ML18093A721
| ML18093A721 | |
| Person / Time | |
|---|---|
| Site: | Salem  |
| Issue date: | 02/26/1988 |
| From: | Eapen P, Murphy K NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I) |
| To: | |
| Shared Package | |
| ML18093A720 | List: |
| References | |
| 50-272-87-38, NUDOCS 8803170232 | |
| Download: ML18093A721 (9) | |
See also: IR 05000272/1987038
Text
U.S. NUCLEAR REGULATORY COMMISSION
REGION I
Report No.
50-272/87-38
Docket No.
50-272
License No.
Licensee:
Public Service Electric and Gas Company
P. 0. Box 236
Hancocks Bridge, New Jersey 08038
Facility Name:
Salem Generating Station Unit 1
Inspection At:
Hancocks Bridge, New Jersey
Inspection Conducted:
December 21-23, 1987
Inspectors:
K.
Approved by:
Q K - f:.~
P. K. Eapen, Ch~EB, DRS
~Id.bl~
'dite I
Inspection Summary:
Routine unannounced inspection of Containment Integrated
Leak Rate Testing (CILRT) and Preliminary Results Evaluation (Report
No. 50-272/87-38).
The inspection verified that the test prerequisites and
test conduct were in accordance with approved procedures.
The test results
indicated an acceptably low containment leak rate.
The test went smoothly
as a result of thorough preparation and use of experienced supervision and
technicians.
QA coverage was thorough.
One unresolved item requires licensee
attention.
During the test the three operating Containment Fan Cooling Units
(CFCUs) tripped ostensively due to motor overcurrent caused .bY increased fan
loading at high containment pressure.
The licensee is verifying the root
cause of these failures and is reviewing fan operation and testing to assure
high confidence that these units are available during a design basis LOCA and
that future fan tripping during CILRTs does not occur.
Results:
No violations or deviations were identified.
One unresolved item .
8803170232 880229
ADOCK 05000272
G
DETAILS
1.0 Persons Contacted
- William Treston, CILRT, Test Director
- George Druffner, PSE&G, R&T Lab
Robert Brandt, Chief, Inservice Testing
Steven Miltenberger, Vice President, Nuclear Operations
Mark Gross, QA
Michael Wita, QA
Jack Curham, Engineering (C.F.C. System Engineer)
United States Nuclear Regulatory Commission
Thomas Kenny, Senior Resident Inspector
Kathy Gibson, Resident Inspector
- Indicates those present at the exit meeting on December 23, 1987.
The inspector also held discussions with technicians and engineers
involved in the containment leak test.
2.0 Containment Integrated Leak Rate Testing
During the period of December 20-23, 1987, a Containment Integrated Leak
Rate Test (CILRT) was performed at Salem 1, as required by 10 CFR 50,
Appendix J.
The test was conducted with containment isolation boundaries
in an
11 condition.
Testing was performed in accordance with
Inservice Inspection Procedure M9-ILP-CT-1, Revision 6, Issue Date 12/18/87.
The Inspector reviewed the test procedure and witnessed portions of the
test.
The purpose of the inspection was to assure that the test was con-
ducted in compliance with the requirements and commitments set forth in
the following section, that test anomalies are identified and acted upon,
and that the test results met the leak rate acceptance criteria.
2.1
Reference Documents
Salem 1 Technical Specifications 4.6.1.2(a), (b) and (c), Type
11A
11 Test
10 CFR 50, Appendix J, Primary Reactor Containment Leakage
Testing for Water Coolent Power Reactors.
Final Safety Analysis Report (FSAR)
ANSI/ANS 56.8-1981, Containment System Leakage Testing
Requirements
3
USNRC I&E Information Notice No. 85-71:
Containment Integrated
Leak. Rate Tests.
2.2 Other Documents Reviewed
M9-ILP-CT-1, Revision 6, Reactor Containment Building Integrated
Leak.age Rate Test Salem Unit 1.
Station QA Surveillance Reports Nos. 87-1785, 87-1786, and 87-1788.
Internal Memorandum for J. Curham to W. Treston,
11 Containment
Fan Coil Unit Operation During Type
11A
11 Testing,
11 dated
December 10, 1987.
Portion of
11 Reactor Containmnet Fan Cooler, Westinghouse
Technical Manual Unit 1 and 2, Salem Nuclear Generating
Station,
11 PSBP No. 139970.
Safety Evaluation Report S-2-M945-NSE-Revision 0,
11Tripping of
Fan Coil Unit (FCU) Motors Durin.g Type
11N 1 Test and Inlet Vain
Damper Settings.
For Fan Coil Unit, Unit No. 2 Salem Nuclear
Generating Station,
11 July 1, 1983.
Nuclear Department memorandum from J. Vargas to H. Berrick,
11 CFCU Fan Peformance/Type A Testing, SRG OPEN Item R-0104,
11
June 26, 1984.
2.3 Administrative Control of CILRT and Procedure Review
The inspector reviewed the test procedure, the CILRT test log and
the PSE&G Research Corp event log to verify that:
Test directors were designated and their responsibilities were
clearly defined.
The procedures were adequately detailed to assure satisfactory
performance.
Test prerequisites were met.
All required parameters were being recorded at the required
frequencies.
The inspector noted that adequate attention was paid to the procedure
as evidenced by keeping the step-by-step sign-offs current.
Test
personnel were being carefully briefed prior to each test evolution.
Plant personnel not directly involved with the test were briefed on
keeping
11 hands-off 11 in the areas affected by the test; this was re-
inforced by restricting personnel from the test areas.
Control of
the test by the test director was adequa~e .
4
The inspector noted several weaknesses in the test procedure itself
as follows*:
a.
Section 5.13 of the test procedure states that if leakage
occurs
11 to the extent that it would interfere with the
_
satisfactory completion of the tests, then leakage paths may be
isolated and the Type A test continued.
11
This statement is in
direct opposition to 10 CFR 50, Appendix J, III.A.l(a). This
disagreement was brought to this attentton of the test director.
b.
Section 6.3 and Attachment 3, step 5.1.C concerning the throttling
of Containment Fan Cooling Units (CFCU) and CFCU data recording
appears to be insufficient as indicated in Section 2.6 of this
report.
2.4 Test Witnessing
The inspector witnessed portions of the following test activities:
1.
2.
3.
4 .
5.
6.
Containment pressurization and licensee walkthroughs for
identifying component leakage.
Four hour stabilization.
Twenty-four hour CILRT.
Metered release in preparation for verification test.
One hour instrument verification test.
Start of containment depressurization.
These activities were witnessed to verify that the CILRT was
conducted in accordance with the test procedure, that test personnel
demonstrated adequate levels of knowledge, that no equipment
anomalies were evident, and the test results were, in fact, within
the requirements of 10 CFR 50, Appendix J.
The inspector attended
briefings with test personnel, interviewed data collection engineers
and leak crew technicians, and made walkdowns with the test director
and other test personnel.
The inspector noted that test
personnel were experienced in conducting the test with senior test
personnel having conducted a number of similar tests in the past.
No adverse leakage paths were identified during the walkdowns and
the installation of the tygon tubing and vent bottles were as required.
The chronology of major events were as follows:
Chronology of Events
December 21, 1987
1400; noted fumes in control room from operation
of air compressors
5
1420; commenced containment pressurization
- 1735; dispatched first leak inspection team
2300; Reached 40.68 PSIA at about 3 psi per hour
December 22, 1987
0030; dispatched leak inspection team
0355; #13 CFCU tripped [at about 57 psia, per
calculation by inspector]
0425; #15 CFCU tripped [at about 58 psia, per
calculation by inspector]
0530; dispatched leak inspection team
0550; began securing air compressors at 62.43
psi a
0619; all compressor secured
0630; began stabilization period
0702; #14 CFCU tripped, all fan coil units out
of service
6
0830; disp~tched leak inspection team, Ele. 100'
airlock appears to have small leak, suspect a l~ak
in inner -ball valve per past LLRT.
1030; stabilization complete
1045; start of CILRT
December 23, 1987
1045; completion of CILRT
1453; start of verification test
1553; completion of verification test
1711; start of depressurization
Startup of the nine portable diesel powered air compressors, that
were located between buildings next to-the turbine building, resulted
in exhaust gases being dra~n into the adjacent building including
the Salem 1 control room.
The licensee determined that this
condition was not hazardous to personnel and the condition subsided
as the compressors were loaded.
A minor_ leak into the vestibule of
the airlock at Elevation 100' was the only component leak detected
as a result of the walkdown by the leak crews.
The only abnormal
event during the pressurization phase was the tripping out of the
three operating Containment Fan Cooling Units (tFCUs) which is discussed
in the next section.
2.5 Tripping of Containment Fan Cooling Units (CFCUs)
Three CFCU's were in operation at the start of pressurization.
At
approximately 57 psia the #13 CFCU tripped followed by #15 CFCU at
approximately 58 psia.
The last unit, #14 CFCU, tripped about one
hour after containment pressure reached the test pressure of 63 psia.
7
The inspector held discussions with the CFCU system engineer to determine
if tripping of these uni ts ca 11 into question their availability
during design basis accidents and whether the CILRT prerequisites
concerning the fans where adequate.
Discussions with the engineer
and subsequent review of Westinghouse design documents and PSE&G
internal memoranda revealed that the licensee had experienced trip-
ping of these fans in past CILRT tests.
The licensee conducted
engineering analyses to determine an appropriate solution to the
problem.
These memos indicate that ihe fans become overloaded under
the test conditions and trip out due to motor overcurrent.
The
engineering documents reviewed did not provide a complete picture of
whether the fans design bases are fully satisfied. Specifically:
1.
The fact that #14 CFCU tripped after operating for one hour with
a constant containment pressure of 63 psia may indicate a cause
other than motor overcurrent.
This should be investigated.
The inspector was not provided with any firm evidence that
motor overcurrent is, in fact, the sole root cause of fan
tripping.
2.
A complete engineering analysis that accurately predicts fan
flow, motor current, component temperatures and the other para-
meters potentially affecting fan availability both during an
CIRLT and the various design basis requirements in Section 6.2.2.2
of the FSAR should be provided.
3.
The above engineering analyses should provide assurance that
fan surveillance testing at normal containment conditions is
adequate to ensure availability during design basis accidents.
4.
The tripping of the fans during the latest CILRT was apparently
caused by an error in adjusting the inlet vane settings.
What
actions will the licensee take to as-sure future errors in setting
the inlet vains do not occur; especially when restoring the fans
to their normal operating condition?
To this end it should be
noted that Westinghouse report PSBP #139970 suggests that motor
amp readings be taken.
Accurate amp readings could provide a
diverse means of verifying that fan load conditions are correct.
In addition, trending amp readings during a C.I LRT would verify
the air density versus load characteristic of the fan units.
5.
Determine whether fan surveillance and CILR test procedures
require modification in view of the answers to the above items.
Pending resolution of the above concerns, this item remains unresolved
(50-272/87-38-01).
2.6
8
CILRT Leak Rate Measurements and Results
The inspector monitored the data acquisition throughout the test and
reviewed the computer data reduction methods.
The calibration
records of the pressure, temperature, flow, and humidity instruments
had been calibrated within 6 months of the test and were traceable
to the National Bureau of Standards as required.
All instruments
were operable throughout the test with no apparent malfunctions.
The computer based data acquisition and reduction activites appeared
faultless with the system being thoroughly tested prior to test ini-
tiation.
The data collection personnel from the Licensee's Research
and Testing Laboratory were certified under the Laboratory's quality
assurance program for leak testing, Level I and Level II.
The
licensee computed the following leak rates at the conclusion of the
test:
24 HR CILRT
(Weight% per day)
Total Time Method
Mass Point M~thod
Calculated (Lam)
0.041
0.044
0.075
0.039
0.043
0:075
Lam + Upper Confidence Limit
Acceptance Criteria
The inspector made independent calculations using an NRC approved computer code
that showed close agreement with the licensee calculations.
3.0 QA/QC Coverage
The inspector interviewed the QC inspectors that provided coverage of the
CILRT and reviewed three QC reports documenting QC findings during the
test.
Instrument calibration, test personnel certification, and test
procedure implementation was covered along with checking 58% of the valve
positions.
The QC inspectors showed thorough knowledge of the purpose and
requirements of the test.
Well prepared QC check lists for the CILRT were
used in assure coverage of important test aspects.
Several valves were
found to be out of position by QC, the QC inspectors followed through to
find the reasons for these discrepancies and were satisfied that the test
director had given permission to Operations to change the valve position
and that there was no effect on the test.
The inspector concluded that
QC coverage was adequate.
4.0
- - - -
9
Exit Meeting
The findings of the inspector were periodically discussed with the test
director and were summarized at the exit meeting on December 23, 1987.
Attendees at this exit meeting are listed in Section 1.0 of this report.
At no time during this inspection was written material provided to this
licensee by the inspector.