seeders also alleged that disciplinary action would be taken against inspectors who did not perform training -- in fact no inspector was ever threatened with or subjected to disciplinary action for failure to train new personnel.
Seeders' statement that inspectors' certifications had lapsed due to Comstock l
negligence was also untrue, i.e.,
CECO instituted an inspector certification upgrade in mid-July of 1984 for all site con-tractors.
Accordingly, Comstock's certification procedure and
_ requirements were being revised and thus, certifications were temporarily halted until the new procedures were implemented.
(Id.)
108.
DeWald believed that Seeders' allegation of being-subject to harassement and intimidation for six months was most likely prompted by numerous audit findings concerning Seeders' area of responsibility and the calibration review instituted in response to the Edison audit.
(App. Ex. 4 at 2 of 11.)
Seeders' testimony revealed that in fact he had not been subject to harassment for the six months preceding August 17, 1984, but that the alleged harassment referred to in his letter was related to actions by Seltmann and Saklak, and that such actions started only about two weeks before he wrote the letter.
(Seeders, Tr. 7855.)
With respect to Seeders' claim that he would not falsify documents, DeWald found that Seeders was never told to falsify any documents.
(App. Ex. 4 at 2 of 11.)
109.
DeWald also concluded that Seeders' claim that he (Seeders) was willing to do anything management required was exaggerated, inasmuch as Seeders was very reluctant to receive l
instruction pertaining to his performance and he became very offensive when such instruction was given.
(App. Ex. 4 at p.2 of 11.)
4 110.
Shamblin found that no further action was re-l quired to address Seeders' concerns based on DeWald's con-l clusions, Gieseker's reports and Seeders' statement to Gieseker.
(Shamblin, Reb. Prep. Test. at 23 ff. Tr. 16274.)
111.
Therman Bowman, a welding inspector, had observed a base metal reduction problem in a structural weld.
When he mentioned this to his lead inspector, Walters, Walters told him he should ignore it because it was not within the scope of his inspection work.
(Bowman, Tr. 6795).. Bowman testified that he told Dyrl Landers, Comstock's welding inspection supervisor, about the problem; and Landers told him to keep up his produc-tion or he would lose his overtime.
(Bowman, Tr. 6796.)
Bowman testified, however, that he believed Landers had merely made a poor joke.
(Tr. 6796-97.)
112.
Upon further examination, Bowman testified that he explained the base metal reduction problem briefly to Landers, gave him a slip of paper on which he had noted the location of the problem and turned to go.
Landers then looked up with a grin and said, " Keep up the good work or we will take you off of overtime."
(Bowman, Tr. 6807-09.)
Bowman felt confused because Landers' manner indicated that he was joking, although his words implied a threat.
(Tr. 6810.)
113.
Later that day, Bowman's supervisor, Worthington, returned the slip of paper Bowman had given Landers, and told Bowman to deal with the problem as he saw fit, including l
writing an NCR, if necessary.
Bowman then issued an NCR.
1 1
(Bowman, Tr. 6812-14.)
114.
Inspector Terry Gorman was at one time in charge l
of Comstock's document vault.
(Gorman, Tr. 5715.)
He l
testified that DeWald would interrupt his regular work assign-ments to request documents from the vault and would indicate l
how and when he wanted the work performed.
(Tr. 5747-49.)
Gorman would then have to let his other work build up and " rush to get it done."
(Gorman, Tr. 5747.)
Gorman's retrieval of QC documents from the vault, which DeWald often requested for auditors and Edison, was apparently a normal part of Gorman's job.
(Gorman, Tr. 5746-47, 5851.)
115.
Gorman sometimes felt harassed and intimidated by DeWald's assignments.
DeWald set deadlines by which he wanted the document.
(Gorman, Tr. 5851-52.)
Gorman sometimes found these time limits unreasonable.
(Gorman, Tr. 5852-53.)
He testified that he could not recall any instance where DeWald asked him to review large quantities of documents and report back to him in a short period of time.
(Gorman, Tr. 5851-52.)
116.
DeWald was demanding and unhappy when Gorman missed DeWald's deadlines, although DeWald would give him more time to finish.
(Gorman, Tr. 5749-51.)
DeWald never disci-plined Gorman nor did he ever give him a warning or threaten him with termination.
(Gorman, Tr. 5751-52.)
117.
Shortly after he had become LKC's QC Manager at Braidwood in August 1983, DeWald instituted weekly Friday meetings between QC management and QC inspectors.
(DeWald, Tr. 1786.)
Those regular meetings were suspended while LKC's fire-damaged QC office was under repair, but otherwise have continued to the present time.
(Id.; Shamblin, Prep. Test, at 27 ff. Tr. 16274.)
118.
These weekly meetings were used to discuss ideas for improving LKC's QC organization and the working relation-l ships among personnel.
(Shamblin, Prep. Test. at 27 ff.
Tr. 16274.)
DeWald also utilized the weekly meetings to respond to comments and questions, and to explain new policies and procedures such as the use of status reports to track LKC progress.
(DeWald, Tr. 1853-54, Seese, Tr. 2639; Bowman, Tr. 6872.)
Once LKC's tracking system was in place, LKC management also provided the inspectors with current information concerning the status of completed and uncompleted inspections.
(Seese, Tr. 2372; Bowman, Tr. 6872.)
J 119.
DeWald testified that he viewed those weekly meetings as an important means of communicating with the QC inspectors.
(Tr. 1786.)
One of the central messages he attempted to communicate was that quality was never to be sacrificed to meet production schedules or goals.
(App.
Ex. 97.)
This was documented an one occasion, a meeting held January 5, 1985.
(cite) 120.
Several QC inspectors testified that LKC manage-ment used those weekly meetings to convey general information to the inspectors.
(Gorman, Tr. 5775; Snyder, Tr. 4244.)
After management had spoken, the inspectors would ask questions
(
about procedures and details.
(Bowman, Tr. 6871-6873; Snyder, Tr. 4245.)
l 121.
During these meetings, LKC management usually described those areas of inspection that were behind and those areas which needed more manpower allocated to them.
(Snyder, Tr. 4241.)
Seese would read the status reports which detailed the progress being made on projects to eliminate inspection
_. backlogs and.the projected dates for completion of those projects.
(Snyder, Tr. 4243; Bowman, Tr. 6871-6873; Perryman, Tr. 9663.)
122.
Seeders testified that at these meetings, DeWald commented about being under schedule pressure from Edison (Seeders, Tr. 7567.)
At those meetings Seeders understood that quantity of inspections was emphasized over quality because the weld inspectors would comment that "DeWald wants numbers again" when DeWald pushed inspectors for greater productivity (Tr. 7566.)
123.
Although Gorman recalled no instance in which DeWald ever established a quota or a norm of inspections that he expected to be completed by the inspectors in a given period of time, Gorman believed that the general meetings were in-tended to urge inspectors to perform more inspections; Gorman interpreted this as emphasizing quantity over quality.
(Gorman, Tr. 5798.)
124.
Gorman recalled DeWald complaining that there were too many people sitting in the office and that they should i
be out in the field performing more inspections, because not l
enough work was being accomplished.
(Gorman, Tr. 5776-77.)
f l
Gorman complained that DeWald instead should have addressed problematic individuals on an individual basis.
(Gorman, Tr. 5798.)
125.
Nevertheless, Gorman agreed that it was proper for management to ensure that the work was being performed, and to ensure that inspectors were not loafing.
(Gorman, Tr. 5779.)
L y
Gorman had witnessed personnel loafing in the vaulti and admitted that he himself had occasion to loaf, although he had never been given a verbal or written reprimand for loafing.
(Gormen, Tr. 5779-80.)
126.
According to Holley, DeWald never expected a certain number of inspections.
(Holley, Tr. 5111.)
Holley attributed to DeWald as relieving tension that Holley felt had been present before. "He (DeWald) really pushed more for the quality of the work."
As with any supervisor or manager, if someone was idly sitting around, DeWald would show concern, but as long as the inspectors were keeping busy and the work was being performed.
Holley testified that DeWald seemed satis-fied.
(Tr. 5112.)
127.
Bowman could not estimate the number of inspec-tions that he performed to determine what an average might be; instead, he gave the examples that he had spent as many as four days inspecting one hanger, and yet he had perhaps inspected as many as 20 hangers in a single day.
(Bowman, Tr. 6865.)
128.
Wicks testified that he believed quantity was emphasized over quality because LKC was trying to meet Edison deadlines.
(Wicks, Tr. 7077.)
Wicks could recall no basis other than shop talk for his belief that Comstock stressed quantity over quality.
(Wicks, Tr. 7087-7088.)
Wicks had l
never personally experienced any instance in which he had been pressured to get more work done.
(Tr. 7088.)
129.
Another inspector testified that management should inform the inspectors how much work existed and charse-l
(
a
1 -
terized the weekly meeting as doing just that.
(Snyder, Tr. 4245.)
Snyder also testified that although DeWald did not talk at the meetings about quality, such statements were not necessary because it was understood that the inspectors had to answer for their inspections.
(Snyder, Tr. 4253-4256.)
130.
Seven inspectors were sure that DeWald, during these meetings, never set a minimum number of inspections to be performed by an inspector in one day.
(Bossong, Tr. 9962-9963; Gorman, Tr. 5782; Holley, Tr. 5111; Hunter, Tr. 8543-8544;
)
Mustered, Tr. 4995, 5050-5051; Perryman, Tr. 9665; Peterson, Tr. 5918.) 'However, three inspectors vaguely remembered DeWald i
talking about such a minimum but understood it to be merely a goal (Martin, Tr. 9240-9241), or an attempt to put the elimina-tion of a backlog in perspective (Bowman, Tr. 6866-6867) or a way of explaining that Comstock wanted a day's work for a day's pay.
(Rolan, Tr. 4757-4758, 4858-4860.)
131.
Gorman, Holley and four other inspectors testi-fied that LKC was exerting pressure to produce because they had heard CECO was going to cancel LKC's contract if the backlog of inspections was not eliminated by certain projected dates.
(Gorman, Tr. 5840-5841; Holley, Tr. 5151-5152; Bossong, Tr. 9857; Hunter, Tr. 8499-8500, 8744; Peterson, Tr. 5950-5951; Seeders, Tr. 7567-7569.)
132.
The only basis that three of these six inspectors could give for believing that Comstock might lose its contract I
was shop talk among inspectors.
(Bossong, Tr. 9857; Gorman, Tr. 5840-5841, 5871, 5884-5885; Seeders, Tr. 7568.)
- Seeders,
\\ however, testified that such shop talk was fairly common when LKC was not meeting its deadlines.
(Seeders, Tr. 7568.)
Perryman testified that there were rumors at LKC "every day" about "anything you want to come up with."
(Perryman, Tr. 9664.)
133.
Three inspectors recalled a meeting at which LKC had suggested that it was in danger of losing its contract.
Holley recalled a meeting in the summer of 1984 at which Seltmann indicated that if the backlog of inspections was not eliminated it could mean that the livelihood of Comstock at Braidwood would be lost.
(Holley, Tr. 5151.) Hunter testified that more than once at the weekly meetings during 1984 DeWald had stated that LKC was in danger of losing its contract if it failed to satisfy certain promised completion dates.
(Hunter, Tr. 8499-8500, 8655, 8744, 8747, 8751.)
Peterson recalled a i
special meeting where Seese indicated that things were looking very critical for Comstock, that they needed to eliminate the backlog, and that everybody's help wculd be appreciated.
(Peterson, Tr. 5950-5951.)
134.
DeWald testified that he had heard a rumor that LKC was in jeopardy of losing its electrical contract in January 1985.
He had not heard such a rumor at any other time.
(Tr. 1347, 1349.)
By January 1985, DeWald had no inspection backlog.
(Tr. 1347.)
The only basis of which DeWald was aware for such a rumor was that at that time, LKC's production department was not keeping up with the site's conduit and cable installation needs.
(Tr. 1348-49.)
Saklak testified that he
- was not aware of any rumor that LKC was in danger of losing its contract in January 1985.
(Tr. 8137-38.)
135.
Shamblin testified that since May, 1984 he was generally satisfied with the performance of Comstock's QC department and was not aware of any threats to terminate Comstock's contract because of insufficient production by Comstock's QC department (Shamblin Reb. Prep., Test. at 30 ff.
Tr. 16274.
136.
Comstock's QC status tracking system was one of its most important management tools to monitor QC progress and allocate resources to effectively address the inspection backlog and to keep current with ongoing installations.
In i
addition, such a system enabled management to project the completion of various inspection tasks, such as elimination of the backlog which existed in the spring of 1984.
Such a projection was based on an estimate of how many inspections, on average, an inspector was likely to accomplish in a day.
(See, e.g.
Int. Ex. 12.)
(Seese, Prep. Test. at 8-10 ff. Tr. 2320; Seese, Tr. 2350-51; Saklak, Tr. 8116-18.)
137.
Comstock's schedules and status reports reflected the goals and progress of inspection groups and areas, not the progress of individual inspectors.
Although the status report figures showing the number of inspections conducted were compiled from Lead Inspector summary sheets based on individual inspectors' reports, those individual reports were not used routinely to monitor individual inspector performance.
The Comstock QC inspectors were informed of the way in which this
, compilation was prepared and the use to which the status reports were put.
(Seese, Tr. 2365-66, 2368-69, 2377-81, 2522-26, 2502-03, 2517-18; Seese, Prep. Test. at 10 ff.
Tr. 2330; DeWald, Tr. 1420-21; DeWald, Prep. Test. at 20 ff.
Tr. 1700.)
138.
Larry Seese, LKC's Assistant QC Manager, compiled the lead inspector summary sheets into a table which tracked LKC efforts to reduce the inspection backlog (before the backlog was eliminated) and keep up with current inspections.
Seese had been a QC Supervisor for LKC at the Perry nuclear plant when LKC developed this tracking system.
(Seese, Prep.
Test, at 7, 10 ff. Tr. 2330; Seese, Tr. 2332-33.)
139.
LKC management explained to the LKC inspectors that those reports were not production quotas and began posting those reports as of October 9, 1984 so that the inspectors could see that the reports were not " sinister."
(Seese, Tr. 2498-99, 2372, 2381-82; Seese, Prep. Test, at 9 ff.
Tr. 2330; DeWald, Tr. 1576-78.)
LKC management hoped that posting the reports would, as it had at the Perry plant, show I
the inspectors LKC's goals and progress toward those goals, fostering a spirit of teamwork and togetherness and a feeling that each inspector was playing a role in LKC's overall success.
I i
(Seese, Prep, Test. at 9 ff. Tr. 2330; Seese, Tr. 2498-99;
)
Saklak, Tr. 8209-10.)
140.
LKC's use of status reports for evaluating group a
progress is confirmed by testimony from LKC's management that the variations in difficulty from inspection to inspection i
--,,------v., - -
-a--
w
- would not even permit formulation of any individual quotas.
(Seese, Tr. 2341-44; Seese, Prep. Test. at 11 ff. Tr. 2330.)
LKC's tracking program allowed management to monitor progress toward those goals.
When informed by the tracking system that actual progress fell behind desired goals, LKC management was able to take actions such as increasing inspector overtime hours, hiring additional inspectors, and transferring inspec-tors from one inspection area to another.
(Seese, Prep. Test.
at 8-10 ff. Tr. 2330; Seese, Tr. 2350-51; Saklak, Tr. 8116-18.)
141.
There are two instances in the record where the use of an inspector's daily status reports to calculate his individual productivity was discussed.
The first concerns the I
circumstances surrounding a written warning given Inspector Hershel Stout; the second concerns Inspector Mark Klatchko's perceptions of how the daily status reports were used.
142.
Herschel Stout did not appear as a witness; instead, the parties designated his deposition for inclusion in the record.
On March 19, 1985, DeWald had delivered a written reprimand to Stout, prepared by Saklak, for excessive ab-senteeism and " poor performance."
(Stout Dep. Ex. 2.)
Saklak had calculated that Stout was performing an average of 1.6 inspections per day, and that Stout's work attendance reflected an unacceptable " Monday-Friday" syndrome (id).
i 143.
Between June and December, 1984, Stout had not i
reported to work on eight Mondays and four Fridays.
His pay was docked for_seven of those absences.
Between January and l
March 19, 1985, Stout called in that he was taking four days of i
--- - -- ~ -
vacation; in addition, he was absent on three Mondays, a Thursday, a Friday, and worked shorter days twice.
(Stout Dep.
Ex. 2.)
144.
Using Stout's daily status reports, Saklak calculated that for the seven weeks between January 28th and March 19th, 1985, Stout had worked a total of 166.3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br /> and performed a total of 34 inspections.
(Stout Dep. Ex. 2.)
Stout agreed that these calculations were accurate.
(Stout Dep. Tr. 116.)
A review of those status reports on the record indicates that Stout partially inspected and reinspected the same hanger on numerous days.
(DeWald, Prep. Test., Attch. 4 ff. Tr. 1700; Simile Tr. 3487-98.)
Larry Bossong, who had acted as Stout's lead, initially testified that the March 19 warning was unfair.
After reviewing it on the witness stand, Bossong agreed that Stout's productivity was " unusually low."
(Bossong, Tr. 9897.)
Although this was the only time his pro-ductivity was calculated, during the 15 months of his employ-ment at LKC, Stout received at least four other warnings for unexcused, excessive absences.
(Stout Dep. Ex. 2; Stout Dep.
Tr. 7-8.)
Several threatened his termination but Stout left LKC voluntarily in July, 1985.
(id.)
145.
In Stout's opinion, the fact that his pro-ductivity had been calculated itself meant that "a quota" was being imposed.
Other than the mere fact of his March 19, 1985 l
l reprimand, no member of LKC management ever led him to believe that a quota of some sorts was being imposed on him.
(Stout Dep. Tr. 85-86.)
In particular, Stout stated that Saklak never
_- spoke to him either about his absences or his productivity.
(Id. at 30.)
DeWald did.not impose any minimum requirement on Stout, nor did the warning identify any numerical expectations; DeWald merely informed Stout that other inspectors were per-forming five or six inspections a day.
(Id. at 94.)
Stout testified earlier that DeWald may have mentioned that others were performing six or eight a day.
(Id. at 32.)
146.
Stout testified that DeWald never suggested that quality should be sacrificed for quality.
(Stout Dep. 96.)
Stout believed that if an inspector performs an inspection, he ought to do the best job he could without regard to the number of inspections.
(Id. at 102.)
He did not feel that at any time he was forced to sacrifice quality during his employment.
Other than his March 19 reprimand, DeWald gave Stout no other indication that he wanted quantity more than quality.
No other member of LKC management led Stout to believe in any respect that quantity was more important than quality.
(Id. at 103-05, t
123-24.)
l 147.
Mark Klachko is employed as a QC inspector.
He was first employed in 1982.
He became a lead welding inspector f
in August, 1983 and resigned from that position in February, 1985 (Klachko Dep. Tr. 20; Klachko Dep. Ex. 2).
148.
As a lead inspector, Klachko was responsible for assigning work to inspectors and collecting their daily status reports for transmission to Comstock management.
(Dep.
Tr. 30.)
Klachko believes that Seese was using the daily v.atus reports strictly to determine the overall status of the
.. _ job.
(Klachko Dep. Tr. 54-55, 193.)
This was confirmed for Klachko when Seese told him to simply send daily totals for the group of inspectors under Klachko's lead and that Seese's needs would be satisfied by such an approach.
(Klachko Dep.
Tr. 54-55.)
Klachko followed this practice for a period of time, although he also attached the individual status reports to the group summary.
(Id.)
Klachko acknowledged that it was proper for management to use the status reports to determine the overall progress of the job.
(Dep. Tr. 51.)
149.
Klachko was approached on a number of occasiens by Saklak and DeWald when they inquired into an individual inspector's productivity.
(Klachko, Dep. Tr. 32-33.)
Klachko believed that Saklak and DeWald relied on the individual status reports to monitor an inspector's productivity.
(Dep.
Tr. 31-32.)
Saklak and DeWald would question Klachko re-garding disparities in inspector performance.
(Id.)
150.
Klachko would tell the inspector that he was being watched and that the status reports were being used as f
" counts" of each inspector's productivity.
(Dep. Tr. 33.)
Klachko never punished any inspector for low productivity, but was aware that inspectors were questioned regarding their productivity by Saklak or DeWald.
(Id. at 33, 35.)
This is apparently what occurred with Inspector Danny Holley, the only inspector Klachko was able to identify by name.
(Klachko Dep.
Tr. 101-102; see findings 160-162, infra.)
l 151.
Klachko himself wasn't concerned about the number of inspections performed, but rather that they were correctly l
l l
performed.
(Klachko Dep. Tr. 105.)
The quality of an inspec-tor's inspections was monitored by Klachko's review of PTL overinspection reports.
(Klachko Dep. Tr. 107.)
Inspectors were concerned about PTL rejections of their work.
(Id. at Tr. 107-108.)
152.
Klachko resigned his position as lead inspector on February 19, 1985.
(Klachko Dep. Tr. 72.)
Because of the tension between Comstock management and the inspectors on a variety of monetary, training and morale issues, Klachko felt as lead inspector he was a buffer between two antagonists.
This role was placing too much pressure on him and he resigned.
(Tr. 72-79.)
153.
Apart from the two proceeding instances in which there was evidence linking the daily status reports and manage-ment statements to inspectors regarding low productivity, there were also other instances in which Comstock management dis-cussed an apparently low number of inspections with individual inspectors.
These instances were also regarded by some inspectors as a push for quantity, rather than quality, of inspections.
154.
QC Inspector Dean Peterson testified that his l
l lead, John Walters, and his supervisor, Ken Worthington, had called him in to note that his productivity was very low.
(Peterson, Tr. 5904.)
Peterson explained his assigned area had welds covered with galvanized spray paint which had not been j
adequately cleaned, and that he needed to clean the welds i
before conducting his inspections.
(Peterson, Tr. 5905-09.)
155.
Neither supervisor gave Peterson the impression that they wanted him to ignore the paint or accept painted welds.
(Peterson, Tr. 5912.)
No one ever directed him not to
~
report any defects he might see in a weld.
(Peterson, Tr. 6072.)
Worthington, however, stated that he couldn't warrant overtime for Peterson.
(Peterson, Tr. 5915.)
Peterson could not recall whether or not he was assigned overtime after this conversation.
(Peterson, Tr. 5915-16.)
156.
QC inspectors typically expected craft personnel to remove paint; QC inspectors did not generally remove exces-sive paint themselves.
(Peterson, Tr. 5907, 5913-14, 6098-99.)
The inspectors carried picks and wire brushes with which to clean those welds which the paint removal crews had not cleaned adequately.
(Peterson, Tr. 5913; 6098-99.)
Peterson and others had complained to their supervisors about the paint on welds left by paint removal crews.
(Peterson, Tr. 6038.
6053-54, 6104-06.)
157.
After this meeting with his supervisors, Peterson no longer cleaned off the remaining paint on the welds and inspected them; instead, he did not inspect the welds but l
merely listed them on his status reports as painted, and continued on to other components which were unpainted and inspectable.
(Peterson, Tr. 6040-42.)
His supervisors then called him in for a second meeting, apparently to discuss his status reports.
He told them he wouldn't spend any more time 1
cleaning welds, and asked that he be transferred to another type of inspection.
(Peterson, Tr. 6042-44.)
r----
g,-m-y, r w m.m.y w
y ww s w.-
.s-
+
u
-m
---~.e.--.---
w.
v+---
..----e-
- - + +
.. _. 158.
Walters told Peterson that he had looked at some of the welds and "they weren't too bad."
Peterson interpreted f
that to m'ean that it shouldn't have taken too long to clean
-them; he did not believe Walters was insinuating that Peterson should inspect them through paint.
(Peterson, Tr. 6048.)
Peterson conceded that it may have been taking him longer to clean the welds than it took for other inspectors.
(Peterson, Tr. 6052.)
Despite these conversations, Peterson did not change his inspection practices.
(Peterson, Tr. 5912-13; 5917; 6100.)
159.
A week or two later, Tony Simile asked Peterson if he would like to be transferred to the "in-process" inspec-tion area.
Peterson said he would, since it was an area in which he had wanted to work.
(Peterson, Tr. 6099.)
We find that this transfer was not " retaliatory," since Peterson i
i requested to be transferred and was pleased with his new assignment.
160.
Holley testified that DeWald had spoken to him once or twice with respect to the time it took him to complete r
his own inspections.
DeWald told him he could perform more inspections than he was performing, and that while he was a slow inspector, he also was a good inspector.
(Holley, Tr. 5112.)
Holley agreed with him that he could probably be more productive, but did not feel in any respect that DeWald was urging him to overlook construction concerns or the quality of a product in order to complete more inspections.
(Holley, Tr. 5115.)
Holley did not feel harassed in any respect by t
I t
,__,en...,.
.n
_n,..,,,, _.
DeWald's conversation with him.
Nor did this cause Holley to change the manner in which he was performing his inspections.
(Holley, Tr. 5116.)
161.
Based on shop talk, Holley understood that an average number of inspections would be.eight or ten per day in the welding and configuration areas; DeWald, however, never indicated to Holley the number of inspections.he expected Holley to perform.
(Tr. 5117-18, 5127, 5251.)
An understand-ing that eight inspections per day were to be performed was also recalled by Inspector Rolan.
(Tr. 4862) 162.
Holley recalled that at various times DeWald, Saklak, Mike'Kast, Mark Klachko, Joe Hii, Chuck Tyler, and perhaps others, had approached him from time to time about his productivity.
(Tr. 5124.)
To the best of his recollection, each of these individuals accepted his explanations about the length of time it took him to perform his inspections, although Saklak tended to become excited.
(Tr. 5124-26.) None of the incidents caused Holley to feel that he ought to overlook construction defects in order to increase productivity.
He felt he was a very conscientious inspector.
(Tr. 5123, 5128.)
163.
Leonard McGregor testified that at some point during the NRC visit by the group of 24 LKC QC inspectors, a request was made for a show of hands to determine how many of l
the inspectors agreed with the allegation that Comstock QC was l
l emphasizing quantity over quality.
(Tr. 17534.)
Stout stated l
l that he both made the allegation and requested the show of hands.
(Stout Dep. p. 23.)
McGregor could not recall whether l
l l
s-
. the question came from the Region or from his fellow NRC inspector, Robert Schulz, or whether the request had come from the NRC at all.
(Id.) He did not recall counting hands, but believed that the inspectors' agreement with the statement was verbally unanimous without abstentions or denials.
(Id.'
McGregor testified that he and Schulz might have told Weil, who was at the other end of the conference call to the Region, or others at the Region later that the inspectors were unanimous in their agreement with the allegation.
(Tr. 17535.)
164.
Snyder testified that although he had raised his hand, he had no personal knowledge of whether or not the allegation was true.
Rather, he based his agreement only on what he had been told by and what he had heard from other inspectors.
(Tr. 4267-69.)
165.
Therman Bowman had raised his hand to reflect his agreement with the allegation by Stout that Comstock was i
emphasizing the quantity of inspections over the quality of inspections.
(Bowman, Tr. 6857.)
His agreement was based only on his understanding that two other inspectors had been pres-sured to do more inspections than they were comfortable with:
Danny Holley and Dean Peterson.
Bowman personally had not been pressured in any such way; rather, Bowman had agreed "in sympathy".
(Bowman, Tr. 6857-60.)
166.
Holley had once told Bowman that a supervisor had questioned Holley's slowness.
(Bowman, Tr. 6858-59.)
167.
With respect to Peterson, Bowman believed he had been singled out by Worthington and Walters for discussion at a l
. _ _ = ___
P group meeting, and that Peterson had been averaging one and one-half welding or configuration inspections per day.
(Bowman, Tr. 6859-60.)
Hunter also recalled that Peterson had been told he was not performing enough inspections, although Hunter believed no such statement was made in front of the QC group as a whole.
(Hunter, Tr. 8545.)
168.
Bowman testified that while it appeared LKC wanted more inspections to be performed by Peterson, there was no suggestion that Peterson sacrifice any QC criteria to t
4 achieve greater productivity.
(Tr. 6861.) Bowman also testi-fied that Worthington had suggested Peterson was not applying himself vigorously enough in performing his research and getting into the field in a timely fashion.
(Tr. 6861.)
t 169.
Gorman had agreed with the assertion at the March 29 meeting that LKC was stressing quantity over quality based on his interpretation of discussions he had with another inspector and from the general meetings.
Gorman recalled that Holley had told Gorman that he was being pushed to be more productive.
(Gorman, Tr. 5832.)
Gorman admitted that Holley I
had never told him that he was required to perform a particular number of inspections per day, or that any kind of quota was being imposed on him, or what the basis was for his feeling of pressure to perform more inspections.
Nor did Holley explain l
to Gorman in any respect that he had been instructed to ignore the quality of his inspections for quantity.
(Gorman, Tr.
l 5874, 5877.)
_ 170.
With respect to the show of hands regarding Comstock's interest in quantity over quality at the March 29th visit, Hunter testified that he had raised his hand in agree-ment.
He agreed with that statement only because he had heard of a program that LKC said had to be completed by a certain date in order to retain the contract from Edison.
(Hunter, Tr. 8655.)
Hunter subsequently amended his testimony to reflect that DeWald's threat regarding LKC being in danger of losing the contract was in connection with an audit; Hunter then retracted that testimony and referred instead to inspec-tion activities.
(Tr. 8745-46.)
171.
Hunter did not believe he had ever heard DeWald tell anyone that they had to do a certain number of inspec-i tions.
(Tr. 8544.)
No one ever criticized Hunter for low productivity or low numbers of inspection performed per day.
f He believed he was one of the top inspectors in terms of the number of inspections he performed per day.
(Id.)
He believed that he probably averaged better than five hangers per day.
(Id.)
172.
Larry Bossong, who was present at the March 29 meeting and ultimately became a union steward, testified that no one had ever suggested in any respect that he was not performing enough inspections.
(Bossong, Tr. 9883-84; Bossong, Prep. Test. at 1 ff. Tr. 16,252.)
He was never told that he was expected to perform a set number of inspections in a given time period, i.e.,
a quota.
(Bossong, Tr. 9884.)
He was aware of only two inspectors who had been approached by management about low productivity:
Herschel Stout, and an individual whose overtime was threatened by Walters (Peterson).
(Bossong, Tr. 9838; 9841; 9883-84.)
173.
Bossong believed that it was management's right to calculate the productivity of an individual QC inspector.
(Bossong, Tr. 9962; 9883-85.)
174.
Bossong had heard from talk among QC inspectors that the " rule of thumb" was one inspection per hour, but that this depended on the complexity and accessibility of the component.
(Bossong, Tr. 9963-64.)
He had also heard this rule of thumb at another nuclear power station construction site.
(Bossong, Tr. 9974.)
175.
Mike Mustered, also present at the March 29 meeting, testified that he had never received any direction to establish a quota for the number of inspections each inspector was expected to perform in a day's work.
(Tr. 4995.)
When he became a lead inspector, no member of LKC supervision or management ever told him that a certain number of inspections were to be performed or expected from the inspectors for whom he was acting as lead.
176.
Mustered was certain that there was no minimum quantity inspection requirement because there had been times when he had taken two or three days to complete one inspection.
(Mustered, Tr. 4996.)
For example, in July of 1984 Mustered spent a week inspecting one hanger for configuration and weld acceptability.
He was not criticized in any respect for taking
1 l so long.
(Mustered, Tr. 4999.) That hanger was approximately six feet wide and upwards of 30 feet in height.
(Id.)
177.
During a period of time when he was not a lead inspector, Mustered was one of the QC inspectors who was involved in reducing the inspection backlog.
(Murtered, Tr. 5012.)
He received.no specific instructions regarding its completion within any specific period of time.
(Id.) During the course of that assignment, he was never informed by his lead or any LKC supervisor that CECO was upset with the backlog or that it needed to be reduced in a prompt fashion.
(Tr. 5013.)
178.
Joe Hii denied that he had raised his hand to signify that he agreed with the assertion made at the March 29 meeting that Comstock was emphasizing quantity over quality.
(Tr. 16647-49.)
179.
While LKC QC worked to complete the backlogged inspections during the spring and summer of 1984, Shamblin and other CECO officials became aware that morale problems were surfacing.
(Shamblin, Prep. Test., Attch. 4 ff. Tr. 16274.)
As reflected below, many of these problems were rooted in events which predated or were unrelated to the workload still facing LKC in the summer of 1984.
We now turn to those in-fluences and the roles they played in LKC QC morale.
180.
As of August 1983, when DeWald became QC Manager for LKC at Braidwood, the Level II QC inspector pay scale ranged as low as $7.00 per hour for new inspectors.
(DeWald, Prep.
Test, at 10-11 ff. Tr. 1700.)
Because higher inspector i
l l
salaries were available at other nuclear plants, a number of QC inspectors left LKC and Braidwood in 1983 and early 1984 for other jobs in the industry.
(Id. at 8; Int. Ex. 9; DeWald, Tr.
1334-35.)
181.
At this same time, LKC was attempting to expand its workforce to meet the challenge of the inspection tacklog.
In order to attract new inspectors, LKC began offering salaries ranging from $11.50 to $12.00 per hour, depending on qualifica-tions and prior experience.
(DeWald, Prep. Test. at 7-8 ff.
Tr. 1700.)- By early 1984, LKC offered new inspectors as much as $16.00 per hour in order to obtain experienced and qualified individuals and to satisfy its unfulfilled manpower needs.
(DeWald, Prep. Test. at 8 ff. Tr. 1700.)
This led to a situa-tion in which more senior inspectors were making less money than new employees.
(DeWald, Prep. Test. at 7-8 ff. Tr. 1700; Saklak, Tr.
8195.)
182.
Faced with dissatisfaction over its salary structure, in mid-April 1984, LKC's corporate office changed all LKC Level II QC inspectors' minimum hourly wages to $12.00, with an additional S.50 per hour for each additional certifica-tion obtained by the iGt;ect>r.
This pay scale became effec-tive retroactively as of.sgell 1, 1984 on a companywide basis.
(DeWald, Prep. Test, at 9-11 ff. Tr. 1700; DeWald, Tr. 1335; Irt.
Ex. 9.)
183.
Since an increase in hourly salary of $.50 over the new $12.00 minimum wage was tied to each additional certifi-cation obtained by an inspector, inspectors were eager to begin training to obtain further certifications.
(DeWald, Prep.
Test. at 11-12 ff. Tr. 1700; DeWald, Tr. 1336; Int. Ex. 274 at 6-7.)
In turn, as inspectors obtained additional certifica-tions, LKC management expected to have more flexibility in assigning inspectors to areas where they were most needed.
(Maiman, Tr. 3920; DeWald, Tr. 1333.)
184.
LKC attempted to arrange inspector training without interfering with its efforts to address the inspection backlog and the regular daily inspection process.
(DeWald, Prep. Test.
at 12-14 ff. Tr. 1700.)
However, the majority of LKC inspectors began demanding that they receive cross-training immediately in order to obtain additional certifications and thereby increase their salaries.
(Id. at 13, Int. Ex. 274 at 7.)
Inspectors also undertook to cross-train each other on their own without scheduling the sessions through LKC manage-ment.
As a result of these factors, inspector productivity and morale suffered.
(DeWald, Prep. Test. at 13 ff. Tr. 1700; DeWald, Tr. 1336; Int.
Ex. 278.)
185.
Further dissatisfaction arose from the fact that in order to continue attracting additional inspectors, LKC had to continue to pay at least some of its new hires $16 per hour, rather than the $12 minimum wage paid to most of the existing cadre of inspectors.
(Snyder, Tr. 4031; Rolan, Tr. 4955; Bossong, Tr. 9946; Klachko, Dep. at 73-74.)
The inspectors were unhappy with the resumption of salary disparities, both as some inspectors received new certifications more quickly than others and as new inspectors were hired at higher salary
i
, levels.
Inspector dissatisfaction increased further because some new inspectors were given priority to obtain certifica-tions so that they could begin performing inspection.
(DeWald, Prep. Test. it 14 ff. Tr. 1700; Shamblin, Prep. Test. at 24 ff.
Tr. 16274.) Because the existing inspector workforce was required to conduct this training, their own cross-training was necessarily delayed.
(Shamblin, Prep. Test. at 24 ff. Tr.
16274.)
186.
Matters became even more complicated when a fire damaged LKC's QC office at Braidwood in June 1984.
Some of the QC inspectors were relocated while the damage was repaired, and their temporary quarters were cramped and generally unsatis-factory.
(DeWald, Prep. Test. at 14 ff. Tr. 1700; DeWald, Tr.
2042-43; Shamblin, Prep. Test. at 24 ff. Tr. 16274.)
The inspectors were unhappy with those working conditions, as evidenced by QC inspector John Seeders' expression of concern in August 1984 meetings with Shamblin and James Gieseker of Edison.
(Gieseker, Prep. Test. at 9 ff. Tr. 2771; CECO Ex. 93; Seeders, Tr. 7471.)
Accordingly, Shamblin directed that additional facilities be provided to improve the inspectors' temporary offices and directed construction forces to expedite the restoration of the LKC offices.
(Shamblin, Prep. Test, at 26 ff.
Tr. 1674.)
By October 1984 the LKC QC office had been restored and all QC personnel were back in the same area.
(Id.
at Attch.
4.)
187.
Dissatisfied with salary conditions, with cramped quarters, and with the slowness of obtaining additional raises through cross-training, LKC inspectors turned to union-ization efforts.
(DeWald, Prep. Test. at 8-10 ff..Tr. 1700.)
188.
In 1983, Local 306 of the United Association of Journeymen and Apprentices of the Plumbing and Pipefitters Industry contacted LKC QC inspectors at Braidwood.
(Maiman, Prep. Test. at 9 ff. Tr. 3806.)
As hiring increased in early 1984, LKC hired 12 individuals with previous union experience.
Eleven of these 12 inspectors openly spearheaded the drive to unionize LKC QC inspectors.
(DeWald, Prep. Test. at 15 ff. Tr.
1700; Saklak, Tr. 8174-75.)
Union activity began in earnest around July 1984.
The inspectors held several evening meet-ings, and small groups occasionally gathered during working hours to discuss unionization.
(DeWald, Prep. Test. at 16 ff.
Tr. 1700; Maiman, Prep. Test. at 9-10 ff. Tr. 3806; DeWald, Tr.
1809; Seese, Tr. 2597-99; Saklak, Tr. 8212-13.)
189.
Unionization efforts created additional tension among the inspectors and between the inspectors and LKC manage-ment.
(Shamblin, Prep. Test. at 28 ff. Tr. 16274.)
Not all of the inspectors had been in favor of unionization, and LKC vigorously opposed the unionization efforts.
(Hii, Prep. Test.
at 5-6 ff. Tr. 16608; Int. Ex. 33 at 121; DeWald, Prep. Test.
at 16 ff. Tr. 1700; Bossong, Tr. 9941.)
Moreover, during the unionization effort, LKC management perceived that the amount of inspection work appeared to slow down, absenteeism was high and some inspectors were openly defiant, attempting to aggravate their supervisor.
(DeWald, Prep. Test, at 16 ff. Tr.
l 1700.)
i 190.
In a vote held in late November 1984, LKC QC inspectors approved the union by a vote of 43-30.
(DeWald, Prep.
Test. at 16 ff. Tr. 1700; Maiman, Prep. Test. at 10 ff.
Tr.
3806.)
However, because appeals against the vote were filed by a member of the voting body and by LKC corporate management, the election was not certified until April 1984, and contract negotiations over salary and working condition.
issues did not commence until late April or May 1985.
(DeWald, Prep. Test, at 16-17 ff. Tr. 1700; Seltmann, Prep. Test. at 11 ff. Tr. 1960; DeWald, Tr. 1809; Shamblin, Prep. Test. at 28-29 ff. Tr. 16274:
Maiman, Prep. Test. at 10 ff. Tr. 3806.)
191.
In the time span between the vote and its certification the following spring, morale, which had improved with the union vote, again began to decline, absenteeism again became a problem, and tempers flared.
(Seltmann, Tr. 2087-89; Shamblin, Tr. 16307; DeWald, Prep. Test, at 16-17 ff. Tr.
1700.) On at least one occasion, the QC inspectors staged a sick-out.
(Hii, Prep. Test. at 6 ff. Tr. 16608.)
Threats of strikes were constant, and in general, LKC inspectors and management observed that the QC inspectors appeared to enjoy taunting the supervisors.
(DeWald, Prep. Test. at 16-17 ff.
Tr.
1700; DeWald, Tr. ;1801, 1809; Seese, Tr. 2639-40, 2724-25, 2765-66, 2601-02; Hii, Prep. Test. at 5 ff. Tr. 16608; Hii Tr.
16639-40, 16645.)
192.
After the ratification, contract negotiations between LKC and the union over salary structure and working conditions began on April 30, 1985 and continued without
. agreement through July 21, 1985.
During this time, LKC manage-ment and the QC inspectors were working in a difficult at-mosphere created by the stalled labor negotiations.
Both sides were trying to show positions of strength and tempers were growing shorter and shorter.
(Maiman, Prep. Test. at 10 ff.
Tr.
3806.)
193.
Finally, on July 23, 1985, CECO contracted with a job shop organization, BESTCo, to supply QC Inspectors, document clerks, document reviewers, and QC engineers to LKC.
Messrs. Marino and Paserba of LKC's corporate office held a meeting on that day with the LKC QC department and ordered an immediate layoff of all personnel except managers, supervisors and secretaries.
The personnel who were laid off were rehired by BESTCo the next day.
CECO employed BESTCo because it was a party to a previously negotiated national labor contract with the union, which could be used at Braidwood with only minor modifications.
In this manner CECO resolved a problem which had appreciably hampered LKC's workforce relations at l
Braidwood.
(DeWald, Prep. Test. at 17 ff. Tr. 1700; Maiman, Prep. Test, at 10-11 ff. Tr. 3806; DeWald, Tr. 1799-1802.
BESTCo was replaced by GE-MCIS sometime in July 1986.
- Martin, Tr. 9146-47.)
194.
Given the bitterness that existed as a result of the unionization effort at Braidwood, we must give some credence to LKC management's belief that discontent over the delay in certification of the union vote and the resulting
(
union-related tension partially motivated 24 LKC QC inspectors l
I to visit the NRC on March 29, 1985.
The allegations made in that meeting that, among other things, LKC management was promoting quantity over quality of inspections,.may have been made at least partly in hope of speeding the ratification process and forcing the completion of contract negotiations.
(DeWald, Tr. 1399-1400, 1809; Seese, Tr. 1500, 1665-71; Saklak, Tr. 8058-60, 6070-73, 8169-79, 8247-48.)
CECO management shared this view.
(Shamblin, Tr. 16411-12.)
240.
Seeders received a written warning in December, 1983, for failing to appear for work after Thanksgiving.
(Seeders, Tr. 7308; App. Ex. 82.)
Upon hi s return, Seeders claimed to have been ill.
When asked to provide a doctor's excuse, Seeders claimed that request was harassment (Id.;
Seese, Tr. 2559-62.)
Seese had called Seeders about twelve times at home over Friday, November 23 through Sunday, November 25, 1984 and no one answered.
(Int. Ex. 33 at 114-114B; Seese, Tr. 2560-61.)
Seese was trying to reach Seeders because Comstock had discovered that, contrary to procedure, Seeders had used photocopied calibrations checklists.
Comstock wanted Seeders to correct his records prior to an upcoming audit.
(Seese, Tr. 2557-63).
i i
241.
Seeders received a second written warning on August 17, 1984, which identified the following problems with his work and attitude:
Seeders failed to appear to conduct a training class on 7-26-84.
The class was rescheduled on 8-2-84 and Seeders ridiculed the material and used foul language.
i l
1 Seeders'
- lead, L. Phillips, had advised Comstock that he was having difficulty in motivating Seeders and that Seeders wasn't following his directions.
Seeders failed to turn in his daily status report for August 10; the ones for August 14 and August 17, he turned in late.
Seeders failed to complete his research of the calibration files due on August 13 in response to a CECO audit.
(Seese, Prep. Test. Attch. 3 ff. Tr. 2330; Seese, Tr. 2480-90.)
242.
In his letter of August 17, 1984, Seeders claimed that Seese had told him it didn't matter how good a job he was doing; if Comstock didn't like him they would get rid of him.
Seese also told him that if he ever wanted to make a living for his family again, he "had better bow-down to their way of doing things."
(Int. Ex. 23; Seeders, Tr. 7442-44.)
Seese denied making any such statements.
(Seese, Prep. Test, at 13-14 ff.
Tr. 2330.)
243.
In September 1984, Seese identified two instances in which Seeders' poor documentation and judgment continued.
l First, Seeders had issued multiple ICRs to document that a tool had not been returned for calibration on time and that the same tool was returned out-of-calibration.
(Seese, Prep. Test. at 16 and Attch. 5 ff. Tr. 2330.)
Seese informed Seeders that he should have issued a single ICR listing both problems so that engineering personnel could address the full scope of the tool's deficiency at once.
(Id.)
244.
Second, Seeders failed to issue an NCR for lost tools.
The work performed with a lost tool is indeterminate l
because its calibration cannot be checked.
An NCR, rather than I
i c
n-
n, an ICR, must be. issued to assure that CECO evaluates the work performed.with the lost tool.
~ Seeders, however, issued ICRs, which merely documented the fact that tools were lost.
(Seese,
' Prep. Test, at 16-17 and Attch. 5 ff. Tr. 2330.)
Seese informed Seeders that he should have issued an NCR for the lost tools.
(Id.)
Seeders' viewed Seese's comments as nitpicking and harassment.
(Seeders, Tr. 7455-56.)
However, one of Seese's duties is reviewing ICR's and NCR's to ensure their clarity and proper documentation.
(Id.; Seese, Prep. Test. at 18 ff. Tr. 2330.)
245.
In September of 1984, further deficiencies in Seeders' calibration documentation records were uncovered by an investigation conducted by Seltmann and by another CECO QA-department audit.
(Seltmann, Prep. Test. at 16 and Attch.
3 ff. Tr. 1960.)
Despite the fact that Seeders had been given additional instruction on the importance of issuing an ICR for all tools out-of calibration, Seltmann identified at least ten subsequent instances in which Seeders failed to issue an ICR for out-of-calibration torque wrenches.
(Seltmann, Prep. Test.
at 16-17 ff. Tr. 1960.)
Seltmann documented his findings in a memorandum to DeWald dated September 25, 1984.
(Seltmann, Prep. Test. Attch. 3 ff. Tr. 1960.)
246.
Seltmann's investigation also determined that an out-of-calibration torque wrench had been issued to the field 17 times since June 6, when it was first documented as being out of calibration.
It was Geeders' responsibility to ensure L
. that out-of-calibration tools were not used in the field.
(Seltmann, Prep. Test. at 17-18 ff. Tr. 1960.)
247.
During the time of Seltmann's investigation Mr. D. Felz of Edison QA was conducting an audit regarding s
Comstock's calibration activities. (Seltmann, Prep. Test.
Attch. 3 at 1 ff. Tr. 1960.)
Seltmann discussed with Felz Comstock's concerns about its calibration documentation and Felz informed him-that a sample survey of Comstock's calibration records showed that the requirement that an ICR be issued for out-of-calibration tools was being violated in 40%
of the cases. sampled.
(Seltmann. Prep. Test., Attch. 3 at 2 ff. Tr. 1960.)
248.
DeWald intended to terminate Seeders based on the seriousness of the calibration record deficiencies identified by Seltmann, the current CECO QA audit and the initial results of Snyder's review of calibration records.
DeWald prepared a memorandum dated September 27, 1984 detailing the reasons for his decision to terminate Seeders.
(App. Ex. 2; DeWald, Prep.
j Test. at 37-38 ff. Tr. 1700.)
The memorandum was never issued, and Seeders was not terminated. (DeWald, Prep. Test. at 38 ff.
Tr. 1700.)
249.
On September 28, 1984, DeWald met with Shamblin, Gieseker, Seltmann and NRC inspector Schultz.
The discussion turned on whether to terminate Seeders or transfer him to a position with less responsibility.
(DeWald, Prep. Test. at 38 ff. Tr. 1700.)
Shamblin believed that termination of Seeders would have been justified.
(Shamblin, Reb. Prep. Test, at 34 l
ff. Tr. 16274.)
However, he recommended that Seeders be transferred to a less challenging position within Comstock.
Shamblin thought that given Seeders' familiarity with operations at Braidwood, :t would be more cost-effective to transfer Seeders than to hire a new person and train him from scratch.
(Shamblin, Reb. Prep. Test. at 34-35 ff. Tr. 16274.)
250.
Later that day, DeWald met with Seeders and informed him that he could either accept a transfer or be ter-minated.
(Seeders, Tr. 7488; DeWald, Prep. Test. at 38 ff.
Tr. 1700.)
Seeders accepted the transfer, which became effective October 1, 1984, and retained the same rate of pay and other benefits that he previously had.
(Seeders, Tr. 7490, 7528; DeWald, Prep. Test. at 39 ff. Tr. 1700.)
251.
Comstock's 100% review of the calibration records resulted, inter alia, in the issuance of NCR 3419.
(App.
Ex. 37.)
The following generic calibration-related problems were documented in NCR 3419:
(1) Missing certifications and/or calibration records; (2) Incomplete or incorrectly completed documents; (3) Duplication of tool identification numbers; (4) Failure to generate corrective action documents when rejectable calibration results were obtained; (5) Failure to perform adequate evaluation and/or QC review of effect of possible use of out-of-tolerance tools on past work; and (6) Release of out-of-tolerance tools for use in the field prior to closure of corrective action documentation.
(Gieseker, Prep. Test, at 16-17 ff. Tr. 2771.)
s 252.
Snyder, who performed the investigatory work that led to the issuance of NCR 3419 (Snyder, Tr. 4160), testified that given the mess Seeders had made of the calibration records it was proper to remove him from that area because otherwise the whole program could have been jeopardized.
(Tr. 4261-4262).
Synder testified that Seeders had not been performing his duties in accordance with the provisions of Comstock Procedure 4.9.1.
Specifically, Snyder stated that several problems, including failure to' initiate documents and missing documents, were evident.
(Tr. 4161.)
Snyder stated that Seeders failed to issue an ICR for an out-of-calibration torque wrench as required by Comstock Procedure.
(Tr. 4106; 4108-4114; App. Ex. 28.)
Snyder testified that, to the extent that such violations of Comstock procedure were obvious, it indicated that Seeders was unable or unwilling to follow such procedures.
Finally, Snyder stated that it took him almost six months to restore order to the calibration records " mess" and to keep pace with his current inspections.
(Tr. 4276.)
253.
Seese testified that, based upon his review of NCR 3419, Seeders' poor work performance contributed to the scope of NCR 3419.
(Seese, Prep. Test. at 19 ff. Tr. 2330.)
With respect to the generic problems identified in NCR 3419, Gieseker testified that LKC's then-existing procedures and instructions were sufficient to assure the integrity of the calibrations and calibration records, if those procedures and instructions had been followed.
(Gieseker, Prep. Test. at 20 ff. Tr. 2771.)
L
3 254.
Comstock has implemented corrective actions to disposition the generic problems and concerns identified in NCR 3419 and taken steps to prevent their recurrence, including transferring Seeders.
(App. Ex. 37; Gieseker, Prep. Test. at 18 ff. Tr. 2771.)
255.
Two other NCR's relating to calibration problems, NCR 3406 and NCR 3472, were also issued during the fall of 1984.
Comstock took corrective action for the deficiencies identified in NCRs 3406 and 3472 and took steps to prevent recurrence.
(Gieseker, Prep. Test, at 16-17, 19 ff. Tr. 2771.)
256.
Worley Puckett was fired as Comstock's Level III welding inspector at Braidwood on August 27, 1984 and replaced by Tony Simile. (DeWald, Prep. Test. at 44 ff. Tr. 1700; Simile, Prep. Test. at 6-7 ff. Tr. 3305; Simile, Tr. 3343.)
Intervenors assert, and Applicant denies, that Puckett was fired because he identified safety concerns.
Puckett told the NRC that he had been fired for this reason and NRC inspector Jerome Schapker investigated the technical concerns that Puckett expressed. (Schapker, Prep. Test, at 3-4 ff. Tr. 10954.)
The following findings weigh the evidence on whether Puckett was fired improperly and on whether he identified construction defects that remain uncorrected.
257.
In the spring of 1984, Tom Vogt, Comstock's Corporate Level III Welding Engineer, reviewed Comstock's weld procedures and identified inconsintencies in them.
(DeWald, Prep. Test. at 40-41 ff. Tr. 1700; DeWald, Tr. 1763-64.)
As a result, Comstock decided that they needed a site Level III weld t
r inspector who could devote full-time attention to the welding program, identify additional problems and implement corrective action.
(DeWald, Prep. Test. at 41 ff. Tr. 1700; DeWald, Tr.
1763.)
258.
On May 15, 1984, DeWald interviewed Worley Puckett.
Puckett's resume reflected experience which'would enable him to fill the proposed Level III position.
(DeWald, Prep. Test, at 42 ff. Tr. 1700.)
259.
There was testimony about this prior experience.
Puckett had previously worked for Kaiser Engineers at the Zimmer Nuclear Power Plant where, in March 1978, he was made Chief Weld Engineer, responsible for developing welding proce-dures and procedures for qualifying welders.
(Puckett, Tr.
5295, 5300-01; App. Ex. 187 at 41.)
In October 1981, Kaiser hired Manfred Goedecke as Weld Manager.
(App. Ex. 187 at 18; Puckett, Tr. 5302.)
Goedecke observed how Puckett ran the welding department (App. Ex. 187 at 33-34), and he appointed a task force which determined that all of the procedures had to be rewritten because none of them satisfied applicable Code requirements.
(Id. at 41-42.)
260.
Goedecke concluded that Puckett lacked the abilities necessary to control welding at a nuclear site.
Puckett did not know how to interpret the codes and could not even determine which code to apply to a given problem.
(Id. at 34, 36-37.)
Goedecke therefore replaced Puckett as Chief Weld Engineer and reduced his responsibilities.
(Id. at 38-40.)
After some retraining, Puckett became better L
at identifying the proper code, but still could not do so consistently.
(Id. at 97.)
Goedecke did not believe Puckett was qualified to perform as a Level III at Braidwood.
(Id. at 102-03.)
The resume reviewed by DeWald does not clearly describe Puckett's reduced responsibilities at Zimmer.
(Int.
Ex. 26; Puckett,.Tr. 5341-42.)
If Shamblin had known that Puckett's responsibilities had been reduced, he never would have approved his employment as a Level III inspector at Braidwood.
(Shamblin, Tr. 16456-58.)
261.
On May 29, 1984, Puckett began working for Comstock at Braidwood.
(Puckett, Tr. 5358; DeWald, Prep. Test, at 42 ff. Tr. 1700.)
DeWald explained to Puckett that he would have to become certified as a Level III inspector.
(DeWald, Prep. Test, at 42 ff. Tr. 1700.)
He was also given a copy of a memo from DeWald which listed the procedural incon-sistencies identified by Vogt and was told that he would be responsible for resolving those problems as well as identifying and resolving any additional problems.
Puckett was also told he would have to resolve difficult inspection interpretations raised by Level II inspectors.
(DeWald, Prep. Test, at 42 ff.
Tr. 1700; DeWald, Tr. 1324-25, 1649; Puckett, Tc. 5360-5361; App. Ex. 1.)
262.
To become certified as a Level III, Puckett had to pass a practical exam to assure that he could identify rejectable welds.
On July 5, 1984 he failed his first practi-cal exam.
(DeWald, Prep. Test. at 42-43 ff. Tr. 1700; Puckett, Tr. 6681-6682; Vannier, Prep. Test. at 2-3 ff. Tr. 3151.)
L Puckett did pass a practical exam on July 19, 1984, but the exam itself was invalidated because none of the items inspected were rejectable, a requirement.for a valid practical under Comstock Procedure 4.1.3.
(DeWald, Prep. Test. at 44 ff. Tr.
1700; DeWald, Tr. 1673; Puckett, Tr. 6428.)
263.
Puckett testified that he took and passed an adoitional practical exam which was graded by inspector Joe Hii and then lost.
(Tr. 6442-46.)
Hii testified that he never graded any of Puckett's exams, written or practical, and does not recall any test being lost or disappearing.
(Hii, Prep.
l Test. at 7 ff. Tr. 16608; Tr. 16640.)
264.
In any event, Puckett took another practical exam on August 22, 1984 using certain test coupons.
(DeWald, Prep.
Test. at 44 ff. Tr. 1700; Puckett, Tr. 6447.)
DeWald graded this exam based on the grading key developed by Vogt for the test coupons.
Comstock Procedure 4.1.3 requires a 100% score for practical exams.
Based on the grading key, DeWald determined that Puckett had failed to identify several re-jectable conditions and failed him.
(DeWald, Prep. Test. at 44 ff. Tr. 1700; App. Ex. 5; Dominique, Prep. Test. at 1 ff. Tr.
16512.)
265.
In December 1985, Joseph R. Vannier, the Level III Welding Inspector Arbitrator at Braidwood, reviewed Puckett's August 22 exam and the test coupons used in it.
(Vannier, Prep. Test. at 2-3 ff. Tr. 3151; Dominique, Prep.
Test. at 4 ff. Tr. 16512.)
Vannier concluded that the practical exam was a good exam for a Level III inspector.
L Additionally, he found that Puckett had made six incorrect determinations.
Vannier would have failed Puckett under the Comstock procedure for any one of these mistakes.
(Vannier, Tr. 3249-3253, 3261-3262, 3266, 3269; Vannier, Prep. Test at 9 ff. Tr. 3151.)
266.
At the hearings, Puckett admitted that on this practical he "did miss a couple of calls" (Tr. 6473), er.d he stated that he had no excuse for that fact.
(Tr. 6475.)
267.
During his employment as Comstock's Level III weld inspector, Puckett documented four basic concerns which he had identified with regard to the welding program:
(1) con-cerns about weld rod control (FF-268 to 272, infra); (2) con-cerns about the welding of stainless steel (FF-273 to 275, infra); (3) concerns about the welding of sheet metal (FF-276 to 290, infra); and (4) unspecified inconsistencies in welding procedures in general (FF-291 to 295, infra).
As described below, (FF-308 to 318, infra.)
Puckett raised additional con-cerns with the NRC Staff regarding various quality issues, including welder qualifications, and we heard extensive testimony regarding them.
Although he did not document these concerns while he was at Braidwood, he testified that he mentioned most of them to DeWald orally.
(Puckett, Tr. 5566, 5577, 5660, 6201, 6212, 6223.)
268.
On July 6, 1984, DeWald assigned Puckett to review weld rod withdrawal slips for the period between May and July 1982.
He was to determine whether heat numbers were traceable to material receipt requests and to certification of L
filler metal.
(DeWald, Prep. Test, at 45-46 ff. Tr. 1700; Puckett, Tr. 5594, 5607; App. Ex. 64).
He was also to identify any locations where incorrect weld rod had been used.
(DeWald, Tr. 1720; Puckett, Tr. 5626-30).
269.
Puckett documented his review in a memo on August 15, 1984.
He found that the heat numbers were satisfac-tory, the welders were qualified, and the rod used met or exceeded requirements.
He found, however, that the tool room attendant violated Comstock procedures and that an NCR should be issued.
Nonetheless, he did not cause an NCR to be initia-ted nor did he take steps to revise the relevant procedure to prevent repetition.
(DeWald, Prep. Test. at 46 ff. Tr. 1700; DeWald, Tr. 1721-1723; Puckett, Tr. 5632-5633; App. Ex. 65.)
DeWald regarded this as Puckett's responsibility.
(DeWald, Prep. Test. at 46 ff. Tr. 1700.)
270.
NCR 3275 was issued to document this problem on September 12, 1984, after it was discovered that Puckett had-failed to issue one.
Subsequently, Tony Simile, Puckett's replacement as Comstock's welding Level III, revised the weld rod control procedure.
(DeWald, Prep. Test, at 46, 51 ff.
Tr. 1700.)
He introduced a traveler system for welding work and required inclusion of a weld installation record in each traveler package.
(Simile, Tr. 3376-77.)
This improved documentation reduced the chance of an incorrect rod being issued. (Simile, Tr. 3381.)
This was done without reference to Puckett's concerns.
(Simile, Tr. 3541.)
L
, 271.
Puckett had also examined weld rod withdrawal slips from different dates, and on August 17 issued a memo stating that he had been unable to locate the Certified Material Test Reports (CMTRs) for three heats of weld rods.
(App. Ex. 66.)
Puckett testified that his inability to locate the CMTRs was not a nonconforming condition, because his search was not complete.
(Tr. 5653.)
Puckett accepted that Applicant's Exhibit 67 contained CMTRs which satisfied his search for documentation for the three heat numbers.
(Tr. 5647-5650.)
272.
Puckett testified that the Comstock program for controlling weld rods was inadequate, in particular because it did not require a QC inspector to oversee the clerk's issuance of the rods.
(Tr. 5620.)
Puckett asserted that he mentioned these concerns to management orally, though he did not document them.
(Tr. 5614.)
Two of the concerns were addressed by a procedure revision initiated without his participation while he was still at Comstock.
(App. Ex. 21 at 4; Puckett, Tr. 5616-17.)
Puckett did not propose QC oversight of rod issuance as a corrective action to the procedural violation by the rod issue clerk noted in his August 15, 1984 memo.
(Puckett, Tr. 5632-33; App. Ex. 65.)
Schapker testified that Puckett never mentioned these weld rod centrol concerns to him.
(Tr. 11194-96.)
Schapker stated that there was no regulatory requirement for QC oversight of issuance.
(Schapker, Tr.
11197-98.)
He found that Comstock's weld rod control procedure was adequate both before and after the improvement introduced b
by Simile, and that the corrective action for NCR 3275 was appropriate.
(Schapker Prep. Test, at 16 ff. Tr. 10954; App.
Ex. 51 at 7.)
273.
On August 10, 1984, Puckett recommended a stop-work on all stainless steel welding.
(Simile, Prep. Test.,
Group Ex. 1 at 14 ff. Tr. 3305.)
He had determined that Comstock weld procedure 4.3.14 had not been qualified in the 2G weld position.
DeWald authorized the stop-work after pointing out to Puckett that it was his responsibility to find solutions to such concerns and on August 17, 1984 the stop-work was issued.
(Simile, Prep. Test., Group Ex. 1 at 6, 14 ff.
Tr. 3305; App. Ex. 54.)
On August 24, 1984, Puckett apparently had QC Inspector John Miner issue NCR 3145 to document this concern.
(DeWald, Prep. Test, at 48 ff. Tr. 1700; Gieseker, Tr. 2972.)
274.
After Simile came on site he assured that all unqualified welds were removed, that the procedure was properly qualified and that all rework was done according to the newly qualified procedure.
(Simile, Prep. Test, at 12-13 ff. Tr.
3305; Simile, Tr. 3362-63.)
Puckett agrees that this was a proper disposition.
(Puckett, Tr. 5541; App. Ex. 51 at 5.)
275.
NCR 3145 also raised the concern that Comstock had no procedure for welding stainless steel to carbon steel.
(App. Ex. 59.)
Puckett admitted that he never saw such a weld at Braidwood.
(Puckett, Tr. 5545-46.)
Schapker talked to Simile and about a dozen Level II QC weld inspectors, including Miner, and they all agreed that there were no such welds within u
Comstock's scope of work at Braidwood.
(Schapker, Prep. Test.
at 13-14 ff. Tr. 10954; Schapker, Tr. 10995, 11166-67.)
276.
On August 9, 1984, Puckett wrote a memo to DeWald recommending that all welding of A-36 steel to A-446 sheet steel using E-7018 electrode be stopped.
(App. Ex. 52.)
DeWald authorized the stoppage and on Auoust 17, 1984, work was stopped.
(DeWald, Tr. 1693; App. Ex. 54.)
On the same date, apparently at Puckett's request, Miner issued NCR 3099, docu-menting the underlying concern.
(Puckett, Tr. 5393-96; App.
Ex. 55.)
The NCR stated that a procedure had never been qualified for welding A-36 to A-446.
(App. Ex. 55.)
Puckett testified that NCR 3099 was intended to reflect the fact that the Comstock procedure did not list A-36 to A-446 as a qualified process.
(Tr. 5397.)
277.
James Gieseker is a CECO electrical Engineer with responsibility for working with Comstock's QC Department.
(Gieseker, Prep. Test, at 2 ff. Tr. 2771.)
Gieseker determined that, in fact, there was no procedure qualification which specifically listed A-36 to A-446.
However, he reviewed the AWS Code and concluded that under paragraph 5.5.1.1, a
Procedure Qualification Record ("PQR") which qualified the welding of A-500 steel to A-446 steel also qualified the welding of A-36 to A-446.
Attachments H and O to Comstock Procedure 4.3.3 had qualified the welding of A-500 to A-446.
However, the Weld Procedure Specification ("WPS") for those attachments did not list A-36 as a qualified metal as required by the Comstock procedure.
Therefore, all that was required to i
l correct.the problem was to revise the procedure attachments to include a reference to A-36 mat'erial.
(Gieseker, Prep. Test.
st 21-22 ff. Tr. 2771; Gieseker, Tr.'2934; Puckett, Tr.
5380-81, 5481; 5423-25.)
278.
Because Gieseker did not consider himself a code expert-he reviewed the matter with CECO QA and S&L.
Both agreed with his interpretation.
Gieseker arranged a meeting for August 22, 1984 to resolve the concern in NCR 3099 so that the stop-work could be lifted.
(Gieseker, Prep. Test, at 22 ff. Tr. 2771; Gieseker, Tr. 2886, 2927; Louden, Tr. 2985.)
279.
The participants at the August 22 meeting in-cluded Puckett, DeWald and Seltmann of Comstock and representa-
.tives of CECO and of S&L, including James Louden, a welding expert.
(Gieseker, Prep. Test, at 22 ff. Tr. 2771; Louden, Prep. Test, at 2 ff. Tr. 2984; Int. Ex. 29.)
At the meeting, the opinion was expressed by either Gieseker or Louden that because there was a PQR for welding A-500 to A-446, under AWS D1.1-1975 there was no need for a PQR for welding A-36 to A-446.
(Louden, Prep. Test. at 3-4 ff. Tr. 2984.)
Puckett agreed with this conclusion.
(Puckett, Tr. 5381, 5397-98.)
280.
All participants at the August 22 meeting, including Puckett, agreed that the appropriate corrective action was revising Attachments H and O to the welding proce-dure by adding A-36 to the list of material which could be welded to A-446.
(Louden, Prep. Test. at 7 ff. Tr. 2984; DeWald, Tr. 1717; Louden, Tr. 2987; Int. Ex. 29.) Attachment H qualified welds from 3/8" to 11/16" in size; Attachment O t
qualified welds from 1/8" to 3/8".
(Puckett, Tr. 5377-78; Schapker, Tr. 10973.)
The NRC Staff subsequently concurred in this conclusion.
(Schapker, Prep. Test, at 8-9 ff. Tr. 10954.)
281.
Gieseker either suggested issuing a speed memo to authorize resumption of the welding pending the procedure revision or asked Puckett what action would satisfy him.
Puckett indicated that a apeed memo from Gieseker would satisfy him.
(Seltmann, Prep. Test. at 19 ff. Tr. 1960; Gieseker, Tr.
2902.)
282.
Apparently, however, Puckett then pointed out that there was another problem with Attachment O.
He testified that he explained Attachment O had been rejected by PTL for a discrepancy unconnected with the A-36 question and that al-though he had resubmitted a revised version on July 5, it had not yet been approved by PTL.
(Puckett, Tr. 5374-77, 6551.)
283.
Although no other testimony indicates that Puckett raised this concern about Attacnment O, the ultimate disposition of NCR 3099, made the following day, appears to have taken account of it.
The disposition did not mention ; it lifted the stop-work only on welds larger than 3/8", covered by Attachment H.
(App. Ex. 55.)
Puckett testi-fied that this disposition was proper.
(Tr. 5427-29.)
284.
At the same meeting, Puckett also expressed his belief that the procedure had been qualified to the wrong code.
He said that the procedure should be requalified using AWS D1.3 instead of AWS D1.1. and that it was wrong for Comstock to be using AWS D1.1-1975.
(Louden, Prep. Test, at 5 ff. Tr. 2984; Gieseker, Tr. 2864, 2886; Louden, Tr. 3001.)
Louden explained to him that the AWS D1.3 Code did not exist prior to 1978.
Therefore, the S&L specification which governed the Comstock welding procedure was based on AWS D1.1-1975, the only appropriate code in existence when the specification was developed.
Louden also explained that Comstock had the option to adopt subsequent codes, such as AWS D1.3, but.that it was acceptable to use the code in effect when the contract was issued.
(Louden, Prep. Test, at 5 ff. Tr. 2984; DeWald, Tr.
1821, 1823; Gieseker, Tr. 2864-66, 2886; Louden, Tr. 3001; Simile, Tr. 3453.)
Louden testified that Puckett appeared to accept this explanation.
(Tr. 3001.)
285.
Puckett testified that his true concern when he recommended that welding of A-36 to A-446 be stopped and issued NCR 3099 was that although Attachment H was qualified, many of the welds being made in the field were the smaller welds, those qualified by Attachment O, which he believed was unqualified at that time.
(Puckett, Tr. 5463-64.)
He also testified that when he raised the concern about AWS D1.3, he did not believe that it was mandatory to qualify the procedure under that Code, but only that since Comstock, in his opinion, had to requalify its procedures it would be better to use D1.3.
(Puckett, Tr.
5453, 5462.)
Numerous facts of record, summarized below, make these assertions difficult to credit.
286.
In his two memos recommending stop-work and in NCR 3099, Puckett did not limit the description of his concern to smaller welds, those covered by Attachment O.
(Puckett, Tr.
5391-5397;' App. Exa. 52, 53, 55.)
Additionally, those'who-talked with Puckett~about his concern did not understand it to be so limited.
The common understanding was that Puckett believed that it was improper to have qualified welding of A-446 under the AWS D1.1 code instead of AWS D1.3.
(Louden, Prep. Test. at 5 ff. Tr. 2984; Gieseker, Prep. Test. at 22 ff.
Tr. 2771; Gieseker, Tr. 2863, 2866-67; Simile, Prep. Test. at 8 ff. Tr. 3305; Schapker, Tr. 10962-67, 10972, 10979, 10982;.
11311, Weil, Prep. Test, at 7 ff. Tr. 11948.)
1 287.
Also, the record indicates that Attachment O was not in fact a rejectable procedure on the date of the meeting.
- On July.5, 1984, Puckett had resubmitted Attachment O for approval by PTL.
The revised Attachment 0 indicates that it was approved by PTL on July 6, 1984.
(Puckett, Tr. 5368-69, 6717; App. Ex.:77.)
Additionally, S&L had observed the same discrepancy in Attachment O prior to Puckett.
(Puckett, Tr.
5417-19; App. Ex. 10 at 3, note E.)
However, S&L had authorized use of the procedure while it was being revised.
(Schapker, Tr. 10977-78.)
Puckett agreed that S&L had such authority.
(Tr. 5418-19).
Although Puckett disclaimed knowledge of the S&L authorization, it appears plainly on page 3 of the procedure.
(Puckett, Tr. 5366; App. Ex. 10 at 3.)
288.
On August 22 after the meeting, DeWald received from Puckett a memo which recommended that all welding, includ-ing A-36 to A-446, be stopped because Comstock was " dangerously approaching a complete breakdown" in its QC program.
In the memo, Puckett stated that procedures involving A-446 "were
l qualified using the criteria of AWS D1.1-1975 and it should never have been done."
That code, he explained, was never intended for thin-gauge materials like A-446, and all proce-dures involving A-446, "should have been qualified using the criteria of D1.3."
(App. Ex. 56; DeWald, Prep. Test. at 49 ff.
Tr. 1700; DeWald, Tr. 1751-52.)
289.
This memo made it clear to DeWald that Puckett had not understood the discussion at the meeting that day.
In particular Puckett had not understood that AWS D1.3 did not exist when work began at Braidwood and that Comstock followed AWS D1.1 1975 as allowed.
(DeWald, Prep. Test. at 49-50 ff.
Tr. 1700.)
290.
Puckett also voiced his concern that the wrong code was being used to qualify the procedure the next day at the Procedure Review Board which was considering the disposi-tion of NCR 3099.
(Gieseker, Prep. Test. at 22 ff. Tr. 2771; Gieseker, Tr. 2863, 2866-67.)
As a result of this incident, Gieseker had an extremely low opinion of Puckett's usefulness as a Level III weld inspector, because Puckett exhibited a poor i
understanding of the AWS Code and erratic behavior once he had identified what he perceived to be a major problem.
(Gieseker, Prep. Test. at 23-24 ff. Tr. 2771, Gieseker, Tr. 2895-96, 2867.)
291.
Other than the concern regarding A-446 and the l
D1.3 Code, the only basis for stopping all welding that Puckett offered in the August 22, 1984 memo was that there were so many
(
_]
l inconsistencies in the other procedures that their qualifica-tion would be considered indeterminate.
(App. Ex. 56.)
292.
Puckett testified that in his opinion the proce-dures would have to be requalified in any event because of the inconsistencies referred to in his August 22 memorandum.
(Puckett, Tr. 5462-64.)
His August 22 memo was meant to convey to DeWald that since the procedures had to be requalified anyway, those dealing with sheet metal should be requalified to D1.3.
(Id.)
293.
Puckett never gave a list of these inconsisten-cies to Comstock.
(Puckett, Tr. 5503.)
He testified that he had conveyed them to DeWald orally.
(Id.)
In the course of a lengthy cross-examination on this subject, Puckett was able'to recollect only two such inconsistencies.
He thought that the grossest inconsistency was that Comstock had qualified a broad range of amperages and voltages for each specially qualified welding procedure.
To do that, it was necessary to have multiple procedure qualification records (PQRs) for each.
(Puckett, Tr. 5472-83.)
294.
We do not understand the nature of Puckett's concern here, since the whole range of variables was in fact qualified by the PQRs.
Schapker testified that having such multiple PQRs to validate a procedure is a common practice'and did not call for corrective action.
(Tr. 10983-88.)
We believe that Puckett was mistaken in this matter.
295.
The other inconsistency Puckett recalled was that the inspection procedure for stainless steel welds stated that in addition to visual inspections, various types of nondestruc-tive examination could be employed, including magnetic particle examination.
(Puckett, Tr. 5493, 5504-05; App. Ex. 57.)
In fact, of course, magnetic particles would not adhere to stain-less steel welds, making this technique useless for them.
It is obvious to us that this " inconsistency" was trovial, and Puckett agreed it was not a safety concern.
(Tr. 5507.)
296.
Puckett admitted that he did not expect DeWald to stop work based on his August 22, 1984 memo.
(Tr. 6250-51.)
Indeed, he conceded that the drastic recommendation to stop all welding was "a little strong" and was designed to attract strong attention within the organization.
(Tr. 6273, 6276.)
Puckett wrote it to get DeWald's attention so that the two men could discuss a broad range of issues troubling Puckett, because Puckett believed DeWald was not paying sufficient attention to concerns he expressed orally.
This discussion never took place.
(Puckett, Tr. 6258-59, 6254, 6273.)
297.
Comstock policy provides that all new er ployees are to be evaluated for a period of three months.
Dur ng that period, employees may be terminated without advance no: ice if they prove unsuitable for the position.
(DeWald, Prey. Test.
at 19-20 ff. Tr. 1700.)
Additionally, under Comstock policy, unsatisfactory job performance is valid grounds for dismissal.
(DeWald, Prep. Test, at 20 ff. Tr. 1700.)
298.
During Puckett's tenure, DeWald became concerned with his performance, specifically with his incomplete analysis of problems and inability to resolve them and his tendency to focus on problems which had already been corrected.
(DeWald, Prep. Test. at 50 ff. Tr. 1700.) -DeWald and Seece discussed dismissing Puckett.
Seese believed that dismissing him wa's better than demoting him to a Level II inspector because such a demotion would have likely affected Puckett's performance.
(Seese, Prep. Test, at 20 ff. Tr. 2330; DeWald, Tr. 1791; Seese, Tr. 2495.)
299.
On August 27, 1984, DeWald met with Shamblin and Gieseker to discuss Puckett's performance and abilities.
They discussed the fact that Puckett was not effective at problem definition or problem resolution, that his understanding of the AWS Code was inadequate, and that he had so far failed to become a Level III inspector.
(Gieseker, Prep. Test, at 23-24 ff. Tr. 2771; Shamblin, Prep. Test. at 32-33 ff.
4 Tr. 16274.)
DeWald decided to terminate Puckett's employment with Comstock pursuant to the 90-day probationary period in his contract because he was incompetent.
(DeWald, Prep. Test. at 50-51 ff. Tr. 1700; Gieseker, Prep. Test. at 23-24 ff. Tr.
2771; DeWald, Tr. 1313.)
Shamblin concurred in this decision.
(Shamblin Prep. Test. at 32 ff. Tr. 16274.)
300.
DeWald met with Puckett and informed him that he was being fired under the 90-day evaluation period.
(DeWald, Tr. 1736-43.)
301.
Anthony Simile, Comstock's Level III weld inspec-tor at the Perry nuclear plant, replaced Puckett as Comstock's Level III weld inspector at Braidwood on August 28, 1984 and has served in that capacity since that time.
(Simile, Prep.
. )
i Test. at 6-7 ff. Tr. 3305; Simile, Tr. 3340-42; DeWald, Prep.
Test. at 51 ff. Tr. 1700.)
302.
Simile was told to review the concerns raised by Puckett to determine whether changes were needed in Comstock's welding program at Braidwood and, if so, to implement them.
(Simile, Prep. Test. at 9 ff. Tr. 3305; Simile, Tr. 3358-59.)
He was instructed by DeWald to review the welding procedures and the welder qualification procedures and records to identify and resolve any problems.
(Simile, Prep. Test. at 10 ff. Tr.
3305.)
DeWald also gave him a package of documents which related to Puckett's concerns.
(Simile, Tr. 3339; Simile, Prep. Test, at 8, Group Ex. 1 ff. Tr. 3305.)
303.
Simile concluded that the welding procedures at Braidwood were adequate, but more cumbersome than necessary.
He revised them by deleting unnecessary material and simplify-ing the presentation of necessary material.
(Simile, Prep.
Test. at 11 ff. Tr. 3305.)
304.
Because Puckett had failed to identify specific problems with welder qualifications, Simile reviewed all welder qualification records for technical compliance with the AWS Code.
Some minor discrepancies were identified on NCR 3710 and resolved.
None of the discrepancies called into question the competency of the welders.
Therefore, all work was allowed to remain.
(Simile, Prep. Test. at 13 ff. Tr. 3305; Simile, Tr.
3364-67; App. Ex. 13.)
305.
NCRs 4649-4656 were initiated when certain discrepancies in the welder qualification records were brought l
i I
to Simile's attention duirng Schapker'c investigation of Puckett's concerns.
(Simile, Tr. 3504, App. Ex. 13.)
NCRs 5014-18 were initiated as a result of Comstock's systematic review of its welder qualification records.
(Simile, Tr. 3515-16, App. Ex. 13.)
NCR 5044 identified the use of an incorrect form to record a welder's qualification.
(Simile, Tr. 3520; App. Ex. 13.)
All of these NCRs identified minor nonconformances in the welder qualification records and none necessitated review of any work in the field.
(Simile, Tr. 3520-3521.)
306.
Based on his review of Comstock procedures Simile believes that on the whole the Comstock welding program was adequate and that Puckett's suggestion to stop work to avoid a
" complete breakdown" of the QC Program was not warranted.
(Simile, Prep. Test. at 17-18 ff. Tr. 3305.)
307.
Simile has always had the freedom to identify, investigate and report deficiencies in Comstock welding programs as a Level III weld inspector at Perry and Braidwood.
(Simile, Prep. Test. at 18-19 ff. Tr. 3305.)
308.
On August 28, 1984, Puckett, in a telephone interview, informed the NRC Senior Resident Inspector at Braidwood that he had been terminated for raising technical concerns.
On August 31, 1984, Puckett was contacted by an NRC staff member from Region III, and he provided further informa-tion regarding his concerns.
On September 11, 1984, Mr. Weil, the Region III allegation specialist, and Mr. Ward, an NRC Reactor Inspector, met with Puckett regarding his technical concerns and alleged wrongful' dismissal by Comstock.
(Schapker, Prep. Test. at 3-4 ff. Tr. 10954.)
309.
On February 25, 1985, NRC Inspector Jerome Schapker was assigned to investigate Puckett's technical concerns.
(Schapker, Prep. Test. at 3.-4 ff. Tr. 10954.)
On March 12, 1985, Schapker reviewed with Puckett a list of Puckett's allegations as described by Weil and Ward.
Puckett agreed that the allegations listed reflected his concerns and that he could think of no additional concerns.
(Schapker, Prep. Test. at 5-6 ff. Tr. 10954.)
310.
Schapker investigated all of Puckett's technical concerns, including reviewing procedures and QC documentation and inspecting installations in the field.
(Schapker, Prep.
Test, at 4-7, 13 ff. Tr. 10954.)
Schapker's conclusions with respect to several of Puckett's concerns are addressed above.
(FF-supra.)
His conclusions with respect to the infra.)
remaining concerns are addressed below.
(FF-311.
Puckett testified he was concerned about Comstock's practice of transferring identifying marks on steel after cutting the steel into pieces rather than before.
(Tr.
5657, 5673-75.)
Puckett could not say that this was a noncon-formance and admitted he knew of no instance when the wrong numbers had been transferred.
(Tr. 5674.)
Schapker's review l
of Comstock's practice indicated that they transferred identi-fying marks prior to cutting and he stated that in any case the practice to which Puckett testified was not a violation of l
regulatory requirements.
(Schapker, Prep. Test. at 19-20 ff.
Tr. 10954.)
312.
Puckett was concerned that a General Electric aluminum welding procedure used by Comstock was not qualified in all positions.
(Puckett, Tr. 5576-77.)
Puckett told LeWald his concern and was told to qualify the procedure and the welders in the remaining position before using it.
(Puckett, Tr. 5575-83.)
Schapker found that the procedure related to non-safety related work and was not even governed by the AWS D1.1 Code.
(Schapker, Prep. Test. at 44 ff. Tr. 10954.)
313.
Puckett raised a number of concerns about welder qualification records.
(Tr. 6137-38, 6149-52, 6155, 6159-63.)
All of these concerns related to inconsistencies in the records identified by Puckett during an admittedly partial review.
(Puckett, Tr. 5679; App. Ex. 68.)
Puckett recognized that the inconsistencies did not mean that the welders were unqualified.
He agreed that to determine whether the welders were qualified it would be necessary to check for objective evidence which could clarify the inconsistencies in the documents.
(Puckett, Tr. 5699, 6136-37.)
l 314.
Schapker determined that none of these dis-crepancies violated AWS Code requirements and that consequently l
none of the qualification records Puckett reviewed were inde-terminate or invalid.
(Schapker, Prep. Test. at 22-36 ff. Tr.
10954; Schapker, Suppl. Prep. Test. at 3 ff. Tr. 10960; Schapker, Tr. 11063, 11276; App. Ex. 51.)
Consequently, no l
- question arose as to the adequacy of the welds performed by the welders.
(Schapker, Tr. 11063.)
315.
Puckett also testified about the " hammer log" that records the identifying symbol given to each welder for stamping welds he completes.
(Tr. 6179-80.)
He was concerned that when a welder quit no one appeared to be responsible for taking possession of his stamp and destroying it, so that stamps were reissued to other welders.
(Puckett, Tr. 6181-82, 6185.)
Inspector Wicks, however, testified that he was respon-sible for assuring that stamps were returned and destroyed when a welder quit.
(Tr. 7232-34, 7247-50.)
Moreover, Schapker testified that in any case there is nothing unacceptable about reissuing such a stamp, and Puckett admitted there was no Code requirement that each welder have a unique symbol.
(Schapker, Prep. Test. at 38 ff. Tr. 10954; Puckett, Tr. 6180.)
Schapker concluded that the Comstock records were adequate.
(Schapker, Prep. Test. at 38 ff. Tr. 10954.)
\\
316.
Puckett also testified to a concern that Wicks, the QC inspector in charge of the welder test booth, who oversees welder qualification tests, was sometimes required to do field inspections.
(Puckett, Tr. 6188, 6193.)
- Again, l
Puckett admitted that there was no Code requirement that there l
be a QC inspector full time in the test booth.
(Puckett, Tr.
6187.)
Schapker talked with Wicks, who knew of no instance where a test was performed without the presence of a QC inspector.
(Schapker, Prep. Test. at 40 ff. Tr. 10954; Schapker, Tr. 11054.)
. 317.
Puckett also testified to a concern that the Comstock welding procedure did not require non-destructive examination ("NDE") for full penetration welds, in addition to visual inspection.
(Tr. 6207-08.)
He believed that the S&L specification required such NDE.
(Tr. 6208-12.)
On the stand, however, he was unable to identify such a requirement in the specification.
(Id.)
Schapker determined that at that time neither the AWS Code nor the specification required NDE for full penetration welds.
(Schapker, Prep. Test. at 42 ff.
Tr. 10954.)
318.
Puckett also testified that he once identified a weld which violated the requirements of the AWS Code because it was not broken where the plane of the weld changes.
(Tr.
6222-24.)
He stated that the S&L drawings did not require the weld to be broken at the corners.
(Tr. 6226-27.)
- Schapker, however, found that the drawings did have this requirement and that all of the welds he observed complied with the drawings.
(Schapker, Prep. Test, at 43-44 ff. Tr. 10954; App. Ex. 51 at 20.)
319.
After ccmpleting his inspection, Schapker con-cluded that Comstock management had addressed every one of Puckett's concerns and taken adequate corrective action where needed.
He would not characterize that action on the part of l
Comstock management as sweeping concerns under the rug.
l l
(Schapker, Tr. 11425.)
320.
Schapker found fault with Puckett not because he raised concerns, but because he did not research problems adequately to determine whether the problems were safety-significant and thus whether there was a need to stop work.
Schapker believes it was Puckett's responsibility-to do a more thorough job in investigating his concerns.
(Schapker, Tr.
11296.)
321.
Puckett admitted that he knows of no instance where the safety of Braidwood was compromised because of any problem he identified.
(Puckett, Tr. 6663.)
322.
Inspector Rick Martin testified that on March 28, 1985 he was at his work station in the outer corral area of Comstock's QC vault when Schulz, an NRC Resident Inspector, sat down next to him.
(Martin, Tr. 9214-22, 9457.)
Schulz asked Martin some questions about welding checklists because Martin was sitting next to him.
(Martin, Tr. 9218, 9458.)
323.
Seese told Martin that afternoon that he would be moved to the inner corral area the next morning because Seese was concerned about control over the documents with which Martin was working.
(Martin, Tr. 9219-20, 9458.)
Martin testified that he believed the move occurred because he had spoke'n with Schulz, and that Seese's rationale was " phony."
(Martin, Tr. 9219.)
l 324.
DeWald testified that Martin was moved into the inner corral area of the QC vault because Martin had a tendency to talk with everyone who passed through the outer corral area; the inner corral had no general traffic flow.
(DeWald, Tr.
1534-35.)
1 325.
When Martin raised with Seese his concern that his transfer resulted from having talked with the NRC, Seese reassured Martin that the reason for the move was to increase control over the documents Martin was using and that it had no relationship to his talk with Schulz.
Seese also assured him that his conversations with the NRC were private and that Comstock management had no way of knowing when Martin had talked to the NRC.
(Int. Ex. 82; Martin, Tr. 9220-21.)
Martin did not believe Seese's assurances.
(Martin, Tr. 9222.)
326.
Martin later admitted that Seese had a "very good basis" for explaining the need to maintain greater document control because documents had been lost in the vault area in the past.
(Martin, Tr. 9458-59.)
Martin also admitted that he had found traffic flow in the outer corral area to be dis-tracting, that conversations would occasionally interrupt his train of thought, and that he would participate in those conversations.
(Martin, Tr. 9459-60.)
327.
We find no basis for concluding that moving l
l Martin's desk was an act of discrimination.
There is simply no l
reason to believe that the move was motivated by a desire to keep Martin from talking to NRC inspectors, rather than by the l
legitimate reasons discussed above.
328.
Inspectors Perryman, Bossong and three others were assigned in March, 1985, to work on a hanger walkdown program.
(Simile, Tr. 3423-25; Perryman, Tr. 9720; App. Ex.
l 106.)
The walkdown program was part of a corrective action involving Comstock, Edison and Sargent & Lundy (S&L), that was i
l t
.~.
...being conducted to disposition Edison NCRs 708 and 709.
(Simile, Prep. Test. at 25, Attch. 3 ff. Tr. 3305; Perryman, Tr. 9805-06; Simile, Tr. 3415-17.)
329.
The task of the Comstock QC inspectors was to compare the as-built configuration of hangers, which were fabricated by Systmas Controls Corporation (an off-site ven-dor), with the design drawings used by the fabricator.
(Simile, Prep. Test. at 20 ff. Tr. 3305; Simile, Tr. 3416-17.)
The QC inspectors, assisted by an S&L engineer, were to mark in red pencil any differences between the as-built hanger con-figurations and the design on the design drawings.
(Simile, Tr. 3417-18, 3421-22.)
These redlined drawings were called "Rev. O" drawings.
(Bossong, Tr. 9864-65; Perryman, Tr. 9680.)
330.
As the next step in the corrective action, S&L engineers evaluated the design significance of any deviations between the as-built condition of the hanger and the design.
(Simile, Tr. 3418; Simile, Prep. Test. at 20 ff. Tr. 3305.)
Acceptable hanger configurations were depicted on a different set of drawings called "Rev. A" drawings.
(Bossong, Tr. 9865; Perryman, Tr. 9680.)
Comstock QC inspectors would then rein-spect the hangers to determine whether the as-built configura-tion of the installed hanger corresponded with that shown on
'the Rev. A. drawing.
The inspectors used the same checklists that, as explained in the next finding, accompanied the Rev. O drawings to indicate acceptance or rejection of the hangers.
(Simile, Tr. 3418, 3421-22; Simile, Prep. Test, at 20 ff. Tr.
3305.)
_ _. 331.
The Rev. O drawings were accompanied by a hanger configuration checklist (Form 7) and the QC inspectors were expected to sign the checklist to document and verify the accuracy of the as-built hanger (s) as redlined in the drawings.
(Bossong, Tr. 9865; Perryman, Tr. 9674.)
Although Perryman and Bossong and other inspectors understood this purpose (Perryman, Tr. 9675-76; 9679; Bossong, Tr. 9866, 9868), they voiced concern that their signatures would be interpreted by others that they were accepting the as-built condition of the hangers before the engineering evaluation had been completed.
(Perryman, Tr. 9676-77; Bossong, Tr. 9866-69; Tr. 9690; Simile, Tr. 3424-25.)
332.
The checklist used for the walkdown program was clarified on May 21, 1985, when a note was added to the check-list to make clear that the QC inspectors' signature verified only the accuracy of the configuration of the installed hanger with the redlined drawing, Rev.
O.
(App. Ex. 106; Perryman, Tr.
9763-64.)
This charge appeared to meet the inspectors' concerns.
(Bossong, Tr. 9844; Perryma-
- r. 9763-65.)
333.
In early May 1985, Perryman, Bossong and the other inspectors requested transfers from the walkdown program to the in-process inspection duties they had previously been assigned.
(App. Ex. 105; Int. Ex. 35; Perryman, Tr. 9720-22, 9682-86, 9759-62, 9678; Bossong, Tr. 9869, 9875; Simile, Tr.
3428-32.)
The requests for transfer were primarily motivated by dissatisfaction with Comstock QC management's slow response to remedy their concerns about possible misinterpretation of f
l the meaning-of their signatures on checklists.
(Perryman, Tr.
9687-82, 9720-22.)
334.
The requests for transfer submitted by Perryman and Bossong were initially denied (Perryman, Tr. 9684-85, 9762; Int. Ex. 35; Bossong, Tr. 9873-75); however, the requests were subsequently reiterated.
(Perryman, Tr. 9685, 9762; Bossong, Tr. 9880; Int. Ex. 35.)
Perryman again demanded his transfer during a training session on an additional, May 29, 1985, clarification to the walkdown procedure which was held on May 30, 1985.
(App. Ex. 107; App. Ex. 108; Bossong, Tr.
9878-80; Perryman, Tr. 9767-69; Simile, Tr. 3547-48)
Perryman stated further that if he was not transferred the quality of his inspection work would be affected.
(Perryman, Tr. 9767, 9769; Simile, Tr. 3547.)
335.
Perryman, Bossong, and other inspectors refused for a time to turn in their completed Rev. A checklists until the procedure was changed or clarified.
(Perryman, Tr.
9679-80, 9689; Bossong, Tr. 9871-72.)
Perryman testified that he intended to throw his held back Form 7s in the garbage unless Comstock changed its procedure.
(Perryman, Tr. 9689.)
Simile's irritation with this practice of holding back check-lists apparently helped prompt the May 30, 1985 training session.
(Bossong, Tr. 9872.)
336.
Moreover, Perryman was involved as a union representative in the labor negotiations between the union and management.
These meetings were held during the day.
(Perryman, Tr. 9780.)
... 337.
On May 31, 1985, Comstock decided to grant the transfer requests.
(Int. Ex. 35.)
Around June 1, 1985, Simile advised Perryman and Bossong and others that the transfers would be granted.
(Perryman, Tr. 9691, 9722-24, 9687; Bossong, Tr. 9877-79, 9872.)
Simile told them they would not.like the reassignments.
( Pe rrym.- n, Tr. 9691-92, 9723-24, 9778; Bossong, Tr. 9870; Simile, Tr. 3434-35.)
One inspector declined and his request was honored.
When he learned that this transfer was to second shift, Perryman also declined; however, his declination was not granted.
(Perryman, Tr. 9723-25.)
338.
Perryman and Bossong were transferred to second shift because that shift was just being staffed and good inspectors were needed.
(Perryman, Tr. 9781; Bossong, Tr. 9883; Simile, Tr. 3448-49.)
339.
Bossong had initially expressed an opinion that these transfers were made in retaliation for raising concerns about the walkdown procedures.
(Perryman, Tr. 9730; Bossong, Tr. 9882.)
However, based on his subsequent managerial experi-ence as a union steward, Bossong expressed doubt as to whether or not the transfer was retaliatory.
(Bossong, Tr. 9882-83.)
Had Bossong been in Comstock management's position, he would l-have basically made the same personnel moves.
(Id.)
l 340.
R. D. Hunter was employed by Comstock and later BESTCo, as a Level II QC weld inspector from October 1983 until he was terminated on March 25, 1986. (Simile Reb. Prep. Test.
[
at 2 ff. Tr. 16036.)
Hunter was deposed in connection with this proceeding shortly before his termination.
Hunter then wrote to Applicant's Quality First program and asked if his termination was "not a way to get back at people who had given a deposition [?)."
(Int. Ex. 70.)
We admitted the issue of l
Hunter's alleged retaliatory firing as a matter in controversy.
(Tr. 3769.)
341.
Hunter testified that in his letter he was not asserting that he was discharged for talking to the NRC or for giving a deposition.
He was merely asking Quality First to investigate this question.
He did not personally know whether it was true.
(Tr. 9032-36, 9079, 9083-84).
He also testified that he was informed by Quality First that their investigation had not revealed that he was discharged for that reason.
He accepted this judgment.
(Tr. 9036-37.)
342.
On the second week of March 1986, Hunter add QC inspector Ron Arndt were performing reinspections of welds on certain hangers and braces.
(Hunter, Tr. 8585, 8598.)
For each hanger he inspected, Hunter signed a weld inspection checklist (" Form 19") verifying that he inspected and accepted each weld thereon as in compliance with each of eight acceptance criteria.
(Simile, Reb. Prep. Test. at 15 ff. Tr.
l 16036; see, e.g.,
Frazier, Reb. Prep. Test., Attch 2.
ff.
Tr. 15930.)
343.
The first criterion on Form 19 is that the weld I
be "sufficiently cleaned as not to impair visual inspection, per para.
3.2."
(App. Ex. 100.)
This criterion refers to Comstock Procedure 4.8.3, 1 3.2, which states in relevant part, the " completed weld surface shall be free of paint, slag, t
scale, flux, dirt, oil and foreign material that would impair the performance of the visual inspection."
(Simile, (Hunter)
Reb. Prep. Test, at 2-3, 14 ff. Tr. 16036; Hunter, Tr. 8592-95.)
344.
If a weld is not clean enough for a visual inspection, inspectors are sut.r.osed either to clean it with a wire brush or to mark it for a crew to clean with a power buffer.
(Hunter, Tr. 8796-8800, 8959.)
345.
If an inspector _ accepts a weld, he marks it as approved on the Form 19 and hammers his unique inspector identification stamp near the weld.
(Hunter, Tr. 8696.)
If the item is too thin for a hammer stamp, the inspector draws his stamp in black marker near th* weld.
(Hunter, Tr. 8696.)
346.
Among the hangers Hunter inspected on the week of March 10, 1986 were 12H35, 2HV14 and 12Hl.
He indicated on the Form 19 checklists for the three hangers that all welds were clean enough for inspection and were acceptable.
Hunter signed the checklists and submitted them to his supervisor.
(Frazier Reb. Prep. Test, at 2-3, 5,
Attch. 2 ff. Tr. 15930; Simile Reb.
Prep. Test. at 8 ff. 16036; Hunter, Tr. 8543, 8598; Int.
Ex. 67.)
347.
On March 18, 1986. two PTL QC inspectors informed Tony Frazier, a PTL Level III weld inspector and supervisor, that an over inspection revealed that welds on those hangers had been inspected through paint by Comstock QC inspectors.
(Frazier (Hunter) Reb. Prep. Test. at 2 ff. Tr. 15930; Frazier (PTL) Reb. Prep. Test.; Attch. 2c (Frazier-1) ff. Tr. 15568.)
- 100 -
Frazier accompanied the inspectors to the field to examine the hangers.
(Id. at 2-3.)
He concluded that welds on the hangers had been' inspected through rust and paint.
(Id. at 3-4.)
348.
Frazier concluded that the painted and rusty welds he saw had been inspected in that condition because on hanger 12H35 Hunter's inspector stamp went through the paint down to the bare base metal, and the paint itself was one continuous coat covering the area of the stamp and the weld.
(Frazier (Hunter) Reb. Prep. Test. at 3 ff. Tr. 15930.)
On hanger 2HV14 welds were so rusty they could not have been inspected properly.
Frazier did not believe that the amount of rust he observed could have formed in the eight days that had elapsed since Hunter's inspection.
(Frazier (Hunter) Reb.
Prep. Test, at 4 ff. Tr. 15930.)
349.
Frazier met with Simile that day to report PTL's findings.
At Simile's request they went to the field to review the hangers.
Simile noted that some welds on one target were covered with rust, and welds on two others were painted.
(Simile (Hunter) Reb. Prep. Test at 3-5 ff. Tr. 16336; Frazier (Hunter) Reb. Prep. Test. at 4-5 ff. Tr, 15930.)
Some of the painted welds on hangers 12H35 and 12H1 had Hunter's stamp next to them through the paint down to the bare metal.
Simile found no indication that the paint could have been touched after Hunter's inspection.
(Simile (Hunter) Reb. Prep. Test, at 5 ff. Tr. 16036.)
Simile concluded that these welds had been i
inspected in an unacceptable condition.
(Id.)
- 101 -
350.
Simile later took 3 other welding inspectors to the field to obtain their opinions of the hangers in question.
Simile testified that all three of them confirmed his belief that the welds had been inspected improperly.
(Simile (Hunter)
Reb. Prep. Test, at 6 ff. Tr. 16036.)
351.
Simile spoke with Hunter on March 18.
Hunter denied inspecting through paint but could not explain how PTL had found his hangers in an unacceptable condition.
Simile then suspended Hunter and Arndt.
(Simile (Hunter) Reb. Prep.
Test. at 6 ff. Tr. 16036.)
352.
Within a day or so after March 18. Simile also took Seltmann to the field for his opinion.
(Simile (Hunter)
Reb. Prep. Test. at 7 ff. Tr. 16036.)
Based on his review of hanger 12H35 Seltmann concluded that Hunter had inspected welds through paint.
Hunter's hammer stamp penetrated paint near welds that were painted over, indicating that the stamps were made after the welds were painted.
Seltmann found no sign of
" touch-up" paint.
(Seltmann. Reb. Prep. Test. at 2-3 ff.
Tr. 16122.)
353.
Bowman testified that in March 1986. Simile requested him to examine 14 hangers inspected by either Hunter or Arndt to determine whether they had inspected welds that were painted or unacceptably rusty and dirty.
(Bowman Reb.
Prep. Test. at 2 ff. Tr. 16000.)
Bowman concluded that Hunter had improperly inspected welds through paint on three hangers, including 32H35 and 12H1.
(Id.)
Bowman had no doubt that the welds had been covered with paint when Hunter inspected them.
- 102 -
(Id. at 3.)
In all three cases Hunter's stamp had been struck through the paint to the bare metal adjacent to the weld.
The paint through which the stamp had been struck was the same.as the paint coveting the adjacent welds.
(Id.)
354.
On March 21, after receiving Bowman's results.
Simile accompanied Hunter and Frazier to the field and showed Hunter two hangers on which he had concluded that welds were inspected through paint.
Simile testified that Hunter admitted he and Arndt had performed inspections through paint.
Hunter said however, that this practice was mainly limited to bracing welds, which had probably been inspected three or four times previously.
(Simile Reb. Prep. Test. at 10, 12-13 ff.
Tr. 16036; Simile, Tr. 16087-88.) When Hunter testified, he denied having made this admission.
(Hunter. Tr. 8975-76.)
355.
On March 24, 1986, Seltmann met with DeWald, Simile, Hunter and Arndt.
Seltmann testified that Hunter admitted inspecting welds through paint, but denied that he had inspected attachment welds through paint.
While Hunter admitted that he knew the procedures governing weld clean-liness, he could not offer any explanation for why he had inspected welds improperly.
(Seltmann Reb. Prep. Test, at 4 ff. Tr. 16,122.)
i l
l 356.
The next day Seltmann met with DeWald, Simile, 4
Michael Dougherty (CECO Project Construction Department engineer) and Thomas Skidmore (BESTCo Site Manager) so that Comstock could inform Edison and BESTCo of the problem.
Seltmann explained at that meeting that Comstock could not i
i
- 103 -
permit either Hunter or Arndt to remain in the QC program because they had falsely represented on inspection checklists that welds were acceptably clean.
(Id. at 5.)
357.
Dougherty testified that he also attended another meeting later that day with Skidmore, Hunter, Arndt, and Bossong, Hunter's union shop steward.
After the meeting all five went to the field and spent several hours looking at the hangers.
(Dougherty Reb. Prep. Test. at 3-4 ff. Tr. 16162.)
Hunter and Arndt could offer no explanation why their hammer stamps appeared through paint.
(Id. at 4-5.)
Thereafter, BESTCo terminated Hunter and Arndt.
(Dougherty, Tr. 16164; Simile, Reb. Prep. Test. at 17 ff. Tr. 16036.)
358.
On the stand, Hunter offered several different, and sometimes contradictory, explanations.
He attempted to distinguish among attachment welds (welds made in the field which attach a hanger to Comstock's supporting structure), fab shop welds (welds made in the fabrication shop), and vendor welds (welds made by the manufacturer prior to shipping).
Hunter explained that the latter two were not his responsi-bility, because they should have been inspected by other QC inspectors.
Therefore, he did not inspect them, although his l
checklists indicate that he inspected all welds.
(Hunter.
Tr. 8941.)
359.
Simile testified that when a new inspection is performed on a weld, it supersedes all previous inspections and becomes the inspection of record.
(Simile Reb. Prep. Test. at 14 ff. Tr. 16036.)
Thus, Hunter's Form 19 became the new
104 -
inspection record and his reliance on prior inspections was improper.
(Id.)
The procedure requires that all welds be clean before they are inspected.
Moreover, Simile testified that some of the painted welds accepted by Hunter were, in fact, attachment welds.
(Id.)
360.
Hunter also explained that at one time he was instructed by DeWald to inspect through paint.
He said that DeWald had told him to ignore Galvanox, a grey paint found on some welds, when doing inspections.
He admitted, however, that DeWald had said that he must be able to clean off the Galvanox sufficiently with his wire brush to make a visual inspection.
(Hunter, Tr. 8810, 8813-16.)
361.
Hunter also opined that the welds he inspected were painted after he inspected them.
(Tr. 8981-82.)
Dougherty and Bossong specifically looked for evidence to support this possible explanation and found none.
(Dougherty Reb. Prep. Test. at 5 ff. Tr. 16162.)
Similarly, Frazier, Bowman, Simile and Seltmann all testified that for the welds they considered to have been inspected in a painted condition, there were no signs of " touch up" or later painting.
(Frazier l
Reb. Prep. Test, at 3 ff. Tr. 15930; Bowman Reb. Prep. Test. at t
l 3 ff Tr. 16000; Simile Reb. Prep. Test. at 5 ff. Tr. 16036; Seltmann Reb. Prep. Test. at 3 ff. Tr. 16122.)
362.
The final explanation offered by Hunter was that, while there may have been "some" paint or rust on the welds on hanger 12H35, for example, it was "a judgment call" as to whether it was clean enough to inspect.
(Hunter. Tr. 8692-93.)
l l
l
- 105 -
I A.large number of qualified people, however, looked at the welds on hanger 12H35 and, without exception, concluded that the welds were unacceptably obscured by either paint or rust.
(Frazier Reb. Prep. Test. at 3-5 ff. 15930; Bowman Reb. Prep.
Test, at 3 Eff. 16000; Simile Reb. Prep. Test. at 5 ff. 16036;
'Seltmann Reb. Prep. Test. at 3 ff. 16122;.Dougherty Reb. Prep.
. Test, at 3-5 ff. 16162.)
363.
We find no evidence of any retaliatory motive underlying Hunter's termination.
None of the individuals who 4-testified apart from Hunter, was even aware of the contents of Hunter's deposition.
Seltmann, who decided to exclude Hunter from performing further QC inspection work for Comstock, knew i,
only that Hunter had been subpoenaed.
(Seltmann Reb. Prep.
Test. at 8 ff. Tr. 16122.)
Simile did not know what Hunter said at his deposition.
(Simile Reb. Prep. Test. at 17 ff.
Tr. 16036.)
Neither Frazier nor Dougherty knew at the time of Hunter's termination that he had been deposed.
(Frazier Reb.
Prep. Test. at 6 ff. Tr. 15930; Dougherty Reb. Prep. Test. at 7 ff. Tr. 16162.)
Additionally, Arndt, who was not deposed, was terminated for the same improper conduct as Hunter.
(Simile, Reb. Prep. Test. at 17 ff. Tr. 16036.)
4 i
I 364.
Gregory Archambeault began working as a QC inspector at Braidwood on January 6, 1986.
(Archambeault, Tr. 12141.)
He was hired by BESTCo and assigned to work as an electrical inspector for Comstock.
(Archambeault, Tr. 12141-42.)
t
- 106 -
365.
Tony Simile, Comstock's General QC Supervisor, is-I responsible for ensuring that each of Comstock's three shifts is staffed with'enough QC inspectors to accommodate inspection 1
needs.
(Simile, Arch. Reb. Prep. Test. at 2 ff. Tr. 16180.)
Beginning in the latter part of 1985 the need for QC inspectors on second shift increased, largely because craft was transfer-ring all cable pulling work to that shift.
Simile asked representatives of BESTCo, GE-MCIS and Union Local 306 to tell new QC hires that they would be working second shift.
(Id. at 3.)
366.
Archambeault, as well as other inspectors' hired in January 1986, was trained in cable pulling on the second shift and remained on that shift to work in cable pulling after receiving his Level II certification in February.
(Id. at 4; Archambeault, Tr. 12228-29.)
367.
Larry Bossong, the first shift steward for Union Local 306, is responsible for signing on new Comstock QC hires as union members.
He met with Archambeault iu January 1986 to brief him about fringe benefits, hcurs of work and related matters.
(Bossong, Reb. Prep. Test. at 1-2.ff. Tr. 16252.)
Bossong noted on Archambeault's work referral slip that he might be assigned to the second shift by Comstock.
Bossong routinely did this for all new hires because Comstock was using several shifts. (Id. at 2-3.)
368.
Archambeault testified that Bossong represented to him that he would be able to work first shift.
(Tr. 12655.)
He said Bossong told him that after he was trained on second
107 -
shift in cable pulling he would be transferred to first shift to be trained in terminations.
(Int. Ex. 118; Tr. 12691.)
Bossong denied making any such promises.
He explained that he had no authcrity over such matters, which were within Comstock's control, and that he didn't make promises he couldn't keep.
(Bossong, Reb. Prep. Test. at 3 ff. Tr. 16252; Bossong, Tr. 16260-61, 16264-65.)
369.
On June 4, 1986, Archambeault contacted Charles Weil of the NRC Staff and related a number of quality concerns to him.
(Archambeault, Tr. 12310-20.)
He subsequently sent the Staff a letter memorializing these concerns.
(Archambeault, Tr. 12320-22; Int. Ex. 122.)
The specific issues that Archambeault raised are addressed below in findings 380 to 389.
370.
On June 2, 1986, Archambeault made a written request to be transferred to the first shift.
He sent it to his union shop steward, George Nemeth, which he understood to be the proper procedure.
He received no response.
(Archambeault, Tr. 12272-75; Int. Ex. 118.)
Simile, who was responsible for making the shift transfer decisions, did not receive this request for consideration.
(Simile, Arch. Reb.
Prep. Test. at 2, 8 ff. Tr. 16180.)
371.
In June 1986 Simile asked Harry Revels, the second shift supervisor, for a list of three inspectors who wished to transfer to first shift.
Revels gave him the names of Ken Willoughby, Ron Nelson and Archambeault.
(Simile, Arch.
l l
- 108 -
Reb. Prep. Test. at 4 ff. Tr. 16180.)
Simile transferred Willoughby and Nelson on June 18, effective June 23.
(Id.)
7 372.
Archambeault believed that he had been passed
~
over for transfer to first shift in retaliation for his having.
raised quality concerns with NRC.
(Tr. 12370.)
Simile did not recall whether he had heard of Archambeault's raising concerns by June 18.
(Simile, Arch. Reb. Prep. Test. at 6 ff.
Tr. 16180.)
He testified, however, that he and Revels had decided that the second shift cable pulling effort could only afford to lose two inspectors.
Simile decided to transfer Willoughby and Nelson, rather than Archambeault, because both of them had previously spoken with him about being transferred to first shift, and he had pr'omised them early consideration for such a transfer.
Simile had no knowledge about Archambeault's shift preferences other than the fact that he had received his name as one of three possible transferees.
(Id. at 5; Simile, Tr. 16215-17, 16219-20, 16223.)
373.
Later in the summer of 1986, Revels told Simile that Archambeault still wanted to be transferred to first shift.
Simile and Revels agreed that the second shift cable
]
pulling effort could not afford to lose any QC inspectors at l
that time.
(Simile, Arch. Reb. Prep. Test. at 6 ff.
4 Tr. 16180.)
374.
In the same conversation, simile told Revels that he felt he was in an awkward position.
He felt that if he transferred Archambeault, the newspapers might misconstrue the action, because of the ongoing harassment hearings, as an l
I
- 109 -
i attempt to remove him from-an area in which he had raised quality concerns.
On the other hand, he thought that not transferring Archambeault would be interpreted as punishing him for having raised those concerns.
(Simile, Arch. Reb. Prep.
Test. at 6-7 ff. Tr. 16180.)
375.
Archambeault heard about Simile's remark from Revels and was unhappy about it.
(Archambeault, Tr. 12359-60.)
Simile testified that neither the perceived predicament nor Archambault's raising concerns to the NRC played any part in his decisions not to transfer Archambeault.
(Tr. 16221-22, 16228.)
We find his testimony credible.
The essence of the predicament, which we think he assessed realistically, was that either granting or denying the transfer request could be interpreted as improper.
Had Simile been disposed to put appearances before business considerations, transferring Archambeault clearly would have posed the lesser risk.
376.
On July 10, 1986, Archambeault sent a second written request for transfer to Nemeth, and a copy to Cartelli, the GE-MCIS Projec Manager.
(Archambeault, Tr. 12327-29; Int.
Ex. 123.)
(In July 1986, MCIS replaced BESTCo as the employer of the Comstock QC inspectors.
See Martin, Tr. 9146-47.)
l Again, Archambeault received no response.
(Archambeault, Tr. 12327-28.)
Again, Simile did not receive this request for consideration.
(Simile, Arch. Reb. Prep. Test. at 8 ff.
Tr. 16180.)
377.
On August 19, 1986, Archambeault and Simile both attended a meeting which had been called at Archambeault's
- 110 -
request to resolve his quality concerns.
Archambeault ex-pressed his desire to be transferred to the first shift and Simile told him it was not then possible because of the con-tinuing cable pulling effort on second shift.
(Id. at 7; Archambault, Tr. 12379-83.)
378.
Archambeault sent a third written request for transfer to Nemeth and Cartelli on August 7, 1986.
(Archambeault, Tr. 12333-35; Int. Ex. 124.)
Simile first saw this memo on August 27, when Cartelli showed it to him.
Simile noted on the memo that a shift change was not possible at that time because of the continuing priority of the second shift cable pulling effort.
He returned the memo to Cartelli.
(Simile, Arch. Reb. Prep. Test, at 8 ff. Tr. 16180; Int.
Ex. 124.)
379.
Archambeault was transferred to first shift in September 1986.
(Archambeault, Tr. 12143.)
This transfer was made possible when the crew of cable pull craftsmen whose work he was inspecting was transferred from second to first shift at that time.
(Archambault, Tr. 12656-57.)
380.
Among the quality concerns Archambeault raised i
with the NRC was one related to deficiencies he discovered in l
l the upper cable spreading room.
(Archambeault, Tr. 12311-12.)
These included damaged cable, bend radius violations, separa-tion violations, and cables falling out of the tray.
(Archambeault, Tr. 12216.) Archambeault believed that the NCR documenting these violations should list every nonconforming l
l cable.
Revels, his supervisor, however, thought that a generic l
l l
- 111 -
NCR, covering all cables in the area, should be written.
Because of this disagreement, Revels told Archambeault to document his concerns in a memo, instead of writing an NCR at that time.
Revels said he would give the memo to Simile and DeWald.
(Archambeault, Tr. 12578-79, 12247-48; Int. Ex. 115.)
381.
Simile examined the affected area with Archambeault and Revels and agreed with Revels that the best way to document the deficiencies was via a generic NCR, cover-ing all the cables in the hrea.
At his direction, Archambeault wrote such an NCR.
(Archambeault, Tr. 12250-52, 12580; Simile, Tr. 16204-06; Int. Ex. 116.)
382.
Archambeault apparently was concerned because he thought the generic NCR was a' method of emphasizing production over quality.
(Archambeault, Tr. 12582; Int. Ex. 122.)
We do not agree.
Simile testified that he preferred the generic NCR because it required all cables in the area to be reinspected and evaluated for rework.
Writing NCRs for individual cables would have run the risk of failing to identify, and therefore failing to require, reinspection of some of the damaged cables because of the congested condition of the area.
(Simile, Tr. 16222-23, 16230-32.)
383.
Edison also issued an NCR addressing the upper cable spreading room deficiencies.
(Int. Ex. 135.)
Archambeault agreed that this NCR covered all his areas of concern and required reinspection of all the cables.
(Tr. 12592-93.)
He also agreed that this satisfied his concern.
(Tr. 12246, 12586.)
- 112 -
384.
A second quality concern raised by Archambeault involved cable separation problems in the vicinity of the remote shutdown control panel for Unit 1.
(Archambeault, Tr. 12279-80.)
The Comstock inspection procedure prescribes minimum distances by which cables in various positions may be separated from each other.
(App. Ex. 125 at 14-15.)
Archam-beault observed violations of what he believed was the applicable criterion involving over a hundred safety and non-safety cables.
(Int.
Ex. 122; Archambeault, Tr. 12281-83, 12615.)
He began to write some 42 Cable Separation Conflict Reports (CSCRs) identifying this condition with respect to safety cables.
Before he could finish them, however, he was called away to perform an in-process inspection of a " hot" cable pull, one that required priority attention.
(Int.
Ex. 122; Archambeault, Tr. 12284-88.)
4 385.
Archambeault believed that assigning him to inspect the hot pull before he could complete the CSCRs was an example of production overshadowing quality.
(Int. Ex. 122; Archambeault, Tr. 12602-03, 12610-12.)
We find no merit in this suggestion.
Management is not obliged to allow QC inspectors to perform their tasks at their own convenience.
Archambeault was interrupted for two to three hours to perform the priority task and completed his documentation the following day.
(Archambeault, Tr. 12287, 12602-03.)
He agreed that he did so just as accurately as he would have the day before and that the delay had no effect on quality.
(Tr. 12611-12.)
We regard his complaint as frivolous.
- 113 -
386.
Moreover, the apparent violations seem to have been attributable to Archambeault's misinterpretation of the cable separation criteria in the procedure.
When shown the procedure and questioned about the criteria, he concluded that in fact he had misinterpreted them.
(Archambeault, Tr. 12597-99, 12618-19.)
387.
A third concern raised by Archambeault involved an incident where footage markers on a length of cable were identified as being repetitive rather than consecutive.
(Archambeault, Tr. 12199.)
The numbers repeated for about 50 feet and then self-corrected.
(Archambeault, Tr. 12199, 12515-16.) The craft noticed this discrepancy and asked Archambeault how to proceed.
(Archambeault, Tr. 12203, 12517-19, 12673.)
He showed the craft lead the applicable portion of the cable installation procedure that the craft were working to.
(Archambeault, Tr. 12518-21.)
Archambeault interpreted this procedure as requiring that the cable pull be stopped while engineering was notified.
(Archambeault, Tr. 12200, 12204, 12530.)
388.
Archambeault consulted his supervisor as to how the situation should be handled.
The supervisor told him that he was to inspect only to the QC inspection procedure, not to the craft's installation procedure.
The supervisor said that the craft could proceed.
(Int. Ex. 122; Archambeault, Tr. 12200-03, 12517-21, 12533-34.)
Archambeault informed an Edison QA inspector about th$s.
(Int. Ex. 122.)
Ultimately, the inspectors were told that they were in fact to monitor the l
- 114 -
craft's compliance with its installation procedure.
(Int.
Ex. 122.)
389.
It is not plain what the nature of Archambeault's concern, was here.
His letter to the NRC asks rhetorically how often inspectors have been given similar bad advice by super-visors.
(Int. Ex. 122.)
We can give no weight tc such specula-tion.
Similarly, there is no evidence that the failure to notify engineering about the nonconsecutive footage markers raises any safety concern.
Archambeault agreed that footage markers are unrelated to.the quality of the installed cable.
(Tr. 12538.)
300.
A portion of Intervenors' direct case rested on the testimony of two expert witnesses, Dr. Richard D. Arvey and Dr. David J. McKirnan.
Both have advanced degrees in psychology.
801.
Dr. Richard D. Arvey is a Professor of Industrial Relations and Adjunct Professor of Psychology at the University of Minnesota (Arvey, Prep. Test. at 1 ff. Tr. 4391; Arvey, Prep. Test. Attch. 1 ff. Tr. 4391) and holds a Ph.D. in psychology from that university (Arvey, Prep. Test. at 1 ff.
Tr. 4391).
Dr. Arvey's field of study is industrial-organina-tional psychology (Arvey, Prep. Test. at 1 ff. Tr. 4391).
However, only one paper in Dr. Arvey's Vita after 1978 reported field work involving an industrial organization.
(Arvey, Prep.
Test., Attch. 1 ff. Tr. 4391; Arvey, Tr. 4392-93.)
He has never worked on a construction site.
(Arvey, Tr. 4392.)
- 115 -
802.
Dr. David J. McKirnan is an Associate Professor of Psychology at the University of Illinois at Chicago (McKirnan, Prep. Test. at 1 ff. Tr. 10267) and holds a Ph.D. in clinical and social psychology from McGill University, Montreal, Canada.
(McKirnan, Prep. Test. at 1 ff. Tr. 10267.)
The major focus of Dr. McKirnan's research is w..cial norms, (McKirnan, Prep. Test. at 2 ff. Tr. 10267; McKirnan, Tr. 10275-
- 76) which he has primarily examined in the context of alcohol use and abuse.
(McKirnan, Tr. 10275-76.)
Only one publication in~his Vita involved research on determinants of behavior in the workplace, and that dealt with management personnel at a Canadian telephone company.
(McKirnan, Tr. 10275.)
Dr. McKirnan has never worked at a nuclear power plant.
(McKirnan, Tr. 10274-75.)
803.
Neither Dr. Arvey nor Dr. McKirnan had any prior involvement with nuclear power plants.
(Arvey, Tr. 4392; McKirnan, Tr. 10274-75.)
Both Dr. Arvey and Dr. McKirnan made limited reviews of the facts surrounding the allegations of harassment of Comstock QC inspectors.
Both were provided with copies of Appendix B to 10 CFR Part 50 and the harassment contention, including the three NRC memoranda prepared as a I
result of the visit of the 24 Comstock inspectors to the NRC on March 29, 1985 (Int. Ex. 42, App. Ex. 109, 110, 111).
(Arvey, Prep. Test at 3-4 ff. Tr. 4391; McKirnan also read brief extracts of depositions of certain QC inspectors (App. Ex. 112, 113, 114 and 115).
(McKirnan, Prep. Test. at 2 ff. Tr. 10267.)
Neither Dr. Arvey nor Dr. McKirnan conducted any interviews of
.s--
- 116 -
Comstock management personnel or Comstock QC inspectors, nor did they review any transcripts of hearing testimony.
(Arvay, Prep. Test. at 304 ff. Tr. 4391; Arvey, Tr. 4394; McKirnan, Prep. Test, at 2 ff. Tr. 10267; McKirnan, Tr. 10279.)
Dr. Arvey spent 12 to 16 hours1.851852e-4 days <br />0.00444 hours <br />2.645503e-5 weeks <br />6.088e-6 months <br /> and Dr. McKirnan 5 to 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> reviewing the materials which were provided.
(Arvey, Tr. 4394; McKirnan, Tr. 10279.)
Dr. McKirnan mainly relied on the events recorded in the NRC memoranda as the basis for his opinion.
(McKirnan, Tr. 10280.)
804.
Dr. Arvey's testimony was limited to an opinion on how, from the perspective of industrial-organizational psychology, it might be determined whether inspector perfor-mance was affected by any alleged harassment or intimidation.
(Arvey, Prep. Test. at 4 ff. Tr. 4392.)
He offered no opinion whatsoever on whether there actually had been any such effect at Braidwood.
Dr. Arvey recommended that additional studies concerning inspector work performance, including what he called longitudinal studies of behavioral traces based on archival
- data, i.e.
reinspections of Constock QC-accepted work, and in-depth interviews be conducted and that information regarding work quality, such as reinspection data, be gatered.
(Arvey, Prep. Test. at 11-13 ff. Tr. 4391.)
Dr. Arvey suggested that reinspection of about 10% of the welds would be convincing.
(Arvey, Tr. 4449.)
805.
Dr. McKirnan's analysis is based on a theory of
" observational learning."
(McKirnan, Prep. Test. at 3-10 ff.
l I
Tr. 10267.)
Employees learn values as well as specific skills L
117 -
on the job.
The learned values influence performance standards by providing " anchors" with which to judge good or bad work.
(McKirnan, Prep. Test. at 4 ff. Tr. 10267.)
The strongest observational learning in the workplace is from models who control resources or otherwise have power.
(McKirnan, Prep.
Test. at 5 ff. Tr. 10267.)
806.
In order to analyze whether adverse observational learning from negative models had occurred at Braidwood, Dr. McKirnan believed that three factors should'be examined:
(1) Whether there was a model for the behavior or values in question (quantity over quality) which would serve as an
" anchor" for the intpectors' criterion for good or bad work; (2) If there was.such a model, whether it was viewed as stemming from or representative of the larger organizational culture; and (3) Whether the persons who embodied the model were perceived as having formal or informal power such that they could affect the rewarde or punishments received by employees, or that conformity to the model would increase positive outcomes for employees while deviance would decrease such outcomes.
(McKirnan, Prep. Test. at 4, 7-8 ff.
Tr. 10267.)
Workplace values of quantity over quality might lower inspectors' " anchors" for performance subtly, so that no inspector might pass a piece of work he identified as defi-cient, but the inspectors might generally understand that the standard for passable work is being lowered.
(McKirnan, Prep.
l Test. at 12 ff. Tr. 10267.)
i
- 118 -
807.
In his general discussion of observational learning Dr. McKirnan testified that various factors influence or set the " anchors" for standards with which an individual will evaluate. work situations.
(McKirnan, Tr. 10289-91, 10394-97; McKirnan, Prep. Test. at 4 ff. Tr. 10267.)
In the case of a quality control inspector, these fac ors would include training in standards for the quality of work inspected and for conduct of inspections.
(McKirnan, Tr. 10293.)
In preparing his testimony Dr. McKirnan made no review of the training received by quality control inspectors during their employment by Comstock, (McKirnan, Tr. 10282, 10293) even though training can be an important influence in communicating values and standards.
(McKirnan, Tr. 10293.)
Dr. McKirnan testified that training that stressed performance standards very strongly would communicate not only the content of those standards, but also that adhering to those standards is impor-tant.
(McKirnan, Tr. 10293.)
The type of training that L.K.
Comstock quality control inspectors received stressed perform-ance standards.
(DeWald, Tr. 1810-12.)
808.
According to Dr. McKirnan, an individual's prior experience, such as a quality control inspector's work at another nuclear power plant construction site, initially establishes the standards of that individual.
(McKirnan, Tr. 10291-92.)
However, although Dr. McKirnan knew that at least some Comstock quality control inspectors had previously worked at other nuclear power plants, (McKirnan, Tr. 10292-293) l
- 119 -
he did not investigate their experience in any systematic way.
(McKirnan, Tr. 10293.)
809.
Although the expectations of other quality control inspectors certainly are a factor that can be expected to influence an inspector's behavior, (McKirnan, Tr. 10297)
Dr. McKirnan did not mention this factor in his prepa.c.d testimony discussing application of the observational learning theory to organizations.
(McKirnan, Tr. 10297-99; McKirnan, Prep. Test. at 5-6 ff. Tr. 10267.)
810.
Dr. McKirnan considered harassment of Comstock QC inspectors to be evidence of a model for valuing quantity over quality.
(McKirnan, Prep. Test, at 8 ff. Tr. 10267.)
But certain activities do not constitute harassment.
Dr. McKirnan did not consider setting productivity standards for inspectors to constitute harassment unless these standards were set unreasonably high.
(McKirnan, Tr. 10284.)
Coarseness of language, particularly at a construction site, also does not constitute harassment.
(McKirnan, Tr. 10288.)
Profanity, which is normal and expected, gets filtered out and would not affect an inspector's " anchors".
(McKirnan, Tr. 10396-97.)
Nor does the existence of an adversarial relationship or an ordinary level of friction between supervisors and subordinates constitute or evidence harassment.
(McKirnan, Tr. 10288-89.)
811.
In many instances, Dr. McKirnan based his conclusions regarding models for emphasizing quantity of inspections over quality on erroneous, misleading or incomplete allegations found in the materials he reviewed.
Many of these
- 120 -
allegations were based on rumors and " shoptalk" which freely circulated among Comstock QC inspectors.
812.
Dr. McKirnan based his opinion regarding the
" anchor" for emphasizing quantity over quality of inspections in part of his understanding that QC inspectors believed.that an inspector had performed 1,000 weld inspections in a single day.
This is a reference to a persistent rumor of a purported inspection checklist signed by DeWald as a Level II weld inspector in 1980-81.
This checklist purportedly documented his inspection of 1,000 or more welds in a single day.
Eight QC inspectors had heard rumors about such a list, but had not seen it; (Hunter, Tr. 8495-98; Martin, Tr. 8294, 9332-33, Mustered, Tr. 5061-62, 5087; Rolan, Tr. 4762-63, 4770-71; 1
Stout, Dep. Tr. 14445; Klachko, Dep. Tr. 192, 265-66; Hii, Prep. Test. at 3 ff. Tr. 16608; Gorman, Tr. 5817-18, 5828.)
l Six inspectors claimed to have seen it, (Bossong, Tr. 9848-50; Bowman, Tr. 6890-91; Holley, Tr. 5154-56; Perryman, Tr. 9653-57; Peterson, Tr. 5933, 5935), but only Holley and Bossong could recall even general details about the location of the hangers they believed had been identified on the checklist.
(Bossong, Tr. 9850-51; 9853; Holley, Tr. 5161; 5164-67.)
813.
No inspector reported having heard DeWald refer to the 1000 weld checklist in any way, nor specifically in the context of his own past productivity, i.e.,
as setting an example for his inspectors.
(Peterson, Tr. 5939-40; Rolan, Tr. 4872; Gorman, Tr. 5819, 5866; Perryman, 9660-61; Bossong, Tr. 9893.)
- 121 -
814.
DeWald denied that such a checklist had.ever existed.
(DeWald, Prep. Test. at 24 ff. Tr. 1700.)
He ex-plained that in the 1980-81 time period, Comstock QC inspection practice was to perform a series of inspections over a period of days and document them on a form later, which was dated the day the document was completed.
(DeWald, Tr. 1481-83.)
His description of early documentation methodology was generally corroborated by others.
(App. Ex. 103; Martin, Tr. 8298, 8343-78, 9384-97, 9583-84, 9606-607; Rolan, Tr. 4764-65.)
815.
We find that the 1000 weld checklist issue is a red herring.
Extensive searches for the DeWald checklist during the course of this proceeding were unavailing.
(Martin, Tr. 9332-35; Hii Tr. 16646-47; Bowman, Tr. 6894-6902; Bowman, Prep. Test at 4 ff. Tr. 16000.)
One inspector in testimony characterized this whole issue as a joke.
(Bossong, Tr. 9890-94.)
816.
Dr. McKirnan was unaware of the facts set forth 1
l in findings 811-815, supra.
(McKiner, Tr. 10321-26.)
He was l
l similarly unaware that other statements made to the NRC on March 29, 1985 were inaccurate.
One inspector had alleged that another was " railroaded" out of his job, knowing that the other inspector actually should have been removed for poor work performance.
(McKirnan, Tr. 10310-15; Snyder, Tr. 4259-62; Appl. Ex. 109.)
Dr. McKirnan relied on the allegation that three inspectors were to be " weeded out" on the basis of number i
of inspections performed, (McKirnan, Tr. 10319; Appl. Ex. 109) l but was unaware that no such removals ever occurred.
He did l
l l
l l
l
- 122 -
concede that peer pressure, as opposed to perceived pressure to perform more inspections, may have contributed to the reported unanimity of inspector responses.
(McKirnan, Tr. 10377-78.)
He-also admitted that it was possible that disputes between management and labor over salary structure and union representation could have contributed to the feelings expressed at the March 29 meeting.
(McKirnan, Tr. 10378-80.)
817.
Dr. McKirnan was also unaware of other factors which would affect a QC inspector's perception of whether there was a model which stressed quantity over quality.
When he formed his opinion, Dr. McKirnan had no knowledge of PTL overinspections and unit concept inspections, Comstock and CECO QA audits and surveillances, NRC Staff inspections or whether QC inspectors had communicated with the NRC prior to March 29, 1985.
(McKirnan, Tr. 10333, 10354-65.)
He was unaware that QC inspectors expressed concern when PTL rejected their inspec-tions, but agreed that such concern indicated a positive quality orientation.
(McKirnan, Tr. 10362.)
Dr. McKirnan was unaware that QC inspectors used documented checklists and were required to sign those checklists.
(McKirnan, Tr. 10350-54.)
Dr. McKirnan agreed that each of these factors would tend to reinforce a model for inspectors which stressed quality of inspections.
(McKirnan, Tr. 10352-54, 10358-65.)
In par-ticular, he agreed that the PTL 10% overinspection program would exert a positive influence on inspectors' " anchors" for accepting work because of the accountability for errors inher-ent in the requirement that PTL rejects be reinspected and
- 123 -
documented by Comstock.
(McKirnan, Tr. 10357-58.)
Although not aware of these factors when he prepared his testimony, Dr. Arvey also testified that these factors would make the QC inspectors more resistant to any pressure to overlook safety concerns or construction discrepancies.
(See Arvey, Tr. 4398-400, 4402, 4405-07.)
818.
While Dr. McKirnan conceded that a fuller understanding of all the available facts surrounding these allegations would be useful, he consistently declined to alter his opinion regarding whether an understanding of all the facts would indicate the existence of a model at Comstock which emphasized quality, contrary to the result suggested in his direct testimony.
(McKirnan, Tr. 10302-03, 10314-15, 10326, 10329-30, 10347-48.)
In fact, he admitted that he made no effort to obtain a balanced picture of life on the job at Comstock.
(McKirnan, Tr. 10399-400.)
In his view, a QC inspector's perception of a situation, rather than the actual facts, would be determinative in establishing whether such a model existed.
(See, e.g.,
McKirnan, Tr. 10303, 10311, 10329, 10333, 10346, 10360, 10364.)
In his opinion, such perceptions could only be uncovered by further investigation, including interviews and a review of the inspectors' work.
(McKirnan, Tr. 10305-306; 10316-17.)
819.
The second factor in Dr. McKirnan's observational learning theory is whether the models for valuing quantity over quality were viewed as stemming from or representative of the larger organizational culture.
(McKirnan, Prep. Test, at 7 ff.
- 124 -
Tr. 10267.)
This turns in part on whether the behavior of Saklak and other Comstock supervisors and managers was viewed as ref1' ng larger organizational values or as deviant or aberrational.
(McKirnan, Tr. 10303.)
Dr. McKirnan was not aware of any other supervisors who were alleged to have been verbally abusive or have physically threatened any inspector.
(McKirnan, Tr. 10303-304.
See McKirnan, Tr. 10318-20; Appl.
Ex. 109.)
He further testified that if this behavior was viewed by the inspectors as deviant then it would not be a substantial influence on their behavior, (McKirnan, Tr. 10304-305.
See McKirnan, Tr. 10405-409) although Dr. McKirnan considered this an empirical question best resolved by a more systematic study than examining memoranda and depositions.
(McKirnan, Tr. 10305-306, 10406.)
820.
When Dr. McKirnan prepared his testimony he was unaware of what had become of Mr. Saklak after he was placed on administrative leave.
(McKirnan, Tr. 10283, 10302.)
When asked how the quality control inspectors might view Mr. Saklak i
being " laid off" and being barred from any safety-related position at Applicant's nuclear power plant _ sites, (see DeWald, Tr. 1565-66, 1570; Saklak, Tr. 8028, 8035-36, Int. Ex. 39)
Dr. McKirnan said that this certainly could be taken as evi-dence that management had responded positively to the expres-sion of quality concerns by the inspectors.
(McKirnan, Tr. 10302-303.)
However, consistent with his view of the importance of QC inspectors' perceptions, Dr. McKirnan would want to investigat) whether the inspectors felt that management
i
-i
- 125 -
had acted only once a large number of inspectors had gone as
-far as contacting the NRC.
(McKirnan, Tr. 10302-303, 10306-308.)
821.
The third factor in the observational learning theory applied by Dr. McKirnan is the formal or informal power-of the persons who embody the models of behavior or values.
(McKirnan, Prep. Test, at 7-8, 9-10 ff. Tr. 10267.)
Such power includes job assignments, pay raises, and termination of employment.
(Id.)
Although Dr. McKirnan rendered an opinion on the power of the models he perceived in the Comstock environment, (McKirnan, Prep. Test, at 7-8, ff. Tr. 10267) he later admitted that he lacked sufficient knowledge to be definitive about the power of the persons who embodied those models.
(McKirnan, Tr. 10295-96.)
822.
In Dr. McKirnan's view, if the quality control inspectors deliberately provoked Mr. Saklak or were amused by his outbursts, that indicates that they didn't see him as
-capable of fulfilling all of his threats.
(McKirnan, Tr. 10337.
See Staff Ex. 17 at 13; Hunter Tr. 8787-89)
Moreover, repeated threats by Saklak to fire various inspectors were not carried out, indicating to Dr. McKirnan that Saklak's power was demonstrably ineffective.
(McKirnan, Tr. 10344-45.)
When one inspector saw Mr. Saklak shake his finger at another inspector, that first inspector said to Saklak:
"If you ever shake your finger at me, I'll tear it off and make you eat it."
(Perryman, Tr. 9757.)
Dr. McKirnan testified that this indicated that the first inspector was not substantially
- 126 -
concerned about Mr. Saklak's power in terms of harassment.or intimidation, at least in that context.
(McKirnan, Tr. 10338.)
Although the inspector's statement to Mr. Saklak was in a deposition which' Dr. McKirnan read excerpts from, Dr. McKirnan did not recall-having heard of it.
(McKirnan, Tr. 10338.)
823.
Dr. McKirnan concluded in his prepared testimony that it was " highly probable" that conditions at Braidwood had-l significantly affected the Comstock QC inspectors' work atti-tudes and their thresholds for approval of work that they inspected.
(McKirnan, Prep. Test. at 10 ff. Tr. 10267.)
However, Dr. McKirnan recognized that this is an empirical i
question that has to be resolved by methods other than those he applied.
(McKirnan, Prep. Test. at 10-11 ff. Tr. 10267.)
On cross examination Dr. McKirnan stated that rendering an opinion on whether there had been an effect on inspector work perfor-mance was inappropriate because the issue could not be deter-mined from the materials he had reviewed.
(McKirnan, Tr. 10381-82.)
He later characterized the conclusion in his prepared testimony as "there is reason to suspect there is an i
issue" as to whether inspector standards had been affected.
(McKirnan, Tr. 10399-400.)
He recommended recourse to engi-neering and behavior trace information to determine whether an I
adverse shift in inspector performance took place.
(McKirnan, l
Prep. Test, at 10 ff. Tr. 10627.)
He asserted that temporal analysis of data should show a relationship between specific 3
incidents of harassment and numbers of inspection misses such that the number of misses goes up when harassment occurs and l
l vw- - - -
-,-,_-,,_--,p r
,,w.-9
-.--...,.,,,--.-.-,,,-.---.---_m-
- 127 -
down when it abates if there is an effect on inspector perfor-l mance at all.
(McKirnan, Tr. 10390.)
824.
Both Dr. Arvey and Dr. McKirnan commented on one aspect of the testimony which was before us.
Although neither one had observed any QC inspector testify, nor had they studied the complete transcripts of hearing testimony or depositions, each urged us to discount testimony by QC inspectors which denied that any perceived harassment had caused them to compro-mise the quality of their inspections.
825.
Dr. Arvey testified that employees are unlikely to admit that they violated a policy governing their work.
(Arvey, Prep. Test. at 9 ff. Tr. 4391.)
However, he later admitted that, at least in the abstract, a statute protecting quality control inspectors at a nuclear power plant who reported defects in their work would make those inspectors more likely to identify such defects.
(Arvey, Tr. 4422.)
Dr. Arvey characterized this issue as very complex, (Arvey, Tr. 4423) and ultimately referred to his opinion that individuals will not easily report their own errors as a " hunch."
(Arvey, Tr. 4424.)
826.
Dr. McKirnan testified that the quality control inspectors were highly unlikely to admit that they or their co-workers had approved deficient work.
(McKirnan, Prep. Test.
at 11-12 ff. Tr. 10267.)
However, Dr. McKirnan stated that the demeanor of witnesses, their body language and the context within which an answer to a question is given is crucial to evaluating the reliability of an answer (McKirnan, Tr. 10389)
J
i 128 -
and agreed that this Board was in a much'better position to judge the inspectors' veracity.
(McKirnan, Tr. 10389.)
827.
A portion of Applicant's rebuttal case consisted of the testimony of two expert witnesses who addressed both the issue of whether the variety of incidents characterized as harassment were likely to have affected the quality of Co stock QC inspectors' performances and whether we should credit QC inspectors' denials that these incidents had no effect on their performance.
These two witnesses were Mr. Robert V. Laney and Dr. Charles W. Hulin.
828.
Dr. Charles L. Hulin received a M.A. in indus-3 trial and labor relations in 1960 and a Ph.D. in psychology in 1963 from Cornell University.'
(Hulin, Reb. Prep. Test, at 2 ff. Tr. 17924.)
Since 1962, with the exception of sabbaticals, he has been a faculty member in the Psychology Department and the Institute of Labor Relations at the University of Illinois at Urbana Champaign.
(Id.)
He is currently Professor of Psychology in the school of Liberal Arts and Sciences and in the Aviation Research Laboratory of the University's Aviation Institute.
(Id. at 2-3.)
829.
Dr. Hulin's speciality is industrial-organiza-tional psychology.
(Hulin, Reb. Prep. Test. at 1 ff. Tr.
17924.)
This field involves the study of the behavior of normal adult individuals in business and industrial organi-zations.
(Id.)
Two of Dr. Hulin's subspecialties are motiva-tion and job attitudes.
(Id. at 1-2.)
Dr. Hulin has been a co-author of four books, sole or co-author of nine chapters in f
4
.. - - - +
- - - - -, ~ - -
,n,.
=, - - -
-m----,,--n
---e--+--n
-,,---,,,-m,
,m,,,,,e,,-
.--r
---,-e
- 129 -
collective research publications, and lead or co-author of almost fifty articles published in technical journals and almost twenty unpublished technical reports.
(Hulin, Reb.
Prep. Test., Attch. 1 ff. Tr. 17924.)
830.
Dr. Hulin has extensive experience studying the environmental, organizational, and individual characteristics thac influence the behavior of individuals working in complex organizations.
(Hulin, Reb. Prep. Test. at 4-5 ff. Tr. 17924.)
His work has been heavily oriented toward blue collar workers.
(Id.)
831.
Dr. Hulin's task at Braidwood was to assess whether the alleged harassment and intimidation of Comstock quality control inspectors set forth in Intervenors' contention may have adversely affected the quality of work of Comstock QC inspectors.
(Hulin, Reb. Prep. Test at 3 ff. Tr. 17924.)
He spent approximately two hundred hours reviewing the available data and preparing his testimony for this hearing.
(Hulin, Tr.
18257.)
In preparation for formulating his opinion, Hulin read various memos and reports and the testimony of a number of individuals.
(Hulin, Reb. Prep. Test. at 16-17 ff. 17924; Hulin, Tr. 18182-83.)
He examined the BCAP CSR and PTL rein-spection data in detail (Hulin, Reb. Prep. Test, at 17 ff.
Tr. 17924) and spent two days at the site interviewing quality management personnel and six QC inspectors.
(Hulin, Reb. Prep.
Test. at 17-18 ff. Tr. 17924 Hulin, Tr. 17897-98.)
He also read Archambeault's letters and summaries of the testimony of the QC inspectors given in the hearings and interviewed seven
-- -=.- =--,,
~-
- 130 -
of the second shift cable pulling QC inspectors.
(Hulin, Tr. 17899.)
832.
Mr. Robert V. Laney graduated from the United States Naval Academy.
(Laney, Reb. Prep. Test. at 1 ff. Tr.
17245.)
He holds a masters degree in Marine Engineering from the Massachusetts Institute of Technology (Laney, Reb. Prep.
Test. at 1 ff. Tr. 17245) and a masters degree in Business Administration from the University of Chicago.
(Laney, Reb.
Prep.-Test. at 1 ff. Tr. 17245.)
Mr. Laney's thirty-eight years of experience in the field of nuclear power have fre-quently required him to become intimately involved with efforts to achieve and confirm high standards of construction quality while meeting demanding produ~ction schedules.
(Laney, Reb.
Prep. Test, at 5 ff. Tr. 17245.)
He has served as a member of a review group appointed by the NCR in connection with an evaluation of quality assurance issues for nuclear power plants under construction submitted to Congress.
(Laney, Reb. Prep.
Test. at 4, 6 ff. Tr. 17245.)
Mr. Laney has kept abreast of changes in nuclear power plant design, construction and quality control methods.
(Laney, Tr. 17316-18.)
Mr. Laney reviewed copious materials in preparing to testify in this proceeding,
(
i (Laney, Reb. Prep. Test. at 7 ff. Tr. 17245; Laney, Reb. Prep.
Test., Attch. B ff. Tr. 17245; Laney, Tr. 17243-44, 17254-55, 17312-16, 17318-19, 17320-31); as well as interviewing certain management personnel of Applicant, Comatock and Sargent &
i Lundy.
(Laney, Reb. Prep. Test. at 7 ff. Tr. 17245.)
Mr. Laney read abstracts of the depositions of a large number
[
- 131 -
of inspectors.
(Laney, Tr. 17313-14; Laney, Reb. Prep. Test.,
Attch. B ff. Tr. 17245.)
833.
Dr. Hulin's research technique is eclectic.
Part of his approach'is to gain as much first hand knowledge of the workers and their jobs as the investigating situation permits, sometimes by interviews and discussions, sometimes by brief stints working in the occupations being studied.
(Hulin, Reb.
Prep. Test, at 4-5 ff. Tr. 17924.)
He believes in using multiple research metho'ds in any given study so that the inherent biases and flaws in the individual methods cancel one another out.
(Hulin, Reb. Prep. Test. at 6-7 ff. Tr. 17924.)
If several methods with different and independent sources of bias lead to the same conclusion, confidence-in the validity of that conclusion is strengthened.
(Hulin, Reb. Prep. Test. at 6-7 ff. Tr. 17924.)
834.
Dr. Hulin's opinion was that it seemed highly unlikely that the alleged harassment and intimidation and undue management pressure adversely affected the quality of the work of the quality control inspectors.
(Hulin, Reb. Prep. Test. at 19, ff. Tr. 17924.)
To form this opinion, he relied on three more or less equally influential types of evidence.
The first was organizational, interpersonal, and psychological factors.
The second was the impact of the society's and inspectors' own values and norms on their job behaviors.
The third was BCAP and PTL overinspection and reinspection data reflecting on the quality of QC inspector work.
(Hulin, Reb. Prep. Test. at 19-20 ff. Tr. 17924.)
- 132 -
835.
The first of the organizational, interpersonal, and psychological factors Dr. Hulin considered was the degree of harassment and intimidation by Saklak of the QC inspectors.
While he considered Saklak's statements to Snyder extreme and undesirable, he did not believe that it would directly affect work behavior of the QC inspectors.
That is so because Saklak's behavior was within what Dr. Hulin perceived as the expected range of supervisor role behaviors, i.e.,
exerting pressure on employees to be productive and using rough and profane language.
Saklak's behavior was reprehensible and above the noise level, i.e.,
noticeable, but not likely to have long-term effects on work behavior of QC inspectors.
(Hulin, Reb. Prep. Test. at 21-23 ff. Tr. 17924.)
836.
A second factor was safeguards built into the system permitting QC inspector access to the NRC and to the CECO Quality First program, e.g.,
via posted phone numbers.
Dr. Hulin also pointed to inspector knowledge that work was to be reinspected or overinspected on a random basis and to inspector belief in the efficacy of those overinspections.
He further noted that the safeguards had worked dramatically when l
Saklak was fired after the inspectors went to the NRC.
(Hulin, Reb. Prep. Test, at 24 ff. Tr. 17924.)
837.
Another factor was the nature of the work.
The i
existence of overinspections and the requirement for inspectors to sign inspection reports made sloppy work perilous.
Moreover, except for a few weld attributes, the accept / reject j
criteria are objective and can be applied consistently by l
l
- 133 -
independent inspectors.
(Hulin, Reb. Prep. Test at 25-26 ff.
Tr. 17924.)
838.
Yet another factor was' lack of hard and fast 2
production standards.
Inspectors indicated in interviews that production standards were informal and flexible, that inspectors could be disciplined for sleeping on the job and the like, but that nothing was done to encourage production at the expense of quality.
(Hulin, Reb. Prep. Test. at 26-27, ff.
Tr. 17924; Hulin, Tr. 17952-53, 17978-79.)
839.
Dr. Hulin did not interpret the Seeders and Puckett incidents as necessarily arising out of an orientation to production as opposed to quality.
Rather, he said, it is plausible that Seeders was a marginal employee who did not do the work he was supposed to.
Similarly, it is plausible to conclude that Puckett simply lacked the abilities for the Level III weld inspector job into which he was hired.
The interpre-tations of other inspectors of the Seeders and Puckett incidents did not demonstrate that these incidents contributed to a pervasive influence of quantity over quality.
(Hulin, Reb. Prep. Test. at 26-30 ff. Tr. 17924.)
840.
Dr. Hulin also concluded that Saklak lacked organizational power.
Interviews indicated to him that Saklak could not translate threats into outcomes -- he did not decide pay raises, could not unilaterally assign suspensions, could only recommend vacation time and could not terminate an inspec-t tor's employment.
The quality control manager had the real power.
(Hulin, Reb. Prep. Test. at 30 ff. Tr. 17924.)
a
,---..n..
134 -
841.
Dr. Hulin also based his conclusions in part on a theory of organizational behavior he labelled the Subjective Expected Utility Theory.
In this theory, workers make choices to maximize their subjective satisfaction based on known or perceived " contingencies," that is, probabilistic relation-ships, between their actions and desirable or undesirable oatcomes.
(Hulin, Reb. Prep. Test, at 7-8, 31 ff. Tr. 17924.)
Workers learn about contingencies through experience, observa-tion of others in situations, direct instruction, and compari-son to past experiences.
Contingencies can be subjective as well as objective.
(Hulin, Reb. Prep. Test. at 7-8, 31 ff.
Tr. 17924.)
842.
A contingency can be attached to the evaluation by co-workers or superiors of an individual is work product.
The influence of an evaluator is related to his power.
(Hulin, Reb. Prep. Test. at 30 ff. Tr. 17924.)
Who controls our behavior comes down to whom we perceive as having a right to evaluate what we do and who has the power to translate the evaluation into positive and negative outcomes.
(Hulin, Reb.
Prep. Test, at 13 ff. Tr. 17924.)
The locus of such power is determined by the employee mainly from experience.
(Hulin, Reb. Prep. Test. at 15 ff. Tr. 17924.)
843.
Hulin found that there were essentially no organizational contingencies for inspectors, objective or subjective, between sheer volume of inspections produced and organizational outcomes.
In particular, Saklak's evaluations had few, if any, impacts on their organizational outcomes.
- 135 -
While there were real and perceived contingencies between sleeping on the job and grossly substandard performance and adverse outcomes, the record of terminations showed nothing beyond the enforcement of normal performance standards.
(Hulin, Reb. Prep. Test. at 31-32 ff. Tr. 17924.)
844.
Dr. Hulin's second major source of support for his conclusion was the impact of social and personal norms and values on the inspectors.
He noted that inspectors bring to the job with them the values and norms of the industry and of the society to which they belong, especially what could be 1
i described as middle class work values:
to do poor work is to be a poor person; work gives identity.
It takes significant force, significant and long-t'erm application of negative contingencies counter to work values to change them signifi-cantly.
(Hulin, Reb. Prep. Test. at 33 ff. Tr. 17924.)
He believed that for these reasons a purported leader such as 4
t Saklak would not be likely to have systematic and significant 4
effect on the overall level of production of QC inspectors.
(Hulin, Reb. Prep. Test, at 34 ff. Tr. 17924.)
)
845.
The final major source of information supporting 1.
Dr. Hulin's conclusion was the overinspection and reinspection data from the PTL and BCAP CSR programs.
He tested the hypotheses suggested by the Intervenors' contention using the best available behavioral trace measures, namely the overinspection/ reinspection agreement rates analyzed by Dr. Frankel.
(Hulin, Reb. Prep. Test. at 17, ff. Tr. 17924; Hulin, Tr. 17927-35, 18231-32, 18238-39.)
See also FF 901 to i
,___m.,
m.
--136 -
954 infra.)
Hulin looked for breaks linked to the arrival or departure of Saklak or the arrival of DeWald.
(Hulin, Reb.
Prep. Test. at 17-18, 34-37 ff. Tr. 17924.)
Hypotheses based on other specific events were tested as well.
A measure of agreement between the original QC inspectors and the over-or re-inspectors was used.
Agreement rates were assentially constant; no significant breaks were seen at any hypothesized discontinuity date.
Thus, neither Saklak nor DeWald had any observable effects on trends in quality.
(Hulin, Reb. Prep.
Test. at 34-37 ff. Tr. 17924,)
846.
Dr. Hulin also ascribed the visit of the 24 QC inspectors to the NRC to a number of conventional causes in addition to the Saklak/Snyder incident.
Candidates for impor-tant precipitating factors included the unionization drive and the frustration over lack of effect after the election; Saklak's. unpopularity; the inability to prosecute a conventional grievance against Saklak; and social pressure and conformity.
With respect to social pressure and conformity, Hulin referred to the Asch and Sheriff effects in social psychology as suggesting that complaints of one or two or three inspectors could have translated into a visit by 24 of them even though not all of the 24 had any complaint.
(Hulin, Reb.
Prep. Test at 38 ff. Tr. 17924.)
The NRC event, he said, was a field demonstration of the Asch conformity experiment in which people agree to statements contrary to observable fact because of internal pressure to conform to prior statemerts of others.
(Hublin, Tr. 18191-92.)
Experimental with the Sheriff effect
=.
- 137 -
demonstrate the same principle.
(Hulin, Tr. 18232-33.)
Dr. Hulin concluded that singling out Saklak as the sole cause was not warranted on the basis of other evidence suggesting multiple motivations for the risk.
(Hulin, Reb. Prep. Test. at 37-38 ff. Tr. 17924.)
847.
Dr. Hulin also concluded that low morale does not affect work quality.
He stated that fifty years of studies indicate a lack of relationship between job satisfaction and productivity, either in quantity or quality.
People either indulge in absenteeism or quit but do not reduce job perfor-mance.
(Hulin, Reb. Prep. Test. at 39 ff. Tr. 17924.)
848.
Dr. Hulin concluded that the data he reviewed indicated that it was unlikely that the alleged harassment and intimidation had no significant effect on the work of the QC inspectors.
Prep. Test, at 40 ff. Tr. 17924.)
He maintained even after heavy cross examination that the overwhelming consistency of the data he had seen supported his conclusions.
(Hulin, Tr. 18256-57.)
849.
Mr. Laney testified regarding the likely effects on quality control inspectors' behavior of the problems con-fronting L.K. Comstock's quality control organization beginning in 1983 and continuing into early 1985.
Mr. Laney categorized these problems as new management, employer-employee relations problems, an inspection and document review backlog, a tight labor market, explosive growth, increased attrition, training lead time, extensive overtime, and salary inadequacy, instabil-ity and inequity.
(Laney, Reb. Prep. Test, at 20-21 ff.
138 -
Tr. 17245.)
Furthermore, a unionization drive was coincident with much of this period.
(Laney, Reb. Prep. Test. at 21 ff.
Tr. 17245.)
850.
Mr. Laney testified that these problems were likely to have several effects.
The quality control inspectors would lose confidence in management because of its apparent lack of preparedness for the substantial increase in work demand and, over time, its seeming slowness to adjust to the increase.
(Laney, Reb. Prep. Test. at 21 ff. Tr. 17245.)
Management handling of recruiting, pay adjustments and training incentives must have appeared to the inspectors as tardy, inadequate and, as to salaries, inequitable.
(Laney, Reb.
Prep. Test. at 21-22 ff. Tr. 17245.)
The inspectors' com-plaints to the NRC resident inspectors on March 29, 1985, and in deposition testimony, show wide dissatisfaction with manage-ment and compensation.
(Laney, Reb. Prep. Test. at 22 ff.
Tr. 17245.)
851.
Greater aggressiveness on quality control inspectors' part in criticizing other management actions was likely to accompany this resentment and loss of confidence.
(Laney, Reb. Prep. Test. at 22 ff. Tr. 17245.)
Not only would expressions of dissatisfaction with supervisory pressures become more common, but minor personnel difficulties would tend to become problems, and problems would tend to become weighty grievances.
(Laney, Reb. Prep. Test. at 22 ff. Tr. 17245.)
The level of sensitivity to all uses of authority and to perceived affronts or inequities would rise.
(Laney, Reb.
l i
- 139 -
Prep. Test. at 22 ff. Tr. '7245.)
The complaints made to the NRC resident inspectors on March 29, 1985 also demonstrate these phenomena.
(Laney, Reb. Prep. Test. at 22 ff.
Tr. 17245.)
852.
Under these circumstances the unionization drive would heighten existing unrest, magnify personal differences between quality control inspectors and supervisors, and elevate the intensity of employer-employee controversies.
(Laney, Reb.
Prep. Test. at 22 ff. Tr. 17245.)
Activism in the support of employee grievances is a customary way of demonstrating the potential of union power.
(Laney, Reb. Prep. Test, at 22-23 ff. Tr. 17245.)
The intensive criticism of supervisor Richard Saklak at the quality control inspectors' meetings with NRC resident inspectors on March 19, 1985 is an example of genuine resentment finally reaching the point of action, possibly under" the stimulus of union encouragement.
(Laney, Reb. Prep. Test, at 23 ff. Tr. 17245.)
In Mr. Laney's opinion the quality -
control inspectors saw Saklak as a tremendous bother and irritant but did not take Saklak's threats seriously.
(Laney, Tr. 17360-61; Laney, Reb. Prep. Test, at 23 ff. Tr. 17245.)
However, Mr. Laney believed that if management erroneously saw those complaints as made solely to advance the cause of the union, then management would have missed other significant messages.
(Laney, Tr. 17362-71.)
The inspectors' complaints to the NRC should not have been unexpected under these circum-stances.
(Laney, Reb. Prep. Test. at 23-24 ff. Tr. 17245.)
- 140 -
853.
That the inspectors complained to the NRC also is consistent with their subsequent testimony that the circum-stances or incidents of which they complained did not adversely affect the quality of their work.
(Laney, Reb. Prep. Test. at 24 ff. Tr. 17245.)
These were group actions emphasizing group complaints.
(Laney, Reb. Prep. Test. at 25 ff. Tr. r,.145.)
Inspectors usually work alone, however.
(Laney, Reb. Prep.
Test, at.24 ff. Tr. 17245.)
They are proud and protective of their technical skills and their personal integrity, for these represent not only their job security but also their badge of acceptance in the workplace.
(Laney, Reb. Prep. Test. at 24 ff. Tr. 17245.)
An inspector working alone will scrupulously protect these attributes, unaffected by external group dissen-sion.
(Laney, Reb. Prep. Test. at 24-25 ff. Tr. 17245.)
Mr. Laney cited several examples of L.K. Comstock inspectors who resisted perceived pressure from management to engage in conduct to which the inspectors objected.
(Laney, Tr. 17335-37.)
This behavior was consistent with Mr. Laney's long experience with quality control inspector behavior.
(Laney, Tr. 17337.)
854.
Mr. Laney would not simply take at face value an inspector's claim that the quality of his work was unaffected by pressure or harassment, (Laney, Tr. 17300) but here, because of all of the evidence corroborating that testimony he found it not only consistent with the complaints to the NRC but credible.
(Laney, Reb. Prep. Test. at 24 ff. Tr. 17245; Laney Tr. 17335-37.)
- 141 -
901.
In response to Intervenors' theory that alleged acts of harassment and intimidation, and " historic, tangible, powerful and pervasive production pressure" impaired the effectiveness of Comstock QC inspectors, (Guild, Tr. 7903-04) and the testimony of Intervenors' expert witnesses (see:
FF 800-26, above), Applicant presented the restats of two large sample reinspection programs.
The first program was the Construction Sample Reinspection (CSR) conducted as part of the Braidwood Construction Assessment Program (BCAP).
This data spans the time period from the start of construction until June 30, 1984.
The second set of data results from the routine overinspection of Comstock QC accepted word by Pittsburgh Testing Laboratory (PTL) for Applicant's Quality Assurance Department for the period July 1, 1982 to June 30, 1986.
These reinspection programs were conceived, designed and carried out independently of each other, for reasons unrelated to Inter-venors' allegations.
They were carried out by qualified inspectors who were independent of Comstock's QA/QC organiza-tion.
(Del George, Reb. Prep. Test, at 6, 9,
ff.
Tr. 16740; Kaushal, Reb. Prep. Test. at 4, ff. Tr. 13068; Marcus, Reb.
Prep. Test. at 7-8, ff. Tr. 15568.)
902.
For purposes of this hearing Applicant created a computerized data base for each program which matched the reinspection results with the names of the Comstock QC in-spectors whose work was reinspected and the dates of such Comstock "first line" inspections.
These data bases permit comparison of Comstock QC inspector performance over time with
- 142 -
i episodes of alleged harassment, intimidation, and undue produc-tion pressure.
The CSR and PTL data bases include an enormous amount of information concerning Comstock QC inspector perfor-mance.
For example, the CSR data base includes reinspections of 733. electrical items including more than 10000 welds.
The PTL data base includes more than 7200 components and 28000 welds, approximately 28% of the total number of components and 10% of the total number of welds completed and QC accepted by Comstock during the four year period from July, 1982 to June, 1986.
(Del George, Reb. Prep.
Test. at 19, 32, ff. Tr. 16740; Marcus, Reb. Prep. Test. at 12, ff. Tr. 15568.)
The CSR data base captures the work of 75 Comstock and E.C. Ernst inspec-i tors, including 19 of the 24 Comstock QC inspectors who ex-pressed concerns to the NRC Staff on March 29, 1985.
The PTL data base captures 100 Comstock inspectors, including 20 of the 24 complaining inspectors.
(Del George, Reb. Prep. Test. at
- 17, ff. Tr. 16740.)
903.
The CSR and PTL data bases show (1) consistently acceptable Comstock QC inspector performance and (2) no trends or variations over time which can be correlated with alleged episodes of harassment, intimidation and undue production pressure.
(Del George, Reb. Prep. Test. at 9-14, ff.
Tr. 16740; Marcus, Reb. Prep. Test. at 33, ff. Tr. 15568; j
Kaushal, Reb. Prep. Test. at 33-34, ff. Tr. 13068.)
904.
Thus, the CSR and PTL overinspection data con-tradict Intervenors' claims that harassment, intimidation and
" historic, tangible, powerful and pervasive production v
w w--
.,r,
- 143 -
pressure" had a significant, adverse impact on performance of Comstock inspectors.
There are three possible explanations:
(1) either the episodes and conditions which have been litigated in this case were not perceived by Comstock QC inspectors as requiring them to sacrifice quality for quantity, or (2) the Comstock inspectors did not allow these e,sisodes and I
conditions to affect their performance, or (3) any effect on Comstock QC inspector performance was constant over time.
Even accepting the third alternative explanation, the adequacy of the Comstock QC accepted work is established by the CSR results.
(Del George, Reb. Prep. Test, at 47-49; ff. Tr.
16740.
See also Del George, Tr. 16945-48; Kostal, Reb. Prep.
Test. at 32, App. Ex. 179; Thorsell, Reb. Prep. Test. at 19, ff. Tr. 14270.)
In the following findings, we describe in detail the design, implementation and results of the CSR and PTL overinspection programs.
905.
The Braidwood Construction Assessment Program (BCAP) was a program of reinspections and reviews carried cut by Applicant in 1984 and 1935 covering safety related construc-tion activities at Braidwood.
The BCAP was comprised of three principle program elements.
These were (1) the construction sample reinspection ("CSR"), (2) the reverification of pro-cedures to specification requirements ("RPSR") and (3) the review of significant corrective action programs ("RSCAP").
The BCAP was designed and carried out for reasons unrelated to intervenors' claims of harassment, intimidation and production pressure on Comstock QC inspectors.
(Kaushal, Reb. Prep. Test.
- 144 -
at 4, ff. Tr. 13068; Del George, Reb. Prep. Test. at 6, ff.
Tr.
16740.)
(Kaushal, Reb. Prep. Test. at 3, ff. Tr. 13068; Staff Ex. 25 at 3.)
906.
The CSR consisted of a visual reinspection of a sample of on-site safety-related construction work which as of June 30, 1984 had been completed and QC accepted.
T'te sample was selected based on engineering judgment, including appropri-ate use of statistical concepts.
The reinspections were carried out from October, 1984 through July, 1985.
(Kaushal, Reb. Prep. Test. at 3, 13-16 ff. Tr. 13068.)
907.
The purpose of the CSR was to assure that no unidentified or unaddressed programmatic design significant construction problems exist at Braidwood.
The term " design significant" relates to the kind of construction deficiency which if uncorrected might prevent the affected portion of the construction from fulfilling its safety function.
The term
" programmatic" indicates that in performing the CSR, Applicant was looking for problems which occurred repetitively and went unidentified or unaddressed through the existing quality program.
(Kaushal, Reb. Prep. Test. at 4-6, ff. Tr. 13068.)
908.
The principle result of the CSR was that no design significant discrepancies - programmatic or otherwise were identified.
(Kaushal, Reb. Prep. Test. at 28, ff.
Tr.
13068 and Attachment 2C (Kaushal-3) (Rev. 2), App. Ex. 177.)
Based on the sampling methodology used, this supports a conclusion with at least 95% confidence that at least 95% of the electrical work completed and QC accepted at Braidwood
- < ~ -
- - + -
- 145 -
prior to June 30, 1984 is free of design significant defects.
(Kaushal, Reb. Prep. Test. at 14, ff. Tr. 13068, Kaushal, Tr. 14190; Frankel, Reb. Prep. Testimony at 9-12, ff.
Tr. 17080.)
The BCAP Organization and the Overview of the BCAP 1
909.
The CSR (and other elements of BCAPs were carried out by an organization called the BCAP Task Force.
The BCAP Task Force Director was a Applicant employee, Dr. N. Kausnal.
He reported to the Braidwood Project Manager, Mr. Wallace.
Most of the BCAP engineering staff was drawn from the Stone and Webster Engineering Corporation and all of the inspection staff was drawn from the Daniel Construction Corporation.
None of these individuals had any prior involvement with the work that was to be reinspected or reviewed under BCAP.
The NRC Staff concluded that BCAP Task Force personnel were " qualified for their assigned tasks with a good balance of education and experience in the nuclear industry."
(Kaushal, Reb. Prep.
Test, at 9-10, ff. Tr. 13068; Gardner, Reb. Prep. Test. at 12-13, ff. Tr. 17606; Staff Ex. 25 at 5.
See also Shevlin, Reb. Prep. Test, at 4-11, 15-16, ff. Tr. 130'68; Shevlin, Tr.
14160-61; Wozniak, Reb. Prep. Test. at 2-3, 9-10, ff. Tr.
13068; Smith, Reb. Prep. Test. at 4-5, 16 ff. Tr. 13068.)
910.
Applicant assigned the task of evaluating the discrepancies found by BCAP CSR inspectors for design signifi-cance to Sargent & Lundy, Applicant's Architect-Engineer.
Sargent & Lundy is responsible for developing the design drawings and specifications to which Braidwood is constructed.
- 146 -
Thus it has the greatest expertise in evaluating the signifi-cance of discrepancies in Braidwood's construction.
(Gardner, Reb. Prep. Test. at 8, ff. Tr. 17606.
Kostal, Tr.
15186-87.)
In these hearings, Intervenors have suggested that S&L could not objectively evaluate CSR observations because of its prior involvement at Braidwood.
However, Intetvenors did not con-sider this to be a fatal drawback at the time BCAP was being planned and carried out.
(Kostal, Tr. 14932-35, 15278-83.)
Moreover, because S&L did not perform construction activities, it was not evaluating its own errors.
(Gardner, Reb. Prep.
Test. at 8, ff. Tr. 17606; Kaushal, Tr. 13717-21; Kostal, Tr.
14310-45, 15146048.)
In addition, S&L was aware that the BCAP Stone & Webster engineers, th'e Independent Expert Overview Group (IEOG) and the NRC Staff were reviewing its calculations and evaluations.
(Kostal and Thorsell, Tr. 14939-46.
See also Gardner Tr. 17608-15, 18304-8.)
911.
BCAP Task Force activities, including the CSR, were overviewed by the BCAP QA group established within Appli-cant's Quality Assurance Department under the direction of Neil P.
Smith.
BCAP QA personnel were drawn.from Applicant, Gilbert Commonwealth, and Pittsburgh Testing Laboratories.
These individuals had no prior responsibility for construction activities at Braidwood.
(Kaushal, Reb. Prep. Test. at 10, ff Tr. 13068; Smith, Reb. Prep. Test. ff. Tr. 13068; Smith, Tr.
14195-7.)
The NRC Staff concluded that BCAP QA was " staffed with qualified individuals who performed in a highly satis-s-
v p,.,
e e
, =---
- 147 -
factory manner."
(Gardner, Reb. Prep. Test. at 13, ff.
Tr.
.17606; Staff Ex. 25 at 5.)
912.
In addition, Applicant established through the I
Evaluation Research Corporation an independent expert overview group (IEOG) comprised entirely of individuals from outside of Applicant.
(Kaushal, Reb. Prep. Test. at 10-11, ft. Tr. 13068:
App. Exs. 128, 137.)
IEOG reviewed the work done by the BCAP Task Force, BCAP QA and S&L.
The NRC Staff concluded that "IEOG audits, reviews, and overinspections were timely and effective.
As a result they provided a valuable overview of I
all aspects of the BCAP program."
(Gardner, Reb. Prep. Test.
at 13, ff. Tr. 17606; Staff Ex. 25 at 5.)
913.
Finally, the NRC Staff assigned Ronald N. Gardner as an on-site inspector to monitor the implementation of the BCAP.
Gardner was at the Braidwood site from August 20, 1984 to June, 1985, during which time he performed daily inspections which focussed among other things on' key aspects of the CSR, including: (i) qualification of BCAP personnel; (ii) training of BCAP personnel; (iii) BCAP Quality Assurance; (iv) Indepen-dent Expert Overview Group; (v) adequacy of CSR procedures, checklists and instructions; (vi) performance of CSR reinspec-tions; and (vii) evaluation of discrepancies identified by CSR inspectors.
(Gardner, Reb. Prep. Test. at 3-7, ff. Tr. 17606; l
Staff Ex. 25 at 6.)
Other NRC Staff personnel (whose conclu-l l
sions were not admitted as evidence in this proceeding) reviewed the design of the BCAP, including the CSR, and re-l viewed Sargent & Lundy's design significance calculations
- 148 -
involving stress analysis.
(Staff Ex. 25 Section I, p. 2; Gardner Tr. 17571-72, 17608-10, 18304-8.)
The NRC Staff concluded that the BCAP program was " satisfactorily imple-mented."
(Staff Ex. 25 at 6.)
CSR Sample Design and Selection 914.
The CSR was a sample program.
For purposes of taking samples, the entire population of on-site contractors' safety related construction work was divided into 30 "construc-tion categories," which were defined as groups of hardware constructed using similar processes or containing similar types of components.
Six of those construction categories contained electrical work: cables; cable pans, cable pan hangers, con-duit, conduit hangers, and electrical equipment installation.
(Kaushal, Reb. Prep. Test. at 11-13, ff. Tr. 13068.)
The total sample for each construction category consisted of three parts.
The first or " random" portion was chosen in such a manner as to support formal statistical conclusions with at least 95%
confidence and at least 95% reliability concerning each sampled construction category.
(Kaushal, Reb. Prep. Test.
at 13-16, ff. Tr. 13068; Frankel, Reb. Prep. Test. at 9-11, ff. Tr.
17082.)
For the second portion of the sample, engineering judgment was used to determine sample size and to select items.
This portion emphasized types of items which are part of safe-shutdown or emergency core cooling systems.
It was initially intended that the " engineering judgment" portion of the CSR sample would also emphasize the types of items which had previously exhibited deficiencies at Byron and Braidwood.
- 149 -
However, for each of the electrical construction categories it was determined that none of the previously identified deficien-
- cies could be limited to a subset of the construction category.
i Therefore, addit'ional items were chosen using random methods.
(Kaushal, Reb. Prep. Test. at 14-15, ff. Tr. 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 were added as the third part of the CSR sample.
2 (Kaushal, Reb. Prep. Test. at 15-16, ff. Tr. 13068; Kostal, Tr.
15074-75.)
Under the provisions of the CSR, if any design significant discrepancies had been found in the initial CSR sample,-the sample size would have been increased.
These sample expansion previsions could have led to a-100% reinspec-tion.
However, since no design significant discrepancies were identified no sample expansion proved necessary.
(Kaushal, Reb. Prep. Test. at 6-7, 16, ff. Tr.
13068; Kaushal, Tr.
13756-57, 14148-49.)
915.
One of the strengths of the CSR sampling program was that it included both a probability (random) sample and a non-probability (engineering-judgment) sample.
(Frankel, Reb.
Prep. Test, at 11, ff. Tr. 17082; Frankel, Tr. 17191.)
The inclusion of items selected on the basis of engineering judg-ment increased the size of the total sample and therefore enhances confidence.
(Kaushal, Reb. Prep. Test. at 18, ff.
Tr. 13068.
See also Frankel, Tr. 17147.)
Increased confidence also arises because this portion of the sample was deliberately
- 150 -
selected from items which were more likely to have design significant defects, if any existed.
(Kaushal, Reb. Prep.
Test. at 17-18, ff. Tr. 13068; Frankel, Tr. 17191-92.)
Thus the use of an engineering sample judgment is a more efficient strategy of seeking to capture rare defects (if one assumes design significant defects will be rare), than merely in-creasing the size of the random sample.
(Frankel, Tr. 17192.)
CSR Inspection Criteria and Inspector Performance 916.
The CSR inspection checklists and instructions were developed by the BCAP Task Force engineers based on relevant design information provided by S&L.
The attributes selected for reinspection were those that (1) are required by applicable codes and standards, (2) potentially have an effect on the item's ability to perform 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.)
Del George compared the CSR and Comstock checklists and concluded that these limitations are of no consequence in using the CSR results to assess the quality of safety-related work or the effectiveness of QC inspector performance.
(Del George, Reb.
Prep. Test. at 15-16, 21-25, 37 ff. Tr. 16740.
See also Thorsell, Reb. Prep. Test. at 18-19, ff. Tr. 14270.)
- 151 -
917.
BCAP Task Force Lead Inspectors, BCAP QA, the IEOG, and the NRC inspector, Gardner, all devoted a great deal of effort-towards assessing the adequacy of the CSR rein-spections.
(Gardner, Reb. Prep. Test, at 5, 6,
13-14; ff. Tr.
17606; Shevlin, Reb. Prep. Test. at 14-16, ff. Tr. 13068; Shevlin, Tr. 13825, Smith,' Reb. Prep. Test. at 4-5, 7-3 4, ff.
Tr.
13068.)
The BCAP Task Force inspectors were very con-scious of this intense scrutiny.
(Shevlin, Reb. Prep. Test, at 14-16, ff. Tr. 13068; Wozniak, Reb. Prep. Test. at 4-5, 9-10 ff.
Tr. 13068, Smith, Reb. Prep. Test. at 4-5, 7-14, ff. Tr.
13068; Gardner, Tr. 18369-71.)
Early in the CSR program the NRC Construction Assessment Team identified deficiencies on i
three of six pipe supports / restraints which the BCAP Task Force inspectors had previously reinspected.
The BCAP Task Force reinspectors had not identified these deficiencies.
In addi-tion, IEOG overinspections identified deficiencies on two conduit hanger installations and four deficiencies associated with a concrete placement which had not been identified during g
the BCAP CSR reinspections.
(Gardner, Reb. Prep. Tent. at 8-9, ff. Tr. 17606.)
In response to these findings and following a meeting with Gardner on January 23, 1985, 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 4
- 152 -
reinspection of previously reinspected mechanical pipe sup-ports, 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 instruc-tions, and the initiation of the BCAP QA overview of ECAP Task Force inspections.
These corrective actions were proven effective during subsequent NRC, BCAP QA, and IEOG overviews of BCAP reinspection activities.
(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.)
918.
Kaushal believed that the root cause of the CSR reinspection errors identified by the CAT and the IEOG prior to January 23, 1985 was a misunderstanding by the BCAP Task Force inspectors of certain attributes on their checklists.
(Kaushal, Tr.
13941-42.)
This view is supported'by Wozniak's testimony (only one error represented an inspector "miss" as opposed to an interpretation error) as well as by a detailed review of App. Ex. 135.
(Wozniak, Reb. Prep. Test. at 5-7, ff.
Tr. 13068.)
Gardner, on the other hand, concluded that the root cause of these CSR reinspection deficiencies was the fast pace at which the BCAP Task Force inspectors were working.
(Gardner, Reb. Prep. Test. at 10, ff. Tr. 13068.)
Although Gardner did not discuss this concern with Kaushal in their meeting on January 23, 1985, or document it in his inspection reports, he continued to monitor BCAP inspectors' attitudes and I
- 153 -
instructions.
(Gardner, Tr. 18369.)
Subsequently the CSR reinspectors were instructed to disregard any pace concerns and take as much time as necessary to perform their inspections.
(Gardner, Tr. 17623-24.
See also Int. Ex. 285 (documenting Kaushal meetings with BCAP inspectors and engineers).)
In Gardner's judgment these instructions, and the signal sent to the entire BCAP organization by the suspension of reinspec-tions, had the effect of reducing the pace of CSR reinspection activities and increasing the quality of subsequent CSR rein-spections.
(Gardner, Reb. Prep. Test. at 10, ff. Tr. 17606; Gardner, Tr. 17623-24.)
Processing of CSR Observations 919.
All CSR reinspection observations reported by the BCAP Task Force inspectors were reviewed by their lead disci-pline inspectors for clarity, completeness and accuracy.
(Kaushal, Reb. Prep. Test. at 22-23, ff. Tr. 13068.)
If suitable for further processing, the observations were evaluat-ed for validity by BCAP engineers.
Under BCAP procedures, (Int. Ex. 143)
C5R observations which had previously been identified by Applicant or its contractors on an existing nonconformance report or other controlled system were consi-dered to be invalid.
Ccnditions 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; Smith, Tr. 13588-603.)
Observations that related to items not within the CSR sample or attributes not on the CSR check-
- 154 -
lists were declared "out of scope."
In 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 l
through existing procedures or other documented plans for l
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; Kaushal, Tr. 13535-38, 13799-802; Int. Ex. 143.) The remaining (valid, in-scope) observations were termed "discrep-ancies" and were transmitted to Sargent & Lundy for evaluation of design significance.
(Kaushal, Reb. Prep. Test.
at 25-26, ff. Tr. 13068.)
920.
Early in the CSR program, Sargent & Lundy engi-neers reviewed each discrepancy sent to them by BCAP for validity as well as for design significance.
However, in March 1985 the NRC inspector Gardner assessed an item of noncompli-ance 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 concern was the invalida-tion itself, rather than S&L's role, after this time BCAP-06 was modified to emphasize that S&L could only recommend inval-idation 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.
J
- 155 -
Test. at 25, ff. Tr. 13068, Kaushal, Tr. 13489-503, 13828-34; Kostal, Tr. 14343-45; Thorsell, Tr. 14476-77[)
921.
Following this early NRC item of noncompliance, all observations that were invalidated had to be reviewed and concurred in by BCAP QA.
(Kaushal, Reb. Prep. Test, at 24-25, ff. Tr. 13068; Smith, Reb. Prep. Test. at 14-15, ff. Tr. 13068; Gardner, Reb. Prep. Test. at 11-12, ff. Tr. 17606.)
The IEOG and Mr. Gardner also reviewed many of these invalidations.
(Gardner, Reb. Prep. Test. at 11-12, 13, ff. Tr. 17606.)
In addition, it was BCAP policy that each time an observation was determined to be invalid it was routed back to the BCAP in-spector who originally conducted the inspection (or in his absence, his supervisor) for acknowledgment, concurrence and comments.
(Kaushal, Reb. Prep. Test, at 23-24, ff. Tr. 13068; Kaushal, Tr. 12974-76; Shevlin, Reb. Prep. Test, at 12-14, ff.
Tr. 13068.)
The BCAP inspectors required proof to support their agreement with invalidations.
(Shevlin, Reb. Prep. Test.
at 12-14, ff. Tr. 13068; Wozniak, Reb. Prep. Test. at 7-9, ff.
Tr. 13068; Wozniak, Tr. 13949-52.)
Whenever an inspector did not concur with an invalidation, a committee was appointed including the responsible BCAP assistant director to review the observation, and based on the committee's recommendation the BCAP Director Dr. Kaushal made the final determination of validity or invalidity.
(Kaushal, Reb. Prep. Test. at 24-25, ff. Tr. 13068; Kaushal, Tr. 13505.)
The record includes each case in which a CSR observation within the electrical construc-tion categories was invalidated despite the nonconcurrence of
- 156 -
the BCAP inspector.
In each case the explanation for the invalidation is reasonable.
(Int. Ex. 150, Kaushal, Reb.
Prep. Test, at 24-25, ff. Tr. 13068; Kaushal, Tr. 13505-33, 13558-63, 13943-53; Wozniak, Reb. Prep. Test, at 9, ff. Tr.
13068.)
922.
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 while reduced an item's capacity by less than 10% but did not impair its ability to perform its safety-related design function were termed "in-significant."
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 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; Kaushal, Reb. Prep. Test. at 28, ff.
Tr. 13068.) Sargent & Lundy's evaluation of discrepancies for each of the six electrical construction categories found no design significant discrepancies.
923.
Sargent & Lundy also performed an evaluation of CSR discrepancies to identify discrepancies which, while not design significant, were repetitive in nature and might in-dicate a design significant problem in the unreinspected
- 157 -
portion of the plant.
(Kaushal, Reb. Prep. Test. at 5-6, ff.
Tr. 13068; Kostal, Tr. 14363-64;.Thorsell, Tr. 14998-99; Int.
Ex. 161.)
This analysis had three components:
(1) tabulation by attribute of CSR results and further evaluation of any attribute in which more than a 1% trend of notable discre-pancies was identified; (2) tabulation by attribute of CSR results and further evaluation of any attribute in which more than a 5% trend (in the case of non-welding attributes) or 10%
trend (in the case of welding attributes) of notable and in-significant discrepancies was identified; and (3) assessment of each discrepancy to determine whether it had potential for design significance had it occurred at other locations.
(Thorsell, Tr. 14999; Kostal, Tr. 14363-64.)
BCAP made two recommendations based on this analysis.
First, it recommended a review of the attachment of conduits to supports, and second, a review of structural steel supporting electrical equipment to determine whether stiffeners were installed where required.
(Kaushal, Reb. Prep. Test at 29-33, ff. Tr. 13068; Kostal, Tr. 14365-14366; Thorsell, Tr. 14367-68, 15000-01.)
These recommendations were incorporated in subsequent walkdown programs.
(Kaushal, Reb. Prep. Test. at 30-31, ff. Tr. 13068.)
924.
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, l
- 158 -
14477-90, 14565-66; Int. Ex. 155, 155A, 155B, App. Ex. 159, 173, 180.)
Sargent & Lundy initially calculated the design margin for CPH 104, taking into account CSR identified weld discrepancies, to be 1.03.
(Kostal, Tr. 14781-83; Int.
Ex. 155B, pp. 14-15.)
Any value equal to or greater than 1.0 is not design significant and therefore acceptable.
(Kostal, 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 used in the initial calculation.
(Kostal, Tr. 14756, 14784-85, 15181-2; App. Ex. 159.)
That calculation resulted in a design margin of 1.89, but a shortcut was taken in the second calcula-tion.
Correcting for the shortcut, the design margin is 1.28.
(Kostal, 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 first specifically pointing that cable out to the manufacturer's representative, or providing a written description of the bend radius violation to the manufacturer.
(Thorsell, Tr. 14482-83.)
However, the circumstances of the error were unique, (Thorsell, Tr. 14482-84, 14568-69, 14609.)
Cable 130 in fact met the technical acceptance criteria in
'i Okonite's letter, (Int. Ex. 162; Thorsell, Tr. 14483), and the cable was independently shown to be acceptable based on the results of a Meggar (insulation resistance) test.
(Thorsell, Tr. 14485, 14609.)
- 159 -
925.
One of the major reasons why none of the CSR discrepancies was found to be design significant was that S&L's design includes substantial margins over and above code re-quirements.
(See Thorsell, Reb. Prep. Test. at 11-15, ff.
Tr. 14270; Kostal, Reb. Prep. Test. at 17-18, App. Ex. 179.)
These design margins arise due to the standardization of components and due to the recognition that construction work is not always perfect.
(Thorsell, Reb. Prep. Test. at 14-17.)
For example, there are approximately one dozen standard designs for conduit hangers at Braidwood.
In reducing the large number of possible designs to about one dozen, a composite of design rules based.on maximum conduit size, maximum cable weight, maximum hanger length and maximum span between hangers was used.
The resulting hangers are able to accommodate all of these conditions within code allowable standards, even though all of these conditions are seldom, if ever, present.
Similarly, peak seismic accelerations are used for designing conduit, conduit hangers, cable pan, and electrical equipment mountings even though mote refined analyses could be performed based on the actual component frequencies which would result in lower seismic design forces.
In addition, manufacturers typically supply materials which exceed minimum strength requirements to avoid the potential expense of scrapping substandard material.
These and other conservatisms overlap and are often cumulative.
(Id.; Kostal, Reb. Prep. Test, at 24-29, App. Ex. 179.)
l
- 160 -
926.
In quantifying the design margin for any compo-nent, the result depends on the method of analysis used.
S&L has developed simplified, conservative methods of analysis to minimize architect / engineering costs for the design of Braid-wood.
More refined calculations are generally available which, by eliminating these conservatisms, will demonstrate higher design margins than initially calculated.
(Kostal, Reb.
Prep.
Test, at 24-29, App. Ex. 179; Kostal, Tr. 15185.)
These refined calculations normally are performed from time to time as plant construction progresses and new load are added to a component, or the component is built differently from shown in the design drawings, or field problems (such as weld disc-repancies) are identified.
Each time S&L will try to use the simplest method of analysis which will show that the affected components remain within Code allowables.
(Kostal, Tr. 15084, 15185.)
Thus S&L's design process for Braidwood is fluid and evolutionary rather than static.
(Kostal, Tr. 15076, 15083-85.)
S&L's responsibility is to show at the end of construction that all electrical components and equipment within its scope of work satisfy code requirements.
(Kaushal, Tr. 13692; Kostal, Tr. 14343-44; App. Ex. 162.)
927.
The standards and methods of analysis used by S&L for the evaluation of CSR discrepancies are the same standards available from the beginning of the project.
(Kostal, Tr.
15174-76.)
Many of these refinements and conservatisms are l
still available and have not yet been used in evaluating CSR discrepancies for design significance.
In particular, S&L's
- 161 -
structural design procedures for BCAP did not require design margins to be calculated when a CSR discrepancy reduced the capacity of a component by 10% or less.
This was based on an engineering judgment that such insignificant discrepancies could easily be accommodated within code allowables by the existing margins and available analytical refinements which were described at length in these hearings.
(See Kostal, Tr.
14399, 15091-122, 15132-36.)
928.
In addition to the design margins available for purposes of evaluating design significance, there is a second kind of design margin applicable to the Braidwood electrical work.
This is code-required margins.
Generally, code writers attempt to obtain a margin of about 2 between code allowable stresses and failure of a component.
By definition, any discrepancy which is not design significant does not encroach upon these code-required margins.
(Kostal, Reb. Prep. Test. at 16-18, App. Ex. 179.)
929.
The QC inspection of an item such as a cable or a cable pan hanger requires the inspector to verify that the l
item confo'rms to design requirements for each attribute on his l
l checklist.
Verification of each such attribute may require one or more inspection judgments.
(Kaushal, Reb. Prep. Test. at l
19-20, ff. Tr. 13068; Shevlin, Tr. 13761-62.)
Moreover, the i
items included in the CSR sample varied greatly in their complexity and thus in the number of inspector judgments required for the initial QC inspection and for the CSR rein-spection.
(Kaushal, Tr. 13758-59; Wozniak Tr.
14166-73; App.
- 162 -
Ex. 143, App. Ex. 144.)
To permit meaningful judgment of inspector performance and meaningful comparison of inspector performance with respect to items of differing complexity, the BCAP Task Force together with BCAP QA and Sargent & Lundy developed the concept of " inspection points" and " discrepancy points."
(Kaushal, Tr. 13758-59; Shevlin, Tr. 13770, 23773; Kaushal, Shevlin, Wozniak and Smith, Tr. 14173-79.)
Each inspector check to determine the acceptability or rejectability of an item or an attribute was identified and termed an "in-spection point."
Each inspection point which resulted in a CSR discrepancy was termed a " discrepancy point."
(Kaushal, Reb.
Prep. Test. at 19-20, ff. Tr. 13068; Shevlin, Tr. 13760-63; Smith, Tr. 13763-64; Kostal, Reb. Prep. Test, at 13-14, ff. Tr.
14270.)
On this basis, over 98% of the inspection points were found to be correct (nondiscrepant) and more than 2/3 of the discrepancy points were insignificant.
(App. Ex. 179 at 16; Thorsell, Reb. Prep. Test. at 11, ff.
Tr. 14270; Kostal, Reb.
Prep. Test. at 22, ff. Tr. 14270.)
930.
Applicant also presented the CSR results for the electrical construction categories on a per. weld basis.
About 85% of the welds had no discrepancies.
(Del George, Reb. Prep.
Test. at 38, ff. Tr. 16740.)
Applicant's weld inspection expert, Mr. Shevlin, did not find this result unusual or alarm-ing.
(Shevlin, Tr. 13774-83, 14066-67.)
The comparable figure for the PTL overinspection data for the period July 1, 1982 to June 30, 1986 is 93%.
For the period in which the two data bases overlap, (July 1, 1982 to June 30, 1986) the agreement
- 163 -
rates are 89% and 90%, respectively, (Del George, Feb. Prep.
Test at 37-38, ff. Tr. 16740; Del George. Tr. 16801-01).
931.
A third way of looking at the CSR results was suggested by Intervenors in this proceeding.
Any item with one or more discrepancies would be termed a " discrepant item."
On this 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% of the cable pan hangers and 72.5% of the electrical equipment instal-lations would be deemed " discrepant items."
(App. Ex. 181.)
Applicant's witnesses did not view this as a reasonable or fair measure of Comstock QC 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 are strike, would count the same as a conduit wall strap support which was totally missing.
(Kaushal, Tr. 13758-59; Shevlin, Tr. 13770; Kaushal, Shevlin, I
Wozniak and Smith Tr. 14173, 14179.)
932.
NRC inspector Gardner agreed that in grading inspector performance he would not equate su'ch dissimilar
" discrepant items."
Moreover Gardner did not believe that App.
Ex. 181, standing alone, 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.
Gardner stated that his own personal standards were high, and he would expect a good inspection program would have resulted in lower rates of l
- 164 -
discrepant items than shown in App. Ex. 181.
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" format.
(Gardner, Tr. 17633-45, 17807-11, 18347-49.)
In Gardner's view, the Comstock Quality Control inspectors were not effective in the " classical" 10 CFR Appendix B sense of identifying all defects, but they were effective and adequate in the sense that they did not miss any design significant defects.
(Gardner, Tr. 17807-9; 17814-15.)
933.
For purposes of this proceeding, Applicant created a "CSR Data Base" which shows for each electrical item reinspected under the CSR, the Comstock QC inspectors who performed the original inspection, the date on which the item was QC accepted, and the results of the CSR reinspections.
(Kaushal, Reb. Prep. Test. at 34-35, ff. Tr. 13068; Kaushal, Tr.
12977-92.)
The CSR results are presented both on an inspection point / discrepancy point basis and on a welds in-spected/ welds found discrepant basis.
(See App. Ex. 178, Del George, Reb. Prep. Test. Attachments 2c (Del. George -1 through-6) Rev.
2, ff. Tr. 16740.)
The creation of this data base required a search by Applicant of Comstock QC inspection records to identify the responsible Comstock QC inspectors and the dates of their inspections, a recount and retabulation of CSR results by S&L to identify discrepancy points and discrepant welds in an appropriate format, and an apportionment of these CSR reinspection results to the responsible Comstock t
- 165 -
QC inspectors.
(Kaushal, Tr. 12977-92; Kostal, Reb. Prep.
Test. at 13-14, App. Ex. 179.)
934.
The CSR data base includes E.C. Ernst inspec-tions, which are distinguishable from Comstock inspections by the date indicated.
(Kaushal, Tr. 12980-81.)
The data base includes Systems Control welding because it was not possible to distinguish reliably between cable pan hangers supplied by Systems Control and those fabricated on-site by Comstock.
(Kaushal, Tr. 13955-60; Orlov, Tr. 14245-50.)
The data base excludes items for which no Comstock QC records could be found.
(Kaushal, Tr. 12986, 12989.) About 10% of the pertinent Comstock QC records could not be found, although this estimate is uncertain for several reasons.
(Kaushal, Tr. 13057-65; Orlov, Tr. 14206-7, 14215-17.)
Where more than one Comstock inspection record existed for the same item, the CSR results were counted for or against each Comstock inspector, unless there was some clear basis for apportionment, or unless there was a clear indication that rework had occurred, invalidating prior inspections.
(Kaushal, Tr. 12985-87, 12988-89; Orlov, t
Tr.
13986-87, 14211-15, 14218-24, 14250.)
(Kaushal, Tr.
12977-92; Kostal, Reb. Prep. Test, at 13-14, ff. Tr. 14270.)
935.
Errors were made in compiling the CSR data base, l
particularly in S&L's counting of weld discrepancy points and discrepant welds.
The CSR data base, Attachment 2C (Kaushal-4), was originally filed with the Licensing Board and the parties in July, 1986.
Revisions were submitted by letter i
l dated September 16, 1986 and September 30, 1986.
The final t
- 166 -
revision (admitted as App. Ex. 178) in dated October 31, 1986.
The reasons for these changes are summarized in App. Ex. 140, App.
Ex. 141, and App. Ex. 176.
(See also Orlov, Tr. 13987-95; Kostal, Tr. 16666-68.)
These errors were corrected and the data base admitted into evidence is accurate.
(Kaushal, Tr.
16668-69, 16682, 16694; Kostal, Tr. 16666-68; orlov, Tr. 16669, 16681-82.)
i 936.
For purposes of evaluating the CSR data base, Applicant's witnesses used " agreement rates" as a reasonable measure of Comstock QC inspector effectiveness.
Agreement rates were calculated in two ways.
First, agreement rates were calculated on the basis of inspection points found nondiscrepant
-by the CSR reinspectors divided by the total number of inspection points reinspected for the same time period.
Alternatively, for CSR weld results (and for the PTL overinspection results)
}
agreement rates were calculated as the number of nondiscrepant welds divided by the total number of welds reinspected for the same time period.
Agreement rates are an imperfect measure of inspector accuracy, because they do not take into account good product erroneously rejected by the Comstock QC inspectors.
(Shevlin, Tr. 13774-78, 14066-67), because the CSR and PTL reinspectors were fallible, (Shevlin, Tr. 14067-69, 14159-61, 4
l 14184) and because agreement rates are influenced by the quality of the work being inspected and reinspected (i.e.
the craft labor error rate).
(Del George, Tr. 16809-11.
See also Int. Ex. 188; App. Ex. 182, 183, 184.)
Nevertheless, agreement rates are a reasonably good indicator of product quality found l
l
- 167 -
~
by the overinspector.
(Del George, Tr. 16946-47.)
More im-portantly, agreement rates are a good indicator of changes in inspector effectiveness over time.
(Del George, Tr. 16945-46.)
This is because there is no reason to believe that the accuracy of the CSR and PTL inspections is related to the dates of the underlying LKC QC inspections or the identities of the LKC QC inspectors.
(Del George, Tr. 16926-27, 16992-96.)
In addition, it is unreasonable to suppose that adverse effects in LKC QC inspector performance due to harassment, intimidation and undue production pressure would invariably be offset by improvements in LKC's craft labor performance, and vice versa, so as to mask continually over a period of years the effects of such harass-ment, intimidation and undue prodi.ction pressure.
(Del George,-
Tr. 16929-46; Frankel, Tr. 17085-92.)
937.
Frankel used objective statistical methods to determine whether the CSR data base agreement rates (calculated on an inspection point basis) differed for the periods before and after July 1, 1982 (the date that Saklak became a QC supervisor for Comstock).
He calculated that the overall 1
agreement rates before and after were 98.893% and 98.925%,
7 l
respectively.
This is not statistically significant dif-ference.
(Frankel, Reb. Prep. Test, at 12-19, ff.
Tr. 17082;
(
Frankel, Tr. 17084-85.)
Frankel also compared the periods l
before and and after August 1, 1983 (the date DeWald became i
l Comstock QC Manager).
Again, he found no statistically significant difference.
(Frankel, Reb. Prep. Test. at 20-21, l
ff. Tr. 17082; Frankel, Tr. 17084-85.)
Because these calcula-l l
- 168 -
tions were based on the portion of the CSR data which con-stituted a probability sample, the results can be projected to the entire population of all Comstock QC inspections covered by the CSR defined electrical construction categories (Frankel, Reb. Prep. Test. at 13, ff. Tr. 17082.) Frankel testified that the power of the test was in excess of 99% for <lifferences in inspection point agreement rates of 1% or more.
This means that there in a 99% probability that there is less than a 1%
difference in the agreement rates before and after July 1, 1982 (or before and after August 1, 1983) for the entire population of Comstock QC accepted work covered by the CSR.
(Frankel, Reb. Prep. Test. at 14-20, ff. Tr. 17082; Frankel, Tr. 17088, 17148-49.)
938.
Del George concluded that the variation in the CSR results over time does not reveal any apparent relationship between Comstock QC inspector performance on the one hand and the incidents of alleged harassment and intimidation and the periods of alleged intense production pressure on the other hand.
(Del George, Reb. Prep. Test. at 27-32, ff. Tr. 16740.)
Del George plotted, as a function of time, the specific allega-I tions of harassment, intimidation and production pressure mentioned in the depositions and testimony in this case.
(Int.
Ex. 191; Del George, Tr. 17020-41.)
He then assumed that each I
incident was 100% effective in that it would produce an effect on the quality of Comstock QC inspections.
(Del George, Tr.
I 16767.)
Five graphs attached to Del George's testimony depict the CSR agreement rates over time (1) on an inspection point
- 169 -
basis for all Comstock inspections; (2) on an inspection point basis for all welding sampled; (3) on a weld basis for all welding sampled; and (4) and (5) on an inspection point basis and on a weld basis for the group of 24 inspectors who ex-pressed concerns to the NRC on March 29, 1985.
(Del George, Reb. Prep. Test at 18-19, and Attachments 2C (Del George -1 to
-5),
ff. Tr. 16740.)
Del George concluded that to the extent that there is sufficient data to draw a conclusion, there is no apparent trend in the CSR data attributable to the alleged undue pressure, harassinent and intimidation.
(Id. at 19-21; Del George, Tr. 16905-10.)
939.
The Pittsburgh Testing Laboratory (PTL) over-inspection program was established at the direction of Walter Shewski, Applicant's QA Manager, in early 1977 to keep ahead of known industry problems such as deficient-QC inspections and failure of welds on the Alaskan pipeline project.
PTL is the Braidwood Independent Testing Agency which reports directly to Applicant's site QA Department.
(Marcus, Reb. Prep. Test. at 5 ff. Tr. 15568.)
Applicant presented the PTL overinspection results for Comstock for the period July 1,.1982 to June 30, 1986.
(Marcus, Reb. Prep. Test, at 14 ff. Tr. 15568; Del l
George, Reb. Prep'. Test. at 13 ff. Tr. 16740.)
940.
The PTL overinspection program consists of routine, continuing, " random" (but not statistically random) inspections of a percentage of contractors', completed QC accepted work.
(Marcus, Rob. Prep. Test at 11, ff. Tr. 15568; l
Frazier, Tr. 15807.)
The inspections performed in this program l
.,0 are called "overinspections" to distinguish them from the "first line" QC inspections performed by the contractors' QC inspectors.
Prior to June 1985, PTL's overinspection of Comstock was limited to visual reinspection of welds.
(Marcus, Reb. Prep. Test. at 8 ff. Tr. 15568.)
Since June 1985, PTL has expanded its overinspection of Comstock to incluele objective type inspections such as conduit / wireway installation, equip-ment installations, and junction box installation.
(D' Antonio, Reb. Prep. Test. at 11, ff. Tr. 15568.)
941.
From July 1982 through June 1986, PTL was over-inspecting a minimum of 10% of the welds accepted by Comstock, although Applicant's QA Department has varied the percentage several times.
(D' Antonio, Reb. Prep. Test., Attach 2 ff. Tr.
15568; Marcus, Reb. Prep. Test. at 8-9 ff. Tr. 15568.) Several brief suspensions in PTL's overinspection of Comstock QC accepted welds also occurred during that period: late Septem-ber, 1982 through November, 1982; late October, 1983 until the end of 1983; (Marcus, Reb. Prep. Test. at 15 ff. Tr.
15568; D' Antonio, Reb. Prep. Test. at 10 ff. Tr. 15568) and the middle of December, 1985 to beginning of January, 1986.
(D' Antonio, Tr. 15730.)
In addition to these suspensions, there is minimal overinspection data for May 1985 due to a transition in the PTL program.
(Marcus, Reb. Prep. Test. at 16 ff. Tr. 15568.)
942.
During the period of July, 1982 to June, 1985, upon completing QC weld inspections accepting a component or components, Comstock would send to PTL a form called an "in-spection request" which identified the components, the number
- 171 -
of welds in each individual component, and the location of the installations.
(Frazier, Reb. Prep. Test. at 13 ff. Tr. 15568; Forrest, Tr. 15576, Tr. 15837-39.)
From July 1982 to Novem-ber 1984, PTL would overinspect 10% of Comstock's QC accepted welds on each Comstock inspection request, rounding up where necessary.
(Frazier, Reb. Prep. Test. at 13 ff. Tr. 15568.)
Although the inspector was free to select the specific welds to make up the 10% sample, he was also told that if, in going to 4
the component, he noticed a bad or marginal weld, he had to include that weld in his sample.
(Id.)
If PTL found any weld to be rejectable, PTL would overinspect an additional 15%
sample of the welds shown on the relevant inspection request.
If any weld in the expanded sample was rejectable, PTL would then overinspect 100% of the welds shown on the inspection a
request.
(Forrest, Reb. Prep. Test. at 8-9 ff. Tr. 15568.)
PTL would notify the contractor of any discrepancies PTL had identified.
The contractor would then handle the identified discrepancies in accordance with its program for dispositioning nonconformances.
(D' Antonio, Tr. 15678; Forrest, Tr.
15679-80.)
943.
From November 1984 through May 1985, PTL's selection process was based on components rather than total welds.
A PTL employee would log in all Comstock inspection requests received.
From this log, the PTL Lead Inspector would select the specified percentage of components and PTL in-spectors would inspect 100% of the welds in those selected components.
(Forrest, Reb. Prep. Test. at 12 ff. Tr. 15568.)
- 172 -
944.
In June of 1985, Applicant's QA Department modified the overinspection program to ensure that PTL over-inspected the specified percentage of welds for each active Comstock QC inspector.
(D' Antonio, Reb. Prep. Test. at 12 ff.
Tr. 15568.)
Under the new method, PTL reviewed Comstock inspection reports on a daily basis and selected samples of the work performed by each inspector.
(Marcus, Reb. Prep. Test. at 11 ff. Tr. 15568.)
In addition, Applicant began routinely to computerize the overinspection data, (D' Antonio, Reb. Prep.
Test. at 11 ff. Tr. 15568) required overinspection within 5 working days after PTL received the Comstock inspection re-quest, (D' Antonio, Reb. Prep. Test. at 11 ff. Tr. 15568) and added a provision for third-party review by Sargent and Lundy Level III inspectors of any discrepant conditions which PTL identified.
(Marcus, Reb. Prep. Test. at 16-17 ff. Tr. 15568.)
i 945.
PTL performed its overinspections of visual welds
- o the requirements of AWS D1.1.
The PTL and Comstock weld inspection procedures have only minor differences, with the exception that prior to June 1985, Applicant's QA Department authorized PTL to overinspect welds through paint.
(Del George, Reb. Prep.
Test. at 7, 16, 37 ff. Tr. 16740; Marcus, Tr. 15,739; Forrest and Frazier, Reb. Prep. Test. at 9-10, ff.
Tr. 15568; Forrest, Tr. 15734.)
A review of PTL's over-inspection records for July 1982 through June 1986 determined that PTL overinspected 7% of the Comstock welds through paint.
(Marcus, Reb. Prep. Test. at 10 ff. Tr. 15,568.) Forrest and Frazier testified that it was the PTL's practice (although not
- 173 -
a written procedural requirement) to indicate in the remarks section of PTL's inspection report which welds were over--
inspected through paint, (Frazier, Reb. Prep. Test. at 9 ff.
Tr. 15568; Frazier, Tr. 15750, 15827;-Forrest, Tr. 15781-82),
and they believed that the PTL inspectors followed this prac-tice whenever they inspected through paint.
(Forrest, Tr.
15781-82; Frazier, Tr. 15827-28.)
The new Visual Weld Accep-tance Criteria for Structural Welding at Nuclear Power Plants
'(VWAC), which was' approved by the NRC staff in 1985, states that overinspection of welds through paint may be permitted.
1 (Marcus, Reb. Prep. Test, at 9 ff. Tr. 15568.)
In any event, there is no substantial difference in the PTL overinspection results if the welds overinspected through paint are excluded.
'(Marcus, Tr. 15826, 15829; App. Ex. 165.)
946.
All PTL inspectors involved in the overinspection program since July 1982 have been certified to ANSI N45.2.6 (1978) as Level II inspectors.
.(Forrest, Tr. 15630; Frazier, i
Tr. 15630.)
As an outgrowth of Byron, Applicant undertook the Braidwood Quality Control Inspector Reinspection Program (QCIRP) to reconfirm the effectiveness of the contractors' programs for certification and qualification of quality control inspectors prior to the fall of 1982.
The Braidwood QCIRP l
looked at samples of the inspectors' first 90 days of inspec-l l
tion work prior to the fall of 1982 to measure the effective-ness of PTL's certification program.
(Forrest, Tr. 15634-37.)
The overall agreement rate for the PTL weld inspectors involved in the QCIRP was about 88.5%.
- 174 -
tors, the overall agreement rate was 92%).
(Marcus, Tr.
15837.) Marcus testified that he does not consider the PTL result to be unacceptable (Marcus, Tr. 15892), and that there was no significant difference between PTL's performance and Comstock's.
(Marcus, Tr. 15837.)
In addition, no design significant discrepancies were identified for the welds which had been QC accepted by the PTL inspectors.
(Int Ex. 181 at 14 of 23.)
Marcus believed that PTL inspectors have sub-sequently performed at least as well, and most likely better.
(Marcus, Tr. 15836.) This is because between the fall of 1982 and the present,.PTL has continuously upgraded its qualifica-tion, attaining certification program, resulting in all PTL in all PTL inspectors receiving additional training, testing and recertification.
(Forrest, Reb. Prep. Test. at 6 ff. Tr.
15568; Marcus, Tr. 15836.)
Forrest and Frazier also testified quality of the PTL inspectors has increased since the fall of 1982, due to these upgrades.
(Forrest, Tr. 15836, Frazier, TR 15837.)
For the period of June 1985 through June 1986, Sargent and Lundy Level III inspectors ultimately accepted 40% of the welds rejected oy PTL, which indicates that the PTL overin-i spectors were conservative in rejecting questionable welds.
(Marcus, Reb. Prep. Test, at 17, ff. Tr. 15568.)
947.
In June of 1986, Marcus directed that the over-inspection results from July 1982 through June 1986 be combined into one data base which correlates the PTL overinspection results with the names of the Comstock QC inspectors whose "first line" inspections were overinspected and with the dates
- 175 -
of such "first line" inspections.
For the period June 1985 to June 1986 this information had already been entered into the computer.
However, for the period of July 1982 through May 1985 it was necessary to gather the pertinent information from PTL and LKC records and enter it into the computer.
(Marcus, Reb. Prep. Test. at 6 ff. Tr. 15568; Marcus, Tr.
15572; Forrest, Tr. 15837-40.)
The PTL overinspection data base created at Marcus' direction is accurate.
(Forrest, Reb.
Prep. Test. at 19 ff. Tr. 15568; Marcus, Reb. Prep. Test. at 6 ff.
Tr. 15568.)
948.
Frankel performed a linear regression analysis on the PTL overinspection data for the period.Tuly 1, 1982 to June 30, 1986.
That analysis showed a very slight increase in weld agreement rates over time.
Since the PTL overinspection data base is not a probability sample, any inferences based on this sample result to the entire population of Comstock QC accepted welding must be based on subject matter expertise, not statistics.
(Frankel, Reb. Prep. Test, at 25-27, ff.
Tr.
17082, Frankel, Tr. 17147-48.)
949.
Del George possesses the subject matter expertise described by Frankel.
(Del George, Reb. Prep. Test. at 1-7, ff. Tr. 16740.)
His testimony explicitly addressed the nature of the data being collected, the limitations of the each reinspection program, and the sufficiency of the information to support conclusions about QC inspector effectiveness.
His engineering review of the PTL overinspection data led him to conclude that no adverse trends actually exist in the popula-
- 176 -
tion from which the sample was chosen.
(Del George, Reb.
Prep. Test. at 7, 15-16, 19-20, 32-40, ff. Tr. 16740; Del George, Tr. 16905-10.)
950.
Del George reviewed the PTL overinspection data for the four year period (July 1, 1982-June 30, 1986) for all Comstock QC inspectors as a class and did not identify any apparent trends or changes over time which would support Intervenors' claims.
(Del peorge, Reb. Prep. Test. at 34, ff.
Tr. 16740.)
Del George also reviewed the PTL overinspection results for the 24 Comstock inspectors who made allegations to the NRC Staff on March 29, 1985.
(Id. at 35.)
A total of 20 of these 24 inspectors were captured in the PTL overinspec-tions, including 19 of the 20 welding inspectors.
(Id. at 17.)
Del George testified that his evaluation of the results for these inspectors did not identify any apparent trends or changes over time which would support Intervenors' claims.
The agreement rates for these inspectors for the months preceding March 29, 1985 are uniformly acceptable.
(Id. at 34.)
951.
Marcus examined the PTL Overinspection results for Comstock weld inspectors on a monthly basis, using 90% as a threshold for further investigation.
(Marcus, Reb. Prep. Test.
at 13 and Attachment 2C (Marcus-2), ff. Tr. 15568.)
Between July 1982 through June 1986, the agreement rate exceeded the 90% threshold for 38 of the 46 months during which PTL per-formed overinspections of welds accepted by Comstock QC in-spectors.
(Marcus, Reb. Prep. Test. at 17 ff. Tr. 15568.) For those 8 months where Comstock did not meet the 90% threshold,
- 177 -
Marcus examined the results of the individual Comstock in-spectors, and the relevant PTL inspection records to determine the cause.
(Marcus, Reb. Prep. Test. at 18-27 ff. Tr. 15568.)
952.
Marcus found that the dips in the monthly agree-ment rate were due to limited, isolated, technical inspection problems, rather than general deterf. oration in Comstock inspec-tor effectiveness.
The problems causing each monthly dip were never repeated.
Each inspector who contributed to a problem, did so only once.
There were no cases where a dip occurred two times in a row.
There were only three dips of any signifi-cance, each for totally unrelated reasons.
(Marcus Reb. Prep.
Test. at 27, 17-26 ff. Tr. 15568.)
The first of these was September,1982, in which the overall agreement rate dipped to about 76% due to numerous weld discrepancies on two control panels inspected by Comstock QC inspector Brown.
This was an unusual case in which the PTL overinspection followed the Comstock QC inspection by almost three years, and the weld discrepancies were due to a change in the weld acceptance criteria for these panels.
(Id. at 18-19.)
In March, 1983 the overall agreement rate dipped to about 75% due to numerous weld discrepancies on three remote shutdown panels inspected by Comstock QC inspector Holley.
Holley incorrectly accepted welds with " start /stop problems" because despite the slag, overlap and incomplete fusion at the ends of the welds, he l
determined that a sufficient length of each weld was accept-able.
(Id. at 20-22.)
Finally, in March, 1986 the overall l
agreement rate dipped to about 85% due to two inspectors, Arndt l
l L
- 178 -
'and Hunter, who improperly inspected welds through rust and paint.
(Id. at 26.)
953.
Marcus and Del George also reviewed the overin-spection results for each of the 100 inspectors who were included in the overinspection program (Id. at 27.) and found no pattern of poor performance for any inspector that might be attributable to the effects of harassment and intimidation.
(Id. at 30.)
Of the 78 Comstock QC inspectors for whom suf-ficient data exists, the PTL demonstrates the acceptability of the work of 75 inspectors.
The performance of only three 7
inspectors (Asmussen, Arndt and Hunter) was unacceptable.
Arndt and Hunter had previously been discharged for inspecting welds through rust and paint (see FF 340-63), and Comstock was directed to reassess Asmussen's training and certification, and PTL's overinspection of Asmussen's work was increased to 50%.
(Id. at 27-29; Del George, Reb. Prep. Test. at 35-36, ff.
Tr. 16740.)
954.
Marcus concluded the quality of the inspections performed by the Comstock QC inspectors was consistently acceptable and, therefore, whatever problems there were at Comstock, whether they included harassment, intimidation or undue production pressure, these problems did not result in any significant adverse inpact on the quality of the inspection work performed by Constock QC inspectors.
(Marcue, Reb. Prep.
Test. at 33 ff. Tr. 15568.)
i
.w.-
.---- --~.- - - - -
I DOLMETEF 39WL UNITED STATES OF AMERICA NUCLEAR REGULPsTORY COMMISSION
'87 JM -8 P4 32 1
BEFORE THE ATOMIC SAFETY AND LICENSING BOARD 0FFin _ -
. 5x 80CXET N 51 F. J L In the Matter of
)
BPMOt
)
COMMONWEALTH EDISON COMPANY
)
Docket Nos.
50-456
)
50-457 (Braidwood Station Units
)
)
1 and 2)
)
CERTIFICATE OF SERVICE I hereby certify that copies of Applicant's proposed Partial Initial Decision, Findings of Fact and Conclusions of Law and Applicant's Motion for Extension of Page Limit for Brief were served on the persons listed below by deposit in the United States Mail, first-class postage prepaid, this 5th day of January, 1987.
Copies are being sent by expedited means to Counsel for the NRC Staff and Interviews and the Licensing Board.
1 Herbert Grossman, Esquire Ms. Bridget Little Rorem Chairman 117 North Linden Street t
Administrative Law Jadge P.O. Box 208 Atomic Safety and Licensing Essex, IL 60935 Board 1
l U.S. Nuclear Regulatory
(
~ Commission Atomic Safety and Licensing l
Washington, D.C.
20555 Board Panel U.S. Nuclear Regulatory Dr. Richard F. Cole Commission Administrative Law Judge Washing, D.C.
20555 Board US.S Nuclear Regulatory Atomic Safety and Licensing Commission Appeal Board Panel Washington, D.C.
20555 U..S Nuclear Regulatory Commission Dr. A. Dizon Callihan Washington, D.C.
20555 Administrative Law Judge 102 Oak Lane l
Oak Ridge, TN 37830
' Stuart Treby, Esquire William L. Clements Elaine I. Chan, Esquire Chief, Docketing and Services Office of the Executive U.S. Nuclear Regulatory Legal Director Commission U,.S Nuclear Regulatory Office of the Secretary Commission Washington, D.C.
20555 Washington, D.C.
20555 William Little Robert Guild, Esquire Director, Braidwood Project Douglass W. Cassel, Jr., Esquire Region III Timothy W. Wright III, Esquire
'U.S. Nuclear Regulatory BPI Commission 109 North Dearborn Street 799 Roosevelt Road Suite 1300 Glen Ellyn, Illinois 60137 Chicago, Illinois 60602
-+
Jo
~
Gdllo '
of the attorneys for Commonwealth Edison Company 4
4 l
1
.. _. _ _ _ _. _ _... _. - _ _ _ _ _ _. _ _ _ _ _ _ _ _ _ _ _ _