IR 05000317/1993021
| ML20057C995 | |
| Person / Time | |
|---|---|
| Site: | Calvert Cliffs  |
| Issue date: | 09/22/1993 |
| From: | Bhatia R, Ruland W NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I) |
| To: | |
| Shared Package | |
| ML20057C990 | List: |
| References | |
| 50-317-93-21, 50-318-93-21, NUDOCS 9309300218 | |
| Download: ML20057C995 (11) | |
Text
.
-.
-
,
e
.
.
.
U.S. NUCLEAR REGULATORY COMMISSION -
-
REGION I
-
REPORT / DOCKET NOS.: 50-317/93-21 l
50-318/93-21
,
-
LICENSE NOS.:
DPR-53 DPR-69 t
LICENSEE:
Baltimore Gas and Electric Company j
MD Rts 2&4, P.O. Box 1535 Lusby, Maryland 20657
,
FACILITY NAME:
Calvert Cliffs Units 1 and 2
!
INSPECTION DATES:
June 28 - July 2,1993
!
INSPECTOR:
W p~lA6 9[Z'J/93 R. Bhatia, Reactor Engineer, Date-i Electrical Section, Engineering Branch, Division of Reactor Safety
!
rf7 72293
APPROVED BY:
W~ iam H. Ruland, Cljfeff'Eledtrical
/ bat </
ction, Engineering Branch, Division of Reactor Safety Area Inspected: This was an announced inspection to review the licensee's corrective action
,
of the previously identined electrical distribution system functional inspection (EDSFI)
Endings.
Results: One violation was identified during this inspection. Six open items were reviewed and five items (one violation and four unresolved items) were closed. One remaining unresolved item pertaining to the HVAC system was updated and remains open. These items are discussed in Section 2.0 and 3.0 of this report,
,
9309300218 930923 7,
PDR ADOCK 05000317 pf G
PDR d2
.
,
.
-
I
!
DETAILS
.
1.0 PURPOSE i
The purpose of this inspection was to review and verify the licensee's corrective actions for findings identified by the NRC during the 1992 Electrical Distribution System Functional
,
In@cction (EDSFI) for Calvert Cliffs Unit I and 2. The review included the licensee's actions, addressed in their written responses to the EDSFI report dated July 8, August 14, and September 9,1992, and to the Notice of Violation, dated September 3,1992. The review also included the licensee's actions, which addressed the subsequent NRC clarification
'
response of March 22,1993, and the licensee's written response to this letter, dated April 28,1993.
j 2.0 FOLLOW-UP OF PREVIOUS IDENTIFIED FINDINGS (UNIT 1 and 2)
2.1 (Closed) Violation Itern Nos. 50-317/92-80-001 and 50-318/92-80-001 i
During the previous EDSF1 inspection, the team identified two procedural concerns pertaining to the surveillance testing performed on the 4.16 kV bus degraded voltage relays and the frequency measurement of the emergency diesel generators (EDGs). First, procedure STP-M-522-2 did not include adequate acceptance criteria for the relay reset settings and also failed to evaluate the documented higher test voltage (*05.77 volt) to assure that the relay would operate adequately. Secondly, in emergency diesel generator surveillance test STP-04, the licensee did not adequately record the highe.st frequency during the EDG start load l
sequencing steps.
- Based on the licensee's response to the above violation, the inspector noted that the licensee,-
by letter dated August 14,1992, and subsequent phone clarification, had justified the actual
"as-found" relay reset value of 105.77 volts and concluded that the relay setting was still
!
within the design margins and could function as designed considering the instrument
,'
tolerances. The licensee had also clarified in their response that the 61 hertz frequency recorded on the surveillance test procedure was appropriately considered between EDG
!
sequencing steps 1 through 7 instead of the assumed initial EDG zero step loading sequence in accordance wa the UFSAR. The NRC review of the additional licensee information resulted in redefining the above violation with a different concern. The inspector noted that the NRC revised violation stated that the licensee had failed to establish an adequate procedure to control relay reset-points in the surveillance test procedure STP-M-522-2. This
change was documented in the NRC letter, dated March 22,1993. In response, the licensee committed to adequately specify all 4.16 kV bus degraded voltage relay reset setpoints
acceptance limits in their surveillance procedures prior to their next use.
i During this inspection, the inspector noted that the licensee had performed the 4.16 kV bus 24 associated relay evaluation of the "as-found" reset voltage (105.77 volts) and
_
concluded that the "as-found" values were outside their relay setting limits; however, they j
I l
_-
.
did not represent a potential challenge to the other systems. In addition, the licensee also i
had reevaluated the last surveillance test results for both units. The relay test data reviewed -
by the licensee was found to be adequate for all 4.16 kV buses. The inspector also noted i
that the licensee had included the degraded relay reset voltage acceptance limits (104.9 volts or less) in the procedure STP-522-2, Revision 8, for Unit 2, 4.16 kV buses 21 and 24.
The inspector's review of the surveillance test results completed on May 24,1993, revealed that all relays associated with the Unit 2, 4.16 kV buses, were reset to less than the 104.9 volts acceptable established settings. The inspector's review of the test results of the Unit I relays associated with buses 11 and 14, performed on July 23,1992, as per the old
procedure STP-M-522-2, Revision 6, revealed that all relay settings were left within the
,
present design guidelines (104.9 volts or less). The hcensee is planning to update the Unit 1 procedure to reflect the new requirements prior to its use by the next refueling outage.
I Based on the above actions completed by the licensee to assure that both units degraded voltage relays were reset within the acceptable established design limits and that adequate
,
reset criteria were included in the Unit 2 procedure, the inspector concluded that the licensee had taken adequate corrective actions to resolve the above concern. This item is closed.
2.2 (Closed) Unresolved Item Nos. 50-317/92-80-008 nnd 50-318/92-80-008
,
During the EDSFI inspection, the team was concerned that the licensee had an inadequate emergency diesel generator (EDG) loading calculation to demonstrate the EDG capability.
l The licensee was committed to finalize calculation No. E88-15, " Diesel Generator Accident
'
Imading," to ensure that the loading was within the diesel generator ratings including
verification of design input data, methodology, and assumptions.
Prior to this inspection, the licensee had revised the above EDG loading calculation. The inspector noted that the licensee's EDG loading calculation had considered and included the
,
various load data based on updated design drawings and walkdown verification. All small
,
motor power factors and efficiency values were considered. The data, in some cases, was obtained assuming similar motor design values and NEMA design standards. The licensee had also considered in this calculation the kVAR loading of power transformers, cable losses, motor loading based on the applicable operating conditions, and load diversity factors addressing various loading combinations allowed by the current operating procedures. The i
'
methodology used to establish large centrifugal pump loads was based on the highest determined flow rate. Brake horsepower values were determined using manufacturers' pump
,
curves at the determined flow rates. If the flow rate was bounded by the curves, the brake
!
horsepower was calculated using the head and the efficiency curves. For EDG loading, the
largest brake horsepower value among those calculated for pumps in a given service (e.g.
component cooling, salt water, etc.) was used for all similar pumps on both units. For example, the brake horsepower value for each charging pump was determined from the manufacturer's curves and the largest value was used for all charging pumps. The licensee also considered the most conservative values for the large fans, HPSI and LPSI pump motor i
i
. - _ - _---_- __- __
.
i
.
i
.
,
.
.
loads. Based on the licensee's calculations, the inspector noted that the worst-case EDG
,
loading would occur during a main steam line break condition in the first minute. During
.!'
' this worst-case condition, and assuming a single failure of the second EDG, the EDG load was determined to be 2732 kW.
The EDGs at the Calvert Cliffs station are rated at 2500 kW for continuous duty,2700 kW
'
for 2000 hours0.0231 days <br />0.556 hours <br />0.00331 weeks <br />7.61e-4 months <br /> and 3000 kW for 200 hours0.00231 days <br />0.0556 hours <br />3.306878e-4 weeks <br />7.61e-5 months <br />. Based on the worst-case EDG loading condition, the inspector noted that the EDG in the first two minutes would exceed its 2000 hour0.0231 days <br />0.556 hours <br />0.00331 weeks <br />7.61e-4 months <br /> rating capability. The licensee stated that they are upgrading the existing EDG ratings to a higher rating to envelope the current loading. In addition, the licensee is also committed to add additional EDGs at the station to enhance the overall onsite power reliability. Upon completion of the existing EDGs rating upgrade, the licensee plans to update the UFSAR and complete the 10 CFR 50.59 safety evaluation and other associated documents that may be affected, including the Technical Specification surveillance requirement. The inspector noted that the licensee had evaluated the Technical Specification Surveillance requirement, as speci6ed in Section 4.8.1.1.2.d.5, and concluded that the initial transient load of no more l
than 2750 kW (short-term two-hour rating) for the first two minutes of an accident could be considered adequate and acceptable for the EDGs operation. The inspector had no further concerns at this time.
The inspector reviewed selected motor input data and compared it with pump curves and nameplate ratings. In addition, the inspector reviewed the licensee's assumptions and veri 6ed the selected plant operating loads during the walkdown to assess the adequacy of this calculation and found no concerns. Based on the completed load calculation, committed actions to upgrade the rating of the existing EDGs, and add additional EDGs at the station, this item is closed.
2.3 (Closed) Unresolved Item Nos. 50-317/92-80-014 and 50-318/92-80-014 During the EDSFI inspection, the team was concerned that the short circuit current contribution from the two battery chargers (due to a fault) was considered only 110% of full l
load capacity. Since these chargers used silicon controlled rectifiers, it is possible that short l
circuit current contribution of each charger could be as high as ten times the full load rating in the initial period of eight milliseconds. The team was concerned that the de bus feeder
,
fuses would have to operate outside their tested rating and that their coordination with the mid-span fuse might be compromised.
-
I
-
-
-.
I u
l
)
.
.
l During this inspection, the inspector noted the licensee had performed a transient circuit
'
analysis documented on an internal memorandum, dated July 1,1993. The study was done by computer modelling the circuits under the worst-case de fault current on the smallest fuse downstream of the dc bus, assuming a bolted fault. The licensee concluded that the fault was less than the tested fuse rating. The inspector's review revealed that the licensee's analysis had considered each battery charger contribution to be ten times the full-load rated capacity including the appropriate battery, cable and battery charger impedances. The model was
,
!
prepared using the guidance provided in the IEEE Standard, 946-1992, Annex B and D,
'
"IEEE Recommended Practice for the Design of Safety-Related DC Auxiliary Power Systems for Nuc1 car Power Generating Stations."
The inspector's review of the analysis revealed no concerns. The evaluation of the licensee's
'
study and coordination curves revealed that the maximum short-circuit current to the smallest j
fuse (100 ampere) downstream of the de bus would be approximately 13,500 amperes at i
!
about six milliseconds. Since the fuse rating of 15.200 amperes bounded the bus fault current, and the adequate contribution of the battery chargers was considered, the inspector concluded that all the fuses could operate, if required, within the tested rating under these conditions and that the fuses were adequately coordinated. This item is closed.
l 2.4 (Closed) Unresolved Item Nos. 50-317/92-80-007 and 50-318/92-80-007 During the EDSFI inspection, the team determined that the cable ampacity selection criteria had allowed the selection of a cable size less than the 125% of the National Electric Code full-load current values for 20,25,50 and 60 HP motors. In addition, the teafn was concerned that the 4.16 kV buses 11 and 12 feeder cables may be overloaded under plant worst-case loading conditions.
The licensee, in their response letter to NRC on September 9,1992, stated that the Calvert Cliff licensing design basis, as documented in the UFSAR Section 8.1.2.0, states that cable sizing is based on the IPCEA standard criteria, which is less stringent than the NEC criterion. During this inspection, the inspector noted that the licensee had completed the evaluation of the plant installed cables. The licensee concluded that all existing plam cables meet the ampacity guidelines of the E-4% (the installation specification), with the aception of 11 cable circuits. Further review of these 11 motor-feeder caules, assuming a degraded voltage of 90% nominal, revealed that all but one motor feeder cable had at least 111% of full-load current carrying capacities. Bascd on this rated ampacity (111%), these ten motor feeder cables could adequately operate continuously at rated full load current. The remaining one cabic feeder pertaining to the purge air exhaust fan ampacity was determined to be 69 amperes by the licensee as per the E-406 guidelines. Based on discussion with the licensee, the inspector noted that they had taken the station operating data of the purge air fan and found this motor running current to be 70.3 amperes, higher than the cable and the nameplate full load amperes rating (65 amperes). The licensee is investigating this high motor current and has postulated that motor vibrations may be the cause. The inspector noted that the licensee is tracking this in Issue Report No IRO-003441, initiated on
.g
,.m,,,,.
.
-
- -.,
- - - - - - -
.
.-r- - - - -
,-r-----
- - - -
-.e
-
,
.
l
.
t
.
June 15,1993. The inspector noted that this fan motor is only operated during plant modes 5 and 6, therefore, the licensee's corrective action would be completed at a later date. Based on the design review and the function of this fan motor, the inspector concluded that the concern regarding this feeder cable, under the above worst-case condition, was not safety significant.
To resolve the future design control concern, the licensee has revised E-4%, " Guidelines for Small Motors," to include the cable rating to be 125% of full load ampere rating. In addition, the "as-found" acceptable ampacity rating of 111% for the above eleven motors was
,
also revised accordingly.
The licensee also performed an evaluation on February 15,1993, and determined that the t
load on the 4.16 kV feeder cable for buses 11 and 12, under the worst-case loading condition and degraded voltage, were 720 and 310 amperes, respectively. Each feeder cable was rated for 1436 amperes. Tne licensee concluded that both feeders were capable of operating at full load with no significant temperature rise under these conditions.
The inspector reviewed selected documents and verified the results by hand calculation. In addition, motor data taken during the walkdown was used to verify input data. Based on the above corrective actions to demonstrate ample cable ampacity, the inspector concluded that the licensee had adequately addressed the above concerns. The remaining one motor circuit I
concern is of low safety significance and is being investigated and tracked by the licensee; this item is closed.
2.5 (Closed) Unresolved Item Nos. 50-317/92-80-012 and 50-318/92-80-012 The EDSFI team identified a concern that, every time a compressor starts, the air supply check valves, which are part of the EDG starting air system, chatter due to pressure transients from the positive displacement pump. These check valves separate the non-lE safety-related compressor piping from the downstream safety-related piping system, however, the performance of the check valves was not monitored as part of the surveillance test i
program.
During a previous EDSFI followup inspection, the inspector noted that the licensee had issued engineering test procedure 92-95R to monitor the performance of the check valves.
I This procedure contained adequate details and instructions for the disassembly, inspection, repair, and assembly of the valves. The licensee had prepared the maintenance work orders to test the Unit 2 valves, and the work was scheduled to be performed during the Unit 2 outage in spring 1993. Based on the results of the Unit 2 valve inspections, the licensee was expected to establish a periodic monitoring schedule for the check valves.
-.
..,- -
. - _ - - - - - - - - _
--- - _...-
.,
.
.
.
.
.
During this inspection, the inspector noted that the licensee had functionally tested four of the six Unit I and 2 check valves during the 1993 Unit 2 refueling outage. Three of the four tested valves (12-DSA-106,12-DSA-Il0,21-DSA-106,21-DSA-110) were found to be satisfactory per test procedure ETP 92-95R. After the initial test, the failed valve (12-DSA-110) was replaced with a similar replacement valve. The replaced valve failed its initial test and was then cleaned, lapped, and reassembled by the licensee. After the rework, the valve was tested and the results were satisfactory. The licensee review of the original check valve 12-DSA-110 revealed that a small amount of rust in the guide area had caused the excessive leakage during the refueling test. In the case of the replacement (new) valve, the licensee determined that it failed the first time because of the shrink wrap on the valve might have trapped moisture condensed on the carbon steel valve internals and caused rusting, which was removed prior to the second test.
Based on the testing and "as-found" condition of the four valves, the licensee was in the process of establishing a preventive maintenance (PM) task for these valves that would be performed every third refueling outage. The inspector determined that the licensee initiated this task PM repetitive request (10242040) on July 1,1993, and upon completion of the final processing, it would become a pan of the automatically generated PM task schedule. The inspector noted that the licensee had also included all (six) valves in their check valve reliability program (MN-1-108). Under this program, the check valve data will be collected, trended, and evaluated for system reliability. Based on the minimum maintenance required on these valves in the past (approximately) 20 years, the licensee concluded that the emergency diesel generator air start systems pressure boundary was adequately secured, and the existing stringent test acceptance criteria will ensure continued valve operability.
Based on the review of the test results performed, the fact that no significant deterioration of the valves had occurred, and the establishment of a routine PM program for the valves, the inspector concluded that the licensee had taken adequate corrective actions to resolve the above issue. This item is closed.
2.6 (Open) Unresolved Item Nos. 50-317/92-80-009 and 50-318/92-80-009 During the EDSFI inspection, the team identified that the licensee analysis / calculations for the ventilation in the 27 foot elevation switchgear room did not consider the total ! css of ventilation due to a postulated fire in the 45 foot elevation switchgear room. The team was also concerned that the HVAC system was susceptible to various common mode failures because of a common supply and return duct configuration. In addition, the various operating procedures lacked adequate instmetions to control the room temneratures under various postulated conditions.
-
-
. _ _ _ _. _ _ _ _ _ _ _
_
O
.
.
..
-
During this inspection, the inspector noted that the licensee had performed an overall assessment of the station HVAC system and had evaluated the heatup condition for the switchgear rooms on the 27 and 45 foot elevations, assuming a total loss of the HVAC system. This evaluation (calculation No. M-92-226, Revision 0) takes into account the compensatory measures currently in place, such as portable fans and a dedicated power supply to reduce the switchgear room temperatures. The inspector noted that the licensee had revised the switchgear ventilation air conditioning procedure, OI-22H, based on this-evaluation. SpeciSc detailed instructions were added to the procedure to assure that the switchgear room temperature limit of 150*F was not exceeded. The inspector's review revealed that, in the event of a total loss of the ventilation, the procedure instructs that the emergency ventilation (a dedicated power supply and fan unit) for each switchgear room be placed in service within 30 minutes. In addition, the procedure states that the unit should be tripped manually and be placed in hot standby within six hours and be in cold shutdown within 30 hours3.472222e-4 days <br />0.00833 hours <br />4.960317e-5 weeks <br />1.1415e-5 months <br /> to avoid cxceeding the maximum operating temperature of the switchgear.
This stringent requirement to shut down the units was chosen by the licensee to resolve the above HVAC concerns. The licensee also revised the alarm procedures (IC10 and 2C10) to take appropriate actions upon receipt of a high switchgear room alarm temperature, which is set at 101 F.
The inspector reviewed the above assessment and evaluations. The dedicated equipment for the switchgear room ventilation was verined during a plant walkdown. Based on the review of the above related documents, the inspector identined a concern that the emergency ventilation equipment installed in the plant was nonsafety-related, meeting only the 10 CFR 50 Appendix-R requirements. In addition, the licensee had committed in their previous response that this will be an interim response, and additional systematic evaluation of other HVAC systems and Appendix R would be completed by the end of 1994. Upon completion of these evaluations, additional long term modi 6 cations may be implemented. Based on the review of the procedures and short term compensatory measures taken, the inspector concluded that this item will remain open pending completion and review of the long term evaluations.
3.0 EDG LOAD SEQUENCER During this inspection, the inspector was informed by the licensee that a loss of coolant incident (LOCI) sequencer timer for the auxiliary feed pump No.13, Unit 1, was incorrectly set at 5 seconds. The inspector determined that this de6ciency on Unit 1 existed from the period from November 30,1992, through July 1,1993. The licensee's planning group found this discrepancy during research being performed to resolve a minor drift concern associated with the similar relay on Unit,
_..
_
_
_.
_
_
-
-
.
A
.
'9 l
-
.
The inspector reviewed the facility change request (FCR) 92-0210 and identified that the Unit 1 and 2 auxiliary feed pump (13 and 23) sequencer relay time settings were revised from 5
seconds to 14.5 seconds and that other logic design changes had been made by the licensee
>
on June 6,1992. These changes were required to resolve a concern that an unacceptable -
voltage drop in EDG output voltage could occur, assuming accident conditions concurrent
,
with the loss of offsite power (LOOP).
l On July 1,1993, while sequencer timer values of both units were being compared to resolve
,
'
a Unit 2 sequencer timing drift concern, the licensce's planning department individual observed that the Unit 1, auxiliary feed pump No.13, time setting had been changed from
,
the existing 14.5 seconds to 5 seconds.
Further review of the documentation by the inspector revealed that the licensee, upon receipt of a vendor letter on March 28,1989, had initiated FCR 904)29 on June 1,1990, to replace
.i
. the station 140 relays with a newer version grade "E" Agastat relays. These relays were t
being replaced on a low priority basis as systems became available. The inspector noted that the auxiliary feed pump No.13 relay was replaced on November 30,1992, and tl.e new
,
relay timer was set at 5 seconds, based upon the old timer setting sheet. The relay time setting sheet, though not fully approved, was attached to the FCR at the time of its original issuance. At the conclusion of this inspection, the inspector noted that the licensee had -
,
promptly corrected the sequencer time for the auxiliary feed pump. The licensee had
!
declared the affected pump inoperable until the time setting was corrected on July 2,1993.
,
The inspector concluded that the licensee's review process, in implementing the above FCRs, had failed to adequately review and assure that the latest design requirement of LOCI sequencer time (14.5 seconds, i 3 seconds) was utilized. The auxiliary feed pump No.13 is an important safety pump and is required to operate during accident conditions. The 5
'
second sequencer time sctting for the auxiliary feed pump could have resulted in an overload of the respective EDG above its designed sequenced load and reduced the bus voltage
-
significantly. However, the review of the voltage drop analysis performed by the licensee undct above overloaded condition was found to be acceptable. The licensee's lack of adequate design review of the above FCRs constitutes a violation of 10 CFR 50, Appendix B, Criterion 111, which states in part, that, " Measures shall be established to assure that
.
applicable regulatory requirements and the design basis, as defined in f 50.2 and as specified in the license application, for those structures, systems, and components tow which this appendix applies, are correctly translated into specifications, drawings, procedures, and instructions...The design control measures shall provide for verifying or checkits 'he adequacy of design, such as by the performance of design reviews, by the use of nternate or-simplified calculational methods, or by the performance of a suitable testing program." (50-317/93-21-001)
-_
_
_
-
,
1
l 4.0 UNRESOLVED ITEMS I
Unresolved items are matters about which additional information is necessary in order to determine whether they are acceptable or they constitute a violation. Several unresolved
'
items are discussed in detail under Section 2.0.
5.0 EXIT MEETING The inspector met with the licensee's personnel denoted in Attachment 1 of this report at the.
,
conclusion of the inspection period on July 2,1992. ' At that time, the scope of the inspection and the inspection results were summarized. At no time during the inspection was written material given to the licensee.
f Attachment: Persons Contacted
,
,
t
.
I i
i
,
P b
f b
b-4
.,,
_
_ _ _
+
...
,,
,
_ _ _____
_ _ _ _. - _ _ _ _ _ _ _ _ _ _ - - - - _ _ _ _ _ _ _ _ _ _ - - _ _ _ _.
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - - - - _ _ _ _
-
h
.
.
.
,
.
I1
.
ATTACHMENT 1 Persons Contacted Baltimore Gas and Electric Company
,
I l
+ Anuje, A., Supervisor Quality Audits
+ Buttner, R., Sr. Engineer
+ DeAtley, R., Sr. Engineer Detter, G., Director, Nuclear Regulatory Matters
Eizenberg, Z., Mechanical Engineer - NES
+ Getty, B., Nuclear Safety and Planning Department Gladey, D., Engineer, DES
,
I Hunter, G., Engineering Analyst - Maintenance Kemper, B., Principal Engineer
+ Maki, W., D., Compliance Engineer McHale, J., System Engineering
- + McLann, E., V., Engineer
- + McVicker, J., Principal Engineer, Nuclear Engineering Design
+ Nolan, C., Mechanical Engineer - NES
- + Pavis, G., Director, Strategic Engineering
- + Sebra, K., Engineering Consultant, Nuclear Engineering Design Sly, C., D., Compliance Engineer
Thurston, B., M., Quality Audit
Vincent, D., R., PES
+ Volkoff, J., Compliance Engineer U.S. Nuclear Regulatory Commission (USNRC)
]
- + Lathrop, K., Resident Inspector
+ Wilson, P., Senior Resident Inspector
+ Those attending entrance meeting on June 28,1993 Those attending exit meeting on July 2,1993
- - _ - _ -__-__ _ _____-______ - __ ____-- ___ ___ __ _ _-_______- ___ _ ____- ___ _