ML20149H085
| ML20149H085 | |
| Person / Time | |
|---|---|
| Site: | Diablo Canyon  |
| Issue date: | 11/04/1994 |
| From: | Bechhoefer C, Kline J, Shon F Atomic Safety and Licensing Board Panel |
| To: | PACIFIC GAS & ELECTRIC CO. |
| References | |
| CON-#494-15901 92-669-03-OLA-2, 92-669-3-OLA-2, LBP-94-35, OLA-2, NUDOCS 9411100086 | |
| Download: ML20149H085 (210) | |
Text
(
~
I 1
(S@/
DOCK _ i t D USHRC 94 my -4 P 218 LBP-94-35 Offlil C bD /l 00CKETUG
.w ;. wni .4 " J1 UNITED STATES OF AMERICA NUCLEAR REGULATORY COMMISSION ATOMIC SAFETY AND LICENSING BOARD i(E3U52 60V - 41994 Before Administrative Judges:
Charles Bechhoefer, Chairman I Dr. Jerry R. Kline Frederick J. Shon In the Matter of Docket Nos. 50-275-OLA-2 50-323-OLA-2 PACIFIC GAS AND ELECTRIC COMPANY ASLBP No. 92-669-03-OLA-2 (Diablo Canyon Nuclear Power Plant, Units 1 and 2) (Construction Period Recovery)
Facility Operating 7' censes No. DPR-80 and DPh ' 2 November 4, 1994 INITIAL DECISION (Construction Period Recovery /Recanture)
Accearances David A. Renka, Esc., Kathryn M. Kalowsky, Esa., Washington D.C., and Christopher J. Warner, Esa. and Richard F. Locke, Esa., San Francisco, California, for Pacific Gas and Electric Co. (Applicant)
Diane Curran, Esa., Washington D.C., and Ms. Jill ZamEk, Pismo Beach, California, for San Luis Obispo Mothers for Peace (Intervenor)
Ann P. Hodadon, Esc., Arlene A. Jorcensen, Esc., and Catherine L. Marco, Esc., for the Nuclear Regulatory Commission Staff 9411100086 941104 'b' PDR ADOCK 05000275 g hD G PDR k
TABLE OF CONTENTS ,
I. Introduction . . . . . . . . . . . . . . . . . . . 1 II. Backcround . . . . . . . . . . . . . . . . . . . . 2 III. Standard of Review . . . . . . . . . . . . . . . . 6 A. General Recuirements. . . . . . . . . . . . . 6 B. Method of Proof. . . . . . . . . . . . . . . 13 C. Asserted Procedural Deficiencies. . . . . . . 16 IV. Contention I (Maintenance / Surveillance Procram) . 21 A. Scone of Contention. . . . . . . . . . . . . 22 I B. Descriotion of Maintenance / Surveillance Procram. . . . . . . . . . . . . . . . . . . 27 '
C. Evaluation of Evidence. . . . . . . . . . . . 35
- 1. Maintenance of Environmental ,
Oualification of Electrical Eauipment. 39
- 2. Check Valves /IST Deficiencv. . . . . . . 50
- 3. Cable Failures. . . . . . . . . . . . . 51
- 4. Wrona Size Motor Installed. . . . . . . 55
- 5. Storace and Handlina of Lubricants. . . 58
- 6. Fuel Handlino Buildino. . . . . . . . . 60
- 7. Tests of Containment Personnel Airlock. 63
- 8. Component Coolina Water (CCW) Heat :
Exchancer. . . . . . . . . . . . . . . . 66 ',
- 9. Auxiliary Buildinc Ventilation System Inocerable. . . . . . . . . . . . . . . 69
- 10. Restoration of Electrical Panels. . . . 71
- 11. Containment Eauipment Hatch. . . . . . . 74
- 12. Manual Reactor Trio Caused by Failure of a Fuse for the Rod Control System. . . . 76 ,
- 13. Limitoraue 2-FCV-37 Failed to Close. . . 78
- 14. Safety Iniection Emeroency Core Coolino l System (ECCS) Accumulator Tanks. . . . . 79
- 15. Corrosion of ASW Annubar. DFO and CO2 Pinina. . . . . . . . . . . . . . . . . 83
- 16. Control of Measurina and Test Eauipment (M&TE). . . . . . . . . . . . . . . . . 87
- 17. Centrifucal Charaina Pumo 1-1; Dearaded Coupling. . . . . . . . . . . . . . . . 90
- 18. Unit Shutdown Due to Inonerable Hich Pressure Turbine Ston Valve. . . . . . . 93
- 19. Diesel Generator 2-2 Failure to Achieve Rated Voltace. . . . . . . . . . . . . . 96 l
- 20. Missed Alert Frecuency STP for Auxiliarv j Salt Water Pumn 1-2 and Component Coolino Water Valve CCW-2-RCV-16. . . . 98 !
l ii j
i l
l 1
- 21. In-Service Prompt Test Data i Ouestionable. . . . . . . . . . . . . . 102
- 22. Hold-Down Motor Bolts On Centrifucal -
Charcina Pumns. . . . . . . . . . . . . 104
- 23. Reactor Coolant System Leakace. . . . . 107
- 24. Reactor Cavity Sumo Wide Rance Level Channel 942A-Inoperable. . . . . . . . . 111 i
- 25. Desian Criterion Memorandum (DCM) l Recuirements. . . . . . . . . . . . . . 115 j
- 26. Pine Support Snubber Damace. . . . . . . 117 !
- 27. Gas Decav Tank Missed Surveillance. . . 120
- 28. Seismic Clips Not Installed. . . . . . . 123
- 29. Containment Fan Coolina Unit (CFCU)
Backdraft Dampers. . . . . . . . . . . . 125
- 30. Control of Foreicn Material / Cleanliness / ,
Housekeepino. . . . . . . . . . . . . . 132 l
- 31. Steam Generator Feedwater Nozzle Crackina. . . . . . . . . . . . . . . . 141 ;
- 32. Procedural Controls Durina Shot Peenina Operations. . . . . . . . . . . . . . . 145
- 33. Unplanned Activation of Encineered Safety Features (ESF). . . . . . . . . . 148
- 34. Limitoraue Valve Failure. . . . . . . . 149
- 35. Motor Pinion Keys in Limitoraue Motor Operators. . . . . . . . . . . . . . . . 151
- 36. Control of Liftino and Riccina Devices. 153
- 37. Main Feedwater Pump Oversceed Trio Due To Failure of Power Sucolv to Soeed Sensina Probes. . . . . . . . . . . . . 158
- 38. Inadvertent Containment Ventilation Isolation. . . . . . . . . . . . . . . . 161
- 39. Reactor Trio on Steam Generator Low Level. . . . . . . . . . . . . . . . . . 164
- 40. Auxiliary Saltwater Pumo Crosstie Valve. . . . . . . . . . . . . . . . . . 167
- 41. Testcock Valve on Diesel Generator. . . 169
- 42. Main Feedwater Check Valve. . . . . . . 171
- 43. ASW Pumo Vault Drain Check Valves. . . . 174
- 44. Motor ODerated Valve Failed to Cycle on Actuation Sicnal. . . . . . . . . . . . 175
- 45. Fire in Electrical Panel. . . . . . . . 176
- 46. Chemical and Volume Control System Diachraam Leakace. . . . . . . . . . . 178
- 47. Conclusion on Maintenance and Surveillance Procram. . . . . . . . . . 179 V. Renewed Motion to Reonen the Record . . . . . . . 183 VI. Contention V (Thermo-Lao Interim Compensatory Measures) . . . . . . . . . . . . . . . . . . . . 190 .
iii P
f.
i r
VII. Conclusions of Law . . . . - . . . . . . . . . . . 198 i
VIII. Order . . .. . . . . . . . . . . . . . . . . . . . 200 r
?
I i
v i
k i
4 I
t r
t B
8
.=
s T
i e I IV i
=__-- -_ _ _ --_.
F LBP-94-35 UNITED STATES OF AMERICA NUCLEAR REGULATORY COMMISSION ATOMIC SAFETY AND LICENSING BOARD Before Administrative Judges:
Charles Bechhoefer, Chairman Dr. Jerry R. Kline Frederick J. Shon In the Matter of Docket Nos. 50-275-OLA-2 50-323-OLA-2 PACIFIC GAS AND ELECTRIC COMPANY ASLBP No. 92-669-03-OLA-2 (Diablo Canyon Nuclear Power Plant, Units 1 and 2) (Construction Period Recovery)
Facility Operating Licenses No, DPR-80 and DPR-82 November 4, 1994 INITIAL DECISION (Construction Period Recoverv/ Recapture)
I. Introduction This is an initial decision on Pacific Gas and Electric Company's (PG&E or Applicant) application to amend the operating licenses for its Diablo Canyon Nuclear Power Plant, Units 1 and 2 (DCPP or Diablo Canyon), located near San Luis Obispo, California, to allow for 40 years of operation dated from the issuance of its operating licenses.
For the reasons set forth herein, we conclude that, to the extent challenged in this proceeding, PG&E satisfactorily justified the license extensions it seeks and, subject to certain directions, as well as normal NRC Staff review, should be granted those extensions, u _ ___ _ _ _ _ _
I l
i l
II. Backcround On July 9, 1992, PG&E submitted a iAcense amendment request by which it sought to extend the life of its l operating licenses for the DCPP by more than 13 years (for Unit 1) and almost 15 years (for Unit 2) .by " recapturing" the period spent in constructing the plants. The licenses, j l
which are limited to a term of 40 years by section 103.c of l the Atomic Energy Act, 42 U.S.C. S 2133 (c) , were issued consistent with a then-extant Commission policy under which 1
that 40-year life extended from the date of issuance of the ,
I construction permit for a particular unit--for Unit 1, a term running from April 23, 1968, to April 23,.2008, and for l l
Unit 2, a term running from December 9, 1970, to December 9, 2010. ;
In 1982, the Commission began issuing the 40-year operating licenses measured from the date of issuance of the operating license. It has also approved license amendments ,
for many reactors conforming the earlier licenses to this new policy. The Applicant is here seeking to amend its operating licenses to take advantage of the newer practice.
As proposed, the extended expiration dates for DCPP would be i September 22, 2021 for Unit 1 and April 26, 2025 for Unit 2.
In response to a notice of opportunity for hearing on the proposed amendments, 57 Fed. Reg. 32,575 (July 22, 1992), San Luis Obispo Mothers for Peace (MFP) timely filed a request for a hearing / petition for leave to intervene.
c.
This Licensing Board was established to rule on the request / petition and to preside over the proceeding in the event that a hearing were to be ordered. 57 Fed. Reg.
43,035 (Sept. 17, 1992).
After a prehearing conference held in San Luis obispo, California, on December 10, 1992, at which we heard argument concerning MFP's petition and the Supplement in which MFP set forth its proposed contentions, together with PG&E's and the NRC Staff's responses each opposing admission of any of the contentions, we granted MFP's petition for leave to intervene and request for a hearing. Pacific Gas and Electric Co. (Diablo Canyon Nuclear Power Plant, Units 1 and 2), LBP-93-1, 37 NRC 5 (1993) (LBP-93-1); see also LBP-92-27, 36 NRC 196 (1992). We determined that MFP had standing and, of the eleven contentions proffered, we admitted portions of two of them: Contention I, challenging the adequacy of PG&E's maintenance and surveillance program, and Contention V, challenging the adequacy of PG&E's interim fire protection measures to compensate for defective "Thermo-Lag" passive fire barriers manufactured by Thermal Science, Inc.
Thereafter, MFP submitted three late-filed contentions.
Following a prehearing conference held on May 11-12, 1993 in NRC's then-Region V office at Walnut Creek, California, we rejected all three contentions. But we determined that portions of two of them could be litigated under the
_4 _
previously admitted maintenance and surveillance program contention, and that portions of another dealing with fire-protection had become moot as a result of steps already taken or planned by PG&E. LBP-93-9, 37 NRC 433 (1993).
We held. evidentiary hearings in San Luis Obispo, California on seven days, August 17-21, 23-24, 1993. The record was closed on August 24, 1993 (Tr. 2295). PG&E and ,
the Staff presented their cases through expert witnesses and i documents. MFP put on no witnesses but presented its case through cross-examination, based in large part on numerous PG&E rH1 NRC documents that MFP offered into evidence.
Thereafter all parties submitted timely proposed findings of fact and conclusions of law,2 and PG&E submitted a timely reply.2 ,
On February 25, 1994, MFP filed a motion to reopen the record, based on material appearing in an NRC inspection report provided to PG&E by the NRC Staff on January 12, <
2 PG&E's Proposed Findings of Fact and Conclusions of
- Law in the Form of an Initial Decision, dated October 8, 1993 (PG&E F0F) ; MFP's Proposed Findings of Fact and l Conclusions of Law Regarding PG&E's Application for a '
License Amendment to Extend the Term of the Operating License for the DCNPP, dated November 19, 1993 (MPP FOF);
NRC Staff's Findings of Fact and Conclusions of Law in the Form of an Initial Decision, dated December 22, 1993 (Staff FOF). j 2
PG&E's Reply Findings of Fact and Conclusions of Law, dated December 30, 1993 (PG&E Reply FOF). j 1
.. _ _ _ . _ _ _ _ _ ___ __ .~ _
l 1
I 1994.3 PG&E and the'NRC Staff filed responses on March 7 ,
and 14, 1994, respectively. By Memorandum and Order dated March 23, 1994, LBP-94-9, 3 9 NRC 122, we denied that motion, primarily because the matters in the inspection report ;
relied on by MFP were at that stage no more than " unresolved items" and because an affidavit of the NRC inspector I
involved in the inspection (one of the Staff's witnesses in this proceeding) stated that nothing in the inspection report was contrary to or inconsistent with his prior l
testimony. Our denial, however, was without prejudice to a l l
later motion to reopen based on any of the unresolved items demonstrated to be significant and to possess substantive I
implications for implementation of the maintenance and surveillance program.
On August 8, 1994, MFP filed such a motion.' PG&E and ]
i the Staff each opposed the reopening.5 We are denying the I motion for reasons spelled out in Part V of this Decision. ;
I l
)
3 This report was entered into the NRC's NUDOCS document storage and retrieval system on February 2, 1994 and hence became a publicly available document no later than that date.
' San Luis Obispo Mothers for Peace's Renewed Motion to Reopen the Record Regarding Pacific Gas and Electric Company's Application for a License Amendment to Extend the Term of the Operating License for the Diablo Canyon Nuclear Power Plant, dated August 8, 1994 (Renewed Motion).
5 PG&E's Opposition to San Luis Obispo Mothers for Peace Renewed Motion to Reopen the Record, dated August 23, 1994; NRC Staff Response to San Luis Obispo Mothers for Peace's Renewed Motion to Reopen the Record, dated August 29, 1994. !
i The Board addresses the contested issues below. We i j
have divided the remainder of'this opinion into six parts.
First, we describe the applicable legal standards for resolving the issues before us (Part III). Next, we address the contested issues in two parts, the first (Part IV) addressing Contention I (Maintenance and Surveillance Program) and the second (Part VI) addressing Contention V (Thermo-Lag Interim Compensatory Measures). These portions of the opinion include various findings of fact necessary to I
our conclusions on the respective issues. In Part V we ;
I spell out our reasons for denying the Renewed Motion, which l relates to the maintenance / surveillance issue immediately preceding it. In Part VII, we set forth conclusions of law, l
and, in Part VIII, our resulting Order. l III. Standard of Review A. General Reauirements.
Since its change in policy in 1982,.when it began measuring the term of operating licenses from the commencement of operation, the Commission has approved more than 60 recapture amendment requests. PG&E F0F, at 3; Tr.
2274 (Peterson). PG&E characterizes NRC's approval of recapture amendments as " routine" and " administrative in nature" inasmuch as it does not involve any alterations in l plant design or operation, or any new environmental impacts l
. -. . .- ~- - -
.? -
l not previously evaluated. PG&E FOF, at 3.' In fact, that description may be more reflective of the number of challenges to such approvals than to the nature of the approval itself: only one such comparable amendment request i
has been challenged. The only contention accepted in that proceeding concerned the adequacy of the surveillance and maintenance program, and it was finally settled (with some additional obligations attached to the extension). See Vermont Yankee Nuclear Power Corp. (Vermont Yankee Nuclear Power Station), LBP-90-6, 31 NRC 85 (1990).
In LBP-93-1, however, we in effect rejected PG&E's
" administrative change" designation. In the context of determining MFP's standing to participate, we noted that the
" risk of an accident with offsite consequences for an additional 13 to 15 years" represented injury in fact (notwithstanding the prior analyses of such accidents) and in substance undercut PG&E's claim that the proposed amendments were not "significant" but virtually ministerial. 1 37 NRC at 10-11. Our evaluation of the record evidence in l this proceeding reinforces our view of the significance of the amendments.
In seeking denial of the license amendments here in issue because of asserted deficiencies in the maintenance / surveillance program (Contention I), MFP points 4
'The Staff makes no attempt to characterize the significance of the recapture proceeding.
out that there are currently no detailed regulatory requirements prescribing conditions for such programs.
Simila:1y, it notes that there are no regulations for evaluating the adequacy of implementation of a maintenance / surveillance program in terms of past performance. MFP FOF $ 13. The Staff essentially agrees, pointing out that the NRC's maintenance rule, 10 C.F.R.
S 50.65, does not become effective until 1996. Staff FOF i I-6.
Normally, in evaluating the adequacy of a program such as the maintenance / surveillance program, a Board would look to standards appearing in regulations. Absent such regulations, MFP would rely generally on Section 182.a of the Atomic Energy Act (AEA) of 1954, as amended, 42 U.S.C.
S 2232 (a) , which provides that an applicant for a reactor operating license must submit sufficient information for the NRC to find that the facility will " provide adequate protection to the health and safety of the public."
Further, MFP references S 103.d of the Act, 42 U.S.C.
S 2133 (d) , providing that the NRC may not issue a license that would be " inimical to the . . . health and safety of the public." These statutory standards are reflected in 10 C.F.R. SS 50.57 (a) (3) and (6), which specify in pertinent part that NRC may issue operating licenses upon finding that there is " reasonable assurance (i) that the activi ties authorized by the operating license can be conducted without i
endangering the health and safety of the public, and (ii) that such activities will be conducted in compliance with
[NRC) regulations" and that issuance of the license "will not be inimical to the . . . health and safety of the public." MFP adds that, although absolute perfection is not required, " reasonable assurance" may not be tainted by cost or risk-benefit considerations, citing Union of Concerned Scientists v. NRC, 824 F.2d 108 (D.C. Cir. 1987). MFP FOF i 12.
For its part, PG&E relies generally upon the
" reasonable assurance" standard. PG&E FOF, at 10. Both it t and the Staff further cite various codes, standards, ,
I regulatory guides and technical specifications dealing with the maintenance of particular equipment.
These standards do not, however, define an adequate ]
maintenance / surveillance program, although the degree to l which they are achieved may constitute a reasonable measure of program adequacy. As the Commission observed (in adopting the S 50.65 requirements to be effective in the future):
[t] he Commission's current regulations, regulatory guidance, and licensing practice do not clearly define the Commission's expectations with regard to ensuring the continued effectiveness of maintenance programs at nuclear power plants.
56 Fed. Reg. 31,306, 31,308 (July 10, 1991). Thus, we will refer to such standards as guidelines to determine whether l
the maintenance / surveillance program is performing its intended function.
MFP further references the guidance provided by INPO 90-008, Rev. 01, Maintenance Programs in the Nuclear Power Industry (March 1990) (MFP Exh. 4), as helpful in defining !
the scope of issues that a maintenance program must address I i
in order to provide adequate protection to the public health l and safety. MFP F0F i 14. PG&E agrees, claiming that the evidence supports PG&E's compliance with those guidelines. l l
PG&E Reply FOF i R9. The Staff points out that, in declining to adopt regulations defining a maintenance program, the Commission specifically declined to adopt INPO 90-008. Staff FOF $ I-8, referencing 54 Fed. Reg. 50,611 (December 8, 1989). The Staff acknowledges, however, the ,
usefulness of the INPO 90-008 standards in looking at the elements of a comprehensive maintenance program. ,
i The general safety provisions of the Atomic Energy Act and implementing general regulations are the ultimate standards against which to evaluate the amendments. The standards of INPO 90-008 and the other material cited by various parties are useful as guidance. In determining the adequacy of PG&E's program, we will refer to all these standards to determine what appears to us to constitute an adequate program sufficient to provide reasonable assurance of public health and safety for the extended operation j period of the proposed amendments.
l l
i With respect to the maintenance and surveillance program, the Applicant has maintained that the Commission's specific criteria for dealing with emergency preparedness exercise issues are germane and that we might give weight only to maintenance deficiencies that indicate a
" fundamental flaw" or " failure of an essential element" of the maintenance / surveillance program. PG&E FOF, at 7. We earlier disagreed in essence with that conclusion and we still do. As we have held, numerous or repetitive incidents i may coalesce to indicate a significant deficiency in the program. LBP-93-1, 37 NRC at 19-21.
" Fundamental flaw" was promulgated to deal with a specific circumstance involving an emergency planning exercise that would occur only at the time a plant was ready for operation. Litigation of the results of such an i 1
I exercise could delay operation, and the concept of
" fundamental flaw" was developed to keep delay to a minimum by limiting the scope of litigation. Lona Island Liahtina Co. (Shoreham Nuclear Power Station, Unit 1), CLI-86-11, 23 NRC 577 (1986). Here, the additional time needed for thorough consideration will not delay operation--
technically, the extension is not needed until at least i l
2008.
For all of these reasons, we have declined to follow the criteria favored by PG&E. Our decision will consider
each of the alleged deficiencies propounded by MFP and give it the weight that we consider it deserves.
With respect to the issue of interim compensatory measures for Thermo-Lag (Contention V), MFP relies on a series of Information Notices issued by the Staff. It also cites several NRC Bulletins. MFP FOF 11 786-87. We see no reason not to rely on this material, interpreted in accord with the " reasonable assurance" standard that also governs this issue.
MFP contends (and we agree) that the burden of proof falls on the Applicant. MFP further asserts that PG&E has not satisfied that burden. MFP claims the asserted deficiencies in the maintenance / surveillance program t1 .t it has demonstrated (as well as asserted deficiencies in PG&E's implementation of interim Thermo-Lag corrective actions) require denial of the proposed amel.dments. MFP FOF $$ 2, 839.
PG&E and the NRC Staff assert, to the contrary, that the maintenance / surveillance program is adequate--indeed, exemplary--and that the statutory and regulatory standards referenced by MFP are nerforce satisfied. PG&E and the Staff claim, and MFP concedes (MFP F0F i 12), that perfection in a program is not required, given the
" reasonable assurance" standards of the Atomic Energy Act.
Although acknowledging that the ultimate burden of proof falls on PG&E, the Staff (and to a lesser extent, PG&E)
further claim that MFP has the burden of going forward with evidence, which (in their view) it has failed to do. Both PG&E and the Staff also find PG&E's implementation of Thermo-Lag interim corrective measures to be adequate.
B. Method of Proof.
PG&E and the Staff each presented witnesses on both the maintenance / surveillance and the Thermo-Lag issues. MFP did not sponsor any witnesses but, instead, developed its case (as is permissible') through cross-examination of PG&E and Staff witnesses and documents offered through them.
The NRC Staff takes the position that the portions of MFP's proposed findings that include technical analyses, opinions and conclusions may not be " adopted" by us, inasmuch as " technical analyses, opinions and conclusions in NRC proceedings must be sponsored by experts who can testify to the soundness of the conclusions set forth." Staff FOF
$ 2. The Staff relies primarily on Rpke Power Co. (William B. McGuire Nuclear Station, Units 1 & 2), ALAB-669, 15 NRC 453, 477 (1982) and Southern California Edison Co. (San Onofre Nuclear Generating Station, Units 2 & 3), ALAB-717, 17 NRC 346, 367 (1983).
' Tennessee Vallev Authority (Hartsville Nuclear Plant, Units 1A, 2A, 1B & 2B), ALAB-463, 7 NRC 341, 356 (1978);
Commonwealth Edison Co. (Zion Station, Units 1 and 2), ALAB-226, 8 AEC 381, 389 (1974); Wisconsin Electric Power Co.
(Point Beach Nuclear Plant, Unit 2), ALAB-137, 6 AEC 491, 504-05 (1973).
m _. . - _
t 4
14 -
However, neither of these cases supports the Staff's- I.
position. In McGuirg, the Appeal Board ruled that a .;
Licensing Board had not erred in declining to admit documents into evidence when there was no competent expert i witness to sponsor them. 15 NRC at 477. Similarly, in San ,
onofre, the Appeal Board would not allow portions of an applicant's Final Safety Analysis Report to be considered as l substantive evidence when the applicant provided-no ,
witnesses for cross-examination on the document. In so i
ruling, the Appeal Board concluded that there was "no basis' l
for allowing applicants to avoid cross-examination on a document of central importance that they themselves had prepared." 17 NRC at 366.
Both of these cases dealt with document reliability and the need for documents to be verified by competent witnesses before they can be admitted into evidence. In contrast, in this case the documents relied on by MFP were accepted into evidence after being introduced through PG&E's or the Staff's expert witnesses. Once admitted into evidence, MFP 1
was entitled to use them in its proposed findings.'
The' result of adopting the Staff's argument would be that, in reaching technical conclusions, the Board would be limited to relying on expert testimony of witnesses--all 1
eThe Staff also cites Louisiana Power and Licht Co. I (Waterford Steam Electric Station, Unit 3), ALAB-732, 17 NRC l 1076, 1088 n.13 (1983), involving prepared direct testimony )
of an expert, which is distinguishable on the same basis. i
i sponsored by PG&E or the Staff and essentially reaching the !
l same conclusion. This would eliminate or seriously abrogate the right of MFP to present its case through cross-examination. We do not read the Commission's rules or i decisions as either requiring or even permitting this result. rederal agencies, and this Board as the delegee of such an agency, are required to do more'than act as an !
" umpire blandly calling balls and strikes for adversaries {
appearing before it." Scenic Hudson Preservation Conference
- v. F.T.C., 354 F.2d 608, 620 (2d Cir. 1965).' This Board includes technical experts who can evaluate the factual material in the record and reach their own judgment as to its significance.
Early during the hearing, PG&E objected generally to MFP's introduction of documents intended to show }
deficiencies in PG&E's maintenance / surveillance program, on l
the ground that the "uncontroverted evidence in the direct t testimony" affirmed the adequacy of such program. Tr. 597.
We overruled that general objection on the basis that we should have the opportunity to evaluate the significance of the documents and the adequacy of the program. Tr. 597-99.
We permitted PG&E to offer objections to specific documents
'See also Texas Utilities Generatino Co. (Comanche Peak Steam Electric Station, Units 1 and 2), LBP-82-87, l 16 NRC 1195, 1199 (1982). '
l
on grounds such as relevance to the operation of the maintenance / surveillance program. Tr. 600.
C. Asserted Procedural Deficiencies.
The NRC Staff raises a question concerning the advance notice it received of exhibite that MFP intended to introduce. The Staff points to the requirement that written testimony be provided the parties fifteen (15) days prior to the commencement of the hearing (10 C.F.R. S 2.743 (b)) and would extend that requirement to apply as well to the identification of exhibits. It faults MFP for not adhering to these standards and concludes that documents that other parties have not had an adequate opportunity to examine, and cross-examination based thereon, may not serve as a basis for our findings. Staff FOF $$ 5-6.
We recognize the difficulties faced by the Staff in attempting to formulate its position in the absence of adequate notice from other parties of their position on issues. The 15-day testimony rule reflects, in part, the Staff's needs in this regard. Nonetheless, we decline to adopt the Staff's suggested approach to the record. No such extension of the testimony-filing rule to documents appears in the rules, either expressly or by implication. Indeed, where credibility of witnesses is at stake (as it was in certain instances in this proceeding, at least one of which was the subject of the Staff's complaint), it would undercut the utility of the cross-examination were the documents to
be revealed in advance to the opposing party. (That is the rationale for. revealing cross-examination. plans only to the Board prior to cross-examination. 10 C.F.R. S 2.743 (b) (2) . )
In addition, we recognize, of course, that MFP's actions in this regard were driven not by any intent to ignore procedural requirements or make it difficult for the parties (or Board) to become adequately familiar with MFP's !
case. MFP was merely forced because of inadequate financial resources to present its case as best it could. A review of i
the procedural developments in this case will place in context the Staff's complaints.
It appears that, until shortly before the start of the evidentiary hearing, MPP had planned to present one or more witnesses. Our early scheduling orders, and MFP's early discovery responses, all anticipated that MFP would provide witnesses and file direct testimony. See, e.g., Memorandum and Order (Discovery and Hearing. Schedules), dated February 9, 1993, unpublished, at 5; [MFP] Responses to First Set of Interrogatories and Request for Production of Documents Filed by [PG&E] and Motion for Protective Order, dated March 22, 1993; Prehearing Conference Order (Late-filed Contentions and Discovery), LBP-93-9, 37 NRC 433, 453, n.42 (1993).
Indeed, during the course of discovery, MFP made use of ;
several technical consultants, whom it identified. Sge, e.g., letter from Jill ZamEk, MFP, to Licensing Board, dated
1 i
1 1
i April 2, 1993. As late as May 6, 1993, it advised that it was " working with limited resources" and "would expect to be l
able to identify its expert witnesses" in the near future.
[MFP]. Reply to [PG&E's] Motion to Impose Duty on MFP to Supplement Responses to Interrogatories and Requests for Production of Documents.
Not until June 21, 1993, when we established the final filing date for testimony as August 2, 1993 (see Memorandum and Order (Notice of Prehearing Conference and Evidentiary 2 Hearing), dated July 8, 1993, 58 Fed. Reg. 589"i4 (July 14, 1993)) did MFP advise that it had "no commitment from any pere a to appear as an expert witness at the hearing."
[MFP] Supplemental Response to First and Second Sets of Interrogatories and Requests for Production of Documents filed by [PG&E]. It then agreed to provide a list of ,
documents on which it would rely and, as a part of that filing, provided a listing that identified many of them."
During the foregoing phases of the proceeding, MFP had not been represented by counsel. On July 28, 1993, counsel representing MFP filed a " Notice of Appearance" (which was provided the Board and parties by telefax). In a telephone conference call on July 29, 1993, the Board, as provided by 10 C.F.R. S 2.743 (b) , directed the parties to provide the "Although the list was provided as a partial response to PG&E discovery, the Staff was on the service list of the response.
.19 -
Board cross-examination plans (covering the first week of the hearing) by August 16, 1993 (the day prior to the start of the hearing). Memorandum and order (Telephone Conference Call,.7/29/93), dated August 3, 1993. All parties did so."
MFP's cross-examination plan indicated that MFP was still in the process of reviewing documents and that its plan would be amended to reflect any additional documentation. Simultaneously, on August 16, 1993, MFP distributed (to all parties and the Board) a document--
termed a " road map" (Tr. 578)--that in effect supplemented the cross-examination plan and identified 219 exhibits that MFP intended to introduce concerning Contention I, grouped in accordance with specified topics." (During the hearing, MFP determined not to offer certain of these documents, and furtner offered a few additional documents that it had not earlier been able to identify. Moreover, at least one series of documents bore on the credibility of "In accordance with 10 C.F.R. S 2.743 (b) (2) , the cross-examination plans were submitted only to the Licensing Board and not made available to opposing parties. In accord with that same regulation, we are providing the various plans to the Commission's Secretary for inclusion in the record of the proceeding.
"MFP further described the " road map" at an August 17, 1993 prehearing conference immediately preceding the evidentiary hearing. See Tr. 578-82. The topics in the
" road map" were not co-extensive with those identified in MFP's June 21, 1993 discovery response.
certain PG&E witnesses and could not have been revealed in advance to other parties without' undermining their utility.)
For a party proceeding by cross-examination only, the ,
"rosd map" would represent a fair substitute for the 15-day filing requirement. Such a document, however, could not have been submitted at the same time as other parties' prepared testimony inasmuch as its formulation depends upon the content of the other parties' direct testimony. Because we heard no. objection to our ruling of simultaneous submission of prepared testimony, we in effect made it impossible for MFP to observe the 15-day time frame for filing. In fact, given the change in MFP's representative, the filing on the date prior to the hearing was about as timely as could reasonably be effected. Additionally, we made it clear that we would afford parties (or the Board) additional time within which to examine documents offered into evidence that they had not previously had an adequate time to examine. Tr. 581-82 (Licensing Board). To the
. extent that additional time for reviewing documents may have been sought, we made every attempt to grant such requests.
Given this history, we are declining to take the course of action sought by the Staff. Although it may have been more difficult for the Staff to prepare its case, we do not believe that the Staff was prejudiced by the late )
identification of certain documents, particularly in view of our offer to provide sufficient time to review the documents
l l
and the circumstance that Staff witnesses were not called upon to testify until late in the hearing, giving them time )
to review documents presented earlier. See, e.g., Tr. 2183-89 (re: MFP Exh. 5); Tr. 2226 (re: MFP Exhs. 105-108).
Moreover, we would prefer to base our rulings on the potential safety significance of the case MFP presented, not' !
l on procedural technicalities that would eliminate from the. !
record essentially all information contrary to the virtually single view being espoused by PG&E and the Staff.
i We turn now to the two contentions before us.
IV. Contention I (Maintenance / Surveillance Procram)
Contention I reads as follows:
The San Luis Obispo Mothers for Peace contends that Pacific Gas and Electric Company's proposal to extend the life of the Diablo Canyon Nuclear Power Plant for more than 13 years (Unit 1) and almost 15 years (Unit i
- 2) should be denied because PG&E lacks a sufficiently effective and comprehensive surveillance and maintenance program.
LBP-93-1, 37 NRC at 14.
This contention was addressed by a panel of witness from PG&E consisting of: Bryant W. Giffin, Manager of l Maintenance Services (DCPP) ; William G. Crockett, Manager of !
Technical and Support Services (DCPP) ; David A. Vosburg, Director of the Work Planning Section, Maintenance Services Department (DCPP); Steven R. Ortore, Director of the Electrical Maintenance Section, Maintenance Services Department (DCPP); Tedd Dillard, Supervisor of Component Programs for the Nuclear Division of Florida Power & Light i
i I
l 1
Company; and David B. Miklunh, Manager of Operational Services (DCPP) ." The NRC Staff presented testimony of a 1
panel consisting of: Pa.ul P. Narbut, Regional Team Leader, l
Region V, Division of Reactor Safety and Projects; Mary H. l 1
Miller, Senior Resident Inspector (DCPP) , Region V; and Sheri R. Peterson, Senior Project Manager (DCPP), Office of l l
Nuclear Reactor Regulation." All of the foregoing Applicant and Staff witnesses were qualified for their particular testimony.
As described earlier, MFP presented no witnesses but instead chose to rely on numerous exhibits (and cross-examination based thereon) consisting of PG&E's internal Nonconformance Reports (NCRs), Licensee Event Reports (LERs) filed with the NRC, PG&E correspondence with the NRC, and NRC Staff Inspection Reports (irs) and Notices of Violation (NOVs). Sg2 Tr. 576-79.
A. Scone of Contention.
In evaluating the adequacy of the maintenance and surveillance program, we initially consider the appropriate
" Testimony of Pacific Gas and Electric Company i Addressing Contention Ii Maintenance and Surveillance, admitted but not bound in, Tr. 590 (PG&E Test.).
P "NRC Staff Testimony of Paul P. Narbut, Mary H. Miller and Sheri R. Peterson Regarding Contention 1: The Surveillance and Maintenance Program at Diablo Canyon, ff.
Tr. 2159 (Staff Test.). As a result of a recent NRC reorganization, Region V has become a part of Region IV.
scope of the program to be evaluated. Here the parties-differ significantly.
PG&E defines " maintenance" as those activities which are performed to assure that structures, systems and components ("SSCs") will continue to operate as designed, as well as those activities necessary to repair or replace SSCs that are degraded or cannot perform the intended function.
Maintenance is considered to be the aggregate of actions at DCFP that (1) minimizes the degradation or failure of SSCs, and (2) promptly restores the intended function of SSCs if they experience operability or functional problems.
PG&E Test., at 4; PG&E FOF $ M3.
It defines " surveillance" as the aggregate of periodic tests and/or inspections that verify that SSCs continue to function in accordance with predetermined specifications or are in a state of readiness to perform their particular safety functions.
Surveillance activitiea can trigger maintenance activities based upon the results of the particular tests or inspections.
PG&E Test., at 5; PG&E FOF i M3.
For its part, the NRC Staff defines " maintenance" and
" surveillance" collectively, as the work of keeping something in suitable condition.
There are basically two kinds of maintenance, preventive and corrective. Preventive maintenance is regularly scheduled work performed on structures, systems or components that keeps failures from occurring due to predicted component degradation.
Industry-wide operating experience is often taken into account in determining what type preventive maintenance is necessary and how often it should be performed.
Corrective maintenance is performed after a failure occurs, or a component exhibits degraded capability.
The surveillance tests are conducted to identify failures or degraded performance that needs to be corrected prior to a system being called upon to perform a safety function.
NRC Staff Test., ff. Tr. 2159, at 2; Staff FOF i I-5.
On the other hand, MFP urges that the somewhat broader !
definition set forth in INPO 90-008 would be helpful in defining the scope of issues that a maintenance program must address. That definition reads:
the aggregate of those actions that prevent the degradation or failure of, and that promptly restore the intended functions of, structures, systems, and components. As such, maintenance includes not only the activities traditionally associated with identifying and correcting actual or potential degraded conditions (that is repair, surveillance, and other preventive measures), but also extends to supporting functions for the conduct of these activities. Examples of these functions include engineering support of maintenance; operator identification of material deficiencies; and some aspects of chemistry control, radiological protection, and training.
MFP Exh. 4, INPO 90-008, Section 1 (" Elements of Maintenance"), at 1.
Given the current lack of any prescribed regulatory definition of maintenance or surveillance, we have considered (against the background of both the witnesses' testimony and the various activities and incidents advanced by MFP) the appropriate scope of maintenance and surveillance in achieving the public health and safety goals set forth in the AEA and NRC regulations. We conclude that the INPO 90-008 definition most suitably defines the scope of a program likely to achieve these results--with some limited exceptions necessitated by the somewhat ambiguous INPO 90-008 definition. Our use of INPO 90-008 in this manner appears to be consistent with the acknowledgement by l
l 1
i I
l PG&E and the Staff that the INPO 90-008 guidance is useful in outlining the subject areas that a program should encompass. As stated by PG&E, "the nuclear industry and the NRC have generally agreed that [INPO 90-008, Rev. 1, March 1990] identifies the requisite elements for a comprehensive .
maintenance program." PG&E Test., at 23 (Giffin).
The INPO 90-008 definition, by including supporting functions, appears to encompass many of the activities and ,
programs not falling within the other definitions but actually employed at DCPP for maintenance or sufveillance purposes, as described below. The INPO definition thus counters the tendency expressed in certain instances by PG&E ,
i and the Staff to " write off" maintenance-type failures or j deficiencies as the responsibility of, e.g., engineering or ;
operations and hence not attributable as a deficiency in l
maintenance or surveillance. In other words, by defining l the " maintenance / surveillance" function narrowly, failures 1
can be allocated to some other " box." The resulting maintenance / surveillance record will thus necessarily appear better than it really might be. By including supporting functions, however, we are only including those aspects of the function (such as operations) . directly encompassing maintenance and surveillance type activities.
That the Commission earlier declined to adopt the INPO 90-008 standard is of no moment, for the Commission declined !
t 4 to adopt any particular standard. It explicitly pointed to
- __-___--______-__A
i E
the INPO 90-008 standard because that was one of the few l i
reasonably well-defined standards that the Commission had considered adopting. It referred generally to other i
standards. The Commission also explained its lack of adoption of INPO 90-008 on the basis not of any disagreement with the standard but rather on its desire to de-emphasize programmatic elements and to require all licensees to monitor the effectiveness of maintenance activities. 56 Fed. Reg. 31,306, 31,308, 31,312 (July 10, 1991). By using i
the INPO 90-008 definition, we are not endorsing any or all -
of the programmatic elements but only considering it in terms of program scope, for which it seems reasonable guidance. Given the current lack of regulatory definition r
of a maintenance and surveillance program, the INPO 90-008 i definition appears to provide a meaningful standard for ascertaining the appropriate scope.(i.e., degree of W
coverage) of such a program."
"No party brought to our attention-in this proceeding NRC Regulatory Guide 1.160, " Monitoring the Effectivenesc of Maintenance at Nuclear Power Plants" (June 1993). That Guide is intended to assist utilities in implementing the requirements of NRC's maintenance rule. Use of the guide is currently permitted to determine methods to comply with :
maintenance requirements. The Guide incorporates by '
reference the definition of maintenance set forth in NUMARC 93-01, " Industry Guideline for Monitoring the Effectiveness of Maintenance at Nuclear Power Plants" (May 1993), at .
Appendix B. That definition closely parallels the INPO-90-008 standard that we are using as a guideline, ,
particularly insofar as it incorporates supporting functions. NUMARC 93-01 directly states (at 6) that its scope includes "SSCs that directly affect plant operations, regardless of what organization actually performs the
B. Description of Maintenance / Surveillance Program.
To understand the significance of the various incidents to which MFP refers, we first describe the scope and extent of maintenance- and surveillance-type activities at DCPP.
Based on the summary provided by the Staff in its proposed findings (Staff FOF 11 I-11 through I-27), which essentially is not challenged by MFP or PG&E, the maintenance and surveillance program at DC"o encompasses aspects of several activities, including (1) surveillance required by Technical Specifications; (2) equipment surveillance not required by the operating license; (3) Inservice Inspection (ISI) and Inservice Testing (IST) Programs; (4) the Environmental Qualification (EQ) Program; and (5) the Maintenance Program.
Surveillance testing required by DCPP Technical Specifications has been developed and implemented by PG&E in accordance with the industry standard ANSI N18.7-1976/ANS 3.2, " Administrative Controls and Quality Assurance for the Operation Phase of Nuclear Power Plants." Such surveillance testing is administratively controlled in accordance with PG&E procedure NPAP-C3, " Conduct of Plant and Equipment Tests." Technical Specification surveillance testing at DCPP is done to assure that safety-related equipment maintenance activities." That neither the Applicant nor Staff may approve of such standard (as reflected in their positions in this proceeding) is not a valid reason for their failing to advise us why its inclusion of supporting functions has not been followed by them.
failures or substandard equipment performance will not remain undetected. PG&E Test., at 10-12 (Giffin, Crockett).
In addition,.there are additional plant activities intended to provide current information about the condition of SSCs at DCPP, such as routine plant operator equipment inspections or walkdowns, predictive maintenance program ;
testing, preventive maintenance program inspections, r
procedure functional checks, performance tests of equipment not controlled by Technical Specifications, erosion / corrosion monitoring, post-maintenance testing, and system engineer walkdowns. These are activities clearly falling within the INPO 90-008 definition, although not ,
necessarily within the definitions advanced by other parties. Maintenance tasks are scheduled and performed on the basis of information derived from these activities in order to maintain equipment performance at the required level for the life of the plant. PG&E Test., at 12-14 (Crockett, Giffin).
The ISI and IST programs are designed to meet the requirements of 10 C.F.R. SS 50.55a(b) (2) and 50.55a(g), as well as plant Technical Specifications, and include inspection, testing, and maintenance of pressure-retaining components as required by the American Society of Mechanical Engineers (ASME) Boiler and Pressure Vessel (B&PV) codes.
Commission regulations (10 C.F.R. S 50.55a(g)) require revision of ISI and IST Programs as necessary to comply (to
l the extent practical within the limitations of design, geometry, and materials for construction of components) with the edition of the ASME B&PV Code and Addenda in effect and adopted by the NRC twelve months prior to the start of each ten-year inspection interval. These programs are in place to ensure that pressure-retaining components will be adequately inspected, tested, and maintained throughout the term of the operating license. PG&E Test., at 14-16 (Crockett).
Post-maintenance and post-modification testing (PMT) also is encompassed within the scope of PG&E's maintenance program. The primary objective of PMT is to ensure that all plant equipment that has undergone maintenance or modification has been demonstrated to be fully functional or operable prior to being returned to service. PMT at DCPP consists of two types of testing: (1) Maintenance or Modification Verification Tests performed by the implementing organization without actually operating the equipment; and (2) Operability Verification Tests to prove l Technical Specification operability. Documented completion of the required PMT is an essential element of the equipment ,
i control process used by the Operations _ Department when $
returning equipment to service. PG&E Test., at 51-53
]
(Crockett, Vosburg). I The planning and scheduling of maintenance tasks at DCPP is part of a work control process to enable maintenance
activities (including both preventive and corrective.
maintenance tasks) to be planned and performed in a safe, timely, efficient and controlled manner. Administrative controls for the work control process at DCPP are integrated into the computerized Plant Information Management System (PIMS), thereby making the data available on a plant-wide basis and ensuring that maintenance tasks receive appropriate levels of review and are tracked through final resolution. PG&E Test., at 43-45 (Vosburg).
6 PIMS provides integrated access to up-to-date information on plant component history, maintenance task instructions and history, problem reports and status, inventory control, and radiation exposure tracking. PG&E Test., at 20-21 (Crockett). It also provides a means by which all personnel working at DCPP can document plant equipment problems and request interdepartmental support.
Id. at 44-45 (Vosburg).
PG&E uses a reporting system to document failures or degradation of SSCs consisting of Nonconformance Reports (NCRs), which are a mechanism for personnel to document certain deficiencies at Diablo Canyon, initiate corrective action and establish a completion schedule for resolution of I
a nonconformance. In Diablo Canyon procedure OM7.ID3, PG&E 1 l
defines a nonconformance as: "A quality problem that constitutes a significant condition adverse to quality."
Staff Test., ff. Tr. 2159, at 3 (Peterson).
J
To be classified as a nonconformance, the quality problem must satisfy one or more of eight criteria outlined in the procedure. The criteria include NRC violations, programmatic or implementation breakdowns, design ;
deficiencies, defects, frequently recurring events and NRC reportable events. If the problem identified is determined to be a nonconformance, the NCR documents the event description, root cause determination, safety analysis, and action taken to correct the nonconformance and prevent recurrence. Staff Test., at 3 (Peterson).
A detailed root cause analysis program provides for the systematic analysis of unplanned occurrences pertaining to maintenance. This root cause analysis program is controlled
~
by Procedure NPAP C-26, " Root Cause Analysis," and provides guidance in several techniques including cause and effect j analysis, change analysis, event and causal factors analysis, barrier analysis, and human factors surveys. PG&E also tracks component histories as part of root cause analysis and component failure trending at DCPP. PG&E Test., at 59-60 (Giffin). There are approximately- 187,000 components in the DCPP Component Data Base, each with its own maintenance history available in PIMS. Component experience also is available from an industry-wide database, the Nuclear Plant Reliability Data System (NPRDS),
l maintained by INPO. Id., at 60-61 (Crockett).
1
l l
I Licensee Event Reports (LERs) are submitted to the NRC pursuant to 10 C.F.R. S 50 73. The threshold for requiring an LER is higher than for an NCR. Typically, the Applicant issues anywhere from 60 to over 100 NCRs each year compared !
i to 20 to'30 LERs each year. Whereas NCRs are used internally at Diablo Canyon, LERs are placed in the Applicant's public docket and used by the NRC for trending purposes and identifying significant events. Both reports are means for licensees to document self-identified problems. Staff Test., ff. Tr. 2159, at 3 (Peterson).
Management of equipment aging is inherent in many of the maintenance and surveillance activities discussed above.
However, PG&E has also initiated other programs expressly directed at aging management issues. Those programs and activities include the preventive, predictive and corrective maintenance programs; surveillance test programs; fatigue monitoring; the Environmental Qualification (EQ) program; the Reactor Vessel Embrittlement Management Plan; the Motor
- Operated Valve (MOV) testing and evaluation program; the Steam Generator Strategic Management Plan; the i
erosion / corrosion program; and the structural monitoring program. Each of these programs and activities produces specific results or corrective actions to maintain and/or restore equipment to its required performance level whether the aging occurred prior to or during plant operation. PG&E Test., at 62-64 (Giffin).
-}
i Since beginning plant operation PG&E has also made a number of major plant modifications to improve reliability or upgrade safety-related equipment that also help minimize 1
the effects of age-related degradation on the plant over its 40-year design operating life. Among these modifications i
are copper removal, including replacement of all feedwater heaters and retubing of all moisture separator reheaters; addition of a Condensate Polisher System; steam generator blowdown rate increase; fuel design improvements; removal of Boron Injection Tanks; reduction of the boron concentration in the Boric Acid System; installation of a digital Feedwater Control System; Chlorination System modifications; and installation of an on-line fatigue monitoring system.
PG&E Test., at 65-69 (Giffin).
PG&E also has established an aging management program pursuant to Program Directive TS1, " Plant Aging Management,"
that addresses age-related degradation over the course of the plant's operating life. This program collects data from new research findings, industry operating experience, the NRC, the Electric Power Research Institute (EPRI), and vendors, for inclusion in appropriate programs. .PG&E Test.,
at 70 (Giffin).
PG&E's aging management activities also include several special maintenance programs which have been established to monitor and manage certain critical components subject to complex aging mechanisms, as well as certain designated t
r
i 1
components with a limited life. For example, steam generator tube degradation is monitored and managed by ;
careful chemistry control during operation and by an extensive cleaning and inspection program during each refueling outage. PG&E Test., at 72 (Giffin).
The reactor pressure vessels at DCPP are also addressed by special maintenance programs. The DCPP Reactor Vessel I
Radiation Surveillance Program is designed to monitor changes in material and mechanical properties of DCPP reactor pressure vessels (RPVs) in order to ensure their l
. continued safe operation throughout the operating life of the plant. Compliance with all NRC regulations governing .;
RPV integrity was documented in PG&E's response to Generic Letter 92-01. PG&E Test., at 75-76 (Giffin, Crockett). )
I Erosion / Corrosion (E/C), which refers to the process of I l
wall thinning in susceptible piping or other pressure boundary components caused by the flow of water or wet steam, is a normal part of the nuclear power plant aging process. The management of E/C is an integral part of maintenance at DCPP. Measures to control E/C include the replacement of certain piping with E/C resistant material 1
such as stainless or chrome-moly steel. PG&E Test., at 77-78 (Crockett). ;
PG&E's testimony also indicates that it has a program for managing the aging of passive, long-lived structural !
concrete and steel at DCPP. Conditions such as spalling or l
= - m w
cracking of concrete, corrosive or caustic attacks from leaks, spills, or exposure to the environment, mechanical damage, and rust are routinely identified and reported by plant personnel. PG&E Test., at 81-82 (Giffin). PG&E's maintenance work at the intake structure is an example of this process. Tr. 1737-38 .(Giffin). This type of maintenance is meant to assure that these structures will perform their intended functions for the life of the plant.
Tr. 1741-42 (Giffin). For safety-relatcd structures, functional surveillance requirements are specifiel in Technical Specifications. Periodic surveillance testing verifies the operability of these structures. PG&E Test.,
at 82 (Giffin).
C. Evaluation of Evidence.
Turning to the merits of the contention, we find that.
all of the expert witnesses--both PG&E and the Staff--
testified as to the adequacy and indeed the excellence of PG&E's maintenance / surveillance program. As summed up by the Staff:
The performance of maintenance and surveillance at Diablo is considered to be superior and clearly supportive of safe facility operation. Their performance has been, at worst, good and has improved over the years. Gradual trends over the past seven years show a reduction in the number of equipment failures, reduction of significant safety problems, increased management involvement in maintenance issues, and more timely identification and resolution of problems. Some examples of poor performance in each of these areas continue to be identified. However, these examples have been of decreasing frequency and safety significance.
i Staff Test., ff. Tr. 2159, at 5-6 (Narbut, Miller).
MFP challenges the adequacy of the maintenance and surveillance program by claiming that the program fails to satisfy a number of broad standards necessary for a satisfactory program. It cites numerous particular incidents to demonstrate how the program fails to satisfy these standards (some incidents are relevant to more than one of the broad standards).
The broad standards set forth by MFP, in outline form, 1
are as follows:" ]
I. Failure or unreliability of important safety systems.
A. Reduction in safety margins. Most of PG&E's maintenance problems in the past several years have disabled or threatened essential safety systems.
B. Inadequate and incorrect analyses of safety significance. PG&E wrongly discounts the safety significance of many of its maintenance deficiencies. This not only results in an incorrect evaluation for pvrposes of evaluating the significance of the incident that occ.urred, but it also raises general questions about the adequacy of PG&E's judgment with respect to safety matters.
"MFP F0F $$ 20-61, at 11-29. Later in the text of this Decision, we shall refer to various outline topics by the paragraph numbers set forth herein.
l i
II. Untimely or Ineffective Response to Maintenance Problems.
A. Untimely Response. PG&E has shown a pattern l of responding to maintenance problems in a lax and untimely manner.
B. Previous Corrective Action Failed to Prevent Recurrence. In many cases, PG&E had the same or similar problem recur after PG&E had attempted to resolve it. This shows an ineffectual maintenance program that is unable to take timely and effective corrective action with respect to maintenance problems.
C. Untimely Detection and Correction of Aging Effects.
III. Breakdown of Multiple Barriers.
IV. Repetitive Patterns of Failure.
A. Lack of Communication and/or Coordination.
PG&E's maintenance and surveillance program is deficient in its communication and coordination between different groups 1
of individuals and/or departments. Examples are provided !
for insufficient communication, insufficient coordination between multiple groups, and insufficient management 1
involvement. l l
B. Previous Maintenance Errors Caused Undetectable Problems. PG&E has demonstrated a pattern of creating undetectable failures through improper maintenance.
- 38.-
C. Inadequate / Improper Surveillance. Routine surveillances, tests and inspections at DCNPP are inadequate to ensure the continued safe operation of the plant.
D. Personnel Errors. PG&E has demonstrated a repetitive pattern of personnel errors which jeopardize the safety of the plant. Examples are cited dealing with personnel errors due to inattention to detail, personnel errors due to failure to follow procedures and personnel failure to self-verify. ,
E. Inadequate Procedures. Procedures or work instructions for personnel are not adequate to ensure that work activities are performed adequately.
F. Manufacturing / Vendor Deficiencies and Internal Defects. PG&E does not have an effective program for detecting manufacturing deficiencies or internal defects. ;
G. Financial Considerations. PG&E's decisions regarding what is needed to maintain the plant in a safe condition have been unduly influenced by economic considerations.
We will individually discuss each of the particular incidents set forth by MFP in its proposed findings and relate them to the broade: standards outlined by MFP. We will discuss them in the order presented by MFP in its specific proposed findings, and will relate them to all the ,
broad standards to which MFP claims they are relevant.
I
- 1. Maintenance of Environmental Oualification of Electrical Eauipment.
MFP asserts that PG&E's maintenance program for the environmental qualification of electrical equipment is fatally flawed because of imperfections in the so-called "telatemp" sticker program." MFP categorizes this claim as an example of a maintenance problem that has disabled or ;
threatened essential safety systems (outline 1 I.A).
In support of this claim, MFP questioned PG&E witnesses on the basis of MFP Exhibits T-1 through T-4. T-1 is a copy I of NRC Information Notice 89-30, High Temperature Environments at Nuclear Power Plants, dated March 15, 1989." T-2 is a copy of PG&E implementing procedures, dated April 22, 1992. T-3 is a consultant's report to PG&E, dated February 27, 1990, concerning "Effect of Localized "In the transcript and in MFP's FOF, the term is spelled "teletemp." In MFP Exh. T-2 (a PG&E document) and the Applicant's and Staff's FOF, the term is spelled "telatemp." We will use the latter spelling in this Decision.
"This document was identified and extensively discussed at the hearing by a PG&E witness but was never formally admitted into evidence. We believe that this omission was due to inadvertence, as claimed by MFP: MFP's questions were directed at a PG&E witness but, because Exh.
T-1 is a Staff document, MFP deferred formally introducing it until Staff witnesses were testifying. MFP neglected to introduce it when Staff witnesses appeared several days later. In any event, the PG&E witness' responses to questions provide an adequate basis for referencing the document herein. See, e.g. Tr. 1844-45, 1861-62 (Ortore)
High Temperatures Upon EQ Components." T-4 consists of numerous sheets recording data from telatemp stickers."
MFP asserts that maintenance of the EQ of electrical equipment that is important to safety is fundamentally important to the safe operation of DCPP, that the qualified life of such equipment is partially determined by assumptions about the normal operating temperatures to which the equipment will be exposed, and that if the normal operating temperature exceeds the assu:ned normal operating temperature the qualified life "must" be shortened. MFP FOF 1 65. MFP contends that, as a result of PG&E's poor management of the telatemp sticker system, the temperatures in many locales are unknown and, for conservatism, the qualified life on many components must be shortened.
The Applicant describes its EQ maintenance / surveillance system as a portion of its program designed to comply with the requirements of 10 C.F.R. S 50.49, which sets forth substantive requirements for an EQ system and defines equipment to be included (but sets no special standards for a maintenance and surveillance program applicable to such equipment). The EQ program is designed to assure that electrical equipment that would be relied on in the event of an accident will be capable of performing its design safety functions to achieve safe reactor shutdown, despite exposure "MFP Exhs. T-2 through T-4 were admitted into evidence at Tr. 2051.
to the harsh environment that could result from an accident.
Among other matters, the EQ program includes.the determination of a " qualified life" based on expected service conditions and identifying and implementing appropriate surveillance, maintenance and procurement requirements to assure that EQ is maintained. PG&E Test.,
at 16-18, 79-81 (Ortore).
At DCPP, the qualified life of a safety component is based on the bulk ambient temperature of the area in which the component is located. However, localized temperatures may be higher than ambient temperatures as defined in the
" binders" which document the basis for the qualified life of each safety component. Tr. 1856-57 (Ortore).
In 1986-87, the Maintenance Department initiated the !
telatemp program, a proceduralized temperature monitoring l
program for EQ equipment. This pre-dated NRC generic '
correspondence (Information Notice 89-30, proposed MFP Exh.
T-1) that raised an issue regarding the potential impact of operating temperatures on equipment performance and qualification. Tr. 2043 (Ortore). )
As a renult of this program, PG&E identified particular
" hot spots" at DCPP. It then contracted with Sargent and Lundy for a report to address " hot spots" and the effects of localized temperatures on the qualified life of EQ equipment. Through these activities, various " hot spots"
.l were identified, allowing calculation of a qualified life l l
for EQ components based on observed environmental conditions. Tr. 1853, 2043 (Ortore); MFP Exh. T-3.
The telatemp monitoring procedure is used principally by the Electrical Maintenance Section at DCPP to monitor electrical equipment and hot spots. PG&E has issued a procedure for implementing that program, MP E-57.8A, MFP Exh. T-2. Tr. 1845, 2045 (Ortore); PG&E Test., at 81 (Ortore). The procedure was last revised in 1992 (Tr. 1891 (Giffin)).
The list of components subject to the monitoring procedure does not include all instrument and control (I&C)
EQ equipment at DCPP. PG&E testified, however, that most I&C equipment is, by design, located in areas that are low in containment or outside the bioshield. Furthermore, most I&C components are low-voltage, low-current equipment that do not generate significant heat. As a result, most I&C equipment is subject to temperatures well below 120 degrees and need not be included in the temperature monitoring program. Tr. 1875, 2045-46 (Ortore). Further, DCPP is a very large and uncongested plant and this factor, along with other design features (e.g., ventilation, routing of power cables), reduces the likelihood of hot spots. PG&E Test.,
at 80 (Ortore).
The Maintenance Department utilizes telatemp stickers to monitor local ambient temperature at EQ components. PG&E Test., at 80-81 (Ortore). These stickers are tabs with
1 mylar faces that. include squares with temperature-sensitive i chemicals which turn color when they are exposed to certain temperatures. The squares record momentary peak- )
l temperatures at the point of installation. During refueling l outages, maintenance personnel read and record the data l provided by the ';elatemp stickers, remove the. stickers, !
affix them to data sheets, and apply new stickers at each location. The temperature is recorded on a form. Tr. 1846-47, 1855, 2041, 2043 (Ortore); MFP Exh. T-2. ;
.)
MFP asserts that the telatemp stickers generally give q readings in 10-degree intervals, and that when a window ;
changes color, that means the component experienced a ;
1 temperature that was between the degree of the window and nine degrees higher. MFP cites PG&E's testimony of the importance of applying conservatism in using the telatemp readings and would have us assume that the safety component being evaluated experienced the highest possible temperature that is indicated by the changed telatemp sticker window.
Thus, if the 150-degree window changed color, .MFP would assume that the component experienced a temperature as high as 159 degrees. Tr. 1861 (Ortore). PG&E's witness was not certain exactly how these_ data would be used but agreed that conservative assumptions would require such an interpretation (Tr. 1855, 1861 (Ortore)).
The Staff initially would have us ignore the documents on which MFP relies, because of lack of adequate advance
l l
1 notice of MFP's intent to use them. Beyond that, the Staff perceives no merit to MFP's claims premised thereon. For the reasons outlined earlier, we are declining to ignore these documents. In particular, we regard as sound MFP's claims that the documents have potential safety significance which could have a material bearing on the adequacy of the maintenance and surveillance program.
The temperature monitoring program provides PG&E with a system to identify any localized areas in which EQ equipment might be exposed to temperatures in excess of the operating temperature previously assumed in the qualified life calculations for that component. If monitored operating temperature exceeds that previously assumed, it may be >
necessary to reduce the component's qualified life. The qualified life of an installed component is then based on the highest temperature data, unless there is reason to believe that a high temperature was only transitory.
Moreover, in performing the qualified life calculation, PG&E generally assumes that the highest temperature registered on the telatemp sticker has been and will be the constant temperature over the service life of the component.
Tr. 1842-43, 2042-43 (Ortore).
The data gathered by the Maintenance Department are provided to the Engineering Department, which, in turn, analyzes such data and determines whether it is necessary to change the qualified life of EQ equipment. Information
1 4
resulting from engineering analyses is sent back to the ;
Maintenance Department, which has the responsibility to change out such equipment prior to the end of its qualified life as part of the preventive maintenance program. -Thus, calculation of the qualified life of EQ equipment is not a Maintenance Department function, although ordering the repair or replacement of EQ equipment based on recalculated l qualified life is such a function. Tr. 1850-51, 2041-42 (Ortore).
PG&E describes the telatemp monitoring program as confirmatory in nature. In other words, it is not the principal means employed by PG&E to ascertain localized peak temperatures. PG&E indicated that, since the original hot spots were identified, it has seen very few changes in normal operating conditions. Tr. 2043 (Ortore).
MPP attempts to disprove this testimony by the telatemp data in MFP Exh. T-4. For example, it claims that readings for selected valves and conduits had temperature variations l l
over a period of years as much as 69 degrees (MFP F0F, l Tables A and B). MFP also introduced numerous data sheets .
into the record, to demonstrate that telatemp stickers are l
sometimes destroyed upon removal, hard to read, otherwise unavailable, or where there either were no telatemp
' measurements or measurements were incomplete (Tr. 1882-84 (Ortore); MFP Exh. T-4). l
It demonstrated that there were no telatemp readings for certain components or general plant areas, that readings of "N/A" or "NA" appeared on many data _ sheets possibly indicating, according to the PG&E witness, that only one sticker was found (when two were required), that a sticker .
was illegible, that there was no sticker below, that a sticker could not be moved without damaging it, and that many stickers could not be found (Tr. 1887 (Ortore); MFP F0F, Tables A and B).
MFP went on to assert that, although MP E57.8A .
generally requires stickers for both the top and bottom of components, many of those dual stickers were not present.
It emphasized the importance of the dual-sticker requirement by demonstrating the considerable temperature variation that .:
could occur between the top and bottom of components., MFP claims that the problem dates from 1988, when the first telatemp measurements were_ recorded, to the most recent refueling outages for each Unit. In addition, MFP claims 1
that an adequate range of temperatures on stickers is not present, setting forth certain components where four rather than two stickers should be used and others where the highest temperatures recorded were not the peak temperatures, only the highest that the stickers could record.
Finally, MFP claims that the procedures for telatemp !
measurements are confusing and hence inadequate. It cites
l
- 47 --
the incomplete lists of equipment to which the program applies (or is intended to apply), the limitation of the procedure to instructions for installation and removal of stickers, without sufficient guidance on how to record the data from the stickers onto the data sheets. Indeed, one of the data sheets contains an explicit complaint that
" procedure should explain how to read stickers."
In short, some of MFP's claims are well founded. The Board finds that PG&E's procedures for telatemp sticker installation are confusing, and that as a result it is difficult to determine exactly where stickers should be installed and monitored. The requisite terminology on data sheets is also confusing--e.g., N/A or NA may refer to the fact that a new piece of equipment was installed, or to the fact that a telatemp sticker was not found or could not be read. N/A or NA on data sheets could also apparently mean either "not available" or "not applicable." Tr. 1886-87 (Ortore). The list of equipment in the applicable procedure to which stickers are to be affixed is also not l representative of all the equipment to which stickers are attached (Tr. 1882 (Ortore)). A failure to list a t
temperature could mean either an erroneous listing or a failure of the temperature to be high enough to warrant a listing (Tr. 1885-86 (Giffin)). Occasionally the PG&E [
witness was not certain what a particular recorded number !
f I
meant (Tr. 1889 (Ortore)). ;
We are also concerned about the level of accuracy of the telatemp measurements, given the many instances in which PG&E recorded only one measurement rather than the required two or, indeed, the four that should perhaps be available.
This pattern leads us to believe that, to the extent PG&E relies on the telatemp program, systemic improvements should !
be made to reduce or eliminate-such inconsistencies.
In sum, it is fundamentally important that PG&E have an adequate program for maintaining environmentally qualified safety equipment. This includes monitoring equipment where temperatures are known to be high, to ensure that the normal operating temperature is not higher than the conditions to which the equipment was originally qualified. If it is, the qualified life may have to be reduced and the equipment replaced. The Board recognizes that the telatemp program is ,
confirmatory only. But to the extent it is used in monitoring these localized high temperatures, it is ;
deficient in that it is not being carried out in a !
consistent and accurate manner and PG&E does not have adequate procedures to ensure that it can be carried out properly. To the extent of that deficiency, MFP's assertion that PG&E's telatemp sticker program reflects a reduction in contemplated safety margins and a potential threat to essential safety systems is well founded.
The documents' relied on by MFP do not, however, constitute a sufficient basis upon which to conclude that l
1
there is an overall programmatic deficiency in the maintenance of EQ equipment at DCPP. As noted, the temperature monitoring program is confirmatory in nature; ;
i.e., it continually confirms the validity of the input data used in qualified life calculations. The si'g nificance of i 1
any imperfections in the data collection process will, to :
that extent, be alleviated. Moreover, the exhibits are replete with examples where more than one telatemp sticker is used on an EQ component. Sgg MFP Exh. T-4. Where more than one is required by procedures, however, those examples do not serve to ameliorate the deficiencies outlined.
In conclusion, although the telatemp procedure is not per se required for PG&E to conform to the EQ requirements i
of 10 C.F.R. S 50.49, we believe that where, as here, PG&E elects to utilize such a program as part of its maintenance / surveillance program, it must have adequate 3
procedures (1) that define all the equipment on which the j stickers are to be utilized, (2) that set forth the number and location of the stickers to be used for each piece of ;
designated equipment, (3) that specify the time, method and ,
precise nomenclature for recording the temperature data, and :
t (4) that assure that the information utilized is not erroneous or misleading. To the extent PG&E places any ;
reliance on such a program (even if only confirmatory, as l
claimed), it should revise its procedures to incorporate i
these changes. We are imposing orders for correction to this effect.2o
- 2. Check Valves /IST Deficiency.
MFP introduced several NCRs and LERs to demonstrate that PG&E's inservice testing (IST) of check valves is deficient (MFP Exhs. 6-11, 13). It claims that, as a result, safety systems have been disabled or threatened (outline i I.A), that routine surveillances, tests and inspections at DCNPP are inadequate to ensure the continued safe operation of the plant (outline i IV.C) and that procedures or work instructions for personnel are not adequate to ensure that work activities are performed adequately (outline i IV.E).
For many years, the ASME code did not require testing of leak tightness of check valves if their position was normally closed (Tr. 602 (Crockett)). However, on August 29, 1988, the NRC issued IN 88-70, " Check Valve Inservice Testing Program Deficiencies," to notify licensees of potential problems with check valve IST. NRC Generic Letter 89-04, " Guidance on Developing Acceptable Inservice Testing Programs," dated April 3, 1989, identified similar generic concerns and required that implementing test procedures be reviewed and revised as necessary within six 2
We are here imposing no conditions or guidelines as to whether or how the Engineering Department uses the telatemp data.
months. The NRC was concerned that check valves included in the IST program were not always tested in both the open and closed positions and that no reverse flow operability tests were being performed on check valves other than those used for containment isolation and reactor coolant system pressure boundary isolation. MFP Exh. 11, at 2. PG&E initiated the review required by the Generic Letter (MPP Exh. 13, at 2).
MFF claims--accurately--that, as a result of PG&E's review, a multitude of deficiencies in the check valve IST program have been (and continue to be) identified. MFP FOF i 98. The finding of these deficiencies does not, however, denominate a current breakdown or failing in PG&E's maintenance program. Rather, the NCRs and LERs cited by MFP demonstrate PG&E's attempts to bring its own maintenance and surveillance procedures in line with recently evolving Staff policy. As characterized by PG&E, the issue is generic.
Tr. 603 (Crockett). Moreover, valves added to the list for testing had never been found to be leaking. Tr. 608 (Crockett). Thus, PG&E's performance in this regard appears to be adequate and cannot serve as a basis for either license denial or conditions.
- 3. Cable Failures.
MFP introduced evidence to the effect that there were five medium-voltage cable failures at DCPP between October 1989 and March 1993--three of which were on 4 kV cable and i I
l
/
two on 12 kV cable--and it claims that PG&E's maintenance and surveillance program was not adequate to detect the degradation of this cable (outline i IV.C) . MFP FOF 11 50, 108, 112. One of these failures was detected only by smoke occurring as a result of the failure. MFP Exh. 15, at 5.
According to MFP, an essential safety system is thus threatened (outline i I.A).
MFP further asserts that PG&E has not identified the cause of the three 4 kV failures and cannot justifiably claim that they are random occurrences. MFP F0F i 115. MFP also faults PG&E for delay in replacing certain of these cables and for replacing them with the same construction material, as to which it asserts there is some question of acceptability for the conditions under which it is operating. MFP FOF ji 119, 126. These failures thus also represent an untimely response to a maintenance problem (outline i II.A) and untimely detection and correction of aging effects (outline i II .C) .
The five failures occurred in two separate sets of underground duct bank conduits, between the turbine building and the intake structure. MFP Exh. 15, at 3. The Applicant asserts first that there is no connection between the failures of the 4 kV and 12 kV cables. See MFP Exh. 21, at ,
l 1-2. (Two of the three 4 kV cables are safety-related, I whereas one 4 kV cable and both of the 12 kV cables are not.
I MFP Exh. 15, at 3; PG&E Test., at 108 (Ortore); Tr. 624 i
l l
)
(Ortore).) It testified that two of the 4 kV failures were ;
t random in nature and time of occurrence (October 29, 1989 ,
and May 3, 1992), whereas one occurred.during a routine i high-potential test during a refueling outage (October 31, . 'j 1992). PG&E Test., at 108 (Ortore). ;
PG&E conceded that, at the time of the hearing, it had ;
not yet completed its root cause analysis of the'4 kV cable failures. But, it had ruled out certain possible causes, including the chemical attack and degradation that had been determined to have caused the 12 kV cable failures. Tr. 625 (Ortore); MFP Exh. 15, at 9. It also determined that the i failed cables were of acceptable quality and design.for
- i their specific applications and service conditions. Both the failed and certain unfailed sections of these cables have been replaced. PG&E Test., at 110. In addition, because of defense-in-depth redundancy, the ultimate safety i significance of a 4 kV cable failure is likely not to be great.
The 12 kV failures occurred, respectively, in February and March 1993. PG&E testified that both were the result of external cable degradation, caused initially by exposure to an unidentified contaminant, probably a cleaning agent, e present in the und.erground conduits for the 12 kV cable.
The contaminant was carried within the conduit by water, :
l which degraded the outer jacket of the cable, exposing the j i
copper shielding. The shielding was then attacked by l
- n. -
54 -
chlorides in salt water present in the conduit, creating uneven electrical stresses on the cable and thus causing a fault. PG&E Test., at 109 (Ortore); Tr. 649, 671 (Ortore).
The failed 12 kV cables and certain other sections have likewise been replaced. In addition, PG&E found a contributory cause of these failures to be inoperable sump pumps in the cable vaults. Prior to the 1993 failures, these pumps were not included in the formal maintenance program. The pumps have now been repaired and are included in the preventive maintenance program. PG&E Test., at 110; MFP Exh. 15, at 11.
With respect to the 4 kV failures, the Staff concluded that PG&E's responsive actions were adequate and reasonably thorough. The Staff further concluded that plant safety had not been significantly reduced by these failures, because of the presence of other, unaffected cables for redundant safety-related pumps. Staff Test., ff. Tr. 2159, at 10 (Narbut). With respect to the 12 kV failures, an NRC inspection verified the repair of inoperable pumps, the initiation of preventive maintenance for such pumps, and the replacement of all failed cable. The Staff would have us find that, with the sump pump added to the maintenance program and the cable replaced, it is unlikely that such a cable failure will recur. Staff F0F j I-84.
We find that PG&E's responses to these failures were reasonable and effective. Even though not on safety related
X cable, however, the fire resulting from one 12 kV failure appears significant, if for no other reason than that fires are per se hazardous. PG&E's corrective actions, approved by the Staff, appear adequate.
- 4. Wrono Size Motor Installed.
Premised on a PG&E draft NCR dated July 28, 1993, MFP Exh. 24, NCR DC2-93-EM-NO31, together with related testimony, MPP asserts that, during a refueling outage, a 10 ft-lb motor was installed on a motor-operated valve (MOV) rather than the required 15 ft-lb motor, that this mistake was caused by multiple personnel errors, and that barriers designed to prevent such errors were ineffective, evidencing a programmatic deficiency. MFP stresses that, although installation of the wrong motor may be attributable to an isolated personnel error, the failure of anot less than three other individuals" responsible for checking the correctness of the installation cannot properly be so designated and, rather, is an indication of a programmatic deficiency in the maintenance / surveillance program. MFP further describes what it perceives to be the safety significance of the erroneous installation. MFP FOF
$$ 144-51. It categorizes the erroneous installation as threatening an essential safety system (outline i I.A), as a series of personnel errors resulting in a breakdown of multiple barriers (outline i III) and as an example of a
repetitive pattern of personnel errors which jeopardize the safety of the plant (outline i IV.D).
PG&E acknowledges the erroneous installation, claiming that an individual preparing a work order to replace the motor on an actuator for a MOV made an error in reading the motor size from a table and hence specified on the work order an erroneous motor, which was later installed.
Tr. 689-90 (Giffin). It characterizes the incident as an isolated personnel error. PG&E FOF i M-A21. The Applicant as well as the NRC Staff, however, deny the incident's actual or potential safety significance. Based on the NCR, they assert that, even with the under-sized motor, the MOV was able tc. shut under design-basis conditions and thus would have performed its intended safety function.22 As for corrective action, PG&E replaced the incorrect motor and checked other similar motors to verify their correctness. It also counseled the work planner and the QC inspector involved with this event, communicated the importance of self-verification to incoming persons through the electrical maintenance bulletin, and held meetings with electrical maintenance engineers to discuss the importance 22 MFP Exh. 24, at 11; Staff FOF 1 I-86. The NCR adds, however, that " [i] t is not likely that the valve would trip the torque switch at reduced voltage and with the Thermal Overload Device (TOLD) sized for a 15 ft-lb motor, once it goes closed it would probably burn out the motor. This would happen only after the valve had performed its safety function."
of the engineer's responsibilities and expectations when they sponsor design control notices. MFP Exh. 24, at 1, 6-8; Tr. 690-91 (Giffen, Vosbourg).
It is clear to us that PG&E took appropriate action to replace the hardware involved, and to assure that incorrect motors had not been inatalled elsewhere. More important, the Applicant also took significant steps to alleviate the maintenance deficiencies. That three individuals were responsible for four personnel errors is not, however, reassuring.22 ,
For these reasons, license ccnditions would not be i appropriate. Nonetheless, because similar incidents might well have safety significance given the wide variety of parts in a nuclear facility that conceivably could be !
i incorrectly utilized, PG&E may wish to explore whether some systemic improvements could be made in this area.
Improvement might be particularly appropriate with respect to the process of self-verification by installers. For example, perhaps the self-verification process could require a second look by the installer at the design change document, as well as at the installed part.
22 MFP's identification of four separate individuals (MFP FOF i 147) appears erroneous--the worker who failed the self-identification appears to be the same as the installer, resulting in errors by three individuals. j l
l l
- 5. Storace and Handlina of Lubricants.
Relying on two NCRs, MFP asserts that failure to control lubricants is a recurrent problem at DCPP and demonstrates a deficient maintenance and surveillance program. MFP FOF i 162. MFP categorizes the alleged deficiency as an example of a previous corrective action that failed to prevent recurrence (outline i II.B) and as a repetitive pattern of personnel errors due to inattention to detail (outline i IV.D).
Specifically, MFP cites " unlabelled and mislabelled grease guns and oil pumps; cross contamination of greases and oils; the use of wrong oils; and failure to maintain log books." MFP FOF i 152, citing MFP Exhs. 27 and 28. MFP notes that the issue was first identified in 1987, that additional problems occurred in 1991, and that, despite corrective actions, similar problems occurred in 1993.
MFP asserts that PG&E received an NOV in 1987 for failure to comply with the procedure governing control of lubricants. In 1990, the wrong oil was added to the " heater 2 drain tank pump 2-1." In 1991, the wrong oil was added to a motor bearing. Also, PG&E wrote an NCR considering other discrepancies in lubrication storage and handling. MFP further claims that, in 1993, an incompatible oil was used to lubricate the Auxiliary Salt Water pumps and that the oil log book did not indicate where or when this oil was
i i
l l
1 l
obtained. MFP Exh. 27. MFP attributes all these l
deficiencies to the maintenance and surveillance program. j The Applicant characterizes the problem (citing the same MFP exhibits) as involving several minor lubrication control issues widely separated in time and dissimilar in nature. It claims that corrective actions have been adequate and existing procedures sufficient. PG&E FOF
$ R-A32. The Staff takes a similar position. Staff FOF ii I-91 through I-93. PG&E further disputes MFP's claims of safety significance, terming them as based on extra-record speculation, inasmuch as only small amounts of oil or lubricants were involved, with no impact on equipment operability. The Applicant further points out that existing preventive maintenance tasks assure that oil is periodically changed and sampled and that equipment is monitored for excessive wear. PG&E FOF $ R-A33. The Staff concurs.
Staff FOF $ I-94.
The Board here supports the position taken by PG&E and the Staff, to the effect that the incidents are essentially isolated and reflect no systemic deficiency in the maintenance and surveillance program. They represent neither a recurrence of a previously corrected problem nor a repetitive personnel error. Given that finding, as well as the circumstance that the amounts of oil or lubricants involved were small and could cause no operational impacts, 1
I
___1
l l
1 we agree that the incidents lack current or potential safety l l
significance. !
- 6. Fuel Handlino Buildino.
In order to assure that all potential releases from the spent fuel pool are exhausted through fuel handling building (FHB) exhaust filters, the fuel pool area must be maintained at a negative pressure. MFP Exh. 39 (LER-89-019-00), at 4; MFP FOF i 167. MFP has cited several instances in which the negative pressure was not maintained adequately, and it attributes these instances to inadequate maintenance or surveillance. MFP FOF $1 170, 173, 177, 181, 184. MFP categorizes these. examples as ones where safety systems were disabled or threatened (outline i I.A), as an untimely detection and correction of aging effects (outline 1 II.C) and as an inadequate or improper surveillance (outline i IV.C).
Two instances in particular, both in 1991, were identified.23 The first was discovered on January 18, 1991, when the FHB ventilation system was declared inoperable after failing to meet the negative 1/8" water gauge pressure requirements specified in surveillance test procedure (STP) M-41. The STP M-41 had last successfully been performed on September 18, 1989, and the January 18,
'MFP Exh. 38. Documents relating to other FHB incidents were not admitted, because they bore no relationship to maintenance or surveillance. See Tr. 827-28.
1991 surveillance was the routine 18-month follow-up surveillance. MFP Exh. 39 (LER-89-019-00), at 2. The second occurred on August 7, 1991, when the Unit 2 FHB failed to pass STP M-41. MFP Exh. 38.
The root cause of the first of these events was determined to be "the degradation of the FHB." Contributory causes were " dirty exhaust fan ducts, failure to maintain a flow difference between the exhaust and supply flows of 19.8 percent, and blocking of a FHB exhaust duct." The root cause of the second event was "an improper understanding of the required calibration frequency of the supply fan inlet vane controller." MFP Exh. 38, at 1, 11-12.
PG&E took corrective action for each of these deficiencies. MFP Exh. 38, at 1. No issue appears to be raised concerning the adequacy of any particular corrective action. MFP, however, attributes the deficiencies to lack of an effective preventive-maintenance program for the FHB.
PG&E points out that in each instance the pressure was negative, although not as negative as required, thus reducing (in its opinion) the safety significance of the )
l incidents. PG&E F0F $ M-A41, citing MFP Exh. 38, at 2-3, "
13-14. The Staff agrees. Staff FOF $ I-95. However, PG&E !
l acknowledr <1 that its preventive-maintenance program covered j 1
FHB doors and ventilation system components but not the '
building as a whole.
l
With respect to the first instance, PG&E promptly investigated the situation, determined the root cause, and ,
implemented corrective actions. PG&E Test., at 104-05 (Crockett). It determined that the cause of the deficient !
negative pressure was the existence of small leakage paths into the building resulting from degradation of building siding and seals. This was an example.of' structural l degradation that was identified by PG&E and corrected.
Tr. 807-08 (Giffin).
Corrective actions included sealing the leaks and re-siding both FHBs. Most significantly, surveillance is now performed within seven days of any fuel movement, rather i
than at the previously prescribed 18-month intervals.
Tr. 812 (Crockett).
The second instance of reduced negative pressure resulted from drifting of the setpoint for the controller, increasing the supply flow into the FHB. MFP Exh. 38, at 3. ;
The drift was identified through surveillance and later corrected. ,
Although the Applicant and Staff deny that these instances reflect any deficiencies in the surveillance 1
program--indeed, they credit the program for detecting the instances of negative pressure--we find it significant that an important corrective action was to increase the frequency of surveillances. To that extent, the previous surveillance program did result in the untimely detection and correction I
of aging effects (as claimed by MFP) and hence warranted improvement. The prompt modification, however, demonstrates the strength of PG&E's overall program. We find no evidence that would permit us to question (as MFP does, at MFP FOF i 181) the adequacy of the current surveillance program for the FHB.
- 7. Tests of Containment Personnel Airlock.
MFP has identified what it describes as several missed surveillances of the airlock door seals and portrays them as failures of the surveillance system. It categorizes them as disabling or threatening safety systems (cutline i I.A) and as examples of a repeat pattern of inadequate performance of routine surveillances, tests and monitoring activities (outline i IV.C) . MFP also faults PG&E for minimizing the significance of the missed surveillances. MFP FOF $ 52.
The first two occurred on September 20 and 21, 1990; another occurred on April 25, 1993. In each case, a personnel air lock gauge was removed (for maintenance) and later reinstalled, but required post-maintenance leak rate testing within the period specified by the Technical Specifications was not performed. MFP Exh. 42 (NCR DC2 WP-N025); MPP Exh. 43 (LER 2-90-011-00); Tr. 830-32 (Vosburg).
MFP also cites a final missed surveillance that occurred on June 11, 1991 (but was not discovered and reported until September 27, 1991). The LER states that
" [a] review determined,that an acceptable leak rate test was not performed following 17 containment entries during the ,
period from June 11, 1991, to September 27, 1991." MFP Exh.
44 (LER 1-91-016-00), at page numbered 1 (2 of exhibit) .
The immediate cause of the missed surveillance was a faulty i
solenoid valve, with root cause attributed to personnel error caused by inadequate knowledge of the leak-rate monitor operation.
The Applicant and Staff each treat the 1990 and 1993 incidents as separate and apart from the 1991 incident (which, as indicated above, was attributed to personnel error). Analytically, however, there appear to be more interrelationships among all of the four separate incidents.
Thus, PG&E initially believed that the 1990 and 1993 incidents were also the result of personnel error. It changed its opinion when it discovered what it deemed to be ;
faulty instructions to the workers who were servicing the gauge. Tr. 830 (Vosburg). And the 1991 incident, although attributed to personnel error, resulted in fact from deficient documentation--i.e., an inadequately documented clearance to take the leak-rate monitor out of service for calibration. MFP Exh. 44, at 3 of LER (fourth page of exhibit).
Collectively, therefore, it appears to us that all of the incidents in question may be properly perceived as resulting, not from individual personnel errors but, rather, i
b from less than complete instructional material for those performing the maintenance-related servicing or calibration.
t Improvement in the-procedures and associated instructions -
seems to be the proper corrective action and, indeed, has already been implemented. MFP Exh. 42, at 9-10; MFP Exh.
43, at 1, 6 of LER (pp. 2, 7 of exhibit); MFP Exh. 44, at 1, 7-8 of LER (pp. 2, 8-9 of exhibit).
Although these incidents all seem to have a bearing on the adequacy of the maintenance and surveillance program, -
and although they have safety significance, they do not appear analytically to cast any strong adverse inferences about the program. In particular, corrective action r
apparently has countermanded the seeming deficiencies that
')
were brought to our attention. We thus decline to include v these instances among adverse information that would detract from the sought extensions of the operating licenses.
One comment on " numbers" is, however, in order. The !
Applicant br.s characterized these missed surveillances as among 55 missed surveillances throughout the 10-year history of the facility and has compared that number to the over 10,000 total tech-spec surveillances that take place I
annually. PG&E adds that in 1992 it missed only three surveillances and in 1993, up to August, it had missed only one. Tr. 836 (Crockett).
PG&E witnesses were questioned about the accuracy of these numbers, given the statement in MFP Exh. 44 (quoted i
1
l above) that a single missed surveillance had resulted in 17 containment entries, each of which would require a surveillance. They explained the.t missed surveillances were grouped by root causes and that all stemming from the same root cause were considered the same missed surveillance.
Tr. 834, 836, 845, 848-53 (Crockett, Vosburg, Giffen).
MFP considers these statistics to be misleading.
Tr. 853 (Curran); MFP FOF ii 52, 190. On the other hand, PG&E supported its methodology by explaining that it helped it to gauge the effectiveness of its program, as well as to correct the root cause of a missed surveillance. Tr. 846-47 (Vosburg). The Staff adds that industry and the NRC accept this practice as appropriate. Staff F0F i I-108, citing Tr. 1149 (Crockett).
We agree that PG&E's methodology for counting missed surveillances is appropriate, given the common industry practice. However, it might be wise for PG&E to add, in the context of statements to other than industry or the Staff, some explanatory preface to avoid the appearance of manipulating statistics to make the record appear advantageous.
- 8. Component Coolina Water (CCW) Heat Exchancer.
The CCW system removes heat generated by various plant systems without releasing radioactive material to the environment. The DCPP has four CCW heat exchangers, two for l
l 1
l
each unit. Each CCW heat exchanger has 1,237 tubes approximately 35 feet long.
In March, 1993, during a Unit 2 refueling outage, testing was conducted on tubes in both Unit 2 heat exchangers. The testing was part of the ISI program that looks at performance to identify and predict early if there is any degradation of equipment. Tr. 857 (Crockett).
Fretting was found on the outside diameter of certain tubes, at the baffle plates. Tubes with damage greater than 20%
were plugged (i.e., removed from service), including ten on one Unit 2 heat exchanger and several on the other. PG&E determined the root cause to be flow induced vibration on the tubes, and accordingly it also revised operating procedures to address maximum flow limits. MFP Exh. 47 (NCR DC2-93-TS-N017, Rev. 00, June 15, 1993).
MFP claims that testing is not being conducted with sufficient frequency, that the ability of maintenance and surveillance activities to assure the efficiency of the CCW j l
heat exchangers is questionable and that the corrective actions, maintenance and design changes may have violated 1
the original design criteria by improperly extending the I original design flow rate. MFP F0F $1 203, 206, 211. MFP portrays this incident as representative of a maintenance problem that has threatened or disabled essential safety systems (outline i I.A) and as an example of inadequate !
routine surveillances (outline i IV.C) .
PG&E counters that the fretting was indeed detected through the surveillance program, that frequency of inspection, which is determined on the basis of expected wear and service life, is sufficient and, in any event, PG&E is studying whether increased inspections are warranted.
Tr. 858 (Crockett). PG&E adds that it incorporated the design basis maximum flow limits into its operating procedures. PG&E FOF $ R-A43, citing MFP Exh. 47, at 1.
The Staff stresses that PG&E identified the problem through its surveillance program, took appropriate maintenance action, and is further studying the appropriate testing interval. It regards the incident not as a weakness but a strength of the ISI program. Staff FOF ji I-109 through I-111.
In evaluating this incident, we find no basis for suggesting that the surveillance program was not properly implemented. The CCW heat exchanger tubes were inspected when they were scheduled to be inspected. Indeed, there had been no previous finding of fretting on inspection of the Unit 1 tubes. Tr. 863-64 (Crockett). We express no opinion, however, as to whether the inspection interval was or is appropriate. The evidence of record does not suggest that more frequent surveillances are clearly neceseary (or, for that matter, clearly not necessary).
Finally, no safety system appears to have been compromised--the regularly scheduled testing is designed to
detect this type of condition so it can be corrected before tube failure (MFP Exh. 47, at 6) and it did so here. PG&E is eppropriately studying whether the frequency of testing should be increased. Given the existing schedular surveillance requirements, we find no undue delay--the condition was discovered five months before the hearing, with no indication that it had occurred earlier. In short, this incident does not reflect adversely on PG&E's surveillance program.
- 9. Auxiliary Buildino Ventilation System Inocerable.
On March 2, 1993, maintenance personnel were preparing to perform a preventive maintenance task relating to the Unit 2 Auxiliary Building Ventilation System (ABVS), and a clearance was placed on the system. In subsequently revising the clearance and implementing the work order, however, personnel improperly closed the wrong damper, activating the ABVS logic to shut down the only operable ABVS fan (the redundant fan was already out of service for ,
1 maintenance). PG&E prepared an NCR and a LER concerning l this reportable event. MFP Exhs. 49 (NCR DC2-93-MM-N012 Rev. 00, dated June 11, 1993) and 50 (LER 2-93-002-00, dated '
April 5, 1993); Tr. 881-83 (Giffen).
l MFP claims that this incident demonstrates inadequate maintenance instructions and poor communication between maintenance and operations staff, creating an unacceptable safety risk (MFP FOF i 220). MFP designates the incident as
l l
1 one which disabled or threatened an essential safety system (outline i I.A), as indicative of insufficient communication between groups or departments (outline i IV.A) and as an example of inadequate or improper surveillance (outline i IV. C) .
PG&E and the Staff regard this incident as an isolated personnel error with no safety significance. This latter conclusion is based primarily on the circumstances that limits set forth in 10 C.F.R. Part 50, Appendix A, General Design Criterion 19, would not be exceeded within the time (two hours) needed to restore system operations, and that 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> is needed before any safety-related equipment would be affected (the ABVS was out of service for only 15 minutes).
MFP Exh. 49, at 8.
We agree with MFP that the incident threatened a safety system, irrespective of the circumstance that it did not last long enough to have immediate safety significance. We also agree that the incident does to some degree represent an inadequacy in communications between maintenance personnel and others. It also reflects a personnel error, caused in part by insufficient instructions. We note that corrective action to improve such instructions is being proposed. Given these factors, we do not consider the incident to have sufficient significance to undermine the I effectiveness of the maintenance program. Nevertheless, given the number of incidents reflecting inadequate 1
l
- 71 -
communications between maintenance personnel and other PG&E departments concerning maintenance-related activities, we are directing PG&E to perform a study of this' problem and provide it to the Staff. In addition, although not warranting a license condition, PG&E might also consider certain general improvements in the preparation of instructions to' maintenance personnel.
- 10. Restoration of Electrical Panels.
MFP points to two incidents in 1993 (one in April, the other in May) involving failures to return electrical panels to their original configuration following work-related activities within the panels.24 On April 1, 1993, the rear hinged panel of the Unit 1 RHF panel was observed with no fasteners installed to secure the hinged panel to the main panel. The fasteners were in a plastic bag in the bottom of the RHF panel. The preliminary safety evaluation was "a potential loss of seismic qualification that could have impacted the operability of vital 4 kV bus F and its associated diesel generator during a seismic event." A preliminary root cause was that responsibility for panel restoration was not assigned to any of the groups performing 24MFP Exhs. 51 and 52; Tr. 888-90 (Giffen). Because of the similarity of the incidents (in PG&E's view), the Applicant is cancelling Exh. 51 and taking action under Exh.
- 52. Nonetheless, two separate incidents did occur, and the cancellation of one of the NCRs represents only a bookkeeping convenience for PG&E, not a lowering of the significance of either incident (or of both considered collectively).
1 concurrent work on the panel. The NCR referred to several
" previous similar events," one of which was a 1989 event to l which MFP alluded. MFP Exh. 52, at 1, 3, 9.
l The second event was reported on June 7, 1993 but occurred earlier (i.e., investigative follow-ups were underway as early as a TRG meeting on May 25, 1993).
Covers were found not to have been installed on the hot shutdown panel for both Units 1 and 2. The covers were observed to be lying in the bottom of the back of the panel, and the mounting screws "were no where [ sic] to be found."
MFP Exh. 51, at 1, 2, 7. The specified root cause was merely a cross-reference to the other NCR and, thus, must be considered by us to be identical.
Immediate corrective actions included (for the first event) replacing the fasteners the same day they were discovered uninstalled and checking fasteners in similar panels and (for the second event) re-installation of the internal hot shutdown panels. Further investigative actions were undertaken but not yet complete at the time of the hearing. MFP Exhs. 51 and 52; Tr. 898 (Giffen).
MFP asserts that PG&E's previous corrective action failed to prevent recurrence of a similar event and that PG&E's safety analysis shows a misunderstanding of or disregard for the safety principles underlying its maintenance responsibilities. MFP FOF $1 227, 231. MFP j categorizes these incidents as involving safety systems
being disabled or threatened (outline i I.A), inadequate or incorrect analyses of safety significance (outline i I.B) and previous corrective action failing to prevent recurrence (outline j II.B), and as an example of insufficient coordination between multiple groups (outline i IV.A).
PG&E determined that both incidents had no safety significance--the first, because the bus and associated diesel would in any event have been operable before and after a postulated seismic event, and because of redundancy, and the second because the as-found condition did not impact safety given other fire-protection features of the plant.
PG&E FOF $1 M-A61 and M-A62, citing MFP Exhs. 51 and 52.
PG&E acknowledges that corrective action is still ongoing.
Tr. 898 (Giffin). The Staff asserts that the incidents do not rise to the level of a fundamental flaw in the maintenance program and, in addition, would discount MFP's findings in this regard as speculative and not supported by expert testimony. The Staff also observes that PG&E's root-cause analysis and corrective actions are ongoing. Staff
, FOF $$ I-119 through I-121.
(
In our opinion, there is no basis for concluding that a safety system was disabled or threatened, even though the l practice of failing to restore equipment being serviced to use could analytically be deemed significant. Further, this is not an example of an inadequate or incorrect analysis of l
safety significance--the analysis is still underway, and no
evidence suggests that final resolution has been unduly delayed. Finally, the 1989 and 1993 incidents are too disparate in time for_us to conclude that the_1993 incidents resulted from a failure in the 1989 corrective action.
We agree with MFP, however, that the incidents are examples of insufficient coordination between multiple groups. This deficiency has appeared in several other cited incidents. Although not sufficient to undermine the adequacy of the maintenance system, this is an area calling for additional corrective action. Wesare encouraged that PG&E is studying this problem, but we nonetheless are directing that this general area be included as part of PG&E's communications study that we are directing, with a report to be furnished the Staff.
- 11. Containment Eauipment Hatch.
On March 12, 1993, the Unit 2 equipment hatch was observed to be not fully closed, during a core offload, thus violating a technical specification. PG&E submitted an LER to the Staff, as required. MFP Exh. 54 (LER 2-93-003-00, dated April 5, 1993). The Applicant also prepared an NCR recording the event, its significance, and corrective action. MFP Exh. 53 (NCR DC2-93-MM-N013 00, dated May 28, 1993).
MFP claims that, despite a previous event and an NRC information notice, the maintenance procedure and personnel preparation was not adequate for the hatch closure activity.
l l
l
MFP FOF $ 238. MFP categor'zes this incident as one where safety systems were threatened or disabled (outline i I.A),
where previous corrective action failed to prevent recurrence (outline i I.B) and as an example of personnel error resulting from inadequate procedures (outline i IV.E). The Applicant and Staff each regard the incident as an example of isolated personnel error not reflecting any systemic maintenance problem. PG&E FOF $ M-A66; Staff F0F
$ I-125.
We agree with MFP that the incident had safety significance: the gap in the cover could, during tuel movement, permit a gaseous release of radioactive material to the atmosphere if an assembly were dropped. Tr. 903 (Giffen). We also perceive the incident to represent, as claimed by the Applicant and Staff, an isolated personnel error. We disagree with MFP's claracterizaticn of the incident as reflecting previous corrective action that failed to prevent recurrence. Although a previous similar ;
l event had occurred ten years earlier (in 1983), the lapse of time is sufficient to assume that the two personnel errors are unrelated.
Finally, there is some ambiguity in the record as to whether the 1993 error resulted from inadequate procedures.
1 PG&E states that " [t] he procedure was adequate, the i journeyman did not follow it." Tr. 904 (Giffen). But PG&E !
l is modifying the procedures to require independent
i
)
)
inspections of the closure from both inside and outside the )
containment. Tr. 904-05 (Vosburg). This appears to us to represent an improvement to already adequate procedures and ]
not a reason to fault the maintenance program.
We note that we believe that it would be impossible to eliminate all personnel errors, as highly desirable as such -
a result might be. We here find not that personnel can never commit errors with significant safety consequences, or that PG&E did not commit such an error here, but only that there appears to be no programmatic reason at DCPP for errors such as this to have occurred. We offer no magic l solution to this endemic problem.
- 12. Manual Reactor Trio Caused by Failure of a Fuse for the Rod Control System.
As set forth in NCR DC1-91-EM-N046, dated June 10, 1991 (MFP Exh. 56), on April 24, 1991, plant operators initiated a manual reactor trip to terminate an increase in reactor power, caused by a failure of the rod control system which t
rendered manual control rod movement inoperable. The immediate cause was failure of a fuse in the bus duct t
disconnect to the rod control power supply cabinet.
Investigation disclosed that 12 of 15 fuses in similar >
locations were of the wrong type. All such fuses were to have been replaced with newer design fuses. PG&E had submitted an LER to report this event to the Staff. MFP Exh. 55 (LER 1-91-008-00, dated May 23, 1991).
l l
l l
MFP asserts that PG&E's previous corrective actions were ineffective and failed to prevent this event. MFP F0F i 245. It categorizes the event as one where safety systems were disabled or threatened (outline i I.A), where previous corrective action failed to prevent occurrence (outline i II.B) and as a personnel error due to failure to follow procedures (outline i IV.D. )
PG&E acknowledged that the failed fuse was of an old style with known reliability problems th't was to have been replaced in 1989. It also acknowledged a personnel error in that the wrong fuses had been replaced. Following the trip, ,
PG&E took steps to replace all of the bus duct fuses for the Unit 1 rod drive control cabinets, as well as other corrective actions. It considers the matter resolved, with no recurring maintenance problems indicated. PG&E FOF i M-A68, citing MFP Exh. 56 at 1, 4, 7, 10. The Staff essentially agrees, perceiving no matter generally relevant to PG&E's surveillance and maintenance program. Staff FOF 1 I-127.
In our view, MFP is correct in its characterization of ,
this event. The failed fuse had safety significance, it was supposed to have been changed and was not, and, predictably, it failed. Indeed, the wrong fuse had been changed.
Although these errors were eventually corrected, the situation does constitute poor maintenance performance.
Nonetheless, adverse circumstances do not appear to be i
recurring with sufficient frequency to disqualify the maintenance program or suggest a readily apparent remedy.
We thus merely observe that continuous vigilance and attention to detail is a worthwhile goal that management should strive for in' implementing its maintenance and surveillance program.
- 13. Limitoraue 2-FCV-37 Failed to Close.
During a routine surveillance procedure in January 1993, one Limitorque motor operated valve (2-FCV-37) failed to close on demand from the control room. MFP Exh. 57, at 3-4. The cause of the failure was determined by PG&E to be a quad ring incorrectly installed during a 1990 maintenance overhaul of the operator, caused by an installation procedure that did not give adequate guidance to the workman. Tr. 913-14 (Giffen); MFP Exh. 57.
MFP would have us find that PG&E failed to perform adequate maintenance on two valves and did not identify the problem in a timely way. MFP FOF i 260. It categorizes the incident as a pattern of responding to maintenance problems in a lax manner (outline i II.A), as untimely detection and correction of aging effects (outline i II.C), as demonstrating a pattern of creating undetectable failures through improper maintenance (outline i IV.B) and as an example of inadequate procedures or work instructions f
(outline i IV.E).
In contrast, PG&E and the Staff portray the incident as an example of the IST/ISI test program performing as intended. PG&E FOF 1 M-A69; Staff FOF i I-128. They stress that PG&E, in developing its NCR, determined the root cause of the error and thereafter instituted corrective action, returning the component to service and revising the maintenance procedure to provide additional guidance. PG&E FOF $ M-A71. In addition, they note that the other "similar" incident referenced by MFP was not related to maintenance and had a different root cause.
We agree with PG&E's description of this incident as only a " single isolated event." Id. Both the problem and its resolution appear to be adequately covered by the NCR.
This does not significantly detract from the adequacy of the maintenance program. We are somewhat concerned, however, about the adequacy of work instructions, a problem that is repeated in several of the incidents before us. In our opinion, but only as a suggestion because no Board order would be warranted here, PG&E should consider whether improved systemic procedures could generally improve the adequacy of work instructions.
- 14. Safety Iniection Emeroency Core Coolina System (ECCS) Accumulator Tanks.
This incident concerns the discovery by PG&E of it sications of intergranular stress corrosion cracking (ISCC) in the Safety Injection ECCS accumulator tanks in
both Units 1 and 2. Relying on a 1993 Staff Inspection Report (MFP Exh. 59, IR 93-08), together with a voluntary LER and a report of PG&E's Onsite Safety Review Group (OSRG)
(MFP Exhs. 60 and 61), MFP claims that PG&E's response to NRC Information Notice 91-05 was untimely and inadequate, that PG&E is not certain about the nozzle material used, and that financial considerations influenced PG&E's decision to delete its corrective action to replace all possible nozzle devices and piping in Unit 2 during its 5th refueling outage (2R5). MFP FOF ii 267, 272 and 276.
The LER relied on by MFP lists three instances in 1985 when ISCC in the Unit 2 accumulator nozzles was detected, two instances in 1986 and one in 1991, also in Unit 2, where ISCC was detected, and one in 1992 where ISCC was detected in Unit 1. MFP Exh. 60 (LER 2-87-023-01), at 4 (of LER).
The record also reveals that all the defective nozzles were replaced but, because of cost considerations and schedule impact, PG&E elected not to replatc all such nozzles. MFP Exh. 61, at 2.
MFP characterizes the deficiencies as disabling or threatening safety systems (outline i I.A), as representing untimely corrective action (outline i II.A), as examples of previous corrective action that failed to prevent recurrence 1
and of untimely detection and correction of aging effects l (outline $$ II.B, II . C) and as an example of PG&E decisions I i
regarding what is needed to maintain the plant in a safe l
condition being unduly influenced by financial considerations (outline i IV.G). For their parts, PG&E and the Staff portray the incidents as a maintenance program performing as it should--in PG&E's terms, "an operating experience that was thoroughly addressed by PG&E." PG&E FOF 1 M-A72. See also Staff FOF 11 I-132 through I-135.
As stressed by the Staff and as reflected in the record, when small nozzle 13aks were first detected in 1985-87, PG&E identified and successfully repaired or replaced all leaking nozzles. The frequency of surveillance inspections was also increased. PG&E further cut out several nozzles for metallurgical analysis. Analysis demonstrated that crack propagation was from the inside and thus not from exterior corrosion. Staff FOF i I-133, citing MFP Exh. 60, at 3 and Tr. 934 (Crockett). Thereafter, following Staff issuance of Information Notice 91-05, concerning the possibility of ISCC in accumulator tanks, PG&E performed further inspections and discovered several other indications of cracking, which were repaired.
With respect to this incident, we agree with the Applicant and Staff that the maintenance and surveillance program performed appropriately. As the Staff asserts, PG&E has been investigating the issue since its initial identification and has implemented appropriate corrective maintenance and increased surveillance. It has committed to an enhanced periodic inspection program for accumulator
nozzles. Staff FOF i I-135, citing Tr. 939-41 (Crockett) and MFP Exh. 60, at 8. Although the matter does have safety significance, as claimed by MFP, there is no evidence supporting the claims that PG&E was slow to initiate corrective action, that its attempted corrective action was unsuccessful or that PG&E fails to know the composition of the nozzles.
Further, the financial claim (MFP FOF 11 276-78), based on assertions that PG&E elected to repair or replace defective nozzles rather than replace all of them (as it had considered doing) because of cost and the structure of California's rate system in dealing with maintenance costs, is not appropriate for us to consider at this point. No systemic compromise of safety standards has been demonstrated, and there is no evidence that, because of California's rate system, PG&E is cutting corners on appropriate maintenance or surveillance activities.2s In short, we concur with the Staff assessment of the examination program for the accumulator tanks, "not only was it acceptable, it was considered very good work to be asWe also note that PG&E testified that "We don't put off required maintenance. This was evaluated, it was determined that we didn't have to do this [ full replacement) at that time and we didn't. . . instead of going with a plan that would automatically just replace all the nozzles we had preparations to inspect . . . all of the nozzles, and replace them if necessary." Tr. 940 (Giffen, Crockett).
1 I
J
looking this hard and finding these things and fixing them."
Tr. 2178 (Miller). ;
- 15. Corrosion cf ASW Annubar. DFO and CO2 Pioina.
This issue concerns corrosion in three types of underground piping: the diesel fuel oil (DFO) line for Unit 2; fire protection carbon dioxide (cardox) piping in Unit 2; and ASW annubar piping (in the form of a 1 1/2" hole).
Corrosion on the DFO piping was first observed in 1990; on the ASW annubar piping in June 1992; and on the cardox piping in January 1993. The corrosion on these three different pieces of equipment is related because of the similar location of the pipes--all of them are located below ground in a concrete trench on the west end or sea side of the turbine building. After each discovery, PG&E initiated corrective action. MFP Exhs. 62, 63, 64, 64A; PG&E Test., j at 99 (Crockett); Tr. 1059 (Crockett).
MFP claims that corrective actions taken after-discovery of corrosion in the DFO piping in 1990 were ineffective and failed to prevent further degradation; that PG&E's maintenance and surveillance program was not adequate to detect and sufficiently control the extensive corrosion ;
that has occurred in the pipe trench /pipeway; that there was inadequate initial application and maintenance of the coal tar protective coating which was intended to prevent corrosion on the piping in the trench /pipeway; that the trench /pipeway was not maintained in an adequate manner to l u
1
prevent the accumulation of water; that PG&E has been unacceptably slow to respond with corrective actions to alleviate the corrosion of pipes in the pipe trench /pipeway; and that PG&E's proposed corrective actions are unsubstantiated and should not be considered in this process.
MFP goes on to assert that PG&E determined that the DFO and ASW annubar piping remained operable despite the corrosion; that PG&E instituted compensatory measures to compensate for the inoperable cardox system; and that PG&E's operability / compensatory determination, however, is not an f
indication of an effective maintenance and surveillance program but rather an indication that PG&E was lucky this time. Further, MFP asserts that PG&E's failure to prevent the accumulation of water in the trench /pipeway contributed to the development of corrosion on the various pipes and is similar to its failure to maintain sump pumps in the vaults; and the submergence of the cables contributed to the severe degradation and eventual failure of the 12 kV cables (see item 3, above). Finally, MFP portrays these developments as an example of inadequate coordination between maintenance and operations personnel.
MFP categorizes these claims and assertions concerning i
underground corrosion of piping as disabling or threatening safety systems (outline i I.A), as an example of responding to maintenance problems in a lax and untimely manner
(outline i II.A), as an example where previous corrective action failed to prevent recurrence (outline i II.B) and of untimely detection and correction of aging effects (outline i II.C), and as an example of deficient communication and coordination between different groups of individuals and/or departments (outline i IV.A). Finally, the corrosion is attributed to inadequate procedures or work instructions (outline i IV.E).
PG&E characterizes these incidents as " operational experience with equipment that is well within the scope of PG&E's maintenance and surveillance capabilities." It goes on to observe that the maintenance and surveillance program
" functioned to find deteriorating piping and then to replace the piping with upgraded design, materials or construction techniques." PG&E FOF T M-A76. It acknowledges, however, that its 1990 corrective actions with respect to the DFO piping corrosion may not have been adequate. "If we had done more then, . . . we may have been able to do something to alleviate the situation." Tr. 1070 (Gif fin) .2' The Staff stressed that the 1990 actions were "not sufficiently comprehensive or conservative to prevent recurrence" but added that, "while further actions arguably could have been taken in 1990, in the vast majority of cases in which a problem has been identified, PG&E has taken prompt action asSee also MFP Exh. 64A, at 1-3 ("the previous corrective actions were ineffective," at 3).
l i
commensurate with safety significance." Staff FOF 1 I-145, citing Staff Test., ff. Tr. 2159, at 13.
Discovery of the DFO corrosion in 1990 arose from normal surveillance. As corrective action, PG&E repaired the pipe coating and increased the frequency of surveillance. Later inspections, however, revealed further corrosion in the DFO lines as well as the other piping.
Eventually, PG&E instituted a corrosion task force, comprising a multi-disciplined organization from engineering to review the material condition throughout the plant of any piping that may be susceptible to corrosion. Tr. 1062 (Crockett). PG&E is also looking at the design of the trench in which all three types of pipes were located and is changing the location of the DFO piping in the trench to minimize its exposure to standing water. The cardox piping is being removed from the trench completely and routed within the turbine building. Tr. 1084 (Giffin, Vosburg).
We here find no fault with the surveillance program--it discovered the problems and none of the pipes were ever j
inoperable (although conservatively some of the ASW piping was declared inoperable pending further review). Tr. 1085-86 (Vosburg). The maintenance program for repairing the DFO pipe in 1990 fell short but now appears to be addressing the corrosion problem adequately--i.e., the broad-based task force, which is studying the " big picture" (Tr. 1088 l
(Crockett)), appears to be an acceptable approach to the
I l
1 1
overall problem. Improvements in surveillance procedures have also been implemented, to facilitate discovery of corrosion in underground piping. Tr. 1076-77 (Giffin).
Thus, we agree with MFP that the corrosion has safety significance and that PG&E's initial corrective action (for the DFO_ piping) fell short. We also agree that communications between various departments could have been i improved. But the current program appears to be following a technically acceptable approach and is likely to represent a permissible means for developing an appropriate program for dealing with underground pipe corrosion for the extended terms of operation. Upgraded surveillance procedures have already been instituted, and one type of piping is being moved to another location. In short, we will not disqualify the maintenance program for past shortcomings that appear to have been recognized and, in our view as well as that of the 1
Applicant and Staff, corrected. We are, however, including this incident as one of those that requires PG&E to perform a study, to be provided the Staff, concerning upgrading of interdepartmental communications for maintenance-related activities. ,
- 16. Control of Measurino and Test Eauipment (M&TE).
Technical specifications for the DCPP provide that .
there be appropriate procedures to ensure that tools, gauges )
l and other measuring and testing devices be properly l controlled, calibrated and adjusted at specified periods to l
I I
i
maintain accuracy. MFP Exh. 66, Enclosure 1, at 1. PG&E has established such procedures.
In a February 1991 inspection, the Staff found both deficiencies in the M&TE programs and that PG&E had previously identified deficiencies but had not aggressively corrected them. MFP Exh. 69; PG&E Test., at 102 (Giffin).
These deficiencies resulted in a single (non-escalated) NRC enforcement action (Severity Level IV violation). MFP Exh.
71.
MFP asserts that the identified M&TE problems are long-standing, recurring and continuing, that PG&E's maintenance and surveillance organization failed to respond promptly to the deficiencies, that corrective actions taken by PG&E were ineffective to prevent recurrence, that PG&E management was insufficiently involved in the resolution of the M&TE deficiencies and that the recurring deficiencies have safety significance. MFP F0F 11 318, 323, 329, 338, 345. MFP categorizes the activities as disabling or threatening safety systems (outline i I.A), as indicating a pattern of responding to maintenance problems in a lax and untimely manner (outline i II.A), as previous corrective action that failed to prevent recurrence (outline i II.B) and as insufficient management involvement (outline i IV.A), and of personnel error due to inattention to detail and to failure to follow procedures (cutline i IV.D).
PG&E acknowledges the previous enforcement action and <
the failure of certain of the corrective actions previously taken. But it maintains that it has taken further corrective action, as to which MFP has nt. indicated any deficiencies. It suggests that current deficiencies are minor paperwork discrepancies that must be differentiated )
from the earlier more-serious problems. The Staff similarly believes that the ongoing corrective actions will be effective. The current NRC Staff senior resident inspector testified that she is satisfied with the current M&TE program based on her own in-depth inspections. Tr. 2192-94 (Miller).
It seems' clear that, in the past, there were both problems with the M&TE program and with maintenance activities designed to alleviate those problems. The record also establishes that the ongoing program appears to be working satisfactorily. That being so, it cannot conclusively be asserted that it casts doubt on the current or future maintenance program. Indeed, it may perhaps be validly claimed that the Staff's enforcement action is producing its intended result. In any event, the evidence of record concerning M&TE problems fails to establish significant weaknesses in PG&E's current maintenance program.
i r
- 17. Centrifuoal Charaina Pumo 2-1; Decraded Couplina, ,
On June 30, 1992, an increase in vibration on ,
centrifugal charging pump (CCP) 2-1 was identified by PG&E ,
Predictive Maintenance (PM) personnel. Investigation into. .
l the cause included taking a gear lube sample from the motor-to-speed increaser coupling. .During the sampling process ;
the coupling sleeve on the speed increaser side of the pump i was found to be stiff due to hardened lubricant. The sleeve was subsequently freed and a work order was issued to replace the coupling prior to any failure of the equipment. l The maintenance organization. initiated an NCR on July 1, l 1992. MFP Exh. 73.
The Technical Review Group (TRG) for this NCR concluded that the root cause of the problem was inadequate motor alignment criteria stemming from ambiguous vendor information. Contributing causes related to inadequate lubrication were also identified. This was the third occurrence involving vibration of CCP 2-1. (However the NCR reviewed only one previous incident involving excessive !
coupling wear that occurred in 1989.) The cause of excessive coupling wear in the 1989 incident was misalignment of the motor with respect to the speed increaser. The role of inaccurate vendor information as a root cause was not recognized in the analysis of the 1989 i
event. Tr. 1120-24 (Ortore). j i
_ _ _ . . , . ~ . , , . . . . . _ . . . . - . . _ - - . _ . . . _ . _ . _ . _ . . --- .
The TRG met five times in the year following the issuance of the NCR. The Group addressed such matters as corrective actions, investigative actions, actions to prevent recurrence, and additional prudent actions. The TRG also considered previous similar events and operating experience at other plants. Closure of the NCR was achieved on July 7, 1993.
The foregoing facts are not in dispute and the parties differ only as to the inference about the predictive maintenance program that should be drawn from this incident.
MFP urges the Board to find that the maintenance program is deficient because previous corrective actions should have prevented recurrence of this event, but failed to do so (outline i II.B), because the program includes deficient procedures or work instructions (outline i IV.E) and because the program for detecting manufacturing deficiencies or internal defects did so only by chance and should have identified the deficiencies before they became self-evident (outline i IV.F). MFP F0F $1 350-53. PG&E and the Staff argue that this incident is evidence of the effectiveness of the company's predictive maintenance program because the degraded condition of the coupling was found before any pump failure occurred. Tr. 1120-21 (Giffin; Ortore); PG&E F0F 11 M-A88-89; PG&E Reply FOF ii R-A63-64; Staff FOF $$ I-154-56.
The Board concludes that the degraded condition of the coupling was found and repaired before any equipment failure occurred and that the inspection was conducted as part of the company's systematic preventive maintenance program i
(PG&E Test, at 38-40 (Ortore); Tr. 1121 (Ortore)), which we l further conclude functioned effectively in this instance.
We base our conclusion not only on the circumstances of detection and correction of the degraded condition but also l on the systematic efforts of the Technical Review Group (TRG) to investigate the incident, find root causes, and develop remedies which could prevent future occurrences.
MFP Exh. 73, at 9.
The TRG investigation revealed that similar degradation had occurred in 1989 on the same pump and that corrective actions for that event had not prevented recurrence. In both cases the root causes were attributed to inadequate alignment criteria; however, in the earlier case it was not recognized that inadequate vendor information contributed to the problem. We have no record basis which might show that the 1989 analysis was lacking in diligence or that it resulted from some weakness in the PG&E maintenance program.
The self critical disclosure of the 1989 results indicates strength in the program because it contributed to a deeper analysis of root causes in the later analysis.
The discovery and correction of the latest pump problem together with the analyses made by the TRG over a period of j
l
nearly one year to identify root causes and preventative measures appear to the Board to have been both reasonable and effective. We rely on the NCR itself for this conclusion since there is no evidence of record to support a finding of weakness in the mair tenance program. MFP's assertion (FOF 11 350-51) that the company should have done better based on prior experience is therefore unsupported and we reject it. We find that PG&E's detection, correction and technical analysis of the degraded coupling on the centrifugal charging pump is substantial evidence of a properly functioning maintenance program at the plant.
- 18. Unit Shutdown Due to Inoperable Hiah Pressure Turbine Ston Valve.
Circumstances related to the failed turbine stop valve are set forth in LER 2-92-003-01 (MFP Exh. 74). MFP relies on the LER and testimony of PG&E witnesses to support its allegation of improper maintenance. There is no controversy among the parties regarding the facts of the valve failure, however. Accordingly we adopt, with minor modifications, MFP FOF 11 354, 355, and 356 in the following paragraph as an accurate factual summary of the event.
On March 22, 1992, a manual shutdown was commenced for Unit 2 when PG&E determined 't hat one high pressure turbine stop valve (FCV-144) was inoperable. FCV-144 is a l hydraulically-actuated swing check valve which protects the high pressure turbine from overspeed. PG&E disassembled
l 1
l 1
I FCV-144 and determined that "the nut that retains the valve disc to the valve swing arm had disengaged from the disc stem, allowing the valve disc to become separated from the valve swing arm." When the valve separated from the swing arm it caused a partial blockage of steam flow through the Main Steam Lead. PG&E has been unable to identify the root cause of this equipment failure. It postulates two modes of failure: (1) unscrewing of the nut off the stem; or (2) a failure of the nut / disc stem threaded joint. MFP FOF ij 354-56.
MFP requests the Board to find that "PG&E may have caused an undetectable failure through improper maintenance." PG&E responded that the immediate cause of the inoperable valve was equipment failure, and that neither the LER nor the testimony of record attributes this problem to a maintenance deficiency. MFP Exh. 74; Tr. 1126-27 (Vosburg). The NRC Staff sees no programmatic deficiency.
NRC Staff FOF 1 I-159.
In our view, the record does not support MFP's belief that the valve failure was caused by a maintenance deficiency. The root cause analyses points either to unscrewing of the nut off the stem holding the disc, or to stripping of the threaded joint possibly due to buffeting of the disc in the steam flow. MFP Exh. 74, at 3; Tr. 1133-34 )
l
1 l
(Giffin). Moreover, no " undetectable failure" occurred."
The valve failure was detected by a rapid load reduction in power of approximately 10%. Thus the valve failure was self-disclosing.
The root causes of failure are likely traceable to manufacture or original installation of the valve.
Tr. 1133-34 (Giffin). Prior maintenance and inspection of-the valve, however, did not disclose any abnormalities that might have warned of potential failure. Had degraded conditions stemming from either possible root cause been present on inspections prior to the failure, they could have ,
(and should have) been detected. Although the valve was manufactured and set up by Westinghouse, which performed at least one prior inspection during an outage, degraded ,
conditions were not discovered prior to failure. Tr. 1131-i 32; 1134-36 (Giffin). The record does not disclose whether degraded conditions were present but not observed on initial installation or on a prior inspection or whether they developed during subsequent operation. However, PG&E has now established enhanced inspection programs specifically to detect possible failures stemming from either of the root causes that have been identified from this event. Tr. 1134-35 (Giffin).
"MFP has not made clear how a maintenance program I could be improved to permit the detection of " undetectable i failures."
There are eight such valves in the two units at Diablo Canyon. The failure described herein is the first of its kind in the life of the plant. Inspection of the seven valves which did not fail did not disclose any abnormal conditions. Tr. 1128-29 (Vosburg). The Board concludes that this failure was isolated and that no adverse inference about the PG&E maintenance program can be drawn from it.
For all of the foregoing reasons, we reject MFP's assertion (outline i IV.B) that this failure supports a general conclusion that maintenance activities may cause undetectable failures and future safety problems. We also reject MFP's assertion (outline i IV.F) that PG&E generally lacks an effective program for detecting manufacturing deficiencies or internal defects for the same reasons. MFP F0F 11 47-48, 57-59.
- 19. Diesel Generator 2-2 Failure to Achieve Rated Voltage.
On December 29, 1992, diesel generator (DG) 2-2 was subject to a post maintenance test. The DG started but did not load because the generator did not achieve rated voltage. MFP Exh. 75, at 1; MFP Exh. 76, at 1; Staff FOF 1 I-160. PG&E determined that the failure to load occurred because all four generator slip ring brushes were out of position after maintenance had been performed. PG&E found that the mispositioning occurred inadvertently when a mechanical maintenance worker loosened some mounting bolts i
i l
l l
to conduct an inspection of the generator shaft, rotated the shaft manually, and then retightened the bolts without inspection of the position of the brushes. MFP Exhs. 75, 76; Tr. 1139-40 (Giffin). The immediate cause for this event was personnel error. Tr. 1139-40 (Giffin). The root cause was inadequate electrical information being supplied ,
to the mechanical technician. MFP Exhs. 75, 76; Tr. 1141-43 l (Giffin). Corrective action was taken to revise the maintenance manual for the DGs to advise technicians on how to prevent recurrence of the error. When the generators are involved in maintenance, procedures will also be revised to involve electrical technicians. Tr. 1144 (Giffin).
MFP requests the Board to find the maintenance program defective because the error outlined herein demonstrates inadequate procedures and a lack of communication or i
supervision among electrical and mechanical maintenance personnel. MFP F0F ji 362-68. The Applicant asserts that this incident is an example of how a maintenance program should work and that MFP's conclusion that the program is globally defective because procedures had to be improved is unwarranted. PG&E FOF i M-A94. The Staff sees no programmatic defect because, even though the error occurred, it was found by surveillance and PG&E took corrective I
actions. Staff FOF $$ I-162, 163, 164.
1 Errors in maintenance and communication occurred as asserted by MFP; however, its exhibits and PG&E testimony
i show that PG&E's surveillance was effective, that it analyzed the problem, and that it took corrective action which appears reasonable and effective. MFP Exhs. 75, 76; Tr. 1142-44 (Giffin). MFP has not controverted this aspect of the exhibits and testimony. Because the errors found here have been effectively analyzed and corrected, this incident cannot serve as cumulative evidence in support of MFP's assertion (outline $1 I.A, IV.B, and IV.E) of general deficiency in the surveillance and maintenance program. The Applicant has carried its burden of proof with respect to the diesel incident.
- 20. Missed Alert Frecuency STP for Auxiliary Salt Water Pumo 1-2 and Component Coolina Water Valve CCW-2-RCV-16.
MFP introduced three exhibits that addressed two separate incidents of personnel error and missed surveillance tests: one on a component cooling water (CCW) valve and the others on an auxiliary salt water (ASW) pump.
MFP Exh. 77 (NCR DC2-93-TS-N005 Rev. 00) ; MFP Exh. 78 (NCR DC1-92-TP-NOS2 Rev. 00); MFP Exh. 79 (LER 1-92-024-00). ,
There is no dispute that the required surveillance tests were missed and that personnel error was responsible.
Missed surveillance tests are the only common factors in the otherwise unrelated incidents invo3ving the valve and pump.
The Board is called upon to decide whether the missed surveillance tests are indicative of a general deficiency in the PG&E maintenance and surveillance program, as claimed by
MFP (outline 11 I.A, III, IV.C, IV.D), or are simply examples of isolated personnel errors that do not suggest a pervasive programmatic breakdown, as claimed by PG&E and the NRC Staff (PG&E FOF i M-A99; Staff F0F 11 I-167, I-168).
Details of the missed surveillance tests and other error are given in MFP Exhs. 77, 78, and 79. There being no factual dispute, only a brief summary of the incidents need be set forth here:
Incident 1. In the first incident, the CCW valve was stroke tested in October 1992, and based on results war placed on alert which required the test frequency to be changed from once every 92 days to once every 31 days. The new test frequency was lost from the plant computer system at the time of entry because of peraonr.el error so the next test was performed on the regular 92 day schedule rather than the 31 day schedule. The required surveillance test was missed twice in the interim. The error was discovered during the 92 day surveillance. PG&E FOF i M-A96; Staff F0F i I-165; MFP FOF $1 375-77.
Incident 2. This incident involved two instances of personnel errors during surveillance testing of an ASW pump.
In the surveillance test of August 21, 1991, the reviewer used an incorrect pump curve to determine the required differential pressure and failed to recognize that the ASW pump should have been declared inoperable. This was later found to be a reportable violation of applicable technical
L
- 100 -
specifications. In the surveillance test of the same pump on November 14, 1991 the reviewer wrongly determined that the differential pressure test was satisfactory when it actually was within the alert range. The pump should have been placed on alert which would have required surveillance on a 46-day testing frequency. The test was missed on January 29, 1992 and PG&E was, on that date, in violation of applicable technical specifications. The error was discovered on October 15, 1992 during review of a similar test that had been performed on October 7, 1992. MFP Exhs.
78, 79; PG&E FOF 11 M-A97, M-A98; Staff F0F $1 I-165, I-166; MFP FOF $1 369-70, 372-73.
MFP requests the Board to find that these examples of personnel error constitute cumulative evidence in support of their broad assertions of programmatic deficiency in the Diablo Canyon maintenance and surveillance program (outline
$1 I.A, III, IV.B, D).
Although the personnel errors occurred as described by MFP, the Board rejects MFP's view that the errors are contributing evidence to an inference of programmatic breakdown. The view expressed by MFP is contrary to uncontroverted direct tectimony of the Applicant. In each case, the testimony describes both preventative and corrective actions within the maintenance program. PG&E's direct testimony elaborates at length on that portion of the PG&E program that provides for root cause analysis, failure
- 101 -
trending and correction of equipment and' personnel failures.
PG&E Test., at 58-62 (Giffin). The Board finds that systematic provision in a maintenance program for the analysis and correction of degraded conditions is an integral part of the overall program and must be considered in determining the quality of the program.
Each of MFP Exhs. 77, 78 and 79 contains analyses not only of the circumstances of equipment and personnel failure but also of the root cause analyses and corrective actions that were recommended by the respective technical review groups. None of the corrective actions were explored or controverted by MFP on cross-examination. The Board finds from its own inspection of the exhibits that the analyses and recommended corrective actions in each case appear to be reasonable. They are indicators of a properly functioning program.
There is no support anywhere for MFP's apparent belief that the worth of the overall maintenance program can be determined by counting unrelated equipment or personnel failures. There is no objective standard for such determinations and MFP ignores the integral role of root cause analyses and corrective actions in such programs. If these actions are effective, we cannot say that the program ,
itself is generally defective even though equipment and personnel failures occur. We find that the examples of failure cited in MFP Exhs. 77, 78, and 79 do not constitute i
- 102 -
contributing evidence that PG&E's maintenance program is generally defective.
- 21. In-Service Promot Test Data Ouestionable.
MFP offered Exhibits 81 and 82 as examples of errors in a surveillance test procedure (STP). MFP Exh. 81 (NCR DCO-92-TN-N055 Rev. 00, 3/1/93); MFP Exh. 82 (PG&E Letter No.
DCL-92-262, 11/25/92). The STP contained a diagram showing an erroneous location for taking vibration' measurements on Auxiliary Feedwater Pumps (AFW). The diagram showed an arrow wrongly pointing to the pump housing instead of the correct location on the pump bearing caps. A PG&E engineer discovered and corrected the error in the diagram after several months but did not initiate an Action Request (AR) to track the matter and to assure that-tests conducted while the diagram error was in place were done correctly. NRC issued a Severity Level IV NOV for failure to issue an AR which was required by procedure. PG&E agreed with the violation. There is no dispute that the erroneous diagram existed and was used in many inspections before the error was found and corrected.
There is no evidence that incorrect data were taken because of the diagram. Tr. 1154-57, 1159 (Crockett). Even if measurements had been taken from the pump housing instead l
of the pump bearing cap, abnormal vibration would have been l detected. MFP Exh. 81, at 7. We adopt the factual j l
l
I
- 103 -
1 description of the NRC Staff as an accurate summary of l events leading to the NOV. Staff FOF ij I-169, I-170, I-171.
MFP calls upon the Board to find that this incident indicates a weakness in PG&E's surveillance testing program (outline, 11 III, IV.C). MFP FOF $$ 387-89; 42 (failed checks and balances or multiple barriers); 49-50 (repetitive pattern of inadequate surveillance). The Applicant and NRC Staff see the error as minor; in the nature of a typographical error. PG&E FOF 11 M-A101, M-A102; Staff FOF
$ I-172.
The Board finds that this minor incident had virtually no safety significance and does not support an adverse inference on the overall quality or effectiveness of the surveillance program. The Diablo Canyon maintenance and surveillance program is premised on the need for both preventive maintenance (including surveillance) and corrective maintenance. Corrective activity is an integral part of the program because preventive maintenance (and surveillance) cannot avoid all corrective maintenance. PG&E Test., at 11-14 (Crockett, Giffin); at 38-42 (Ortore, Giffin, Vosburg).
MFP's own exhibits show that, even for the minor error described here, PG&E undertook a substantial corrective effort which included internal review of its procedures leading to the discovery of the diagram error and icview by
- 104 -
a technical review group which analyzed the event, identified the root cause, and considered whether dots I painted on equipment to guide vibration measuremento should j be programmatically controlled.2s The corrective actions appear reasonable; there is no contrary evidence. The Board concludes that this event does not contribute to an inference of programmatic deficiency in PG&E's maintenance and surveillance program.
- 22. Hold-Down Motor Bolts On Centrifucal Charcina Pumns.
MFP Exh. 83 is a PG&E NCR that reports several l
discrepancies in Centrifugal Charging Pump (CCP) 2-1 hold ]
down motor bolts found by PG&E during preventive maintenance in July 1992. Further investigation by PG&E subsequently revealed hold-down bolt discrepancies on other CCPs as well.
The discrepancies were attributed to flaws in the original procurement specifications and vendor supplied information during plant construction in the 1970s. Some of the conditions now regarded as discrepant such as machined hold-down bolts, elongated bolt holes and stacked washers were done for motor alignment purposes and were accepted field asMFP urges adoption of FOF i 385, alleging that guide dots missing from equipment are an additional indicator of program weakness. We reject this view because the dots were neither regulatory nor plant requirements and served only as informal aids to technicians. They had no safety significance.
- 105 -
practice during plant construction. MFP Exh. 83, at 1-5; Tr. 1161-62 (Giffin).
MFP requests the Board to find that PG&E's maintenance program is defective because it failed to identify the hold-down bolt discrepancies in a timely manner and it has given inadequate attention to identification of discrepancies related to initial installation. MFP FOF 11 393-401. It claims that this issue contributes to a showing that most of PG&E's past maintenance problems have threatened or disabled essential safety systems (outline i I.A). MFP F0F $$ 25-27.
PG&E argaes that MFP's exhibit shows the preventive maintenance program working as it should in finding manufacturing discrepancies that had existed since original procurement. No defect in the maintenance program is shown by this occurrence. The Staff agrees that it does not show any defect in the maintenance program. Neither does it show any current problem in procurement which is, in any event, outside the scope of the admitted contention according to both the Applicant and Staff. PG&E FOF i M-A103; Staff FOF
$1 I-173 through I-176.
The Board finds that this occurrence had no public health and safety effects and does not support MFP FOF $$
25-27, which allege a general deficiency in maintenance of safety systems. We also find that any flaws existing in the 1
original procurement program are outside the scope of contention I and need not be considered further.
- 106 -
MFP's concern for timely discovery of flaws in hold-down bolts is supported by its exhibit which shows that PG&E had not implemented a Westinghouse technical bulletin and thus did not find the problem earlier. MFP Exh. 83, at 5.
We adopt MPP's FOF 11 393-96 which assert that the problem could have been found earlier.
The truth of MFP's proposed findings applies to this particular incident but does not assist it in proving programmatic deficiency in PG&E's maintenance program. We know of the missed opportunity to discover the problem earlier from the self-critical analysis of the PG&E Technical Review Group. Self-critical analysis is an indicator of integrity in the program. The scope of TRG investigation included consideration of similar problems on other components and the formation of a hold-down bolt " hit team" to track corrective actions and assure completion of inspections of all components identified as a result of this TRG. MFP Exh. 83, at 8, 9. The analyses and corrective actions of the TRG appear reasonable; there is no contrary evidence.
We find that root cause analysis and broadly based corrective action are indicators of strength in the maintenance program. The deficiencies described herein could have been discovered earlier; however, that has no implication for the safety of future operations because the lessons from this occurrence were learned and incorporated
- 107 -
into the maintenance program. The Board finds that MFP Exh.
83 demonstrates a now-properly functioning maintenance program and is not now supportive of an adverse conclusion on that program.
- 23. Reactor Coolant System Leakace.
MFP Exhs. 84 and 85 are an NCR and LER respectively that describe an incident of excessive reactor coolant system (RCS) leakage that occurred at Diablo Canyon Unit 2 on August 13, 1991. MFP Exh. 84 (DC2-91-MM-N069 D14, 2/2/:13 ) ; MFP Exh. 85 (LER 2-91-004-00, 9/16/91). There is no dispute among the parties that excessive RCS leak rates existed; that the leakage was not promptly detected because of personnel error in calculation of leak rates; that the error resulted in a violation of a Technical Specification (reported to NRC in the LER); and that the leakage occurred in the body-to-bonnet joint of check valves in the Chemical Volume Control System (CVCS). The valves had degraded studs from boric acid corrosion. The full description of this event is given in MFP Exhs. 84 and 85 and the Board adopts MFP FOF 11 402-408 as uncontested findings of fact that summarize the foregoing occurrences.
MFP asserts that PG&E failed to establish an effective surveillance program that would have prevented or detected the degradation and leakage despite prior industry communications on this subject. It asserts further that I
l 1
i I
- 108 -
PG&E's corrective action is of questionable effectiveness, ,
l l l vague, indefinite and lacking in commitment. MFP F0F l $$ 409-20. It also claims that this incident is contributing evidence to a general conclusion of l l
programmatic deficiency in the surveillance and maintenance I
l program because safety systems were threatened; there was 1
untimely detection and correction of aging effects; and there was generally inadequate surveillance and testing (outline 11 I.A, II.C, IV.C). MFP FOF $$ 26-27, 36-37, 49-50.
MFP filed no proposed findings on the personnel error that led to failure to meet a Technical Specification as set forth in MFP Exh. 85, and we treat that issue as abandoned.
PG&E asserts that its corrective actions were effective and its existing surveillance procedures are adequate to detect leakage. PG&E F0F i M-A109. The Staff asserts that this occurrence does not evidence a breakdown in the overall program. Staff F0F $ I-181.
The Board finds that prior to this incident PG&E had effective procedures for detection of leaks from the RCS but did not have a preventive maintenance program for the inspection and retorquing of valve bolts or for inspection of gaskets or for the detection of boric acid corrosion on bolts. MFP Exh. 84, at 2-5, 9; Tr. 1185-86 (Giffin, Vosburg).
l
- 109 -
As a result of this incident PG&E inspected and maintained 17 additional Unit 2 valves and developed additional corrective actions to replace nuts and studs showing corrosion and to retorque all nuts on valves. A
" hit team" program was developed for replacing bolting on valves that have carbon steel (B7) bolts that are exposed to boric acid with stainless steel bolts. The bolt replacement has been accomplished. This eliminated the problem of bolt corrosion from boric acid because stainless steel bolts are resistant to corrosion. Tr. 1184 (Giffin).
The Board rejects for lack of evidence MFP's assertions that the corrective actions adopted by PG&E are vague, indefinite or lacking commitment. The corrective actions summarized above and set forth in detail in MFP Exh. 84 and in PG&E testimony appear to the Board to be clear and understandable. Contrary to MFP FOF 11 418-20, we find it commendable rather than sinister that PG&E took " prudent actions" in addition to " corrective actions to prevent recurrence" as a result of this incident. There is no evidence that " prudent actions" will not be implemented.
Although MFP is correct in its assertion that PG&E had no preventive maintenance program that would have directly prevented or detected degradation of bolts prior to the incident described herein, the Board rejects MFP's assertion ,
of PG&E negligence (MFP F0F $$ 409-11). Industry and regulatory documents cited only by title in support of that ;
1 l
- 110 -
assertion are not before us and, without Board knowledge of their contents, are not adequate evidence to support such a finding. We also reject MFP's assertion that no information exists on the condition of Unit i valves (MFP F0F i 412).
They have been inspected for leakage. Tr. 1182 (Giffin).
Moreover, MFP's exhibit is an account of events that took place at Unit 2; it is neither alarming nor significant that it contains no information on the condition of valves in Unit 1.
There is no basis for concluding that a preventive maintenance program targeting valve bolts and gaskets should have been in place earlier. Reasonable minds may well have decided initially that leak detection was adequate to protect the reactor from excessive loss of coolant, as proved to be the case here. Staff FOF i I-180. MFP simply disagrees; however, the dispute is academic. PG&E has now adopted the additional inspection and maintenance procedures for retorquing of bolts, and for detection and prevention of corrosion as advocated by MFP. These actions resulted in improvements in the effectiveness of the maintenance and surveillance program. The Board does not find the prior program defective simply because it was later improved as a result of operating experience. The program improvements adopted by PG&E in this case contribute to a finding of reasonable assurance of safety in future operations. The I l
Board rejects as unsupported MFP's assertion that this
- 111 -
L
?
incident is evidence of a general 2y defective surveillance :
and maintenance program.
- 24. Reactor Cavity Sump Wide Rance Level Channel 942A Inocerable.
The issue of inoperable channel 942A requires the Board to decide whether the particular failures of equipment and personnel that occurred are also evidence supporting a general finding of programmatic deficiency in the PG&E surveillance and maintenance program.
MFP Exhs. 86-89 describe events related to two occurrences of an inoperable reactor cavity sump wide range level channel. MFP Exhs. 86 (DC2-91-TI-NO96 D8); 87 (PG&E Letter No. DCL-92-090, 4/20/92); 88 (NRC Letter to PG&E with ,
NOV attached, 2/28/92); 89 (PG&E letter No. DCL-92-071, 3/30/92). These channels are instruments used to provide post accident water level data inside containment. The data are used to verify the occurrence of a loss of coolant accident (.OCA).
L Two channels are required by Technical Specifications to be operational when the reactor is in one of several specified operational modes. MFP Exh. 89, at 1. ,
The channels are not the only way of detecting a LOCA; other ,
redundant means for doing so exist at DCPP. Tr. 1200-01 (Vosburg).
The first equipment malfunctions occurred in 1990 when two channels became inoperable and were not detected by ,
operators from August 21, 1990 to November 6, 1990. The ,
-im , - . , ,-
- 112 -
failure to detect the malfunctions resulted in a violation of a Technical Specification which was reported to NRC in an LER. The second incident began with a channel instrument malfunction on October 10, 1991 that went undetected until October 22, 1991, when an NRC Inspector found it by reviewing the Safety Parameter Display System (SPDS). The failure to detect the malfunction for more than seven days with the reactor in mode 2 (startup) or mode 3 (hot stendby) was a violation of a Technical Specification and NRC issued a Severity Level IV NOV. PG&E acknowledged the violation in its response to the NOV. MFP FOF $$ 421, 423, 424; MFP Exh.
89, at 1.
NRC's cover letter with the NOV was critical of PG&E for taking inadequate corrective actions in the 1990 event that failed to preclude the 1991 undetected failure. MFP Exh. 88. NRC's main concern was for inadequacy of the operator surveillance program that resulted in failure to detect equipment malfunction. NRC has only minor concern for the equipment failure itself, which it regards as a common occurrence. Tr. 2199 (Narbut).
MFP requests the Board to find with respect to this incident that PG&E's corrective actions to train operators after the 1990 event were inadequate to prevent recurrence of undetected equipment failure. MFP FOF 11 425-27, 429-30, 431-33. It also asserts that the SPDS system is not maintained adequately. Finally it argues that this issue is
- 113 -
I i
relevant to maintenance and surveillance even though the !
i personnel failure that led to the NOV was by control room operators rather than maintenance personnel. MFP also asserts that this incident supports a conclusion of general -
deficiency in the PG&E surveillance and maintenance program because it is part of a more general pattern: it involved safety systems; PG&E's response was lax and untimely; the problem recurred; and surveillance was inadequate (outline,
$$ I.A, II.A, II.B, IV.C).
PG&E minimized the significance of the equipment and personnel failure because the incident is isolated and now resolved; it involved operator failure which does not reflect on PG&E's ability to maintain equipment; failed indicators are not uncommon; and the majority of its corrective actions are effective. The NRC Staff sees no general deficiency in the maintenance program because operator knowledge problems are not widespread and subsequent corrective actions were effective.
The Board finds that MFP's assertion of inadequate corrective action in 1990 that permitted recurrence of equipment malfunction and operator failure to detect a malfunctioning channel in 1991 is true and uncontested in this case. No root cause was found for the equipment failures in 1990 or 1991 and correction was finally achieved by replacing the affected equipment and cables. Cause for the operators' failure was their reliance on a misleading l
l l
I t
i i
- 114 -
P chart record and their failure to understand indications of i failed channels shown on the SPDS despite instructions i
issued in 1990. Tr. 1191-96 (Vosburg, Crockett). However, ,
corrective actions on equipment have been effective and [
chere have been no channel failures since the 1991 incident.
Tr. 1199 (Crockett). Surveillance of the SPDS by control ,
room personnel is now adequate. The issue is closed. Tr.
2200-2201 (Miller). [
Corrective action is an integral part of the surveillance and maintenance program. Even though equipment and personnel failures occurred, the program cannot be found generally deficient if corrective action was prompt and effective. The fact that correction took two tries in this case was not unreasonable under the circumstances.
Resolution of the intermittent equipment problem was difficult but there was no evidence of laxness, lack of diligence or lack of commitment in addressing it. Between the 1990 and 1991 incidents, four 942A channel failures t
occurred but none exceeded the 7-day technical specification I
action statement. These failures did not go undetected and there is no evidence of prior warning of deficient surveillance by operators during that period. The issue is now closed and there is nothing left of it that undermines 1 l
reasonable assurance of safe operation during the recapture period.
I 1
1
- 115 -
The Board rejects for lack of evidence all of MFP's assertions alleging that the SPDS system is unreliable. No record exists to support a conclusion about the reliability of the SPDS system. MFP cites unrelated equipment failures that do not support its proposed findings on the reliability of the SPDS. Contrary to MFP's claim, there is no regulatory requirement for the SPDS to be seismically qualified (Tr. 1197 (Giffin)). Allegations of unreliability in the SPDS are without merit.
Contrary to MFP's claim, the failure of control room operators to perform adequate surveillance in this case has no general adverse implication for the adequacy of the surveillance and maintenance program which is the subject of the admitted contention.
The Board concludes that the particular failures of equipment and personnel occurred as alleged by MFP but these failures do not support an inference of programmatic deficiency in the PG&E surveillance and maintenance program.
- 25. Desian Criterion Memorandum (DCM) Reauirements.
The issue raised by MFP based on Exh. 90, MFP POF, $1 446, 450, requires the Board to decide whether the review of 1
(es.lg" documentation undertaken by PG&E shows a programmatic l l
wear. ness in the surveillance and maintenance program
)
- 116 -
(outline i IV.C) . MFP Exh. 90, DCO-93-TN-N006 D1, February 12, 1993."
PG&E conducted an audit (as part of an upgrade project) to determine the adherence of its maintenance. program to its administrative requirements (Design Criterion Memorandum Category I (DCM)). The purpose of the audit is to remove discrepancies between design documents and the surveillance and maintenance program. PG&E regards this as an important purpose, as does NRC. The upgrade program for category I devices was expected to be completed by the end of 1993.
Tr. 1204-09 (Crockett).
In the course of the audit, PG&E encountered a discrepancy wherein no test exists to provide verification of the emergency diesel generator fuel oil day tank low level switch transfer pump start signal actuation. MFP Exh.
90, at 2. Although the equipment discrepancy in this instance should be removed, the finding itself is of minor significance because, when the diesels start automatically, operators start the pumps manually and do not rely on the switch. Tr. 1203-04 (Crockett).
MFP FOF $1 448-49 appear to agree with PG&E as to the necessity for the design document review, the necessity for it to be completed, and the need to make changes that result from it. There is no controversy in those assertions in "See also PG&E Exh. 23 (DCO-93-TN-N006, August 23, 1993), which is an updated version of MFP Exh. 90.
- 117 -
need of resolution inasmuch as PG&E testimony shows that is its intent.
We reject, for lack of evidence, MFP's assertion that the review should be extended to include " Class 2"-
equipment. There is no mention of that matter anywhere in l the record and the Board has no basis to decide it. We also reject MFP's concern that there is uncertainty as to PG&E's l
- 1. commitments to NRC. No record was developed on that matter at the hearing and NRC has not addressed it.
The Board concludes that there is no evidence at all related to the DCM issue that contributes to a finding of general programmatic weakness in the surveillance and maintenance program. Critical self-assessment, voluntarily undertaken by PG&E, indicates program strength not weakness.
The Board rejects any possible implication by MFP that l
l l voluntary program improvement by PG&E, per se, supports an inference of prior program deficiency. MFP's assertions in the DCM issue are without merit and are rejected in their .
l entirety.
- 26. Pipe Suncort Snubber Damace.
The issue raised by MFP based on Exhs. 91 and 92 calls upon the Board to decide whether an incident involving a damaged pipe support snubber is evidence of a generally deficient surveillance and maintenance program at DCPP. MFP Exh. 91 (NCR DC1-92-MM-N021, Rev. 00, February 12, 1993);
MFP Exh. 92 (PG&E Letter No. DCL-92-264).
i
_____------_----_-_-------_-----_------------_-----_-------__--__--_--J
- 118 - i 1
I MFP claims the snubber failure contributes to a !
I conclusion of general programmatic deficiency because it shows that PG&E lacks a program to detect manufacturing deficiencies or internal defects and it thereby did not prevent the snubber failure (outline i IV.F). MFP FOF ii 57, 58, 59, 454, 455. PG&E claims that the failure is due to a manufacturing defect that does not implicate maintenance and that the defect could not be detected absent disassembly of the snubber. PG&E FOF i M-124. The Staff agrees with PG&E. Staff FOF i I-196.
PG&E found the damaged snubber during a system walkdown in May 1992. The snubber was locked and buckled. Root cause analysis showed that the failure was due to an out-of-tolerance condition on the interior stainless steel verge wheel combined with stress and chloride exposure from salt air at an outdoor location. The out-of-tolerance condition resulted from a manufacturing defect which led in turn to excessive stress in service. The wheel failed under excessive tensile stress due to stress corrosion cracking.
PG&E tested and overhauled or replaced other snubbers similarly situated. MFP Exh. 91, at 3-4.
MFP is correct in its claim that PG&E did not have a program in place that would have prevented this failure.
However, MFP cites no regulatory requirement that would obligate PG&E to devise such a program, and we know of none.
- 119 -
MFP Exh. 91 discusses the regulatory. scheme-governing snubbers. There is a Technical Specification requirement that all snubbers be operable during specified plant operations with some exclusions for non-safety related i systems. Technical Specification 3.7.7.1 requires that when a snubber becomes inoperable it be replaced or restored to operable status within 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> from the time of discovery.
The regulatory scheme therefore relies on corrective action and obligates PG&E to correct snubber failure within a specified time period. MFP Exh. 91, at 2.
The PG&E corrective maintenance program is recognized with approval by NRC. Staff Test., ff. Tr. 2159, at 2 (Peterson), 5-7 (Narbut, Miller). Corrective maintenance is a normal part of the surveillance and maintenance program. ,
PG&E Test., at 38-41 (Ortore, Giffin). The program therefore cannot be found generally deficient solely because equipment failure was not prevented or because corrective action was necessary to restore. failed equipment. PG&E has a program for the inspection and testing of snubbers. The program would not detect internal stresses on components or -
manufacturing defects before failure. There is no evidence that effective preventive maintenance could have been I'
devised to prevent failure from manufacturing defects. Such a program would require disassembly of snubbers for the purpose of determining manufacturing tolerances of components. There is no assurance that component stress or l
- 120 -
corrosion would be present and visible when disassembly was undertaken. Tr. 1218 (Giffin). No expert has advocated that PG&E devise such a preventive maintenance program.
MFP has not cited any deficient or inadequate corrective action taken by PG&E in this incident and we found none in our review of the record. We find that PG&E had no regulatory obligation to undertake a preventive maintenance program for the purpose of preventing the snubber failure that occurred. There is no evidence that such a program would improve safety because the failure was a unique and isolated case. MFP Exh. 91, at 4. The Board finds that the snubber failure has no adverse implications regarding the programmatic adequacy of the PG&E surveillance and maintenance program.
- 27. Gas Decav Tank Missed Surveillance.
The issue of the missed gas decay tank surveillance is based on MFP Exhs. 95 and 96 and requires the Board to decide whether this incident is part of a pattern of missed surveillance tests that reflects a programmatic weakness in the surveillance testing program.
MFP asserts that this incident is part of a pattern that demonstrates a weakness in PG&E's surveillance testing program. This incident is said to be evidence of a general program deficiency because it is part of a pattern of disabled or threatened safety systems and part of a repetitive pattern of missed surveillances (outline ji I.A,
- 121 -
IV.C). MFP FOF $$ 25-27, 49-52, 461. PG&E asserts that this incident does not reflect a pervasive problem and does not represent a programmatic breakdown. PG&E FOF 1 M-A128.
The NRC Staff asserts that this incident has nothing to do with the maintenance program at Diablo Canyon and that it does not represent a programmatic breakdown. Staff F0F 1 I-201.
In this incident, a 24-hour gas decay tank surveillance required by technical specifications (TS) was not performed within required time limits. The test was performed about two hours later than required. The late test counted as a missed surveillance and a TS violation that resulted in the preparation of an NCR and an LER submitted to the NRC. MFP Exhs. 95, 96.
The root cause of the missed surveillance was an inadequate instruction given to the technician. The instruction said the test should be performed daily when it should have said the test was required by technical specifications to be performed every 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />. Contributing to the error was the fact that the errant chemistry technician had forgotten the requirements although he had previously received 20 hours2.314815e-4 days <br />0.00556 hours <br />3.306878e-5 weeks <br />7.61e-6 months <br /> of instruction on them. MFP Exh. 95, at 5-6, 13. When circumstances required that he defer the surveillance, he did so without regard to the expiration of the TS time requirement. Corrective actions I
- 122 -
included revising the instructions and counseling the
- technician.
The Board finds that root cause determination and corrective action taken by PG&E were adequate as stated in ,
MFP Exhs. 95 and 96 because there is no contrary evidence and our own review' revealed no deficiency." Corrective action is an integral and necessary component of a f
surveillance and maintenance program. When performed adequately, corrective action provides reasonable assurance of safety in future operations with respect to the specific incident at issue. The fact that reliance is placed on corrective rather than preventive action is not per se an indication of programmatic deficiency. It is the failure of corrective action that leads to concerns for the adequacy of a program. Tr. 2202-03 (Miller). Thus, there is no regulatory requirement to prevent all unwanted events even though MFP would have it so. Moreover, where corrective $
action has been adequate and the incident is properly brought to a close, the incident itself cannot logically be used as cumulative evidence of programmatic deficiency. For all of the foregoing reasons, the Board concludes that the incident of the missed surveillance is not evidence of a "MFP did not cross examine PG&E witnesses on any aspect of the missed surveillance and it did not controvert ,
the adequacy of PG&E's cause determination and corrective action.
l 4
i l
- 123 -
general programmatic deficiency in the PG&E surveillance.and maintenance program.
- 28. Seismic Clios Not Installed.
MFP Exh. 98 is an NCR that describes and analyzes events surrounding the discovery on December 3, 1992 that Unit i reactor trip and bypass breakers did not have seismic clips installed as required by plant procedures. The nonconforming condition existed from the time of Unit 1 restart on November 3, 1992 to December 3, 1992, when it was corrected. MFP Exh. 98, at 2-5 (NCR DC1-92-OP-N062, Rev. O, January 27, 1993).
MFP asserts that this event contributes to a conclusion of general programmatic deficiency in the surveillance and maintenance program because: PG&E root cause analysis identified a programmatic reficiency; prior corrective action in a similar event did not prevent this one; other contributing causes such as personnel inattention further demonstrate program weakness (outline 11 I.A, II.B, IV.A.,
PG&E claims that it corrected the problem promptly; no inoperable conditions existed without seismic clips; it is making programmatic corrections. PG&E F0F ji M-A132 -
M-A134. The Staff agrees with PG&E and claims that this isolated event is not indicative of pervasive programmatic failure. Staff FOF $$ I-202 - I-205.
- 124 -
There is no dispute concerning MFP's assertions of procedural error and degraded condition of equipment associated with this incident. Each error is cited directly from MFP Exh. 98, which is PG&E's self critical NCR. The Board is called upon to decide whether these errors support an inference of general programmatic deficiency in the PG&E surveillance and maintenance program. MFP Exh. 98, at 7; MFP FOF $1 467, 468.
l The record contains no factual evidence related to this event other than that in Exh. 98 and the testimony of PG&E's experts. As has been MFP's practice, it cites PG&E's documents with precision for the purpose of demonstrating error but takes no account of the analyses and corrective actions that are also set forth in the NCR and testimony.
The Board accepts MFP's assertion that error in procedure and degraded condition of equipment existed at the time of the incident. However, PG&E's unrebutted description of corrective actions carries the same weight as its description of flaws, and there is no evidence that PG&E's corrective actions were inappropriate or ineffective. The Board finda them reasonable on its own review. We therefore conclude that with respect to the seismic clips the degraded condition has been corrected and the associated root cause has been identified. No f?aw in equipment or procedure is now known to exist on that subject. Therefore there is no basis for an adverse conclusion relative to PG&E's
1 l
- 125 -
surveillance program or to question the safety of future operations.
The Board rejects MFP's claim (MFP FOF $1 469-70) that prior corrective action developed for a similar event did not prevent this incident. The previous case dealt with missing clips on initial installation, whereas the one before us is concerned with reinstallation of clips after routine testing. MFP Exh. 98, at 11, 14; Tr. 1241-43 (Vosburg). The two cases have different causes and the previous corrective actions would not have prevented the event cited here.
The Board concludes that the raissing seismic clips incident was resolved and that the incident does not suggest a general programmatic deficiency in the PG&E maintenance and surveillance program.
- 29. Containment Fan Coolina Unit (CFCU) Backdraft Dameers.
MFP asserts that failures associated with Licensee's corrective maintenance of Containment Fan Cooler Unit (CFCU) backdraft dampers over a period of time is evidence of general deficiency in the Licensee's surveillance and maintenance program (outline ij I.B, II.A, II.B, II.C, III, IV.A, IV.C, IV.G). The last of these items (outline i IV.G, i
MFP FOF 15 60-61) refers to financial matters involving a l ratepayer settlement between PG&E and the State of California and is outside the scope of Contention I.
- 126 -
PG&E and the NRC Staff acknowledge that performance failures in corrective maintenance of backdraft dampers occurred in both units at Diablo Canyon. PG&E Test, at 88-89; Tr. 1261-62 (Giffin); MFP Exh. 100; Staff Test., ff. Tr.
2159, at 8-9. The Applicant and Staff claim that the failures were limited to difficulties in correcting backdraft damper problems and were not indicative of a general breakdown in the PG&E surveillance and maintenance program. PG&E FOF 1 M-A144; NRC Staff F0F i I-217. The Board must decide if the backdraft damper problems outlined herein were of limited scope and significance or if they contribute to a conclusion of general breakdown in the licensee's surveillance and maintenance program.
Each of the DCPP units has five containment Fan Cooler Units (CFCUs) within the containment to provide ventilation and cooling during normal operations and during accident conditions. Each CFCU has a backdraft damper downstream of the fan. The backdraft damper is designed to close on reverse air flow and prevent non-operating fans from rotating in & reverse direction. PG&E Test., at 88 (Giffin).
When a fan is not running, reverse rotation from air flow in the discharge duct is an indicator that the backdraft damper is not closed completely. Reverse rotation of a CFCU fan is significant because at sufficient reverse speed the fan may trip on overload upon receiving a start
I 1
l
- 127 -
signal in the event of a LOCA. Containment overpressurization during a LOCA is a possible consequence in certain analytical scenarios involving three non-operational CFCUs with stuck-open backdraft dampers. MFP Exh. 102, at 8-9.
The Applicant has a history of finding reverse CFCU fan rotation in one or both units during inspections dating back to 1986. Corrective maintenance taken between 1986 and 1991 on failed backdraft dampers included replacement of parts, design changes in damper linkages, and addition of springs to assist damper closure. Nevertheless reverse rotation was observed in a Unit 1 CFCU during an inspection in March 1991. PG&E decided that no corrective action was needed.
In January 1992, two backdraft dampers were found in Unit 2 with counterweights that had fallen off. Other counterweights were found to be too loose.
Inspection of counterweights in Unit 1 determined that they were attached but some were installed too tightly. The Applicant concluded by analysis that this would not affect performance of the CFCU safety function. However, three CFCU backdraft dampers were found stuck open in a February 19, 1992 inspection of Unit 1. It was also found i
that some damper linkage bars had previously been installed incorrectly on two of the dampers, and that there were l l
degraded bolting problems on some dampers. The Applicant 1 i
concluded and reported to NRC that three Unit 1 backdraft i
1
- 128 -
dampers were inoperable, that the condition was outside the design basis of the plant, that it had entered the applicable technical specification, and that the condition ,
had potentially existed since March 1991. Operability of all three CFCUs was restored by February 26, 1992. MFP Exh.
102, at 4; MFP Exh. 103, at 5. The Applicant subsequently determined in consultation with Westinghouse that the stuck-open dampers did not render the CFCUs inoperable and that the plant had continuously met its technical specification requirements for the number of operable CFCUs between March 1991 and March 1992, when the correct design configuration was restored. MFp Exh. 103, at 9.
The Applicant reported to NRC in early April 1992 that all five Unit 1 backdraft dampers and one in Unit 2 had been overhauled and inspected. All Unit 2 dampers had been inspected and found to be correctly assembled.
Subsequently, in April 1992, the Applicant found reverse rotation in a Unit 2 fan, counterweights installed too tightly in Unit 2 dampers and washers installed in Unit 2 dampers contrary to approved design. Corrective work was completed on these problems near the end of April 1992. MFP Exh. 102, at 2-7.
The Applicant reported to NRC that the root cause of its problems with the backdraft dampers was failure to l
perform proper maintenance. Contributing causes were found to involve management underestimation of safety l
1
4 1
1
- 129 - J significance, poor planning of damper work, inadequate work l instructions, inadequate job turnover, no Quality Control involvement, inadequate Plant System and System Design engineer involvement, inadequate post maintenance testing, and missed opportunities from prior problems. MFP Exh. 103, at 8.
NRC issued an NOV to PG&E in June 1992, listing three Severity Level IV violations related to the CFCU backdraft damper problems. The violations were: A. Work order instructions were not implemented; B. Inspection of Unit 2 CFCUs done without appropriate procedures; C. Failure to correct reverse rotation in CFCU 1-5 from March 27, 1991 to February 22, 1992. NRC accepted PG&E's analysis of operability in the matter of the three Unit 1 CFCUs with stuck-open backdraft dampers and did not cite the company for violation of the applicable technical specification.
Later during the period September 25, 1992 to November 13, 1992, the Applicant found cracked backdraft damper blades in both Units. It determined that this was a condition potentially outside the design basis of the plant.
None of the blades failed in normal service. However, PG&E concluded that longitudinal cracks were of sufficient length to result in blade failure if challenged by the postulated design basis LOCA pressure wave. The Applicant reported in its LER of November 17, 1992 that it had replaced the blades in Unit 1 with blades made of fatigue-resistant material and
- 130 -
had plans for Unit 2 blade replacement at the next scheduled outage. MFP Exh. 101. All blade replacement has now been completed. PG&E Test., at 89 (Giffin). 3 The Board finds that the matter of cracked backdraft damper blades was resolved promptly and effectively.. There is no record of improper maintenance. This matter is unrelated to the issues previously discussed in this section. The Board concludes that this episode does not contribute to a concern for possible programmatic deficiency in the Applicant's surveillance and maintenance program. ,
Tr. 1254-55 (Giffin).
The Board concludes that the problems cited in the Staff IR, the Staff NOV and in Staff testimony show that there were deficiencies in PG&E management and in the performance of the Applicant's system engineers, maintenance engineers, corporate engineers and maintenance personnel regarding the repair and maintenance of backdraft dampers.
However, there is no evidence that any of these deficiencies are related to improper financial considerations, as alleged ,
by MFP. The Staff concluded that there was inadequate implementation of basic engineering instincts. Tr. 2210-11.
(Miller); NRC Exh. 2; MFP Exhs. 102, 140. PG&E was concerned about organizational deficiencies and later undertook a general inquiry into maintenance practices on the HVAC system. MFP Exh. 100; Tr. 1255-56 (Giffin).
i l
- 131 -
The Board concludes that repetition of this poor performance in other aspects of the Applicant's maintenance program would raise serious questions about. programmatic effectiveness. The Staff characterized PG&E's performance in this instance as "the biggest black mark in'the past few years." However, this performance has not been widely repeated in other aspects of the Applicant's maintenance program. The Staff evaluated the Applicant's performance in the CFCU problem in context with its overall performance in maintenance and found it to be an isolated event. The Staff gave a superior SALP rating to the maintenance program in full consideration of the CFCU problems. Tr. 2214-16 (Miller, Peterson, Narbut).
The Staff's testimony is credible and entitled to substantial weight. There is no credible contrary evidence.
The Board therefore concludes that the Applicant's poor performance in the CFCU problem was isolated and not repeated throughout the maintenance program. We find that ,
the CFCU problem at Diablo Canyon does not support an inference of general programmatic deficiency in the maintenance program and we reject MFP's FOF to the contrary.
However, we are including the interrelationship of engineering and maintenance, as reflected in the CFCU maintenance problems, as one of the series of examples giving rise to the need for a study to improve i
- 132 -
interdepartmental communications regarding maintenance l activities that we are directing PG&E to perform.
l
- 30. Control of Foreign Material / Cleanliness /
Housekeepina.
MFP asserts that a variety of incidents involving unattended debris and foreign materials left inside containment have occurred at both units of Diablo Canyon allegedly starting in 1985 and extending to December of 1992. According to MFP, problems involving debris and foreign material have safety significance, have occurred q repeatedly, and have not been effectively resolved by PG&E despite numerous opportunities to do so. It urges the Board to find that PG&E's control of foreign materials is inadequate and unacceptable.32 MFP also urges adoption of findings of general deficiency in the PG&E surveillance and l maintenance program (outline $$ I.A, II.A, II.B, IV.E).
There is no dispute by PG&E or the NRC Staff that the incidents of debris and foreign material in containment occurred as claimed by MFP. PG&E responds that the incidents cited by MFP are only loosely connected to one
MFP relies on the following exhibits consisting of NRC and PG&E regulatory documents in support of its assertions: MFP Exh. 105, NRC IR 92-31 (12/11/92); MFP Exh.
106, NRC Diablo Canyon Shutdown Risk and Outage Management Inspection, NRC IR 50-275/92-201 (12/8/92); MFP Exh. 107, NRC IR 88-10 and 88-11 (6/17/88); MFP Exh. 108, NRC NOV in IR 88-07 (5/5/88); MFP Exh. 109, NCR DC2-91-TN-N102 R2 (11/18/92); MFP Exh. 110, NCR DCO-91-MM-N042 (5/19/92); MFP Exh. 111, LER 2-91-012-00 (3/5/92); MFP Exh. 113, PG&E reply to NOV in NRC EA 89-241 (3/12/90) ; MFP Exh. 35, PG&E Self-Evaluation of DCNPP (7/93).
- 133 -
another and they collectively do not show a pervasive problem of foreign material or debris control in containment-, PG&E F0F 11 M-A145, M-A146. PG&E claims further that past problems with debris control have been corrected and current housekeeping in the plant is unassailable. PG&E FOF i M-A156. The NRC Staff claims that the events cited by MFP are few in number, unconnected with one another, and are not evidence of any general breakdown in the Applicant's surveillance and maintenance program.
Staff FOF 1 I-228.
The Board is called upon to decide whether the incidents cited by MFP of debris or foreign material periodically left in containment collectively indicate a programmatic deficiency in the PG&E surveillance and maintenance program.
A. Foreion Material Exclusion MFP Exhs. 107 and 108 document a series of violations involving loose tools in containment, untimely personnel actions to correct foreign material exclusion deficiencies, failure of corrective actions to prevent repetition, and loss of cleanliness controls. The violations occurred over a short period of time from March through May 1988. NRC issued NOVs citing PG&E with first a Severity Level V violation (MFP Exh. 108) and later a Severity Level IV violation (MFP Exh. 107).
i j
i
- 134 -
PG&E's replies to the NOVs are documented in PG&E Exhs.
25 (PG&E Letter No. DCL-88-150,. June 6, 1988) and 26 (PG&E I 1
Letter No. DCL-88-184, July 18, 1988). The Applicant outlined its corrective actions which included development of foreign material exclusion procedures, revision of administrative procedures to assure compliance with ,
cleanliness controls, revision of procedures to require Quality Control Department surveillance of housekeeping in ;
containment when the reactor vessel is open, and additional training for personnel. PG&E acknowledged that corrective actions stated in Exh. 108 (issued May 5, 1989) did not j prevent the loss of cleanliness control cited in Exh. 107 l (issued June 17, 1989).
MFP Exh. 110 is an NCR issued May 19, 1992 that analyzes programmatically several instances of loss of foreign material exclusion area (FMEA) controls of the type 1 described in MFP Exhs. 107 and 108. FMEA controls are for the purpose of preventing entry of loose parts into the ;
reactor coolant system during refueling or maintenance activities. Root cause for several instances of loss of FMEA controls was found to be due to management failure to implement the FMEA program as described in applicable administrative procedures. Contributing causes included
" lack of ownership" on FMEA jobs, inconsistent interpretation of requirements, procedures not user friendly, management expectations not communicated, and
- 135 -
insufficient FMEA boundary identification. The Technical ,
Review Group (TRG) conducted a thorough investigation and-recommended many corrective actions. There is no evidence that the corrective actions were ineffective and they appear reasonable to the Board. MFP Exh. 110, at 5-14.
MFP Exh. 113 (March 12, 1990) is PG&E's reply to an NRC NOV in which it acknowledges three violations that in the aggregate were categorized as a Severity Level III violation applicable to Unit 1 and 2 containment recirculation sumps.
One of the aggregated violations was for an inadequate inspection of the Unit 1 containment sump for loose debris.
Debris that could cause restriction of sump suction during a LOCA was found inside the upper grating assembly of the Unit 1 sump by an NRC inspector. The inadequate PG&E inspection was due to individual failure to implement a clear procedure. PG&E Test., at 106 (Crockett). PG&E took several corrective actions including initiating video probe inspection of sump pipes and valves for debris, revision of technical and administrative procedures to assure attention to recirculation sump cleanliness, implementation of foreign material exclusion requirements for sump activities, and establishment of foreign material exclusion requirements for sump suction piping during refueling outages. There is no evidence challenging the effectiveness of the corrective actions and the Board sees no reason to do so. This violation has not been repeated. Tr. 1508-09 (Crockett).
r
'r,
- 136 -
B. Unattended Material in Containment MFP Exh. 111 is an LER issued in March 1992 reporting unattended tools and debris found in Unit 2 containment in October 1991 and Applicant failure to meet the containment
- inspection requirements of TS 4.5.2.c after containment integrity was established. MPP Exh. 109 is an NCR issued November 18, 1992 documenting the debris control problems in Unit 2 containment cited above. The root cause for this event was determined 'o be " lack of a comprehensive program for control of material after containment integrity has been established." A contributing cause was lack of understanding of requirements by some individuals who failed to complete a required data sheet certifying that a visual j inspection had been performed and no loose debris were present in containment. To correct the problem the 1
Applicant established a comprehensive new program for control of material in containment and trained personnel in the procedures. MFP Exh. 109, at 9-12, 23-25.
MPP Exh. 105 is an NRC NOV issued December 11, 1992, l
citing PG&E with a Severity Level IV violation for corrective actions (described above for Unit 2) that were ineffective to prevent uncontrolled materials inside Unit 1 containment when containment integrity had been established.
Loose, unattended materials were found near the Unit 1 containment sump on November 5, 1992 after containment integrity had been established. The safety significance of
)
I l
l
- 137 -
this event was low and Applicant's proposed corrective-
. actions (offered immediately at the exit meeting) were accepted by the NRC Staff as adequate to correct ~the deficiency. MFP Exh. 105, at 6-8.
C. Other Material Control Issues MFP Exh. 106 is an NRC IR documenting a 1992 Staff shutdown risk and outage management inspection conducted in part by headquarters Staff. A deficiency was found in which a 3/8 inch instrument tubing disconnected for maintenance was not capped to prevent entry of foreign material. The Applicant had no procedures to control entry of foreign i material into disconnected instrument tubing. This was the only FMEA deficiency cited and the team found in general that the housekeeping and material control throughout the plant were a strength. MFP Exh. 106, at 20-21, A-3. PG&E added the missing requirement to its procedures for instrumentation. Tr. 1516-17 (Crockett). This was an isolated event of low safety significance that was corrected by minor procedural changes. It has no implication for the
- programmatic adequacy of Applicant's surveillance and t
. maintenance program. Tr. 2237-38 (Miller). The Board gives it no weight in its decision.
MFP Exh. 35 is a PG&E self-evaluatian of Diablo Canyon issued in July 1993. Deficiencies in the erformance of some supervisors with responsibilities for housekeeping and !
I programmatic deficiencies in implementation of housekeeping
I
- 138 - !
practices were found. MFP Exh. 35, MA.1-1, MA.2-1.
Deficiencies related to uncontrolled debris refer specifically to material found in the Turbine and Auxiliary ,
Buildings for which we have no evidence on safety significance.32 The report concludes that " minor housekeeping discrepancies remain high." The NRC Staff, however, finds the plant clean and in a general state of good housekeeping. Tr. 2239-40 (Miller, Narbut). The Staff inspector occasionally finds material on the floor but the plant staff takes care of it quickly. i We conclude that there is likely a continuing flurry of 1 minor housekeeping discrepancies at Diablo Canyon. This has no adverse implications for the surveillance and maintenance i
program because there is no evidence that the discrepancies cited in MFP Exh. 35 have safety significance. The overall housekeeping program produces generally good results. We j find the self-critical analyses displayed in MFP Exh. 35 to be a programmatic strength because they focus attention of i
plant personnel and supervisors on the continuing need to be alert for developing problems. Minor housekeeping -
deficiencies at DCPP are entitled to no weight in the 32 All previous evidence on the safety significance of debris relate to its significance in containment where sump screens could become clogged during an accident or foreign material could inadvertently enter the reactor coolant l system. There is no similar evidence in the record for the Turbine Building or Auxiliary Building. I i
l
- 139 -
licensing decision before us and we reject MFP F0F to the contrary.
D. Board Analysis The Board has considered the cluster of exhibits on material control in the light presented by MFP (i.e., that in the aggregate they would show a programmatic deficiency in material control in containment which in turn would contribute to an aggregate finding of pervasive deficiency in the plant surveillance and maintenance program) . Based on Staff testimony and the exhibits themselves, we conclude that the Applicant had programmatic level problems both in the control of debris in containment and implementation of FMEA requirements during the 1988 time frame and thereafter.
In both' areas, PG&E and Staff analyses found. programmatic !
deficiencies. There was repetitive occurrence of deficiencies after corrective actions had been taken. Root causes were related to adequacy of management, adequacy of instructions to personnel, adequacy of technical procedures, and comprehensiveness of corrective actions. The difficulties were corrected, however, and recent. inspections show little or no deficiency with material control in containment. Tr. 2236-40 (Miller, Narbut).
It does not aid the inquiry to further aggregate deficiencies related to FMEA issues with those related to i debris in containment absent evidence, which is lacking I
here, that deficiencies in different categories are
- 140 -
traceable to the same procedures, personnel or management.
We conclude that the Applicant and Staff are correct to distinguish deficiencies into separate categories because this is the way they are identified, analyzed and corrected in practice. Tr. 2235-36 (Miller).
In presenting its case through exhibits prepared by PG&E or the Staff, MFP ignored corrective actions that appear in the same or associated exhibits that are relied upon to show deficiencies. However, we do not read these l
documents selectively but consider both the deficiency and ,
the corrective actions they describe to be equally credible.
Therefore, in general, a deficiency such_as this one that has been resolved for the purpose of enforcement will not rise to the'importance of mandating denial of the license extension. We do not sit to sanction PG&E for past i
deficiencies or violations of NRC regulations.32 Our inquiry is related only to whether the plants are being sufficiently well maintained that they can be operated with reasonable assurance of safety in the requested license recapture periods. Thus, for MFP's case to be persuasive, 23 Commonwealth Edison Company (Braidwood Nuclear Power Station, Units 1 and 2), ALAB-890, 27 NRC 273, 278 (1988).
The Appeal Board found on the matter of possible violation of Commission regulations: "But this is not an enforcement ,
proceeding and the issue at hand is thus not whether a i sanction should be imposed against the utility because of I its asserted noncompliance with a Commission regulation. ,
Rather we are concerned in this licensing proceeding with whether the Licensing Board correctly authorized the ;
issuance of operating licenses. . . .
b
1
- 141 - ,
there must be evidence that there are current serious defects in the surveillance and maintenance program which the Applicant is either unwilling or unable to correct."
The record does not support that view. ,
The Applicant's and NRC Staff's testimony were persuasive that such defects or impediments to correction do not exist in matters relating to debris in containment and we find that the past safety-significant deficiencies cited by MFP have been adequately corrected. For all of the foregoing reasons'we reject MFP's FOF on debris issues in their entirety and find that PG&E has carried its burden of proof on the issues of debris and foreign material exclusion in containment.
- 31. Steam Generator Feedwater Nozzle Cracking.
A steam generator feedwater nozzle is a 20-inch i diameter piping connection through which feedwater flows ,
" Commonwealth Edison Company (Byron Nuclear Power Station, Units 1 and 2), ALAB-770, 19 NRC 1163, 1169 (1984). 4 The Appeal Board observed on-the matter of denial of en 6 application "such a result would depend for its validity upon a supported finding that it is not possible for the ascertained quality assurance failings either to be cured or. .
to be overcome to the extent necessary to reach an informed judgement that the facility has been properly constructed." 1 Here the issue is not whether the plant has been properly .i constructed but whether it has been properly maintained. We conclude, however, that with respect to the maintenance and surveillance program, the application before us could not be- ;
denied unless there was a supported finding that maintenance l deficiencies could not or would not be corrected. We may
+
and in fact are imposing conditions to correct certain aspects of the maintenance and surveillance program that we deem deficient. ;
i
)
- 142 -
into each steam generator. The nozzles and immediate upstream piping are susceptible to interior surface cracking ,
caused by thermal stress that occurs when on infrequent occasions cold water flows through a hot nozzle. Ultrasonic inspections of Unit 1 feedwater nozzle welds performed earlier than scheduled during a refueling outage.in September 1992 showed some surface cracking. A short piping section and the pipe to nozzle welds were replaced on all four Unit 1 steam generators. Later metallurgical j investigations showed that the repairs were unnecessary ,
because the cracks were actually smaller than originally thought and within the allowable range of the ASME code. ;
Calculations showed that the plant could have been operated at least to the end of another refueling cycle, when regular inspections were scheduled, without hazard from feedwater nozzle weld cracking. PG&E Test., at 91-92 (Crockett);
Tr. 1536-37, 1540-41, 1551 (Crockett). PG&E has now decided to perform non-destructive testing on main feedwater piping during each refueling outage rather than on the 10 year schedule allowed by the ASME code. Tr. 1552-53 (Giffin).
MFP Exh. 117 is a voluntary LER (Indications On The l
Main Feedwater Piping Near The Steam Generator Feedwater i i
Nozzles Due To Thermal Failure, 10/30/92) submitted to the NRC for information purposes that describes the steam generator feedwater nozzle cracking problem. As part of background, it discloses that Unit 1 feedwater nozzle
- 143 -
radiography performed in 1986 during the first refueling outage in response to NRC IE Bulletin 79-13 was incomplete.
It also discloses that some radiography techniques employed at Unit 1 in 1979, 1986 and 1987 may not have been in full compliance with Bulletin 79-13 requirements. Radiographs on pipes with cracks from the 1992 repairs suggest that the cracks would not have been detected by radiography. PG&E concluded that the errors had no safety significance.
MFP's complaint in this matter arises from its interpretation of the early radiography errors disclosed by PG&E in the LER (MFP Exh. 117). In MFP's view, the errors that occurred in 1986 and 1987 had safety significance because they were blatant, they remained uncorrected for a long time, and they could have led to serious safety risk had the rate of cracking been more rapid. This exemplifies a maintenance and surveillance program not functioning as it should, according to MFP. MFP FOF T 552.
According to MFP, most maintenance problems disable or threaten essential safety systems (outline i I.A); there is untimely detection and correction of aging effects (outline
$ II.C); and there is inadequate routine surveillance, tests and inspections (outline i IV.C) . The foregoing failures are claimed in turn to be contributing evidence of a generally inadequate surveillance and maintenance program at Diablo Canyon.
- 144 -
PG&E concludes that the discovery and resolution of the nozzle cracking problem is an example of the proper functioning of the DCPP maintenance and surveillance program. Tr. 1538 (Crockett); PG&E Test., at 92-93 (Crockett). According to PG&E, the NRC Staff thought that the PG&E analysis of the nozzle cracking problem was reasonable. Tr. 1556 (Crockett); see Staff FOF 1 I-231.
There is no disputed material fact related to the discovery and correction of feedwater nozzle weld cracking in 1992. However, MFP calls upon the Board to agree with its adverse opinion of the 1986 and 1987 errors in radiography. We decline to do so because the issue was not ventilated at hearing; NRC Bulletin 79-13 is not in the record; there is no record of regulatory obligations created for the Applicant by Bulletin 79-13; the safety significance of the errors is not self-evident, coming as they did early in the operating life of the plant; and PG&E's conclusion of no safety significance is uncontradicted. We find this issue, developed for the first time in MFP's proposed findings of fact, to be speculative.
It is undisputed, however, that in 1992 PG&E found and ;
i corrected small surface cracks in the Unit 1 steam generator i i
feedwater nozzle connection welds well before crack growth l l
could have exceeded code allowable dimensions. Tr. 1553-54 (Giffin). This was effective preventive maintenance.
Contrary to MFP proposed general findings of fact for this 1
4g 4 +AA- m
- 145 -
incident no essential safety-system was disabled or threatened; the detection and correction of the cracks was timely because it occurred before any code allowable was exceeded; the inspection leading to the discovery of cracks was not routine but proactive and it resulted in early detection.
The Board concludes that nothing in this incident undermines confidence in the surveillance and maintenance program at Diablo Canyon. The Board rejects MFP's contrary claims and finds that PG&E has proved its case on the steam generator feedwater nozzle weld cracking issue.
- 32. Procedural Controls Durino Shot Peenino .
Operations.
Three incidents of unanticipated spread of radioactive contamination and/or airborne radioactivity occurred during incpection and maintenance operations on steam generator hot and cold legs, one on September 25, 1992, one on September 26, 1992 and one on October 2, 1992. Circumstances in all i
three incidents were similar. In each case a cold leg manway door was opened while shot peening was being carried ,
out on the hot lag of the steam generator and eddy current testing was being carried out on the cold leg side. The hot leg was pressurized with dry air and the air from the shot peening operation, while the cold leg was supposed to be l
under negative pressure from an exhaust system using a high )
efficiency particulate air filter, but the open manway door I
l 1
. .- -, -- - l
- 146 -
provided a direct outlet for the contaminated air, bypassing the filter. The contamination incidents resulted in an NOV of Severity Level IV. MPP Exh. 118, at 1, 8.
Between the first and second incidents, between the second and third, and after the third, the Applicant took corrective actions in an attempt to forestall recurrence.
These included additional instructions for those performing the maintenance and inspection work, and the addition of a checklist to the procedure for the eddy current testing and further instructions for the shot peening procedure.
Finally, the air flow direction was reversed after the last incident, moving the dry air input to the cold leg and the filtered exhaust to the hot leg. PG&E Exh. 22, at 9-10.
MFP would have us conclude from this series of incidents that "these factors, taken together with the other deficiencies described [herein), indicate an inadequate maintenance and surveillance program at DCNPP." MFP FOF
$ 566, at 202. MFP categorizes these incidents as ones that disabled or threatened safety systems (outline i I.A), as a previous corrective action that failed to prevent recurrence (outline i II.B), as representing insufficient communication (outline i IV.A) and as an example of inadequate work l instructions (outline i IV.E).
Both the Staff and PG&E would have us find that "MFP Exhibit 118 can be accorded no weight." PG&E FOF $ M-A166, at A-70, Reply FOF $ R-A102, at A-50; Staff F0F $ I-237 at I
- 147 -
- 95. Both state that the document is "only peripherally" related to maintenance. PG&E FOF 1 M-A166, at A-70, Reply F0F 1 R-A104, at A-50; Staff FOF T I-237, at 96. The Staff and PG&E interpret the meaning of the terms maintenance and surveillance far too narrowly. These incidents were all the results of improper actions on the part of persons performing maintenance and surveillance activities. PG&E sought to correct all of them by improving the training of maintenance and surveillance personnel and by altering or adding to maintenance and surveillance procedures. To say that such incidents do not bear upon the maintenance and surveillance program is to stand logic on its head.
Although it is true that the primary purposes of the procedures, eddy current testing and shot peening, may have been accomplished, in our view doing the job includes the notion of doing it safely, without putting plant personnel or public at risk.
We do not, however, find that the incidents noted weigh strongly enough to condemn the entire program or even to call it to question. These were three incidents within the space of about a week in which a one-time operation was attempted for the first and only time. We do not find it surprising that errors of this degree were made, nor are we shocked to note that it took three tries to correct them. ,
i The evolution of these incidents reflects somewhat poorly l upon the maintenance and surveillance program, but not so l I
1
- 148 -
poorly as to condemn that program, and the careful analysis and prompt actions taken to correct the flaws weigh heavily in the entire program's favor.
- 33. Unclanned Activation of Enoineered Safety Features (ESF).
MFP submitted as exhibits a series of documents (MFP Exhs. 119, 120, 121, 122, 122A, 123, 124, 12 ti and 127) describing five incidents in which unplanned ESF actuations occurred. PG&E's analyses of these events and its reports of them to NRC found that personnel error lay at the root of each, as all three parties agree. PG&E FOF $$ MA-169 through MA-173, at A70-A72; Staff FOF $$ I-240 through I-244, at 96-97; MFP FOF $$ 573-77, at 205-07.
MFP would have us find that " [i] nadvertent ESF actumtions are significant occurrences" and that "the number of personnel errors involving unplanned ESF actuations reflects poorly upon the adequacy of the maintenance and surveillance program at DCNPP" (MFP FOF i 580, at 207, i 587, at 209). We cannot agree.
The number of surveillance and testing operations performed annually runs into the thousands. Tr. 834-35 (Crockett). The five instances noted in MFP's exhibits were !
indeed " benign": careful analysis in each case concluded i
that no threat existed to the public safety. Although it is l l
true that inadvertent actuations are to be avoided (Tr. 1576 (Vosburg)), they are scarcely the " absolutely horrible" (Tr.
[i
- 149 -
1578 (ZamEk)) matters that MFP, in cross-examination, tried to make them out to be. Indeed,.we agree with PG&E and the Staff that, generally speaking, ESF actuations move the plant to a more conservative condition. Tr. 1576 (Vosburg);
Staff FOF i I-239, at 96; PG&E FOF i 168, at A-70. For obvious reasons, tripping the plant is to be avoided, and any inadvertent ESF actuation causes undesirable reporting and paperwork. Tr. 1577-78 (Vosburg). But these are not inherently unsafe incidents and they do not appreciably contribute to the wear-out or breakdown of safety-related equipment. Tr. 1580 (Vosburg).
Having read MFP's exhibits in some detail, we are favorably impressed first with the apparent paucity of the mistakes compared to the total number of actions taken per year and second with the diligent and analytical dissection of each error that PG&E carries out, even whan the subject error is of little safety significance. We find the overall impression borne by this particular set of exhibits to be very favorable to the maintenance and surveillance program at DCPP.
- 34. Limitoraue Valve Failure.
MFP offered two documents concerning a single failure of a simitorque valve operator. These are respectively an NCR and an LER reporting on and analyzing the failure of a Limitorque valve operator during a test of that operator.
MFP Exhs. 128, 129; Tr. 1589-90 (Ortore).
1 1
l
- 150 -
Analysis indicated that the operator failed because a locknut and the setscrew intended to secure the locknut had not been tightened properly. The immediate situation was corrected, an investigation was performed to determine the reasons why the locknut was not properly secured, the INPO network was informed of the occurrence, crews were tailboarded on the importance of the proper attention to details on such valves, other similar valves were inspected for loose nuts and set screws, the hardness of the worm gear shafts in other valves was measured and it was found that this factor varied, suggesting that part of the cause was an inability of the set screw to properly hold the locknut, (indeed, contact with the manufacturer disclosed that the material of the worm shaft had been changed), and the procedure for assembly of the locknut was revised to specifically include instructions for tightening the setscrew. It was revealed in the course of the investigation that loose nuts were to be found only on valves worked on by one specific technician. MFP Exh. 128 cassim.
MFP would have us find that the DCPP maintenance and surveillance program is deficient because it did not catch this problem before it occurred by supplying instructions on the tightening of set screws, measuring the hardness of the worm shafts, and developing a " proper method for securing j the lock nut to the worm shaft." MFP F0F $ 594, at 211. We i
- 151 -
note that even in the state revealed by the test, any
~
suspect valve operators would have been-capable of carrying out their safety functions. MFP Exh. 129, at 7. We also note the diligence and completeness with which the investigation was pursued and the sensible steps taken to prevent recurrence. In our view, far from showing any fundamental flaw in the maintenance and surveillance program, this incident and its follow-up demonstrate the program's strength. This problem, however, represents another example that leads to our conclusion that PG&E should perform a study seeking improvement of interdepartmental communications for maintenance-related activities.
- 35. Motor Pinion Keys in Limitorgue Motor Operators.
MFP submitted an exhibit (MFP Exh. 132) concerning failed motor pinion keys in motor operated valves, and both direct and cross examination were carried out on witnesses familiar with the exhibit (Tr. 1615-25 (Ortore, Vosburg)).
MFP Exh. 132 is a voluntary LER submitted by PG&E concerning an incident in which a motor pinion key failed. The analysis did not show any safety problem, and the LER was submitted on a voluntary basis in order to keep NRC and the industry informed of potential mechanical failures. Indeed, other similar valves were found to be operable even with their motor pinion keys sheared, inasmuch as the friction between motor pinion and shaft is sufficient to operate the
- 152 -
valve without the key. MFP Exh. 132, at 3, 6, 7; Tr. 1616, 1618 (Ortore), 1621 (Vosburg).
The key sheared off during the incident because, through a miscommunication between the maintenance personnel and operating personnel, it was subjected to "short stroking" while in a manual mode of operation, an overstressing condition that it would generally not encounter in service. The vendor, in fact, believed that the key was of sufficient strength, but has now changed the design to specify a stronger material. MFP Exh. 132, at 3, 1.
MFP would have us find that the miscommunication whic?
revealed the low strength of these keys " reflects a pattern of miscommunications between maintenance and operations, which has caused other problems in these plants" and that the low strength of the keys and the failed keys in certain valves reflect a "significant number of safety defects in safety components that are found only through luck," and that "PG&E's inability to detect and correct these hidden defects in a timely way could have a significant adverse effect on safety. . .
" MFP F0F ji 604, 605, at 214.
In our view, the fact that the valves were still operable with the internal defect suggests that the hidden defects were not serious safety matters, and the thorough !
I l
review and analysis of the situation, once discovered, reflects very positively upon PG&E's willingness to analyze i
l 1
- 153 -
and improve its procedures when chance reveals hidden flaws, i however inconsequential.
Communications can, of course, always be improved between organizations sharing responsibility for plant maintenance and operation, and in this case we believe they have been. Surely the entire episode does not reflect any .
fundamental problems in maintenance and surveillance.
- 36. Control of Liftina and Riccina Devices.
MFP presented two instances that it alleged showed failure on the part of DCPP maintenance and surveillance
~
personnel to properly control and implement lifting'and :
rigging procedures. The first occurred on March 7, 1991.
It was an incident in which a crane was being used to lift a relief valve into position on a main steam line outside containment. MFP Exhs. 135, 136. The boom of the crane .
came too close to a 500 kV power line and the line arced to ground, causing a loss of offsite power. The 230 kV startup ,
system had been cleared for maintenance and was not ,
available. The emergency diesels started and loaded the vital busses, constituting an ESF actuation, and there was a momentary loss of residual heat removal. !
Refueling was in progress at the time and one assembly was in the manipulator crane and positioned over the core.
Several systems were affected by the loss of power, and not all equipment functioned as intended. For example, the auxiliary building ventilation fans could not be restarted
- 154 -
until certain components were replaced; emergency diesel EDG 1-1 started, but only after 19 seconds rather than the 10 seconds that is the tech spec limit; and the Unit 1 control room emergency lighting failed to function. MFP Exh. 135, at 5-8; MFP Exh. 136, at 5-9.
Safety analyses concluded that the health and safety of the public were not endangered by this occurrence. MFP Exh.
135, at 11; MFP Exh. 136, at 12. However, personnel safety was endangered by the arc to ground of a 500 kV line. Tr.
1635 (Giffin).
The root cause of the event was human error compounded ;
by ineffective use of existing management systems.
Apparently the crane crew did not even know that the 500 kV i
line was energized; they assumed that since the plant was !
shut down the line was dead; and they did not know it could be used co backfeed power into the plant. The foreman was apparently distracted by the many other activities he was coordinating at the same time. MFP Exh. 136, Att. 2 passim. ;
Although this incident seems to the Board to indicate a number of flaws in the maintenance and surveillance program, it does not seem focussed upon the sort of knowledge and experience expected of the rigging and lifting crafts. It scarcely takes an expert rigger to know that one should stay away from high voltage lines with crane booms. The problems seem again to center around communication (a knowledge of b
- 155 -
the state of the system one is working'on), training, and attention.
The second incident involved a radwaste container being prepared for shipping. On May 28, 1992, the primary and secondary cask lids were being installed on this container and a mechanical maintenance foreman was supervising the operation. He left the area. The lids had initially been lifted with three slings and placed on the cask. The clearance between'the lid and the cask was very close and the lid required some alignment. To facilitate this alignment, the crew decided to use two chainfalls and a sling in the operation. They mistakenly chose two one-ton chainfalls to lift a weight that was close to 2400 pounds for each chainfall, because they misjudged the weight. MFP Exh. 137, Att., at 4; PG&E Exh. 27, Enc. 1, at 1-2. The error became a subject of an NRC inspection report and resulted in a Severity Level IV Violation. MFP Exh. 137, Cover Letter, at 2.
The work of loading the cask was being carried out under Maintenance Procedure M-50.23, and was being supervised (at least at the start) by a mechanical maintenance foreman. When he left the scene, the rigging crew, in violation of rules and of the scope of the tailboard that had introduced the work, improvised in an area where their knowledge was insufficient. PG&E Exh. 27, Enc., at 2.
l
- 156 -
There appears to be a difference of opinion as to the significance of this incident with respect to personnel safety: the NRC inspector noted "that this was a personnel safety issue and that the individual standing on the cask lid could have been seriously hurt if the chain had parted" (MFP Exh. 137, at 4), whereas PG&E management opined that the incident "did not present a threat to personnel safety" since "the lid would have dropped a maximum of two inches and could not have slipped off the ]ip of the container...."
PG&E Exh. 27, Enc. 2, at 1. It seems to the Board that, whichever opinion one adopts, it is clear that an incident like this one could have personnel safety significance, and it is important that measures be in place to prevent such happenings.
MFP would have us find that "these two incidents share some pertinent characteristics, and thus they demonstrate a deficiency in PG&E's ability to control lifting and rigging devices for heavy loads." MFP FOF $ 633, at 227.
PG&E believes that the two incidents are unrelated, or at least that they do not "have any commonality." PG&E Exh.
27, Enc. 2, at 1: PG&E FOF $ M-A190, at A-77. Staff witness Miller said that the question whether there might be a common causal factor between the two incidents had not yet been decided (Tr. 2248). Both Staff and PG&E would have us find that the later of the two incidents, the chainfall l l error, was not related to maintenance (PG&E Reply F0F
- 157 -
1 R-Alli, at A-52; Staff FOF 1 I-258, at 103-104). Both cite our previous ruling on contentions (LBP-93-1, 37 NRC at
- 23) wherein we found that these and other incidents did not have a common focus with respect to a failure of the personnel and training programs. PG&E FOF i MA-191, at A-78; Tr. 1637-38.
Although the two incidents may not reflect in a common manner on the personnel and training programs, we do see certain parallels. Further, since the second incident was being carried out under a procedure that PG&E itself styles a " Maintenance Procedured under the direction of an employee titled " Mechanical Maintenance Foreman," we believe that both bear upon the adequacy of the maintenance program, and that both reflect poorly upon that program's ability to communicate both vertically (from supervisor to supervised) and horizontally (from itself to other organizational components).
Nevertheless, we do not see in these incidents any basic flaw of sufficient proportions to warrant denying the license amendment. These incidents, however, are being included 7.s part of the basis for our requirement that PG&E 1
conduct a study seeking improvement in interdepartmental communications regarding maintenance-related activities.
- 158 -
MFP points out several failures of the feedwater pump speed controllers. MFP FOF i 634-57, at 227-35; MFP Exhs.
138, 139, 140, 140A, 142. On March 6, 1992, a reactor trip occurred because of a low-low water level in the 1-3 steam generator. The low-low level was the result of a trip of the 1-1 feedwater pump. The inverter feeding the speed controller on that pump had failed, an automatic transfer to a second inverter also failed, and the loss of the speed control channel caused the pump to go to a maximum speed condition, which resulted in a pump trip. MFP Exh. 138, at 1-4.
PG&E's analysis of the event concluded that the root cause of the failure was that the original (pre-1989) design of the speed probe system was a single channel design and hence incompatible with the later-installed Lovejoy system that failed. The problem had been earlier identified in a letter from a technician, a letter that never received a response. MFP Exh. 138, at 10. The failure of the probes to transfer to the alternate power supply was caused by a small piece of insulating debris, which had fallen between l 1
the points of a relay contact. MPP Exh. 140A, at 4. In the course of investigating this event, PG&E discovered that the type of inverter whose failure had set off the sequence had 1
failed nine times between 1990 and 1992, each of those l l
l I
- 159 -
failures having resulted in repair or redesign in consultation with the manufacturer. MFP Exh. 138, at 2-4.
On April 23, 1991, a failure of main feedwater pump speed control had also occurred and had led to a turbine trip and a reactor trip. The root cause of that event was found to be failure of an amplifier in the Lovejoy control system. MFP Exh. 142, at 2-5.
The inverter difficulties all arose from a set of new inverters installed in November of 1989 in an effort to give the pump speed control a more reliable power supply.
Tr. 1651 (Giffin); MFP Exh. 138, at 2. Obviously the effort was less than fully successful.
MFP would have us find that inverter failure pas a "long standing problem" at DCPP, and that PG&E's actions were ineffective in preventing the problem's repeated recurrence. MPP F0F 1 643, at 232. Even PG&E concedes that
" [wj e waited too long and continued to try to fix (the problem) instead of just putting in a new design" and "we should have written off the power supply." Tr. 1652 (Giffin); MFP Exh. 138, at 18.
The Staff and PG&E do not see the problem as a maintenance and surveillance problem. The Staff says, ;
\
"MFP offers no connection between these findings and PG&E's l l
maintenance and surveillance program . . . . The issue of concern to PG&E and NRC related to the timeliness of PG&E's design engineering program efforts, not to the maintenance
- 160 -
program." Staff FOF i I-261, at 104-105; Tr. 2246-47 (Miller). PG&E says the sequence of events "did not reflect directly on either maintenance or surveillance." PG&E FOF i M-A195, at A-79; Tr. 1653 (Giffin).
Once again we believe that the Staff and PG&E read the scope of the contention too narrowly. If maintenance efforts fail or are untimely because of a failure to properly coordinate with the engineering department or because of a failure of engineering to properly support maintenance, that is in itself a failure of the maintenance program viewed in a holistic sense.
MFP would also have us find that " [f] inancial consideration influenced PG&E's corrective action." MFP FOF i 648, at 233; MFP Exh. 138, at 18. That may well be so.
There does not, however, seem to be any tendency to compromise safety for financial reasons. Indeed, both PG&E's technical witnesses and the technical witnesses for the NRC Staff characterize the pump speed controller failures as matters of little or no safety significance.
Tr. 1653 (Giffin); Tr. 2216 (Miller). We agree. It appears that the greatest significance for the main feedwater pump failure is the effect it has on operability and availability of the plant rather than on safety. Although PG&E's response to the failures may have been as " untimely and inefficient" as MFP would characterize it (MFP FOF i 650, at
l
- 161 -
233), there seems to have been no substantial effect on safety.
MFP is correct that any problem with a feedwater pump can introduce the possibility of a transient, and that transients are not desirable from either a safety or availability standpoint. But we accept-the professional opinions of the technical witnesses presented by the Staff and PG&E to the effect that the pumps are not safety-related ,
equipment. Tr. 1653 (Giffin); 2219 (Peterson).
We are again confronted with a situation in which the response of the maintenance and surveillance program as a i
whole has been less than perfect. The delay in fixing the inverters in particular demonstrated poor engineering support for the maintenance effort. But the matter is not i
one of substantial safety significance, and we cannot see it as reason to deny the license extension. We are, however, ;
including this matter as one of the bases for our !
requirement that PG&E perform a study seeking to improve !
i coordination of the maintenance depoartment with other departments (such as engineering) engaged in performing maintenance-type activities.
- 38. Inadvertent Containment Ventilation Isolation. !
MFP introduced a series of NCRs and LERs purporting to
.show instances of containment ventilation isolation (CVI) resulting from personnel error. MFP Exhs. 144, 145, 146, 146A, 147, 148, 149, 149A, 150, 150A, 151, 51A. In the
_ .. . . _ _ _ . . _ -. . ~ . .
- 162 -
views of the Staff and the Applicant (which coincide), these )
events are unrelated to one another except that they all led to CVI. Staff FOF 1 I-269, at 107; PG&E FOF 1 M-A202, at 1
A-82. Both Staff and Applicant point out that CVI is inherently a " benign" event, that is, simply a change to a more safe (conservative) condition, and that the wear and tear on safety devices resulting from CVI is minimal. Staff FOF 1 I-267, at 107; PG&E FOF i M-A199, at A-80; both citing Tr. 1670-71 (Vosburg).
MFP would have us find that these incidents show several major flaws in the DCPP maintenance program: ineffectiveness of attempted corrective actions, lack of communication and coordination between maintenance and other departments, financial considerations influencing safety decisions, failure to adhere to procedure and policy guidelines, and untimely response to deficiencies in the radiation monitoring system. MFP FOF $$ 663, 665, 670, 674, 680, 684.
The Staff and Applicant would have us find that, there being no real nexus between t.hese events, the charge of ineffective corrective actions is not warranted. PG&E FOF
$$ M-A202, M-A203, at A-82; Staff FOF $$ I-270, I-271, at 108. Both of these parties point out that recent regulatory 1
changes have been made in reporting requirements to downplay the importance to safety of CVI events. PG&E F0F $ M-A199, at A-81; Staff FOF $ I-267, at 107; both citing Tr. 1668-69 l (Vosburg) Both the Staff and Applicant also point out that
- 163 -
the radiation monitoring system is in the process of being upgraded to a digital system that should be less sensitive to electrical ncise in the plant. PG&E FOF i M-A200, at A-81; Staff FOF $ I-269, at 108, both citing Tr. 1673-74 (Giffin).
Neither the Staff nor PG&E addresses the matter of financial considerations, but we find that the portion of the record cited by MFP (MFP Exh. 148, at 7, 15)--when taken in context--does not suggest that financial considerations played any strong role in the decisions made on methods to address the CVI problem at hand.
We do, however, find two matters troubling. The first is the fact that " [t] here is no plant or GC procedure specifically regarding work on energized equipment" (MFP Exh. 150A, at 12). This lack of procedures is particularly serious inasmuch as four of the CVI events resulted from errors on the part of personnel who were working on energized equipment. These numerous errors warrant a specific list of precautions for affected personnel. We do not accept PG&E's argument that these precautions are
" common knowledge to journeymen electricians" (id.) since, if this were the case, there would not have been the four CVI events.
The second matter is the timeliness of PG&E's response to the oversensitivity of the radiation monitoring system to electrical noise. The system has exhibited this l
l
, ,_ - _ . _ = _ _ ._
- 164 -
characteristic "since the Units have been operating" (MFP Exh. 149A, at 5), yet PG&E is.only "in the process of ,
installing" a new system, and while the specific monitors ,
involved in the incidents listed by MFP.no longer give trouble, other instruments in the system still do. Tr.
1673-74 (Giffin). Despite the avowed benignity of a CVI incident, as we noted in Section 33, suora, unneeded ,
actuation of ESF should be avoided. It stresses the crew and diverts resources. Tr. 1577-78 (Vosburg). The Board feels that it is high time the steps (apparently well known) that are needed to further suppress this undesirable ESF actuation should be taken.
Accordingly, we. direct that the conversion of the radiation monitoring system be completed and that a set of rules for working on energized equipment be promulgated.
- 39. Reactor Trio on Steam Generator Low Level.
Two exhibits offered by MFP (MFP Exhs. 155, 156) were the NCR and LER concerning a react';t trip occasioned by low steam generator level that occurred on February 1, 1991. A carpenter was carrying planks intending to erect a scaffold for inspection and repair of a feedwater valve. The planks struck an instrument air valve, closing it and shutting off the operating air supply to two feedwater valves and their bypasses. The feedwater valves failed closed, the feedwater l
supply to two steam generators ceased, the steam generator level dropped, the reactor tripped on low steam generator
. _ ~ - . ~ - - - . _ . .-
- 165 -
level and steam flow /feedwater flow mismatch and the. turbine tripped. .MFP Exhs. 155, at 2-3, 156, at.1; Tr. 1692-93 ,
(Giffin).
When the reactor trip and turbine trip occurred and the ,
plant transferred from its own generated power to outside power, four pieces of equipment failed to operate: a circulating water pump failed to restart, a 25 kV motor-operated disconnect failed to open, a control rod drive mechanism cooling fan failed to start, and a main turbine stop valve failed to close fully. MFP Exhs. 155, at 3-4, 156, at 3-5; Tr. 1694-95 (Giffin). PG&E's safety analysis of the event indicated that such a string of failures was, i
in fact, bounded by previous analyses and that the health and safety of the public was not adversely affected. MFP d Exhs. 155, at 8, 156, at 9-10. The cause of each of the four subsidiary failures was identified: the circulating -
water pump failed because of a malfunctioning relay (Tr. ,
1699 (Vosburg)), the 25 kV disconnect failed because outage workers had wrapped plastic sheeting around one of its ,
operating shafts, the control rod drive cooling fan failed because of temperature and age-induced changes in its magnetic starter, and the main turbine stop valve failed to ' ;
close because of under-lubrication of a bushing on its actuator spring. MFP Exhs. 155, at 3-4, 156, at 4-5.
Corrective actions were taken to prevent recurrence of the initiating event, including a revision of the procedure .
- 166 - 1 l
l governing erection of scaffolding. Corrective actions were :
1 also taken to forestall recurrence of the subsidiary events. 1 MPP Exh. 156, at 11-12; Tr. 1693, 1702 (Giffin); Tr. 1699, 1701 (Vosburg).
MFP would have us find that this incident represents another case in which multiple maintenance deficiencies occurred and previous corrective actions were ineffective (MPP FOF 11 693, 702, at 248, 251). It would also have us find that the incident presents " substantial and multiple inadequacies in PG&E's maintenance and surveillance program" (MFP F0F i 704, at 252). PG&E would have us find that this was an isolated event, that it has been adequately addressed to minimize future occurrences, and that it is an example of how a "living" maintenance program incorporates operating experience in order to improve. PG&E FOF $1 M-A209-211, at A-84-85, citing Tr. 1694-96 (Giffin), 1700 (Vosburg). PG&E sees no programmatic concerns stemming from this incident.
PG&E Reply FOF i R-A120, at A-55-56. The Staff, too, sees this as an isolated incident that has been properly resolved and that has actually resulted in enhancing the maintenance picgram. Staff FOF $ I-275, at 110.
In this instance we agree with PG&E and the Staff. We see this as an incident in which a virtually unforeseeable random event with no serious safety implications has resulted in a thorough and intensive analysis, ultimately improving the maintenance and surveillance program. '
- 167 -
- 40. Auxiliary Saltwater Pumo Crosstie Valve.
MFP presented an exhibit (MFP Exh. 168) in which was described an incident where an auxiliary saltwater pump crosstie valve was found with its manual handwheel inoperable because of extensive rust buildup. MFP points out an ostensible inconsistency in the testimony of PG&E witnesses concerning the safety status of this valve and its capacity for manual operation (MFP FOF i 705, at 253). It contrasts the statements of witnesses Giffin and Ortore (Tr.
1725) to the effect that the function of closing the valve is not safety significant with that of witness Vosburg (Tr.
1718) that the closing of the crosstie valve is a cafety function.
We see no real inconsistency. Witness Vosburg merely testified that the design of the system takes account of the manual operation capability of the valve in order to make it unnecessary to design the remote operator to Class I (safety grade). Witnesses Ortore and Giffin noted that the entire capacity of the valve to be operated--both electrical and i manual--is itself Class II because operation of the valve is not required for mitigation of an accident. (The body of 1 l
the valve is Class I since it is part of the primary pressure boundary.)
MFP's own exhibit shows that the operating capability is only needed if, at a time some thirteen hours after an
- accident, it becomes necessary to separate two redundant I
- 168 -
trains of auxiliary saltwater cooling in the event that a leak develops in a passive component of one of the two trains (MFP Exh. 168, at 7-8), and the manual operator would be needed only if the electrical operator simultaneously failed. A second valve in series with the valve whose manual operator failed was available in any event, and although that valve's manual operator was clogged with paint it was quickly and readily freed for service. Id. at 3-4, 15-16. The maintenance program has been changed to require inspection of these manual handwheel operators. Id. at 18-19; Tr. 1730 (Giffin).
MFP would have us find that the failure to include regular inspection of the handwheels in the original maintenance and surveillance program " reflects a basic inadequacy in the program" (MPP FOF 1 706, at 254). Because MFP's own exhibit concludes that, even with the valves in the condition found, "the ASW system would have functioned as designed to support post-accident operation" (MFP Exh.
168, at 8), we cannot agree with MFP on this point. Clearly the analysis and review of the incident has improved the program, but clearly also it was adequate before the incident occurred.
MFP makes a second point: that the response to the original discovery of the stuck handwheel was intolerably tardy (MFP FOF $$ 715, at 257, 722, at 259). This is because, while the flaw was discovered in June of 1990 and
)
i l
j l
- 169 -
I an attempt was made to correct it in September of 1990, actual repair was not accomplished until the NRC resident ,
inspector called the outdated repair tag to the plant j manager's attention in January of 1991. It appears that the first attempt to correct the situation was abandoned because-plant management knew that spare parts were not available ,
and was reluctant to disassemble the valve without such parts. MFP Exh. 168, at 2-3. We have no record evidence of !
the time needed to get spare parts for such valves, so we P
cannot judge whether the period was too long or not.
Certainly the fact that the valve was ultimately repaired without spare parts suggests that a sufficient effort could have fixed it more quickly. Nevertheless, we see no serious programmatic fault in PG&E's behavior in this instance. The matter was a flaw in a system that was not directly safety-a related, there existed adequate redundancy during the period before repair, and PG&E has taken adequate steps to prevent ,
recurrence.
- 41. Testcock Valve on Diesel Generator. ;
MFP introduced an exhibit (MFP Exh. 172) that concerned ;
the failure of a testcock valve on a diesel generator. This !
I exhibit describes the following occurrence: Maintenance was ;
being performed on a diesel generator, and a testcock, a device to permit checking cylinder pressurer, had been replaced. Mechanical Maintenance performed an initial leak check on the testcock, and the engine was run to bring it to l 1
- 170 -
operating temperature, when another leak check was to be performed. When Mechanical Maintenance tried to perform this second check, the mechanic inadvertently checked and tried to tighten a testcock other than the one that had been replaced, and that testcock broke off. Analysis and investigation revealed that the broken testcock had failed because of high cycle vibration fatigue caused by loosening in service. Id. at 2-3, 5; Tr. 1746-47 (Giffin).
MFP would have us find that the degradation of the broken testcock was identified because of an error in post-maintenance testing (MFP FOF i 726, at 260), and that seems true enough. But MFP would also have us find that, when taken in conjunction with the incident of the failed motor operator keys described above, this, being a second instance wherein a flaw was discovered when a maintenance worker made an error, indicates some basic infirmity in the maintenance program. MFP F0F i 727. That we cannot find. Both the Staff and PG&E would have us find that this is a very minor incident with no real bearing on the fundamental soundness of the PG&E maintenance and surveillance program. Staff FOF
$ I-286, at 113; PG&E FOF 1 M-A219, at A-87. We note that, even had the damaged testcock gone undetected and broken off while the diesel was running, it would not have caused the diesel to fail its intended function. Tr. 1749-50 (Giffin).
i P
- 171 -
This, even taken with the matter of the failed motor operator keys, is surely no indicator of any significant !
flaw.
-42. Main Feedwater Check Valve.
Four of MPP's exhibits (MPP Exhs. 190, 191, 192, 193) allegedly relate to the malfunction of a main feedwater j check valve at Unit 1. Although it appears to the Board that three of the four documents (MFP Exhs. 191, 192, 193) ;
involve that valve only peripherally, we are considering them together in order to correspond to the treatments given ;
in the proposed findings of the parties.
The first document reports the finding of a condition !
of leakage in a main feedwater check valve and the considered decision, based upon evaluation of the safety significance of this leakage and upon the fact that .
operators could readily take account of the condition of the valve, that there was no need to repair the leakage immediately. The fundamental cause of this leakage was a failure on the part of the valve's vendor to make clear the {
proper procedure for reassembling the valve after servicing.
That problem has been corrected. MFP Exh. 190, at 2, 5-9. i The subject matter of the other three exhibits (MFP Exhs. 191, 192, 193) was PG&E's treatment of, and response to, an actuation of the P-14 ESF at Unit 1. That actuation l i
occurred at a time when the feedwater check valve was still leaking, and it was at first thought that that condition f
- 172 -
might have contributed to the event, but later analysis showed that the check valve leakage did not really contribute. MPP Exh. 190, at 3; Tr. 1781 (Vosburg). The ESF actuation was primarily brought about by leakage through the main feedwater control valve and its bypass valve, both of which malfunctioned because of a drift in the valve position controller. MFP Exh. 191, at 5. That condition has been corrected by increasing the frequency of surveillance of the position controller's condition. Tr.
1783-84 (Giffin); 1786 (Vosburg).
The P-14 interlock trips the main feedwater pump and the turbine and closes the main feedwater isolation and regulating valves in order to protect the turbine from damage by water intrusion. MFP Exh. 192, at 4-5; Tr. 1773 (Vosburg).
MFP would have us find that inadequate maintenance was performed on the main feedwater check valve during Unit l's third refueling outage (MFP FOF i 735, at 263). That is apparently correct. As we have noted above, the material supplied by the vendor of the valve apparently did not contain adequate instructions for its proper assembly after servicing. That matter was corrected, and it apparently did not lead to any serious consequences. Further, careful safety analyses showed the leakage would not lead to any serious safety problems. MFP Exh. 190, at 6-8. As to MFP's proposed finding that PG&E's response to the leakage was
- 173 -
untimely (MFP FOF i 738, at 264), we see no reason to fault the management decision to leave the leakage uncorrected.
MFP would also have us find that poor communication played an important part in the series of events herein described, inasmuch as MFP believes that no proper notification was given to the reactor operating staff that leaky valves might occasion trouble during a startup. MFP F0F $ 743, at 265, citing MFP Exh. 193, at 1. The NRC comment that MFP cites in support of this postulate is, in fact, apparently in part a simple misunderstanding of the role the check valve played in the P-14 actuation incident.
Although we do not have the original report of that incident sent to NRC, the revision introduced shows, as we noted above, that the leaking check valve played little part in the ESF actuation. The valves primarily to blame for the ESF actuation were not known to be faulty at the time. No serious lack of communication between maintenance and operations is apparent to the Board here.
MFP repeats that untimely action after the discovery of the leaking check valve was to blame for " numerous reactor trips, an ESF, and a degraded feedwater system." MFP FOF ji 746-750, at 266-67. We see no instances here recorded where leaking valves of any sort led to a reactor trip--the ESF occurred during recovery from a trip due to other causes (Tr. 1781 (Vosburg))--nor do we see a substantially degraded
- 174 -
feedwater system. We see no evidence here of a significant deficiency in the maintenance and surveillance program.
- 43. ASW Pumn Vault Drain Check Valves.
MFP introduced an exhibit (MFP Exh. 196) that describes an incident in which the check valves in both Auxiliary Salt Water (ASW) pump vault drains were simultaneously removed for servicing. Because both chains of a redundant system were being worked on at the same time, PG&E witness Giffin, out of an abundance of caution, initiated the NCR process, but engineering analysis showed that the removal of the valves did not affect the operability of the ASW pumps and that the incident did not, in fact, present a nonconformance. Tr. 1795-96 (Giffin). Thus the document is not really an NCR, and, indeed, it contains within it the information that the incident was neither nonconforming nor reportable (MFP Exh. 196, at 6-7, 12).
Nonetheless, MFP would have us consider certain findings in the report as matters that reflect adversely upon the maintenance and surveillance program of the DCPP.
MFP FOF i 754, at 268, citing MFP Exh. 196, at 5-6; MFP FOF
$ 755, at 269, citing MFP Exh. 196, at 11. Both the Staff and PG&E would have us find that, inasmuch as the report does not suggest any hazard to the public health and safety (indeed, it specifically discounts any such hazard (MFP Exh.
196, at 6)), and because it is not actually an NCR, it
- 175 -
l l
cannot lend any support to MFP's contention that the maintenance and surveillance program is flawed. PG&E F0F i M-A233, at A-92-93; Staff FOF $ I-296-97, at 116-17.
Although we cannot endorse the reasoning of PG&E and the Staff in this--clearly, a recounting of an egregious failure of the maintenance and surveillance program, whether in an NCR or elsewhere, whether it directly hazarded the public health and welfare or not, could cast doubt on that program--we find, after carefully considering the matters MPP has pointed out, that none of them rises to the level of a significant challenge to the program.
- 44. Motor Operated Valve Failed to Cycle on Actuation Sional.
MFP introduced an exhibit (MFP Exh. 210) concerning a motor-operated safety injection valve that failed to cycle closed and open when tested. The valve opened properly on a signal to do so, then closed on a signal to close, but failed to open a second time on a second attempt to open.
Investigation revealed that the operator's declutch fork had been installed upside-down during maintenance eight years previously, prior to the start of plant operation, and that the excessive stresses induced by this improper installation ultimately caused the failure. Id. at 2-3. MFP would have us find, by a tortuous chain of reasoning interpreting 10 C.F.R. Part 50, App. A, that this chain of events violated the so-called " single failure criterion" (MFP FOF $ 762, i
l l
ny
- 176 -
763, at 271-72), and it would have us find this despite the fact that the valve in its as-found condition would have operated in an emergency. MFP Exh. 210, at 5; Tr. 1809-10 (Ortore).
We note that, at the time the erroneous installation of the clutch fork occurred, the maintenance instructions and training were not equivalent to present standards. MFP Exh.
210, at 6; Tr. 1810 (Ortore).
MFP would further have us find that this is a case wherein a " deficient installation went undetected until the component failed . . . .
MFP FOF j 764, at 272. We note that in fact the failure was detected by a routine surveillance intended to detect just such failures, and that during the eight-year period the valve was regularly tested and performed as it should have. Tr. 1810 (Ortore).
We see this occurrence as an isolated instance that has been dealt with properly, an instance that does not suggest a significant flaw in the main".enance and surveillance program.
- 45. Fire in Electrical Pane.'.
MFP introduce an exhibit describing the occurrence of a fire in an electrical panel. The fire was discovered both through the activation of a smoke detector annunciator and by a mechanic's helper who saw smoke and spread the alarm.
The fire brigade responded and soon extinguished the fire.
The damage was such that the exact cause of the fire could
- 177 -
not be determined, but analysis by knowledgeable people led to the conclusion that the fire resulted from overheating of a loose terminal on a 480v breaker. It was further believed that the terminal was loose because the termination was of the compression type, a type that is dependent upon the skill of the installer for acceptable results. MFP Exh.
216, at 1, 2, 4, 6.
The exhibit notes that such terminations were replaced in the 1970s on Class I high voltage equipment, but that a
" conscious, economic decision" was made not to replace them on Class II equipment, a condition that accords with accepted IEEE practice. MFP Exh. 216, at 5.
MFP would have us find that PG&E's behavior in this instance is a case of putting financial considerations above safety and hence shows that financial considerations have undermined the maintenance and surveillance program at DCPP.
MFP FOF ij 767-68, at 273-74. It does not seem so to us.
The financial consideration led only to economizing on non- ;
1 safety matters. Indeed, the panel that failed "has no l I
effect on the safe shutdown capability of either unit." MFP -
Exh. 216, at 6. Improved terminations have been installed and a system of infrared thermography inspections has been l
instituted to reduce the probability of recurrence. Id. at 7; Tr. 1823-24 (Crockett). We see no basic flaw here.
l i
- 178 -
1
- 46. Chemical and Volume Control System Diaphraam Leakace.
MFP cross-examined PG&E's witnesses on certain instances of leakage from valves in the Chemical and Volume Control System (CVCS) of the plants. Tr. 1826-39; PG&E Test., at 100-02. To clarify matters addressed in that testimony and cross examination, PG&E introduced two exhibits (PG&E Exhs. 28, 29). Although the technical specification for leakage from the CVCS was exceeded in each case, careful analysis concluded that there was no effect on the public health and safety, inasmuch as the leakage limits are set by consideration of the doses in the control room and at the site boundary over the entire 30-day course of a very severe accident. PG&E Exhs. 28, at 5-6, 29, at 4-5.
These two instances of leakage occurred in 1991 and 1992, but they stemmed from different root causes and the precautions taken to prevent repetition of the first occurrence would not have prevented the second. Tr. 1832-33 (Giffin).
MFP would have us find that multiple deficiencies led to these incidents and that "this broad array of deficiencies implicates the overall adequacy of PG&E's maintenance and surveillance program." MFP FOF $ 784. We l do not agree. The incidents of leakage stemmed from different causes, those causes were in each instance carefully analyzed, and proper steps were taken to improve i
~
i l
l i
1
- 179 - l the program. 'PG&E Exhs. 28, at 6, 29, at 5; Tr. 1831-33 ]
(Giffin). We see no indication in these-events thac the maintenance and surveillance program is seriously flawed.
- 47. Conclusion on Maintenance and Surveillance Procram.
We began our discussion of the Maintenance and Surveillance Program with the general conclusion of the Staff witnesses that the program was adequate and, indeed superior. We then tested this overall general conclusion against the incidents cited by MFP.
There is additional testimony and other evidence dealing with a general evaluation of the Maintenance and Surveillance program and how it compares to other such programs throughout the industry. PG&E performs approximately 14,000 preventive maintenance tasks and 7,000 corrective maintenance tasks annually at DCPP. PG&E Test.,
at 38, 40 (Ortore). The incidents referenced by MFP, some of which (as detailed above) are quite serious, represent but a small percentage of those tasks. PG&E claims, and we have no reason to doubt, that the nonconformances, although typical of the range of matters that confront the maintenance and surveillance system, are "not the normal.
They're the exception of how we do business." Tr. 2072 )
(Giffin).
- 180 -
Collectively the incidents demonstrate-that PG&E has l
not reached perfection--but no one (including MFP) expects that it would or could do so. As a-PG&E witness observed, "we are not perfect. We would like to be, but we're not."
Tr. 2071 (Giffin). ;
Clearly perfection--in context, " error free maintenance"--is not required to provide the " reasonable assurance" necessary for us to approve the license extensions here sought. Tr. 2275 (Peterson). Compared with nuclear power plant industry norms, however, PG&E ranks quite favorably. The NRC has frequently commended PG&E for achieving a high level of safety performance at the DCPP.
PG&E Exh. 19 (NRC commendation letters dated June 22, 1993, February 5, 1993, June 30, 1992 and February 3, 1993).
Moreover, the Commission periodically conducts a Systematic Assessment of Licensee Performance- (SALP) - which, in effect, evaluates the performance of operating reactors ,
in specified disciplines, one of which is maintenance and surveillance. PG&E Test., at 182-83 (Giffin). Ratings are
- currently categorized into three levels--1 (" superior level of safety performance"), 2 (" good level of safety performance") and 3 (" acceptable level of safety performance") . "
"NRC Management Directives System, Handbook 8.6, Part I, approved July 14, 1993, at 6 (replacing former NRC Manual Chapter and Appendix 0516), referenced in PG&E Test., at 182-86 (Giffin). A fourth category ("N") represents
i t
- 181 -
For the most-recent SALP review period as of the time .
of the hearing-(from July 1, 1991 through December 31, t
1992), PG&E received six "1" ratings and'one "2 and improving" rating. Maintenance and surveillance received a- !
"1" rating. PG&E Test., at 183 (Giffin); PG&E Exh. 20. A plant need not receive a "1" rating to qualify for a license extension of the type sought here--indeed, a "3" rating would be sufficient. Tr. 2275-76 (Peterson). <
In addition, PG&E presented the opinion of Tedd A. >
Dillard, a maintenance expert employed as Supervisor of Component Programs for the Nuclear Division of Florida Power
& Light Co. (FP&L) and, previously, from May 1983 to November 1988, the Manager of Maintenance for FP&L's St. ,
Lucie Nuclear Power Plant. PG&E Test., at 116. Mr. Dillard testified (as had other PG&E witnesses) that excellence in operating records of a plant directly stems from an adequate maintenance and surveillance program and that PG&E's record in this respect is among the leaders in the nuclear industry. The Staff expressed a similar view. ;
Notwithstanding such accolades, MFP has brought to our attention several aspects or areas of the maintenance and surveillance program that warrant some corrective action.
We commend MFP for its efforts in this regard. Although not functional areas for which insufficient information exists to support an assessment of licensee performance. Handbook 8.6, at 7.
l
- - -- i
1
- 182 -
sufficient to warrant denial of the license application, they are sufficient to warrant orders for correction.
Specifically, we direct the following corrections:
- 1. The telatemp sticker program must be improved, as described under item IV.C.1 above. These improvements need i
be made only to the extent PG&E determines to use such a program in fulfilling its EQ requirements.
- 2. Conversion of the radiation monitoring system must be completed and a set of rules for working on energized equipment be promulgated (item IV.C.38 above).
- 3. PG&E must undertake a study, to be submitted to the Staff for review, concerning methods for improving communications between maintenance and other departments, to the extent maintenance elects to use those departments in implementing its maintenance and surveillance program (see ,
7 items IV.C.9, 10, 15, 29, 34, 36,.37).
We delegate scheduling and confirmation of satisfactory completion of these matters to the NRC Staff. Such delegation of post-hearing matters is appropriate where, as here, they involve deficiencies that should be corrected but which do not pertain to the basic findings necessary to the issuance of a license. Public Service Co. of New Hamoshire-(Seabrook Station, Units 1 and 2), CLI-90-3, 31 NRC 219, 230-31 (1990); Consolidated Edison Co. of New York. Inc. !
l (Indian Point Station, Unit No. 2), CLI-74-23, 7 AEC 947 )
(1974).
- 183 -
V. Renewed Motion to Reopen the Record MFP's August 8, 1994 Renewed Motion to Reopen the Record seeks to include in the record on Contention I an NOV issued by the NRC Staff on July 14, 1994, together with various materials included in two inspection reports dealing with the subject of that violation as well as with certain other matters covered in those reports. The motion is opposed by both PG&E and the NRC Staff.
Such motions are governed by 10 C.F.R. S 2.734, which requires that a motion to reopen a closed record be timely (except in circumstances not here pertinent), that it address a significant safety or environmental issue, and that it demonstrate that a materially different result would be or would have been likely had the newly proffered evidence been considered initially. Further, the motion must be accompanied by one or more affidavits.
On July 22, 1994, the Staff advised the Board and parties of the issuance of the NOV. Letter from NRC Staff counsel to Licensing Board, dated July 22, 1994, transmitting copy of NOV dated July 14, 1994. The NOV cover letter to PG&E, dated July 14, 1994, characterizes the NOV (based on IR 94-08) as "a significant violation involving the failure of your engineering staff to fully recognize or correct operational deficiencies in the Auxiliary Saltwater I
i l
i
-. .__ - - ~.
- 184 -
(ASW) System, despite several opportunities to recognize the existence of.these deficiencies.""
The letter went on to note that NRC had also
" considered separate citations for failure to implement adequate design control measures to assure that ASW-specifications and procedures were adequate to properly limit maximum Component Cooling Water (CCW) ~ temperature during a design basis accident, and for failure to provide complete and accurate information to the NRC regarding the results of ASW system testing." The NRC explained that those failures were a " direct consequence of the poor engineering work" and, as a result, " separate citations are not warranted." NRC letter to PG&E, dated July 14, 1994.
In its Renewed Motion, MFP notes that it previously had sought to reopen the record to introduce IR 93-36 (January 12, 1994), to demonstrate tha'; the surveillance program for the ASW system was inadequate and PG&E failed to perform needed maintenance, and that in LBP-94-9 we dismissed the ,
motion without prejudice because the issues raised were as yet " unresolved." In that Order, we also established a threshold showing for any renewed motion, that the information be demonstrated as "significant and possessing substantive implications with respect to implementation of l
"The Staff transmitted copies of IR 94-08 to the Board and parties by Memorandum dated March 17, 1994 (Board jl Notification 94-06).
- 185 -
the maintenance / surveillance program" at DCPP. LBP-94-9, 39 NRC at 125.(footnote omitted).
MFP characterizes the NOV as confirming the existence of problems initially raised in IR 93-36. MFP goes on to claim that the motion is timely filed, that it raises significant safety and environmental issues, and that it is likely to affect the outcome of the case relative to Contention I "because it contradicts testimony at the hearing by PG&E and the NRC staff and because it corroborates many aspects of [MFP's] position that PG&E's maintenance and surveillance program is inadequate." MFP adds that "the fact that PG&E appears-to have misrepresented the operability of the ASW system and the status of its maintenance and surveillance program implicates the integrity of PG&E's entire maintenance and surveillance program'and the reliability of PG&E's testimony in this case." Renewed Motion, at 3.
PG&E and the Staff do not contest the timeliness of the Renewed Motion, at least insofar as it seeks to add the NOV to the record. (They question the timeliness of other matters arising from IR 94-08.) Nor at this time do they contest the safety significance of the contention to which the new inspection findings may be relevant. Rather, they question the significance and indeed relevance of the
" engineering" matter to the maintenance and surveillance contention. They mention PG&E's August 11, 1994 response to
l i
l
- 186 -
i the NOV, including corrective actions to prevent recurrence l
(copies of which were included in PG&E's response to the j motion) and NRC's acceptance of that response (also included 1
in PG&E's response to the motion). They also each stress that the matter addressed in the NOV was an engineering performance issue, not a maintenance, surveillance or equipment operability issue and thus has little bearing bn Contention I.
Finally, they argue that the other issues that MFP attempts to raise were derived from IR 93-36 but were closed out by the Staff in IR 94-08, without giving rise to any enforcement action. They thus assert that MFP's motion does not meet the threshold established in LBP-94-9 for reopening the record.
In support of its response, PG&E presented the affidavit of Michael J. Angus, PG&E's Manager of Nuclear Engineering Services, who is responsible for overall management of PG&E's engineering support and design engineering activities at DCPP. He affirms that, with the exception of the one NOV, all inspection issues cited in the Renewed Motion have been closed out. He also describes the limited scope of the NOV and the method of resolving or closing out the other issues derived from IR 93-36 that were raised by MFP. ,
For its part, the Staff's response included affidavits of three of the witnesses who appeared before us during the I
i
i
- 187 -
evidentiary hearing--Paul P. Narbut, Sheri R. Peterson, and Mary H. Miller. Most important, Mr. Narbut, the Staff Senior Inspector who was the author of both IR 93-36 and IR 94-08, confirms that PG&E has a sound maintenance and surveillance program and that he does not have any " current f
concerns" for the operability of the ASW system. He expresses his belief that PG&E has shown that the ASW system was "at all times operable" and that it "has done a credible job of addressing my technical and management inspection issues." He also describes a number of technical inadequacies and factual errors in the Renewed Motion.
For their part, Ms. Peterson and Ms. Miller reaffirm their earlier testimony that PG&E's maintenance and surveillance program is clearly supportive of safe facility operation. They too differentiate an engineering deficiency from a deficiency in the maintenance and surveillance program. They note that the single violation does not equate to overall inadequate performance and that it would be taken into account in future SALP evaluations. Ms.
Miller explicitly states that " [t] he f act that specific problems and findings were identified is not unusual or unexpected and is not inconsistent with NRC's SALP evaluation." Affidavit of Mary H. Miller in support of NRC Staff Response to MFP's Renewed Motion to Reopen the Record, dated August 25, 1994, at i 5.
- 188 -
In reviewing the motion, we must recognize the rigorous standards that the Commission has imposed to warrant ;
reopening of a closed record. And we find that those-standards have not here been satisfied.
~
First, we agree with MFP that the NOV concerning [
engineering activities is sufficiently related to maintenance and surveillance of the ASW system to be within the scope of Contention I. We are unwilling to thrust the j violation into a separate organizational box in order to dispose of it artificially on that basis. Instead, we view- !
i' it as another aspect of the relationship between maintenance and engineering that is included in the INPO definition of ;
maintenance that we adopted as a guideline earlier in this Decision. Further, a Severity Level III violation is also i
significant enough to constitute information pertinent to the contention and to satisfy that aspect of the reopening criteria. Finally, the Renewed Motion was clearly timely filed, at least insofar as it was based on the NOV.
We are denying the Renewed Motion for one basic reason:
t the proffered information could not result in a different -
decision from the one we otherwise are reaching. The NOV undoubtedly constitutes a " black mark" on PG&E's record, comparable in some respects to the CFCU matter. Cf. Tr.
2214 (Miller). That incident was taken into account in l
NRC's SALP evaluation, but it did not preclude PG&E from receiving a superior rating. In the SALP report, it was !
i
- 189 -
deemed to be "an isolated example [that] didn't really l warrant a programmatic problem." Tr. 2215 (Peterson). In ;
F the words of Ms. Miller, " [il t's possible to have superior performance and not be perfect." Tr. 2166 (Miller) . On the i basis of the Staff affidavits submitted in response to this motion, we view this NOV similarly--a deficiency that, particularly with respect to maintenance and surveillance, does not reflect a programmatic inadequacy.
By the same token, based on the Staff affidavits,_none of the other matters arising out of irs 93-36 or 94-08 warrants reopening of the record, either separately or as a group (even considered in conjunction with the NOV). None f
has thus far resulted in escalated enforcement action by the ,
Staff, although several " apparent violations" were still under review as of the time IR 94-08 was issued.
In the context of licensing rather than enforcement ;
significance, and of Contention I in this proceeding, the ,
most serious apparent violation is the failure to provide complete information to the Staff. As set forth in IR 94- !
08, i the inspector concluded that the licensee failed to provide complete and accurate information to the NRC in regards to the CCW 1-2 heat exchanger's ability to meet the design basis heat load. This failure is considered j an apparent violation (Apparent Violation 50-275/94 '
02).
Inspection Report 50-275/94-08; 50-323/94-08, Details, at 5. .
i According to Mr. Narbut, this " apparent violation" concerns
- 190 -
" failure to report material details, i.e., incompleteness" of a report, not " misrepresentation" as characterized by MFP. Narbut Affidavit i 2. Significantly, the Staff has not issued an NOV on this matter and thus does not seem to deem it significant enough to warrant further enforcement action. As set forth by Ms. Miller, PG&E's " integrity in responding to maintenance problems . . . is not raised in the NOV, nor is . . . integrity considered a concern."
Miller Affidavit i 9.
We conclude that the incident cited by MFP (either alone or in conjunction with the other matters raised by the motion) could not alter the result we are reaching and the motion is therefore being denied.
VI. Contention V (Thermo-Lac Interim Compensatory Measures)
Contention V, as initially submitted, read as follows: i It is the contention of the San Luis Obispo Mothers for l Peace that the Thermo-Lag material fails as a fire ;
barrier and, in fact, poses a hazard in the event of a fire or an earthquake. Until this situation is adequately resolved, the license for Diablo Canyon Nuclear Plant certainly should not be extended.
LBP-93-1, 37 NRC at 26.
Thermo-Lag is a fire barrier material that has been 1 used at DCPP. Recent testing of Thermo-Lag material has raised questions as to its ability to perform satisfactorily i
as a fire barrier for the rated durations specified for l 1
certain applications. The problem with Thermo-Lag fire j barriers is generic in the nuclear power plant industry i
__m___*-_
.1 l
I
- 191 -
1 because a majority of nuclear power plants have used the.
material to satisfy NRC fire protection requirements. The asserted deficiencies with Thermo-Lag are that 1) Thermo-Lag fire barriers may not provide the fire resistance necessary to satisfy NRC fire protection requirements; 2) Thermo-Lag may burn more readily than originally believed; and 3) the ampacity derating factors used by licensees to derate power cables may not be great enough to account for the insulating effects of the Thermo-Lag material. PG&E Test., ff. Tr.
1277, at 3, 4; NRC Staff Test., ff. Tr. 1417, at 2 (Madden).
The NRC has issued a series of Information Notices regarding deficiencies found in Thermo-Lag 330 fire barrier ;
material. Additionally, it has required that nuclear power plants implement interim compensatory-measures, pending determination by the Staff of possible additional corrective steps that may be required. Egg NRC Bulletin 92-01 (6/24/92) and Supplement 1 to NRC Bulletin 92-01 (8/28/92).
PG&E's response to this NRC request is documented in a September 28, 1992 Response to NRC Bulletin 92-01, Supplement 1. PG&E Exh. 3. In this response, PG&E identified eleven specific Thermo-Lag fire areas at DCPP that are subject to these interim compensatory measures.
PG&E's compensatory measures include: (1) a roving fire watch where fire detection devices are employed; or (2) a continuous fire watch where fire detection devices are not available. Tr. 1287, 1288 (Cosgrove, Powers). PG&E's
- 192 -
interim compensatory measures have been accepted by the NRC Staff and documented in a letter dated October 27, 1992.
PG&E Test., ff. Tr. 1277, at 13; PG&E Exhs. 3, F-1.
In admitting this contention, we limited it to the portion dealing with the adequacy of PG&E's resolution of the Thermo-Lag issue. Specifically, we accepted the ,
i allegation that PG&E has failed to implement adequately and )
abide by the Commission's 2nterim compensatory measures required for the use of Thermo-Lag fire barriers. LBP-93-1, 37 NRC at 27-28, as clarified by " Memorandum and Order (Discovery and Hearing Schedules)," dated February 9, 1993, at 2 (unpublished). The scope of this contention does not include whether fire watches, as a compensatory measure, are an adequate substitute for Thermo-Lag fire barriers declared inoperable. To reiterate, the sole issue is whether PG&E has adequately implemented and will continue to implement adequately the Staff-approved compensatory measures at DCPP.
Tr. 1297, 1299, 1430; LBP-93-1, 37 NRC at 27-28, s clarified by " Memorandum and Order (Discovery and Hearing Schedules)," dated February 9, 1993, at 2.
As litigated, Contention V also does not concern ;
1 inside-containment applications (radiant energy shields) of l Thermo-Lag material. In rejecting a late-filed contention in this proceeding, we specifically ruled that, as a result 1
of PG&E's decision to replace Thermo-Lag material in this application with shields of a different manufacturer, the i
l 1
- 193 -
issue of radiant energy shields and the allegation that the Thermo-Lag material is itself a fire hazard in these applications no longer raised an issue creating a genuine dispute. LBP-93-9, 37 NRC at 444-45.
With respect to this contention, PG&E offered testimony from David K. Cosgrove, Supervisor of the Safety and Fire Protection Group at DCPP, and Robert P. Powers, Manager of the Nuclear Quality Services Department of PG&E's Nuclear Power Generation Business Unit. The NRC Staff presented testimony from Patrick M. Madden, Senior Fire Protection Engineer, Office of Nuclear Reactor Regulation, and Mary H.
Miller, Senior Resident Inspector at DCPP. All of these witnesses were well qualified for their testimony.
MFP asserts that the interim compensatory measures have been inadequately implemented and hence are ineffective. It advances essentially three reasons: (1) that PG&E has not demonstrated reasonable assurance that its interim compensatory measures can and will be reliably impleraented until such time as the generic Thermo-Lag issue is resolved; (2) that inoperable fire detection / suppression equipment, coupled with the failure by personnel to implement or perform compensatory fire watches, compromises ':he critical detection / suppression component of PG&E's defense-in-depth fire protection program and jeopardizes the safe operation of the plant; and (3) that human error and inadequate understanding jeopardize the adequacy of PG&E's i
.- . ._ __ -_ __ _ _ J
- 194 -
implementation of compensatory fire protection measures at DCPP. MFP FOF ij 791, 793, 797.
Only a moderate amount of Thermo-Lag is installed at DCPP. " Moderate" describes an installation incorporating between 100 to 1,000 square feet or between 100 and 1,000 linear feet of fire barrier material. PG&E has already i
replaced the Thermo-Lag in the Unit 2 containment and has I advised that it would replace all the Thermo-Lag in the Unit 1 containment during the refueling outage that was then scheduled for February 1994. NRC Staff Test., ff. Tr. 1417, at 2-3 (Madden).
PG&E has utilized a roving fire watch program throughout DCPP " essentially since Units 1 and 2 have been in operation." PG&E Test., ff. Tr. 1277, at 6, 7.
Therefore, implementation of the fire-watch portion of the interim compensatory measures required only that "the tour route [be) slightly modified to encompass the additional fire areas." Id. at 14. In response to a PG&E objection that attempted to limit inquiry into fire watches solely to i those established for Thermo-Lag purposes, we ruled that "since the fire watch program is merely an extension of an existing program . . . inquiry into potential deficiencies ,
in the existing program, the fire watch program, [is) i 1
l permissible." Tr. 1297. j We also ruled that inquiry into other aspects of the .
l implementation of interim compensatory measures (beyond fire
- l l
1
i 1
- 195 -
watches) was permissible. That does not mean, however, that i inquiry into all aspects of the fire protection program in all fire areas at DCPP is permissible, as claimed by MFP '
(MFP FOF $ 789). There must be a demonstrated relationship to implementation of the interim compensatory measures.
MFP has specified a number of particular incidents or conditions that are said to demonstrate that PG&E is not properly implementing the interim compensatory measures (MFP Exhs. F-1A, F-2, F-3, F-5, F-6). Most involve missed hourly or continuous fire watches, relating to fire watches initially established prior to the Thermo-Lag watches. We find these are relevant because of the administrative similarity between those watches and the Thermo-Lag watches.
We will deal with each specifically.
First, MFP cites an LER that reflects a 1991 technical-specification violation for failure to perform an hourly fire-watch patrol for areas that had inoperable fire barriers (not related to Thermo-Lag deficiencies). The fire watch was not performed because the hourly roving fire watch was unable to exit from the radiologically controlled area of the plant and exchange duties with another fire watch in the turbine building. The root cause was determined to be a lack of adequate instructions to fire-watch personnel. MFP ,
i Exh. F-1.
The next document is a 1992 LER representing a continuous fire watch missed through personnel error.
- 196 -
Corrective action included clarification of procedures and additional. personnel training. MFP Exh. F-2.
The third-document is also a 1992 LER reporting another :
continuous fire watch missed through personnel error. The sprinkler fire water to certain areas was isolated in accordance with an equipment tagout request-without the Shift Foreman noting that a continuous fire watch was needed. An hourly watch ended up patrolling in the affected area. The root cause was determined to be personnel error on the part of the Shift Foreman, and corrective action included counseling the Shift Foreman and operators involved and issuing an Operations Coordination Instruction concerning equipment tagout requests affecting the fire protection system. MFP Exh. F-3. The incident did not involve a Thermo-Lag fire area. Tr. 1330 (Cosgrove).
The next document was also a 1992 LER, involving two i separate events. In the first, fire detectors in a particular area were inoperable for more than an hour, with no compensatory measures in place. In the second, fire detectors in another area were inoperable for more than an i hour without a continuous fire watch in place. Both events were attributed to personnel error. Corrective actions included revision of procedures, preparation of an incident summary outlining the events, and clearly stating the expectations for dealing with spurious alarms and enhancing on-shift training of plant operators. MFP Exh. F-5.
- 197 -
The final document cited by MFP also concerned two events, occurring in late-1992. In the first, the fire detection computer was inoperable for more than an hour, without initiation of required compensatory measures. In the second, the fire detection computer also malfunctioned, but the malfunction was not detected until the following day. Consequently, the required roving hourly fire watch was not instituted in a timely manner. (During investigation, PG&E discovered another computer malfunction.) The December 28, 1992 LER reported that the investigation was still in progress. MFP Exh. F-6. None of these events involved Thermo-Lag barriers. Tr. 1325 (Cosgrove).
PG&E maintains that, since the initiation of the interim compensatory measures for Thermo-Lag fire-areas, it has successfully completed 100% of the hourly fire watches.
PG&E FOF 1 T23, citing PG&E Test., ff. Tr. 1277, at 14 '5 .- l (Cosgrove, Powers), and Tr. 1320 (Powers). It defines a successful tour of Thermo-Lag areas to be one entry into the defined area within the appointed hour and. concludes that the interim compensatory measures have been successfully implemented. PG&E F0F $ T23, citing Tr. 1307-08 (Powers).
The Staff agrees. Staff FOF 11 V-30, V-54. MFP does not dispute the record concerning Thermo-Lag watches but continues to assert that the other record is more representative.
+ - - - - - - - - - - . - - - - - _ - - - _ - - - - - - - - - . -
. ~
- 198 -
Although we agree with MFP that the various missed fire watches are relevant for evaluating the likelihood of missed Thermo-Lag watches, we do not regard PG&E's overall record as flawed or as posing a threat to the adequacy of its compliance with fire-protection requirements. Reasons for various missed watches have been identified, and PG&E has taken adequate steps to prevent their recurrence. ,
During the hearing, the Board also explored claims to the effect that PG&E had improperly altered the records of fire watches (Tr. 1282-87, 1322-24, 1389-1409 (Cosgrove, Powers)). No party has pursued that matter in its FOF, and we thus regard it as abandoned.
In short, nothing in the record would suggest that PG&E's implementation of the Thermo-Lag interim compensatory measures was so flawed that license denial (or even license conditions) are warranted.3' VII. Conclusions of Law The Licensing Board has considered all of the evidence .
submitted by the parties as well as the entire hearing i i
3'By letter dated March 9, 1994, the Applicant advised us that, as it notified the Staff on February 14, 1994, l "PG&E will complete replacement of the Thermo-Lag . . . as appropriate with other fire systems by December 31, 1994.
Compensatory measures will be maintained as appropriate for each Thermo-Lag installation until its replacement is complete." PG&E Letter No. DCL-94-034, at 4. Although our resolution of Contention V is not based on this communication (which is not in the record and has not been subject to cross-examination), we note that, when accomplished, it would render Contention V moot.
l l
i
- 199 -
record. That record consists of the Commission's Notice of Hearing, the pleadings filed herein, memoranda and orders issued in this proceeding, the transcript of the hearing, ,
and the exhibits received or deemed to be received into evidence. Based on the findings of fact set forth in Parts IV and VI above, which are supported by reliable, probative ;
and substantial evidence as required by the Administrative Procedure Act'and the Commission's Rules of Practice, we conclude that:
(1) PG&E has met its burden of proof with respect to both Contention I (subject to certain limited conditions) i and Contention V. .
(2) With respect to the matters placed in controversy by these two contentions, and subject to the limited conditions set forth with respect to Contention I, there is reasonable assurance that: (a) the Diablo Canyon Nuclear ,
~
Power Plant can and will be operated without endangering the i
public health and safety for the requested 40-year operating life; (b) such activities will be conducted in compliance with the Commission's regulations, and (c) such activities will not be inimical to the common defense and security.
(3) All issues, arguments or proposed findings presented by the parties but not addressed herein have been found to be without merit or unnecessary for this decision.
m - * * ' - w
- _ _ _ _ _ _ _ _ _ -
- 200 -
VIII. Order WHEREFORE, on the. basis of the foregoing, in accordance with 10 C.F.R. S 2.760, it is this 4th day of November 1994, ORDERED:
- 1. The Director, Office of Nuclear Reactor Regulation, is authorized, upon making requisite findings with respect to matters not at issue in this Initial Decision, and subject to the conditions specified on pages 180-81 with
[ respect to Contention I, to issue the amendment proposed by PG&E in its application of July 9, 1992. ;
- 2. This Initial Decision shall become effective and constitute the final action of the Commission forty (40) days after the date of its issuance, subject to any review pursuant to the Commission's regulations.
- 3. MFP's Renewed Motion to Reopen the Record, dated August 8, 1994, is hereby denied.
- 4. In accordance with 10 C.F.R. S 2.786, any petition for review of this Initial Decision rust be filed within l fif teen (15) days after service of the decision. Any other party may file, within ten (10) days after service of a petition for review, an answer in support of, or in opposition to, the petition for review. The petition for review may be granted or denied in the discretion of the e , ,,, _ . . _ _ _ _ _ _ _
- 201 -
Commission, giving weight to the considerations of 10 C.F.R.
S 2.786 (b) (4) .28 THE ATOMIC SAFETY AND LICENSING BOARD 1 u r e v Charles Bechh'oefer, Ch/irman ADMINISTRATIVE JUDGE f,f/f.41 0
- r. Jertly R. Kline MINISTRATIVE JUDGE
/?
f r frederick J. S n ADMINISTRATIV sJUDGE Rockville, MD.
November 4- 1994 l
l I
l 1
i
In the near f uture, we shall issue transcript corrections for the evidentiary hearing.
1 UNITED STATES OF AMERICA NUCLEAR REGULATORY COMMISSION In the Matter of PACIFIC GAS AND ELECTRIC COMPANY Docket No.(s) 50-275/323-OLA-2 (Diablo Canyon Nuclear Power Plant, Unit Nos. I and 2)
CERTIFICATE OF SERVICE I hereby certify that copies of the foregoing LB INITIAL DECISION--LBP-94-35 have been served upon the following persons by U.S. mail, first class, except as otherwise noted and in accordance with the requirements of 10 CFR Sec. 2.712.
Office of Comission Appellate Administrative Judge Adjudication Charles Bechhoefer, Chairman
/ U.S. Nuclear Regulatory Comission Atomic Safety and Licensing Board Washington, DC 20555 U.S. Nuclear Regulatory Comission Washington, DC 20555 Administrative Judge Administrative Judge Jerry R. Kline Frederick J. Shon Atomic Safety and Licensing Board Atomic Safety and Licensing Board U.S. Nuclear Regulatory Comission U.S. Nuclear Regulatory Comission Washington, DC 20555 Washington, DC 20555 Ann P. Hodgdon, Esq. C. J. Warner, Esq.
Office of the General Counsel Richard F. Locke, Esq.
U.S. Nuclear Regulatory Comission Pacific Gas & Electric Company Washington, DC 20555 77 Beale Street San Francisco, CA 94106 Robert R. Wellington, Esq. Joseph B. Kn.atts, Jr., Esq.
Diablo Canyon Independent Safety David A. Repka, Esq.
Comittee Winston & Strawn 857 Cass Street, Suite D 1400 L Street, N.W.
Monterey, CA 93940 Washington, DC 20005 t
Docket No.(s)50-275/323-OLA-2 LB INITIAL DECISION--LBP-94-35 Truman Burns Peter G. Fairchild, Esq.
Robert Kinosian California Public Utilities Commission California Public Utilities Commission 505 Van Ness Avenue 505 Van Ness, Room 4103 San Francisco, CA 94102 San Francisco, CA 94102 Nancy Culver, President Diane Curran, Esq.
Board of Directors Counsel for SLOMFP Mothers for Peace c/o IEER P. O. Box 164 6935 Laurel Avenue, Suite 204 Pismo Beach, CA 93448 Takoma Park, MD 20912 Dated at Rockville, Md. this 4 day of November 1994 i b_ N-hM Offiggofthe/5ecretaryoftheCommission
)
l
[
i l
l l
J I
l l
1
- 201 -
Commission, giving weight to the considerations of 10 C.F.R. !
I S 2.786 (b) (4) .38 1 i
THE ATOMIC SAFETY AND LICENSING BOARD l
r a fbr n,- a Charles Bechh'oefer, Ch/irman ADMINISTRATIVE JUDGE l' bCO4
- r. Jerdy R. Kline MINISTRATIVE JUDGE
/l frederick J. S n ADMINISTRATIV < JUDGE Rockville, MD.
November 4, 1994 l
l I
1 l
I l
3*In the near future, we shall issue transcript corrections for the evidentiary hearing.
r-UNITED STATES OF AMERICA NUCLEAR REGULATORY COMMISSION In the Matter of
~
PACIFIC GAS AND ELECTRIC COMPANY Docket No.(s) 50-275/323-OLA-2 (Diablo Canyon Nuclear Power Plant,
. Unit Nos. I and 2)
CERTIFICATE OF SERVICE I hereby certify that copies.of.the foregoing LB INITIAL DECISION--LBP-94-35 have been served upon the following persons by U.S. mail,. first class, except-as otherwise noted and in accordance with the requirements of 10 CFR Sec. 2.712.
Office of Commission Appellate- Administrative Judge Adjudication Charles Bechhoefer, Chairman 7 U.S. Nuclear Regulatory Commission Atomic. Safety and Licensing Board Washington, DC 20555 U.S. Nuclear Regulatory Comission Washington, DC 20555 Administrative Judge Administrative Judge Jerry R. Kline . Frederick J. Shon-Atomic Safety and Licensing Board Atomic Safety and Licensing Board U.S. Nuclear Regulatory Commission U.S. Nuclear Regulatory Commission Washington, DC 20555 Washington, DC 20555 Ann P. Hodgdon, Esq. C. J. Warner, Esq.
Office of the General Counsel Richard F. Locke, Esq.
U.S. Nuclear Regulatory Commission Pacific Gas & Electric Company Washington, DC 20555 77 Beale Street j San Francisco, CA 94106 Robert R. Wellington, Esq. Joseph B. Knotts, Jr., Esq.
Diablo Canyon Independent Safety David A. Repka, Esq.
l Committee Winston &'Strawn 857 Cass Street, Suite D 1400 L Street, N.W.
Monterey, CA 93940 Washington, DC 20005 l
l E
1 t
II
- Docket No.(s)50-275/323-OLA-2 LB INITIAL DECISION--LBP-94-35 Truman- Burns Peter G. Fairchild, Esq.
Robert Kinosian California Public Utilities Commission California Public Utilities Commission 505 Van Ness Avenue 505 Van Ness, Room 4103 San Francisco, CA 94102 '
San Francisco, CA 94102 Nancy Culver, President Diane Curran, Esq.
Board of Directors Counsel for SLOMFP Mothers for Peace c/o IEER i, O. Box 164 6935 Laurel Avenue, Suite 204 Pismo Beach, CA 93448 Takoma Park, MD 20912 4
l r
l 1
1 i
j
- -. ..- ,.