ML030730771
| ML030730771 | |
| Person / Time | |
|---|---|
| Site: | Salem  |
| Issue date: | 03/14/2003 |
| From: | Lanning W Division of Reactor Safety I |
| To: | Keiser H Public Service Enterprise Group |
| References | |
| IR-02-010 | |
| Download: ML030730771 (22) | |
See also: IR 05000272/2002010
Text
March 14, 2003
Mr. Harold W. Keiser
Chief Nuclear Officer and President
P. O. Box 236
Hancocks Bridge, NJ 08038
SUBJECT: SALEM GENERATING STATION - NRC SPECIAL INSPECTION REPORT
50-272/02-010, 50-311/02-010
Dear Mr. Keiser:
On January 30, 2003, the U.S. Nuclear Regulatory Commission (NRC) completed a special
inspection at your Salem Generating Station (SGS) Unit 1. The enclosed report documents the
inspection findings which were discussed on January 30, 2003, with Mr. J. Carlin and other
members of your staff.
This special team inspection reviewed your actions regarding repetitive fuel oil leaks on the
emergency diesel generators (EDGs) at SGS and the failure of the turbocharger on the 1C
EDG. Given that the turbocharger was replaced and the 1C EDG was returned to service, we
do not perceive this to be an immediate safety issue. However, given that the cause of the
failure of the turbocharger has not yet been determined, we are concerned that the possibility
exists that the failure mechanism could apply to other EDGs at the site.
Based on the results of this inspection, a non-cited violation was identified and is discussed in
Section 4OA2.1 of the report. Because of its very low safety significance, and being entered
into your corrective actions program, this matter is not being cited for formal enforcement, in
accordance with Section VI.A of General Statement of Policy and Procedures for NRC
Enforcement Actions, NUREG 1600. If you deny this non-cited violation, you should provide a
response with basis for your denial, within 30 days of the date of this inspection report, to the
Nuclear Regulatory Commission, ATTN: Document Control Deck, Washington, DC 20555-001;
with copies to the Regional Administrator, Region I; the Director, Office of Enforcement, United
States Nuclear Regulatory Commission, Washington, DC 20555-0001; and the NRC Resident
Inspector at the Salem Generating Station
In accordance with 10 CFR 2.790 of the NRCs Rules of Practice, a copy of this letter and its
enclosure will be available electronically for public inspection in the NRC Public Document
Room or from the Publicly Available Records (PARS) component of NRCs document system
(ADAMS). ADAMS is accessible from the NRC website at http://www.nrc.gov/public-reading-
room/adams.html (the Electronic Public Reading Room).
Mr. Harold W. Keiser 2
If you have any questions, please contact James C. Linville of my staff at 610-337-5129.
Sincerely,
/RA/
Wayne D. Lanning, Director
Division of Reactor Safety
Docket Nos.: 50-272; 50-311
Enclosure: Inspection Report 50-272/02-010 and 50-311/02-10
cc w/encl:
M. Friedlander, Director - Business Support
J. Carlin, Vice President - Engineering
D. Garchow, Vice President - Projects and Licensing
G. Salamon, Manager - Licensing
T. O'Connor, Vice President - Operations
R. Kankus, Joint Owner Affairs
J. J. Keenan, Esquire
Consumer Advocate, Office of Consumer Advocate
F. Pompper, Chief of Police and Emergency Management Coordinator
M. Wetterhahn, Esquire
State of New Jersey
State of Delaware
N. Cohen, Coordinator - Unplug Salem Campaign
E. Gbur, Coordinator - Jersey Shore Nuclear Watch
E. Zobian, Coordinator - Jersey Shore Anti Nuclear Alliance
Mr. Harold W. Keiser 3
Distribution w/encl:
Region I Docket Room (with concurrences)
R. Lorson, DRP - NRC Resident Inspector
H. Miller, RA
J. Wiggins, DRA
G. Meyer, DRP
S. Barber, DRP
W. Kane NRR
B. Sheron, NRR
D. Dambly, OGC
D. Holody, EO, RI
R. Urban, ORA, RI
F. Congel, OE (OEMAIL)
S. Figueroa, OE (SLF)
T. McGinty, OEDO
J. Clifford, NRR
DOCUMENT NAME: C:\ADAMS\Cache\ML0307307710.wpd
After declaring this document An Official Agency Record it will be released to the Public.
To receive a copy of this document, indicate in the box: "C" = Copy without attachment/enclosure "E" = Copy with attachment/enclosure "N" = No copy
OFFICE RI/DRS RI/DRS RI/DRS RI/DRP RI/DRS
NAME RFuhrmeister JLinville ECobey GMeyer WLanning
DATE 03/07/03 03/14/03 03/13/03 03/13/03 03/13/03
OFFICIAL RECORD COPY
U.S. NUCLEAR REGULATORY COMMISSION
REGION I
Docket Nos.: 50-272, 50-311
Report No.: 50-272/02-010 and 50-311/02-010
Facility: Salem Generating Station, Units 1 and 2
Location: PO Box 236
Hancocks Bridge, NJ 08038
Dates: September 16, 2002 - January 30, 2003
Inspectors: R. Fuhrmeister, Senior Reactor Inspector, Electrical Branch
R. Barkley, Senior project Engineer, Projects Branch 3
K. Young, Reactor Inspector, Electrical Branch
T. OHara, Reactor Inspector, Electrical Branch
Approved by: James C. Linville, Chief
Electrical Branch
Division of Reactor Safety
i
SUMMARY OF FINDINGS
IR 05000272/2002-010, 05000311/2002-010; 09/16/2002 - 01/30/2003; Salem Generating
Station, Units 1 and 2; Special Team Inspection.
The report covered a special inspection of PSEG Nuclear (PSEG) actions in response to
repetitive fuel oil leaks and the failure of the turbocharger on the 1C emergency diesel
generator (EDG). One Green non-cited violation and one unresolved item with potential safety
significance greater than Green were identified. The significance of most findings is indicated
by their color (Green, White, Yellow, Red) using IMC 0609, Significance Determination
Process (SDP). Findings for which the SDP does not apply may be Green or be assigned a
severity level after NRC management review. The NRCs program for overseeing the safe
operation of commercial nuclear power reactors is described in NUREG-1649, Reactor
Oversight Process, Revision 3, dated July 2000.
A. Inspector Identified and Self-Revealing Findings
Cornerstone: Mitigating Systems
@ GREEN. A non-cited violation of 10 CFR Part 50, Appendix B, Criterion XVI, Corrective
Action, was identified for the failure to identify the cause and prevent recurrence of
EDG injection pump fitting fuel oil leaks. This resulted in repetitive EDG outages to
repair fuel oil leaks. (Section 4OA2.1)
This finding is greater than minor because it adversely affected the equipment
performance attribute of the mitigating systems cornerstone objective, in that the
necessary repair activities for these fuel oil leaks resulted in an increased unavailability
for the EDGs. The finding was of very low safety significance because the fuel oil leaks
did not result in an actual loss of safety function for the EDGs.
@ TBD. Ineffective implementation of corrective actions failed to prevent a recurrence of
EDG turbocharger failure.
Failure to identify the cause and prevent recurrence of the EDG turbocharger failures
was a performance deficiency. This finding is greater than minor because it adversely
affected the equipment performance attribute of the mitigating systems cornerstone
objective, in that, the 1C EDG was incapable of performing its safety function for a
period of time in excess of its technical specification allowed outage time. This finding
was determined to have potential safety significance greater than very low because the
likelihood of core damage due to a loss of AC power was significantly increased while
the 1C EDG was not available to mitigate a loss of offsite power event. This finding is
unresolved pending completion of a significance determination.
B. Licensee-Identified Violations
None
ii
Report Details
4. OTHER ACTIVITIES [OA]
4OA2 Problem Identification and Resolution
.1 Fuel Oil Leaks From Fuel Injector Pump Inlet Fittings
a. Description and Chronology of Condition
The Salem Generating Station (SGS) emergency diesel generators (EDGs) experienced
a number of fuel oil leaks from the injection pump inlet fittings. These fittings are often
referred to as banjo bolts due to their physical configuration resembling a bolt through
the body of a banjo. The 1C EDG suffered four separate leaks on the 1R cylinder banjo
bolt during the time period from April 2002 through September 2002 (April 15, July 8,
September 2, September 7). On July 8, 2002, the leak rate was estimated to be about
1-2 liters per minute (lpm), on September 2 about 50 gallons over a one hour period,
and on September 7, 2002, about one lpm. On each occasion, PSEG Nuclear (PSEG)
repaired the leak by installing new concentric grooved copper gaskets.
Historically, the Salem Unit 1 EDGs have experienced numerous fuel oil leaks at the 1,
2, 8 and 9 cylinder locations on both the left and right cylinder banks. Leakage at this
location has been a known problem for ALCO engines in both nuclear and commercial
service for some years. In an attempt to resolve the problem, Fairbanks Morse Engine
Division of Coltec Industries changed the gasket design in February 1998, and again in
July 1999. The original design was a flat, dead-soft copper gasket (Type 1), the first
change was to a concentric grooved copper gasket (Type 2), and the second change
was to a steel-backed O-ring gasket (Type 3).
All Salem Unit 1 EDG banjo bolt gaskets were replaced with the concentric grooved
design in 1999. Based on field observations and work order reviews, PSEG believes
that the Salem 2 EDGs have the original design flat gaskets installed. Since 1999, the
Salem 1 EDGs have experienced 20 banjo bolt fuel oil leaks. The Salem 2 EDGs have
experienced 1 such leak.
b. Probable Contributing Causes
1. Inspection Scope
This inspection was conducted in accordance with NRC Inspection Procedure (IP)
93812, Special Inspection, to assess PSEGs response to a continuing problem of fuel
oil leaks from the fuel injector pump inlet fitting (banjo bolt) on the 1C EDG 1R cylinder.
The team assessed PSEGs actions and evaluation of the banjo bolt fuel oil leaks as
they apply to all the Salem Generating Station EDGs.
The team reviewed applicable sections of the Updated Final Safety Analysis Report
(UFSAR), Technical Specifications (TS), engineering evaluations of previous events,
and held discussions with various engineering disciplines, maintenance, operations, and
site managers to determine the technical and regulatory aspects of the banjo bolt fuel oil
leaks.
2
The team reviewed applicable documents generated to address the banjo bolt fuel oil
leaks, subsequent trouble shooting and corrective actions. Additionally, the team
interviewed plant personnel involved with the root cause analysis team and personnel
associated with the testing, troubleshooting and repair activities.
The team reviewed some aspects of the maintenance history of the EDG to determine
the extent of the banjo bolt fuel oil leaks. The review included: a list of all work orders
for EDG work since January 1998, and a list of notifications generated for Unit 1 and
Unit 2 EDGs. The documents reviewed are listed under Supplemental Information in
this report.
The team reviewed the EDG vendor manual and a sample of preventive maintenance
procedures to determine if PSEG was maintaining their vendor manual current and that
the sampled maintenance procedures reflected the current guidance in the vendor
manual.
The team assessed the testing of the 1A and 1C EDGs during the week of September
16, 2002, to ascertain the operabillty of the diesel generators.
2. Findings
Introduction. The team identified a Green NCV for the failure to identify the cause and
prevent the recurrence of a significant condition adverse to quality, as required by
Criterion XVI of Appendix B to Part 50 of Title 10 of the Code of Federal Regulations (10 CFR 50).
Description. PSEG corrective actions for banjo bolt leaks have historically consisted of
retorquing the banjo bolt. Since the installation of the concentric grooved gaskets in
1999, the Salem Unit 1 EDGs have experienced approximately twenty instances of
banjo bolt leakage on sixteen separate occasions. The leaks have been anywhere from
several drops per minute up to an estimated one gallon per minute. Between April and
September 2002, the 1C EDG experienced four leaks at the same location, the banjo
bolt connection for the 1R fuel injector inlet.
PSEG initiated a Transient Assessment Response Plan (TARP) Team to evaluate the
problem after the July 8, 2002, 1C EDG leak. The TARP team goals focused on
repairing the leak, restoring the engine to service, and determining if any other similar
components warranted test or inspection. Based on a determination that this type of
leak was common throughout the industry, the TARP Team recommended running the
other Salem 1 EDGs and checking for leakage. The TARP Team determined that the
cause of the leak was a relaxation of the pre-load torque on the banjo bolt. The torque
relaxation was attributed to the bolt backing out under vibration during engine operation.
The banjo bolts for all Salem 1 EDGs and the Salem 2 A and B EDGs were retorqued
while the engine was running. This activity was scheduled for the Salem 2 C EDG
during its next run.
PSEG conducted a formal root cause evaluation of the condition following the
September 7, 2002, 1C EDG fuel oil leak. PSEG determined the cause of the leaks to
be the inability of the concentric grooved copper gasket to maintain preload when
3
compressed by the fuel supply tube under thermal transients and misalignment of the
fuel supply tube due to manufacturing tolerances. Proposed corrective action was to
replace the concentric grooved gaskets with the steel-backed O-ring gaskets during the
1R15 outage. Interim action called for retorquing the banjo bolts during monthly
surveillance test runs. The root cause team identified several contributing causes,
including: failure to implement the 6 month hot retorque preventive maintenance
developed as a previous corrective action in early 1999, and organization and staffing
changes during the period 2000 through 2002, which permitted 16 banjo bolt leaks to
occur without review or evaluation by engineering personnel.
Analysis. The failure to identify the cause and prevent recurrence of EDG injection
pump fitting fuel oil leaks was a performance deficiency. This finding is greater than
minor because it adversely affected the equipment performance attribute of the
mitigating systems cornerstone objective, in that, the necessary repair activities for
these fuel oil leaks resulted in an increased unavailability for the EDGs. Using Phase 1
of the Significance Determination Process (SDP) for Reactor At-Power Situations, the
inspectors determined that this finding is of very low safety significance (Green)
because the fuel oil leaks did not result in an actual loss of safety function of the EDGs.
Enforcement. Criterion XVI, Corrective Actions, of Appendix B to 10 CFR 50 requires
that for significant conditions adverse to quality, the cause of the condition be
determined and corrective actions be taken to prevent repetition. Contrary to the above,
prior to September 2002, PSEG Nuclear failed to determine the cause of repetitive fuel
oil leaks on the banjo bolt connections, and failed to prevent repetition of the leaks.
Because this failure to implement effective corrective actions was of very low
significance and has been entered into PSEGs corrective action program, this violation
is being treated as an NCV, consistent with Section VI.A of the NRC Enforcement
policy: NCV 50-272/02-10-01, Failure to Implement Effective Corrective Actions.
4OA3 Event Followup
.1 Failure of Emergency Diesel Generator 1C Turbocharger
a. Description of Event
On September 13, 2002, during a 24-hour endurance run, the 1C EDG experienced a
failure of its turbocharger. This is the same EDG which had been experiencing
repetitive fuel oil leaks as discussed in Section 4OA1 of this report. Approximately 45
minutes into the loaded run, personnel in the EDG control room heard a loud noise in
the diesel room, and observed a drop in the engine load. Personnel on site observed
black smoke coming from the 1C EDG exhaust stack. PSEG personnel performed a
field inspection and identified damage to the turbocharger and associated air inlet piping
as well as debris scattered around the room.
PSEG assembled a TARP Team and a root cause evaluation team to investigate the
cause of the failure and develop a corrective action plan. PSEG replaced the
turbocharger, retested the 1C EDG, and restored the 1C EDG to operable status.
Examination of the failed turbocharger rotating element showed that an inducer blade
suffered a fatigue failure. The failure initiated at a pit like surface defect on the concave
4
side of the blade along the hub to blade fillet. At the end of the inspection, PSEG was
still attempting to obtain additional design information on the turbochargers in order to
determine why the blade experienced a failure. PSEG was also making preparations to
have a turbocharger dynamically tested by the EDG manufacturer.
1. Inspection Scope
The team assessed PSEGs immediate actions and subsequent evaluation of the EDG
turbocharger failure.
The team reviewed applicable section of the UFSAR, TS, engineering evaluations of
previous events, and held discussions with various engineering disciplines,
maintenance, operations, and site managers to determine the technical and regulatory
aspects of the 1C EDG turbocharger failure.
The team reviewed applicable documents generated to address the 1C EDG
turbocharger failure, subsequent trouble shooting and corrective actions. Additionally,
the team interviewed plant personnel involved with the root cause analysis team and
personnel associated with the testing, troubleshooting and repair activities and
interviewed plant management knowledgeable of the decision making processes that
transpired during the 1C EDG turbocharger failure event.
The team reviewed some aspects of the maintenance history of the EDGs to determine
if there were previous failures similar to the September 13, 2002, 1C EDG turbocharger
failure. The review included: a list of all work orders for EDG work since January 1998,
a list of notifications generated for Unit 1 and Unit 2 EDGs. The documents reviewed
are listed under Supplemental Information in this report.
The team reviewed the EDG vendor manual and a sample of preventive maintenance
procedures to determine if PSEG was maintaining the EDG vendor manual current and
that the sampled maintenance procedures reflected the current guidance in the vendor
manual.
The team assessed the testing of the 1A and 1C EDGs during the week of September
16, 2002, to ascertain the operabillty of the diesel generators.
The Special Inspection Charter required the team to confirm the adequacy of PSEGs
investigation, root cause evaluation, planned corrective actions and extent of condition
review for the 1C EDG turbocharger failure on September 13, 2002. PSEG had not
completed these activities at the end of the inspection. These activities will be reviewed
after completion by PSEG. Pending review of the completed PSEG actions, this item is
identified as URI 50-272/02-10-02, Evaluate PSEG Actions for Turbrocharger Failure.
2. Findings
Introduction. An unresolved item was identified relating to ineffective corrective actions
for earlier turbocharger failures which did not prevent subsequent failures. The finding
is under the mitigating systems cornerstone, and has potential safety significance
greater than green because the likelihood of core damage due to a loss of all AC power
5
was significantly increased while the 1C EDG was not capable of mitigating a loss of
offsite power event.
Description. The inspectors verified that Salem EDGs have experienced previous
turbocharger failures. PSEG identified previous fatigue failures of both the compressor
and turbine blading of the turbochargers. PSEG developed a new maintenance
procedure after the 1990 inducer blade failure which required either magnetic particle or
dye penetrant inspection of the rotating element every four refueling outages. This work
was never scheduled. In 2000, the portion of the procedure requiring the non-
destructive examination (NDE) was changed to Detailed Turbocharger Inspection, and
a note was added that this should only be performed upon direction from Engineering.
The every four refueling outage turbocharger inspection procedure never had a
requirement to perform NDE. As a result, no NDE was performed by PSEG on the
rotating elements of the turbochargers.
Following the 1998 turbocharger turbine blade fatigue failure, PSEG implemented
vibration monitoring for the diesel generator turbochargers. PSEG had some vibration
data, which was recorded during each operation of the EDG, but did not have a
comprehensive monitoring and analysis program which could indicate degrading
turbocharger performance. The data was inconsistent and did not show a readily
identifiable trend. This is due, in part, to the data not being recorded at a consistent
engine operating condition.
Oil analyses did exist which indicated increasing wear products in the 1C EDG oil,
however the data was not forwarded to the site for review in a timely manner. There
does not appear to be a comprehensive program of oil sampling and analysis,
completed in a timely manner, to predict the onset of problems with the EDGs.
Analysis. The failure to identify the cause and prevent recurrence of the EDG
turbocharger failures was a performance deficiency. This finding is greater than minor
because it adversely affected the equipment performance attribute of the mitigating
systems cornerstone objective, in that, the 1C EDG was incapable of performing its
safety function for a period of time in excess of the TS allowed outage time. This finding
was determined to have potential safety significance greater than very low significance
because the likelihood of core damage due to a loss of all AC power was significantly
increased while the 1C EDG was not capable of mitigating a loss of offsite power event.
Enforcement. Criterion XVI, Corrective Actions, of Appendix B to 10 CFR 50 requires
that for significant conditions adverse to quality, the cause of the condition be
determined and corrective actions be taken to prevent repetition. Contrary to the above,
prior to September 2002, PSEG Nuclear failed to implement effective measures to
prevent repetition of the turbocharger inducer failures, e.g.: NDE not performed,
ineffective vibration monitoring, and untimely reporting of oil analysis results. Pending
receipt of information to permit performance of a significance determination, this item is
identified as URI 50-272/02-10-03, 1C EDG Turbocharger Failure SDP.
4OA5 Other Activities
6
.1 (Closed) Unresolved Item 50-311/2002-06-02: Failure to Report Fault Exposure Hours
for the 2B EDG in January-February 2002
Introduction. An unresolved item was identified regarding fault exposure hours for an
EDG. The issue was unresolved pending determination if there was a performance
issue.
Description. This unresolved item involved the need for NRC review of 314.2 fault
exposure hours for the 2B EDG in February 2002. Under NEI 99-02, Revision 2,
effective January 1, 2002, such fault exposure hours are not included in the PI
calculation, but are required to be reported to the NRC in the PI comment field. The
NRC will then evaluate whether the incident was caused by a performance deficiency
and evaluate the risk significance of the event using the Significance Determination
Process (SDP). In the April 2002 PI submittal to the NRC, PSEG failed to report these
hours as required; these fault exposure hours were subsequently reported to the NRC in
the July 2002 PI submittal. PSEG indicated that the failure to include this information
was due to an personnel oversight by the staff who collates this data due to the recent
change in the NEI 99-02 reporting rules.
Analysis. The inspectors reviewed the technical details surrounding this event.
Specifically, the EDG was removed from service on February 12, 2002, approximately
13 hours1.50463e-4 days <br />0.00361 hours <br />2.149471e-5 weeks <br />4.9465e-6 months <br /> into a 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> run of the engine due to a fuel oil leak at the upper part of the
lower mechanical connection of the fuel oil supply line from the fuel oil booster pump
into the No. 3 Left cylinder. This leak could not be terminated by maintenance
personnel tightening the fitting. This short section of fuel line is often referred to as the
gooseneck due to its long, curved shape. While PSEG indicated that this unquantified
leak was not a significant fire hazard and the spray from the gooseneck was contained
by an equipment operator placing a rag over the leak, PSEG elected to shutdown the
EDG as a precaution and repair the leak versus continue the test to its conclusion.
Following replacement of the gooseneck three hours after engine shutdown, the EDG
was satisfactorily retested.
The inspector reviewed the maintenance history on the EDGs for the last three years.
The last time the fitting was disturbed on this engine was sixteen months prior to this
test to inspect the fuel injectors. In the past, PSEG has noted leakage at the lower part
of this fitting during the post-maintenance testing run but not during operability runs; the
leakage was readily corrected by tightening the fitting. No leaks of the type noted in the
February 12, 2002, failure were noted by PSEG nor evident during the inspectors
records review. Due to indications by equipment operators that the leak did not
functionally impact engine operation and engineering information that the EDG is
capable of performing its intended function with a loss of fuel to this one engine cylinder,
this failure was not classified by PSEG as a functional failure per its maintenance rule
program. PSEG concluded, and the inspector agreed, that this condition did not
represent a performance deficiency and thus would not be a finding that the NRC
processed through the SDP.
.2 Fault Exposure Hours Incurred on the 2A EDG in January 2002
a. Inspection Scope
7
In PSEGs July 2002 PI submittal, PSEG acknowledged that it also failed to report 16.1
fault exposure hours on the 2A EDG for the reasons mentioned above. While the short
duration of this fault exposure time indicated that the risk significance of this event was
very low (i.e., Green), the inspectors reviewed the technical details surrounding this
event to determine whether a performance deficiency existed in this matter.
On January 9-10, 2002, during an unloaded post-maintenance run of the 2A EDG,
excessive oscillations of the fuel rack linkage and generator field voltage were observed.
The cause of these oscillations were later attributed to a poor connection from the
potential transformer which converts the generator terminal voltage to a manageable
control signal. During maintenance activities, a drawer must be removed and
reinserted. During this evolution, the output signal from the potential transformer (PT) is
connected to the external circuitry via connectors that must be carefully aligned. The
connector age, relatively poor design and reliance upon precise drawer alignment to
prevent connector damage contributed to this problem. This failure was classified as a
system functional failure since it impacted the operability of the EDG. The inspector
noted that a similar problem was noted on the 1B EDG approximately one week later.
Following this event, PSEG initiated corrective actions to ensure that any time the PT
drawer is opened, a maintenance continuity check of the PT contacts is performed. The
PT connectors were also inspected on the other EDGs. In addition, a minor design
change package was approved to replace these connectors with more reliable
connectors such that this problem would not recur. As of the date of this report, the
DCP has been completed on the majority of the EDGs.
A review of PSEG maintenance records for the prior three years did not indicate a
similar problem with these connectors, although there were anecdotal references to
problems with these connectors in the 1993-1995 time frame. Existing procedural
requirements ensure that any time this drawer is opened, EDG operability is
subsequently demonstrated via running and loading the EDG. In both instances of EDG
oscillations in January 2002, the problems with the PT connection were identified by
PSEG during EDG testing prior to declaring the EDG operable. Thus based on existing
controls to ensure the EDGs were operable following manipulation of the PT drawer,
and PSEGs planned and completed corrective actions based on recent problems with
the connectors in question, no performance deficiency was identified in this matter.
b. Findings
No findings of significance were identified.
4OA6 Meetings, Including Exit
.1 Exit Meeting Summary
On January 30, 2003, the NRC team presented the inspection results to Mr. J. Carlin
and other members of the PSEG staff.
8
Some of the information reviewed by the team was identified as proprietary. All
proprietary documents were returned to PSEG Nuclear at the end of the inspection.
9
CHRONOLOGY OF EVENTS
Turbocharger Failures
1973 EDG 2B - Compressor Blade Cracking During Factory Testing
1977 EDG 1B - Turbine Blade Failure
1987 EDG 2C - Compressor Blade Failure
1990 EDG 2B - Compressor Blade Failure
1998 EDG 2A - Turbine Blade Failure
9/13/2002 EDG 1C - Compressor Blade Failure
Turbocharger Replacement History
Unit 1
EDG 1A - 12/95 EDG 1B - 4/01 EDG 1C - 2/91
Unit 2
EDG 2A - 2/98 EDG 2B - 12/94 EDG 2C - 4/02
Banjo Bolt Fuel Oil Leaks and Turbocharger Issues
1987 - 1998 Corrective action documents revealed various banjo bolt fuel leaks at Salem.
Examples include the following:
EDG 2C - Cylinder 9R Fuel Oil Leak, Replaced Fuel Pump.
EDG 2B - Fuel Oil Leaks
EDG 2C - Fuel Leak on an Injector
1991 Vendor specifies banjo bolt torque value of 225 ft/lbs.
Unknown Vendor changed banjo bolt torque value from 225 to 125 ft/lbs.
1998 Vendor changed banjo bolt torque value from 125 to 150 ft/lbs.
EDG 1B - Fuel oil leak was noted on the 9L cylinder.
PSEG established a repetitive task for hot re-torque of banjo bolts every 6
months.
10
11/4/98 EDG 1C - Cylinder 1R, fuel oil leaking at copper gaskets on supply side of the
fuel injection pump. Personnel could not tighten bolt to stop leak. Personnel
estimated fuel leak rate to be approximately 22 drops per minute.
9/22/99 EDG 1C - PSEG replaced all copper gaskets with concentric grooved copper
gaskets. (Vendor recommended replacement.)
3/20/00 EDG 1C - Cylinders 1L and 2L fuel injectors were found leaking 1 or 2 drops per
minute.
2000 Several notifications (20009208, 20021598, and 20022490) were generated
regarding increased vibration trending for EDG 1C turbocharger.
1/27/01 EDG 1C - Cylinder 2R fuel oil drip observed.
1/29/01 EDG 1C - Cylinders 9L and 8L had fuel leaks after 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> run (approx. 250 ml)
1/25/02 Notification (20089940) described that over the last 2 months, the NBU has been
severely challenged with repetitive operability issues with the EDGs at Salem
and Hope Creek.
4/15/02 EDG 1C - Cylinder 1R banjo bolt fuel oil leak occurred. Approximately 1 to 1.5
pints of fuel oil leaked during one hour of operation
5/9/02 EDG 1C - Cylinder 1R leaks during surveillance run.
5/16/02 EDG 1C - Maintenance replaced cylinder 1R banjo bolt copper gasket only and
applied 145 foot pounds of torque to all of the bolts.
6/10/02 EDG 1C - No leaks were experienced during one month surveillance run.
7/8/02 EDG 1C - Cylinder 1R leaks at approximately one liter per minute. PSEG
replaced the banjo bolts, copper gaskets, fuel supply inlet bolts and fuel supply
inlet flexitallic gaskets. Additionally, the fuel injection pump was realigned.
PSEG performed post maintenance testing, the monthly surveillance test,
inspection and hot re-torque of the banjo bolts. All items were completed
satisfactorily. Turbocharger data showed increased vibration levels.
8/5/02 EDG 1C - No leaks occurred during one month surveillance run.
9/2/02 EDG 1C - During a 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> surveillance run, cylinder 1R leaked. This leak
created a 9 foot by 3 foot puddle of fuel oil on the floor as observed by an
operator. PSEG declared the EDG inoperable. PSEG replaced banjo bolt
washers and performed re-torque of all banjo bolts. Additionally, the fuel injector
inlet pipe was reversed. PSEG began an eight hour run on EDG 1C following
hot re-torque on 1R cylinder banjo bolt to prove operability. The eight hour run
was successful. No leaks were observed during eight hour run.
9/3/02 Completed the eight hour run an declared EDG 1C operable.
11
9/5-6/02 EDG 1C was tagged out for scheduled maintenance.
9/7/02 EDG 1C was started for post-maintenance retests. Operator noted cylinder 1R
was leaking from repaired banjo bolt. PSEG tagged out EDG 1C and replaced
fuel injector pump, fuel supply tube, fuel inlet bracket, banjo bolt, copper gaskets
and snubber valve.
9/8/02 PSEG started EDG 1C and successfully run. EDG 1C was declared operable.
9/11/02 PSEG ran EDG 1C for approximately 2.5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br /> to obtain engine operating data
for use in root cause determination. No leakage or banjo bolt movement was
noted during run.
9/13/02 At approximately 1030, EDG 1C was started for a 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> endurance run. No
leakage or bolt movement was noted for approximately one hour into the run. At
approximately 1130, EDG 1C apparently experienced a turbocharger failure.
The failure was characterized by a loud noise, loss of load and smoke. PSEG
tripped the EDG.
9/14-15/02 Replaced failed turbocharger with spare stored in warehouse. PSEG inspected
the failed unit and established a root cause team to perform a root cause
evaluation. Began 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> run following turbocharger replacement.
9/16/02 Completed 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> endurance of EDG 1C after replacing turbocharger. No
banjo leaks were noted. PSEG sent the failed turbocharger to lab for analysis.
The preliminary cause of the failure was a fatigue induced fracture of the
compressor blade.
12
SUPPLEMENTAL INFORMATION
KEY POINTS OF CONTACTS
J. Carlin V.P. Engineering
D. Carpenter Maintenance Manager
G. Delp Supervisor of Preventive Maintenance Manager
M. Dzuba Training Maintenance Manager
V. Fragonese Inspection Support Manager
C. Fricker Operations Manager
R. Labott Principal Engineer (Metallurgist)
M. Mosier Licensing Engineer
J. Nagel Licensing Supervisor
G. Salamon Licensing Manager
M. Schimmel Maintenance Manager
NRC Personnel
F. Bower, Resident Inspector, Salem
M. Ferdas, Resident Inspector, Hope Creek
W. Lanning, Director, Division of Reactor Safety
J. Linville, Branch Chief, Electrical Branch, DRS
R. Lorson, Senior Resident Inspector, Salem
ITEMS OPENED, CLOSED, AND DISCUSSED
Opened and closed
50-272, 311/02-10-01 NCV Failure to implement corrective actions for repetitive fuel oil
leaks on EDGs
Open
50-272, 311/02-10-02 URI Evaluate PSEG Actions for Turbocharger Failure
50-272, 311/02-10-03 URI Failure to implement effective corrective actions for
turbocharger failures
Closed
50-311/2002-06-02 URI Failure to report fault exposure hours on 2B EDG
13
LIST OF ACRONYMS
EDG Emergency Diesel Generator
IP Inspection Procedure
lpm Liters per Minute
NCV Non-Cited Violation
NDE Non-Destructive Examination
PSEG Public Service Electric and Gas
PT Potential Transformer
SDP Significance Determination Process
SGS Salem Generating Station
TARP Transient Assessment Response Plan
TBD To Be Determined
TS Technical Specifications
UFSAR Updated Final Safety Analysis Report
URI Unresolved Item
14
DOCUMENTS REVIEWED
Procedures
SC.MD-PM.DG-0003(Q), Diesel Generator Every Two Refueling Preventive Maintenance,
Rev. 12
SC.MD-PM.DG-0004(Q), Diesel Generator Every Four Refueling Preventive Maintenance,
Rev. 8
SC.MD-PM.DG-0007(Q), Diesel Generator Turbocharger Inspection and Changeout, Rev. 2
SC.MD-PM.DG-0032(Q), Periodic Diesel Engineer Inspection Maintenance, Rev. 5
Notifications
20009208 20010710 20021598 20022490 20023817 20032157
20047937 20060775 20063061 20089027 20089044 20089055
20089940 20095790 20096933 20105157 20105160 20105379
20106767 20108436 20111656 20111721 20111891 20112191
20112377 20112907 20112985 20091260 20013354 20013941
20095160 20008745 20088423 20088423 20088443 20088721
20088861 20089841 20089940 20097502 20097946
Condition Reports
981222221
Work Orders
CM990225172 PM000107013 60027856 60031656 30000139
60026149 70022345 70024517 70022455
Miscellaneous
EB-11267, Equipment Bulletin - Fuel Pump Inlet Gasket, 7/99
EDG Mechanics Qualifications List for Salem and Hope Creek, 9/17/02
List of Notifications for EDGs Unit 1 & Unit 2, 1999 - 2002
Summary of Known ALCO 251 Engine Turbo Charger Failures
Summary of 1C DG Fuel Leaks, July 1999 - June 2002
NEI 99-02, Revision 2, Regulatory Assessment Performance Indicator Guideline
Technical Specification for Unit 1, Salem Generation Station
UFSAR, Salem Generating Station
TARP Report, Salem Diesel 1C Fuel Oil Leakage Notification 20112191, 9/27/2002
MPR Letter, ALCO Owners Group Revised Maintenance Instructions, 4/16/1997
PSEG Letter, No. 2 Unit 2B Diesel Turbocharger Blade Failure, Salem Generation Station,
8/8/90
301103, Vendor Manual for ALCO Engines, 8/27/91
September 16, 2002
MEMORANDUM TO: James C. Linville, Team Manager
Division of Reactor Safety
Roy L. Fuhrmeister, Team Leader
Special Inspection
FROM: Wayne D. Lanning, Director /RA by J. R. White Acting for/
Division of Reactor Safety
SUBJECT: SPECIAL INSPECTION CHARTER - SALEM NUCLEAR
GENERATING STATION
A special inspection has been established to inspect and assess fuel oil leak on emergency
diesel generators (EDGs) that occurred at Salem Nuclear Generating Station, Unit 1 between
September 2 and 7, 2002 and the turbo charger failure that occurred on the Salem 1C EDG on
September 13, 2002. The special inspection team will include:
Manager: James C. Linville, Chief, Electrical Branch, DRS
Leader: Roy L. Fuhrmeister, Senior Reactor Inspector, DRS
Members: Keith A. Young, Reactor Inspector, DRS
Richard S. Barkley, Sr. Project Engineer, DRP
Eugene W. Cobey, Senior Reactor Analyst, DRS
(partime)
This special inspection is in response to notification by telephone on September 3, 2002 from
the senior resident inspector of the third fuel oil leak on EDG 1C over a five month period. A
fourth leak occurred on September 7, 2002. The basis for the special inspection is to monitor
and assess PSEG Nuclears root cause evaluation and corrective actions, independently
evaluate the risk significance of the EDG fuel oil leaks, and determine possible generic
implications. In addition, the team will monitor and assess PSEG Nuclears root cause
evaluation and corrective actions for the EDG 1C turbo charger failure which occurred on
September 13, 2002. The team will evaluate its risk significance and determine possible
generic implications.
The special inspection was initiated in accordance with NRC Management Directive 8.3, NRC
Incident Investigation Program. The inspection will be performed in accordance with the
guidance of Inspection Procedure 93812, Special Inspection. The report will be issued within
45 days following the exit for the inspection. If you have questions regarding the objectives of
the attached charter, please contact James Linville at (610) 337-5129.
Attachment: Special Inspection Charter
Distribution w/Attachment:
H. Miller, RA
J. Wiggins, DRA
S. Richards, NRR
A. Blough, DRP
R. Crlenjak, DRS
J. Linville, DRS
G. Meyer, DRP
R. Lorson, DRP
E. Cobey, DRS
H. Nieh, OEDO
S. Richards, NRR
Special Inspection Charter
Salem Nuclear Generating Station
EDG 1B & 1C Fuel Oil Leaks
Regarding the fuel oil leaks which occurred on the Salem EDG 1C on September 2 & 7, 2002,
and on Salem EDG 1B on September 3, 2002, and the 1C turbo charger failure on September
13, 2002, the special inspection should:
! Confirm the adequacy of PSEGs investigation and root cause evaluation.
! Confirm the adequacy of PSEGs planned corrective actions and extent of condition
review.
! Evaluate the effectiveness of prior corrective actions to the EDG 1C fuel oil leaks in May
and July 2002.
! Review the PSEG risk analysis, including the bases for their determination that the
EDGs remained available to fulfill safety functions for all initiating events..
! Evaluate appropriate risk assumptions and independently determine the NRC risk
analysis of these incidents.
! Evaluate the risk significance and PSEG activities associated with the fuel oil leak on
EDG 2B in February 12, 2002 and EDG 1B on May 8, 2002, as reported in performance
indicator data.
! Evaluate the effectiveness of maintenance and engineering activities on the diesel
generators and support systems of Salem Unit 1 and Unit 2 EDGs, including industry
operating experience.
! Determine possible generic implications.
! Document the inspection findings and conclusions in an inspection report within 45 days
of the inspection exit meeting.