ML20097H194
| ML20097H194 | |
| Person / Time | |
|---|---|
| Site: | Beaver Valley |
| Issue date: | 06/12/1992 |
| From: | Sieber J DUQUESNE LIGHT CO. |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| NUDOCS 9206240313 | |
| Download: ML20097H194 (26) | |
Text
m-a h
Q
<a a ley Power $tation Shtpegport PA 15077-(XX)4 (412) 391525$
JOHNO SIEBLR V<a President * %cear Group
)g U.
S.
Nuclear Regulatory Commission Attn:
Document Control Desk Washington, DC 20555
Subject:
Beaver Valley Power Station, Unit No. 1 and No. 2 BV-1 Docket No. 50-334, License No. DPR-66 BV-2 Docket No. 50-412, License No. NPF-73 Combined Inspection Report 50-334/91-80 and 50-412/91-80 Electrical Distribution System Punctional Inspection In response to the NRC correspondence dated April 1, 1992 which transmitted the above referenced inspection report, attached is a schedule for the resolution of the Unresolved Items resulting frou the inspection.
An extension of the due date for this response to June 12, 1992 was agreed to by Region I.
If there are questions concerning this response, please contact Mr.
K.
E. Halliday at (412) 393-5600.
Sincerely, D.
Sieber Attachment cc:
Mr.
L.
W.
Rossbach, Sr. Resident Inspector Mr.
T.
T. Martin, NRC Region I Administrator Mr. M.
W.
Hodges, Director, Division of Reactor Safety Mr.
A.
W.
DeAgazio, Project Manager Mr.
M.
L. Bowling (VEPCO)
I
/jk0 p
l noc
.I b d 'v 5; h 9206240313 92'0612 PDR ADOCK 05000334 G
. ~
_tt t,,
- i
[ Beaver [ Valley Power Station,_ Unit.No._1 and:No. 2 BV-_1_Dpcket No.-=50 334,-License:No. DPR-66 BV-2 Docket-No. 50.-412, License-No, NPF.
e
_ Combined Inspection Report 50-334/91-80 and 50-412/91-80:
3
- Electrical 1 Distribution System Functional Inspection Page'2
~bec:
ORC' Members-AIS Members G.
S.: Thomas 1
- T. P. Noonan
- D..E."Spoerry.
- N.--R.
Tonet~
.F.=J.
Lipchick-K. E. Halliday C.'D.'Schmitt
- R. L. Hansen
.S.
A. Nass-K.-L. Ostrowski
- F.-D._Schuster--
K. D.-Grada D. R.
Dic3 nan T. A. Mayers Central File (2),
.ND3NSM:5595
References:
4 ND1MNE:6175 m
v 1
9 3
w
.ec.,-.
m.
we 3
-,.si,y
-,w,--
~
i9, w
r
4 Page 1 of 24 DUQUESNE LIGHT COMPANY Nuclear Group Beaver Valley Power Station Unita 1 &2 1
Combined NRC Inspection 50-334/91-80 and 50-412/91-80 Electrical Distribution System Punctional Inspection 1
Unresolved Items i
OVERVIEW Duquesne Light Company has reviewed the results of the Electrical Distribution System Functional Inspection of Beaver Valley Units 1 &
2, Combined Inspection Report 50-
-334/91-80 and 50-412/91-80.
This response addresses the 14 unresolved items identified in the inspection report.
The level IV violation and the deviation which were part of this report have been addressed separately in a previous submittal to the NRC (dated May-8, 1992).
1 As stated in the Inspection Report, it is agreed that several issues contained within the 14 unresolved items represent potential desicJn weaknesses and areas of concern, but not i
' plant operability issues. Most of these weaknesses are the result of the calculation methodology applied for Beaver Valley Unit-1-(BV-1) in the early 1970s.
If issues of safety significance develop during the effort to address these unresolved items, appropriate review / corrective measures will
'ba taken to resolve the issues.
Our scheduled response times reflect the extent of analytical efforts required to completely resolve the issue.
This recognizes that a substantial amount of this work can only be performed sequentially (i.e.,
with the performance of one task dependent upon the completion of another).
In the Executive Summary of Inspection Report 91-80, two areas of concern in the technical support area were identified.
To address the concern of the small number of supervisors relative to the size of the electrical engineering staff, an interim organization has been implemented that increased the number of supervisors from 2 to 4.
To-address ~the issue of limited system knowledge of the engineering personnel, the overall training requirements for all Nuclear Engineering Department personnel is currently under review by department management.
The resolutions of each of the unresolved items is prescribed i
as-follows.
\\
l-i I
^
Page 2 of 24 i
1.
Setting of Degraded Grid Relays 50-334/91-80-04 and 50-412/91-80-04
==
Description:==
A review was performed of degraded grid relay settings and reset capability, coordination with the EDG start and fast bus transfer
- schemes, and applicable schematics.
This review determined that the degraded grid relays on the 4160 V and 480 V safety related buses were set at 90% +/-1.6% of their respective nominal bus voltage.
Therefore, the minimum theoretical voltages allowed on the switchgear and on the load conter buses, before the appropriate automatic action takes place, are 3677 V and 424 V, recpectively.
To ensure that the specified settings adequately prstect the safety related motors from undervoltage conditions, the continuous ratings of the motors were also surveyed.
A sampling of several 4160 V and 480 V motors revealed a continuous rating of 90%
of the nominal (nameplate) voltage i.e.,
3744 V
and 414 V
(460 V
x 0.9),
respect 1vely.
A comparison of the above values shows that, under degraded voltage conditions, the 4160 V motor would be operating at a voltage below their minimum continuous rating and that a 10 V margin exists for the 480 V motors.
In addition, when the cable voltage drop from the bus to the motors is-taken into consideration, the voltage at the motors' terminals could be considerably less than the motors' continuous rating.
The 90% relay setting was verified through a review of several Relay Setting Sheets and is in accordance with the guidelines contained in - BV-1 and BV-2's " Protective Relay Philosophy and' Practices for 4160 V and 480 V Systems",
Engineering Standard No.
ES-E-004, Revision 0,
dated September 11,
- 1989, and ES-E-003, Revision 0,
dated February 14 1989, respectively. The +
tolerance was calculated i,n a Westinghouse analysis o/-1.6%
f the relay loop.
The team discussed the concern with the licensee who pointed out that the settings were in agreement with Item 6 of Table 3.3-4 of BV-1 and BV-2's Technical Specification.
The-licensee also indicated that the transformer tap settings kept the bus voltage near the nominal values.
However, they were unable to provide an analysis to show that the motors could be operated - below their continuous rating should a degraded voltage' condition exist.
The setting of -the degraded grid relays and/or the capability of the safety related motors to operate below their continuous setting is unresnived pending appropriate analysis or justification by the licensee (50-334/91-80-04, 50-412/91-80-04).
i
\\
4 Page 3 of 24 Proposed-Resolution:
50-334/91-80-04 and 50-412/91-80-04 The current degraded grid relay setpoints were established-recognizing that equipment will operate below 90%
of nameplate voltage for periods of. time with minimal impact on the equipment and that corrective actions would be taken to improve degraded grid conditions. The electrical grid in the BVPS -. area -is very strong and does not operate at degraded voltage conditions or normally experience voltage
-excursions of a magnitude to-cause low voltage concerns.
The degraded grid relay setpoints will be addressed by the upgrade of applicable electrical calculations.
Pending completion-of this - ef fort, the use of interim etpoints will be investigated and implemented, if appropria_e. Final setpoint determination will be made after completion of the necessary calculational efforts.
Scheduled Date:
50-334/91-80-04 Interim setpoints implemented 9th refueling outage Final setpoints imp? t.nented lith refueling outags 50-412/91-80 Interim setpoints implemented 4th refueling outage Final setpoints implemented 5th refueling outage 1
2.
--4 KV Breaker Interrupting 50-334/91-80-05
==
Description:==
The. team was unable to make an assessment of the adequacy of-the circuit breakers to interrupt the Unit 1 system fault: currents.
However, the--team determined that the 4 KV breakers had an interrupting capacity of only 30,000 A.
-When this - is compared to the-46,200 A interrupting rating of the Unit-2 circuit breakers and when the small considered the adequacy of - Unit 2
circuit breakers is interrupting margin - of the the Unit 1
breakers is
)f concern.
Therefore this issue is unresolved pending appropriate calculations by the licensee.
-Proposed Resolution:
50-334/91-80-05 The Unit 1 4kv circuit breakers were supplied from the ITE Corporation. These breakers are all Type 5HK250 except for the-feeder breakers to the non-1E 4kV buses which are Type 5HK350.
Per manufacturer's information, the Type 5HK350 l
~_
t Page 4 of 24 breakers are rated 46,200 amps symmetrical at 1.05 per unit voltage-The Type SHK250 breakers have symmetrical ratings of 30,000 amps at 4760 volts, 35,000 amps at 4160 volts and
~
o
-37,500 amps at 3850 volts.
The preliminary results of a design analysin indicates that the worst case calculated short circuit generated on any of the Unit 1 4ky buses
-1 class lE or non-class 1E - is 27,600 amps symmetrical. Based upon this preliminary analysis the interrupting capacity of the 4kr circuit breakers appears to be adequate.
A thorough review of the calculation inputs, methodology and assumptions will be performed, and
~
the results recalculated prior to issuance.
Scheduled Date:
50-334/91-80 Final calculations will be issued by April 1, 1994
-3.
125 Vdc Short Circuit Calculation 50-334/91-80-06 and 50-412/91-80-06
==
Description:==
The-team' examined BV-2 Calculation No.
10080-E-062, Revision 4 which analyzed-the short circuit ~ currents available in battery systems 2-1 to 2-6.
The results of this calculation were then used to assess the interrupting capability of the circuit breakers and the-ability of t$e cables to withstand the maximum, predicted short circuit currents.
The team focused on the Class 1E battery system 2-2, but all Class 1E and non Class 1E systems were considered.
Pertaining to'the calculation, the team noted that it had not included the - contribution from battery charcJers and the dc motors.
The team considered the 125 A contribution from the battery
- chargers, by
- itself, to be of minor significance to the results.
However, the combined effect of. this1 and the energy feedback from the applied motors, a 60 hp motor on battery system 2-5, could have a e.g., ficant' impact on signi the calculation results and should have been considered.
No short circuit calculations were available for the Unit 1
-battery sy tems.
Therefore, no conclusions could be reached.
i l
Based upon - the above, the analysis for available short
-circuit current in the battery system is unresolved pending
.the licensee's revision -of the Unit
.2 calculations and l
preparation of the Unit 1 calculations (50-334/91-80-06)
(50-412/91-80-06).
Page 5 of 24 Proposed Resolution:
50-334/91-80-06 and 50-412/91-80-06 Prior to 1991, short circuit currents were evaluated for i
Unit 1 safety related battery
- systems, and appropriate modifications were madu to existing circuits.
Open items which-remain will-be resolved prior to issuing the calculation.
A short circuit-calculation for the Unit I non-1E battery system will be performed.
Significant Unit i battery charger and motor contributions will be factored into these calculations.
Unit 2 calculation 10080-E-062 was reviewed to address the concern of DC motors not being considered It was found that the contribution of a 30 hp_( Oil Backup Pump) motor had in fact been considered for battery system 2-5, as found on page 11 of the calculation 1008 0-E- 0 62 Rev 4, and a 60 hp pump motor was considered for battery system 2-6 as found on pacJe 17 of this-calculation. Our review indicates i
that significant loads have been considered and that no additional action is required on this issue.
t Scheduled Data:
50-334/97-80 December 31, 1992:
Completion of calculations for 125 VDC safety related systems December 31, 1993:
Completion of calculations for 125 VDC non-safety related systems 50-412/91-80-06 No additional action is required.
4.
Steady State. Loading of EDG 50-334/91-80-07 Desc.ription:
Calculation 8700-DEC-E-048, Revision 0,
dated January 13, 1989, using the spread sheet method, evaluated the steady i
state loads for the.BV-1 Emergency Diesel Generator (EDG)
-No.
1.
The study identified the loads imposed on the EDG at each step of the automatic sequence and for the period after the automatic loading under three scenarios: Design Basis Accident, Loss of Normal Power, and Safety Injection.
l l
l- -.
Page G of 24 The team's review of this calculation revealed that the acceptance criteria specified that the maximum coincident (short time) load should not exceed 90% (2745 kW) of the 30 minuto diesel generator's ratingwaxi(mum 3,050 kW).
Based upon the load summary tables, the coincident load for the' worst case scenario (Safety Injection) was 2741.3 kW or slightly below the value stated in the acceptance criteria.
In addition, the team determined that the maximum calculated continuous load was 2579.3 kW, also slightly below the continuous rating (2,600 kW) of the EDG, Although both values were well within the 2000 hour0.0231 days <br />0.556 hours <br />0.00331 weeks <br />7.61e-4 months <br /> rating of the machine, 2850 kW, the team noted that minimal margin existed between the calculate' loads and the imposed limits.
- However, the licensee responded that, since the maximum coincident and continuous loading occurred after the automatic sequencing, potential overloads could be handled administratively.
The licensee also indicated that the calculation was undergoing revisions.
In support of this, they provided an internal memorandum, dated February 25, 1991, which identified incorrect entries found during a review of mechanical inputs in EDG Load Study Calculation 8700-DEC-E-048.
The summary sheets of this memorandum were an-updated version of Attachment F to Calculation 8700-DEC-E-048.
The team's review of the revised loads list identified several areas of concern:
l.
The worst case loading occurs under the Loss of Normal Power scenario
- and, for this
- case, the maximum steady state load is 2754 kU, which slightly exceeds the acceptance criteria of 2745 kW of Calculation 8700-DEC-E-048.
The licensee reiterated that the loads are limited by administrative controls and provided operating procedures to show how certain loads are cycled.
However, this was not clear from the body of the calculation.
2.
For the motor loads on pages 2A, 3A, and 4A, the memorandum identifies the nominal horsepower, "HP";
the flow curve brake horsepower, " Curve BHP" and a calculated brake -horsepower,
" Calculation BHP".
Since, in some cases, the EDG loading uses the
" Calculation BHP" which is less conservative than the " Curve BHP",
the team asked the licensee to provide an analysis or calculation identifying the bases for the calculated brake horsepower and the criteria for selecting these values instead of the ones derived from the flow curves.
The team was unable to determine the availability of such data.
M
-qn4---
4 A-
-FA 5 - * -
- L As-(
+r-n*
A---->=-4
- > - * +
4-m:e-m-
A4 m4-ae 34_
4 Page 7 of 24 3.
The motors for Auxiliary Feed Water Pumps FW-P-3A and 3B appear to be underrated for the intended functions.
This issue is discussed in-details under Section-10 of this report.
4.
Several discrepancies exist between the lo7ds as identified in the FSAR and in the memorandum.
The FSAR did not appear to reflect the changes identified in Attachment F
of the calculation, dated January 1989.
In order to verify that the 2745 kW load at the running
-powc; factor would be carried by the EDG, the team corpared it to the Reactive Capability Curve included in the EDG Data, 8700-1,30-32, Page 15, but d2termined that this curve was generic.
The licensee was not able to supply. the documented basis and_ applicability of this curve for Unat 1 EDGs during the inspection period.
E The team observed that calculation-8700-DEC-E-048 only addressed EDG.No.
1 The reason for this was that EDG No. I was more heavily loaded.
In view of the February 25, 1991 memorandum, the team calculated the EDG Noc 2 loads and found them to be less than those-on No.
1.
The team also observed that swing pump load had not been considered to be carried-either by EDG No. 1 or by EDG No.2.
The above issues were identified and discussed with the licensee who agreed that the revised calculation would
-include necessary clarifications.
In view cf the above, the steady state loading of the diesel generator is unresolved pending revision of the calculation by the licensee and review of tha results by the NRC.
(50-334/91-80-07) l 3
- Proposed Resolution:
50-334/91-80-07
-1.
The Unit 1 EDG loading has been investigated and it was found_that, by more detailed evaluation of the generator loads, the 2745 kW limit would nct be exceeded.
The team also expressed concern that the calculation did not specifically mention administrative requirements for shutting down the auxiliary feedvater pump prior to manually loading the residual heat removal pump
-o n the EDG.
A review of the operating procedures to address loading on the EDG has been
-performed and appropriate statements addressing all similar-administrative. controls will be incorporated into the
'O Scenario' section of the revised calculation.perating I
Page.8 of 24 2.
The mechanical load inputs'for the diesel generator-load study were developed by - calculating the pump or fan motor input BHP
-requirements. based on the calculated system flows and - head and efficiency data from the manufacturer's certified test - curves.
The
_ calculated loads should - agrae with the ' curve BHP' in all cases where data is derived from the same source.
Minor differences may-occur due to difficulty in accurately, reading the~ manufacturer's curves which tvpically plot BHP on a very small scale.
lor the
. diesel generator load study, _ the ' calculated value was always used in lieu-of the curve value.
A review of
-the calculated and L curve values will be performed to assure that the values are consistent.
3.
The results of a preliminary analysis indicates that the existing motor is adequate to provide the BHP requirements of the auxiliary feedwater pump.
The calculation will be: verified prior to completion.
4.
Upon completion of the above items, an FSAR change will be incorporated into the next UFSAR update.
Specific-reactive capability curve data for the Unit 1 EDGs is not available at present.
A transient analysis is being performed-that.will demonstrate the reactive capabilities of the - Unit-1 EDG.
See unresolved item 5 for further-
. discussion and schedule.
Swing pump loads-were not specifically discussed in calculation 8700-DEC-E-048 Rev.
O.
However, the swing pump loads on either train are enveloped by the Train
'A' pump
. analysis._ Therefore, the calculation results are not effected. A~
statement will be added to the revised calculation to document this consideration of the swing pump loads.
Scheduled Date:
50-334/91-80-07 Actions-for resolutions 1
thru 4
listed above will be complete by-September 1 1992.
Refer to unresolved item 5 for the scheduled 6
completion date on the EDG transient c
analysis.
The UFSAR vpdate will be incorporated into the 1993 annual update.
t
.. ~ -. -.
w Page 9 of 24 l
'5.
. Dynamic Loading of the EDG 50-334/91-80-08 and 50-412/91-80-08
Description:
I An: analysis to demonstrate the transient loading, capability of the BV-1 emergency diesel generators was included in Calculation 10080-E-048.
The team's review of the i
. applicable portions of this calculation revealed that the analysis was based upon
-a generic Dead Load Pickup t
Capability Curve and upon a manufacturer's letter, dated
, ;=
December 4, 1972 to Stone & Webster Engineering Corp.
The
-Dead Load Curve was used to analyze Stop 1 of the EDG's automatic loading, whereus the manufacturer's letter was used to analyze the other steps.
This lett.r included a summary.cf; sample EDG loading cases to which the licensee was _to compare the postulated accident loading steps.
As long as these were enveloped by a sample loading case, it was concluded that the voltage drop and its recovery time to 90% were acceptable.
The team's evaluation of the analysis indicated that there was no1 assurance that the curve was applicable to the Unit 1 EDGs - and - no back up calculations to support the design basis of the sample cases.
In
- addition, no diesel generator test as described in Gections 8.6.2 and 8.6.3 of the FSAR was available for review at the time of the inspection.
Unit 2_ Calculation 10080-E-048, similarly, did r.ot include a transient analysis.
Based on the above, the
- =
team concluded the transient loading _ capability _of the Unit 1 and Unit 2 -emergency diesel generators is unresolved l.
pending the licensee's retrieval of applicable tests or their' preparation _of an appropriate analysis.
(50-334/91-80-08) (50-412/91-80-08).
2 Additionally, the team reviewed surveillance testing of BV-1 EDGs for the-ability to supply the required real and reactive power during auto sequencing.
One of the tests simulated h LOCA with loss of offsite power.
The 'se.cond verified the responst of the machine but the loads added fail =to envelope DBE loads.
It was also noted that critical perforn.ance parameters were not adequately recorded (i.e.,
voltage, frequency and rack position).
\\
Proposed Resolution:
50-334/91-80-08 -- A transient analysis to evaluate EDG loading capability is being performed utilizing a
computerized program combining both analytical-and test l
information.
Testing is being planned for the 9th refueling outage to gather performance data not currently available.
The modeling validation will be completed within three (3) months after the Gth refueling outage.
l l
w.--
~
.1 Page 10 of 24 Additional validation effort, if determined to be required, could extend completion of the task to the loth - refueling outage.
Appropriate corrective action, if necessary, will be taken pending the results of this-analysis.
Performance testing procedures will be evaluated and revised, if - necessary, assuring that all of the pertinent performence parameters are recorded.
50-412/91-80-08 -- The seme - program will be utilized to evaluate loading -capabill y of-the -Unit 2
EDGs with modeling validation to be completed-within 3 months after the 4th refueling outage.
Appropriate corrective action, will be taken pending the results of this if necessary,ile analysis.
Wh a transient analysis for Unit 2 is not available at present, it should be-noted that the Unit 2 EDG manufacturer's loading. test was conducted utilizing larger motors than the actual connected loads.
Scheduled Date:
50-334/91-80-08 Unit 1 EDG analysis to be completed by 3-months after the-end of the 9th refueling outage.
50-412/91-80-08 Unit 2 EDG analysis to be completed by 3 months after the end of the 4th refueling outage.
Performance testing procedures (OSTs) will be revised by November-2, 1992.
6.
EDG Mode Change 50-334/91-80-09 and 50-412/91-80-09 i
Description-To address ' the sequencing of safety related loads on the emergency diesel generators following a loss of offsite power, the team reviewed Drawing 8700-RE-21 CL-4, Revision 4, dated March-21 1989, for Unit 1,
and T.rawing 12241-E-s 12A, Sheet 1, Revdsion 12, dated June 9, 14
for Unit-2.
The review included the control schemes fn stripping the 4160 V bus and sequencing the safety related loads on the bus', the type-and_ setting of.the sequence timers, and the
-setpoint drift.
For' Unit 1,
sequencing was accomplished using an electro-mechanical timer with a cam actuated contact.
With this
-type of-timer, the came are assembled on the same shaft and
'are rotated by the same motor.
Therefore, the time between load addition remains essentially constant and the possibility of two motors being started at the same time 7
y
,t
-e
Page 11 of 24 because of ' drift is non existent.
For Uni-2, the automatic sequence was accomplished using indivi ual solid state timing relays with negligible drift.
t The review of the Beaver Valley-1 electrical schematic revealed that, when the EDG is in parallel with the offsjte transmission system, a degraded grid condition or a loss of offsite power would cause the tripping-of the normal breaker and the immediata addition of emergency' bus loads, before the governor could change from the droop to the isochronous operation, and the voltage regulator could change from the _. parallel to the isolated mode.
This is caused by the fact that a set of contacts associated with the tripped breaker, along with the already closed EDG breaker, signal the load sequencer to load the emergency bus.
The estimated time for this occurrence could be-0.5 seconds or less.
This condition exists every time the EDGs are tested, including those times when they are tested to support Limiting Conditions for Operation.
The licensee was unable to -provide an analysis fcr this event by the end of the inspection.
The licensee indicated that they would review their design bases documents to see i
if the. issue.had been addressed.
This item is unresolved pending appropriate review and evaluation by the licensee (50-334/91-80-09) (50-412/91-80-09).
Proposed Resolution:
50-334/91-80-09 The BVPS Unit No.1 EDG has a Woodward type UG-8 governor.
This governor is a mechanical governor and does not have separate parallel and isochronous modes o'f operation. If a scenario occurs as described above, the voltage regulator will change from parallel to isochronous mode of operation, but the - governor does not have.to change its mode of operation.
An evaluation of this event will be performed.
50-412/91-80-09 The BVPS Unit No. 2 EDG has a Woodward EGB-50 governor with EGA Box.
The EGA Box electrically switches the governor from parallel to isochronous moGa of operation when required. The voltage regulator also changes from parallel to isochronous mode of operation.
An evaluation of this event.will be performed.
Scheduled Date:
50-334/91-80-09 An evaluation will.be completed by October 31,-1993.
50-412/91-80-09 An evaluation will be completed by March 31, 1994.
1
= w
--t,w
--r t----
=
Page 12 of 24 7..
penetration Heat Loads 50-334/91-80-10
==
Description:==
No calculation was available for review to establish the
-suitability of the Unit 1
electrical containment penetrations to carry continuous load currents without exceeding the allowable temperature rating.
Although the licensee furnished Specification No.
BVS-384, Revision 3, which referenced the IEEE 317 Standard, in the absence of a relevant calculation, the team had no basis for concluding
'that the Unit 1 penetrations were adequately sized and protected for the continuous loads.
Therefore, the capability of the Unit 1 penetrations is unresolved pending appropriate calculations by the licensee (50-334/91-80-10),
A review of the effects of heating due to short circuits as well as the protective device scheme for the ability to protect conductors against prolonged overcurrents found that:
1.
Short circuit values stated in BVS-384 are similar to those for Unit 2 penetrc. tion assemblies which were found to be acceptable, f
2.
Protective devices for the 300 hp Residual Heat Removal Pump (RH-P-1B) indicated satisfactory protection based on review of motor curves.
Based on the unavailability. of calculations the team concluded that, although the design appeared satisfactory,.
they had no basis for making an appropriate determination.
The team was also unable'to conclusively determine whether the appropriate protection had been provided.
Proposed Resolution 50-334/91-80-10 BV-1 penetrations were procured in accordance with specification BVS-384 from an approved vendor with a 10CFR-50,-Appendix B program. - The penetration continuous loads specified were utilized by the penetration vendor as design inputs. The continuous loading limits for the BV-1 penetrations are specified on the shop drawings provided by the vendor.
The BV-1 penetration vendor is no longer in business, ao additional data from the manufacturer is unavailable.
Resolution of this item will require development of additional information not currently available.
Since an immediate concern is not apparent, resolution will be a long term action.
._m_.-_.__.-
Page 13 of 24 Scheduled Dato 50-334/91-80-10 An ection plan will bo developed by l
October 31, 1993 to addrors this issuo.
8.
Cable bizing 50-412/91-80-11
==
Description:==
To address the sizing of the fondor cables used with both safoty and non-safety related loads of Unit 2,
the team r e v i e w e d C a l c u l a t i o n ti o.
10080-E-072 Revision 2.
This review revealed that the calculation, allowed the use of 550' 7 upper limit for - insulation temperature, instead of the usual 250' C required by the IPCEA Standards, when the i
cable is subjected to short circuit currents.
The team expressed concern regarding the finding since the allowed temperature was close to the 577'c auto-ignition temperature of the cable' jacket material supplied by the Korite Company.
The -team also found that an associated Ct.lculation llo.
10080-E-020, Revision.
3, produced even higher temperatures than the allowed limit.
Ilowever, in this caso, the uso of larger sizes cables, offectively reduced the maximum predicted temperature to below the imposed limit.
t The team was particularly concerned for the absence of a station procedure to inspect cables after a short circuit, a practico specified as important by the architect-engineer in 1985.
Additionally, thoro was no information available regarding molting and flow of the insulation and cascading offects..on adjacent squipment.
In consideration of the observations portaining uo - ha shmt circuit current available and of the fact ttat no 4
proceduro existed requiring a full ir.spection of the calle after an overload trip of.
the feedor
- breaker, the complet.onosa of the calculation for the BV-2 4.16 kV canlos is unrosolved pending appropriato analysis and corrective action by the licensee (50-412/91-80-11).
Preposed Rasolution:
50-412/91-80-11 The 550' C
upper temperature limit for short circuit considerations is mentioned in calculation 10080-E-020, Rev.3, not in calculation 100BO-E-072,- Rov.
2.
The basis i-for this - acceptance criteria is contained in letter 2DLS-23991, dated 1/7/85.-
The
~550*
C maximum temperature-ensures that a three phase bolted or a phase to phaso fault current cannot cause a cable to L mite since it is below l
the 577* C auto-ignition temporst'a o for Kari to Co.
cable jacket material.
l
. ~ _..
.._~. -
Pago 14 of 24 We have reviewed the existing documentation and Duquonno Light company bolloves that the 550*C critoria stated in calculation 10080-E-020 Rev. 3 is acceptable.
Calculation 10080-E-020, Rev.
3 was reviewed to determino if certain cables. exceeded the 550' C temperaturo limit.
Although the condition was indicated in a portion of the calculation, tho. conclusion section indicated the issue had boon resolved for each caso.
Additional reviews will be conducted to confirm the adequate resolution of this issue.
Critoria for cable tasting or inspection following short circuit conditions is boinq developed.
Once established the Opo'.1 ting Proceduros will be ruvined accordingly.
Scheduled Dato:
50-412/91-80-11 The above actions will be completed by December 31, 1992 Unit 1 Dosign Documents 50-334/91-80-12 9.
==
Description:==
The team noted that much of the design documentation for Unit 1 was not readily available for review during the inspection.
The licensoo indicated that the documentation T
is retained in doo storage and would require additional timo for retrieval.p Some specific unrosolvud issues, e.g.,
short circuit available at the 125 Vdc bus and electrical penetration heat
- loads, were identified in the section above.
However, the team also identified other arcan where an adequate evaluation of the Unit 1 electrical system could.iot be fully evaluated because of the lack of documents.
Thoso ~ areas include (1)- sizing of HCC cables for power and control circuits; (2) acceptability of the t
fast bus transfer schomo; (3) short circuit current avai' able - at the 120 Vac buses and (4) coordination of dc j
protectivo devicas.
Those issues are unresolved and will n
be revjewed when appropriate documentation can be mado available by the licensee (50-334/91-80-12) l I
l I
w-,
,,--,,--,..n.-
-- n
~,,.-...n--.._,--
n-.-,
-,--.~,-.,.n--
,. m n, n
__,-.we--.----...-
~ ---,,,, -
~.
l 1
Pago 15 of 24
-Proposed-Rosolution:
50-334/91-80-12 i
(1)
Sizing of MCC Cables for Power and Control Circuits Pro 11minary DV-1 calculations 8700-DEC-E-082, Revision 0,
" Cable Sizing Analysis For Loads Fod By Safety Polated Motor Control Contors" and 8700-DEC-E-ll3, Revision 0, "lE MCC Control Circuit Veltage Drop" woro performed, but have not yet been 13 sued.
A more i
detailed review and validation of those calculations is
~
required prior to their approval and uso.
Part of the validation requires input f rom-other calculations not yet complete.
The 8700-DEC-E-082 calculation will be superseded by
{
other calculations to be developed, while the 8700-DEC-l E-113 calculation will bo issued.
Scheduled Dates
- Safety Related MCC Power -Cable Sizing calculation i
will be completed by March 31, 1993.
Safety Related MCC Control Circuit Voltago Drop calculation will be completed by June 30, 1994.
(2)
Acceptability of Fast Bus Transfer Schemo During the audit the NRC reviewed in detail the Fast Bus Transfer Study for BV-2 and found it adequate.
The i
review of the Unit i fast bus transfer concluded that the summary of results for the study were lost and only a printout of the analysis data was retrievable.
Due to the time that has transpired betwoon the original BV-1 transfer study and the present, the study will bo updated.
It is expected that the revised BV-1 transfer study will yield acceptable results consistent with the existing BV-2 transfer study due to the following parallels:
Both u'its experience similar inputs due to the influenc.3 of connected loads and generators on the l
system grid.
Both units utilize the same transfer scheme and the i
same model 4KV circuit breakers to accomplish the transfer.
BV-1 has a smaller cumulative load to transfer than BV-2.-
l f
. _ _.. -.. _. _. _ _ _ _. _. - _.. _. _ _ =. -.. - -. _... -... _, _......,,,. _ _ -
.,.__.,_._..._,,-,....,m...___.,..,_
)
page 16 of 24 s
Scheduled Date:
An updated BV-1 fast bus transfer study will be completed by September 30, 1994.
(3) 120VAC Short Circuit Analysis A preliminary BV-1 calculation addresses the magnitude of the fault currents available at the 120V AC system.
The calculation han not been approved because the results obtained depend on design inputs from two other calculations not yet approved.
Scheduled Tsts. The analysis will be completed by June 30,
-1994 (4)
Coordination of D.C.
Protective Devices Unit 1 preliminary calculation 8700-DEC-E-062, " Safety-Related 125--VDC-Short Circuit -Analysis",
has been reviewed and it has been concluded that existing system independence and redundancy envelopes any breaker coordination concerns.
This condition will be further reviewed and correctivo measrres implemented if necessary.
Scheduled Date:
Safety Related system studies will be completed by December 31, 1992.
10.
Capability of Auxiliary Feed Pumps 50-334/91-80-13
==
Description:==
A review was performed of the manufacturers' pump and fan characteristic curves to determine the power demand on the emergency diesel generators for the three accident (1)ithDesign basis accident scenarios listed in the FSAR:
w unit trip; and (3)
-(DBA); (2) Loss of normal power Safety anjection signal with coincident loss of power and unit trip.
Major pump loads on the 4160 V emergency system, according to Table 8.5-1 of the FSAR, included: the charging high head safety injection pumps (Cil) with a nameplate rating of 600 IIP; the 250 IIP low -head safety injection pumps (SI),
the-300 11P outside recirculation spray pumps (RS), the-500 llP-- river water pumps (RW), the steam generator auxiliary feed pump (FW) with a 400 llP rating, the 300 llP residual heat removal pumps (Rit!, and the primary plant component cooling water pumps (Cc) rated at 400 11P.
On the 480 V
Page 17 of 24 system, the team identified the major pumps to be the quench pumps, the laside recirculation spray pumps, the containment air circu4 4 tion fans, and the Icak collection exhaust fans, each pump rated at between 150 and 3' 0 llP.
In conjunction with the pumps head / flow curves, the team reviewed a recent reassessment of the Unit i
diesel generate electrical loads (study 8700-DEC-E-048) and the mechanical engineering review of the same, documented in an internal memorandum, dated February 25, 1991.
The team noted that the steam generator auxiliary feed pump load had been changed - to 495 BilP, a 23.7% increase above the motor nameplate, 400 HP with 15% service factor, and a 28.9%
increase over the 384 BilP specifled in the FSAR.
The pump motor appeared to be operating above_its continuous rating, even when the service factor was considered.
Significantly, in Unit 2, flow restricting devices had been installed in the foodwater lines to protect the auxi.'.iary food pumps from runout conditions.
Pertaining to the FSAR, the licensee stated that.__it was outdated and that it-did not reflect runout conditions and
- maximum power demand for a postulated pipe break in the pump discharge line.
The auxiliary feedwater system comprises two motor operated pumps, _ each powered by a redundant emergency bus, and one turbino driven pump.
The turbino driven pump is not considered available in an accident involvir.g steam generator or supporting systems.
Following a feedwater pipe break coincident with a loss of normal power, if one diesel failed to
- start, the remaining redundant motor driven pump would automatically start and operate at runout
~
conditions.
In the initial phase of the accident, the pump is not essential, therefore, this pump could be shutoff without consequences.. However, later, when the heat sink capacity of the affected steam generator begins to deplete, the pump is needed to maintain-a minimum flow through the steam generators and must be available after the break is isolated.
During the estimated 10 minutes, minimum, required by the operator to diagnose the accident and temporarily stop the pump, the pump would be subjected to runout conditions with consequent cavitation and potentially serious damage.
Similarly, the - motor could suffer damage because of its operating beyond its rating.
Discussions with the licensee pertaining to the pu:aps '
operation in the above mode indicated that a series of high capacity tests simulating the runout conditions had been conducted to evaluete the pumps behavior.
The licensee stated that no visual or audible abnormalities were observed at the time except for a noise reduction where the niik alis iu p
Pago 18 of 24 pump reached its runout condition.
Flows, pressures, and motor amperes had also been measured.
The team ovaluated the not positive suction head availablo at the testod runout flow and found it to be below the required value indicating that during the runout tests, the pump had operated in cavitation.
there was no Ilowever #j ected evidence that the pump had been sub to a
comprehensive damago assessment program and, therefore, no positive indication of the present conditions of the pumps.
The team also noted that the li';enoco had f ailed to check the motor efficiency against the given curves.
On the other hand, the team verified that the hydraulic to shaft horsopower ratios for both motor driven pumps matched the curva efficiency value for the tested flow, r
The licensoo recognized the possibility of having run tho pump in cavitation for the duration of the tests and that they could not prove absence of damage to either the pump or the motor, flowever, they indicated that recent monthly performance tests, conducted at the pumps' rated flow (350 GPM), showed no performanco degradation.
Following the inspection, the licenseo further evaluated the conditions of the pumps and concluded that apparently no damage had occurred during the runout tests that had been previously-conducted.
11oweve r, the capability of-the pump to operate at runout conditions, in the event of a feedwater line break is unrosolved pending appropriato analysis and correct 1vo actions by the licenson.
This analysis should considor the effects on the motors' operability and environmental qualification if the motors aro operated above their nameplate rating under worst environmental and voltage conditions.
In addition, the analysis should address the setting of the breakers' protective devices to ensure that the breakers do not trip on overload (50-334/91-80-13).
Other than the considerable AFW pump load increase.on the diesel generator,'the load study, calculation lio. 8700-DEC-048, identified other minor load increases over the FSAR values for ar. estimated total of 296 IIP.
The effects of these added loads on the operability of the diosol generators are discussed elsewhere in the report.
Proposed Resolution:
50-334/91-80-13 The auxiliary feedwater pump concerns _ have been evaluated, and'it has been concluded that the auxiliary feedwater pumps-at Beavor Valley Unit 1 can experience a runout condition under postulated accident and test conditions.
Based on a review of past and current test data,'the pumps show no signs of degradation and current performanco is
-.~
_ _ _. _ _. _ _. _ _ _ _ _ _ __ - -.m. _ ___ _ _.
4 Page 19 of 24 acceptable.
Test procedures will be revised to assure that runout conditions are not experienced during future testing.
The manufacturer has confirmed that the pumps are capable of providing acceptable operation following a ten minuto period at runout conditions.
The motor - loads have been evaluated for the postulated maxinum loading condition.
Based on worst caso conditions of temperature rise and ambient temperature, the winding temperatures remain well below the allowabic value.
Relay trip settings will be further ovaluated to determine tho appropriate setting considering pump protection and maximum postulated loads Scheduled Date.
50-334/91-80-13 The electrical evaluation will be completed by September 1,
1992.
The test procedures will be revised by August 31, 1992.
11.
Switchgear Seismic Qualification 50-334/91-80-14 and 50-412/91-80-14
==
Description:==
l During a walkdown, the team noticed an unusual amount of
-480 V breakers in~ the racked out position and expressed concern regarding the impact of such configurations on the seismic qualification of the switchgear.
The licensee stated that the
" racked out" configuration had been evaluated by way of "in-situ" testing of safety related
.MCCs and that this had showed virtually no change in vibratory response, ieveddespite the numerous racked out pans.
Therefore, they bel that the 480 V.switchgear would I
respond in a similar manner.
The licensee-also indicated that the issue was under review and it-would be resolved-by analysis or test or a combination of both.
This item is unresolved pending completion of the licensee's evaluation:(50-334/91-80-14) (50-412/91-80-14).
Proposed Resolution:
50-334/91-80-14 and 50-412/91-80-14 A Duquesne Light. Company internal " Safety System Functional Evaluation" performed prior to the EDSFI inspection 4
questioned the effect of breakers left in the " racked out" position on the seismic qualifications of-the 480 V Motor-i Control Centers in Unit 1.
As a result, "in-situ" testing
-was performed which demonstrated no appreciable effect on the. seismic qualification.
i
. - -. - - ~. -. -.
. - -. -... ~- -
i Page 20 of 24 The Mccs used in Unit 2 are of a different manufacturer, khich can be " racked out " without opening the compartment door, and will be evaluated for that condition.
A similar review will be performed for the 480 V switchgear for both Unit 1 and Unit 2 to determine if there is any appreciable effect on seismic qualification due to the racked out configuration, Scheduled Date:
50-334/91-80 The Unit i evaluation will be complete 6 months after the 9th refueling.
50-412/91-80 The Unit 2 evaluation will be complete 6-montha after the 4th refueling.
j 12.
Rating of Diesel Generator pts 50-334/91-80-15 I
==
Description:==
A review of BV-1 drawings 8700-RE-21BT revealed that the Potential Transformers (pts) for the voltage regulator and the static exciter were rated at 2,400/120 V and 2,400/240 V,
respectively.
These rating are adequate when the EDG is operated in the test mode with its "Y"
point grounded.
L However, when the EDG is operated with the "Y"
point ungrounded, as in Design. Basis-Accident modo, a ground on one phase would drive the other two phases to 4,160 V with respect to ground.
The concern was that ground detection
- relays, in this application, are normally set at approximately 21 amps to eliminato nuisance trips.
Therefore, a small ground on any i
phase would_go undetected.
This ground, however, would be adequate to elevate the potential of the ungrounded phases to 4,160 V
above ground.
The potential transformers associated with these phases would then be exposed to a bushings,V between live parts and the core steel potential of 4 160-with potential damage to the pts.
and case
-Damage to the. PT 's insulation would ultimately adversely impact the operation of the voltage regulator and the static exciter.
-By the end of the. inspection, the licensee was not able to provide design bases documents to show that the insulation rating of these pts was adequate for operation with a postulated grounded phase.
Tnis issue is unresolved pending appropriate review and analysis by the licensee
-(50-334/91-80-15).
+
a
... ~... -.
Page 21 of 24 Proposed Resolution:
50-334/91-80-15 In order to demonstrate that'the insulation rating of the pts in question are adequate to operate with a postulated grounded phase, it is essential that the name plate data be retrieved, Since a bus outage is required, the nameplate information will be obtained and evaluated during the next refueling outage.
Scheduled Date:
50-334/91-80 This will be resolved prior to restart from the 9th Refueling Outage.
I i
13.
Generator Bearing Cooling 50-412/91-80-16
==
Description:==
During an inspection of the BV-2 EDGs, plastic pipe was usrad for-the cooling water supply to the rear bearing of the EDGs.
This pipe appeared to have been replaced and, in one case, the use of a toothed tool was evident.
The team was concerned that a
failure of these lines could ultimately render the EDGs-inoperable as a result of a rear bearing failure, loss of jacket cooling water, or shorting of the generator from the broken pipe water spray.
The licensee had no analysis clearly demonstrating the capabilities of the pipe.
However, they indicated that, to their knowled the units had been supplied with the plastic pipe ge-Apparently,.the reason for the pipe was that an electrical insulating material was required to totally isolate the rear bearing from the-rest of the EDG.=
Regarding the observation that the pipe appeared' to have been replaced with one of a different color the licensee noted that -they had replaced six. of the eight installed pieces of pipe over a period of several years, including l
l some that had been damaged and replaced during start-up.
The last replacement occurred in October 1990.
According to Maintenance Work Request No. 909461, the pipe had been broken in euch a manner as to allow the jacket cooling l
Water =to go into the bearing oil.
During this replacement, the Maintenance Department-requested Engineering to review and approve-the-use
-of a
material with physical l
characteristics different-from the originally specified ones.
The new material was approved.
However, the team i
found no' evidence that' an evaluation had been done of the new material's performance in the environment of the EDG l
i l
e tl.
.a.
i Page 22 of 24 room and compared to the original material's specification requirements.
In view of the above,-the acceptability of the new plastic pipe is unresolved pending appropriate analysis by the licensee (50-412-91-80-16).
Proposed Resolution:
50-334/91-80-16 An analysis was performed to justify the installed PVC piping.
The analysis conservatively assumes the following:
2 the worst case i
- 1) lowest strength PVC piping installedseismkc ) and operating environmental temperatures, 3) condition loading and 4) dicateconservative dos,ign values _ on the piping.
The results in that the PVC p; ping is suitable for-this application.
Additionally, the PVC pipe is compatible uith and recommended for use with the HALCO 39M coolant additive,.
per NALCO's product bulletin.
[
Therefore, the installed PVC pipe is acceptable for use in the EDG bearing cooling water connections for the life of j
the plant.
Scheduled Date:
No additional action is required.
14.
Relay Testing 50-334/91-80-17
==
Description:==
i The calibration was witnessed of a Class 1E undervoltage relay used to start the Unit l
'A' diesel generator.
During this test, the Asea. Brown Boveri relay exhibited a
?
setpoint drift which appeared to be temperature related since the test cart was located-in an area where a cold draft was blowing on the cart and on the relay.
The testing personnel stated that a letter would be sent to Nuclear Plant Engineering requesting an evaluation and 'a determination of the impact of this condition on plant
-operation.
By the end of the inspection, the licensee had not completed its evaluation.
The team also reviewed letter RBRB142, dated September 6,
t
,,-,,.4
_,J.
r-----
,,_..u
...,...Ay
,.._---4
i i
Pago 23 of 24 The team also reviewed lottor RBRB14 2, dated September 6,
l 1991, which discussed a sotpojnt problem with relays 27-VB100 and 27-VC100.
Those relays aro used to detect undervoltage on the supply to the reactor coolant pumps.
l
- The writor of the lottor had suggestod that, the day before i
their testing, the relays should be set outsido their Technical Specification limits so that, by the time they woro tested, they would have drifted into the correct band.
The team requested a copy of Engineering. Memorandum flo. Eh l
101626 which responded to the letter, but the licensco was not able to retrieve the response by the end of the inspection.
Thoroforo, the responso of the undervoltago i
relays to temperature changos is unrosolved pending 5
i appropriate ovaluation by the licensoo and review by the NRC (50-334/91-80-17),
i Proposed Resolution:
50-334/91-80-17 The relay in question is an ASEA Brown Boveri (formerly ITE) Type 47H, class lE undervoltage relay. The setpoint for-this relay is 99 volts with an acceptable range of 99 i
to 102 volts per the applicable Relay Calibration Proceduro i
(RCP).
The applicable procedures for this relay do not currently address the environment in which the relays aro tested, and will be revised to reflect this.
The relay
' l testing personnel are currently ensuring that prior to testing, the relays are in a stable environment. Subsequent i
relay testing has demonstrated. that the relay sotpoint remains in its range when calibrated in a more stable test environment, t
Relays VB100 and 27-VC100 are G.E.
Typo CTV.
During the pt..tmance of monthly Maintenance Surveillanco Procedures (HSPs),
the relay. "as found" setpoints were found outside the setpoint range specified.
These past setpoint shifts have since been attributed to cooling of the coils due to a time delay betwoon the de-energization i
of the coil and checking the "as found" setpoint.
The i
manufacturer's instruction manual GEI-15536G states that the rolay setpoint -will increase after heating from energization.
This effect corresponds to that seen from reviuving past relay "as left" and "as found" setpoint
. data.- -Based upon this review, the applicable surveillance procedures will be superseded with a new procedure which will requiro checking the relay setpoint immediately after de-energization.
The relay technicians are aware of this situation and are implementing the proposed changes.
8 P
r
..r_-,-..
,.m
_.._ _ _ _.,-....m.-
.___.m.,~-._
- 4. e--
._._______________.____._____m_..__
Page 24 of 24 i
l n reference to letter RBR1314 2, the letter suggested
.owering the relay sotpoint prior to performing the monthly calibration to attempt to compensato for this suspected heating offect.
Note that this was never done or seriously considered.
- Also, note that it was never suggested to lower the sotpoint below Technical Specification limits, only to a
limit slightly below that contained in the applicable survoillance procedure which 10 considerably higher than the Technical Specification limit.
Scheduled Dato:
50-334/91-80 The applicablo relay procedures will be revised and personnel re-trained by September 30, 1992.
h r
4 f
9 2
9 l
i i
l I
l.. _. _ -,._
-. _ _. _ _. _, _. -.,. _, _ ~... _ _,
, _. _ _ _ _ _ _. _ _ _ _. _. _. _ _ -. -. _.. _.. _. _... _ _ _ -