ML20079R134
| ML20079R134 | |
| Person / Time | |
|---|---|
| Site: | Diablo Canyon  |
| Issue date: | 06/17/1983 |
| From: | Kaufman P CALIFORNIA, STATE OF |
| To: | NRC ATOMIC SAFETY & LICENSING APPEAL PANEL (ASLAP) |
| Shared Package | |
| ML20079R137 | List: |
| References | |
| ISSUANCES-OL, NUDOCS 8306220045 | |
| Download: ML20079R134 (45) | |
Text
.
O s
C,.
I Jutt aa 1983 >
'R s
T 10 1
i UNITED STATES OF AMERICA
/
$d!'%s
'E l
NUCLEAR REGULATORY COMMISSION r?
g -
,s BEFORE THE ATOMIC SAFETY AND LICENSING APPEAL BOA.
K 6 I
)
In the Matter of
)
)
Docket Nos. 50-275 0.L.
PACIFIC GAS AND ELECTRIC COMPANY )
50-323 0.L.
)
j (Diablo Canyon Nuclear Power
)
l Plant, Units 1 and 2)
)
)
REPLY OF GOVERNOR GEORGE DEUKMEJIAN TO THE RESPONSES OF THE NRC STAFF AND PACIFIC GAS AND ELECTRIC COMPANY TO THE GOVERNOR'S MOTION TO REOPEN THE RECORD ON CONSTRUCTION QUALITY ASSURANCE JOHN K. VAN DE KAMP, Attorney General h
of the State of California
.9 ANDREA SHERIDAN ORDIN, Chief Assistant Attorney General MICHAEL J. STRUMWASSER, Special Counsel to the Attorney General SUSAN L.
- DURBIN, PETER H. KAUFMAN, Deputy Attorneys General 3580 wilshire Boulevard, Suite 800 Los Angeles, California 90010 Telephone:
(213) 736-2130 Attorneys for Governor George Deukmejian 8306220045 830617 PDR ADOCK 05000275 Q
O PDR c
TOPICAL INDEX Pages I
INTRODUCTION 1
II THIS BOARD HAS BEFORE IT SIGNIFICANT NEW EVIDENCE OF VIOLATIONS OF NRC SAFETY REGULATIONS GOVERNING THE DIABLO CANYON CQA PROGRAM 4
A.
Neither PG&E Nor the NRC Staff Can Escape the CQA Implications of the Design QA Failures 4
B.
The Latest Information From the IDVP and NRC Staff Confirms the Inadequacy of the Diablo Canyon CQA Program and Establishes That the IDVP Is Inadequate i
to Cure Those Deficiencies 5
1.
The IDVP Has Demonstrated That the CQA Program of the Contractor Responsible for Construction of the Primary Coolant Loop Was Wholly Ineffective 8
f 2.
The IDVP Can Provide No Assurance That the CQA Breakdowns Already Uncovered Are Not Harbingers of l
Construction Errors of Major Safety l
Significance 10 l
l 3.
The IDVP Has Itself Demonstrated That Its Findings Are Inapplicable to the Ten Contractors Not Examined 14 4.
There Is No Comfort in the Fact That PG&E, the IDVP and the NRC Staff Have Detected Each Others' Errors 15 C.
PG&E's and the Staff's Comments to the l
Tennyson-Roam Statements and the Previously Identified Breakdowns in CQA Provide Further New Evidence That PGEE Has Not Had and Does Not Have a CQA Program That Meets the Commission's Regulatory Requirements 16 1.
The NRC Staff and PG&E Have Confirmed That CQA Inspectors Have Not Been Qualified in Accordance With Regulatory Requirements 16 1.
~
d TOPICAL INDEX (Continued)
Pages 2.
PG&E and the NRC Staff Have Substantiated the Fact That One of Its Major. Subcontractors Did Not Comply With the Requirements of Criteria 3 and 5 of Appendix B 19 D.
The Evidence Submitted With the NRC Staff's Response Supports the Governor's Call For A Hearing to Determine the Adequacy of CQA For the Extensive and Rapid Paced New Construction at Diablo Canyon 22 E.
PG&E Has Revealed That It Has Never Developed A Systematic QA Program For SS&Cs Which, Though Important to Safety, Are Not Safety Related 27 III PG&E AND THE NRC STAFF FUNDAMENTALLY MISCONSTRUE THE LAW BY PROPOSING THAT l
A DISPUTED ISSUE CAN BE RESOLVED WITHOUT AN ADJUDICATION ON THE BASIS OF THE STAFF'S OPINION ON THE ISSUE 36 IV CONCLUSION 39 l
l l
[
e ii.
i s
s TABLE OF AUTHORITIES Cases Pages Metropolitan Edison Company ALAB-729 (1983) 17 NRC 31 Pacific and Gas Electric Co.,
CLI-81-5 (1981) 13 NRC 361 2,7,37 Vermont Yankee Nuclear Power Corporation, ALAB-138 (1973) 6 AEC 520 2
Statutes 10 C.F.R. Part 50 Appendix A 1,6,7,10,23,28,29,32,33,34,35 Appendix B 1,6,7,10,16,17,18,19, 20,21,23,25,27,33,35,37,38 10 C.F.R.'Part 100 Appendix A 28,29,32 10 C.F.R. SS 2.715 37 50.49 30,31 50.57 35 Miscellaneous ANSI N45.2.6 16,17,18 ANSI N45.2.10 29 48 Fed. Register, p. 2728 31 Regulatory Guide 1.58 16 O
WASH 1309 16,18,19 4
111.
i UNITED STATES OF AMERICA i
NUCLEAR REGULATORY COMMISSION BEFORE THE ATOMIC SAFETY AND LICENSING APPEAL BOARD
)
In the Matter of
)
)
PACIFIC GAS AND ELECTRIC COMPANY
)
Docket Nos. 50-275 O.L.
)
50-323 0.L.
J (Diablo Canyon Nuclear Power
)
Plant, Units 1 and 2)
)
)
)
REPLY OF GOVERNOR GEORGE DEUKMEJIAN i
TO THE RESPONSES OF THE NRC STAFF l
AND PACIFIC GAS AND ELECTRIC COMPANY
(
TO THE GOVERNOR' S MOTION TO REOPEN l
THE RECORD ON CONSTRUCTION QUALITY l
ASSURANCE Pursuant to leave granted by the Atomic Safety and Licensing Appeal Board's June 7, 1983 order, Governor George t
Deukmejian hereby replies to the responses of Pacific Gas and Electric Company ("PG&E") and the Nuclear Regulatory Commission
("NRC") staff to the Governor's Motion
(" Motion") to reopen the record on construction quality assurance ("CQA").
I INTRODUCTION The Governor's motion to reopen on Construction Quality l*
Assurance is based on significant new evidence demonstrating I
violations of the requirements of General Design Criterion I
("GDC-I")of Appendix A to 10 C.F.R. Part 50
(" Appendix A") and of the eighteen criteria of Appendix B to 10 C.F.R. Part 50
(" Appendix B").
This evidence establishes that PG&E and its 1.
l
major subcontractors did not timely develop a program for CQA which met the requirements of Appendices A and B and did not fully implement the programs they did have in a manner consistent with the requirements of those regulations.
Under the rule enunciated by the commission in Pacific Gas and Electric Co. (Diablo Canyon Nuclear Power Plant, Units 1 and 2) CLI-81-5 (1981) 13 NRC 361, a party moving to reopen a record meets its burden by timely making an initial case that i
an NRC safety regulation would be violated."
(13 NRC at pp. 362-363.)
Such a case is made when the moving party has presented evidence which creates a triable issue of fact as to whether a violation of a commission regulation has occurred.
Vermont Yankee Nuclear Power Corporation (Vermont Yankee Nuclear Power Station) ALAB-138 (1973) 6 AEC 520, 523.)
The Governor has made the required showing by presenting evidence of:
a)
Major deficiencies in PG&E's quality assurance program / for Diablo Canyon's design and 1
construction as identified by the Independent Design Verification Program ("IDVP") ;
b)
Design errors stemraing from that deficient program as identified by PG&E, its contractors and the IDVP; c)
Construction errors stemming from that deficient l
1.
As used herein, " program" includes both development and implementation of the QA program.
1 l
2.
i l
program as identified by PG&E, its contractors and the IDVP; and d)
Deficiencies in the QA program of a major PG&E construction contractor.
In their responses, PG&E and the NRC staff have not only failed to refute this evidence but in fact have either corroborated it, significantly added to it, or simply failed to respond to the issues raised.
Repeating the now familiar 4
litany -- there was a problem but we solved it, there is no problem but we're investigating -- PG&E and the NRC staff continue studiously to ignore the significance of the CQA failures that are constantly being uncovered at Diablo Canyon.
Both PG&E and the NRC staff have avoided dealing with the specific requirements of. Appendices A and B.
Instead of examining the QA failures they have found in relation to the requirements of the commission's regulations, they have l
substituted their own judgment as to what type of quality l
assurance program is necessary in order to avoid construction errors with major safety significance.
Because of this, they
-have barred the IDVP from judging CQA at Diablo Canyon against the requirements of the regulations and have ordered it instead to conduct a review of what they, and not the regulations, regard as significant.
The discussion that follows will demonstrate that PG&E and the NRC staff have not only failed to disprove any of the factual allegations upon which the Motion was based but in fact have made the Governor's case for him.
3.
l l
l
II THE BOARD HAS BEFORE IT SIGNIFICANT NEW EVIDENCE OF VIOLATIONS OF NRC SAFETY REGULATIONS GOVERNING THE DIABLO CANYON CQA PROGRAM A.
Neither PG&E Nor the NRC Staff Can Escape the COA Implications of the Design QA Failures In their responses, PG&E and the NRC staff strive mightily to separate the CQA process at Diablo Canyon from the implications of the total design QA ("DQA") breakdown that occurred.
PG&E argues that DQA was really done by an organization so removed from the CQA apparatus an to be virtually a " separate" company.
PG&E does not dispute the fact that each of these
" companies" used the same QA manual and reported to the same PG&E
[ Response of PG&E to Motions to Reopen the Record on CQA ("PG&E Response"), p. 74] senior management.
PG&E also relies on the fact that the NRC staff has traditionally put a greater effort into CQA inspections than DQA inspections, so that more faith can be placed on the staff's review of PG&E's CQA process.
(PG&E l
Response, p. 75.)
The Motion pointed out that the commission relies on the highest utility management to create an atmosphere that insures l
vigilance throughout the QA process.
Neither PG&E nor the staff has disputed that the same senior PG&E management which oversaw 1
the grossly deficient DQA program had ultimate responsibility for the CQA program.
In relying on the fact that the NRC staff purportedly 4.
devoted more effort to CQA than DQA inspections, PG&E is asking this board to ignore the NRC staff's consistent position that its inspections alone cannot be expected to guarantee quality assurance and that such assurance can only come from the existence of a top utility management having a clear and aggressive commitment to QA.
(Motion, p. 8.)
The staff asks this board to ignore its own actions in recommending the initiation of an IDVP review of CQA in order to ensure that there had been no "spillover" of DQA problems into the CQA area.
(Faulkenberry Affidavit, p. 2.)
The staff acknowleuges that the DQA failures raised " questions" about CQA that led the staff to call for an IDVP audit of CQA.
[PRC staff Response to Motion to Reopen the Record on CQA (" Staff Response "), p. 11. ]
But, the staff insists, the DQA breakdowns do not guarantee CQA failures.
True enough, they do not prove CQA errors.
All'the Motion claimed was that the DQA failures give rise to an inference of COA inadequacy -- an inference confirmed by the other evidence cited in the Motion.
By resisting the inference, the staff asks this board to reject an inference the staff itself drew.
B. The Latest Information from the IDVP and the NRC Staff Confirms the Inadequacy of the Diablo Canyon CQA Program and Establishes That the IDVP Is Inadequate to Cure Those Deficiencies The Motion demonstrated that, as narrow and limited as the IDVP was, it had uncovered:
a) significant contractor violations of its own QA requirements; 5.
a
b) the failure by PG&E adequately to control the activities of its construction contractors; and c) a lack of aggressive corrective and preventative action by PG&E and its contractors in the face of these breakdowns.
Since the filing of the Motion, the IDVP has made revisions to its reports on CQA.
These reports confirm.the previous conclusions.
With their responses, neither PG&E nor the NRC staff have seriously disputed that breakdowns in the CQA process have occurred; rather, they have attempted to argue that such breakdowns are minor and insignificant because they have so far not been discovered to have resulted in a construction error with major safety significance.
The responses to the Motion evince an indifference to the specific requirements of Appendices A and B.
In its place they have substituted their judgment and the judgment of the IDVP as to what type of quality assurance program is necessary in i
order to avoid construction errors with major safety significance.
Such a position is both unresponsive to the Motion and at odds with the law.
The Motion is founded upon the premise that PG&E has an obligation-to comply with the requirements of Appendices A and B and that a showing that PG&E has significantly deviated from those requirements is sufficient grounds for this board to reopen the record on CQA.
PG&E and the staff apparently believe that it is 6.
acceptable to violate the requirements of the commission's regulations (append ices) if no construction error of major safety significance can be proven.
The commission has ma6e it abundantly clear that a record will be reopened upon a showing that there has been a significant violation of an NRC safety regulation irrespective of whether that violation has yet resulted in the discovery of an error of major safety significance.
(Pacific Gas and Electric Company (Diablo Canyon Nuclear Power Plant, Units 1 and 2), supra, 13 NRC at 362-63.)
Because both staff and PG&E have been looking only for construction errors creating a significant safety hazard, rather than violations of the requirements of Appendices A and B (see Morrill Affidavit, p. 5), the IDVP was not designed to look for QA program inconsistencies with regulatory requirements or to report on items uncovered based on their inconsistency with those requirements.
Rather it was designed for the purpose of determining compliance with whatever program the contractor had in place and then to ascertain whether the discovered inconsistencies with that program involved or would lead to construction errors with major safety implications.
(See the program description for ITRS 36 & 38.)
In short, the IDVP was not designed to shed light on the adequacy of the QA programs of the contractors it was evaluating.
As a result, the IDVP's conclusions properly have no bearing on the issues raised by the Motion.
The only significance of the IDVP lies in the raw data it provides on the number of QA program failings uncovered and the implications those failings have for 7.
the adequacy of the other CQA programs at Diablo Canyon.
- And, l
indeed, the raw data uncovered by the IDVP amply demonstrate a significant breakdown in the CQA program for at least one PG&E construction contractor.
Moreover, although the NRC staff and PG&E consistently proclaim that the IDVP results can be regarded as conclusive on the question of whether construction errors with major safety significance can now be expected to be found in the future at the site, the IDVP did not and could not provide such assurance.
To date, the IDVP has limited its conclusions to the specific contractors examined - not to the overall CQA program.
1.
The IDVP Has Demonstrated That the CQA Program of the Contractor Responsible i
for Construction of the Primary Coolant Loop Was Wholly Ineffective In reporting ~the results of the IDVP evaluation of Wismer and Becker's work on the primary coolant loop (ITR 38),
the NRC Staff and PGEE emphasize how relatively few discrepancies were uncovered.
For example, PG&E states that only 16 "open items" were filed with respect to the review of 3528 documents and only 9 such questions were raised after their inspection of 2298 plant items.
(PG&E Response, p. 55.)
Moreover, it reports that only one construction error of major safety significance was l
~ uncovered after the IDVP review.
(PG&E Response, p. 47.)
However, in choosing to cite the IDVP figures in this fashion, both PG&E and the NRC staff were comparing apples and oranges.
And in choosing to focus on the fact that only one construction error of major safety significance was uncovered, 8.
they have demonstrated their failure to appreciate the CQA significance of what has been found.
In fact, the IDVP has shown that the Wismer and Becker CQA program was wholly effective.
The following table summarizes the results shown in Appendices G1 and G2 to the latest revision of ITR 38:
QA Program
- Unsatisfactory
- Observed Percent Items (attributes) Evaluated 18 80 22.5 Observations 573 2841 20.2 Verification Of Physical Installation Items (attributes) Evaluated 21 45 46.7 Observations 1101 3176 34.7 P
(See attached Hubbard Affidavit, p. 6.)
In short when one compares the number of QA program areas in which one or more errors was discovered against the total number of program areas examined, approximately 20% of those areas were found to contain errors.
And when one compares the actual number of items observed in the QA program areas examined and the number of observations within the program where there were errors, approximately 22% of those QA program elements observed were found to contain errors.
With respect to the actual physical inspection of items within the plant, the results are even more striking.
Fully 46.7% of the classes of plant items looked at had one or more errors in them.
Moreover when the IDVP looked at the individual installation items themselves, nearly 35% failed to meet acceptance criteria.
I 9.
Such results' simply are not consistent with a finding that Wismer and Becker had a properly functioning CQA program.
PG&E, the NRC staff, and the IDVP, were not looking for the root cause of the QA errors they were uncovering.
Instead they were looking for construction errors of major safety significance.
Instead of assuring that no such construction r
errors existed by insisting on strict compliance with the
(
requirements of Appendices A and B, PG&E and the NRC staff sought
~
to reassure themselves that all was well by examining the Wismer and Becker QA program against their own standards of what an adequate OA program should be.
Thus, a significant QA discrepancy was not one which established that the program was j
not functioning in compliance with the regulations, or even that the program was not meeting its own standards.
Rather PG&E and l
the NRC staff found significance only if the immediate error detected created a direct hazard.
A program so structured inevitably falls far short of the task which PG&E and the NRC staff have set for it.
l 2.
The IDVP Can Provide No Assurance That l
The CQA Breakdowns Already Uncovered l
Are Not Harbingers of Construction Errors j
of Maior Safety Significance According to the NRC staff, the IDVP's goal is to assure that no construction errors of major significance remain undetected.
(Morrill Affidavit pp.3-4; Faulkenberry Affidavit,
- p. 2.)
Yet nothing in its scope, nothing in its results to date, and nothing in the CQA discoveries made by other organizations provide any assurance that this will in fact be the case.
10.
As the NRC staff admits, the IDVP has looked at only two of twelve contractors whose work was of major safety significance.
All that the IDVP looked at were the contractors for containment concrete and the reactor coolant system, completely ignoring the contractors responsible for activities such as:
a) electrical work; b) heating, ventilation and air conditioning; O
c) fire systems; d) equipment monitoring; e) cranes; and f) platforms and gratings.
In short, there is no possible conclusion one can draw about the quality of work of the unexamined contractors based on a look at the two contractorc who were picked.
This fact has been shown by the recent disclosures about the H.P.
Foley Company.
Foley's CQA program was clearly not functioning as it was supposed to.
As a result its management structure was changed.
Its inspectors are in the process of being requalified.
Its workers are being retrained and a significant portion of its work is being redone.
But the IDVP caught none of this.
Indeed, even the contractors examined by the IDVP were not scrutinized thoroughly.
The IDVP missed a significant i
construction error in the work of a contractor it was reviewing.
PNO-V-83-21 (Attachment II to NRC Response) involves the discovery that the wall thickness of piping for the reactor 11.
coolant system was below required standards.
The IDVP completely missed this CQA failure simply because its job was defined too narrowly.
Consequently even its conclusions about the safety significance of errors uncovered in Wismer and Becker work are suspect.
As Mr. Hubbard states in his attached affidavit, the IDVP's emphasis:
. on reviewing QA/QC records and conducting visual examinations as contrasted to conducting tests or physical inspections almost assured the eventual finding that no single IDVP observation in and of itself would constitute a significant safety matter.
This is true since in general the IDVP CQA review fails to address important attributes significant to safety.
For example, the IDVP CQA review did not address such significant safety indications as might have been revealed by a Reynolds probe nondestructive examination of concrete strength, by a repeat of radiographs of l
safety related welds or at least an independent review l
l of previous radiographic film results, by an independent l
test of material composition to verify the accuracy of the original material certifications, and by an i
independent review of pre-operational test results and a repeat of selected portions of such tests.
Such j
omissions of significant areas of product compliance are l
each serious weaknesses in the IDVP CQA review."
(Affid., pp. 6-7.)
12.
l
I The limitations of the IDVP in this regard are no better exemplified than with the error uncovered in PNO-V-83-21 (NRC Response, Attachment II).
Here, a cracked weld was found in the component cooling water system.
The weld may not have been properly performed, but this was discovered solely by accident, as a result of the movement of this piping during the new construction at the plant.
The discovery of this construction error highlights the fact that the IDVP overemphasizes visual inspection of the work.
Such a check would not have uncovered i
the deficiency in the weld in this system.
Only an actual retesting of the weld at issue could have verified its failings.
Thus, the IDVP's failure to perform such tests as part of its review demonstrates that it cannot be relied upon to assure there I
are no significant construction errors in the Wismer and Becker work.
Finally, if there were any question about the IDVP's effectiveness that doubt is surely put to rest by the i
"Superstrut" revelations.
Here, completely outside the scope of the PG&E, NRC and IDVP QA review process, it was discovered that Midland-Ross, a supplier of safety-related material to the Foley Company, had no formal QA/QC program prior to 1979.
The failure of the Midland-Ross Company to have a QA/QC program for the l*
equipment it was supplying to the Foley Company was in turn compounded by the fact that the Foley Company itself had no procedure for auditing the QA/QC program of Midland-Ross and thus no way of assuring itself that the paper certificatione it was receiving reflected the actual quality of the materials being 13.
---.-, -...~ -.- _.._. -
used.
In short none of the procedures in existence in the past and nothing the IDVP is doing now can provide assurance that the quality of the materials in use for at least the Foley Company are of the quality they are required to be.
3.
The IDVP Has Itself Demonstrated That Its Findings Are Inapplicable To The Ten Contractors Not Examined In selecting two of the twelve major contractors, the l
IDVP sought to draw conclusions about the entire construction program.
The Governor has expressed reservations about such a l
procedure -- not because the contractors were not drawn at l
random, as the staff seems to suspect, but because sampling inappropriate to exclude whole populations. /
2 itself is The IDVP's results dramatically demonstrate this point.
While, as noted above, error rates over 45% were found in the Wismer and Becker work, the corresponding rates for Guy F.
Atkinson, the other contractor examined, were below 20%. /
3 When one contractor has an error rate of 46.7% and another has a rate of 17.6%, one inescapable conclusion is that the error rates vary among contractors.
What then do we know about the error rate for, say, the electrical work, which was wholly ignored by the l
l 2.
There should be no confusion on this point.
Where sampling is to be done from a given population, the samples i
should be drawn at random, assuming an adequate sample size can be had.
But the Governor has never suggested that populations l
should be sampled - randomly or otherwise.
The random sample l
should be drawn from all relevant populations and subpopulations.
3.
This should not come as a great surprise.
One would expect concrete work to have far fewer errors (particularly errors susceptible to visual detection) than piping work, 14.
i l
IDVP?
Plainly, nothing.
The IDVP's own findings preclude generalization of its conclusions to any contractors not examined.
In summary, the IDVP has not sampled the work of most of the relevant contractors, has not examined the sample it did take with sufficient thoroughness, and has not applied the proper standard to the CQA program it seeks to validate.
The IDVP simply fails to demonstrate that no construction errors of major safety significance exist at Diablo Canyon.
l 4.
There Is No Comfort In The Fact That PG&E, The IDVP, And The NRC Staff Have Detected i
l Each Others' Errors Seeking to make the best of the litany of errors, PG&E and the staff point out that, after all, the errors were detected, albeit not precisely in the manner they should have been found.
While a PG&E contractor may have had a deficient program, some of the deficiencies were discovered by PG&E, others l
by NRC enforcement personnel, still others by the IDVP.
So there's really nothing to worry about.
l l
The problem, of course, is that there is no way of l
l knowing how many errors have run the gauntlet and remain undetected.
There is no evidence in this record from which to conclude that any of the programs was adequate to plug the other program's leaks.
What is known is the fact that the NRC has only a handful of inspectors monitoring the QA work of the applicant and its contractors.
For them to detect any errors is a proper source of concern.
15.
I
C.
PGEE'S and the Staff's Comments to the Tennyson-Roam Statements and the Previously Identified Breakdowns in CQA Provide Further New Evidence that PG&E Has Not Had and Does Not Have a Construction Quality Assurance Program That Meets the Commission's Regulatory Requrements The responses of PG&E and the NRC staff with respect to the Tennyson-Roam statements and the previously identified breakdowns in CQA, far from discrediting the Motion have in fact provided new and even stronger evidence in support of it.
1.
The NRC Staff and PG&E Have Confirmed That CQA Inspectors Have Not Been Qualified in Accord with Regulatory Requirements By virtue of the Tennyson-Roam Statements, the Governor also demonstrated that until December 1982, the Foley Company was not required by PG&E to have QA inspectors who met the inspector qualification standards set forth in ANSI N45.2.6.
(Exhibit B to Motion, pp. 14-17, 21; See also the Hubbard Declaration, pp. 13-18.)
Inasmuch as ANSI N45.2.6 was adopted in 1973 as Regulatory Guide 1.58 and included as part of the WASH 1309 procedures which PG&E had testified to the Licensing Board in 1977 that its CQA program incorporated, /
4 the Governor l
concluded that this PG&E contractor had violated the requirements of criteria 2 and 5 of Appendix B.
(Hubbard Declaration, at
- p. 18.)
/
4.
Direct testimony of Russell P. Wischow pp. 6-7, referenced in ASLB transcript pp. 3597-98.
(See also PG&E's Response at p. 21 for confirmation of this fact.)
16.
PG&E's response to this statement by the manager of the Foley CQA program was to rely upon the investigation subsequently conducted by the NRC staff.
However this investigation, far from disproving the validity of Mr. Tennyson's statement, actually verifies it and goes on to provide further evidence of the Foley Company's continued inability to meet the requirements of criteria 2 and 5, and, until very recently, of the failure of the entire PG&E CQA program to meet the requirements of these criteria.
The report of this investigation is attached as an exhibit to the affidavit of John D. Carlson.
In this report, the following findings are made (Exh., pp. 7-8):
a.
ANSI N45.2.6, specifying qualifications and educational requirements for QA inspectors was not a part of the Foley CQA program; c.
This standard was not a part of the Foley CQA program because it was not a part of PG&E's CQA program; d.
PG&E did not commit to complying with the requirements set forth in this standard until July 14, 1981; e.
As of August 1982, the Foley Company still had not made this standard a part of its QA program; f.
Not until January 1983, was this standard actually made a part of the Foley QA program;
/
17.
g.
Even after it became'a part of the Foley l
program, it was discovered that QA inspectors working in the field did not have resumes
^
supporting their certification in all areas; s
and i
e h.
It was determined that from December 7, 1983, to March 10, 1983 Level I inspections were not being co-signed with a Level II signature.
In short since 1974, when WASH 1309 was issued, the qualification standards for QA inspectors set forth in ANSI N45.2.6 have been the accepted means for carrying out the i
requirements of criteria 2 and 5 of Appendix B.
Moreover, PG&E t
testified to the Licensing Board in 1977 that its COA program incorporated the standards set forth in WASH 1309.
However, the NRC staff's investigative report now establishes that PG&E's programadid not incorporate the portion of WASH 1309 which included ANSI N45.2.6 until eight years after it was established i
i l
as a regulatory standard and five years after PG&E purportedly committed to meet it.
It also establisnes that the Foley Company's QA program did not attempt to meet these requirements until January 1983 or nine years after it became a regulatory standard and that even now Foley is having difficulty implementing those requirements.
l These findings by the NRC staff are remarkable not only because of PG&E's contrary testimony to the-Licensing Board in 1977,3 but also because the staff had found in 1975 when it completed Supplement No. 3 to its Safety Evaluation Report 18.
i f
6
- 7..,..y
, /.
~T s
4 7
~ ("SER") for Diablo Canyon that the QA program for plant operation 1]
incorporated the procedures contained in WASH 1309.
(See p. 21 4
.i.
4:
.g
,of PGEE's Response.)
i a
e Jd.
In summary, the Staff's investigation of the 7
Tennyson-Roam statements not only corroborates their. testimony 9
! ' - 4[
about the failings of the Foley Company's QA program but also
~
4 M,, ~ yk, M -
/4 s
demonstrates that, until very recently, PGEE's entire CQA program gr x.
p, j '.
Way..have had a similar failing.
Moreover, not only did these
,y 3 failings exist, with their consequent implications for the
>A
?
/,, quality of the plant's. construction, but they existed hidden from Q
n
', the view of the Lic' nsing Boar 6\\ when it made its decision in 1977 e
'il j_
\\ n' t
/hapd from the staff 'wheht itaprepare'd'its SER in 1975.
r of(.
f aN ~
As a Q< /q 3. consequence; the staff and'then the Lihensing Boatd' reached their
<4 3
I
};sconclusionsabouttheconsistenchofPG&E'sCOAlprogramwiththe p'
V}
w/
\\
,rgquirements of Appendix B based on their rdisperceptioh that the '[
^ )i program complied with all.the QA implementation procedures set l
^
?
,r, L
)
?
9,.
6 L
'for,th in WASE.,1309.' Inasmuch as compliance with those procedures
+ a
/t
~]
is essential to a finding that a CQA program is consistent with 7
d~ the, requirements OD,} Appendix '3',<PG&E's _ failure to do so for i
.J J1
.,s
/ '/s -*jie, elf and for itu dontract' ors provides a clear basis upon which
/
lr f
+
7
/
to or' der the record on CQA reopened.
j t
g
{
2.
PG&E an60the NRC Staff Have Substantiated jf the Fact That One of its Maior Contractors i4*
Did Not Comply with the Requirements of l
CriterlC3 and 5_ of Appendix B j.
f T'
2 v
j A
s In referencing PG&E's. response to I&E Bulletin
,d t3No. 79-14 and.the subseque.3t R'.L.
Cloud report o7 the same piping i
l
'(
,p s
j systems (Motion, ppel0-11I,' the, Governor demonstrated - that the
\\
nt V
p j
f /
19.
'j!
, ' ;1 3
4-
,+ f
.\\
yN bl('+2l\\
V g"
s 1
l',
e v,
(u sr IS 1
!, ~
,,ul - l 5.. - -- -
s'a&
interface between contractor CQA departments and the PG&E engineering department did not always function in a closed loop manner as it was supposed to.
Relying on the Tennyson-Roam statements, the Governor also established that Foley Company QA inspectors were not furnished with installation instructions and 3
design drawings of sufficient detail to enable them to confirm whether actual construction met design requirements (Motion, Exh.
B, pp. 60, 63).
As a result of all the above, the Governor concluded that PG&E's CQA program was not in compliance with the requirements of criteria 3 and 5 of Appendix B.
In the NRC staff investigative report (at pages 11-12),
which is attached to the affidavit of John D. Carlson, the staff concedes that the Foley Company's QA department did not always s
have specific drawings and installation standards for electrical conduit supports and instrument tubing supports (supra, p. 12).
Furthermore, with reference to I&E Bulletin No. 79-14 both the NRC staff and PG&E admit that discrepancies existed between the plant piping as built and the " record drawings" of it because of a failure by the PG&E engineering department to expeditiously process drawings depicting modifications made during construction.
(PG&E Response, at pp. 56-57.)
Both PG&E and the NRC staff have substantiated the fact that the Foley Company CQA department did not operate with j
l det' ailed and descriptive drawings and instructions describing the i
standards which the work to be inspected had to meet.
They confirm that the Foley CQA department and the COA department of l
.at least one other contractor, did not properly interface by I
i 20.
(
l
closing the loop with the PG&E engineering department in confirming that "as built drawings" or " record drawings" (as PG&E would term it), accurately reflected actual construction in the plant.
In so doing, PG&E and the NRC staff have confirmed that the Foley Company did not have a QA program which met the requirements of criteria 3 and 5 of Appendix B.
To be sure, in making the concessions they have, neither PG&E nor the NRC staff have conceded that violations of these criteria have occurred.
They claim that nothing of consequence happened and that any possible problem has been remedied.
Both PG&E and the NRC staff are wrong.
PG&E's CQA program allows CQA inspectors to sign off work without verifying that the record drawing has timely incorporated any modifications made to the original design specifications during construction.
As a result PG&E's engineering-department, to the extent it fails appropriately to process construction changes to design, may continue to utilize " record drawings" that do not accurately reflect that plant as constructed.
Such construction changes are likely to be quite frequ'ent occurrences when as in the case of the Foley Company, generic rather than unique instructions are being provided QA inspectors.
Timely review and incorporation of any field changes into the documents are required to assure that the field change has not resulted in design criteria being violated inadvertently.
Such a prompt reconciliation of design documents l
l l
/
21.
1
i has not routinely occurred in the past. /
5 If these inaccurate record drawings are utilized as the basis of the seismic design analysis which is currently being undertaken or the System's Interaction Program (SIP), which has not yet been completed, then either or both of these analyses will be just as invalid as the seismic design evaluation which PGEE_ conducted prior to the I&E Bulletin No.
79-14 review.
In summary, PGEE's open loop CQA program, which permits the use of non-specific installation instructions for QA inspectors and poor and untimely liasion between the CQA-and design OA departments with respect to as-built and record drawings, may well result in a seismic analysis and a SIP report which are next to useless.
D.
The Evidence Submitted with the NRC Staff's Response Supports the Governor's Call For A Hearing to Determine the Adequacy of COA For the Extensive and Rapid Paced New Construction at Diablo Canyon In light of the Tennyson and Roam statements concerning the imp'act the furious pace of new construction was having on the already deficient Foley Company CQA program, the IDVP's findings concerning CQA breakdowns, and PG&E's past history of breakdowns in quality assurance under relatively normal conditions, the Motion called for a new hearing to verify whether PG&E had 5.
PG&E acknowledges that as late as 1981, nearly four years after the Hosgri design changes, the discrepancies between the "as builts" and actual field conditions resulted from the use of unrevised record drawings which had not yet been updated by the PG&E engineering department.
(PG&E Response, p. 67.)
22.
-.. - -. - ~ -
developed and implemented a CQA program for the new construction at Diablo Canyon which met the requirements of Appendices A and B.
The Tennyson-Roam statements made it clear that (a) the rapid build up in the construction work force had led to the use-of unskilled and unqualified. workers and inspectors, (b) that understaffing in the Foley CQA department had led to up to 80 hour9.259259e-4 days <br />0.0222 hours <br />1.322751e-4 weeks <br />3.044e-5 months <br /> work weeks for inspectors, and (c) that unusually large numbers of errors in construction including welding work were occurring.
Given these circumstances the Motion concluded that new hearings on CQA were in order.
In its response, the NRC staff includes affidavits from its inspectors as well as reports from PG&E concerning the state of the Foley CQA program which corroborate the Tennyson-Roam statements and support the Governor's call for a new hearing to determine whether PG&E's CQA program is adequate to the task of i
assuring the quality of the new construction at Diablo Canyon.
In the documents supporting the affidavit of Gonzalo Hernandez, Jr., which are attached to the NRC staff brief, the l
l NRC supports the Tennyson-Roam statement that the Foley Company was encountering problems with the quality of welding in the fuel handling building.
Based on inspections of the welding performed, the staff issued a Notice of Violation (Hernandez l
l
[
Af fd., p. 1).
PG&E responded to this Notice by re-inspecting ten l
percent of the fillet welds in the Unit 1 portion of the fuel i
l handling building (.'xh.
D to Hernandez Affd., p. 1).
As a result of this re-inspection, 40 of the 489 welds were found to 23.
l i
L
.p be at variance with QA program specified criteria but could be accepted anyway.
Forty-seven other welds were found to be completely. unacceptable and had to be redone (Exh. D to Affd.,
- p. 2).
Thus, 17.8 percent of the welds looked at in the sampling program were found to be inconsistent with QA program criteria and 9.6 percent of them were in fact completely redone.
In its May 19, 1983 letter to the NRC staff explaining the corrective action program it had undertaken in response to this breakdown in CQA, PG&E indicated that the QA program:
. deficiencies could not be isolated to specific connection types, individual welders or inspectors, or locations that would have allowed f
pinpointing any further discrepancies." (Exh. D to Affd., p. 3.)
j As a result, PG&E ordered a 100 percent re-inspection of the structural steel fillet welds in the Unit 1 fuel handling building.
This re-inspection showed that 14 percent of the welds had to be redone.
(Exh. D to Affd., p. 3).
i In its report to the NRC staff, PG&E explained its l
understanding of the cause of what can only be termed a complete breakdown in quality assurance as follows:
"PG&E has investigated the findings of the reinspection program and has determined that there l
l are two principle causes of the structural steel i
l welding deficiencies.
The first is insufficient i
training of new welders and new welding inspectors,
(
in that they did not receive sufficient P
24.
l
familiarization with all the acceptance criteria and that strict compliance with the criteria was required.
The second is an insufficient ratio of QC inspectors to production workers; this resulted from an undesirable distribution of inspectors, lack of good supervision of the contractor Quality Control group, and was to some degree influenced by the' build up in the number of workers.
Additionally, it was also found that three QC inspectors had not been properly qualified." (Exh.
D, supra at p. 3.)
In short, PG&E itself has corroborated what Tennyson and Roam themselves stated about the Foley Company l
l CQA program.
Workers and inspectors were insufficiently trained, inspectors were forced to work overtime to make up for insufficient staffing and inspectors were unqualified.
However, the Motion did not and does not rest solely
'upon the occurrence of these breakdowns in the PG&E-Foley CQA program.
Rather, it rests equally upon the response of PG&E and the NRC staff to these b'reakdowns.
Contrary to the requirements of criterion 16 of l
Appendix B, PG&E has simply not taken the kind of aggressive corrective and preventative action required by the nature of the breakdowns which have been uncovered.
Similarly, the NRC staff, while finding breakdown after breakdown in the Foley CQA program, continues to treat each identified
/
25.
.~..a.
l breakdown as a separate and isolated instance of no generic significance.
In'this regard while PG&E, to its credit, has undertaken corrective and preventative actions with respect to the fillet weld errors which the NRC staff uncovered by j
increasing training classes for its welders and inspectors, by increasing the number of inspectors, and by reviewing some of the welding work inspected by the unqualified inspectors, it has failed to recognize that the problem in the Foley CQA program is bigger than simply welding and that the problem area with respect to past work may cover more than just fillet welds.
As Hubbard explains in his attached affidavit:
"The preceding corrective action measures in my opinion do not provide the necessary level of safety assurance as required by criterion 16 of Appendix B.
l In my opinion, sampling is no longer appropriate.
Thus, all welding performed and inspected by Foley personnel during the 1983 big push should be 100%
reinspected.
In addition, welding work performed in 1982 should be included in the reinspection program on
~
a sampling basis to confirm that the time period when l
l inadequate welding and weld inspection occurred was i!*
restricted to Foley's 1983 activities.
The preceding complete review is required since PG&E and Foley have failed to demonstrate that the widespread welding l
deficiencies could be isolated to particular types of welds, individual welders or inspectors, or particular 26.
~
. _ = _ - _ -
locations in the plant.
Further, a tabulation-summarizing all inspection and test activities conducted by Foley QA/QC personnel during 1983 should be prepared, and a suitable sample of these activities l
should be reverified to determine the extent to which the inadequate inspection activities can be isolated to weld inspections."
(Hubbard Affid., pp. 4-5.)
i In short, neither PG&E nor the NRC staff have exhibited any appreciation of the generic nature in QA terms of the breakdowns which have occurred in the Foley QA program and that appropriate corrective action within the meaning of criterion 16 of Appendix B requires more than what has been done.
Therefore, because of the import of the breakdowns in CQA which have now been corroborated by PG&E and the NRC staff, and because both the staff and PG&E have failed to take the kind of action required by Appendix B, it remains for this board to reopen the record for a review of the adequacy of CQA for the current round of new construction at Diablo Canyon.
E.
PG&E Has Revealed That It Has Never Developed a Systematic QA Program for SS&Cs Which, Though 1
Important To Safety, Are Not Safety Related With the Tennyson-Roam Statements, the Governor demonstrated that one of PG&E's major construction contractors, the H.P. Foley Company ("Foley"), has a CQA program which does not systematically inspect structures, systems and components l
(SS&Cs) which, though important to safety, are not safety related.
(Motion, Exh. B, p. 53.)
Thus, it was concluded that the Foley CQA program failed to meet the requirements of 27.
Appendix A.
(See Hubard, March 26, 1983, supplemental affidavit,
- p. 30.)
In its response to the Motion, PG&E has turned this demonstration of a violation of Appendix A by one of its
. construction contractors into a demonstration by PG&E of a violation of Appendix A by its entire CQA program.
PG&E's failure to comply with the requirements of Appendix A is confirmed with the revelation at page 26 of its brief that PG&E regards and has always regarded the terms " safety-related" and "important to safety" to be synonymous.
As a consequence it did not regard the requirements of Appendix A as applicable to any SS&Cs other than those which are " safety-related" (which it defined at page 3.2-1 of its FSAR using the 10 C.F.R. Part 100 Appendix A definition).
PG&E's beliefs about the correct interpretation of the terms "important to safety" and " safety related" are contrary to established law, and its decision to apply the provisions of Appendix A only to SS&Cs which are " safety-related" establishes
~
that it has failed to meet the Appendix A requirement to develop and implement a systematic CQA program for important-to-safety SS&Cs other than those which are safety-related.
Appendix A provides, in pertinent part, that SS&Cs:
important to safety shall be designed, fabricated, erected, and tested to quality standards commensurate with the importance of the safety functions to be performed.
"A quality assurance program shall be established and t
L implemented in order to provide adequate assurance that these 28.
i structures, systems and components will satisfactorily perform their safety functions."
ANSI N45.2.10 (1973) defines the term quality assurance l
as follows:
"All those planned and systematic actions necessary to provide adequate confidence that an item or facility will perform satisfactorily in service."
(Emphasis added.)
In short, Appendix A requires the development of a systematic program to assure that SS&C's important to safety will satisfactorily perform their safety functions when in service.
i Since 1971 the term "important to safety" has been defined in the first paragraph to the Introduction to Appendix A as:
" Structures systems and components that provide reasonable assurance that the facility can be operated without undue risk to the health and safety of the public."
On the other hand, the term " safety related" is defined in the Commission's regulations as those SS&Cs necessary to assure:
1) the integrity of the reactor coolant pressure boundary; 2) the capability to shut down the reactor and maintain it in a safe shutdown condition; or 3) the capability to prevent or mitigate the consequences of accidents which could result in potential off-site exposures comparable to the guideline exposures of 10 C.F.R. Part 100.
(See 10 C.F.R. 100, Appendix A sections III(c), VI(a)1 and VI(b)3.)
29.
t The distinction between these two terms was confirmed in a November 20, 1981 memorandum from Harold Denton, Director of Nuclear Reactor Regulation, to 'all NRR personnel (attached as an l
Exhibit to the Affidavit of Walter P. Haass accompanying the NRC staff response).
In that memorandum, Mr. Denton explains that safety-related is a subset of the class of important-to-safety items.
Furthermore, contrary to PG&E's suggestion that the i
distinction between these two terms is a departure from past practice and a wholly new interpretation of the meaning of these terms, the Denton memorandum makes it abundantly clear that this is not the case.
Though he acknowledges that these terms have sometimes been used " interchangeably," Denton specifically states that such usages were not:
" consistent with the definitions and usage of such terms in the regulations and
. do not fully reflect the intent of the regulations or current licensing practice."
The commission itself recently reiterated the fact that safety-related is a subset of important to safety.
On January 6, 1983, the Commission unanimously approved a i
revision to 10 C.F.R. S 50.49
(" Environmental qualification of electrical equipment important to safety for nuclear power plants").
In the statement of consideration accompanying the new rule, the commission stated:
"The scope of the final rule covers that portion of equipment important to safety commonly referred to as
' safety-related' (which the Commission interprets as essentially ' Class lE' equipment defined in IEEE 323-1974),
30.
f s-, - - -,
,-,-----.-,---,-r
---.-.--------.-----,n,.-
,~en
-,-.r~-.n.-.
.-,w---.-
r
--w,w,
and nonsafety-related electric equipment whose failure under postulated environmental conditions could prevent the satisfactory accomplishment of required safety functions by safety-related equipment."
(Emphasis added.)
(48 Fed. Reg. 2728, 2730 (1983).)
The language of the rule itself also makes clear that the class of important to safety is broader than that of safety related.
The new rule states in section 50.49 (b) that:
"[E]lectric equipment important to safety covered by the section is (1) the safety-related equipment and (2) the nonsafety-related equipment whose failure under postulated environmental conditions could prevent satisfactory accomplishment of specific safety functions and (3) certain i
post-accident monitoring equipment."
(Emphasis added.)
Significantly, in determining that the Regulatory Flexibility Act of 1980, U.S.C.
605(b) was not applicable, the commission stated that "this rule codifies existing requirements."
Were any doubt to remain that the terms." safety-related"
\\
and "important to safety" are not synonymous, that doubt was eliminated by the recent decision in Metropolitan Edison Company (Three Mile Island Nuclear Station, Unit No. 1) ALAB-729 (1983) 17 NRC.
In that case the appeal board rejected the contention l
that the terms "important to safety", " safety-related" and
" safety grade" were synonymous (Slip Opinion at pp. 130-131).
To the contrary, it found that
/
31.
l
" General Design Criteria 1 [in utilizing the term "important to safety"] plainly contemplates a range of safety requirements dependent on function, rather than a single requirement to which all structures, systems and components must conform.
. Appendix A to 10 CFR Part 100 illustrates the concept established in GDCl (i.e. gradations in quality level corresponding to relative safety importance) by identifying explicitly a select sub-class'of structures, systems and components from the broad class of those important to safety that is required for the performance of specific critical safety functions."
(Emphasis added.)
(Slip Opinion, p. 132.)
In short, a CQA program such as PG&E's which treats the terms important to safety and safety related as synonymous and thereby fails to require a systematic quality assurance program for important-to-safety SS&Cs which are not safety related,' fails to comply with the requirements of Appendix A.
As the appeal board indicated in Three Mile Island, supra, Slip Opinion, pp. 131-132, Appendix A contemplates that various SS&Cs which are important to safety will have different requirements imposed upon them depending upon their safety significance.
With respect to quality assurance, it is clear that the more stringent standards set forth in Appendix B are to be applied to safety-related SS&Cs because of their more significant safety function.
However, this does not mean that no systematic COA program is required for the remaining important-to-safety SS&Cs.
Though the program requirements may 32.
= -
beLless stringent, a systematic program is nonetheless required.
In light of the fact that PG&E has revealed that it had
.no systematic CQA program for non-safety related SS&Cs important to safety, PGEE'has confessed that it failed to meet the requirements of Appendix.A.
As a result this board must order the record reopened so that PGEE may demonstrate how it will satisfy the requirements of this regulation.
The NRC staff's position with respect to this issue is novel.
While reccgnizing that the term important-to-safety is broader than' safety-related (Staff Response, p. 14), and that there is a need to separately provide quality assurance for important to safety SS&Cs not covered by the CQA requirements of i
Appendix B (see affidavit of Walter Haass, p. 3), the staff j
argues that PG&E's failure to develop and implement such a h
program does not constitute a basis for reopening the record in this case.
Its argument in this regard-is twofold.
First, says the staff, when the commission ordered the IDVP, it did not require PG&E to demonstrate CQA for important-to-safety SS&Cs which are not safety-related; therefore, this board is not required to do aso.
Secondly, since the Licensing Board made no specific finding aon PG&E's compliance with the CQA requirements of Appendix A, b*
such a topic is beyond the proper scope of a motion to reopen the record in the proceedings below.
Both of these arguments are unsupported by any authority and are otherwise devoid of merit.
The fact that the commission did not order the 3
verification of the quality of non-safety related SS&Cs important 33.
i
to safety before it would lift the suspension of PGSE's lower power license has nothing to do with the question of whether this board should reopen the record on CQA.
This board is not being asked to lift the suspension of PG&E's low power license, rather it is being asked to reopen the record on the question of whether PG&E was entitled to the license in the first place inasmuch as its CQA program had failed and continues to fail to meet the requirements of Appendix A.
The staff's position in this regard is particularly remarkable in light of the fact, as disclosed in the affidavit of Walter P. Haas (p.3) that the staff now intends to "re-review l
PG&E's commitment to GDCl" and to require such action as is i
necessary to ensure that the requirements of Appendix A are met.
l If it is appropriate for the staff to now "re-review" l
PG&E's commitment to the QA requirements of Appendix A in order to ensure compliance with the commission's regulations, the Governor would respectfully submit it is even more appropriate for this board to do so.
If the staff now believes there is l
reason to doubt its original conclusion that PG&E had met its commitment to Appendix A, why should this board not also doubt the validity of the Licensing Board's conclusion with respect to l =
that regulation?
The second argument, that the record cannot now be reopened because the Licensing Board didn't address Appendix A, is even more remarkable.
The staff suggests that if a Licensing l
Board tries to make a finding and gets it wrong, it would be easier to reopen the record than if the Licensing Board simply l
l 34.
refuses to address the issue altogether.
No authority is cited for this proposition, which comes propelled by no logical force.
The Motion seeks to reopen the record, not to edit the finding.
While a failure to address an issue might be a pertinent factor to weigh in reopening if causad by a failure of the moving party to raise the issue in a contention, that fact would scarcely be dispositive of the matter.
In any event, the defense can be rejected on its own t
terms.
Contrary to the staff's claim, the Licensing Board made an implicit finding on Appendix A.
Before an operating license may issue, 10 C.F.R. section 50.57 requires that the commission l
find, among other things, that the applicant has complied with I
all applicable commission regulations.
Included within those commission regulations are the provisions of Appendix A.
In its July 17, 1981 decision the Licensing Board specifically concluded that:
" Construction of the facility has been substantially completed in conformity with the construction permit and the application as amended, the provisions of the Atomic Energy Act of 1954, as amended, and the rules and regulations of the Commission."
(Emphasis added.)
(Decision, supra, at p. 6.)
Inasmuch as the Motion has demonstrated that PG&E's CQA program did not meet the requirements of Appendix A, that l
board finding is just as erroneous as its specific finding that-the requirements of Appendix B were met.
Thus, just as the I
record must be reopened to correct the Licensing Board's erroneous specific finding, it must also be reopened to correct 35.
i
'the error contained in its general conclusion.
Just as PG&E must be required adequately to verify the quality of safety-related SS&Cs in Diablo Canyon it must also be required to demonstrate i
how it intends to verify the quality of non-safety-related but important-to-safety SS&Cs in that plant.
III PG&E AND THE NRC STAFF FUNDAMENTALLY MISCONSTRUE THE LAW BY PROPOSING THAT A DISPUTED ISSUE CAN BE RESOLVED WITHOUT AN ADJUDICATION ON THE BASIS OF THE STAFF'S OPINION ON THE ISSUE PGEE has responded to the evidence of violations of NRC safety regulations governing the Diablo Canyon CQA program by asserting that the NRC staff has said: (a) that such violations have been satisfactorily resolved; (b) that it could not substantiate thru; or (c) no such violations existed.
As a consequence PG&E concluded that there was no reason to reopen the record on CQA.
The NRC staff, citing its own investigations reached the I
same conclusion.
The reliance the NRC staff and PG&E place on the staff's investigative conclusions is wholly misplaced.
A difference of opinion between the staff and other parties to an NRC adjudicatory proceeding does not automatically result in a decision in favor of the staff.
If, as in this proceeding, a triable issue of fact has~been raised concerning a violation of the commission's safety regulations affecting.the CQA programs at Diablo Canyon, then a hearing must be held and a decision reached concerning the conflicting evidence.
In the 36.
i
course of that hearing the board would be free to reject the staff's conclusions.
If the staff position on an issue were conclusive there would never be a need for a hearing or for this board for that matter.
Under 10 C.F.R. section 2.715(c) an interested state has a right to participate in all adjudicatory hearings held to determine, among other things, an applicant's ability to design, construct and operate a utilization facility in accordance with the commission's regulations.
In the present case, the Governor has presented evidence that CQA programs at Diablo Canyon have not met the requirements of Appendices A and B.
PG&E and the NRC staff have failed to present any facts that the Governor's evidence, if true, would not establish such a violation.
All they have presented are their own opinions which, if believed, could dispose of some of these claims.
As a consequence a triable issue of fact has been presented over whether violations of the commission's safety regulations on CQA have occurred.
Inasmuch as the presentation of a triable issue of fact on such a question is a proper basis for reopening a record (Diablo Canyon, supra, 13 N.R.C. at pp. 362-363), this board is required to reopen the record in this proceeding.
The staff's analysis of the PG&E CQA program itself demonstrates the propriety of reopening the record.
The staff does not dispute that past violations have occurred.
It simply differs with the Motion on the implications of those violations, the adequacy of remedial action, and the likelihood that l
37.
undetected violations remain to be discovered.
Indeed, the substance of the staff's disagreement often amounts to no more than the unsupported, unexplained conclusion that the staff believes that no significant construction errors exist at Diablo Canyon.
Nowhere are the shortcomings of the staff's opinion more evident than at page 18 of the Staff Response, where the staff e
seeks to explain a way the implications of the inadequate IDVP review of the CQA program.
On the question of the adequacy of the IDVP -- an issue on which the staff has yet to state its official position -- the staff says:
. Although the Staff's review of the SWEC audit is not yet complete, it has made a tentative conclusion respecting the audit findings.
" Subject to confirmation in subsequent revisions of l
ITR's 36 and 38, the Staff is satisfied that the SWEC audit has properly documented and evaluted the 29 open items consistent with the criteria in 10 C.F.R. Part 50, Appendix B.
Of these 29 open items only 1 finding, I
EOI 9026, required some action.
Overall, the Staff is preliminarily satisfied that none of the SWEC items reveal a significant deficiency in the construction of Diablo Canyon.
./
6 6.
The tortured language also illustrates the difference between the issue here and the issue the staff seeks to address.
The present issue is whether or not the Diablo Canyon quality assurance program meets the requirements of commission regulations and thus has ensured safe construction.
The staff 38.
What-exactly is one supposed to do with the staff's preliminary satisfaction?
Does this constitute a prediction by the staff of how it will decide the issue?
What happens if the board denies the Motion and the staff changes its mind?
Will the staff then move to reopen?
Plainly this pronouncement -- which purports to be staff's evaluation of the only evidence on CQA produced by the IDVP -- pulls at itself from opposite directions, leaving the staff's opinion incapable of serving the function for which it was intended.
IV CONCLUSION The time has come to recognize that a serious problem with respect to both DQA and CQA exists at Diablo Canyon and that an effective and comprehensive remedy must be achieved.
In early October, 1981 after only two of the hundreds of i'
design QA failures had been uncovered, Harold Denton was prepared to admit that PG&E's DQA program and the product of that program did not meet the commission's regulatory requirements.
Inexplicably, though far more than two CQA failures have now been uncovered at Diablo Canyon, the NRC staff still refuses to draw a similar conclusion.
The fundamental problem at Diablo Canyon is the refusal of the applicant to take seriously the NRC's safety regulations.
I instead opines that its review of the "SWEC items" does not demonstrate "a significant deficiency in construction."
The Motion cites the SWEC review as evidence the COA program was deficient, a point the staff does not dispute.
39.
The commission relies on its regulations to assure the safety of design and construction.
Perhaps in the mistaken belief that the regulations are over-conservative, perhaps in reliance on the redundency of " safety in depth," PG&E has failed to ensure compliance with the NRC's quality assurance regulations in designing and building Diablo Canyon.
Now,_ faced with irrefutable evidence of its failure e
to meet the QA requirements for construction, PG&E again asks this board to judge it not by the regulations but rather by its own standards of safety.
The law requires more of an applicant for an NRC license; it requires careful compliance with all applicable regulations.
And the law requires of this board, when confronted with evidence this requirement has not been met, to ask not merely whether the known construction errors have been remedied but what can be done now to provide the assurance of safety the regulations require.
It is too late to ensure compliance with the regulations in-the construction of Diablo Canyon; too much construction has l
taken place without adherence to them.
Therefore some substitute must be found for their function.
Such a substitute cannot be determined unilaterally by the applicant -- nor by the applicant and the staff.
Any approval of the quality of construction at l
Diablo Canyon should be made only by an adjudicatory panel, in public hearings, based on evidence in the record, and subject to the full examination of all parties.
/
40.
l l
[
1 In the end, the Governor's Motion remains founded on sound common sense: the need to provide a valid assurance of safety in a power plant plagued by errors that continue to be detected.
The Governor urges this board to grant the Motion in order to restore public confidence in the plant-and in the regulatory procesa.
DATED:
June 17, 1983.
Respectfully submitted, JOHN K. VAN DE KAMP, Attorney General of the State of California ANDREA SHERIDAN ORDIN, Chief Assistant Attorney General MICHAEL J. STRUMWASSER, Special Counsel to the Attorney General SUSAN L.
- DURBIN, PETER H. KAUFMAN, Deputy Attorneys General By
~ K MAN
/
PETER H.
Attorneys for Gove or George Deukmejian 6
41.
mL i
i 0
l 1
0 I
h l
AFFIDAVIT OF RICHARD B. HUBBARD
.-.