ML20098B073
| ML20098B073 | |
| Person / Time | |
|---|---|
| Site: | Grand Gulf  |
| Issue date: | 07/25/1984 |
| From: | Kammerer C NRC OFFICE OF CONGRESSIONAL AFFAIRS (OCA) |
| To: | Rehn T NRC OFFICE OF THE EXECUTIVE DIRECTOR FOR OPERATIONS (EDO) |
| Shared Package | |
| ML19272B132 | List: |
| References | |
| NUDOCS 8409250395 | |
| Download: ML20098B073 (1) | |
Text
[:..s uu%,
@Fn&WTJRAU' 1
Action:
Denton, NRR l
UNITED STATES
- 8 k
NUCLEAP. REGULATORY COMMISSION Suspense: 8-3-84
- y WASHINGTO N. D. Cl 20355 Q's 1, 2, 3 and 6 correspond
- :
,, g.
July 23, 1984 Q's in Udall 's 7-26-84 letter noted in margin.
O's 4 & 5 (r numbered 16 & 17)should be answered as part of the Udall letter.
j Cys: Dircks Roe Rehm Stelic DeYounc MEMORANDUM FOR:
Tom Rehm, Ass.istant to the ~.cutive Director for O'Reili Operations GCunnir.'; l FROM:
Carlton Kammerer, Direct r Office of Congressional '. nirs
SUBJECT:
REQUEST FOR INFORMATION ON GRAND GULF The majority staff of the Committee on Interior and Insular Affairs (Chairman Udall) has requested that we respond to the following questions concerning the NRC's review of Grand Gulf.
p/ fR 1.
Has NRC reviewed past inspection reports and/or LERs to detect
~
patterns of errors and subsequent corrective actions and to determine if utility management only addressed symptoms or took more exhaustive actions to determine the cause of problems discovered at Grand Gulf?
ffS 2.
Has NRC staff compiled a list of materiaI false statements made by representatives of MP&L?
pp 3.
Is a new SALP report on MP&L's performance being prepared?
h HAS NRC staff prepared a report describing how NRC determined ' '
that Grand Gulf operators were adequately trained?
h I Does NRC staff have a listing of current MP&L managers describing their qualifications and the date of employment at MP&L?
6.
What inspections and/or assessments have been performed by p
MP&L to fulfill criteria 18 of Appendix B of Part 50?
4
'E' 6/C e
i s. M.. <..!..
e Mississippi Power and Light Company f
ATTN:
Mr. J. B. Richard Senior Vice President, Nuclear P. O. Box 1640 Jackson, MS 39205 Gentlemen:
SUBJECT:
REPORT NOS. 50-416/83-55 AND 50-417/83-09 The NRC Systematic Assessment of Licensee P oard has completed Gulf facility was evaluated for.the period September 11982 through y.
The Grand September 30, 1983.
The results of the evaluation are documen,ted in the enc i
SALP Board Assessment.
This evaluation will be discussed with you at your offices in Jackson, Mississippi on January e
I 19, 1984 tional areas of plant operations, radiological controlThe 3
i uated in the func-lance and preoperational testing, fire protection, emergency prepared s, maintenance, surveil-The SALP Board's evaluation of your performance in the assurance progr,am.
contained in the SALP Board Assessment se functional areas is Ssveral significan which is enclosed with this letter.
evaluation prucess.t weaknesses were identified by the SALP Board during It is the opinion of the Board that these concerns concerted management attention to correct.
The Board require resource commitments have Operational Enhancement been made by recognizes that major you in the appaars that these prog, Program and the Operator Recertification Program. i rams will result It if they continue to receive proper management attentiin significant performance imp rovements on and the necessary resources.
The SALP Board evaluation process consists of categorizin functional area.
of your facilities are defined in section II of the enclosed S g performance in each m:nt.
Any comments which you have concerning our evaluation of th nce oard Assess-of your facility should be submitted to this office within twenty d e performance i
the date of our meeting in Jackson, Mississippi ays following Your comments, if any, and the SALP Board Assessment sures to the Region II Administrator's letter which issues the SALP B, will bo ment as an NRC Report.
letter will, if appropriate, state the NRC position on matterIn additio oard Assess-this status of your safety programs.
s relating to the In accordance with 10 CFR 2.790 (a), a copy of this letter your response, if any, will be placed in the NRC's Public Docu, the enclosure and you notify this office, by telephone, within ten days following th ment Room unless meeting in Jackson, Mississippi and submit written e date of our applicationdto. withhold W 22*0;i5's40111 "Q' a. Ij e,
PDR ADOCK 05000416
t G
~'
./
o L..
~
Mississippi Power and Light Company 2
t.
9 information contained therein within twenty days following the date of our meeting.
Such application must be' consistent with the requirements of 10 CFR 2.790 (b)(1).
Should you have any questions concerning this letter, we will be glad to discuss them with you.
Sincerely, l
i Richard C. Lewis, Director Division of Project and Resident Programs i
j SALP Board Chairman
Enclosure:
l SALP Board Assessment for Mississippi Power and Light Company cc w/ encl:
Ralph T. Lally, Manager of Quality Middle South Services, Inc.
J. E. Cross, Plant Manager bec w/ encl:
NRC Resident Inspector NRR Project Manager, NRR D. S. Price, RII i
C. A. Julian, RII 4
~.k &
y,,..
F
)
y N'^N. RII, RIJp" RI RI RL(hS-h RII HCC.etC MC
(,
Gu 0 rice:ht CJuT
..ce PStohr JAG shinski RClewis 3
1/,,
+--,r-,--,,,--..-
a-
~
11 2.
Radiological Controls 9
a.
Analysis During the evaluation period, reutine inspectiors were performed by the resident and regional inspection staffs.
NRC inspection effort in this area was primarily directed towards preoperational and startup procedures; tests of the radwaste systems; training and qualifications of personnel; and licensee respon:e to LERs, Inspection and Enforcement Information Notices, and NRC inspector identified concerns.
The licensee was responsive to the inspection effort.
No major weaknesses were identified in the radiation protection program.
The licensee has maintained a training prograr.. for the health physics technicians and has established a qualification, testing, and acceptance program for contract technicians.
These programs have been instrumental in upgrading the technical competence of the health physics staff. The actual experience of the health physics and chemistry technicians in nuclear power plant operations was low.
However, the qualifications of the health physics staff were acceptable and met regulatory requirements.
The effectiveness of the radiological control program has not been tested, as the plant has not operated above the five percent power level. Considerable work remained to make the radioactive waste handling systems fully operable. However, the licensee had the capability to dewater radioactive resin wastes, and solidification capabilities were available through contract services.
Modifications of the radwaste facilities were underway to provide additional storage capabilities of liquid wastes, to provide better sampling capabilities, and to achieve ALARA goals.
It was not apparent, however, that significant progress was being made on these radioactive waste handling systems.
Therefore, increased management attention should be given to this area to assure that adequate capabilities are available prior to full power operation.
The violations identified during the evaluation period were not indicative of a programmatic breakdown in the radio-logical safety program.
The two violations were:
(1) Severity Level V violation for failure to collect samples at the required frequency to make up the monthly composite sample of the liquid waste discharge basin in accordance with the technichl specifications.
(2) Severity Level V violation for failure to post an area where radioactive materials were stored.
t L'
12 b.
Conclusion I
Category 2 c.
Board Comments Performance in this area was evaluated as Category 2 during the previous SALP assessment. Continued management attention should be devoted to this area to ensure the successful completion and operation of the waste management system.
Implementation of the radiological controls program, during power operations, will be closely monitored by NRC to determine performance trey.s.
No decrease in licensee 1
management attention, in this area, is recommended.
f 3.
Maintenance a.
Analysis a
During the evaluation period, the area of maintenance was routinely inspected by the resident and regional inspection staffs.
A large portion of this SALP review period was spent in a maintenance outage.
During the outage, modifications t
necessary to improve plant performance and reliability, and correct design deficiencies were installed.
Management involvement with the actual conduct of maintenance in the plant was minimal. There appeared to be a need for Itcensee management to review the manner in which the maintenance i
department conducted safety-related activities, and to ensure i
that a clear line of authority, direction, and responsibility existed.
Licensee schedules, generally, were overly optimisitic,.
i which required them to be frequently reviewed, revised, i
and reissued.
Being overly optimistic imposed an apparent schedule pressure on the plant maintenance technicians.
However, there appears to be an adequate staff to perform the required tasks, if adequate time has been provided for accomplishment of the tasks.
Administrative procedure training was completed during the evalulation period just prior to the repair effort of the fire damaged diesel generator. This maintenance training was i
not totally effective to assure that the maintenance staff would meticulously adhere to procedures.
Minimal management involvement, unrealistic schedules, and the lack of effective training were the major contrioutors to i
the short-cuts of technical and procedural requirements taken i
by maintenance personnel which resulted in the significant number of violations identified below.
These problem areas e
,.. - +..
.~.,,-n-.-,--
n-,.-~w,.
,,.-,,,c----.r.--._,,.,,,wrm,-
,.mp ww
-,nen.w,w,n.-,-e-n.-
i 13 i
were readily apparent during the repairs to the fire damaged diesel for which violation 3, below, was issued, as well as
- the proposed assessment of a civil penalty for repetitive violations for, failure to follow procedures, and control temporary alterations.
The licensee's performance in some areas of maintenance did indicate management involvement and responsiveness.
For example, when the generic aspects of seven Hydraulic Control Unit (HCU) solenoid failures were questioned, management involvement and followup resulted in the complete overhaul and cleaning of the entire HCU system.
Six violations and one deviation were identified as follows:
(1)
Severity Level IV violation for failure to follew procedures for the control of maintenance work.
(2)
Severity Level IV violation for failure to provide a procedure for smoke testing of the control room, and for l
failure to provide cleanliness controls during maintenance on safety related systems.
1 (3)
Severity Level IV violation for fa11ura to obtain proper authorization for safety related equipment repairs, failure to perform a safety evaluation, and failure to properly authorize a temporary alteration.
(4)
Severity Level IV violation for failure to provide adequate justification for determining that no unreviewed safety question existed on temporary alterations to the Division III diesel generator.
~
(5)
Severity Level V violation for failure to control measuring and test equipment used on safety related i
systems, and for failure to properly mark restricted use equipme.nt.
(6)
Severity Level V violation for failure to complete masonry wall modifications in accordance with drawing requirements.
L (7)
Deviation for failure to provide maintenance personnel training.
i 1
p I
O c
w r, -,,
..------.-,---,e--...
,--.-m_r-..------..-,r-
14 b.
Conclusion s
Category 3 c.
Board Comments Lack of sufficient licensee management involvement in maintenance is evident.
The use of overly optimistic l
schedules has, at times, been a contributor to poor maintenance practices. This scheduling problem was discussed in the previous SALP assessment and continues to be an area of concern.
Increased management attention should be devoted to this area by the licensee.
Increased NRC inspection activity should also be peiformed in this area.
4.
Surveillance and Preoperational Testing a.
Analysis During the evaluation period, the area of surveillance including preoperational testing, was inspected by the resident and regional inspection staff at a level greater than normal, based on the weaknesses identified in this area du~ ring the previous SALP assessment.
1 Surveillance A special team inspection was conducted in this area from September 27, 1982 to October 8, 1982.
This inspection revealed that the procedures in use failed to properly implement all technical specification requiremerits; several required surveillances were not being performed; significant weaknesses existed in the administrative controls for surveillance procedures and changes; and there was no formal quality assurance audit program in this area.
The licensee met with the NRC staff on October 14, 1982, to discuss the actions neces:iary to correct these deficiencies.
As discussed above under Licensee Activities, NRC issued a i
Confirmation of Action Letter on October 20, 1982 documenting that the licensee had committed to revise the surveillance program, revise technical specifications, retrain operations personnel in technical specifications, and perform a quality assurance audit of these corrective actions.
The licensee completed the corrective actions as documented in their letters to NRC of August 29, 1983, September 1,1983, and September 13, 1983.
The NRC staff inspected the corrective actions which were required to be completed prior to reactor restart.
Additional actions remain to be completed during l
power escalation and/or in accordance with long term commftments.
i m
_.,_.-.-~-- -
4v-.-
_-r,
,---,_--,.m.__.-.,,.
y.
.,,.. ~,..
-.mm.~_
i-15 To distinguish revised procedures resulting from this effort, all newly approved surveillances were designated as Revision
- 20. A massive licensee effort was put forth in this under-taking.
Procedures were prioritized as tc their impact on plant restart and those required for recriticality and i
low power testing were reviewed first.
Some surveillance procedures which would not be needed for some time, such as refueling, remained to be reviewed and revised.
NRC has inspected samples' of the revision 20 surveillance procedures and found them acceptable prior to restart.
The containment isolation valve local leak rate test surveillance program was one area of concern.
During NRC inspections the procedure for local leak rate testing was determined to be inadequate because valve alignments were not included to provide objective evidence 6n how the local l
testing was performed.
During preoperational
- testing, 2
containment penetration drawings were marked to indicate the proper valve alignment for testing.
At that time the licensee stated that these drawings would be used to develop valve alignments for the future surveillance procedures. The approved procedure did not, however, include this information which had been developed from the preoperational testing effort. A violation, item 6 below, was issued. The liccnsee 4
conducted a thorough review of the previous containment leak rate tests prior to recriticality.
Retests were performed wnere appropriate.
j Preoperational Testing During the licensee's performance of the Loss of Coolant Accident simulation, coincident with a Loss of Offsite Power (LOP /LOCA), a design defect was discovered by the licensee in its custom. load shedding and sequencing (LSS) panels.
The licensee promptly modified the equipment to correct the design problem.
The licensee's program for modifications i
4 and tests of the LSS panels was found, during routine NRC inspections, to be effective and performed in accordance with appropriate administrative controls and procedures.
During this same LOP /LOCA test sequence, the license.
uncovered a number of problems with improper operation of Emergency Safety Feature ~ (ESF) valves and with the per-i formance of the Division I diesel engine.
These problems, most of which were not related to each other, were corrected and the equipment was tested in an orderly manner.
During l
testing a fuel line rupture caused a fire in the Division I diesel generator.
A valve in the fire protection deluge system initially failed to function. After recovery from the fire and completion of repairs,'the diesel generator was run continuously for seven days to demonstrate reliability.
4 O
s. -- -- -
~ - - ' ' ' ' ' ' = ' ' ' ' ' ' ' ' " " " " ' ' ^ ' ' ' ' ~ ' ~ ~ ~ ~
16 During the evaluation period the conduct of preoperational testing and the test results were reviewed and analyzed by 8 the resident and regional inspection staffs. One such test was the thermal expansion test performed subsequent to issuance of the low power license, during non-nuclear heat up. The test procedures were technically adequate and the licensee's performance of the thermal expansion test program indicated adequate prior planning, assignment of respons-ibilities, decision making at appropriate levels, involvement of appropriate personnel, and understanding of the issues.
Test results were reviewed and discrepancies evaluated and/or corrected prior to continuing heat up.
In addition, the licensee performed the drywell bypass leakage test.
The test initially failed to meet the acceptance criteria because construction perstnnel had opened two previously sealed electrical conduits penetrating the drywell and had not resealed these leakage paths.
This indicated a failure to monitor changing plant conditions and review these conditions relative to pre;oerational test requirements.
Nine violations, involving multiple examples, were identified as follows:
(1) Severity Level IV violation for failure to perform relay calibrations within the required frequency.
(2) Severity Level IV violation, with nine examples, for failure to provide adequate acceptance criteria for tests.
(3) Severity Level IV violation for failure to perform a required surveillance procedure.
(4) Severity Level IV violation for failure of the containment and drywell ventilation exhaust radiation monitoring system to meet the technical specification trip logic requirement.
(5) Severity Level IV violation for fa11ura to maintain 4
service water valves locked closed as.equired by technical specifications.
(6) Severity Level IV violation for an inadequate test procedure for performing Type C containment leak rate tests.
(7) Severity Level IV violation, with three examples, for failure to provide a procedure to implement technical specification requirelients for safety related valves.
17 (8) Severity Level IV violation for failure to perform an independent verification of a hardware modification.
(9) Severity Level IV violation for failure to perform a j
safety evaluation for test equipment installed in operable safety related systems.
b.
Conclusion Category 3 c.
Board Comments Performance in this crea was evaluated as Category 3 during the previous SALP assessment.
Licensee resources appear to be strained.
Significant problems were identified in this area early in CN evaluation period.
A major program was implemented to correct the problems which should result in improved performance during the next SALP assessment.
Increased licensee management attention should be directed to i
the area.
Although the overall rating in this area is a Category 3, there was a period of time when performance was unsatisfactory.
Ccrrective action were taken to bring performance up to the present level.
NRC inspection effort should be increased in this area.
s 5.
Fire Protection a.
Analysis During this assessment period, limited inspections were performed by the regional inspection staff to review the licensee's implementation of the operational fire protection i
and prevention program.
The licensee's fire protection program adhered to the NRC i
guidelines during the SALP period except for the violations listed below, which have been corrected.
The administrative procedures for control of the fire protection program appear adequate and meet NRC requirements.
Adherence to these procedures, based on limited inspections, appeared to be satisfactory considering that the plant was in transition from the construction to the operational phase.
- However, several temporary construction structures remained within i
safety related areas of the plant.
These structures obstructed the permanent plant fire protection systems provided in the area but were scheduled to be removed or provided with appropriate fi're protection features prior to plant operation.
[,
Maintenance and tests of the fire protection systems were L
satisfactory with the exception of the control room Halon extinguishing system, battery penetration seals, and several i
i.
18 fire detection systems, as noted below.
Also, one of the e pre-action sprinkler system control valves for the diesel i
generator building did not operate properly during a diesel generator fire on September 4,1983. An investigation was in progress, at the end of the evaluation period, to determine the cause of the problem.
The plant fire brigade appeared to be well organized and adequately trained, as evidenced by their response to, and performance during, the diesel generator fire discussed above.
A review of the plant fire brigade training and drill records indicated that each reviewed brigade member had participated in the required training sessions and drills during the assessment period.
Sufficient fire fighting equipment was available to equip the brigade. Maintenance and care of this equipment was satisfactory.
~
Reporting of fire protection discrepancies, for the most part, has been timely and very comprehensive. A large number of fire protection related discrepancies and items for which construction had r.ot been properly completed were identified by the licensee and promply reported to the NRC as required.
For these self identified discrepancies, the licensee responded to the correct limiting condition for operation as specified in the technical specifications.
Staffing for the fire protection program appeared marginal.
Although the fire protection coordinator appeared to be well qualified for his position, sufficient personnel are not permanently assigned to this area to assure that* the program I
will continue to be adequately administered.
Important fire protection tasks were assigned as collateral duties to a nt.mber of different personnel who were not under the control of the fire protection coordinator.
1 The following four violations and one deviation were identified:
(1) Severity Level V violation for failure to implement the technical specification requirements to submit a special report to the NRC on inoperative fire rated assemblies, and failure to assign a designated fire watch for a l
removed fire rated hatch cover and several blocked open fire doors.
(2)
Severity Level V violation for failure to implement the fire protection program in that maintenance and test procedures had not been established for the battery power emergency lighting units, fire detection zotes Z-15 and Z-18, and fire barrier penetration seals.
(3)
Severity Level V violation for failure to conduct the semiannual weight and pressure verification of the
, - -, + - - - -. - - -., _ -. -
--. - + -,,, - -.. _,,,, -
..-s
.,w, e---,
,,,,-,e,y,,-m
,,,-,-,.,.._,,,-m--.-,__m.,.:,--,,,-,,
-,-n-.
l 19 Halon fire suppression system for the control building complex as required by the. technical specifications.
e (4) Severity Level V violation for failure to implement the technical specification requirements to post a fire watch for two blocked open and inoperable fire doors.
(5) Deviation for failure to provide skid mounted portable air compressors for breathing air applications as committed to in the FSAR.
b.
Conclusion Category 2 c.
Board Comments Performance in this area was evaluated as Category 2 in the previous SALP assessment.
Concerns were raf sed by the SALP Board, in the previous assessment, regarding the use of temporary structures, and the minimal staffing level in this area.
These issues continue to be concerns during this assessment.
Management attention should be directed to this area to assure adequate fire protection staffing.
6.
Analysis During this evaluation period, a full scale emergency exercise was monitored, and a routine inspection was conducted.
No violations, deviations or deficiencies were noted during the exercise or inspection. Minor problems were identified during the exercise for followup action.
The licensee has generally shown timely response to NRC identified initiatives.
All deficiencies identified during the emergency preparedness implementation appraisal were resolved, and there were few other concerns identified by NRC inspectors which required action by the licensee.
There appeared to be continued management commitment to the emergency preparedness program.
Senicr corporate management representatives were personnally involved in the annual emergency exercise. Management commitment was also evidenced by the prompt manner in which the licensee filled the plant Emergency Preparedness Coordinator (EPC) position when the former EPC was transferred.
A reorganization of the corporate office resulted in increased visibility and manage-ment access to the emergency prbparedness program. At the end of the appraisal period, key positions in the emergency preparedness program were filled.
Staffing levels appeared to be adequate to handle the emergency preparedness workload.
20 A training program had been established and implemented. The e training. program for key personnel in the emergency organization appeared to be effective as demonstrated by participant performance during the 1983 annual emergency j
exercise.
However, during this exercise, offsite monitoring teams appeared to lack experience in radiological surveil-lance methods. The licensee committed to provide additional training to monitoring team members.
As a result of the licensee's efforts, the emergency preparedness program has been shown to function effectively during the exercise and during an actual event.
During the exercise the licensee properly assessed accident conditions, correctly classified the accident, took prope'r remedial
- action, and recommended appropriate offsite protective actions.
The licensee's ability to evaluate accident conditions and take action was also demonstrated during the assessment period when a diesel generator fire occurred.
Licensee assessment actions, mitigating actions, and accident classification appeared to be consistent with approved emergency plan. procedures.
The incident was properly reported to NRC.
b.
Conclusion Category I c.
Board Comments Performance in this area was evaluated as Category 2 in the previous SALP assessment.
Licensee ranagement attention and involvement in this area are aggressive.
No decrease in licensee management attention is recommended.
7.
Security and Safeguards a.
Analysis i
Routine inspections were performed in this area by the resident and regional inspection staffs.
During this evaluation period, the licensee's physical security program was implemented in accordance witn approved regulatory requirements. However, some weaknesses were noted during the assessment period with regard to electronic security system failures, and personnel adherence to security procedures.
With regard to the electronic security system, the removal of the anti passback features from the computerized access control system has resulted in a significant improvement in the performance of the system.
Regarding the implementation of security procedures, the majority of the violations and reportable events reflected below were caused or contributed to by the failure of 1
k
-l
23 structure, from the Vice President down to Plant
- Superintendents.
In regard to the submittals for equipment qualification, a comparison of the May 20, 1983 submittal to the corrected ' version of August 25, 1983, showed that the earlier version had 366 deficiencies for 1,160 pieces of equipment. The overall decline in quality is attributed to the extensive scope of these issues, and to a perceived limitation of resources in attempting to meet unrealistic schedules set by the licensee.
In response to NRC initiatives, the licensee provided timely responses for those issues which were believed to impact full power licensing.
Specifically, the licensee provided extensive reports within a short time span for the issue?
involving hydrogen control, diesel generator reliability, and electrical relay performance.
In regard to staff requests for prototype details on this, the first Mark III plant, the licensee was very cooperative and made available the resources of their architect-engineer, Bechtel-Gaithersburg.
During meetings with the NRC, the licensee provided appro-priate technical and management level personnel to make the meetings productive.
On the other hand, a number of responses for significant issues were provided months later than the date specified by the licensee, or the licensing action was not completed as anticipated.
In the matter of Safety Relief Valve Test Results, a submittal required by a license condition, the licensee provided the response in May 1983, some seven-and-one-half months after t' e report's known availability.
Other responses, such as issues under review for closure by the first refueling outage (e.g., soil structure in,teraction), and some requests in generic letters, j
were provided three to five months later than expected and l
only then after NRC prodding of the licensee.
In regard to l
late responses to generic letters, the licensee appeared to have an internal problem of routing the generic letters to the appropriate group for timely response.
b.
Conclusion Category 3 c.
Board Comments
~
Performance in this area was evaluated as Category 2 in'the previous SALP assessment.
The magnitude of the licensing activities during the appraisal period appears to have contributed largely to the degradation in rating since the previous SALP appraisal.
Increased licensee management attention should be devoted to this area.
k' 24 10.
Quality Assurance Program e
a.
Analysis During this inspection period, routine and special inspec-tions were performed by the resident and regional inspection staffs.
One special inspection was performed to review Quality Assurance audits relative to a Confirmation of Action letter i
issued by Region II on October 20, 1982.
The reviews performed during both the special and routine inspections indicated that the licensee's Quality Assurance (QA) audits j
were complete, timely, and thorough. Corrective actions for audit findings have improved.
The Plant Manager, Operations i
QA Manager, and applicable plant personnel freely communicate to reach mutual agreement on audit finding corrective actions.
This open communication has improved the I,
identification of audit problem root causes; consequently, corrective action is more meaningful.
I Procurement activities were generally well controlled and documented.
The licensee's responsiveness and corrective actions on previously identified NRC concerns improved in i
some areas.
A special design control inspection was conducted during this assessment period.
Due to the large backlog of design changes, the licensee established a design change task force j
to specifically identify the status of approximately 2000 outstanding design changes.
Management controls were considered adequate to assure proper completion of significant design changes prior to plant restart.
The NRC conducted extensive interviews with design change task force l
personnel. Drawing control, procedure updating, and *. raining l
in selected design changes was ongoing.
Improvements have been made ir. the control of plant drawings, and the licensee was devoting significant resources to this 4
~
task.
At the end of the appraisal period, however, a problem still existed with the inability to provide legible copies of control drawings for use in the control' room and other important work stations.
The delay in solving this problem was stated by the licensee as a logistical' difficulty in obtaining legible drcwings from vendors. This matter was identified for future followup during operational readiness
~
l inspections by NRC.
l Although the audit program performed by the Quality Assurance staf.f appeared complete, NRC was' concerned about the overall effectiveness of the operational Quality Assurance program.
An unusually large number of significant problems were identified by the NRC and licensee in the surveillance
-nw-,-t,-mv-c,--,,-------,,--wre-,e-.-,,wm-,n.,
..-na,--
,,--n
-.,, - -, - - +
-,,.-~.w.-,
w v
-~~..-,mv
25 testing program and in the operator training program during the appraisal period.
These problems had either not been i
' identified by the Quality Assurance program or they had not been pursued in the manner necessary to result in satis-i
]
factory corrective actions.
Criterion XVIII of 10 CFR 50 Appendix 8, and the -topical quality assurance program require
{
planned and periodic audits to verify compliance with all aspects of the Quality Assurance program and to determine the i
effectiveness of the program.
Surveillance testing and operator training are important aspects of the overall quality assurance program, but audits by the quality assurance staff did not reveal the lack of compliance with NRC requirements in these areas.
A precise cause for this lack of effectiveness was not identified.
However, licensee management attention is needed to reevaluate the scope and i
depth of quality assurance audits to assure a more meaningful overview of NRC requirements.
Another apparent weakness in the implementation of the operational Quality Assurance program involved the line organization's direct responsibility for quality. The Plant -
i Quality section' was responsible for the quality control j
function and reported to the Plant Manager.
They performed most of the direct observations of plant activities and 4
therefore played an important role in the overall quality program.
Quality Assurance, in turn, audited the Plant Quality section but Quality Assurance did not routinely observe the performance of licensed activities in the field.
Violations 1 and 7, described below, were cited against performance of Plant Quality section personnel.
Violation 1 i
was caused by repeated time extensions granted by the Plant j
Quality section for the completion of corrective actions for identified problems.
An example of a documented deficient area which.was given repeated extensions, involved training discrepancies and, in particular, incorrect statements submitted to the NRC on operator licensing examination applications. -These problems were documented in January 1983 j
and, as of August 1983, month-by-month extensions of correc-tive action deadlines were granted by the Plant Quality section to the training departrent.
Had plant and corporate j
management been responsive to the problems identified by the i
Plant Quality section, these training deficiencies could have l
been promptly corrected and may not have resulted in the j
critical path delay to plant startup.
i A further NRC concern relates to the frequent uscage in' plant procedures of the words "should" and "must" as substitutes i
for "shall". These permissiv'e verbs are often used inappro-priately in procedures such that plant personnel are not i
provided with sufficient guidarft:e as to the conser.vative action which should be taken. Violation 7, below, was issued i
as a result of a failure to implement regulator.) requirements t
,,,.r,-
m-.,,,,,-.w,,n--,,--
--,-c-,~c,-,v,,-n
,,,-c -
,wa -w,rv.--m-
---..<~,,----en-n.yn,,e-
-, - - ~ -
-.,,,n, an,,.
,+-
i 26 involving the use of these verbs.
Licensee management attention should continue to be applied to this area tn
, resolve this concern.
Seven violations were identified as follows:
(1) Severity Level IV violation for failure to correct conditions adverse to quality in a timely manner.
(2) Severity Level V violation for failure to issue an audit within the technical specification required time frame.
(3)
Severity Level V violation for failure to store records properly.
(4)
Severity Level V violation for failure of responsible personnel to initiate an incident report to management as required by procedure.
(5)
Severity Level V violation for failure to have proce,
dures in place prescribing the methods and duration of the appointment of alternates to the Safety Review Committee as specified in the technical specifications.
(6)
Severity Level V violation for failure to provide prompt and adequate documentation of quality related deficiencies.
(7)
Severity Level V violation for failure to implement regulatory requirements in accordance with the Quality
~
Assurance program.
i b.
Conclusion Category 3 c.
Board Comments Performance in this area was evaluated as Category 2 in the previous SALP assessment.
It appears, on the surface, from a review of the QA program and its implementation at the Grand Gulf facility, that the program is effective. However, when an overall evaluation of the facility's history for this period is conducted, it becomes readily apparent that the implementation of the QA program at Grand Gulf is inadequate to identify problems and/or ineffective in bringing about adequate corrective actions.
Increased licensee management attention is required in this area to assure tL*.t licensee personnel are effective in performing the QA functions as required by NRC regulations.
Increased NRC inspection effort should be directed to this area.'
,p--
m.
-,,.yeym._
,-.,,,,y,,,,-,-.,,,,---n-,_egm..,-m.,..v.m,,,y.-,.,.,
,,,,-,___.g%-,,pw,ww,n.%q..p-,,.yg---,---~
27 11.
ConstructionActivities(Unit.2) a.
Analysis During the assessment period, the licensee has continued the construction of Unit 2,
utilizing a work force of approximately 900 personnel.
The decision to continue construction was based on an investment protection study which showed that extended storage of the equipment already site could best be accomplished by finishing the major on buildings, and by storing the equipment in place.
Inspection effort during the assessment perind has been minimal because of the announced extension and possible
. deferral of the project.
There have been eight inspections of constructicn activities during this period. Three of the inspections were primarily involved with piping and structural installation, and welding activities; and three of the inspections were primarily involved with follow-up of licensee-identified items. The installation of equipment and construction of the buildings appeared to be progressing very smoothly. The licensee has reported that construction work l
was ahead of schedule and below budget.
The use of a small crew and an. extended schedule appeared to have eliminated many of the bottlenecks and other problems which normally occur during a large construction project. There appeared to be a positive attitude displayed by those involved with construction activities.
There was one violation identified during this inspection period involving improper curing of concrete.
The problem was resolved by retraining of the personnel involved, and did not appear to be indicative of a QA breakdown.
Oce violation was identified:
Severity Level V violation for improper concrete curing.
b.
Conclusion l -
Not rated c.
Board Comments L._
An assessment of a licensee's performance in the overall l
c.stegories of operation and/or construction is achieved by appraising their performance in the numerous functional areas that make up the associated overall category.
Since the licensee and NRC activity in the construction functional l
areas was min.imal, insufficient data existed to properly.
evaluate performance in this area.
l
-e,m--.,w e,--. - -. -
28 B.
Supporting Data e
1.
Reports Data
\\
a.
Licensee Event Reports (LERs)
During the assessment period, there were 254 LERs submitted.
The distribution by Licensee Cause Code and SALP Functional Area is as follows:
Cause Code No. of LERs Personnel 52 Design
_19 External 17 Procedure 16 Component 41 i
Other 109-TOTAL 254
~
SALP Functional Area No. of LERs Operations 59 Radiological Control 1
Maintenance 37 Surveillance 61 Fire Protection 70
_ Quality Assurance 17 Other 9
TOTAL 254 Twenty percent of the events were attributed to personnel 1
error, sixteen percent were due to component failure, and forty-three percent of the events were classified as "other".
Of thb' personnel error LERs, fifty percent were caused by licensed or unlicensed operators, twenty-one percent were caused by maintenance personnel, and twenty-eight percent were caused by personnel from other organizations. Of the 1
~, - -
/
gg events classified as "other", fifty-six percent were reported 8 because the licensee was in an Action Statement caused by the propping open of fire doors for the purpose of construction activiti,es.
Overall, the rehtively large number of LERs was due, in part, to the facility having standard technical specification reporting requirements which involved a high level of required reports.
Additionally, the ongoing construction activities at the. plant resulted in many reports that would not have been required had the construction activity been complete.
b.
Construction Deficiency Reports (CDRs)
Twelve Unit 1 CORs, and thirteen Unit 2 CDR1 were reported for this assessment period.
The distribution of these reports into cause related categories is as follows:
Category Unit 1 Unit 2 Mechanical 2
2 Electrical 2
1 i
Design 5
5 Quality Assurance 2
2 Supports and Anchors 1
2 Welding 0
1 Total 12 13 c.
Part 21 Reports Unit 1 - 11 Unit 2 - 0 2.
Investigation and Allegation Review One allegation concerning an emergency preparedness issue was closed during the SALP assessment period.
Additionally, one allegation concerning diesel repair maintenance was ongoing at the end of t.he SALP period; and two investigations, one concerning 1
i missing rebar and one concerning operator licansing applications, were in progress.
e
-,--,,w y,
.,-,_--,_,---.-,-.v
-,-.m,,w,.
..-.,.cm
,.y-m.--
,,_,.,.yy.
--r,---.-.--
.._ y --. -., - -. -. -,, -. - -, - - - -.,.,.
,e,---
30 3.
Enforcement Actions 9
a.
Violations Unit 1 Unit 2 Severity Level III '-
I violation Severity Level IV 24 violaticas Severity Level V 19 violations 1 violation Deviations 6
deviations b.
Civil Penalties There was one civil penalty assessed for the failure to provide positive access control to a vital area.
Subsequent to the evaluation period, one civil penilty was proposed for repetitive violations for failures to follow procedures and control of modifications that occurred during the evaluation period.
c.
Orders None d.
Administrative Actions - Confirmation of Action Letters There was one Confirmation of Action Letter dated October 20,
- 1982, concerning surveillance procedures, technical specifications, training on technical specification require-ments, and quality assurance reviews of regulatory require-ments.
l l
e.
Management Conferences October 14, 1982, Enforcement Conference concerning surveil-j lance procedures, technical specifications, and training on l
technical specifications which led to the issuance of a Confirmation of Action Letter dated October 20, 1982.
November 2,
- 1982, Management Meeting concerning staff attrition, qualification of staff, use of consultants, and overtime.
l l
December 2,
- 1982, Enforcement Conference concerning a security violation.
I l
January 17, 1983, Enforcement Conference concerning failure to follow procedures.
t April 8,1983, Management Meeting to discuss the Operations Enhancement Program.
k_
j
'..s 31 April 20, 1983, Enforcement Conference concerning control of f modifications and maintenanca, and failure to take effective corrective actions in the area of temporary alterations.
May 11, 1983, Management Meeting to discuss changes to technical specifications required prior to startup of Unit 1.
May 11, 1983, Enforcement Conference concerning a security l
violation.
June 2, 1983, Management Meeting related to proposed Physical Security Plan modifications at Grand Gulf.
July 15, 1983, Management Meeting to discuss problems encountered with scram solenoid valves in unit 1.
July 29, 1983, Management Meeting related to the licensee's investigation on rebar in the standby service water basin.
August-12, 1983, Management Meeting to discuss the licensee's actions to satisfy matters of a Confirmation of Action letter dated October 20, 1982.
September 9,1983, Manager.ent Meeting related to changes in the Physical Security Plan.
i September 23, 1983, Management Meeting to discuss Agastat Relay failures, proposed corrective actions, and other topics of current interest.
l l
1 v
i i
e I
e r
w -,
r,
---w
, m v. w, n w.w-.,r-wmme,--v---,
w,, e w m,
,w__
-ww,-we--,ew,-,-a---
-en-,,n,--
,,w,------w,-,-._n a.w-a--,
cp a,
- 5. Ko bin
'o,,
, UNITED STATES NUCLEAR REGULATORY COMMISSION 8
g WASHINGTON, D. C. 20555 b
>j
%'....Y.
JUN 171983 MEMORANDUM FOR:
Karl V. Seyfrit, Chief AE0D/T323 Reactor Operations Analysis Branch Office for Analysis and Evaluation of 0.perational Data THRU:
Stuart D. Rubin, Lead Engineer Reactor. Systems 4 Reactor Operations Analysis Branch, AEOD FROM:
Sagid Salah Reactor Systems 4 Reactor Operations Analysis Branch, AE00
SUBJECT:
TECHNICAL REPORT FOR, GRAND GULF UNIT 1 Forwarded herewith is the subject Technical Report for your infonnation.
As a result of this investigation, no further evaluation of this problem is considered necessary.
Therefore, this technical review is complete.
S id Salah Re ctor Systems 4 Reactor Operations Analysis Branch, AE00 Encl osure:
-As stated cc w/ enclosure:
CHeltenes
/
1 e
O
AEOD TECHNICAL REVIEW REPORT UNIT:. Grand Golf Unit 1 TR Report No. AE0D/T323 DOCKET NO.:
50-416 DATE: Juris 17,1983 LICENSEE:
Mississippi Power & Light Company EVALUATOR / CONTACT:
S. Salah
' NSSS/AE:
GE/Bechtel 4
SUBJECT:
TURBINE TRIP BYPASS DELAY EVENT DATE:
November 24, 1982
SUMMARY
Mississippi Power and Light Company has reviewed the Architect Engineer's request which recommended a design change to initiate a direct turbine trip upon loss of the main circulating water pumos.
According to the request, '
incorporation of the proposed change vould satisfy a GE criterion which requires that the steam line bypass be, ppen for at least 5 seconds following a turbine trip during a loss of condenser vacuum transient.
After 'a detailed review of this issue, the licensee concluded that the proposed design change is not necessary.
The 5 second bypass operation was not a GE triterion but rather a pressurization transient response based upon an assumption of 2 inches of Hg/second vacuum decay rate, considered to be conservative at the time the FSAR was implemented.
DISCUSSION Technical Specification 6.9.1.1.2.h requires prompt notification of errors discovered in the transient or accident analysis or in the methods used for such analyses as described in the safety analysis report, or in the basis for the technical specification that have or could have permitted rea'ctor operation in a manner less conservative than that assumed in the analyses.
. he' lice'nsee reviewed Bechlei'~s~~regilest _to All'isiChalmers Power Systems,IInc'. for T
~
- ~ ~
~ ~ ~
~
~
l
. design change to initiate a direct turbine trip upon loss of main circulating l
water pumps.
In the process of review it was discovered that General Electric in their FSAR accident analysis used a plant condition which is less conservative than the. design change requested, by Bechtel.
In the analysis General Electric assumed five seconds of steam bypass operation after a i
t'urbine trip.
Such a requirement is not explicitly stated in the system l-design specification.
After a further review of this issue, the licensee concluded that the proposed design change is not necessary.
The 5 second bypass operation was not a GE assumption but rather a pressurization transient response based upon an assumption of 2 inches Hg/second vacuum decay rate, considered to be conservative at the time the FSAR was implemented.
l 6h
% '{
YN
e
.2 -
h The loss of condenser vacuum transient is categorized as a reactor pressure increase event.
For the loss of condenser vacuum transient, based on a FSAR assumption of 2 inches Hg/second vacuum decay rate, the steam bypass would be availahle for 5 seconds provided that the bypass is signaled to close at a vacuum of 10 inches Hg less than the stop valve closure.
The results (maximum vessel pressure of 1179 psig and MCPR greater than 1.13) are expected to be less severe than the limiting transient of this category.
The Architect-Engineer determined by analysis that the present Grand Gulf Nuclear Station (GGNS) logic will not allow the steam bypass valve to renain open for 5 seconch because the vacuum decay rate was estimated to be about 10 inches Hg/second.
Therefore, the actual bypass operation perioc is only about one second.
With one second bypass operation, a new analysis for loss of condenser vacuum transient would result in the maximum vessel pressure greater than 1179 psig quoted for 5 second bypass, but still less than the limiting 1234 psig of the load rejection without bypass transient.
FINDINGS
- Actual steam bypass operation followingsloss of Condenser Vacuum transient is approximately 1 second rather than 5 seconds due to vacuum decay rate of about 10 inches Hg/second instead of 2 inches Hg/second.
CONCLUSIONS Technical Specification 6.9.1.1.2.h requirements were fulfilled by the licensee by immediately issuing an LER comparing the differences between the actual design logic with GE's FSAR analyses.
To narrow this difference i
the Architect-Engineer has requested a direct turbine trip upon a loss of main circulating water pump.
In addition licensee has pointed out in their follow up report that FSAR assumes a vacuum decay rate of 2 inches Hg/second compared to real value of around 10 inches Hg/second.
This difference in vacuum decay rate results in one second for the steam bypass valves to remain 4
open instead of 5 seconds.
Since the proposed design change would merely improve the system. behavior for a less limiting transient, it is not necessary to implement the change from the viewpoint of improving overall safety and operating margins.
Loss of condenser vacuum transient is categorized as an increase in reactor pressure event and since there are other more severe limiting pressurization transients, such as, load reject 'without bypass, consequences of loss of 4
condenser vacuum does not limit the GGNS component design.
Therefore this technical review is complete.
REFERENCES:
1.
LER 82-105/OlT-0 1
2 L.ER 82-105/01X-1 '(Supplementary infoVmation) 3.
Grand Gulf Nuclear Station Unit i FSAR Chapter 15.
1
...-....--,,.---,m..-v,
Yt m
o
.. g,,
UNITED STATES
. g NUCLEAR REGULATORY COMMISSION t
a wAsamorow, o. c.2osss
%.,,,/
NOV 151983 MEMORANDUM FOR: Karl V. Seyfrit, Chief AE0D/T334 Reactor Operations Analysis Branch Office for Analysis and Evaluation of Operational Data THRU:
Stuart D. Rubin, Lead Engineer Reactor Systems 4 Reactor Operations Analysis Branch q
FROM:
Sagid Salah Reactor Systems 4 REACTOR Operations Analysis Branch
SUBJECT:
TECHNICAL REPORT FOR GRAND GULF UNIT 1 Forwarded herewith is the subject Technical Report for your infonnation.
As a result of this investigation, no further evaluation of this problem is considered necessary.
Therefore, this technical review is complete.
s S id Salah Reactor Systems 4 Reactor Operations Analysis Branch, AE00
Enclosure:
l
-As stated cc w/ enclosure:
C. Heltemes, AE0D
.-s
-,----..-,--,.-,---_w-c---
.--,,--.-m.-_.e
--,_r
+,,-----.m_
-,,,,v,..----%.-y-r.
.--w,----
c.
o
- o..
[
AE00 TECHNICAL REVIEW REPORT
- UNIT: Grand Gulf Unit 1 TR REPORT NO. AEOD/T334 r
i.
~ DOCKET NO.: '50-416 DATE: November 15, 1983 LICENSEE: Mislissippi 'ower a Light Company EVALUATOR / CONTACT: S. Salah
[
NSSS/AE: GE/Bechtel
SUBJECT:
REACTOR VESSEL DRA!kAGE EVENT DATE: April 3, 1983
SUMMARY
When the RHR mode of operation was changed from Low Pressure Coolant Injection (LPCI) to' shutdown cooling, a drainage path was created from the reactor vessel'to the supprestion pool. This caused 10,000 gallons of vessel water to be drained into the suppression pool before action was taken to stop the flow. The consequences of this event were within tolerable limits because
[
draining was stopped before reactor vessel level reached the low level alarm setpoint.
4 DISCUSSION i
J On April 3,1983, around 9:45 a.m. approximately 10,000 gallons of water drained from the reactor vessel to the suppression pool through the RHR-l-
system. This drainage was caused by two RHR valves (F004 and F006) being open simultaneously. At the time, the reactor was at atmospheric pressure, with vessel water temperature approximately 100*F.
!i' Prior to this incident, loop "A" of the RHR system was lined up for the LPCI mode and loop "B" was lined up for the Shutdown (50) cooling mode. Figure 1 shows this lineup of the RHR system.
In order to change the lineup of loop "A" from LPCI to SD cooling mode the operator went to close valve F004.
Full closure of the valve would normally result in the red valve position r
indicating-light bulb (on the control panel) being extinguished. The i
operator, observed that the light bulb was not illuminated and assumed that the valve was fully closed. However, unknown to the operator at the time, the indicatir.g light bulb was burnedout which resulted in a faulty indication of full valve closure. The operator, 'not realizing this, assumed F004 was fully closed when it was in fact still partially open.
Then the operator
. opened the valve F006 to put loop "A" in the 50 cooling mode.
Opening valve F006 resul'ted in an unintended open flow path from the 4
f reactor vessel to the suppression pool which drained 10,000 gallons of water out of the reactor vessel.
Water drained out because of the higher reactor vessel elevation.
The control room operator noticed the indicated reactor vessel water level going down, and took immediate action to stop the flow of reactor vessel water before reactor vessel level reached the low level alarm setpoint. The water drain rate from the reactor vessel to the sup-pression pool was approximately 18,000 gpm.
Since the LPCI system dis-charges into the reactor vessel above the top of the active core, there was no danger of core uncovery at any time, 2,
pp y,I h l'f (V M s 1
j aJ t
i "This report supports ongoing AE00 and HRC activities and does not represent l
the position or requirements of the responsible NRC program office.
L
-2 There was a similar incident at LaSalle on September 15, 1983.
Water drained from the reactor vessel to the suppression pool when an operator opened the LPCI B su,:foW valve in accordance with the approved surveillance procedure.
The static head of water in the reactor vessel caused water to drain through the check valve.
The check valve failed to seat properly.
This caused the draining of 5,000 to 10,000 gallons of water from the reactor vessel to the suppression pool in approximately three minutes.
The control rcom operator noticed indicated reactor vessel water level going down and ter-minated the drainage manually while the automatic system isolated the primary containment.
FINDINGS For the event at Grand Gulf a burned out indicating light bulb caused a misleadingposition status for the RHR pump suppression pool suction valve which resulted in a brief drainage path from the reactor vessel to the suppression pool.
CONCLUSIONS When the operator changed the moda of operation of the "A" RHR train from LPCI to 50 cooling, a flow path was created from the reactor vessel to the suppression pool.
This caused reactor vessel water to flow into the sup-pression pool.
The main cause of the problem was created by the burned out 1,ight bulb which erroneously indicated that the F004 valve was closed.
As a result of this incident, there were no significant consequences except that 10,000 gallons of reactor vessel water was drained into the suppression pool. The operator took immediate action to stop the flow of reactor vesscl water before reaching the automatic initiation setpoint for the standby low pressure core cooling system.
There was no danger of core uncovery at any time. This is due to the fact that the LPCI line comes into the reactor vessel above the top of the active core.
e
I k,
U
' *J t
9
" ~
C I -
4 6
0 U N
==
2 C 2 2
m L
U m+
v ed 4
O O
3.
.H d Z e
'f Q M
1 3'"'.z.
Z O
4H 3C U
3' i
.c w
O&
C M M e
O U e-t M
3 C A>
M
.h 3
=W Lt M L
A 0 0
,=
% O u
LA M
g C
L es U
8 E
'r
- sy.
g
'A C
O 2
o{
Z e
M C
~
O U
)
\\
ZC
'e ~ -
L*h
)
emM 9
)
Lo C
O d
b I
/\\ O O
La W M
A T
\\
N C
NC A
el sw rw g(~
g o
I
.o 9
o R
.?
e
~-
~
~
_ _E C C %
l A
- s. ox
%= = '4 4
g q
g j
H
.g so c w
[ S N
C Ce 4
4 d S
W W M
L O
O O w SS
}
/
W A
Ls.
= 0 C
ii A 4,.-
p.e.
z
/
o, C <
&=
d
'm 63 N
w.
wm 7
t wc
.4 W
1 6*
m O
>; [* 7
~
UNITED STATES
^
[ [..e,
..JCLEAR REGULATORY COMMISbuY
..% g' h,h[/
~" *
- E V-WA SHING TON. C. C. 20555
. m, - t, _ 108.:
a o..*
1 MEMORANDUM FOR:
v Thomas Novak, Assistant Director for. Licensing
' Division of Licensing FROM:
Frank J. Miraglia, Nssistant Director for Safety Assessment Division of Li, censing SUBJEbT:
GRAND GULF OPERATTNG EXPERIENCE In response to your rea'uest (menorandum 6f Octobe' 6, 1983) the Operating r
Reacters Assessment Branch. (ORAB) has reviewed operating experience during the past year at the Grand Gul# facility and prepared the attached report.
The ORAB review included a survey of reported events at Grand Gulf during
. the pest 15 months (i.e. the low power license period) and a comparison of the event reports with reports from two other'recently licensed RWRs (LaSalle, and Susouehannal filed during their Inw pnwer license ceriods.
The~-
sources of event reports included prornot (telephone) notifications filed'per 10rCFR 50.72 is well as Licensee Event Reports (LEP) recuired by the Technica' Specifications.
Operating reactor evants briafing sur aries were ~
aise examined to identify the more significant events.
AEnD prnvided us with
~
substantial support in obtaining event reports.
In general the review revealed that high number of promot reportable events (10 CFR 50.72\\ nave occurred at Grand Gulf in the past year.
The rate of occurrence of these events has been at least three Times'grerter than that of the two other recently licensed BWRs used for ccmparison.
The large number of pronpt reports are concerned for the most part with inadvertent actuations of engineered safety features.
According to the 50.77. reports, equal nur.bers of these events have been caused by e,quipnent failure and errors on the part of ooerators and technicians.
~
.. Review of operating reactnr ev'ent briefing sur= aries indicates that five "significant" events have been reported for Grand Gulf during tha year.
They include a low ter,erature vessel pressurintinn incideat, e' ectr'rd systen 11'uncticn causing in.=duertent RPS trips, 7 diesel cenera cr
-. ' ire incideht, simultar. ecus ' malfunction of both Transamerica DeLaval diesel gene-ators, and an' operator error yhich resulted in 10,000 gallons of water beine c' rained f ren the rcretor vessel to the suppression pool. The nurher' of sipr.ificant events at Grand Gulf durino the inw pcwer license perind is Figher thar that for the twn other recently licensed EURs consicered in the review.
L3571'.r had only crm event significant encugh to be rernrted 7.i. a brie # ire ar.d f urcuehanna had It should also be noied that the periods of icw pm eJ lic?rse for nMe.
LPSaile anc SuscuehannA were 'nuch shorttr than G-P.nv Gulf.
l,C~/f 3 N 1 W
/
I D.?
~?,)SI
- Thomas M. Novak Based on our review we have concluded that operating experience at Grand Gulf durino the past. fear has been atypical.
Comparison of Grand Gulf experience with that of other SWRs indicates that the period of operation with the low power license at Grand Gulf has been abnormally long (greater than 12 months versus 4 months for Susouehanna and LaSalle) and that the rate of prompt reportable events has been much greater then expected.
Based on discussions with Region 11 we believe that the high rate of reported events is at least in -
part related to the large amount of construction and testing activities which have gone on during the past year.
This construction and testing activity is the result of design changes being implemented at the piant.
The fact that many events which have occurred are related to personnel errors may indicete a. lack of experience, on the part of plaht personne.l.
The rate at which events have occurred at Grand Gulf has rot decreased steadily over the lono term as the plant has moved closer tn comnercial operation.
Rcwever, a sudden she.rp decrease in the rate did occur in November 1983 which' may be attributed to site inactivity following conoletien of inw power testing in October.
On this basis it would be reasonable to expect the incident, rete to continue this decreasing trend as the plant r.nves closer to conmercial operation, and testing and constructinn activities are completed.
We have discussed the results of our review with IE Regier !!, and they have. informed 'us that our conclusions are censistar.t with their cost recent SALP review.
Regi appropriate action,on 11 will continue tn monitor plant perforr.ance and.take s should problems continue to occur at a high rate..
GL E fW' Frank J. MiEIgl17
.t. ~is tant Di're'ctor for Safety Assessnent Division of Licensing 4
Enclosure:
As Stated O
s e
e 4
e 4
h e
o
.,r r.,.v.-
,._-,___,--y
_.y
.__,._y_,
.,.,__,_w.
OPERATING EXPERIENCE REVIEW AT GRAND GULF UNIT el
,r INTRODUCTION I The staff r'eview of operating experience inclu.'ded a.' survey of reported events at Grand Gulf during the past 15 months (i.e. the low pnwer license period) and
.a comparison of the event reports with reports from two other recently licensed BWRs flaSalle and Suscuehanna) fil.ed during their low power ' license periods.
The sources of event reports include prompt (telephone) notifications filed per 10 CFR 50.72 as well as Licensee Event Repbrts (LER) required by the Technical Specifications.
Operating reactor events briefing surr. aries were als'o examined to identify the more significant events.
These briefings are regularly scheduled meetings among NRC management to discuss'recent events at operating reactors.
SURVEY OF EVENT REPORTS In the period between mid-August 1982 and September 1, 1983 160 incidents requiring prompt notification were rep'orted a,s required by 10 CFR part 50.72 One hundred and twenty-two of these events involved plant systems.
The reir.aining 38 events. involved the plant physical security syster.
Th.is review has focused on the non-security related events.
The security related ev'ents wore not considered significant and were expected based on the testing and construction occurring at the plant. -Thirty-five percent (35C) of the non '
security related events have root causes related to operator and technician actiyities (e.g. testing, troubleshoorting).
Ecuiprent problems (mostly electrical) account' for thirty-two (32%) of the events. The direct causes of the renaining oneathird of the events are unknown or not apparent ' rem the brief.50.72 reports.
140st.nf the events involve inadvertent actuations.of safety Jystems with. the plant shutdown (e.g. standby gas' treatnent system, control room fresh air system, reactor trip,, diesel oenerator start).
The average monthly rate at which these' events have been rrported is approximately 10 events / month.
This rate is compared with rates for LaSalle and Suscuehanna in Table 1 and appears to be abnormally high.
Region 11 inspectors attribute the high rate to the large amount of testing and construction goinc on at*the plant.
A review of the data by month does not reveai any particular trend in
..the incident rate.
Data for the past three months shows a rate of occurrence close to the average in September and Octrsber with a sharp decrease in !!cvamber to 3 events / month.
The sharp decrease is attributed tr site ir7.c-'vity folicwing completion of low power tests.
A sterdy reducticr. in 'N rr.te of ocgurrence is e.xpected as the plant nears comnercial cperatier, since cesign changes and associated tests are expected te be ccr.pio+cd.
In the period beginning August 1,1982 and ending July 1, )gM a total of 227 LERs were issued frco Grand Gulf.
The averr.ne r.nrthly rr.te at wF# eh LERs have been issued is shown in table i r.icnp wi+5 cer.parabic-rates 'er tr!alie ar.f Susovehanna.
The Grand Gulf. rate is similar to the rates fnr tr.57.ii.c end Ses c.veha nna.
This is in sharp contr'ast tiith the 20 CFT. part 50.77 reports ciscussed abcve where the Grand Gulf rate 5:as s'ignificartly higher than the other two plants.
Review of the Grand Gulf LERs irdicates that tbnut ene-hr.is" of the reports relate to problens dith fire crotection systens.
These prcblens include many instar.ces of snnke netector altres causec by dust 'rc crmtruction; ando re.5oyal ef fire barriers ior coeurgfdalar;ma SeJutaiu mstrLew
TABLE 1 RATE OF REPORTED EVENTS AT THREE BWR PLANTS DURING-LOWPdHERLICENSEPERIOD Period of Low Rate of Reported E' vents.
b.
Facility n
Grand Gulf 12*
10 21 L
LaSalle 1 4
1 19 Susouchanna 1 4
3 12 ',.
o The study period consists of the first 12 months of the icw pnwer 'licerse period.
The actual period of the low power license will be icncer than.
12 months.
S 9
9 9
O W
8 6
6 D
S t
0 O
e L
--m--
n u
-e-,
v.
,,,,-,,,---,--,vn---
...--,-----,,----n---.
n-e--,.-,---e-
i.
deficiencies.
Other causes of reported events include eouipment pr'oblems and planned entry of }echnical specification action statements for purposes of testing or cons truction.
~
REVIEW 0F SIGNIFICANT EVENTS Significant events which ha've occurred at Grand Gulf during the past year have been identified through a review of issues raised at the regularly scheduled briefings of NRR management on operating reactor experienca.
The review consisted of a review of the Operating Reactor Event Briefing meeting minutes.
For purposes of comparison a similar review.has been performed for LaSalle and Susquehanna for the periods they held low newer licenses.
Events which are discussed at operating reactor event briefings have been subjected.to e screening process in which' five or six two weeks for discussion based on the rsignificant events a,re selected every eview of 100 to 150 events reports during the two week period.
The purpose of identifying those events here is to provide a measure of the severity and extent of significant operational problems.
Durino the Grand Gulf low power license period, five significant problems at Grand Gulf were reported.
Our review indicates.that only one significant event was repnried for LaSalle durino the period of its low power license.
No events were reported for Suscuehanna.
The Grand Gulf events are summarized below.
Violation of RTNdT Heatino limits Durino ECCS In.iection October 5,19F.2 During surveillance testing with the plant in cold shutdewn a high DC voltage spike occurred which initiated an ECCS.in.iection.
Low pressure ccre spray injected and caused the reactor vessel to become water solid (extending to the MSIVs).
The resulting pressure transient violated the Technical Speciffcation on nil-ductility reference temperature, RTNOT.
' ' Reacter Protection System (RPS) MG-Set Outout Breaker Tries l'av 19. lep3 Ir. advertent trippint. of the RPS l'.G-se.1 eutput t eaters has nerurree repetitively
.resulting in isolation of the instrument air system and a reactor scram signal.
The cau'ses of the trips have been ider.tified as thPreal overlCad due to insufficier cabinet ventilation, and low voltage due to vnitage swings while the F.PS but is fed frcm the alternate power supply.
To recuce the number cf r.utput breaker trips the licensee increr. sed cabinet venti.iatien, installed voltage reculators to secoth cut voltage fluctuaticns, and insta11ec' a ren 'statier ciectrical l
transmissien line frem off-site.
In additien instrur.er.
Air.s. stem, isolation relays have been re-aligned t'o an interru. table crve supply.
This problem t
- 7..
l.
e 4.
re-occurred in January 1984.
Upward voltage spikes remaining above the setpoint longer than.1 second have caused the protective MG-set output breaker to trip, resulting in de-energization of containment isolation system i
. logic circuits followed by isolation of the RHR system.
The licensee has been unable to identify the source of the voltage spikes.
To correct the problem, the ~1icensee has increased the output breaker delay tine from.1 second to 1.4 seconds.
The new delay time is based on measurements of spike duration and censultation with suppliers of the electrical covipment.
The modification essures that spikes lasting less than 1.4 seconds will not result in a trip of the protective breaker.
between the licensee and Region II. Additional corrective actions are also vnd Inadvertent Reactor Yessel.Drainace Durine Shutdown' Aoril 3,1983 On April 3,1983, approximately 10,000 gallons of water drained from khe reactor vessel to the suppression pool through the residual heat removal (RHR)
This drainage was caused by two RHR valves (F004 and F006) being open system.
simul taneously.
At the time of the event, the reactor was at atmospheric pressure with vessel water temperature approxinat.ely 100*F (cold shutdown' conditions).
The vessel water level continued te decrease until the icw level isolatien signal was received and shutdown coolinn isolation valves closed to terminate the leakage.
Diesel Generator Room Fire Sectember 4,1983 A diesel generator engine fire was caused by a ruptured fuel oil suppky line which sprayed oil on the hot exhaust ma'nifold of the diesel.
.The diesel which caught fire was running at 25, percent load for testing at the. time.
Two other diesel generators were 'not affected by the fire.
The water deluge system failed to function automatically, but was manually retivated to extinguish the fire.
The diesel generator governor and turbo chargers were' damaged.
in addition some electrical equipment in the room suffered, water damage.
E Inocerabil'ity of Delaval Diesel Generatnrs October TS.1053 On Octcber 28, 1923, a Technical Spec'ficatier Ict'.en Sta t en was % rad when two of the three diesel generators becar:0 '.rrnerabic.
The Division i
. diesel generator was inoperable due to gasket faili:re en a lube oil line.
The Division 11 diesel generator became increrable due to a locse base plate nut on the turber.hseor:r which ror.ulted in a trip cf the vibration senser which tripped the diesel.
Corre;tive actinn was taken to repair both diesel-gene ra tors.
Both of the diesel pencrators were rerufacturet by Transarerica Deiaval Inc. (TD1).
TD1 cierel generaters have erer.tly cc e veder,ciese scrut'i; 1.,
l.'
.d~ 't '. ' "u h.e :.i n':*'t fr % r in a trei /'n q i cr.corA+,cr at the Shnraham plant.
Staff review of the Transamerica Delaval c:'esel generater prcblen at Grand Gulf is still onpr.ine.
(
CONCLUSIONS g
Based on our review', we have concluded that ope' rating exnerience at Grand Gulf I during-the' low power license period has.been atypical.
Comparison of Grand Gulf experience with that of other SWRs indicates that the period of operation with the low power license at Grand Gulf has been abnormally long (12 months a months for Susqachanna and LaSalle) and that the ra'te of prompt versus repor' table events has been' much greater than expected.
Based on discussions with Region II we believe that the high rate of reported events is related, at.
least in part, to the large amount of testing and construction activities which have gone on during the past year.
This construction and testing activity is the result of design changes being implemer.ted at the plant.
The fact that many of the events are related to personnel errors'may indicats a lack of experience on the part of plant personnel.
The rate at whi.ch events have occurred at Grand Gulf has not decreased steadily over the long term as the plant has moved closer to consnerical operation.
However, a sudden sharp decrease in the ra'te did occur in November 1983 which may be attributed to site inactivity following completion of the low p6wer test.ing in October.
On this basis, ye believe it is reasonable to P.xpect the incident rate to continue..
this decreasing trend as the plant moves closer to cornercial operation,' and testing and construction activities cease.
Should an abnornally high rate of ir:cidents re-appear, appropriate actin.'s such es initiatinc a review of
. personnel training programs and plant procer'ures should be initiated *,o identify the root cause of the continuing problem so that necessary corrective, r.easures measures can be taken.
O 6
a 6
e
,a f
e b
e e
b O
9 0
g
f
(
Enclesure 3 UNITE 3 STATES
'/;
NUCLEAR REGULATORY COMMISSION
- 3
E
- I WASHINGTON. D. C. 20555 o
N FEB 2 41984 9
MEMORANDUM FOR:
Chairman Palladino Commissioner 'Gilinsky Commissioner Roberts Commissioner Asselstine Commissioner Bernthal FROM:
William J. Dircks Executive Director for Operations
SUBJECT:
PERSONNEL ERRORS AT SELECTED OPERATIh'G PLA!TS The Office of Inspection and Enforcement and the Office for isnalysis and.
Evaluat. ion of Operational Data were requested by Commissioner Gilinsky's,
staff to provide information on the frequency of personnel errors at selected -
operating plants (i.e., Grand Gulf, Sequoyah, and Quad Cities).
The Commission should understand that the information presented here is strictly a staff effort based on information available to the staff and, has not been verified with the individual licensees.
The !G.C Operatior.s Center data base contains information en events that are equired to be reported under the provisions of 10 CFF. 50.72.
I'.a ny differer.t types of events are reported, inclucin; all plant trips and safety system actuations.
The following characteristics of the IE data base should be kept in mind
~
when using the information presented here:
1)
The information is called in to the llRC shortly after the event, and at that time an accurate determination of the cause may not be available.
2). Corrections to original reports are not reutinely made if later infor.a.
tion would indicate a different event cause.
3)
Eecause the search capability of the system relys partially en a text search routine, some events which involve cperator error may be missed.
This search used " operational f ailure" and " personnel error."
'r.'e beli eve these to be the most frequer:tly used categories for. labeling cperatic.al errcrs.
PhO Ccr.t a ct :
n g g,f y M > r F.'J.
- -:ebdon, AEOD
'*' Y ' '
-1 T
492-4420 G. Lanik, IE m
HTh'e Commissioners ~
Table 1 provides a summary of oar findings.
The tabulated events were reported as operajor errors, personnel errors, or procedural errors.
Some events were judged to affect the combined units.
These are counted separately and not included as Unit 1 or Unit 2 events.
Table 1 Personnel Errors Reported to the NRC Operations Center 1983*
' Personnel Errors Site Total
. Quad Cities, Unit 1 4
lQuad Cities, Unit 2 -
1 Quad Cities (both) 4 9
Grand Gulf, Unit 1 27 27 Sequoyah, Unit 1 6
Sequoyah, Unit 2 3.'
Sequoyah(both).
1 10
-These reports. are from calendar year 1983.
'~
In addition, A'EOD searched the'. Sequence Coding and Search System (SCSS) for LERs from. Grand Gylf, Quad Cities, and Sequoyah that stated or implied that a personnel action was involved in the event.
.Etcause of the extensive amount of information from each LER that is ' coded in-the 'SCSS, 'i was r.ot necessary.to rely on text searches for particular words (e.g., "pers'ennel error") or to rely on the data coded by the licensee on the LER fo?m.
Thus, if the LER text expressly stated that a " personnel error" occurred, or. if the LER implied that a personnel ' error occurred (e.g., " Inadvertently he operated an incorrect valve"), the information was coded into SCSS and was captured by the subsequent search.
The results of this search were manually reviewedlto identify personnel errors
.!that* could be 'attri,buted to plant personnel (e.g., design errors and f abrication/
canuf acturing errors were excluded).
A rather broad definition of " personnel errcr" was used which included both errors of cr missica (e.g., inadvertent operation of the wrong valve) and errors of omission (e g.., missed surveillance requi reme.nts).
' The results of this analysis are sumcarized in Table 2.
D 0
a
. The Commissioners
-3' Table 2
~
Personnel Errors Reported in' LERs
~
Personnel LERs Plant / Unit Period Errors Received Quad Cities, Unit 1 1983* -
7+**
36 -
- Quad Cities, Unit 2 1983*
4 20
' Grand Gulf, Unit.1 1983*
58 162 Grand Gulf, Unit 2 1983*
0 0
Sequoyah, Unit 1 1983*-
18 85 Sequoyah, Unit 2 1983*
9 64 Sequoyah, Unit.1
. June 1, 1982-7 90
~
June 1, 1983**
Sequoyah, Unit 2 August 1, 1981-18 "61 August 1, 1982**
Some LERs for 1983 have' not'yet be'en received and added to the data base.
However, the period is essentially the same for all units.-
First year of commercial operation.
~
Many of the personnel errors reported to the Operations Center were i
also reported in LERs.
Therefore the numbers in Tables 1 and 2 should not be added.
Clearly from Tables 1 and 2, Grand Gulf has reported nore personnel errors than the other ' units analyzed.
However, care should be taken,in reaching conclusien$
frca this data.
As the ACRS discussed in Appendix E to RU?.EG-0572 (enciesed) there are many reasons for non-randemness (e.g., outliers) in operational data, including differences in reporting requireraants, differences in reporting philosophies, etc.
It should be noted that many of these differences,have been reduced by the recent publication of 10 CFR 50.73, " Licensee Event System"-
and 10 CFR 50.72, "]cmediate Notification' Requirement for Operating Nuclear Reactors," wnich became effective on January 1,1954.
In addition, a review of a count of personnel errors does not consider the severity of the error or its cohsequences.
For example, many of the errors reported by Grand Gulf were missed surveillance requirements that did not directly affec plant cperation.
, Finally, because of the time available to prepare this analysis and the size of
.the computer prirtout, we were not able to make copies of the printout.
Consequently; the prirtouts have been provided (separately) only to Ccemissioner Gilin, sky's office and have not been provided to other interested parties and have not been 9
-,e.-
n
The Cc=.issioners 4
retained by the staff.
If other interested parties want a copy, copies can be made from the enclosed original, or the search strategy can be rerun on the compu additional printouts produced.
e t
O l
.E,'3, $b 5 !. TIE. 14 $Q William J. Dircks Executive Director for OpeYations
Enclosures:
As stated cc w/ enclosures:
g a
e
?
[
e s
e e
a j'
S l
S 9
9 4
4
- o. $
O 8
a O
e 4
9 6
9 e
O e
r 1.
I o
I.
t.
ASPENDIX E y
STAT I ST IC A.L ANI.L YS I S C.:
LERs:
A.TRI AL STUDY O
A e
W y
3 O
g
i
/
i introduction
- Approximately 8700 LERs were submitted by the licensees of U.S.
- commercial
- . nuclear _ power plants during the years 1976, 1977, and 1978.
For several l
reasons, the number,of' LERs. varies.*from unit to, unit.
These variations are l
Important, because, rightly or wrongly,,,they are *of ten ' viewed by government '
l agencies and the public as indications of relative salaty.
Whlis such variations may be. Indicative of actual di f f erences,in saf ety among nuclear power units, they,may have, other explanations, it is therefore Icportant to understand
- all possible explanations and their centributions to variations in the numbers of LERs from unit to unit.
Certain dif ferences in the frequency of submission of LERs from unit -
to unit will occur as a resul.t of the apparent random nature of the events boing reported.
Because.of.this "raridomness", it is *possible--in f act, probable--that, even among identical nuclear power plant f acilities,wlth Id:ntical f ailure probabilities, there wilI be variations in the reporting rate for LERs.
In rgality, however, variations beyond those due,to "rando. ness" will frequently be observed.
The r.easons for such non-random var)ations include the facts thatt
-(l)
Technical Specifications and license provi,slons very among nuclear power plant f acilities, because of di f f erences in reactor suppliers, architect / engineers, and constructors, and changes in designs over
. the years.
These variations cause di f f erences in the resorting requira. ants a.cng f acilities.
(2)
There may be a. tendency at, some f acilltles to report events rnere,
readily than at cthers in cases of marginal repertability.
This censideratien pertains to events other then cavleus)"r.eportable '
'c:currences" (Ros), which all licensees must repcrt *.
This tendency can also change with time..
(3)
The c:currence of 'an event may af f ect.the prcbability of future events.
Repair of a f acility component or 1r.preverent of a deficient precedure may significantly reduce the likellhecd of an asseclated event.
On the other hand, Inef f ective corrective acticn icl'Ic< lng an '
event.ay result in its repeated o:currence.
(4)
The ecde cf c;eraticn (e.g., on-line er shutccani af f acts the frecyan:y cf varicus kinds of Inspections and the suscepta llity cf systems to ra ndem f a i l ur e s,.
The a ount of reteter (c<n-time, for example, may
', affect the frequency with hich LERs are submittsc, s
b
- !aa ref6rence list following Chapter'4 E-l e
(5)
Mis. Interpretations by licens' e o.- NRC per sonnel involved in the e
preparation submission, and processing of LERs can affect relat4ve s
reperting if equencies among reactor systems.
(6)
At some' multi-unit power stations,' cuch as Oconge and Browns Fe s
events which involve p'la*nt' systems
. components common to all, units, such as swing diesels and electrical switchyards, are~ filed in the'NRC data bank under the..' docket number of the first' unit.
r.
The actua'l presence of more safety-related deficiencies in a system at (7) an Individual facility should result'in more frequent submission of LEP.s.
Differences in the numb'er of LERs due to this cause would be a measure of relative safety.
Although the above f actors af f ect the frequency with which LERs are reported, their ef fects are of ten relatively small.
Frequently the variations produced by these ef f ects are too small to be distinguished from these occurring on a random basis.
For example, the Point Beach lluclear Staticn in 1976 had 11 reportable occ,yrrence LERs for Unit I and
,16 for Unit 2.
Does this necessarily Indicate that one or a combinatio'n of,
the caeses listed above produced this dif ference, or is it possible that a deviation of this magnitude could have been expected if both units had the same average probability of occurrence of reportable events?
Statistical analysis indicates that 11 and 16 in one. year are both consistent with average cecurrence rates In the range of one per 20 days to one per 37.
days (10-18 per year).
In f act, the pair of numbe. s, il and 16, is the mest prcbable one-year outcome for two units with an average rate of one
- e.- 27 days 113.5 par year),
in 1975, the Zlcn Nuclear Staticn had 55 re;crtable c:currence LERs for Unit I and 37 for Unit 2.
In this case, the cavlation in the number of LERs bat een the two units is too large to te attributed solely to random effects, if randemness alone were invci'ved, Unit i pretatly could net have had a r'eporting rate less than ene per 5.2 days 170 per year), while Unit 2,probably could not have had a rate greater than one per 7.2 days ($1 per year),
in, fact, if both Zlon units had Identical prcbabilities of repcrtable events, ibere is no r.cre than ene chance in ene millien that a deviatien this large could c: cur by chance.
Naturally, there are dif ferences between the Feint Ete:h units.
Unit I is 1.c years oncer than Unit 2.
During 1975, Unit 2 prece:ec 111 mere electrical energy than Unit 1.
The results in this example incicate
~
that ene should not necessarily conclude that dif f erences in the rates of
' LER suteissien bet.een the tdo units are significant.
At 21cn, he.svar, cne sr.culd eepect to find that the two units reported at significantly-different rates fc." reascns other than rendemness.
a s
E-2
~
Methodolooy.
Methods from probabrillt' theory can be used to calc *ulate the impact of y
randemness on the distribution of the number of LERs emong identical nuclear p:wer units.
Of ten, probability tables from reference, textbooks are suf ficient to perform the analyses.. Computer.stmulations are necessary for the more complicated analyses.
In interpreting the resulting data, it is important to note sevekal basic facts:
(1)
The numerical size of expected random variations in event rates increases as the average event rate increases.
Deviations of 10 or more are readily expected on a randem' bag ts for an average yearly rate of 100, but are unlikely for en average yearly rate of 20.
The relative size or percentage variation, however, decreases as the average rate increases.
(2)
The chance of large random variations among units increases as the -
. number of units being compared increases.
For two, units with an assumed everage annual LER submission rate of 100, there is only.a smalI chance that one rate vilI deviate by, more 'than 20 frcen the
. average because of randomness.
For a comparison among 30 units, hcwever, there is a good chance that at least one will deviate by more than 20 f. rem the annual everage rate of 100 because of render ness.
A selected set of LERs was used here to demonstrate the application of this methodolo;y.
The sources of the LERs were the.22 EWAs that achieved ce.mer-clal c;eratien peler to 1975.
Records show 1 hat this grcup submitted a tctal of 27 LERs fer 30-day reportsble occurrences in auxillary peccess syste:s (vring 1976, 1977, and 1978.
Thus, for inn s group of units, the average was about ene LIR of this type for the three-year perloc.
It is first assumed that all units in the gecup were identical with res;ect to their chances of generating LERs of this type.
Further, it is assumed that if a nuclear pc.ar plant experiences a repertable occurrence in an auxillery process syst6m, the chance of another recurrence is unaf f acted.. Throughout. this study a Folsson distr.lbetlen ef.
, events is assui.ed.
Pectability theory Indicates that,.hlle the average Is ene, it is very Onlikely that each Individual unit wculd experience exactly cne, in fact, the prebt.bility that all 22 units would each re;cet
?his r.u.ber is less than cr.e In ten billion.
T h e n.: s t likely. result is
? hat about eight units will have no LERs, about eight will have cne LER, abdut fcur will have two LERs, and about two will nave three LERs.
Further, it is unilhely (Si chance) that any one unit will have six or cre LERs.
Cc.;arlser, to actual LER data sho..s nine units *lth no LERs, s e'ven viih ene LER, t wo with two 1.ERs, one. wlin three I.Eas, t=c.lth fcur g
E-3
1 l
l LERs, and one with five LERs.
consistent Yith the' assumptions stated above.The distribution.of LERs for the 22 SWRs is 1
~
k This axample, does not, prove, however,ttnat the 22 EWRs are identical to ecch other_wi,Th regard to the causes,,of auxiliary process systems failures.
It simply indicates that one should not_ expect 'to find significant dif ferences among these units, even though some submitted as few as zero and others as many as five LERs.
The value of this an,alysis is that,it provides a methodology through which significantly high dcviations can be readily identified among a population of expected random deviations.
Analyses For purposes of this study, the LERs from 67 nuclear power' plants were reviewed.
For purposes of analyses, these were divided into FWRs (total = 42) and BWRs.(total i 25) and each of these groups was further separated into
" older" and " newer", power plants.. In this c.ase, " older" was arbitrari ty defined as those power plants that went into ope' ration prior to 1975 (see Table E-1).
_ Fur this group, all LERs. submitted during calender years 1975.
thr'ough 1978 represent events that occurred during commercial coeration.
Dat a - u.s e d in.these analyses were, based on the NRC computer bank and included' repertable occurrences only.
The,Ros were separated into those required to be submitted en ~ e prompt or two-week basi's and those submitted on a tnirty-day basis.
These were analyzed separately since there did not appe'ar' to be any correlation in the relative ' numbers of each type as reported by licensees at, the 57 ' power plants.
Lastly, the LERs were further separated a:ccrding to the system to which they pertained.
A listing of these systems is shown in Table E-2.
The prima.y goal in the analyses was to. identify significant deviations cr variaticas in the number of LERs repcrted from plant to plant and system to system.
A deviation was c'onsidered to be significant if inere wts c 8T chance or less that it coeld have resulted f rom random variations.
Cenc l us i o_n_s l
~
Cn t*.e basis ci these analyses, the following CenClusiens and/c.* Cb5ervatiOns were made:
(1)
The frequencies of reportable occurrence LERS among the varicus nu: lear
' - pc.er units were signi f icantly di f f erent.
There vere no identifiable groups of reacter units whose members generated the same average netter
. cf. reportable c:currence LERs during each of the ihree years in the study.
e f
I.
E-4 e
,-4
.., ~.,
(
N (2)
Considering,the three-year period as a whole,' 5 units among the 29 older PWRs deviated significantly, from the others.in terms of the total number of two-week R0s.
The numbers of LERs from Calvert Cli f f s-1, Palisades, Rancho, Seco,, and Three Mile Island-l were high; Raine Yankee was low.
The remaining 24 PWRs repo,rted numbers of LERs
' consistent with an average of about 20 per unit for the period from -
~
1976.through 1978.
(3)
For the same 29 older PWRs,. considered year by year, the data.showed that the total number of ;two-week R0s. steadily decreased in each successive year.
The averages were ten per unit in 1976, six in' 1977, and four in 1976. - Significant deviations from these occurred at Calvert
- Clif f s-1 in 1977, Palisades in 1977 and 1978, Point Beach-l in'1978, Rancho Seco in 1977, and Three Mile Island-l in,1978.
All had higher than normal reporting' rates.
Maine Yankee had a rate in 1976lsigni-ficantly lower than normal.
These results indicate that the'high three-year totals for the four units listed in paragraph 2 above were basically due to high reporting rates in just one of the three years, while the rates for the other two years appear to be normal.
(4). Further analysis of the data showed that the high totals of two-week ROs in four of the older PWRs were attributable ~ to abnormally high-numbers of LERs concerning specific systems.
Calvert Cliffs-l had significantly high three-year totals for electric power systems and-for reactor s.ystems.
Palisades. reported high totals for the same.two systems, in addition to engineered safety features.
Rancho Seco reported a high total f or elect.ric. power systems.
Three Mile Island-l had high totals fer radiation protection systems and fcr events -
cl ass' d as " systems code not applicable."
Many of the electric power e
system LERs were related to cif-site power systems and emergency I
diesel generators.
Reactivity centrol systems ware the scurce of l
most of the reactor system 1ERs from Palisades.
}.
l (5)
,e.mong the older PWRs with normal yearly totals for twc-week ROs, some l
nevertheisss reported significantly higher th'an normal totals ci LERs for specific systems.
The number of LERs in reacter systems was higher than normal at Arkansas Nuclear One-1, Ocenee-2 and -3, and H.5. Rcbinsen-2.
The number for Zicn-1 as higher than ncrmal,
ice radiatica prctection systema.
LERs icr electric pcver systems were higher than normal at Fort Calhoun, Occr.se-1 and -3, Fr air ie Island-1, and Turkey Point-3.
The systems mentioned here, hewever, did not cer. tribute significantly to the icial numFer ci LERs, since LERs f rom engineered saf ety f eatures ar.d reacter ccciant l
systems dot.i nated t he twc-week ROs f r om ol der FWRs.
As a result, devi at ions f rot. ncrmal in -he less often re; cried systems did not have a significant i-l pact cn the total
- r. ember cf LERs,icr these plants.
l E-5 l
i
(6)
The data show that newer Ph'Rs', after they achieved conmercial
~
operation, hjd significantly higher LER sutmissi.on rates for two-week R0s than did older PWRs.
The exception was indian Point-3.
As with the older plants, eng'Ineered saf ety features and reactor coolant
. systems were responsible for a large fraction of the LERs.
(7)
With regard to 30-day Ros, there. were no.. identi fiable units among the 29 older PWRs that deviated significantlp from the average totals for the' three-yeaf'per16d.~" If *is possible, however, to identi f y three separate subgroups among the units in this category.
A first~ subgroup includes seven.un'its with an average reporting rate of.about twenty 30-day Ros for the three years.
These were Oconee-2, Point Beach-l and 9-2,. Rancho Seco, San Onofre-!; and Turkey Point-3 and -4.
Anothsr group' had an average of about fort.y'-f ive 30-day RO's for,the three years.
The 10 units. in this group were H.B. Robinson-2, Haddam Neck, Indi an Point-2, Maine Yankee, Oconee-1 and -3, Prairie Island-l and -2, R.E. Ginna, and Three Mile Island-1.
A third group of 5 units with a normal reporting rate o.f about 70 for ine three-year period included Arkansas Nuclear One-1, Kevanee, Palisades, and Surry-l and -2. Signi ficant deviations from these groups occurred 'in 7 units with high reporting rates. "These were Calvert Clif f s-l,' O.C'.
Cook-1, Fort Calhoun, Millstone-2, Yankee Rowe, and Zicn-1 and -2.
is interesting to note that three of the five operating Combustion It Engineerin'g reactors are in this category.
These are Calvert Cliffs-1,
.ert Calhcun, and Millstone-2..
In addition, this category includes' all three ci the older PWRs.having a power level of 1000 MWe er more.
These are D.C. Cock-l and Zion-! and -2.
(5)
The data show that the ene yesi totals for thirty-da~y R0s in older FN?.s were similar to the three-ye,ar totals in that definite subgroups can be identified.
In general, a unit that was in a low or higher reporting subgroup in one year remained in the same subgroup in' later The exceptions were Yankee Rowe, which.was in a higher re-years.
porting subgroup in 1977, but in lower reporti.ng subcroups in the
~
4 other two yeurs, and surry-1 and -2, which vere in a Icwer reportihg subgroep during the first two ~ years but in the higher se.bgrcup in 1975.
Several signi ficant corre,lations were fcund.
These units which tended to remain in the icwest reporting subgrcu;s nevertheless in-creased _their reperting rates for Thirty-day F.Os from year to year.
The sum of their thirty-day and two-week Ros, however, rsmained ess'entially ccnstant in time, since the two-wask RO total steadily decreased during the three-year period.
Large units cf 1000 MWe er more reported higher numbers of 30-day Ros, except when the plant factor for the year was icw (less tnan cne-third)..Later Cc-bestion
' E-s
.4 n -,,.
~g, w
x I
Engineering ' nits (not including Maine Yankee) also submitted higher u
numbers of LERs for thirty-day R0s, except when the plant availability factor was 15w (less than one-half).
~
\\
-(g)
Newer PWRs reported thirty-day R0s at ret s cons,istent with the higher t
reporting subgroups among older PWRs.
(10) The syst' ems most r ssponsible 'for the higher LER submission., rates for thirty-day ROs 1n Combustion Engineering units were auxiliary process systems, electric power systems, instrumentation systems, and steam and power conversion systems.
These ' units usual ly devi ated f rom the_ normal reporting rate for these systems, in large units the systems involving a higher than normal number, of thirty-day ROs were auxiliary process systems, engincered safety f e'atures, instrumentation, systems, and radiation protection systems.
(11) With ' regard to two-week ROs among the 22 older BWRs, eight units deviated f rom the normal reporting rate,d,uring the three year period.
- These were Cresden-2, Duane Arnold, E.1. Hatch-1,. Fitzpatrick, and Peach Bottom-2 and -3, with higher rates than nor~ al and Dresden-1 m
and Lacrosse with lower rates than normal.
The remaining units reported an aver ~ age rate of about twenty-f our two-week R0s for the three-yee'r period.
The ' rate remained constant at about eight per year.
c (12) E. I. Hatch-l ' reported two-week Ros at a comparatively hich rate for each of the three years.
The numb,er of reports pertaining to nearly every system-deviated f rom normal reporting rates for those systems.
(13) Duane Arnold reported two-week R0s at a comparetively high rate in 1976 and 1977.
The sysicms with higher than normal' numbers of reports were related to electric power.
For~Fitzpatrick, the number of two-week RCs for 1976 was high.- This unit also had a high number of R0s in instrumentation systems. - For Peach Bottcc-2 and -3, the nu-ber cf two-week ROs for 1976 and 1977 wasibigh.-
Unit 2 had an
~
abnormally high number of R0s for reacter coolant systems. and steam.
and power conversion systems.
Unit 3 repor' ed a high nc=ber in, t
engineered saf ety f eatures and for cther auxiliary syste s.
D:esden-3 repcrted a higher-than-normal number of LERs in 1977.
Further, this unit reported an abnormally high number cf R0s in electric power systems.
Nine Mile Point-l reported higher.than-ncrmal totals of LERs concerning instrumentation systems.
Quac Cities-l reported a hich incidence of two-week ROs in steam and pcvar ccaversion systems.
E-7 8
e
-y y
m-.1-
,__,_ve,.,m
-,,c,,_~,,.,,_,,y-
_%_.,,.._,,,e
.m3 r
-~,-.7--
(14) Amohg the tMree newer BWRs, only Browns Ferry-3 reported abnormally high numberse of two-week ROs in reactor systems af ter the unit began commercial oper ati on.
~
' (15) Two BWR units, Fitzpatrick and Brunswick-1, reported abnormally high numbers of thirty-day R0s in nearly every system.
As an, extension to. the above, 'LERs,, pertaining,to' set point dri fi[Yere analy' zed using as a data source the ownputer_ bank.at the. Nuclear Safety'
. information Center l(see' Appendix ~ D 'l li)'. ' ~These analyses showed that. fnere was - no. significant deviation in the. total annual LER submittal rate for setpoint dri f t among older BWRs or among older PWRs.
The average rate for BWRs, however, was apprcximately five times as largi as that for PWRs.
Six older PWRs reported rates higher than normal for the thr'ee-year period.
T,hese were Zion-1 and
-2, Fort Calhoun, Millstone-2, Palisades, and,,,
Kewanee.
It 'is interesting to note that three of these are Combustion
, Engineering units.
Among newer PWRs, four units reported at high' rates in
~1978.
These were J.M. Farley-1, Indian Point-3, North Anna-l, and Salem.
Three older BWRs reported set point drif t events at abnormally high rates for'the, entire three year period.
These were Duane Arnold, Brunswick-2, and Nine Mile Point-l.
Six older BWRs reported at abnormally low rates.
Tnese were Big Rock Point, Browns Ferry -1,
-2, and
-3, Lacrosse, and Monticello.
Co amentary
~
This portion of. the study has clearly demonstrated the potential usef ulness of statistical analyses in ' he evaluation of LEP.s submitted by licens~ees.
t Euch analyses make it possible to distinguish deviations' in the numbers of LERs which would be expected on the basis of randomness from those that.
almost certainly wculd not.
The latter can be used as a means for the identi f ication of areas for possib.le f urther inve st igati cas.
While,the deviations noted in this study do not necessarily imply saf ety-related
~
prcblems, they shouid ncnetheless be purseed in or. der to determine the true implications.
~
i
~
lt wculd prcbably be desirable to' computerize these analyses for autcImatic pro:sssinc of reports as they are' legged into the LER data base.
Ut.iliza-
~
tica of the data base in this manner would make it pessidis to catect
. sioni f i cant devi aTions f rom normal.
Further, an automated system could be
. prEgraE.med to obtain detail beyond the syste:n level, in' order to' identify reperting rate deviaticas fcr relevant subsystems and cc ponents.
t
'e l
t
~
E-8 l-l
Table E-l
'e Number of Reportable Occurrence l'ERs frcm Ceanercial Nuclear Power Plants (1976-1978)
GROUP I:
01 der PWRs (commercial operation prior..to 1976) Total = 29 Nuclear Reportable' Occurrences,'
Nuclear Reportable Occurre'nces
,Pnwer Plant-
--30 day 2-weck Power Plant 30-day 2-week..
' Arkansas Nuclear One-1 71 17 Point e.each-l 15 30
~
Calvert Cl1ffs-l 169 35 Point Beach-2 18 20 D.C. Cook-l 147 20 Prairie Island-1 51 17
-For.t Calhoun
,109 24 Prairie Island-2 36 18 H.B'. Robinson-2 53 26 Rancho Seco 17
~,
40 Hadda.m Neck 41 19 R.E. Ginna 44 24 Indian Point-2 57 26 San Oneire-I 19 '
11 Kewanee 75 19 Surry-l 79 19 Maine Yankee 47 6
Surry-2 71 8
Millstone-2 118 21 Three Mile l's l a nd-l 44
'41
.Oconee-1 42 34 Turkey Point-3 24 11 Oconee-2 21 -
26 Turkey Point-4 20 16
.~
,' 'Oconee-3 41 21 Yan'<ee Rc e 9s.
15 i
Felisades 64 55 Zion i 185
.25 Zion 2 122 15-Averace 65.6 22.7 E-9 l
Table E-1. Continued e-GROUP 11:
Newer PWRs (commercial operation after Jehuary 1, 1976) Total = 13
.Nucleer ReDortebie Occurrences Nuclear Recortabla Occurrences Power. Plant 30-day 2-week Power Plant 30-cay __
2-week.
Arkansas Nuclear One-2 21 7'
indian Point-3 85 15 Snaver Valley-1 216 27 J.M. Farley-1 138 23
. Calvert Cliffs-2 135 25 North Anna-l 98 29 ~
Crystal River-3 154 32 S.t. 1. uc l e-I 123 22 D.C. Coo'4-2 95 7
Salem-l 118 6B Davis-Besse-1 220 32 Three Mlle island-2 '42 17 '
Trojan 63 44 Average 116.5
6.8 9
g O
e e
E-lO e
It w
,y..
r m-,
g
-~
~
4
\\
Table E-l' Continued
=
GROU? IIi:
01 der SWRs.fcoxnercial cperatico pcfor to'1976) Total = 22 Nuclear R'eportable Occurrences.
Nuclear Reco'rtabi'e Occurrences Prwer P lant ---
30-day 2-week Power Plent-30-day 2-week Big. Rock Point 105 M
taCrossu 27 10 Browns Ferry-l-55 26 Mi 1:Istone-l.
Bd-
, 27 Browns Ferry 33, '
la Montl.ce 1 Icr 65 30 Brunswick 261:
3'4.
Ni ne Mi l'e Point-l, 93 27 122.
la OystheCreek.-l Cooper 56.
35 Dresden-1 70-ICP Peach Bo.tton-2 1.46. '
, 56 Dresden-2 153
, ST.
l'eech' Sottcen-3 107 5'6 Dresden-3 105 29 Pilgrim-l 103 25 Duane Arnold 120 SB:-
Quad Cities-1 94 31 E.1.. Hatch-!
94 162?
Quad Cities-2 75 "
14 Fitzpatrick 181 41 Yermont Yankee 96 l'8 Average' 102.0 38.0 e
D' E-Il
+-
y...
Table E-l. Continued t
GROUP IV:
Newer Bh'Rs (commercl31 operation after January 1, 1976) Total = 3
' Nuclear Reoortable Occurrences Power. Plant 30- d a y-2-week Browns Ferry-3 58
'12 Br.unsw i ck-l 211 9
E.1. Hatch-2 65 12 Average lLl.3
.i.i, o O
e S
e O
g e
=
O e
e O
e 9
I' e
9 G
0 e
O e
W O
e G
6 E-12 e
q v
w
.,-w_.g.
r.
c-y Table E-2 System Codes for 1.ERs System System I,
Auxiliary Process System's 8.
Other Major Sys ems 2.
Auxiliary Water Systems 9.
Radiation Protection' Systems
.3. - Electric Power Systems
- 10.. Radioactive Waste Manageme.nt Systems 4.
Engineered Saf ety Feaf ures 11.
Reactor Systems 5.
Fuel Storage and Handling Systems 12.
. instrumentation and Control Systems 13.. Steren and Power Convers)on Systems 7.
Other Auxi1iary Systems 14.
System Code Not Applicable
+
e W
9 9
g E-l'3 e
,en e
4,,n en--
HQ%
t
\\
APR 2 5 sg4 AEOD/T406
!!iE:10RA!!DUM FOR: Karl V. Seyfrit, Chief Reactor Operations Analysis Branch, AEOD THRU:
Stu$rtD. Rubin,LeadEngineer Reactor Systems 4, ROAB ThomEsR. Wolf,ReactorSystemsEngineer FROM:
Reactor Systems 4, ROAB
SUBJECT:
TECHNICAL REVIEW REPORT OF Ati IliPROPER SPARE PARTS PROCUREME!!T EVElfT AT GRAND GULF U"IT 1 Ecclosed is a technical review report of.an improper spEre pErts procurement event which occurred at Grand Gulf iluclear Station Unit 1 on September 19, 1983. While the. actual. event was minor, the main concern of proper equipment quality level classification is generic to Grand Gulf and the industry.
However, URC Generic Letter 83-28 sufficiently addresses the problem. Therefore, no additional AEOD/ROAB review and actions are necessary at this time.
Thomas R. Wolf, Reactor Systems Engineer Reactor System 4 Reactor Operations Analysis Branch AEOD
Enclosure:
As stated Distribution:
ROAB RF Dbg -
ROAB SF c,,, f,,,, W 7 /
3 AEllD CF 0
r/<i
- 7. Wolf S. Rubin T. Ippolito C.J. Heltemes RS4/ROAB C/RS4/ROAB rr,cep fi S
D,
\\
AE00 TECHNICAL REVIEW REPORT
- UNIT:
Grand Gulf 1 TR REPORT NO.: AEOD/T406 DOCKET NO.: 50-%16 DATE:
April 23, 19s4 LICENSEE:
Mississippi Power Light Company EVALUAIOR/ CONTACT: T.R. Wolf NSSS/AE:
GE/Bechtel l
SUBJECT:
EVALUATIONOFANIMPROPERSPAREPdRTSPROCUREMENTEVENT AT GRAND GULF UNIT 1.
EVENT DATE:
September 19, 1983
/
Sumary:
Licensee Event Report 50-416/83-147 documents that due to improper quality level specifications, incorrect spare parts.were procured and installed in the control room chlorine detection system.
Review of other industry documents indicates that there may exist a generic problem in the programs designed to assure that proper equipment quality levels are maintained over the plant life span. NRC Generic Letter 83-28 " Required Actions Based On Generic Implications of Salem ATWS Events" properly and sufficiently addresses this problem.
Consequently, no further AEOD/ROAB action is necessary at this time.
Orthn,its/,ri-b% +
u., y v e r w T i o
' W e\\
- This document supports ongoing AE00 and NRC activities and does not represent the position or requirements of the responsible NRC program office.
1
\\
Discussion and Findings Asaresultofaninternalplantauditwhichindicatedthattheremight exist a generic problem in the quality levels specified in spare part procurements, Grand Gulf personnel initiated a review of all procurement documents.
It was determined during this review that spare parts purchased and installed in the control room chlorine detection system had, indeed, been procured utilizing erroneous quality level specifications.
Consequently, on September 19, 1983, the detection system was declared inoperable and a Limiting Condition for Operation was entered.
This condition remained in effect until October 11, 1983.
During this time span properly qualified components, i.e., a gasket, a washer, a spring and an indicator pipe, were purchased and installed. To help preclude similar problems from happening to any other systems and components, procedures were modified to include' the engineering design group in the review of all procurement documents.
This event was documented in Licensee Event Report (LER) 50-416/E3-147 and closed out by NRC Rer, ion II in Inservice Inspection Report 50-416/83-52.
See Refs. 1 and 2, respectively.
A similar occurrence was discovered in September of 1982 at the Quad-Cities Nuclear Power Station. As documented in LER 50-254/82-027 (Ref. 3),
replacement main valve guides and pis. ton rings in the electromatic relief valves for the main steam system were procured as non-safety related. This' 3
problem was attributed to inadequate implementation of procurement require-ments.
One of the corrective action:s taken to prevent recurrence of such a problem was the revising of station procedures governing component classification.
Although not specific to any failure, it is noted in a March 1984 NRC Region IV inspection report for Arkansi.s Nuclear One Units 1 and 2 (see Ref. 4) that similar procedure problems exist.
It is stated in the inspection report that the licensee prese'atly has no documented program to ensure that the maintenance program incorporates the technical require-ments contained in the ASME Boiler and Pressure Vessel Code, manufacturer's 4
technical manuals and instructions, and other sources. The licensee, the report further notes, has proposed a program to rectify the problem.
The Comission recognizes the vital role which a proper operational quality
. assurance program has on the safe operation of each nuclear power plant.
Recently, this was demonstrated in the findings of an NRC Task Force which studied the generic implications of the 1983 Salem Nuclear Power Plant,
. Unit 1 anticipated transient without scram event (Ref. 5).
These findings were transformed into required licensee actions and transmitted to all power reactor licensees and applicants in NRC Generic Letter 83-28 (see Ref. 6).
Specifically, this generic letter requires all licensees and applicants to provide the NRC.with descriptions of their programs which assure that all safety-related system components are identified in all 3
documents used in the plant to control safety-related activities. These activities include maintenance, work orders, and replacement parts.
As presently planned, each submittal will be individually and generically reviewed with individual safety evaluation reports produced and issued.
e
-m-w
-w e
w--,---
- ---,wr--
w,-.--..-
+ -,
,---,n--,--,----
w------<--,--er
-m e-
+w-,-u---,
,---m,-.r,--
,-+--i------
w----e
)
7 g
- o Conclusions The occurrenceg examined.in this study indicate that a generic problem may exist ig the programs designed to assure that proper equipment quality levels are maintained throughout the plant life. Without adequate programs, including items such as replacement parts, proper functioning of essential systems and components cannot be assured.
NRC Generic Letter 83-28 properly and sufficiently addresses this problem.
Actions.taken by each licensee and applicant to respond to this letter should
- be sufficient to preclude this problem in the future. Consequently, it is concluded that no additional AE00/ROAB review of this matter be taken at this time.
References 1.
Mississippi. Power and Light Company (L.F. Dale) letter to Nuclear Regulatory Commission Region II (J.P. O'Reilly).
Subject:
Grand Gulf Nuclear Station Unit 1.
Docket No. 50-416, " Inadequate Quality Level Parts Installed in the Chlorine Detection System".
LER 83-147/03 L-0. October 19, 1983.
2.
Nuclear Regulatory Commission, Region II (D.M. Verrelli) letter to.
Mississippi Power and Light Company (J.B. Richard).
Subject:
Grand Gulf Nuclcar Station Unit 1.
NRC Inspection Report:
50-416/83-52.
December 14, 1983.
3.
Commonwealth Edison (N.J. Kalivianakis) letter to Nuclear Regulatory Commission Region III (J. Keppler).
Subject:
Quad-Cities Nuclear Power Station Unit 1.
Docket No. 50-254.
LER 82-027/03 L-0.
September 29, 1982.
4.
Nuclear Regulatory' Commission, Region IV (J.E. Gagliardo) letter to Arkansas Power and Light Company (J.M. Griffin).
Subject:
Arkansas Nuclear One Units 1 and 2.
NRC Inspection Reports: 50-313/84-06; 50-368/84-06. March 8, 1984.
5.
Nuclear Regulatory Commission (NRR), " Generic Implications of ATWS Events at the Salem Nuclear Power Plant", NUREG-1000/Vol.1, April 1983.
6.
Nuclear R.egulatory Commission (D.G. Eisenhut) letter to All Licensees of Operating Reactors, Applicants for Operating License, and Holders of Construction Permits.
Subject:
Required Actions Based on Generic Implications of Salem ATWAS Events.
July 8,'1983.
I 1
e
- T e-w-
-y t
- - - - i-yv-t t -
-=i w me o e- -a m,. - -.
-m-rm--,,,-,
---w--i--------v-i,*
v--
\\
/
o UNITED STATES g
NUCLEAR REGULATORY COMMISSION s
g wAswmaTon,o c. cssa
%.*/
FEB271984 MEMORANDUM FOR:
Thomas Novak, Assistant Director for Licensing Dtvision of Licensing FROM:
Frank J. Miraglia, Assistant Director for Safety Assessment Division of Licensing
SUBJECT:
GRAND GULF OPERATTNG EXPERIENCE In response to your request (memorandumr of' October 6,1983) the Operating Reactors Assessment Branch ;0RAB) has reviewed operating experience during the past year et the Grand Gulf ficf Tity and prepared the attached report.
The ORAB review included a survey of reported events at Grand Gulf during the past 15 months (i.e. the Tow power Ticense period) and a comparison of the event reports with reports from two other recently licensed SWRs (LaSaTie and Susquehanna) filed during their Tow power license periods.
The sources of event reports included prompt (telephone) notifications filed per 10 CFR' 50.72 as'we1T as-Licensee Event Reports (LER) required by the TechnicaT Specifications. Operatin'g reactor events briefing sumari'es were
-also examined te identify the more-sfgnificant events. AE00 provided us withr substantial supvcet in obtaining event reports.
In generai the review reveaTed that higit number of prempt reportable events (10 CFR 50.72) have occurred at Grand Gulf in the past year.
The rate of occurrence of these events has been at least three times greater than that of the two other recently licensed BWRs used for comparison.
The ia'ge number of prompt reports are concerned for the most part with inadvertent actuations of engineered safety features. According to the 50.72 reports, equal numbers of these events have been caused by equipment failure and errors en the part of operators and technicians.
Review of operating reactor event briefing sumaries indicates that five "significant" events have been reported for Grand Gulf during the year.
They include a low temperature vessel pressurization incident, electrical system malfunction causing inadvertent RPS trips,. a diesel generator room firi incident,
. simultaneous malfunction of both Transamerica DeLaval diesel generators, and
- an operator error which resulted. in 10,000 gallons of water being drained frem j
the reactor vessel to the suppression pool. The number of significant events at Grand Gulf during the low power ifcense period is higher than that for the two other recently Ifcensed BWRs considered in the review.
LaSalle had only one event significant enough to be reported at a briefing and Susquehanna had none.
It should also be noted that the periods of low power ifcense for LaSalle and Susquehanna were much shorter than Grand Gulf.
e%
lbf b'
J I
RB 2 71984 f
Thomas M.' Novak -
t Based on our review we have concluded that operating experience at Grand Gulf Juring the past year has been atypical.
Ccmparison of Grand Gulf experience with.that of other BWRs indicates that the period of operation with the low power ifcense at Grand Gulf has been abnormally long (greater than 12 months versus 4 m:nths for Susquehanna and LaSalle) and that the rate of prompt reportable events has been much greater than expected.
Based on discussions with Region II we believe that the high rate of recorted events is at least in part related to the large amount of const.ruction and testing activities which have gone on during the past year. This construction and testing activity is the result of design changes being implemented at the plant.
The fact that i
many events which have occurred are related to personnel errors may indicate a lack of experience, on the part of plant personnel.
The rate at which events have occurred at Grand Gulf has not decreased steadily over the long term as the plant has moved closte to commercial operation.
However, a sudden sharp decrease in the rate dio occur in November 1983 which may be attributeu to site inactivity following completion of low power testing in October.
On this basis it would be reasonable to expect the incident rate to continue this decreasing trend as the plant moves closer to ccmmercial operation, and testing and construction activities are completed.
We have discussed the results of ou'r review with IE Region II, and they have informed us that our conclusions are consistent with their most recent SALP review. Region II wi1T continue to monitor plant performance and take appropriate actions should problems continue to occur at a high rate.
POr!gtzCL sig=ed t?
FrarJs J. Miraglia l
Frank J. Miraglia, Assistant Director
\\
for Safety Assessment Division of Licensing
Enclosure:
As Stated DISTRIBUTION Central Files ORAB r/f FMiraglia, GHolahan EButcher MCWWto -
DHoustan AWagner(RII)
TIppolito PFarron DVerrilli (RII)
CPREVIOUS CONCURRENCE SEE DATE l
l ORAB:DL*
ORAB:DL*
ORAB:DL3 DL j
NCaruso EButcher GHolahan lia 01/11/84 01/11/84 02/S9f84 02/;qf/84 i
---r e
.n.
,-_.-,._,..m
..-._,%-,_m,___._,,--.-...,~.-.--__,_-.--,-,-..._..,--_--e.r-,
,,,.m
(
1.
o.
OPERATING EXPERIENCE REVIEW 9
AT GRAND GULF UNIT #1 INTRODUCTION The staff review of operating experience included a survey of reported events at Grand Gulf during the past 15 months (i.e. the low power license period) and a comparison of the event reports with reports frem two other recently licensed SWRs (LaSalle and Susquehanna) filed during their icw oower license periods.
The sources of event reports include promot (telephone) notifications filed per 10 CFR 50~.72 as well as Licensee Event Reports (LER) reouired by the Technical Specifications.
Ocerating reactor events briefing sumaries were also examined to identify the more significant events. These briefings are regularly scheduled meetings among NRC management to discuss recent events 'at operating reactors.
SURVEY OF EVENT REPORTS In the period between mid -August 1982 and September 1, 1983 160 incidents requiring prompt notification were reported as required by 10 CFR part 50.72 One hundred and twenty-two of these events involved plant systems.
The remaining 38 events involved the plant physical security system.
This review has focused on the non-security related events.
The security related events were not considered signiffcant and.were expected based en the testing and construction occurring at the plant. Thirty-five percent (35%) of the non-security related events have root causes related to operator and technician activities (e.g. testing,. troubleshoorting). Equipment problems (mostly olectrical) account for thirty-two (325) of the eventa. The direct causes of the remaining one-third of the events are unknown or not apparent from the brief 50.72 reports. Most of the events involve inadvertent actuations of safety systems with the plant shutdown (e.g. standby gas treatment system, control room fresh air system, reactor trip,, diesel generator start).
The average monthly rate at which these events have been reported is approximately 10 events / month. This rate is compared with rates for LaSalle and Susouehanna in Table 1 and appears to be abnormally high.
Region II inspectors attribute the high rate to the large amount of testing and construction going on at the plant. A review of the data by month does not reveal any particular trend in the incident rate. Data for the past three months shows a rate of occurrence close to the average in September and October with a sharp decrease in November to 3 events / month. The sharp decrease is attributed to site inactivity following ccmpletion of low power tests.. A steady reduction in the rate of occurrence is expected as the plant nears comercial operation, since design changes and associated tests are expected to be completed.
In the period baginning August 1,1982 and ending July 1,1983 a total of 227 LERs were issued from Grand Gulf.
The average monthly rate at whicn LERs have been issued is shown in table 1 along with comparable rates for LaSalle and Susquehanna. The Grand Gulf rate is similar to the rates for LaSalle and Susquehanna.
This is in sharp contrast with the 10 CFR part 50.72 reports discussed above where the Grand Gulf rate was significantly higher than the other two plants.
Review of the Grand Gulf LERs indicates tnat about one-half of the reports relate to problems with fire protection systems.
These problems include many instances of smoke detector alams caused by dust from construction; and, removal of fire barriers for construction purposes.
Only nineteen percent
.l (19%) of the 227 reported events invalved persannel errors and/cr procedural,_
. r TABLE I RATE OF REPORTED EVENTS AT THREE BWR PLANTS DURING LOW POWER LICENSE PERIOD Period of Low Rate of Reported Events Power License (Avg. No. reports / month)
Facility (months 50.72 LER Grand Gulf 12*
10 21 LaSalle 1 4
1 19 Susquehanna 1 4
3 12
- The study period consists of the first 12 months of the low power license period. The actual period of the low power license w"1 be longer than 12 months.
i
e.
deficiencies. Other causes of reported events include equipment problems and planned entry of technical specification action statements for purposes of testing or construction.
REVIEW OF SIGNIFICANT EVENTS Significant events which have occurred at Grand Gulf during the past year have
' been identified through a review of issues raised at the regularly scheduled briefings of NRR management on operating reactor experience.
The review consisted of a review of the Operating Reactor Event Briefing meeting minutes.
For purposes of comparison a similar review has been performed for LaSalle and Susquehanna for the periods they herd Icw power Ticenses.
Events which are discussed at operating reactor event briefingc have been subjected to a screening process in which five or six significant events are selected every two weeks for discussion based on the review of 100 to 150 events reports during the two week period. The purpose of iden' tifying those events here is to provide a measure of the severity and extent of significant operational prohlems During the Grand Gulf low power license period, five significant problems at Grand Gulf were' reported. Our review indicater that only one significant event was reported for LaSalle during the period of its low pcwer license No events were reported for Susquehanna. The Grand Guif events are sunnarized below.
Violation of RTNDT Heatina Limits Durina ECCS Infection October 5,1982 During surveillance testing with the piant in cold shutdown a high DC voltage spike occurred which initiated an ECCS injection.
Low pressure core spray injected and caused the reactor vessel to become water solid (extending to i
theMSIVs).
The resulting pressure transient violated the Technical Specification on nil-ductility reference temperature, RTNDT.
l Reactor Protection System (RPS) MG-Set Outout Breaker Trios, May 19, 1983 Inadvertent tripping of the.RPS MG-set output breakers has occurred repetitively resulting in isolation of the instrument air system and a reactor scram signal.
The causes of the trips have been identified as thermal overload due to insufficient cabinet ventilation, and low voltage due to voltage swings while the RPS bus is fed from the alternate power supply.
To reduce the number of output breaker trips the licensee increased cabinet ventilation, installed voltage regulators to smooth out voltage fluctuations, and installed a new station electrical j
transmission line from off-site.
In addition instrument air system isolation relays have been re-aligned to an interruptable power supply.
This problem l
i
~~
,,)~
r-e re-occurred in January 1984 Upward voltage spikes remaining above the setpoint longer than.1 second have caused the protective MG-set output breaker to trip, resulting in de-energization of containment isolation system logic circuits followed by isolation of the RHR system.
The ifcensee has been unabTe to identify the source of the voitage spikes. To correct the problem, the licensee has increased the output breaker delay time from.1 second to 1.4 seconds. The new delay time is based on measurements of spike duration and consultation with suppliers of the electrical equipment.
The modification assurer that spikes lasting Tess than 1.4 seconds will not result in a trip of the protective breaker. Additional corrective' actions are also under discussion between the Ticensee and Region II.
Inadvertent Reactor Vessel Drainage Durine Shutdown Aoril 3,1983 On ApriT 3,1983, approximately 10,000 gallons of water drained frem the reactor vessel to the suppression pooT through the residual heat removal (RHR) system'. Utis drainage was caused by two RHR valves (F004 and F006) being open simultaneously. At the time of the event, the reactor was at atmospheric pressure with vesseT water temperature approximately 100*F (cold shutdown conditions)_ The vessel water level continued to decrease until the low leveT isolation'sfgna1>was received and shutdown cooTing isolation valves closed to. terminate the leakage.
Diesel Generator Room Fire Seatember 4,1983 A diesel generator engine fire was caused by a ruptured fuel oil supply line (which sprayed oiT on the het exhaust manifold of the diesel. The diesel which caught fire was running at 25 percent Toad for testing at the time.
Two other diesel generators were not affected by the fire.
The water deluge i
system failed to function automatically, but was manually activated to extinguish.the fire. The diesel generator governor and turbo chargers were damaged.
In addition some electrical equipment in the room suffered water damage.
Inocerability of Delaval DieseT Generators October 28, 1983 On October 28, 1983, a TechnicaT Specification Action Statement was entered when two of the three diesel generators became inoperable.
The Division I diesel generator was inoperable due to gasket failure on a lube of f line.
4 The Division II diesel generator became inoperable due to a loose base plate i
nut on the turbocharger which resulted in a trip of the vibration sensor which tripped the diesel. Corrective action was taken to repair both diesel generators.
Both of the diesel generators were manufactured by Transamerica Delaval Inc. (TDI).
TDI diesel generators have recently ecme under close scrutiny by the staf# following a crankshaft failure in a TDI diesel generator at the Shoreham plant. Staff review of the Transamerica Delaval diesel gInerator problem at Grand Gulf is still ongoing.
---+==,-_____._%m---
+--.m
.,e_-,-
e,.-n.-,m,.-c.mm..
.-,w-.
_e-,-,+,..gew,....-.---,-.m.,-y,,.-
,mg.
,r,- -, -
,-,y.,
,y
2 5-e
_ CONCLUSIONS Based on our review, we have concluded that operating experience at Grand Gulf during the low power license period has been atypical.
Comparison of Grand Gulf experience with that of other BWRs indicates that the period of operation with the low power license at Grand Gulf has been abnomally long (12 months versus 4 months for Susquehanna and LaSalie) and that the rate of prompt reportable events has been much greater than expected. Based on discussions with Region II we believe that the high rate of reported events is related, at least in part, to the large amount of testing and construction activities which have gone on during the past year. This construction and testing activity is the result of design changes being implemented at the plant. The fact that many of the events are related to personnel errors may indicate a lack of experience on the part of plant personnel. The rate at which events have occurred at Grand Gulf has not decreased steadily over the long term as the plant has moved closer to comericai operation. However, a sudden sharp decrease in the rate did occur in November 1983 which may be attributed to site inactivity following completion of the Tow power testing in October.
On this basis, we beTieve it is reasonable to expect the incident rate to continue this decreasing trend. as the piant moves closer to connercial operation, and testing and constructfon activities cease Should an abnomally high rate of incidents re-appear,. appropriate ac'tioni such as initiating a review of personnel training programs and piant procedures should be initiated to identify the root cause of the continuing problem so that necessary corrective measures measures can be taken.
-s S
-~
~
v
,----.-w---
,--.ww-nwm,
,m---erma
,,-w,
--r--
cwn-e
,w-en.g--w,---,,-m m - r --se--,er m ww w w w,4-g--m+
vsy,p----emw--.m s-w
.. <.9 i{
UNITED S TEs E'
NUCLEAR REGULATORY COMMISSION c.
E y
j wAswmcTow, o. c.2osss g,ms/
oT a
FEB 2 71984 MEMORANDUM FOR:
Cai1 H. Berlinger, Manager AEOD/P401 TDI Project Group FROM:
Frederick J. Hebdon, Chief Program Technology Branch Office for Analysis and Evaluation of Operational Data-
SUBJECT:
OPERATING HISTORY OVERVIEW FOR DIESEL GENERATORS IN NUCLEAR SERVICE Enclosed is our comparison between problems experienced with Transamerica Delaval, Inc. (TDI) diesel generators and di-esel generators from other manufacturers.
If you have any questions concerning this material, please call Bob Dennig (x24491) or Matt Chiramal (x24441).
i
\\
W M gb on, Chief. LA Frederick J. He Program Technology Branch Office for Analysis and Evaluation
_ of Operational Data
Enclosure:
As stated cc w/ enclosure:
P. Baranows.ky, RES
- 4. '
^
t 1,,. w ^ '.
=
f b
l i
i L
Report No. AE00/F401 Date February 1984 e
OPERATING HISTORY OVERVIEW FOR DIESEL GENERATORS IN NUCLEAR SERVICE Prepared by Office for Analysis and Evaluation of Operational Data 5
I t
The ' subject matter is under continuing review.
This report supports ongoing NRC activities and does not represent the position or requirements of other NRC program offices.
I 6
A m
N O
A
)
} ( YMfI g
i
., -. _ -.,, - ~. - - - - -. -. -... - -. -..,,. - - - - - -
=
Background===
As a result of the evaluation of the failure of the main crankshaft in a Transamerica Delival, Inc. (TDI) diesel engine at the Shoreham Nuclear Plant on August 12, 1983, the staff raised questions concerning the reliability of TDI diesel engines used in nuclear service.
An operating history of TDI diesel engines in both nuclear and non-nuclear applications identified operational problems which were believed to be unique to TDI diesels.
AE00 was asked to review operational data to determine if the TDI problems are, in fact, unique to TDI diesel engines.
We have not attempted to address the broader issue of diesel generator reliability at sites with TDI supplied engines vs. the operational reliability at sites with engines by other manufacturers.
The subject of onsite AC system reliability at operating plants is treated in great detail in NUREG/CR-2989 " Reliability of Emergency AC Power Systems at Nuclear Power Plants" July 1983.
For operating plants, manufacturing defects are just one contributor to system unavailability.
There are contributions from operation and maintenance deficiencies and other component failures wnich have not been found in the TDI experience.
(See for example NRC meeting summaries dated 8/10/81 and 11/9/81 on diesel generator reliability at Joseph M. Farley.)
Approach AEOD focused on the TDI experience for nuclear applications as listed in This listing was an enclosure to a draft Commission paper prepared by R. Caruso, NRR.
We subsequently grouped the failure-related items in into six categories based on the subsystem in which the failure occurred.
The subsystems include:
2-Engine Subsystem Turbochanger Subsystem Lebe 011 Subsystem Fuel 0il Subsystem Cooling Subsystem Air Start Subsystem
- For the six categories, we reviewed Section 9.4 of NUREG/CR-2989 " Reliability t
of Emergency AC Power Systems at Nuclear Power Plants," July 1983 to determine if there were problems with non-TDI diesels that were comparable to the TDI experience.
Section 9.4 is a table of operational events associated with emergency diesel gene'rators for the years 1976-1980.
The events were compiled
, from Licensee Event Reports, station blackout questionnaire responses, and responses to a questionnaire for NUREG-0737 " Clarification of TMI Action Plan Requirements," November, 1980.
For 'each category, Enclosure 2 lists the experience for TDI from Enclosure 1, followed by comparable experience for other diesel engines from Section 9.4 of NUREG/CR-2989.
Items were selected as " comparable" if they concerned the same L
failure mode of the same or similar components or they described a similar
~
situation (e.g., a design error, use of wrong material); and they were possibly associated with the design and fabrication of the diesel engine [i.e., they could not be readily ascribed to maintenance-related problems (e.g., dirty
- oil, sticking components, maladjusted setpoints, gasket leaks, minor oil leaks, minor cooling leaks)].
The last section of Enclosure 2 is entitled " Modifications." The TDI items listed _ here are items in Enclosure I which were not themselves failures or modifications undertaken as a result of a failure elsewhere.
In order to s
-e--
,y,a-4, p.n-_
.---.e,,--.--,,ng,,v,
,y
-,,..,g--_,,
.,,,w.
m
,w_wp..n,,,.y,.m,,vn--ww,-
- provide comparable information f.or other manufacturers we have included as to Enclosure 2 the entire Table E.4 from the draft version of a
AE0D also reviewed LERs associated with diesel engines for the periud 1981-1983 and selected events which described major problems in the categories of Engine Subsystem, Turbocharger Subsystem, Fuel Subsystem, Cooling Subsystem and Design Error.
The results are provided in Enclosure 3.
Finally, for your information, population data for the TDI and non-TDI diesels is provided as Enclosure 4 as a function of manufacturer and continuous power rating.
Discussion The operating history of the TDI diesels and non-TDI diesels, and hence the-data generated, may not be comparable, depending on the issue being examined.
Speci fically:
1.
The operational experience for Grand Gulf and Shoreham, where the vast majority of TDI failures and deficiencies have occurred, was generated during a period of preoperational testing.
During such a time, the contribution of the manufacturer to difficulties should be easier to recognize and should dominate any contribution by the operating and maintenance staff.
Als'o, one expects a relatively high number of " bugs" or deficiencies to crop up during early operation.
In contrast, the information available to us for other manufacturer's engines comes frcm the operational phase, wherein the contribution from operational and maintenance personal can reasonably be expected to increase.
The data supplied to us for San Onofre 1 engines, which reflect few manufacturer problems, come from the operational period.
4 i
2.
We suspect that operating hours, loadings, number of demands, and the spacing of t$ese demands all play a role in diesel generator performance.
1 Again, the Grand Gulf and Shoreham engines where the majority of tne problems have occurred appear to have accumulated a large number of j
l operating hours in a short period of' time as compared with most diesels in
{
nuclear service. A good comparison would require selection of engines from other manufacturers which have seen similar service. We did not have enough information to make such a selection, and the influence of operation and maintenance still might be difficult to isolate and exclude.
i l-3.
.The testing of both TDI diesels and non-TDI diesels varies considerably depending on whether the diesel was tested in accordance with Regulatory Guide 1.108.
We know, for example, that the TDI diesels at Grand Gulf j
were tested in accordance with Regulatory Guide 1.108 while the TDI diesels at San Onofre were not.
4.
We have'no data on the maintenance practices for TDI and non-TDI engines.
This makes it difficult to determine which failures are attributable to the manufacturer's design and which were caused or exacerbated by i
licensee operation and maintenance practices.
S.
We assume the information for the TDI engines is complete.
While NUREG/
CR-2889 contains the most complete record on diesel experience assembled I
to date, the information was provided via LERs and questionnaires and variation in completeness of reporting is still present to some extent in the non-TDI information.
i 3
-e r-,w+v----
,i&,mm ry +w e -
.w.,-,myy--e,wwwwm w c._
- emr--,-.u--,.--y-
-ywm~,,e
_. Thus, we do not have complete or necessarily comparable data for the TDI diesels or the nop-TDI diesels.
Consequently, statistical analyses of diesel reliability (e.g., failure rate analysis), and even qualitative analysis of the
-prevalence or magnitude of problems based on available data should be approached with caution.
We have been able, however, to review the experience and note whether or not
- similar difficulties have been reported for engines of differeat manufacturers.
The following sections attempt to summarize this information but the reader is urged to read the enclosures for himself, since his notion of " comparability" and ours may differ.
Engine Subsystem Experience Instances of piston crown separation and catastrophic crankshaft failure which have occurred in some TDI engines were not found in non-TDI experience reviewed.
Non'-TDI diesels have recorded incidents of damage or failure of basic engine components such as bearings, cylinder heads, pistons, and bolts.
Some of the failures have been repetitive (e.g., six incidents of cylinder head cracks in the same diesel generator at Surry (GM) which ultimately resulted in
~
replacement of all cylinders) and some have been quite serious (e.g., at least three incidents where engines were replaced - Millstone 2 (Fairbanks Morse), Arkansas-2 (Fairbanks Morse) and Quad Cities 1 (GM).
A key concern of the TDI project appears to be whether or not the experience with basic engine parts in non-TDI diesels reflects a " generic" problem with basic engine components.
While we cannot unequivocally rule out the possibility due to the quality and completeness of our information, the evidence for the
t 6-most part suggests isolated difficulties.
For example, the GM experience in the ~ basic engine, area is comparatively sparse in contrast with the GM experience in the turbocharger subsystem, where it appears a generic problem existed for GM engines.
Turbocharger Subsystem Problems with non-TDI diesels have been principally associated with GM diesels.
Most failures were ass' ciated with bearing failure that caused the turbocharger o
to fail.
In some cases fires resulted.
In several cases the turbocharger was replaced.
These problems seem comparable to the TDI problem of bearing wear due to lack of lube oil.
Lube Oil Subsystem The only lube oil problem at a TDI diesel was the oil leak and fire at San Onofre 1.
Several non-TDI diesels have had lube oil problems, however, most of these problems were water leakage int the lube oil.
Xewaunee reported a single instance associated with a GM diesel which was very similar to the San Onofre event in failure mode and mechanism.
Fuel Oil Subsystem Instances of fuel oil leaks or replacement of fuel oil supply lines were noted for GM, Fairbanks Morse, ALC0 and Cooper Bessemer diesels.
In at least two cases fires resulted.
The TDI experience does not appear unique as far as mode, mechanism, or consequences are concerned.
Cooling Subsystem j
Shoreham reported that the engine jacket water pump of a TDI diesel failed by i
fatique.
While licensees with non-TDI diesels have experience pump failures,
7-none reported the mechanism (i.e., fatigue) experienced at Shoreham.
A few instances of jaclet water leaks (that were not associated with major engine damage) were noted for non-TDI engines.
Air Start Subsystem Air start valve failures have occurred with GM, Fairbanks Morse, Cooper Bessemer and Worthington diesels.
However, ongoing maintenance by the licensee may be a more important factor here than original equipment manufacturer; specifically, cleanliness of the air start system.
The specific Grand Gulf and Shoreham problem with air start valve capscrews was not noted in non-TDI experience; however, problems of a similar magnitude with other air start system parts have been noted in non-TDI engines.
Modifi cations A review of' Enclosure 3 shows numerous reliability improvements made throughout the operating lives of non-TDI engines.
These appear to be comparable to the TDI experience listed in the modifications section of Enclosure 2.
8 1
e n
,,n n.-
-,,,-.,~,,.,,--n
,,,,,,n,,,
.e,-n_,-,
-n.,--,,,,-
0
\\
ENCLOSURE 1 U.S. NUCLEAR EXPERIENCE WITH TDI ENGINES e
f
{
l t '
_U. S. Nuclear Experience with TDI Encines e
San Onofre 1
- .Two TDI Diesel Engines Instal. led - Model DSRV-20 Serial No. 75041/42, Rated at 6000KW (nominal) 8800KW (peak)
Problem Cause/ Solution Excessive Turbocharger No Tube oil during standby.
th rus t - bea ring ' wea r.
Lube oil system modified.
10 CFR Part 21 report issued because problem generic.
Lube oi,1 leak and fire.
Excessive vibration.
Line re-supported.
Piston modification to Pistons reworked by TDI to prevent crown separation.
respond to Part 21 report.
Problem identified at Grand Gulf.
. Shoreham
*" Three' TDT DieseYtiiine's inst'aTTed,'ModeT;057-48 '
Serial No. 74010/12, Rated at 3500KW Probiem Cause/ Solution Excessive turbocharger. thrust No lube oil during standby.
, bearing wear Lube oil system modified.
e Piston modifications to prevent Pistons reworked by TDI to crown separation.
respond to Part 21 report.
Problem identified at Grand Gulf.
Engine jacket water pump, Water pumps reworked by TDI in modifications.
response to Part 21 report.
Air starting valve capscrews Response to Part 21 report.
replaced.
Too long for holes.
Enginela~bket water pump shaft
~
' Pump shafts redesigned and fa.iled. by fatigue.
replaced.
Cracks in engine cylinder heads.
Fabrication flaws.
All heads replaced.
e e
=
,,-ev.r,.,.
-,,-. =..
m,-
.,v.
.w.,-------
e,---v
,y
,2 -
Problem Cause/ Solution Two fuel oil injection lines Manufacturing defect in tubing.
ruptured.
T ing replaced with shielded design.
Engine rocker arm shaft bolt High stress cycle fatigue.
Bolts failure.
replaced with new design.
Broken crankshaft.
Cracks in Inadequate design.
Replaced with remaining crankshafts.
larger diameter crankshafts.
Cracked. connec. ting rod. bearings.
Inadequate design and substandard material.
Replaced with new design.
Cracked piston skirts.
Replaced all pisten skirts with new design.
Generic problem.
Broken cylinder head stud nuts.
Repliced all head stud nuts.
-Cracked bedplates in area of ICracks evaluated by LILCo and main journal bearings.
-- detennined to not be significan.
Unqualified instrument cable.
Replaced in response to Part 21 report.
. grand $Tt * ' '
~~"N"^
' * ~ ~
Two TDI engines instaIIed - Model DSRV-16 Serial No. 74033/34,. Rated at 7000KW i'
Problem Cause/ Solution
~
Pistori crowY~ssfaratibn 'durihg Heiddown studs fafTed.
Pistens operation.
returned to TDI for rewcrk.
Generic problem.
Excessive turbocharger thrust No lube oil during standby, bearing wear.
Lube oil system modified.
Air starting valve capscrews Response to Part 21 report.
replaced Too long for holes.
. Flexible driye.ccupling material Replaced with different material.
incompatible wit?r operating enyirenment.
Latching reTay failed during Relay replaced.
testing.
4
,,-y-1,-
.-ev
,e-,,--,w.
,,-,---e
.------.-.,----.--.----.-,..-.-m,----,r--m
y 3-Problem '
Cause/ Solution Air start serfsing line not seismically supported.
Sensing line relocated and properly supported.
Governor lube ofl cooler Lube oil cooler relocated to located too M.g Possibility lower elevation.
of trapping air in system.
Engine pneumatic logic Pneumatic logic design corrected.
improperly designed.
Could result in premature. engine shutdown.
Non-Class IE motors supplied with EDG auxiliary system Motors replaced with Class lE qualified motors.
pumps.
Crankcase cover capscrew failed.
Head lodged in Capscrews replaced with higher strengh screws.
Lock tab washers generator and shorted it out.
installed.
Generator screens installed.
High pressure fuel injection Manufa'cturihg defect in tubing, line failed.
.. Tubing replaced.
o.. :.
~
Fuel oil line failed.
Causic.
High cycle fatigue of Swagelock major fire.
fitting. Additional tubing supports to be installed.
~
Cracks in connecting push All push rods replaced.
rod welds.
3 s.,
~
Turbocharger vibration.
Turbocharger replaced.
Cracked jacket water welds.
Excessive turbocharger vibration.
- Cracks re-welded.
Turbocharger mounting bolt' Excessive turbocharger vibration.
failures.
Bolts replaced.
Air start valve failures.
Cause unknown.
System cleaned and several valves replaced.
More frequ'nt maintenance scheduled.
e
.,,..,,,_-_.-_--,,---,--.--_---.----.-y-,,.,,-,y m.
Problem
Divisi,on II pistons replaced.
Division I pistens to be inspected.
Cylinder head cracks.
Two cylinder heads replaced.
e e
9 e e.
. =.,
.... f, ?
,e,
a.
w...<.4 6
4.
. g.
4
. r... g e
O
. e I
l m
" ~
=
9 ENCLOSURE 2 COMPARISON OF EXPERIENCE 1976 - 1980 E
i
y Plant LER # Date Diesel #
Cescription Shoreham e-Broken cylinder head stud nuts.
Replaced all head stud nuts.
Shoreham Cracked bedplates in area of main Journal bearings.
Cracks evaluated by licensee and determined to not be significant.
i Grand Gulf Piston crown separation during opera-tion.
Hold-down studs failed. Pistons returned to TDI for rework.
Generic problem.
Grand Gulf Cracks in connecting push rod welds.
All push rods replaced.
Grand Gulf Cracks in piston skirts on Division Il diesel.
Pistons replaced.
Grand Gulf Two cylinder heads replaced.
Grand Gulf Crankcase cover capscrew failed.
Head lodged in generator and shorted I
it out.
Capscrews replaced with higher strength screws.
Lock tab washers installed.
Generator screens installed.
B.
General Motors Experience Davis Besse 80-52 7/9/80 All Exhaust supports received too much stress.
Supports added during refueling outage.
3-Plant LER # Date Diesel #
Description a
Prairie Island 1* 80-30 10/8/80 02 Eductor hose broke and diesel tripped on high crankcase pressure.
Surry 1 76-7 7/2/76 1
Heat stress caused cylinder head crack.
Surry 1 76-6 5/8/76 1
Crack in cylinder head.
Surry 1 76-04 5/12/76 1
Crack in cylinder head, bent rod, and broken piston.
Engine not turned over before testing.
Water in cylinder.
Surry 1 76-10 9/4/76 1
Heat stress caused gylinder head crack.
Water in cylinder.
Sixth failure.
All cylinders to be replaced.
Vermont Yankee
- 77-17 6/23/77 B
Eductor hose came loose.
Diesel tripped on high crankcase pressure.
Improved hose clamps to be installed.
C.
Fairbanks Morse Experience Arkansas 2 79-32 4/19/79 2
Engine bearings failed.
Engine was replaced.
Design / Manufacturing error -
see Attachment 2 for details.
Duane Arnold 76-64 10/4/76 IGal Vertical drive coupling hub broke.
Wrong material (cast iron instead of ductileiron).
NUREG/CR-2989 shows these plants have GM engines; NUREG/CR-1362 shows manufacturer as Fairbanks Morse.
- + - -
,--m-ww>v
,e,-
--n,.
---,,-e-
-,rn,
,m--
-,.---.,---,----,.,r
i plant LER # Date Diesel #
Description Duane Arnold
' 76-12 3/26/76 Front cover plate on engine leaked oil.
Oil caught fire, but was quickly extinguished.
Millstone 2 76-63 12/18/76 13U No. 3 upper piston connecting rod l
bearing capscrews sheared and ejected rod through the upper crankcase cover.
Diesel was replaced.
Probably failed from a series of unlubricated or dry starts.
Millstone 2 76-08A 2/23/76 12U Piston failed.
Overhauled engine.
Hatch 2 80-159 11/26/80 2C The cotter pins for the rod cap retaining nuts on two cylinders were broken per-mitting excessive clearance between the connecting rod bearings and the crank-shaft.
One of the connecting rods separated from the crankshaf t and caused -
engine failure.
D.
ALCO Experience Salen 2 80-31 Coupling connecting two sections of camshaft had eight of its attaching bolts sheared.
New camshafts and bolts installed.
d Plant LER d Date Diesel #
Description E.
Cooper Bessemer Experience 1
Cooper.
80-27 5/8/80 2
Piston rod pins broke.
Damaged parts replaced.
All piston bolts were replaced.
A Cooper 79-36 11/10/79 2 Four cylinder sleeves were damaged.
All damaged parts were replaced.
F.
Worthington Experience None noted.
G.
Nordberg Experience
' hone noted.
H.
Allis Chalmers Experience None noted.
I.
Caterpillar Experience None noted.
2.
TURBOCHARGER SUBSYSTEM A.
TDI Experience 1
San Onofre 1 Excessive turaocharger thrust bearing wear.
No lute oil during standby.
Lube i
oil system modified (Part 21).
Plant LER # Date Diesel #
Description Shorehem Excessive turbocharger thrust bearing wear.
No lube oil during standby.
System modified.
Grand Gulf Excessive turbocharger thrust bearing
\\
wear.
No lube oil during standby.
Lube cil system modified.
Grand Gulf Turbocharger vibration caused cracked water jacket welds and mounting bolt failures.
8.
General Motors Experience Arkansas 1 79-6 6/7/79 All Vendor design error.
Rapid start after shutdown could damage turbocharger bearings.
Arkansas 1 78-17 7/15/78 2
011 leak into turbocharger.
Diesel could have operated with leak.
Turbocharger replaced.
Arkansas 1 78-8 3/20/78 2
Searing failure in turbocharger.
Exhaust caught fire.
Turbocharger replaced.
Diesel could have continued to operate in an emergency.
Davis Besse 80-69 9/2/80 1-1 Bolt fragment found in crankcase during oil change.
Bolt was from turbo-gear assembly.
Plant LER # Date Diesel #
Description Davis.Besse
' 79-46 3/30/79 1-1 Turbocharger bearings failed.
The turbocharger was replaced.
Davis Besse 78-18 2/8/78 1-1 Turbocharger failed and was replaced.
Dresden 2 77-051 10/30/77 2/3 Clutch and shaft bearing failure.
Fitzpatrick 76-65 10/11/76 A 011 leak in turbocharger caused fire.
Turbocharger replaced.
Maine Yankee 79-066 10/16/79 1B Catastrophic failure of turbocharger due to bearing failure.
Fire resulted.
(Turbocharger in DG-1A also replaced).
Conn. Yankee 79-09 8/31/79 All Design error could cause turbocharger failure if started within 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br /> of being shutdown.
Kewaunee 77-23 9/20/77 1A Fire in exhaust, but diesel was operable.
Monthly tests changed to 4 hour4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> duration.
Monticello-79-010 4/26/79 All Design error.
Lack of turbocharger lube oil after shutdown.
Point Beach 79-7 4/24/79 All Design error.
May cause turbocharger failure if there is a start 15 to 100 minutes after a diesel shutdown.
..-.,--.,--n-Plant LER # Date Diesel #
Description Saint Lucie
, 79-21 6/25/79 All Design error.
Insufficient turbocharger lubrication may occur on a second start within 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br /> of diesel shutdown.
Saint Lucie 77-42 9/20/77 1A Diesel loaded to full emergency load.
Attempted to pick up full design load.
Turbocharger thrust bearing and clutch 4
were damaged.
Turbocharger replaced.
Saint Lucie 77-2 1/18/77 IB Turbocharger failed.
New unit installed.
Surry 1 79-44 12/30/79 3 Turbocharger failed and was replaced.
Surry 1 79-17 5/2/79 All Design error.
Turbocharger bearing damage may result from start too soon after shutdown.
C.
Fairbanks-Morse Experience Crystal River 3 80-30 Turbocharger ductwork separated from turbo-charge r.
Diesel unavailable for 95 hrs.
Duane Arnold 76-21 Exhaust gases leaked onto engine and burned.
Gasket and insulation replaced.
D.
ALCO Experience Salem 1 77-80 Turbocharger and exhaust gas expansion joint failed.
Cause determined to be turbine blade failure.
Modifications made to turbine to improve blade reliability.
Plant LER # Date Diesel #
Description E.
Cooper Besse&er Experience None noted.
F.
Worthington Experience None noted.
G.
Nordberg Experience None noted.
H.
Allis Chalmers Experience None noted.
I.
Caterpillar Experience
' None noted..
3.
LUBE OIL SUBSYSTEM A.
TDI Experience San Onofra 1 Lube oil leak and fire.
Caused by excessive vibration.
Line re-supported.
Shoreham None noted.
Grand Gulf None noted.
B. ' General Motors Experience Arkansas 1 79-16 6/27/79 2
Lube oil cooler leaked water into oil.
Replaced cooler.
n -Plant LER # Date Diesel #
Description Arkansas 1 e 79-17 9/11/79 1
Lube oil cooler leaked water into oil.
Replaced cooler.
Kewaunee 79-25 9/22/79 Broken lube oil line.
Copper tube replaced with stainless steel.
Vibra-tion caused break.
C.
Fairbanks-Morse Experience None noted.
D.
ALCO Experience None noted.
E.
Cooper Bessemer Experience Cooper 78-31 9/12/78 2
Insufficient oil to bearings during engine,aastdown.
Bearing replaced.
Zir n 1 78-09 Oil cooler tube leak caused high
~
pressure across filter.
Zion 1 78-65 7/17/78 1A Lube oil cooler tube leak of water into oil.
High velocity water eroded tube.
t l
Zion 2 77-67 11/10/77 0 Water leaxed in oil through lube oil cooler.
Zion 2 o0-2G 41/1/80 2A Lube oil leak at cracked weld in pipe.
. Plant LER # Date Diesel #
Description F.
Worthington Experience None noted.
G.
Nordberg Experience None noted.
H.
Allis Chalmers Experience None noced.
I.
Caterpillar Experience None noted.
4.
FUEL OIL SUBSYSTEM A.
TDI Experience San Onofre None noted.
Shoreham Two fuel oil injection lines ruptured.
Manufacturing defect in tubing.
Tubing replaced with shielded design.
Grand Gulf High pressure fuel injection line failed.
Manufacturing defect in tubing.
Tubing replaced.
Grand Gulf Fuel ott line failed.
Caused major fire.
High cycle fatigue of Swagelock fitting.
Additional tubing supports to be installed.
~
=. -
.. l Plant LER # Date Diesel #
Description l
e B.
General Motors Experience Beaver Valley 78-32 4/18/78 1
Fuel oil pump leak.
Quad Cities 78-27 9/28/78 1
Fuel supply lines replaced.
Turkey Point 3 79-15 4/26/79 B
Fuel starvation caused by cracked nipple in fuel line.
C.
Fairbanks Morse Experience Ouane Arnold 76-75 11/4/76 IG-21 Crack in fuel line leaked fuel which caught fire.
Supports added for fuel lines.
Calvert Cliffs 79-69 11/27/79 11 Leaking fuel line.
10/23/79.-
Crack in fuel line.,It was resoldered.
Crystal River 10/5/76 1C Fuel line repaired.
Hatch 1 12/14/77 A Fuel oil line modified.
H. B. Robinson Millstone 1 77-29 9/27/77 DG Nipple in cylinder 12 was cracked and leaking.
Millstone 1 77-7 2/1/77 DG Nipple in cylinder 12 was cracked and leaking.
Millstone 2 78-19 8/3/78 13u Leaking fuel injectors.
Diesel could have continued to run in emergency.
Millstone 2 78-19A 1/28/79 13u Leaking fuel injection. Manu facturing defect.
Other assemblies checked okay.
- Plant LER # Date Diesel #
Description Millstone 2 e 76-52 9/1/76 13u Injector leaking fuel and small fire resulted.
Peach Bottom 2 and 3 4/27/79 2
Fuel oil line replaced.
5/18/79 2
Leak in fuel line.
10/5/79 2
Fuel oil line repaired.
D.
ALCO Experience Palis.ades 79-5 1/3/79 1~- 1 Fuel line broke.
One hundred eighty gallons of fuel sprayed out.
Pilgrim 80-62 9/3/80 A
Fuel line to cylinder 9R had broken.
E.
Cooper Bessemer Experience Cooper 77-47 9/12/77 1
Fuel line to day tank vibrated and broke.
Support was improved.
Cooper 76-34 8/23/76 1
Fuel line to injector broke.
F.
Worthington Experience None noted.
G.
Nordberg Experience None noted.
H.
Allis Chalmers None noted.
Plant LER # Date Diesel #
Description I.
Caterpillar 2kperience None noted.
5.
COOLING SYSTEM A.
TDI Experience San Onofre 1 None noted.
Shoreham No failure.
Engine jacket water pump reworked by TDI in response to Part 21 report.
Shoreham Engine jacket water pump shaft failed by fatigue.
Pump shafts redesigned and replaced.
Grand Gulf None noted.
B.
General Motors Experience Conn. Yankee 6/21/76 Diesel fresh water pump leak.
Pump rebuilt.
Dresden 3 77-38 9/14/77 3
Outboard bearing worn on pump.
Quad Cities 1 80-26 10/11/80 1/2 Cooling water pump motor shorted.
C.
Fairbanks Morse Experience Crystal River 79-108 12/1/79 1B Shutdown cooling water pump failed.
Bearing failure.
_ Plant LER # Date Diesel #
Description Millstone 2 5/5/78 Jacket water pump failed.
Prairie Island 1 79-2 1/26/79 02 Cooling water pump did not start because of a speed switch failure.
Hatch 1 3/5/76 Coolant jacket system modified.
D.
ALCO Eroerience Indian Point 2 77-29 8/29/77 Jacket water leaks repaired.
lE.
Cooper Bessemer Experience None noted.
F.
Worthington Experience None noted.
G.
Nordberg Experience None noted.
'H.
Allis Chalmers Experience Lacrosse 5/20/77 IB Cooling water leak.
Rewelded bad weld.
I.
Caterpillar Experience None noted.
\\
_. _ _, 1 6.
AIR START SUBSYSTEM-
- t..
TDI Experiende pl ant-LER # Date Diesel #
Des cription San Onofre None noted.
'Shoreham Air start valve capscrews replaced.
Too long for holes.
Response to Part 21 report.
Grand Gulf Air start valve capscrews replaced.
Too long for holes.
Response to Part 21 report.
Grand Gulf Air start sensing line not seismically supported.
Sensing line relocated and properly supported.
- Grand Gulf Air start valve failures.
Cause unknown.
System cleaned and several valves replaced.
More frequent maintenance scheduled.
8.
General Motors Experience Dresden 2 79-014 3/5/79 2
Bendix air solenoid failures.
Scheduled modifications should improve performance.
Dresden 2 77-071 12/3/77 2/3 Air regulator diaphragm ruptured.
.- Plant
_LER # Date Diesel #
Description Quad Cities 1
' 79-5 1/23/79 1/2 Air start solenoid stuck open.
Quad Cities 1 76-5 2/11/76 Air start solenoid stuck.
C.
Fairbanks Morse Experience Farley 1 78-18 3/8/78 IC Air start solenoid valve failed.
corrosion prevention improvements being studied.
Farley 1 77-26 9/13/77 IB Air start solenoid stuck open.
Farley 1 77-27 9/16/77 1-2A Air start solenoid stuck open.
Farley 1 77-15 8/17/77 18 Air start solenoid stuck open.
Farley 1 77-27 8/28/77 IB Air start solenoid stuck open.
Hatch 1
~
76-24 5/15/76 1A Air start solenoid stuck closed.
Millstone 2 1/13/76 13U Air pilot valve failed.
Calvert Cliffs 1 79-061 10/24/79 11/12 Diesels started and left running until seismic supports installed (Air start).
Calvert Cliffs 2 80-035 7/30/80 All Design error.
Tubing not seismically qualified (Air start).
H. B. Robinson 5/26/76 Leaking air start solenoid repaired.
H. B. Robinson 7/03/78 Air start solenoid replaced.
4 Plant LER # Date Diesel #
Description H. B. Robinson 10/18/78 A Air start solenoid replaced.
Vermont Yankee 77-18 7/26/77 A
Air start valve failed to open.
Debris in line.
Valves to be replaced by improved valves.
D.
ALCO Experience None noted.
E.
Cooper Bessemer Experience Zion 1 78-72 8/17/78 18 Air start pilot valve leaked.
Zion 2 79-34 5/11/79 0
Air valve leaked air from reservoirs.
F.
Worthington Experience l
Cook 2 78-13 3/19/78 200 Air start check valve on cylinder
- 5 broke.
k G.
Nordberg Experience None noted.
H.
Allis Chalmers Experien'ce None noted.
'I.
Caterpillar Experience None noted.
MODIFICATIONS e
A.
TDI Experience Plant LER # Date Diesel #
Description San Onofre 1 None noted.
Shoreham Unqualifted instrument cable replaced.
Grand Gulf Flexible drive coupling material incompatible with operating environment.
Replaced with different material.
Grand Gulf Governor lube oil cooler located too high.
Possibility of trapping air in system.
Lube oil cooler relocated to lower elevation.
~
Grand Gulf-Engine pneumatic logic improperly designed.
Could result in premature engine shutdown.
Pneumatic logic design corrected.
Grand Gulf Non Class 1E motors supplied with diesel auxiliary system pumps. Motors replaced with Class 1E qualified motors.
8.
Other Manufacturers See Attachment 1 to this enclosure.
t 1
l
\\
l ATTACHMENT 1 I
f TO
(
ENCLOSURE 2 1
DIESEL GENERATOR M00!FICATIONS I
f y
{
f r
r I
i 1
i I
l' S
I i
i I
r i
(
6 r
Tabla E.4 Diesel Cenerator Hodifications Plant Date Subayates Hodificattoa
/ Arkansae kclear One 1 6/22/74 Covernor Overspeed trip changed from 940 to 1035 3/77 Cooltag water Replaced DG heat exchanger with one of larger capacity.
J Beaver Valley 1978 Air-start Stagger test on air-start notera.
i 1978 Air-start Blowdown air receivers 3 times / day.
Upgrade fuel oil standards..
~
2/4/81 Atr-start lastalled air dryers.
l 4/3/79 Annunciatloa Add alarm on DG status - control room.
l 9/20/79 Control Prevent bresier closure for ao field and y
. change under.~21tage relays to solid state.
l 8/23/79 Sequencer Sequencer receives power from #1 or #4 vital buses.
Provida separate control transformer
~!
~
from 480V energaacy basea.'
10/3/78
' ~Euciter Improve manual field fissh - bypass some contacts.
10/13/78 1.ube and cooling Install isolatloc val?res for instrumente water instruments for calibration.
~
10/22/80 Fuel lastell tank to temporarily hold fuel far Add sampling connections.
test.
. 7.-
,"/
I i
v.
W,
,A
\\e, *
~
Table E.4 Continued,
Plant Date Subayatem Hodification
~~
neaver Valley 5/18/81 Air-start Install union to permit check valves to be (continued) replaced.
5/20/81 Alt-start Install horizontal check valves.
j 8/5/81 CH design changes CH has-a design that will eliminate turbo bearing damage in the.25 to 3 hour3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br /> period after a run greater than one hour.
Replaced crankahaft vibration damper.
Replaced soldered lube oil cooler with rolled tube cooler.
How acreen with trap to remove material that would damage turbo.
New turbo gear.
Idler gear assembly.
Lube oil compling connection.
Big Rock Point 10/21/76 Electric-atart Change starter from crank / pause to continuous 25 second crank.
1/10/77 Governor New oil line to governor to insprove start time..
(.
('i.
. ~4 t
)
s n
Table E.4 Continued' Plant Date Subayatem flod ' fica tion
' Big Rock Point 5/27/77 Governor (continued)
Install throttle aroi.to replace governor control.
Brunowick 1 & 2 1/12/75 Annunciatora Fuel oil alarina.
2/7/75 Dainpe ra Dampers modified.
Did not work correctly.
2/25/75 Cabica cable separation.
11/25/75 Air-atart Air compresnor relief valve.
2/22/79 Regula'to r Auto-transfer from ac to de regulator upon loan of voltaCo regulator potential transformer.
5/24/76 Room temperaturu
, Individual temperature switchen in each bG room.
5/6/77 Lube and cooling Increase lube and jacket water heater water actpointo.
4/21/77 Instrumento Increase lobe oil preneure trip teca with valven for instrumento.
2/23/79 Annunciator Separate alarina for low air preneure and bearing gear engaged added to llot Availabic.
8/4/80 Fire Protection
!!odify ventilation to remove fumen.
8/29/79 '
Control Change tank pack actpoint from 510 to 500 rpm.
Controla service water valve.
rm s~:,
- (,
~~
.)
t I
i.
I-Tabic E.4 Continued Hodification Plant Date Subayatca Calvert Cliffa 1&2 5/15/79 Blower Procedure to prevent blower damage.
11/13/79 Covernor Semi-annual flush of governor.
Hanufacturer recommendations.'
1/20/79 Fuel Inapect injectors every 6 montha instead of 18 months.
7/24/79 Exciter Potential transformer was grounded and' amoked.
It was replaced.
6/21/79 Procedure Procedure review mado to assume steps for returning DG are included.
8/8/79 Vent fan Added. relay to provide positive start of fan.
5/5/80 Atr-start Press nenoorn vibrated.
Relocated.
Also moved acase line upstream of check valve.
Root valves on pressure switch.
Cooper 10/10/78 Air-start 3/2/74
, Control Speed oenae modif.ted.
3/17/74 Exciter Two parallel contacts to assure field flashes.
1
j Tabis E.4 Continued Plant Date.
Subayaten Hodification 1
Cooper 4/30/75 Add air-operated valve in oil line to turbo.
(continued)
Prevents oli accumulation in inlet header.
,.m 11/2/76 Air-start Installed timo delay in air-start to assure start.
4/29/79 Annunciator Added annunciator.,
5/22/79 Luba Time dolay relay keeps oll' pump working for 30 acconda after DC trip.
It la for crankahaft lube during conotdown.
5/25/00 Logic Protect DC if offaite power lost while testing DG.
Trip non-easential loads.
I 6/10/00 Hodify DC allencer bypass control to assure'
..edequate air pressure to fuel racks during 4
start.
6/6/00 Cooling water Added continuous vent on Jagket water pump to prevent air vapor lock. '
Crystal River 3 3/16/01 cooling water Vent valve for refilling coolant.
11/21/00 Generator Generator stator temperature relay-change for relay with higher'actting.
e 10/9/00 Luho Lube oil alarm actpoint too low.
7/3/79 Logic DC could not be roset in normal manner i
after low lube pressure trip.
1 1
y_.
G
' -l)
~
(.g q%..-
c' s Tabic E.4 Continued Plant Date Subayatem Hodification Crystal River 3 5/22/79 Annunciator Add alarma that will. annunciate any (continued) condition that will prevent an ado-start.
5/3/79 Annunciator Eliminate battery ground alarm while flashing field, f
b 11/15/79 Engine Replace cais rollers - colt.
/ resden 2 & 3 D
10/77 Covernor Change governor speed with engine operating.
7/70 Air-start Add capability to blowdown starting air.
10/70 Logic Install tachometer for display.
0/79 Annunciator Alarm bus tie and omergency bus breaker test poaltion.
9/79 Annunciator Alarm air shutoff valve in closed position.
9/79 Air-atart Install multiple air-start ayaten.
0/79 Annunciator Install droop nlarm in control room.
11/79 Air-start Provide more positive relay action for two air-start ayatems.
8/70 Annunciator Separate alarm for 2 and 2/3 DCs.
10/30/79 Cooling water Hake cooling water valves inotor-operated from control room.
1
Teble E.4 Continued Plant Date Suboystein Hodification Farley 1 & 2 5/7/77 1.ube Increase pre-lube time on 2850kW.
2/26/79 Sequencer Hodify manter tent switch no acquencer will not remain in test (opring return switch).
7/26/81 Procedure Load reject - open DG, breaker instead of supply breaker.
12/00
. Fuel Fuel oil consumption was low so installed a connection to auxiliary boiler fuel tank.
11/78 Air start Installed stainicas steel pipeo and non-regenerative air dryer.
9/79 Inverter puring DG atart inverter ac breakers tripped on tranolent.
Upgraded breakers.
2/79 Sequencer Test inode select switches are wrong type -
removed and installed specified switches.
10/79 Synchronizer Auto oynchronizer doco not work.
It was removed.
Fitzpatrick 10/0/76 Fuci Hake fuel oil low level switch independent of pump motor control circuit.
15/15/76 Logic Replace UV relaya.
10/15/76 Air-start Primary and accondary air-start motora start simultaneously.
Simplify design.
e m
c
~
.,u..
Tchle E.4 Continued Plant Date Subsystem Hodification i
Fitzpatrick 11/24/76 Control DC not adequately protected.
Add (continued) droop-normal switch in control room.
10/31/70 Hove panel from DG to eliminate vibration.
i 12/15/70 Annunciator Honitor control power to DG.
i 12/6/76 Logic Block low lube oil and high Jacket water temperature tripa for LOCA.
Nort Calhoun 10/81 Logic Add interlock to sequencer react and DG breaker.
Prevent complate train failure because of ningle relay failure.
11/01 Control Filter tachomoter output with 0.1 uf cap.
Cinna 10/11/81 Fuel Install water-tight doors on fuel oil tanka.
Ilatch 1 & 2 1976 Air-start Ped. oxide on valve causes failure.
Replace valve.
1976 Logic Hake UV logic 1/2 taken twice.
6/77 Fuel Synthetic hooes are falling.
Replace with steel piping.
7/77 Cooling water Cauge for Jacket water.
8/77 Distribution Replace 4100V cables to bus bar.
Cables are lienting.
S e
.i Table E.4 Continued Plant Date.
Subaya tem.
Hodification p
llatch 1 & 2 10/77 Logic Konet undervoltage relays to prevent l
(continued) continuous operation at' low voltage.
4 l
11/77
' Air-atart Hake moisture detection indicdtor lights j
function.
i 11/77 Fuel Replace veut line with stainicas steel.
l 11/77 Fuel Eliminate injector leakaSc.
Install clean fuel oil drain tanka oa cach DG.
t 4/70 Control Hove voltage regulator adjustseent inside cabinet.
6/70 logic
- Hake undervoltage relay contacts normally.
open.
Eliminate vibration sensitivity.
6/70 Logic Lockout DG breaker for.overcurrent, etc.
j 7/70 Cooling water Low pressure chutdown switch too high.
l 1
10/70 Distribution Provide capability for DC 18 to acrve units 1 or 2.
Cooling water Add a acrvice water pump with local and remote control.
I
.6/00 Logic Eliminate relay.
It may hang up in emergency mode.
j i
(
i ~)
'45
47-' ' ' --
~
3
{
2 Table E.4 Continued Plant Date Subsyntem Hodification.
Ilatch 1 & 2 10/80 Control Add synchro check re ay.
(continued) 1/01 Governor Replace governor booster servometer.
11'il Seni conduit to keep cold, moist air.out.
1/01 Covernor Alarm if DG is not synchronous speed.
9/77 Lube' Change pre-lube time.. See I.ER 77-62.
2/8L Fuel Clean spilled fuel.
This is a procedure f
change.
d-11/00' Procedure Open llCC circuit breaker for RilR test valva.
Jumper replacement for coolant high temperature trip.
2/81 Procedure Renet LOCA algnal after jumpera are removed to allow auto RIIR pump start.
4 j
Indian Point 3 1978 Annunciator Alarm shutdown, lockout, loss of de, or not auto-start.
1979 Control Isolate control circuita of DG 31.
1980 Intake air Ioolate air intakes to prevent DC breathing CO2 if CO2 actuates.
La Cronac 8/76 Engine Add DG 18.
'6/77 Annunciator Add low water temperature and low voltage alarms.
t
Tabl6 E.4 Continued Plant Date Subsyntesa Hodification La Crosse 8/77 Control Power "on" liglit added for DG 15.
j (continued)
~
10/78
' Annunciator DC 1A not in auto-alaris.
4/80 Annunciator Alares lose of inverted IC.
8/77 Annunciator Install hydrogen alarma in battery roosa.
Illl1s tone 1 11/6/74 Cooling wateE DC can get einergency cooling water from the fire s'ys tein.
3/17/78 Engine C,'ra rollers were replaced.
7/5/79 Annunciator Alarra when the DG is not ready for start.
bynchrometercheckrelay.
9/7/79 Synchrometer Ill110 tone 2 1/20/76 Prevent DC start when output, of primary transforener is open.
t 12/1/77 llent exchanger Corronlon.
Add additional zinen.
3 1/19/76 Control T1:ne delay relays unreliabic.
They were replaced.
~
4/21/76 Air-start Hake the air-start system more reliabic.
3/29/76 Fuel Replaced flex hone to injectora with copper.
8 k
3 5.)
o
- (..
p.
c
.c ;.
!l Table,E.4 Continued I'lant Date Subsystesa Hodification H111 stone 2 0/11/76 Logic Delete non-emergency signals in DC logic.
(continued) 12/9/76 Air-start Hodify air compressor menne lines.
10/19/76 Air-start Install isolation valves in air compressor pressure acnse lines.
7/22/77 Lube Remove low lubo oil level trip.
y 1/21/77
' Engin'a Add stiffeners to prevent DG' vibration.
4/30/79 Logic Remove OTL, CTL, and 'CLL tripa.
See LER 77-32.
S/22/70 Annunciator
'Alaria loss of control power to circuit breaker or circuit breaker racked out.
4/25/01 Air-start Put unions in air lines.
5/11/70 Annunciator Alarm auto-start.
11/0/70 Exhnust Add allenceca in exhaust line.
4/30/79 Logic Pecycnt DC ntact on reactor trip.
i 5/12/79 Synchrometer Synch.. check relay.
1/20/0L Lube Isolate lube oil filter lines.
5/12/79 Annunciator Remote annunciator of DC trouble alarm.
e.
Table' E.4 Continued Plant Date Subsystem Hodification Hillatone 2 8/20/79 Annunciator Alarm on fuel oli valves closed.
(continued) 1/20/01
' Lube Install vent lines on lube oil atrainor.
l 3/77 Logic Undervoltage relaya were not sufficient.
~
Added additional re.laya.
l l
3/00 Annunciator,,
separate alarms for low fuel level.
Exhaust Install tornado misalle shields on exhaust.
North Anna 1 12/79 DC fan
. Increase rating from 180 to 230 hp by strengthening power takeof f.
n 3/B0 Logic Change crankcase vacuum trip setting for
.teore reliable starts.
d l
7/01 Logic Auto-react in governor motor operated pot circuit.
Pa11 andes 3/00 Procedurca Hodified procedurca to reduce operator errors.
3' Peach "llottom 2&3 Logic Solid state relaya wege sannitive to voltage spikea.
Installed aga'atot time delay relays.
i Logic IIFD relaya were not reliable.
Installed agnatot relaya.
ino Improve fuct header and com follower.
1 c:'
(.# '.-)
(,'j
,-~
.,q.
s.
5 Tahic E.4 Continued Plant Date Subsystem Hodification Peach' Bottom 2 & 3 Governor
' Install governor EG-D10G.
(continued) l Fuel / Fire liigh temperature switch on fuel tank.
i Fuel Fuel sample lines added.
J oint Beach 1 & 2 9/79 Exhauct Exhaust manifold inspection post-examine exhauct screen.
11/80 Fire Vent DG during turbine hall fire.
Reverse vent fan direction.
i t/Prairic Island 1 & 2 8/13/76 Fuel Replaced fuel hoses with pipes.
11/26/80 Logic Remove 2 minute delay after DG atop.
9/1/77 Fuel Replaced unreliable fuel oil level switches.
8/20/78 Logic Revised circuit to prevent burn out of lockout relays.
8/17/80 Turbo Installed screens at inlet to turbo.
5/26/77 Fucl Changed power supply fcr D2 clenn fuel pump.
Robinson 2 7/71 Cooling water Alarm for expanalon tank level.
Early warning of leaka.
~
2/72 Logic Install key switch to bypass DC tripa (alarmed).
l f
~
Tabic.' E. 4 Continued Plant Data Subaystem Hodification Robinson 2 10/75 Air-start Add a accond air start solenoid on ecch,,DC.-
(continued) 10/74 Exciter / battery Replace Icad acid battery with Nicad and locate in DC room.
i l
8/74 Fuel Replace synthetic hoses with areal tubes.
8/74~
Switches Replace componenta in defective switches.
6/78 Lube Time delay starts.
Prelube changed from 15 acconds'to 2 minutes.
2 8/80 Annunciator Alarm DG out of service.
2/8L Start logic Hake start algnal last 10 acconds instead of 1 accoud.
8/81 Lube Change prelube time from 2 to 4 1/2 minutes.
In progreso Air-start Service water piping to air dryer is being
]
cbcaged from carbon steel to atainloca ateel.
/
t/St. Lucic 5/78 Turbo Turbo soak back pump used to stop at 200 rpm.. Now pump continuco to run (acens to have climinated_ problem).
l 5/78 Pr'ocedure To prevent turbo problem operate DC at 100% instead of 37% (eliminate on turbo generator).
l i
s T
t
=g
.3 g
(.' L
- t, Table. E.4 Continued Plant Date Subsystem Hodification St. Lucic 5/79 Cooling fan Issproved crankshaft coupling to fan.
Similar marine couplings have failed. ~
6/79 Lube No non-emergency starts until lube oil cools ao pressura could be maintained.
3/00 Exciter Install larger sized exciter leada.
One failure had resulted.
10/81 Cooling water Add vents to cooling water high points.
id/80 Cooling water Proceduco to ensure p' roper venting.
8 V rojan 9/77 Logic Voltage permissive relay could reset on T
voltage dip.
Add a seal-in contact.
5/77 Immersion heaters Hotor overload devices tt.p heatera -
i now bypassed.
9/80 Procedure Honitor fan filter delta p' for volcano-proof systems.
l 9/80 Air-start Revised test procedure to have independent j
and siisultaneous test of each air starting ayates.
See INPO SOElt 80-1.
I
!!aine Yankee 9/78 Fuct Prevent fuel oil transfer pumps from i
operating when fill valves are open.
)
6/78 Annuncia' tor Improved alarms.
4 l
i i
e.
~
~ (
l I Tabla E 4 Continued Plant Date Subsystem Hodification linine Yankee 7/81 Cooling water Provide cooling for DG 1A from primary,
(continued).
component cooling water and DC IB Trom accondary component cooling water.
f Quad Cities 1&2 2/11/78 Exciter Replace exciter transformer supprease s
^
state-of-the-art devices.
4 1/20/80 Logic Trip 4kV breaker if shutdown relay operatea.
Prevent motoring.
l 9/15/80 Air-sta'rt Install check valve for the C-D DG receiver upstream of the tie with A-B act.
I i-9/28/78 Fuel Put sleeven on fuel transfer lines to protect.
2/23/79 Dreaker Install test owitches foe 4kV undervoltage functional tent.
l t/Surry 1 & 2 7/2/80 Lube Check motor-driven lube oil' vibration -
several failures. Also there were turbo vibration checks.
3 i
4/2/80 Procedure Insure cafety-related valve poaltions are independently verified urkey Point 3&4 6/11/79 Exciter Removed connection of neutral from exciter to DG transformer.
1 l
+
s
+. :./
~
u
..m.
' ?.
'.,-l
'..,,) '
Tebia E.4 Continued Subayatem Hedification Plant Date Turkey raint 3 and 4 1/27/01 Indicator R splaced iniicator linkl. that coulti cauco q
(continued) starc fa11cre, 3,*21/78 Fuel Steel accmed cubing was n.placed uith l
stainican eteel.
j Vermoat Yankee
'1d/14/80 Vent Damper will fail open or loss of air or l
power.
j 10/20/80 Lube Improved lube oil temperature indication.
I 1/16/80 Damper Shut vent fans off on low temperature.
Prevent governor malfunction.
a 1
10/6/79 Exciter Honitor auto and manual theostata to ensure sufficient excitation.
2/7/76 Exciter Hove exciter to the station batteries.
t j
1 I
- Additional detail on Arkansas 2 Diesel Failure 4/19/79.
The following material was quoted from " Nuclear Power Experience":
o Arkansas One 2 - fiov. 78 (prior to initial criticality)
During a test, Fairbanks Morte DG B tripped from 100% load.
Inspection revealed damage to bearings (rod and main), crank-shaft and several pistons.
The cause of failure was postulated as oil aeration; however, analysis showed the oil to be within specs.
They changed oil with Mobilgard 445.
They found a loose baseplate mounting screw which could have contributed to the failure by not allowing uniform expansion.
The DG was load tested successfully following repairs.
(gzs)
In Apr. DG #2 engine failed during a routine surveillance test.
The unit developed a severe vibration after being unloaded and was immediately tripped by an operator observing the test.
Investigation revealed 4 upper crankshaft bearings wiped and 3 piston skirts cracked.
Repair of the diesel engine continued throughout the remainder of the month.
See XI. A.323 for additional information.
(hua, hub)
Arkansas One 2 - Apr. 79 - hot standby While performing a surveillance run of the "B" Emergency DG, the engine exhibited excessive vibration.
DG "A" was proven operable immediately and the unit was brought to cold shutdown.
Investigation revealed failure of the forward half of the upper main bearings.
Damage was found at the rod bearings and crankshaft.
The failure was caused by poor lubrication due to bearings being improperly located relative to the position of the journals of the crank.
The main bearing caps were.
relocated t>y redwelling.
Extensive load testing was ' success-fully completed following repairs.
(ibe) m-
EilCLOSURE 3 NON-TDI EXPERIENCE 1981 - 1983
.f
.1.
ENGINE SUBSYSTEM 3
General Motors Experience Plant
-LER #
Description Fort Calhoun 82-07 While testing DG 2, a leak was discovered in the
-copper tubing vent line from the thermal mixing valve to the coolant expansion.
The vent line was found to be cracked at the point where it was connected to a fitting.
The defective tubing was replaced.
An engineering evaluation has been initiated to determine if flexible hose can be used.
Quad-Cities 1 81-22 1/2 DG during maintenance found babbitt in the oil pan.
Further inspection revealed that No.11 bearing cap was warped.
The cause was a pin-hole leak found in a cross-over fuel line which possibly diluted the lube oil to the bearing.
The bearing apparently overheated and warped the bearing cap.
The diesel engine was replaced with an identical model.
Sequoyah 1 83-70 1A-A EDG tripped on high crankcase pressure.
After
~
trouble-shooting, the engine oil cooler and the No. 8 cylinder power pack were replaced.
The probable cause 4
of the oil cooler failure was normal end of life.
The cylinder head was sent to GM for analysis.
Fairbanks Morse Experience Duane Arnold 81-015 During annual inspection of DG 21, the lower crankshaft 81-016 bearing of crankshaft No.14 was found wiped on the journal surface.
Redundant 1G-31 revealed a similar problem -
y
,,m-r_...e,--e-c,-
-c,
,,---,-,,-s e
Plant LER #
Description lower crankshaft thrust bearing #13 and adjacent bearing No.12 were found wiped on the journal surface.
The bearings were replaced and the crankshafts relapped.
Calvert Cliffs 1 81-36 On 12 EDG it was discovered that the upper crankshaft thrust bearing was worn excessively, due to inadequate pre-lube of the engine prior to starting.
The thrust bearing was replaced.
A test was run to determine pre-lube requirements of each DG.
The'results of the test were used as the basis for establishing minimum times for pre-lube on all non-emergency starts.
Calvert Cliffs 1 81-78 DG 12 was taken out for PM.
Two cylinder injectors and a water jacket relief developed leaks.
The injectors were replaced and the relief was installed with new 0-rings.
Eight airblower discharge flange bolts were discovered broken - the failure of the bolts were determined to be by material analysis to be fatigue.
All 14 bolts and their inserts were replaced.
' Farley 1 53 On 7/28/81, DG IC tripped under load.
On 7/30/81, while attempting to perform DG IC operability test, it failed to start.
Investigation revealed that an 0-ring between No.11 cylinder lining and cylinder had failed allowing water to enter the No.1 cylinder and overflow to reveal other cylinders via the intake air manifold and caused a " hydraulic lock" Plant LER #
Description of the engine.
This resulted in damage to Nos. 1 and 11 piston inserts and bushings, the lower thrust bearing and vertical drive assembly.
All damaged parts were replaced.
Farley 1 81-32 DG IC failed to start. The cause was a leaking seal between the inner and outer cylinder which caused the No.10 cylinder to fill with water.
The cylinder liner was replaced.
Farley 1 81-67 DG 2C tripped under load due to high crankcase pressure.
No. 8 cylinder 'iner was found scored and the No. 8 cylinder 0-ring was faulty which allowed water to enter the cylinder.
The scoring of the cylinder caused localized heating and exhaust leakage into the oil sump which caused the high crankcase pressure.
Due to the No. 8 cylinder 0-ring failure a decision was made to replace all 12 cylinder liners.
Farley 2 81-43 DG IC tripped under load due to high crankcase pressure.
t;os. I and 11 cylinder liners were found to be scored.
The engine was repaired.
Hatch 2 82-79 DG 2C tripped after 37 minutes of operation, was restarted and tripped again.
The cause was found l
tc be bearing failure.
This engine had multiple
[
manual starts (an estimated 120-150 fast. starts) as a result of increased surveillance.
The first
.._.. _ _ _, ~ _. _,. _. _ - _. _ _ _ -...
4-plant LER #
Description bearing to fail was the No. 8 connecting rod bearing, with other bearings showing damage.
During the 20 hour2.314815e-4 days <br />0.00556 hours <br />3.306878e-5 weeks <br />7.61e-6 months <br /> run-in check, one main bearing showed minor scoring.
The bearing was replaced.
The multiple manual starts revealed that a longer pre-lube time would allow the bearings to be better lubricated before the diesel was started to avoid bearing failures.
The operating procedures were revised to incorporate a new pre-lube time as recommended by the manufacturer.
Hatch 2 81-127 2C DG failed during testing.
Investigation revealed one of two rod cap retaining bolts had come out, allowing engine torque to break the second retaining bolt which caused the rod to separate from the crankshaft.
The engine was repaired and returned to service.
2C DG has an identical failure in November 1980.
' North Anna 2 83-50 EDG-2J tripped during surveillance testing.
Internal cooling water leakage resulted in a high crankcase pressure trip and caused a cracked piston and cylinder liner.
The unit was repaired and returned to service.
Cooper Bessemer Experience Cooper 82-20 During testing the No. 2 DG shutdown with no alarms or indications.
DG was declared inoperable when e
e.
yw-
.__y_-. -,. _..
p.-
c_
p
,7_
-.,p7,-_w,g,..,w__.,,,-g
.,sy,.r_--,,,,-,._,--------,,,.y,#,.
.,--e--
- Plant LER (
Description water was found in the lube oil system.
A 3-inch L-shaped rupture of the No. 8 left cylinder liner expansion seal was found.
The seal was replaced.
Cocper 82-16 During surveillance testing 1 DG tripped with no other alarms or indication.
The unit tripped due to drift in the holding mechanism of the safety trip valve cverspeed device.
The valve was replaced.
A section of the 125 psi control air line to the trip valve was also replaced after a small hole in the line was observed.
Zion 1 81-36 While testing 1A DG, an abnormal amount of lube oil was seen leaking from No. 6 right cylinder head covers as well as an unusual noise from the same cylinder.
The engine was manually shutdown.
The intake rocker arm broke due to binding between it and the rocker stand.
The engine was repaired and returned to service.
Zion 2 82-20 2B DG was declared inoperable when it failed to start.
A broken coupling between the camshaft and the starting air distributor prevented the engine from cranking.
The coupling was replaced.
Worthington Experience Cook 1 and 2 81-38 During a routine inspection of 1AB EDG, a taper pin 81-45 in the fuel ack assembly was found to be loose and 4
the pin was found to be broken.
All other taper pins
Plant LER (
Description a
of the Unit 1 EDG's were cracked and found to be tight.
Some loose pins were found on Unit 2 diesels and corrective actions were taken.
2.
TURB0 CHARGER SUBSYSTEM General Motors Experience
~
ANO 1 82-05 EDG 2 failed to start during testing.
The turbo-charger had failed and was replaced.
The failed turbocharger was returned to the manufacturer for repair.
An evaluation will be made to determine the root cause of failure and long term corrective action.
Beaver Valley 1 81-30 During monthly surveillance testing of 2 EDG, the
~
unit tripped apparently due to overspeed.
The diesel would not restart.
The diesel failure was attributed to a failed taper pin and bent lever in the governor.
The turbocharger also failed.
The turbocharger and taper pin and lever were replaced.
Saint Lucie 1 82-24 During testing, the IB EDG turbocharger failed.
Subsequent inspection revealed a deteriorated soak back oil pump not providing sufficient lubrication to the turbocharger guide and thrust bearing.
The turbocharger and oil pump were replaced.
. Saint Lucie 1 82-33 IB EDG turbocharger failed - caused by a broken coupling of soak back pump.
Plant LER #
Description Saint Lucie 1 81-47 During a retest following modifications,1A EDG turbocharger failed.
Cause not determined, however, in the weeks prior to failure there were approximately 60 engine starts and a great deal of light load operation associated with maintenance and modification retesting.
Cooper Bessemer Experience Zion 1 83-02 0 DG failed to accept greater than 507, load.
The turbocharger had seized which reduced load capacity to 50%.
The turbocharger was replaced.
(Second failure of this type since 1973.)
3.
FUEL OIL SUBSYSTEM Fairbanks Morse Experience Farley 2 81-13 On 5/5 and 5/6/81, DG 2B was declared inoperable due to a lube oil leak and a fuel oil leak respectively.
The cause of the lube oil leak was a leaking 0-ring on the lube oil strainer.
The cause of the fuel oil leak.was a fatigue failure, due to vibration of a compressor filling on a copper line.
The line was replaced with stainless steel and re-routed to reduce vibration.
1 i
Plant
._LER #
Description Cooper Bessemer Experience Cooper 81-21 During testing of 2 DG, an injector line failed.
Cause of failure of fuel injector supply line is believed to be metal fatigue and vibration.
Tubing severed completely near injector compressor fitting.
Component was replaced.
-4.
COOLING SUBSYSTEM Nordberg ' Experience Brunswick 1 82-78 2 EDG tripped due to low jacket water pressure.
The two dowel pins and eight capscrews in the flex drive coupling drive plate had sheared, allowing the piste to separate from the engine crankshaft.
A new drive plate and new dowel pins and capscrews were instal ed.
The same will be done for the other EDG's.
ALCO Experience
~
Salem 1 81-18 DG 1A was declared inoperable because of a cooling 81-53 83-04 water leak.
Similar occurrences:
76-12, 77-59, 77-77, 77-80, 80-02, 80-22, 80-31, 80-60 and 81-02.
The nipple connecting jacket water valve 1DA45A and jacket water pipe was cracked and leaking.
The nipple was replaced in kind.
-g-Plant LER #
Description 9
5.
DESIGN ERROR GM Experience Browns Ferry 1 83-09 Design review of EDG engine coolers showed that the coolers may not be capable of maintaining the engine cooling water below the 190*F hot alarm setpoint when the dieselis at full power.
Cooling water maximum temperature necessary to maintain a jacket water temperature of 190*F has been con-4 servatively calculated to be 76*F.
Apparent design error made in sizing engine coolers for inlet cooling water at 95'F.
Browns Ferry 1 83-24 Design review of EDG room ventilation showed system may not keep electrical components below maximum temperature limits when ambient air temperature is above 87.3*F and diesels are at full load.
Deflectors installed to direct exhaust air away from components.
Sequoyah 1 83-38 DG's would become inoperable when outside air temperature is greater than 88'F.
The heat load of the DG is higher than that originally used.
Fitzpatrick 82-39 EDG A & C were declared inoperable.
Overheating of l
a ventilation cowling within the generator due to 4
an original ~ design error was the cause.
Evaluation by v.endor and licensee continues.
l
^ g s
ENCLOSURE 4 I',
POPULATION DATA FOR DIESEL GENERATORS IN NUCLEAR SERVICE r.
t M
/
t l
l l
s'
Tablo 1 shows the population of non-TD1 diesel engines for licensed nuclear service as a function of generator output and engine manu fa cturer.
The data for non-TDI engines was taken from Table A.1 of NUREG/CR-2989 (which includes active LWR's licensed through 1981 with the exception of McGuire).
Although some discrepancies in the generator output rating were noted between information found in LERs, staff memoranda, NUREG/CR-1362 (" Data Summaries of Licensee Event Reports of Diesel Generators at U.S. Commercial Nuclear Power Plants,"
March 1980) and Table A.1, the discrepancies do not significantly affect the general distribution of the population.
Table 1 Generator Output (XW) 500-1750-2500-3500-4000-Manufacturer Total
<500 1749 2499 3499~
3999 4999
>5000 General Motors 52 3
8 41 Schoonmaker 10 8
2 Bruce 4
4 Fairbanks Morse 42 40 2
Alco 18 8
10 Cooper Bessemer 7
7 Worthington 4
4 Nordberg 4
4 Allis Chambers 2
2 Caterpillar 1
1 Total 144 6
0 24 97 8
9 Transamerica Delaval 7
3 4
As Table 1 shows, most of the diesel engine experience from manufacturers, other than TDI,is with engines having output ratings between 1750 KW and 3000 KW.
By contrast the TDI engines are generally much larger machines.
l
-