IR 05000334/1978031
| ML19269D228 | |
| Person / Time | |
|---|---|
| Site: | Beaver Valley |
| Issue date: | 12/19/1978 |
| From: | Kohler J, Tanya Smith, Streeter J NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION III) |
| To: | |
| Shared Package | |
| ML19269D227 | List: |
| References | |
| 50-334-78-31, NUDOCS 7903070506 | |
| Download: ML19269D228 (8) | |
Text
.
.
U.S. NUCLEAR REGUIATORY COMMISSION OFFICE OF INSPECTION AND ENFORCEMEhT
REGION III
Report No. 50-334/78-31 Docket No. 50-334 Priority --
Category C License No. DPR-66 Licensee: Duquesne Light Company 435 Sixth Avenue Pittsburgh, Pennsylvania 15219 Facility Name:
Beaver Valley Power Station Inspection at:
Shippingport, Pennsylvania Inspection condt cted:
November 1-6, December 8, 1978 b [bbk*/
e g
Inspectors:
J. E. Kohler Y87 Rcastor Inspector f
T t
T a ning Capacity)
'/2 /79 Reactor Inspector o _.
F.NStreeter, Chief d[ct/78 Approved by:
huclear Support Section 1 Insoection Sum ury:
Inspection on November 1-6, December _8,1978 (Report No. 50-334/78-31 Artas Inspected:
Routine, announced inspection by a regional based inspec-tor of the containment integrated leak rate test.
The inspection involved 50 inspector-hours onsite by one NRC regional based inspector.
Results: No items of noncompliance or deviations were identified.
7903 070 SOG
.
.
DETAILS 1.
Persons Contacted The below listed technical and supervisory personnel were contacted:
- J. Werling, Station Superintendent
- H. Williams, Chief Engineer
- R. Zabowski, Technical Supervisor
- L. Schad, Operations Supervisor
- W. Robinson, Test Engineer
- R.
Balcerek, Maintenance Supervisor D.. Schultz, Shift Supervisor
- R.
Mafrice, Results Coordinator
- denotes those present at exit interview on November 6, 1978.
- denotes those present at exit intervf ew on December 8,1978.
- denotes those present at exit interviews on November 6, and December 8, 1978.
2.
Licensee Action on Previous Inspection Findings (0 pen) Unresolved item (334/78-27-01) CILRT procedure:
The below items assoc sted with procedure number 1.47.2 were collectively designated unreso.ved item (334/78-27-01).
Prior to performance of the initial CILRT, each item was discussed with the licensee.
The status of each of these items is indicated below.
(1)
Computer program: Mass point technique data for analysis was used.
Hourly data was used with trend data as a backup as additional data to lower the deviation (closed).
(2) Acceptance criteria: The acceptance criteria used by the licensee was that the 95% upper confidence level must be less than 75% La (Closed).
(3) Test Pressure: The test pressure was held greater than 38.3 psig at the start of the test (closed).
(4) Volume Changes: Pressurizer level and sump level changes were monitored and logged by the licensee (closed).
(5) Leak Rate Repair:
The licensee understood and complied with the NRC position on leak rate repair (closed).
(6)
Instrument Failure:
The licensee amended the procedure to require at least one dewcell throughout the test (closed).
'
.
.
(7) Volume Weighting Factors: The procedure was amended to incorporate the weighting factors used during the preoperational CILRT (closed).
(8) Pump Back:
The licensee decided to perform an induced leak rate test as the supplemental verification in lieu of the pump back test (closed).
(9) Venting of Weld Channel Boxes:
The containment liner fabrication was field tested by welding channel boxes over each containment liner weld seam and performing a pressure test.
After the integrity of the particular weld seam was verified, the channel box was plugged.
Prior to the performance of the 1978 CILRT, the licensee was informed that the NRC had taken the following position regarding containment test channels of the liner welds: During the performance of the CILRT, the weld channels should be vented to containment test pres-sure because the unvented channels present an artificial barrier to leakage which has not been assumed to exist af ter a hypothetical design basis accident.
At the beginning of the 1978 CILRT, the licensee informed Region I and NER that the weld channel boxes would remain unvented for this test.
The licensee and his agent, Stone and Webster, intend to present evidence to NRR to prove that the weld channel boxes are extentions of the containment liner and, therefore, are not re-quired to be vented.
The inspector informed the licensee that the validity of the CILRT with the weld channel boxes unvented is in question and will be resolved between Region I, NRR, and the licensee.
Either the 1978 CILRT must be deemed to be valid or another leak rate test must be performed prior to expiration of the 40 + 10 months interval (or approximately October 1979) specified in Technical Specification 4.6.1.2.
This matter remains unresolved.
(Closed) Inspector followup item (334/78-27-02):
The below items were designated as inspector follow items and were resolved to the inspector's satisfaction prior to performance of the first of two attempted leak rate tests.
(1) Cloud cover:
The licensee recorded weather conditions at the meteorological station located in the control room.
(2) Data rejection:
The licensee rejected no dat.
.
(Closed) Unresolved item (334/78-27-03): Review of resistance tempera-ture detqctor calibration information.
The canputer input calibration for RID's received careful revicw prior to performance of the second containment integrated leak rate test.
The licensee's review resulted
- .n changing the 0-10 mv span of a typical RTD from several hundred degrees to about a hundred degrees.
(Closed) Unresolved item (334/78-27-04):
Development of dewcell calibra-tion procedure.
The licensee's calibration of the deweell instrumen-tation was acceptable.
(Closed) Inspector followup item (334/78-27-05): Review of pressure instrumentation calibration information.
The inspector reviewed the calibration of the CILRT pressure instrument and found it acceptable.
3.
Initial CILRT - Unsuccessful 11/2-12/78 The first periodic CILRT performed November 2-12, 1978, was classified as a failed test.
The test was repeated prior to startup (Paragraph 4).
The following items deal with the specific events that took place dur-ing the initial CILRT.
a.
Airlock Bypass Leakage During the CILRT pressurization phase, the airlock pressure began increasing.
This indicated a leak through the inner door.
Since the outer door was closed, containment integrity was maintained.
The licensee terminated CILRT pressurization and repaired the inner door.
Investigation revealed that one of the four solenoid pressurizer equalization valves was permitting containment pressure to leak into the airlock.
The valve in question only allows flow from containment to the airlock and therefore was not tested on any routine 6-month integrated airlock test or 72-hour seal pres-sure test.
The licensee stated that he would include the valves in his surveillance testing program.
The most recent 6-month integrated airlock test was perf ormed on October 30, 1978, just two days prior to the CILRT. Based on the successful integrated airlock test, con tainment integrity would be maintained by only the outer door.
Consequently, the licensee's decision to depressurize and repair the airlock equalization valves was discretionary and has no bearing on the characterization of the initial CILRT attempt and subsequent depressurization as a passed or failed CILRT.
The inspector informed the licensee that the item was reportable under provisions of Technical Specification 6.9.
The licensee classified the event as 30-day reportable event.
The inspector had no further questions regarding this ite.
.
b.
Weighting Factors The inspector reviewed the weighting f actors used for the CILRT listed in the approved CILRT procedure and a computer printout of the weighting factors used for calculational purposes.
It was determined that the computer printout weighting factors did not agree with the weighting factors listed in the proce-dure.
The inspector determined that the weighting factors used for the computer for calculational purposes were the same as those described in the preoperational test report.
The licensee revised the procedure to agree with the computer weighting factors.
The inspector had no further questions regarding this item.
c.
CILRT Valve Lineup The inspector reviewed the valve lineup during the test.
It was noted that lines such as high head and low head safety injection were isolated, drained, and vented.
The licensee was informed that there were no requirements for exposing the isolation valves in lines that were assumed to be inservice following design basis accident.
The licensee stated that the valve lineup would be reviewed with respect to 10 CFR 50 Appendix J, Section III.A.1.d.
The inspector had no further questions regarding this item.
d.
Type A Test Peaaltv
.
The licensee determined that a penalty must be added to the measured CILRT leak rate for lines that were not isolated, drained, and vented for the test.
The inspector stated that lines included in the Type A penalty should be reevaluated based on the require-ments of 10 CFR 50, Appendix J, Section III.A.1.d.
The licensee performed this reevaluation prior to pumpup for the second CILRT attempt.
The inspector had no further questions regarding this item.
e.
Computer Program Prior to the initial pumpup, the inspector reviewed the procedure to determine the method in which the leak rate calculations were being made by the plant computer.
The following data processing deficiencies were noted by the inspector during the review:
(1) The computer program only accepted leak rate calculations based on one data set per hour frequency.
(2)
Only 24 data sets could be analyze.
(3) The small number of data sets able to be analyzed caused the statistical effect of random data to be maximized.
(4) There was no way to delete an incorrect data ser.
(5) Only one pressure gauge was used for the CILRT, and, there-fore, there was no way to verify its readings by comparison to snother gauge.
(6) There was no clock to calculate the elapsed time of the test.
The licensee was informed of the above conditions and made the decision to commence the initial test attempt with these de-ficiencies.
f.
Test Results The inspector obtained weighted average containment temperature dewpoint and containment pressure.
From this data, independent calculations were made. Based on an analysis of data collected for 35 hours4.050926e-4 days <br />0.00972 hours <br />5.787037e-5 weeks <br />1.33175e-5 months <br />, the inspector independently calculated a leak rate of.114 weight percent W/0 per day with.a 95% upper confidence level of.013.
This containment integrated leakage rate is in excess of the maximum allowable containment leak rate of.1 weight percent per day specified in Technical Specification 3.6.1.2.
The licensee calculate! the containment leak rate based on 24 data sets to be.094 6 ight percent per day with a confidence level of
.056 weight percent per day.
This also is in excess of Technical Specification 3.6.1.2.
4.
Second CILRT - Successful 11/20-21/78 Af ter the initial CILRT attempt showed leakage in excess of Technical Specifications the containment was depressurized and repairs were made.
The following su= mary lists the major test changes made after the initial unsuccessful attempt.
a.
Major revision to valve lineup whereby all systems classified as being inservice post-LOCA were in their normal station configura-tion; ECCS penetrations which were initially isolated, drained, and vented were flooded and isolated; modification to the valve lineup of 15 additional penetrations was made.
The Type C test penalty initially determined to be.012W/0/ day was lowered to approximately.004W/0/ day due to the revised valve lineup.
b.
Repair of the butterfly isolation valves in subatmospheric con-tainment evacuation penetration.
These valves were warped by exposure to steam while the valves were closed. This caused the valve seats to defor _,
.,,
.,
,- -.
-
- - - - - - -. - ~ ~ ~ ' - - = =
.
~
Recalibration of all RTD's was made so that the 0-10 av output c.
signal was equivalent of a temperature span of approximately
=
1000F instead of several hundred degrees.
'
d.
The licensee performed an induced Icak rate test instead of
=
the planned pump back test as the supplementary verification of the leak rate test.
=
e.
The results of the second CILRT at the upper 95% confidence level showed a leak rate of.0362 weight percent per day.
Allowing for the Type C penalty of.004 weight percent pa-day, the total containment leak rate was calculated to be app-ximately.04 weight percent per day, which is below the ma: e tra allowable containment leakage rate of.1 weight percent per day.
.
yT A supplemental test was performed for 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> in which a 2.5
'
SCFM leakage rate equivalent to about.054 weight percent per day was induced.
Af ter 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> and 25 data sets the licensee
'
measured a leak rate of about.077 weight percent per day.
This is within the allowable limits set forth in 10 CFR 50 Appendix J and is confirmatory of the 24-hour leak rate.
5.
Future Since the first CILRT attempt failed to meet the accep tance criteria due to leaking isolation valves in the containment evacuation system
'
as described in paragraph 4.b., the test schedule applicable to sub-sequent Type A tests will be reviewed and approved by the NRC in accordance with Section III.6(a) of Appendix J.
6.
Exit Interviews - 11/6/78 and 12/8/78
-
Exit interviews were conducted with licensee representatives (denoted in paragraph 1) on November 6 and December 8,1978.
"
The following items were discussed during the November 6,1978 inter-view:
'
(1)
Leak rate in excess of technical specifications based on 35 hours4.050926e-4 days <br />0.00972 hours <br />5.787037e-5 weeks <br />1.33175e-5 months <br /> worth of data.
(2)
NRC position that venting of weld channels is required for an acceptable CILRT.
(3)
The necessity to review the valve lineup based on 10 CFR 50, Appendix J, Section III. A.1.d.
(4) The necessity to analyze intermediate data points and inclusion of their results in their calculated leak rate.
-
-.
.
_., _ _ - _ _. -. _ _
-
,
.
.
(5) Obtaining additional CILRT pressure gauges to monitor the results of the single instrument used during the initial test.
(6) Verif ying the operability of the dewcells because their operation was suspect.
The licensee stated that a retest would be performed and consid-eration would be given to the above mentioned items.
The following items were discussed during the December 8, 1978 exit interview:
(1)
Successful CILRT retest performed on 11/20-21/78.
(2)
Consequences of steam jet air ejector damage to valves in containment evacuation penetration.
(3) Weld channel box venting status.