ML16342D432
| ML16342D432 | |
| Person / Time | |
|---|---|
| Site: | Diablo Canyon  |
| Issue date: | 08/26/1996 |
| From: | Dyer J NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION IV) |
| To: | Rueger G PACIFIC GAS & ELECTRIC CO. |
| References | |
| NUDOCS 9609230026 | |
| Download: ML16342D432 (18) | |
See also: IR 05000275/1996012
Text
+gS REQII,
4'4
O
Cy
n
Y/
0
'~ ++*++
UNITEO STATES
NUCLEAR REGULATORYCOMMlSSlON
REGION IV
611 RYAN PLAZAORIVE, SuITE 400
ARLINGTON,TEXAS 760114064
August 26,
1996
Gregory M. Rueger, Senior Vice President
and General Manager
Nuclear Power Generation
Bus. Unit
Pacific Gas and Electric Company
Nuclear Power Generation, B14A
77 Beale Street, Room 1451
P.O. Box 770000
San Francisco, California 94177
SUBJECT:
NRC lNSPECTION REPORT 50-275/96-12'0-323/96-12
Thank you for your letter of August 9, 1996, in response to our letter and Notice of
Violation dated July 10, 1996.
We have reviewed your reply and find it responsive to the
concerns raised in our Notice of Violation. We willreview the implementation of your
corrective actions during a future inspection to determine that full compliance has been
achieved and will be maintained.
Sincerely,
Docket Nos.:
50-275
50-323
License Nos.:
DPR-82
J. E. Dyer, Director
Division of Reactor Projects
CC:
Dr. Richard Ferguson
Energy Chair
1100 lith Street, Suite 311
Sacramento,
California 95814
960923002b
960826
ADOCK 05000275
8
Pacific Gas and Electric Company
-2-
Ms. Nancy Culver
San Luis Obispo Mothers for Peace
P.O. Box 164
Pismo Beach, California 93448
Chairman
San Luis Obispo County Board of
Supervisors
Room 370
County Government Center
San Luis Obispo, California 93408
Mr. Truman BurnshMr. Robert Kinosian
California Public Utilities Commission
505 Van Ness, Rm. 4102
San Francisco, California 94102
Robert R. Wellington, Esq.
Legal Counsel
Diablo Canyon Independent
Safety Committee
857 Cass Street, Suite D
Monterey, California 93940
Mr. Steve Hsu
'adiologic Health Branch
State Department of Health Services
P.O. Box 942732
Sacramento,
California 94234
Christopher J. Warner, Esq.
, Pacific Gas and Electric Company
P.O. Box 7442
San Francisco, California 94120
Robert P. Powers, Vice President
and Plant Manager
Diablo Canyon Power Plant
P.O. Box 56
Avila Beach, California 93424
Pacific Gas
and Electric Company
-3-
E-Mail report to D. Nelson
(DJN)
E-Hail report to
NRR Event Tracking System
(IPAS)
bcc to
DMB IE01
'cc distrib. by RIV:
L. J. Callan
DRP Director
Branch Chief (DRP/E,
WCFO)
Senior Project Inspector
(DRP/E.
WCFO)
Branch Chief (DRP/TSS)
Leah Tremper
(OC/LFDCB,
MS:
TWFN 9E10)
Resident
Inspector
DRS-PSB
HIS System
RIV File
M:
Hammond
(PAO.
WCFO)
DOCUMENT NAME:
R:t DCNDC612AK.FRH
To receive copy of document,
indicate In box: "C" = Copy without enctosures
"E" = Copy with enctosures
"N" = No copy
C:DRP/E
HJWong
08
96
D:WCFO
I
D:DRP
KEPerkins
JEDyer
08@696
08/ /96
08/&/96
OFFICIAL RECORD COPY
08/ /96
0
Pacific Gas
and Electric Company
-3-
E-Mail report to D. Nelson
(DJN)
E-Mail report to
NRR Event Tracking System
(IPAS)
.bcc to DMB IEOl
bcc distrib. by RIV:
L. J. Callan
DRP Director
Branch Chief (DRP/E.
WCFO)
Senior Project Inspector
(DRP/E.
WCFO) .
Branch Chief (DRP/TSS)
Leah Tremper
(OC/LFDCB,
MS:
TWFN 9E10)
Resident
Inspector
DRS-PSB
MIS System
RIV File
M.
Hammond
(PAO.
WCFO)
DOCUMENT NAME:
R:K DCNDC612AK.FRH
To receive copy of document,
indicate In box: "C" = Copy without enclosures
"E" = Copy with enctosures
"N" = No copy
C:DRP/E
HJWong
08
96
D:DRP
D:WCFO
KEPerkins
JEDyer
08/&/96
08/ /96
08@&6
OFFICIAL RECORD COPY
08/ /96
0
Pacific Gas and Hectric Company
August 9, 1996
PG8 E Letter DCI=96-167
245 Market Street. Room 937-N9B
San Francisco, CA 94105
.ila///ug ilrlclnm
Mail Code N9B
P.O. Box 770000
San Francisco. CA 94177
415/973-4684 Fax 415/973-2313
Gregory M. Rueger.
Senor Vice President and
General Manager
Nuclear Power Generation
U.S. Nuclear Regulatory Commission
ATI'N: Document Control Desk
Washington, D.C. 20555
Docket No~ 50-275, OL-DPR-80
Docket No. 50-323, OL-DPR-82
Diablo Canyon Units 1 and 2
Re
I to Notices of Violation and Notice of Deviation in NRC Ins ection Re ort Nos. 50-275/96-
12 and 50-323/96-12
Dear Commissioners and Staff:
NRC Inspection Report Nas. 50-275/96-12 and 50-323/96-12, dated July 10, 1996, and
amended on July 11, 1996, included two Severity Level IVviolations regarding implementation
of the Unit 1 Main Steam Safety Valves (MSSVs) Augmented Testing and Inspection Program.
The inspection report also included a notice of deviation regarding PG8 E's failure to obtain
valve-specific correction factors during the Unit 1 seventh refueling outage (IR7) for 11 of 20
MSSVs contrary to a commitment made in PG8 E letter DCL-95-241, dated November 1, 1995.
PG8 E's response, summary cause analysis, and corrective actions associated with the
violations and deviation are enclosed.
The specifics associated with events leading to these
violations and the deviation have already been discussed
in detail at the July 1, 1996,
Predecisional Enforcement Conference held in Arlington, Texas.
PG&E wants to emphasize its recognition and appreciation of two points expressly mentioned in
the inspection report.
First, PG8 E as the licensee has the continuing and ultimate responsibility
to ensure that the Diablo Canyon Power Plant is operated at all times in accordance with its
Technical Specifications.
In this instance, PG8 E did not fulfillthis responsibility, because of a
lack of formality in developing and implementing the augmented MSSV testing program.
Second, PG&E recognizes and willcontinue to emphasize the absalute need for assuring that
all communications between PG8 E and the NRC are made in a prompt and accurate manner,
are made to the appropriate NRC persannel, and are properly documented.
Sincerely,
CC:
Steven D. Bloom
L. J. Callan
Kenneth E. Perkins
Michael D. Tschiltz
Diablo Distribution
Enclosure
WEC/2237/N0001 327
Enclosure
PG8,E DCL Letter 96-167
REPLY TO NOTICES OF VIOLATIONIN
NRC INSPECTION REPORT NOS. 50-275/96-12 AND 50-323/96-1 2
Qn July 11, 1996, as part of NRC Inspection Report (IR) Nos. 50-275/96-12 and
50-323/96-12, NRC Region IV issued two Severity Level IV Notices of Violation (NOVs)
and one Notice of Deviation (NOD) to Diablo Canyon Power Plant (DCPP), Units 1
and 2. The statements of violation and deviation and PG&E's responses
are
documented below.
STATEMENT OF VIOLATIONA
10 CFR Part 50, Appendix B, Critenon XVI, states, in part, that measures shall be
established to assure thaf condifions adverse to qualify are promptlyidentified and
corrected.
l.
Confrary to the above, on April2; 1996, fhe licenseeidentified a condition adverse to
quality wifh respect to the liftsetpoinfs on 3 of5 Unit 1 main steam safefy valves tested
(main steam lead 1), and did nof take prompt acfion to determine whether the same
condition existed on the main sfeam safefy valvesin main steam leads 2, 3, and 4.
Specifically, fhe main steam safety valves on main steam lead 2 were not tested until
April 11, 1996, when 3 of5 were found ouf-of-tolerance, and the main steam safety
valves on main steam leads 3 and 4 were not fested untilApril14, 1996, when 5 of5
valvesin main-steam lead 3 and 1 of5 valvesin main steam lead 4 were found ouf of-
tolerance.
This is a Severity Level /Vviolafion (Supplement I).
REASON FOR THE VIOLATION
PG8 E agrees with the violation as stated in the IR.
PG8 E has determined that the cause of the delays in testing the additional steam leads
was a lack of formality and rigor in developing and implementing the Augmented
Testing Program for the main steam safety valves (MSSVs). Though the procedure for
performing the augmented testing was well defined, PG&E did not adequately define
the actions that should be taken should adverse test results be obtained.
PG&E failed
to recognize that although it had committed to discuss test findings with the NRC before
expanding the scope of testing, its primary responsibility was to assure compliance with
the Technical Specifications (TSs).
PG8 E should have promptly expanded the testing
to assure that additional out-of-tolerance (QQT) conditions were identified and
corrected.
The circumstances
surrounding this failure to promptly expand testing are
addressed
in the following sections.
Enclosure
PG8E DCL Letter 96-167
Back round
In September of 1995, PG8E submitted to the NRC License Amendment Request
(LAR) 95-06, "Request for Emergency Review and Approval of Change to Technical Specification 3.7.1.1, Table 3.7-2- Increase in Setpoint Tolerances for Main Steam
Safety Valves." LAR 95-06 explained that despite efforts to resolve the problem, MSSV
testing results were continuing to show a pattern of high initial lifts. The LAR requested
relaxation of.the TS tolerances and presented the necessary supporting safety
analyses.
On October 1, 1995, the NRC issued License Amendments (LAs) 108 and
107 to PG8 E. The new TSs relaxed the MSSV as-found tolerances from+/-1 percent
to +/- 3 percent (+3/-2 percent for the lowest set valves).
As a condition of the LAs, PGKE was required to perform augmented inspection and
testing on the MSSVs on a more frequent basis than that required by the ASME Code
.
for at least one operating cycle. The NRC required that PG8 E provide additional
details on the program by November 1, 1995.
On November 1, 1995, PG8 E submitted
letter DCL-95-241, "Main Steam Valve Augmented Testing and Inspection Program."
In the letter, PG8E explained that the testing program would provide information on the
time dependency aspects of high initial liftphenomenon observed during past on-line
testing.
PG8 E stated itwould perform augmented testing on the Unit 1 valves one
steam header at a time on a staggered test basis to provide varying time periods
. between header tests.
PG8E also stated that testing to meet the requirements of
TS 3.7.1.1 and the ASME Code for all the Unit 1 MSSVs would be conducted just prior
to the eighth refueling outage (1R8). PG8 E further stated that ifthe MSSV lifts are not
within +/-'
percent, but are within+I-3 percent (+3/-2 percent for the lowest set
MSSVs), the valves would be reset.
No test expansion beyond the five valves on the
test header would be done in an effort to preserve the time related aspect of the test
data.
Should a valve or valves liftoutside the as-found TS tolerance, the valve(s)
would be returned to+/-1 percent tolerance and the'scope of additional testing would
be discussed with the NRC staff prior to expanding the testing.
The NRC responded to PG8E's proposal on December 26, 1995. The NRC response
reiterated PG8 E's proposal and concluded by stating that the NRC staff had reviewed
the licensee's submittal dated November 1, 1995, and had determined that the
augmented testing and inspection plan described therein was acceptable.
T~iP
Qn April 2, 1996, PG8 E conducted its first round'of augmented testing of Unit 1 Steam
1 in accordance with approved work orders and Maintenance Procedure M-4.18,
'Verification of LiftPoint Using Ultra Star Assist Device for the Main Steam Safety
Valves."
Enclosure
PG&E DCL Letter 96-167
When testing found three of five MSSVs OOT high on the initial lifts, PG8 E reset the
valves, documented the as-found OOT conditions on action requests (ARs), and
requested analyses to be performed on results by the San Francisco based Regulatory
and Design Services'R8DS) Transient Analysis (TA) Group. The R8DS TA Group
performed a preliminary analysis (projecting results over 20 valves), determined the
results were within the design bases of the plant, and forwarded the findings to the site
on April4, 1996.
On April 8, 1996, PG8 E initiated a conference call with the NRC to
discuss the Unit 1 MSSV as-found data, analysis results, and the possible expansion of
testing.
PG&E informed the NRC that data on five additional valves would be collected
and that PG8 E would notify the NRC of the test results.
On April 10 and 11, 1996, PG&E tested Main Steam Lead 2. The initial lifts of three of
five MSSVs were high OOT. PG8 E reset two of the OOT valves and the other was left
at+ 1.4 percent.
(See response to NOV "B"contained herein.)
PG&E documented the
as-found OOT conditions on ARs and submitted the data to the TAgroup for analysis..
On the morning of April 12, 1996, PG8E and the NRC held another conference call to
discuss the latest findings and further testing.
PG8 E informed the NRC that testing on
the remaining ten valves on Steam Leads 3 and 4 would resume on April 14, 1996, and
that results should be available on April 15 or 16, 1996.
At 1730 PDT, on April 12, 1996, PG8 E made a 10 CFR 50.72 report to the NRC that
the as-found liftpressures ofsix of ten Unit 1 MSSVs (Main Steam Leads
1 and 2)
exceeded the:TS setpoint tolerance of +1- 3 percent and that preliminary analysis of the
data indicated that the condition constituted a condition outside the design basis of the
plant. At approximately 2130 PDT, on April 12, 1996, another conference call was held
between the NRC and PG8 E in response to the NRC requesting additional information
on MSSV testing, the basis for the 50.72 report, and the operability determination for
the untested ten MSSVs on Steam Leads 3 and 4. PG8 E indicated at this time that
arrangements
were being made to test Steam Leads 3 and 4 valves on April 14, 1996.
On April 14, 1996, PG8 E conducted testing on Steam Leads 3 and 4. On Steam
Lead 3, five of five MSSVs lifted high OOT. PG8 E reset four of the five MSSVs and left
one at 1.2 percent.
(See response to NOV "B"contained herein.)
On Steam Lead 4,
one of five MSSVs lifted high OOT. PG8 E reset the one OOT MSSV. The test results
were documented on ARs and the data was submitted to the R8 DS TA group for
analysis.
A conference call was held with the NRC on April 16, 1996, to discuss the
latest test results.
On April 18, 1996, the R&DS.TA Group forwarded preliminary analysis results to the
site that indicated peak steam pressure remained within the 110 percent design
'ressure
if credit was taken for either direct reactor trip from turbine trip or for the 40
percent steam dump valves.
PG8 E discussed the analyses results with the NRC on
April 19, 1996 and reported the results April 25, 1996 in LER 1-96-003-00, "Technical
. Enclosure
PG8 E DCL Letter 96-167
Specification 3.?.1.1 Not Met During Testing of MSSVs Due to Setpoint DriA; Unit 1
MSSVs Outside Design Basis Due to Setpoint Drift."
Lessons Learned
While acknowledging in the lR that PG8E did pursue testing of all Unit 1 MSSVs, the
NRC noted that its concern was with the aggressiveness'and
timeliness of PG8 E's
pursuit of the potential OOT conditions. The lR further stated that it appeared from the
NRC's perspective that PGRE engineering personnel had lost sight of PG8E's ultimate
responsibility to maintain the facilitywithin the requirements of the TSs and that this
occurred largely because of a lack of formality and rigor in developing and
implementing the Augmented Test Program.
The IR states that a number of important
decisions in this matter were made by one individual, including decisions regarding
operability, and that these decisions were not appropriately documented nor were they
subjected to further evaluation.
The IR further stated that in regard to this violation, the NRC noted that a similar
condition was discovered in September 1995, when 19 of 20 Unit 1 MSSVs were found
OOT on their initial liftand an analysis showed that this condition would have resulted
in the steam generators exceeding 110 percent of their design pressure.
Thus, PG8 E
should have been more sensitive to the possibility that this same condition existed
when Main Steam Lead 1 was tested on April2, 1996, and three offive MSSVs were
found OOT high, and should have expanded testing to the other leads more promptly.
PG8E agrees with the statements above.
PG&E acknowledges that mistakes and
errors in judgment were made in implementing the MSSV Augmented Testing Program
and has learned several important lessons from this experience.
First, PG8E inappropriately allowed an unacceptable degree of "informality"to be
associated with this Augmented Testing Program and thus did not apply its normal rigor
to the process.
PG8 E now recognizes that the "augmented" nature of this test program
does not relieve it of its responsibility as a licensee to promptly and correctly address
any test results that call into question system operability. Because of the potential for
the discovery of inoperable valves during testing, PG8 E should have been better
prepared for, and should have vigorously pursued, identification and correction of any
additional OOT conditions to assure that the plant continued'to operate within the limits
of the TSs.
Second, where uncertainty existed, prior to beginning the tests PG8E should have
taken the Initiative to clarify its underStanding of the test process to be used with
appropriate NRC personnel.
Enclosure
PG8E DCL Letter 96-167
P
CORRECTIVE STEPS TAKENAND RESULTS ACHIEVED
'I.
PG8E has conducted performance counseling with the individual who made
several of the key decisions during th'e test program.
2.
3.
4.
PG&E has now formalized its Augmented Test Program into Surveillance Test
Procedure (STP) M-77A, Revision 0, "Augmented Test Program for Main
Steam Safety Valves (MSSV)'." The PG&E letter to the NRC (DCL-96-156)
dated Aly16, 1996, transmitted this procedure and Maintenance Procedure
M-4.18, Revision 14, 'Verification of LiftPoint Using Ultra Star Assist Device for
the Main Steam Safety Valves."
STP M-77A includes provisions for scope expansion.
It also states that MSSV
testing shall be considered "in progress" until'all of the valves, including any
expansions,
have been tested.
It states that it is management's
expectation, due
to the importance of the testing, that the testing should continue 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> per
day, 7 days a week, unless test results indicate a design basis concern.
Ifa
design basis concern is indicated, testing willbe completed within 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> of
the identification of the concern.
Though we believe communication
inadequacies regarding augmented testing expectations existed between PG&E
and the NRC, in this situation,. as the licensee it is PG&E's responsibility to
ensure that a common understanding exists.
's
prescribed in M-77A:
a)
Operations and management willbe informed of OOT conditions.
')
Prompt operability assessments
will be generated as necessary.
c)
Time limits have been set on the determination of operability of each
steam lead and test expansion to other steam leads.
A project manager as been appointed to specifically manage the MSSV
Augmented Testing Program.
DATE WHEN FULLCOMPLIANCEWILLBE ACHIEVED
PG8E is currently in full compliance.
The procedure was approved and became effective on July 16, 1996.
It was successfully used on July
18, 1996, to test the MSSVs on Unit 1 Lead 3.
Enclosure
PG8E DCL Letter 96-16T
STATEMENT OF VIOLATION8
Diablo Canyon Technical Specification 3/4.7. 1, Table 3.7-2 requires, in part, that steam
line safety valve liftsettings be returned "Withins1% following main steam line Code
safety valve testing.
Contrary fo the above, on April 11, 1996, and on April14, 1996, the licensee found
main steam safety valves MSS V-223 and MSS V-14 ouf oftolerance by 1.4% and 1.2%
respectively.
Following the testing, the licensee did not reset them within the allowed
tolerance of +1% ofthe specified seffing before returning fhem to an operable status.
Thisis a Severity Level IVviolation (Supplement I)
REASON FOR THE VIOLATION
PG8 E agrees with the violation as stated in the IR.
The setpoints of the valves, as measured with the Ultra Star (or AVK)test equipment,
were not altered at the tim'e of testing because PG8E believed that adjusting the valves
would have resulted in less conservative setpoints than those previously recorded
when the valves were set on live steam at the Westinghouse Western Service Center
(WSC). However, PG8E failed to document in a formal 10 CFR 50.59 safety
evaluation that this decision met the requirements of the TSs.
Further explanation of
this event follows.
PG8 E stated in LER 1-96-003-00 that test personnel were only able to develop
correction factors (CFs) for 9 of 20 Unit 1 valves during the October 1995 testing at the
WSC.
(Also see the response to the NOD contained in this letter.) Of the nine valves
for which CFs were obtained, the data collected showed a maximum AVKsetpoint
variation of +l- 1.4 percent from the settings achieved on steam alone. Therefore,
PG8 E judged that during in-situ testing, valves without CFs that are returned to +l- 1
.percent of their measured setpoint using AVKwould be susceptible to the+I-1.4
percent error introduced by the AVKtest methodology.
That is, a valve returned to
within+i-1 percent could have an additional error of up to+i-1.4 percent added to its
liftsetpoint.
During the in-situ testing, PG8 E considered the foregoing information and the as-found
data in deciding whether or not to reset the valves.
PGBE reasoned that it would b
m re conservative to leave valves unadjusted for which the as-found setpoints were
i wou
e
less than + 1.5 percent and chose to accept instead the more accurate live-steam
settings recorded at the WSC.
Similarly, PG8 E reasoned
it would be more
conservative to adjust valves for which the as-found setpoints were'greater than+ 1.5
percent.
As MSSV-223 and MSSV-14 were two of the eleven valves without CFs and
Enclosure
PG8 E DCL Letter 96-167
P
their as-found setpoints were less than+1.5 percent, they were left unadjusted.
Other
valves without CFs forwhich as-found setpoints greater than +1.5 percent were
accordingly reset to within+/-1 percent.
Though PG&E continues to believe the
reasoning used to not readjust the two valves was correct and that the requirements of
the TSs had been met, PG&E failed to document the decision and the basis in a formal
10 CFR 50.59 safety evaluation.
PG8 E believes this event occurred in part because of the failure to have a formal,
documented Augmented Test Program plan considering all the potential issues that
would develop during testing.
Had PG&E developed a more formal test plan, the issue
of when or when not to reset the valves would have been thoroughly thought through
and formally documented prior to test results being received and time pressures
being
felt.
From this experience, PG&E has learned that it should have documented in a
10 CFR 50.59 safety evaluation that the accepted settings for MSSV-223 and MSSV-14
were those established at the WSC and not those recorded as "as-left" in Maintenance
Procedure M-4.18. This experience also underscores the importance of more formal
test planning, as discussed
in PG&E's response to Violation A.
CORRECTIVE STEPS TAKENAND RESULTS ACHIEVED
1.
MSSV-223 and MSSV-14 were retested on April21, 1996. The valves were left
with settings within +/- 1 percent of setpoint.
2.
STP M-77Awas written and approved to formalize the, test program.
It explicitly
states the acceptance
criteria for as-left settings using the AVKtechnology.
3
The principal individuals involved with the in-situ testing of MSSV-223 and
MSSV-14 participated in the preparation of STP M-77A and.in the preparation or
reviews of the response
to this NOV. Therefore, they are cognizant of the
lessons learned.
DATE WHEN FULL COMPLIANCEWILLBE ACHIEVED
PG&E is currently in full compliance.
Enclosure
PG&E DCL Letter 96-167
STATEMENT OF NOTICE OF DEVIATION
PG8 2 Letter DCL-95-241 dated November 1, 1995, stated thaf "The following
additional MSSV testing willbe completed dunng 1R? and 2R? oufages at fhe
Westinghouse fest facility:
1)
Valve signature
profiles
willbe defermined on live sfeam and with fhe AVKtest
equipmenf
2)'he magnitude ofthe AVKfest equipmenf bias willbe determined."
Confiary to fhe above, on November 24, 1995, the 1R7 outage was complefed, but the
licensee had determined the magnitude ofAYKtest equipmenf bias on only 9 of 20
Unit 1 MSS Vs. As ofJuly 1, 1996, the licensee had not submitted a requesf forrelief
from ifs commifmenf fo defermine AVKfest equipment bias during 1R7.
'I
REASON FOR THE DEVIATION
PG& E agrees with the deviation as stated in the IR.
PG&E failed to obtain the AVKtest equipment bias and Unit 1 valve-specific CFs for 11
of 20 valves due to equipment problems at the test facility. ln addition, PG&E
personnel failed to document and effectively communicate internally and to the NRC
that the commitment would not be met as required by procedure.
Testin
Back round
ln October 1995, during the Unit 1 seventh refueling outage (1R7), all 20 MSSVs were
removed and shipped to.the WSC in Beaumont, Californi. Testing was to be
performed during a nominal two-week window (October 8, 1995, to October 20, 1995)
to obtain AVKtest equipment bias and to perform Code testing to satisfy the
requirements of the TSs.
Near the end of the first week of testing (approximately October 13, 1995), pG&E
engineering personnel collecting the AVKbias information had determined that none of
the data collected was valid because
the steam accumulator on the test stand for this
data collection was too small in capacity to give valid data.
This information was
relayed to the'esponsible
director at DCPP..
The director, after discussions with test personnel at WSC, recognized that only limited
scope testing could be accomplished at WSC during the remaining week of the
available testing window and directed that half of the valves (those for Steam Leads
3 and 4) be returned to DCPP once the Code safety testing and steam signature testing
Enclosure
PG8E DCL Letter 96-167
had been completed using a larger test stand.
It was anticipated that the remaining
AVKvalve-specific bias data could be collected at the site when the valves were re-
installed at DCPP.
It was later determined, however, that in-situ testing for collection of
the bias information could not be done.
Afterthe valves for Steam Leads 3 and 4 were shipped back to DCPP, personnel at the
WSC devoted the remaining time, using the larger te'st stand, to testing and obtaining
complete data for the valves on Steam Leads
1 and 2. On October 20, 1995, the
remaining ten'valves for Steam Leads
1 and 2 were returned to DCPP. At this point in
time, because of the time constraints for testing the valves at the WSC and data being
lost for one valve, PG8 E had collected AVKdata for only 9 of 20 valves.
On November 1, 1995, PG8 E sent the letter (DCL-95-241) containing the commitment
which is the subject of this NOD.
PG8E Investi ation
The investigation into this event revealed that there were deficiencies in PG8 E's
performance in addition to its failure to collect the AVKequipment bias data for all 20
valves.
PG8 E failed to modify the November 1, 1995, letter to the NRC when it was
known to PG8E that testing had been completed at WSC and that only a portion of the
AVKvalve-specific bias data had been collected. Atthe time engineering continued to
believe that the data in question could be obtained once the valves were reinstalled at
DCPP.
PG8 E also failed to notify the NRC of the missed commitment and to
implement corrective action to obtain the data before the end of 1R7 on
November 26; 1995.
The reason for PG8 E's failure to obtain a complete and valid set of AVKequipment
bias data is addressed
in. the testin'g background presented above.
PG8 E believes the
other failures were attributable to a breakdown of proceduralized communications.
PGBE procedure XI1.ID1, "Regulatory and Correspondence
Processing," directs that
the responsible individual or department should contact R8 DS ifthey cannot complete
a commitment action prior to its firm due date.
PG8E has determined that prior to the
end of 1R7 on November 26, 1995, full and effective communication of the status of the
MSSV Augmented Testing Program did not occur among engineering personnel,
R8 DS, and management.
Had this communication occurred as prescribed, PG8 E
would have communicated this information'to the NRC, and:PG8E management would
have had an opportunity to implement corrective action.
This breakdown of communication was also the reason why PG&E failed to generate
an AR to track the commitment to obtain the Unit 1 AVKtest equipment bias information
as stated in the November 1, 1995 letter. PG8 E stated at the Enforcement Conference
Enclosure
PGB,E DCL Letter 96-167
that RB DS had failed to generate an AR. PGB E has since determined that. an AR was
purposely not written for the Unit 1 valves because of a miscommunication between
R&DS and Secondary Engineering.
On November 6, 1995, when a Unit 2 tracking
document was generated,
RBDS personnel were informed that the Unit 1 testing had
b
been completed.
RBDS then concluded that the gathering of all AVKtest equipme t
ias information was also completed.
Consequently,
RBDS did not generate a tracking
ip
en
document for a task that supposedly had already been completed.
CORRECTIVE STEPS TAKENAND RESULTS ACHIEVED
PGB E has addressed
the operability of MSSVs without valve-specific CFs in
Operability Evaluations (OEs) 96-04, "Operability of Main Steam Safety Valves
that Do Not Have Valve Specific Correction Factors," and 94<2, "Operability of
Main Steam Safety Valves with Potentially High Initial LiftSetpoints." These
OEs conclude that using AVKssuggested
CF (mean seat area) to set the valves
is appropriate and that the plant can operate safely.
2.
PGB E willcollect AVKequipment bias data for the remaining 11 valves during
1R8.
PGB E has generated
a commitment tracking document to track this
commitment.
I
4.
A nonconformance report has been generated to address missed commitments.
5.
PGB E has appointed a project manager to specifically manage the MSSV
Augmented Testing Program.
Reca
in
Current MSSV Commitments
Units 1 and 2 MSSV augmented testing involving one'steam header at a time will
be conducted on a staggered test basis to provide varying time periods between
header tests.
Ifthe MSSV lifts are not within +/- 1 percent, but are within +/- 3
percent (+3/-2 percent for the lowest set MSSVs), the valves willbe reset.
Should a valve or valves liftoutside the+/- 3 percent tolerance, the valve(s) will
be returned to +/- 1 percent tolerance.
2.
It is intended that augmented testing provide information on the time
dependency aspects of the high initial liftphenomenon observed during past
on-line testing.
This testing willprovide a set of on-line test data for each of the
five valves on each lead prior to the eighth refueling outages (1R8 or 2R8) on
the respective units. However, ifthe results of the augmented testing on a
specific lead indicate that test expansion to other leads is necessary
to conform
to TS requirements, test expansion willpromptly occur, and PGB E willtreat the
results obtained for operability evaluation as ifthese tests were TS
Enclosure
PG&E DCL Letter 96-167
surveillances.
Should testing'of valves on additional leads be necessary,
the
further expansion willpreclude PG&E from collecting the time-deperidency data
needed for analyzing the high initial liftphenomenon.
3.
PG8 E's efforts to determine the root cause of the high initial lifts is on-going.
(Reference PGBE letter DCL-95-241 dated November 1, 1995, "Main Steam
Safety Valve Augmented Testing and Inspection Program.") A summary of the
results of the root cause evaluation activities and corrective actions willbe
providad upon completion of the analyses and evaluations.
4
PG&E willtest all MSSVs upon return to Mode 3 (Hot Standby) from any Mode 5
(Cold Shutdown). This testing willcontinue until the analysis is complete and
corrective actions (replacement of valve seating material, etc.), ifany, are taken.
NRC concurrence for discontinuation of this testing willbe obtained.
5.
Commitments made at Enforcement Conference on July 1, 1996:
a)
Regarding "AVKTest Methodology Validation," PG&E willcomplete the
uncertainty analysis.
b)
c)
d)
e)
PG&E willcomplete the seating materials replacement evaluations.
PG&E willconduct MSSV-specific performance evaluation.
C
PG&E willconduct potential MSSV Parts Improvement Program.
i
PG&E willmonitor industry MSSV experience.
f)
Technical staff training:
1)
As one of the "CF Corrective Actions," PG8 E willreview
appropriate licensing commitment implementation and NRC
communication requirements.
2)
As one of the "Conduct of Test Program Corrective Actions," PG&E
willconduct training on the missed commitment event focusing on
technical data'evaluation,
formal communication, and compliance
with commitments.
3)
As one of the "SV Operability 8 Timeliness Corrective Actions,"
PG8 E will conduct training on operability determination
procedures.
(Partially or fully completed as this response
is
prepared.)