IR 05000373/1993036
| ML20059K619 | |
| Person / Time | |
|---|---|
| Site: | LaSalle  |
| Issue date: | 01/25/1994 |
| From: | Falevits Z, Gardner R, Winter R NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION III) |
| To: | |
| Shared Package | |
| ML20059K610 | List: |
| References | |
| 50-373-93-36, 50-374-93-36, NUDOCS 9402020183 | |
| Download: ML20059K619 (16) | |
Text
-,
.
._
.
_
_
'
.
,
t U.S. NUCLEAR REGULATORY COMMISSION
REGION III
Reports No. 50-373/93036(DRS), 50-374/93036(DRS)
Dockets No. 50-373, 50-374 Licenses No. NPF-ll, NPF-18 Licensee:
Commonwealth Edison Company
.
Executive Towers West III
'
1400 Opus Place, Suite 300
,
Downers Grove, IL 60515
'
facility Name:
LaSalle County Station, Units 1 and 2 Inspection At:
LaSalle Site, Marseilles, Illinois inspection Conductr:d: dovembcr 29 through December 21, 1993
,
/
Inspectors:
f, 7
/ 2M7</
,
Z. Fale'vits Da' te '
&.2e$ti$-:
//2s/94 R. Winter
'Date'
Approved By:
d[ / J d'
'
fronald N. Gardn#, Chi Date Plant Systems Shction e Inspection Summary:
,
inspectj_on on November 29_through December 21. 1993 (Reports No.
Areas Inspected: Announced routine inspection of electrical and
,
instrumentation and control (l&C) modifications conducted in accordance with NRC Inspection Procedure 37700.
Results: The team determined that the electrical modification process was
,
generally acceptable. Four apparent violations and one deviation were identified:
(1) failure to identify the root' cause(s) and take corrective-action to prevent repetitive failure of the secondary containment isolation (VR) dampers to close, (2) inadequate and failure to follow procedures, (3)
'
failure to demonstrate by testing that a 3 second time delay setting on the RPS EPM assemblies' overvoltage (0V), undervoltage (UV), and underfrequency
,
(UF) relays would perform its intended design function, (4) failure to assure l
that measuring and testing devices are properly calibrated to maintain
,
accuracy within necessary limits. Also, one deviation from the FSAR was noted relative to relay setting trip range for the RPS EPM assemblies.
l 9402020183 940125 PDR ADOCK 05000373'
O PDR,
-. -..
.. -
.
.
-
-
- - - -
-
-
-
..
.
..
DETAILS
.
1.0 Principal Persons Contacted Commonwealth Edison Company
- W. Murphy, Site Vice President
- J. Schmeltz, Operations Manager
- J. Gieseker, Site Engineering and Construction (SEC) Manager
'
- J. Miller, SEC - Plant Support Supervisor
- D. Bockman, SEC - Modifications Supervisor
- M. Cray, Master Instrument Mechanic
- C. Silich, Quality Control Supervisor
- M. Santic, Maintenance Supervisor
- R. Ragan, System Engineering Supervisor
- T.
Shaffer, Executive Assistant
'
- J. Lockwood, Regulatory Assurance Supervisor
- B. Bowers, Operational Analysis Department - Site Supervisor
- W. Kirchoff, SEC - Site Support Engineer
- C. Caskey, Station Quality Control
- E. McVey, Regulatory Assurance Assistant U. S. Nuclear Reaulatory Commission (NRC)
- R. Gardner, Chief, Plant Systems Section l
- D. Hills, Senior Resident-Inspector
'
- Denotes those present during the exit meeting on December 21, 1993.
<
2.0 6_ction on Previously Identified Inspection Findinas (0 pen) Unresolved item (373/93020-02(DRP)): (374/93020-02(DRP))
,
The installation of filtering capacitors in the control room ventilation-monitors has increased the time delay of the detector trip creating.the possibility of radiation exposure to control room personnel before ventilation j
isolation during an accident.
The inspectors could not easily determine whether GDC 19 limits on Control Room Operator radiation dose (30 REM thyroid)
i would be exceeded. Administrative controls are in place to preclude a control
'i room ventilation (VC) purge. This would mitigate the worst case scenario.
.i
'
This item will be referred to NRR. This item will remain unresolved..
3.0 Review of Failure History of Reactor Buildino Ventilation (VR) System I
Secondary Containment Isolation Dampers
.;
)
On November 29, 1993, with Unit 1 at 95% power the alternate power feed, which was' supplying the.H RPS bus tripped when the operator started.the Unit 2 reactor recirculation (RR) pump. The loss of this RPS bus caused both. Unit 1-
& 2 Secondary Containment Isolation Reactor Building Ventilation (VR)~ systems
2 i
- -
-
.
..
-
.
-
-
.-
'
.
,
to trip (actuate). On recovery, a full scram occurred on' Unit 2'when an operator, attempting to re-energize the H RPS bus, operated the RPS MG selector switch and turned it too far, deenergizing the other RPS bus.
The secondary Containment Isolation Reactor Building Ventilation' supply damper IVR04YA did not fully close as expected.
Thirty three minutes later, the damper was cycled by operators several times before the close indication occurred. The inspectors noted that although the damper showed open indication for 33 minutes, the operating crew did not declare the damper inoperable and did not enter the appropriate Technical Specification (TS)
action statement.
Technical Specification 3.6.5.2 requires that the VR damper close within'10 seconds and in the event of an isolation damper failure, the other of the two series dampers be maintained operable. Within eight hours, either the inoperable damper must be restored to operable status, or the affected penetration shall be isolated.
The secondary containment isolation system (VR) dampers (6' diameter) are designed with two half blades, each of which is actuated by a pressurized air cylinder. When its solenoid valve is de-energized, the ventilation air flow and return spring pressure close the damper.
The dampers close to prevent release of radioactive material while maintaining a 1/4 inch H,0 negative pressure in the reactor building using the SBGT system.
Interviews with the operating crew on shift indicated that VR secondary containment isolation damper failures to properly close or open have been a-problem.
The Station Control Room Engineer (SCRE) on shift during the event, stated that he thought it was a limit switch adjustment problem and did not declare IVR04YA inoperable because similar problems had occurred in the past.
.
During the event review and interviews, the following concerns were raised:
'
-
A work request was not issued (by the operating crew) to determine the root cause of the failure (and if parts replacement was required).
-
The equipment failure was not entered into the control room degraded equipment list to alert other operators.
-
An' operability determination was not made (in a timely manner as required by procedure).
-
The PIF written for the November 29, 1993 event (scram) did not document the isolation damper failure in the proper section.
No specific PIF was written to document and determine root cause of this failure.
-
Unit 1 Control Room Log stated "lVR04YA at least 99% closed."
(This damper is 6' in diameter and an accurate estimate per cent of closure was not possible). Also, the shift engineer's log stated that "1VR04YA did not show closed indication until subsequent cycling".
,
.
.
.
,
,
-
Step F2 of procedure LAP-220-5, " equipment operability determination"
. requires that operators determine if a previous evaluation exists for this damper by reviewing the equipment operability binder (s) in the control room. This was not done.
Instead, the control room operator stated that he reviewed the surveillance evaluation book which contained an analysis for a quarter (1/4) inch gap. However, there was no correlation between a quarter inch gap and an opening equal to 99%
closed.
,
-
Dampers ace tested monthly under no flow condition; however, during
-
operation the dampers are required to close under full flow conditions which apparently results in a pressure buildup condition between the inboard and outboard dampers.
In some cases, this condition prevented'
the damper from fully closing. Also, operability test LOS-CS-Ql, dated October 12, 1993, Section 11, states, "In the past, dampers have failed to fully close against a confined space due to the force exerted by.the compressed air.
If a full closure fails to occur, perform the following:
A.
Consider the damper inoperable..., and F.. Tech Staff will perform an operability evaluation per LAP-220-5." These actions appeared to be more stringent than the specific actions taken by operators when the same dampers fail to function during plant operations.
The inspectors selected for review VR equipment failure history Problem Identification Forms (PIFs), discrepancy reports, DVRs, AIRS,-and Operator
!
logs to evaluate licensee's effectiveness in identifying and resolving recurring equipment problems. The failure history file for the VR dampers was not well organized and could not be easily retrieved. The inspectors determined that by December 1992, 35 DVRs were issued on VR damper problems, most for failure to close or open. No root cause analysis (RCA) was.available for review.
Also, within the last 12 months, 24 PIFs were issued to document VR damper problems. None were classified level 3 (requiring RCA) although in at least 9 of the 24 instances, the dampers failed to function properly in 1993.
The inspectors noted that 32 work requests have been issued to perform corrective maintenance work on damper IVR04YA in a five year period; many were issued for the damper's failure to fully close.
In addition, various work requests have been issued to address similar problems noted with the other Unit I and 2 secondary containment isolation dampers. NRC report 373/89012 documented additional VR damper failures that occurred prior to 1989.
Some of the VR damper failures noted in the work requests were attributed to:
-
Solenoid valves not operating properly (most frequent cause of VR damper failure)
-
Damper actuators not working properly (misaligned, binding)
-
Excessive friction between limit switch and damper seal
-
Limit switches not set properly
c
-~
.
-
Chains severely corroded
-
Loose electrical termination
-
Damper blades sagging
-
Damper blades rubbing on inside surface of body
-
Return springs weak and severely corroded
-
Speed Regulator filters dirty
-
Hardened gummy and dry grease like substance on cylinder stem (stem was seized)
-
Rubber air line hose problems (worn, old, ruptured)
-
Pneumatic line blockage
-
Embrittlement of the diaphragm in the solenoid valve
-
Solenoid valve "o" rings (dry / brittle)
-
Stiffness on damper seals Dampers failed from many different causes, but the licensee's preventive maintenance program was very narrow in scope and only addressed two of the damper problems.
Although several vendor recommendations and proposals have been issued to the licensee to address the VR dampers design and maintenance problems, licensee action to implement these recommendations has been very slow.
A Techno Corporation letter dated September 29, 1989, stated that the damper blades were rubbing on the bottom inside surface of the body as they
=
approached the closed position. This friction force opposed the spring closing force and stopped the blades just short of complete closure.
To resolve this problem, the vendor recommended addition of hangers from the hinge area to the blade and the addition of quick exhaust valves into the air cylinder feed lines.
The licensee did not follow the vendor's recommendations.
In March 1990, the vendor had sent to the licensee hanger brackets to be installed on the dampers to eliminate sagging. The hangers were not installed.
The Tech Staff engineer felt that the benefit would be marginal and the hanger brackets were returned to the vendor.
A CECO followup letter from corporate design engineering dated September 10, 1991, states that the. licensee elected to shave the bottom of the damper blade to remove the blade to body contact and reduce friction force. The letter further stated that if the station continued to have problems with the dampers
i
.
.
-
.
'
-
J l
not closing completely, that the station should then modify the dampers as originally recommended by the. vendor.
In a Techno Corporation letter dated March 17, 1992, the vendor further recommended that the dampers actuators be replaced with larger one. to provide an actuator with an integral return spring, that will give more than twice the return force from the constant - force spring arrangement (500 lbs vs 180 lbs).
The increased return spring force would overcome the dampers friction
,
force.
The inspectors were informed that the review for this proposal is still being performed by site engineering.
As of January 3,1994, the VR dampers continue to have problems to fully close during normal plant operation on demand and during the monthly surveillances.
The inspectors were informed that prior to performing the monthly surveillance a team of licensee electrical and mechanical maintenance personnel would be ready expecting damper failure to properly operate (so that prompt action can be taken to correct the expected problem when it arises).
I In November 1992, damper support brackets were installed on IVR05YB on a trial
basis. However, this fix did not address the problem since the damper
'
subsequently failed to properly function in June, July and October 1993. The engineers failed to trend these failures and issued a work request to add the same style brackets to IVR04YA.
In September 1993, the licensee was unable to install the brackets because of installation problems and wrong components.
'
The inspectors determined that a 10 CFR 50, Appendix B, Criterion XVI violation had been issued in NRC Report 373/89012 for failure to promptly identify the cause of failure, and take corrective action to prevent l
recurrence, of the IVR05YA reactor building isolation damper failure to close
'
on April 5 and 12, 1989, (within the 10 second required by the Tech Specification and specified in surveillance LOS-CS-Ql).
The apparent cause of the failure was subsequently determined to be degraded actuator lubricant which was found to be gummy in consistency and contaminated with dirt. This caused increased resistance to the actuator's piston movement.
Following this failure, the licensee replaced the actuators every refueling outage
'
thereafter.
The licensee also committed to evaluate other options that would resolve the dampers failure to close.
In September 1992, 2VR05YB failed to fully close. During the licensee
'
inspection, an inch and a half (1 1/2") gap was observed.
The technical staff was requested to perform an evaluation to determine acceptability.
The evaluation was not available for review.
The inspectors noted that on October 27, 1993, damper IVR05YB failed to cycle (close) properly, apparently due to friction within the damper actuator.
An engineering memo dated December 14, 1993, stated that a visual inspection of the actuator cylinder by site engineering found hardened and a dry grease like substance on a portion of cylinder stem. The stem vent port contained a gummy grease like substance and the stem was found seized within the cylinder.
The oil inside the cylinder was observed 1o be " sticky" and a rusty surface was noted. The licensee concluded that the failure mode of IVR05YB to properly close appeared to be the binding or seizing of the actuator stem due to dirt,
,
.
,
.
.
contamination, or changes to the lubricating oil viscosity. The licensee contacted the vendor to assist in identifying the true failure mode.
The vendor requested that the actuator be sent to their shop for determination of root cause. As a result of IVR05YB failure on October 27, 1993, the main steam tunnel EQ trigger point temperature of 144 F was exceeded for the MSIV limit switches and solenoid valves for 71 minutes, affecting the EQ life for these components.
Further review of failure history of the VR dampers by the-inspectors revealed that the dampers have failed to properly close on first attempt on various occasions during the monthly surveillances. The inspectors determined that whenever a damper would fail to close, the licensee would cycle the damper several times until satisfactory operation was achieved as shown by damper.
open/close indication lights.
.
In a letter from the LaSalle Mechanical and Structural Design Group (M&S) to Mr. J. D. Williams, Engineering Supervisor, dated September 18, 1992, the design engineers presented management with a proposed' upgrade including design change options and maintenance improvements to eliminate the closing failures of the secondary containment isolation dampers. However, as of the end of this inspection, no action has been initiated to address this issue.
The inspectors were informed that the proposed upgrade package was sent to site engineering'who were " currently not pursuing" this issue.
On November 10, 1992, Technical Staff engineers performed a field inspection
,
of dampers IVR04YA and IVR05YB to observe the damper blades condition and operation (See LaSalle letter No. 118028). Various observations and deficiencies were noted, examples included:
-
Upper spring rubbing the retaining bracket due to misalignment causing the right blade to close slower than the left blade.
The licensee documented that this condition " decreases the blade momentum and may contribute to future closing problems." NED recommended that spring alignment be checked and corrected by maintenance at the earliest opportunity.
-
Corroded retaining brackets, hoses and conduit fittings.
-
Metal to metal contact (blades to housing) as the blades approach
'
the closed position. Previous trims to the blades were not contoured perfectly.
The letter stated that NED does not recommend any further shaving of the damper blades because it would likely degrade the dampers's ability to maintain secondary containment boundary.
The NED engineers recommended that support brackets be installed on IVR05YB to determine its effectiveness and reliability.
The brackets were subsequently installed, but the licensee failed to evaluate and trend its effectiveness and reliability.
Licensee short term corrective action to address the deficiencies noted above was not available for review.
i
, -
-
-
-
.
_
-
l
.
,
To track the proposed long term upgrades and corrective action for the
'
IVR04YA and IVR05YB dampers, the licensee assigned Nuclear Tracking System (NTS) numbers 925-251-92-E-00500, 00501 and 00502. However, when the inspectors requested to see the licensee's documented progress on the corrective actions, no data was available for review on these NTSs.
During discussions with the VR system engineer, the NRC was informed that a silicon spray was used on the damper components.
The inspectors determined that the spray was not approved by the vendor for use.
Also, the licensee has not been documenting the use of the spray.
The inspectors determined that the LaSalle Station Priority Technical Issues
Report, dated October 31, 1993, contained 106 items needing increased attention.
The VR damper issue was not one of the 106 issues. Also, the licensee's self assessment program failed to identify problems in this area.
The inspectors determined that engineering evaluations of recurring VR damper
'
problems to fully close and open was ineffective in preventing recurrence.
i There appears to be a lack of assessment capability to identify the root causes of VR repetitive problems in aggregate. The licensee failed to perform a systematic evaluation and trending of the multiple damper failures. The licensee corrective action plan did not appear to be effective nor timely to address these equipment operability problems.
In addition, management overview and engineering involvement in the resolution of multiple VR failures were not aggressive. Trends were not being adequately identified relative to~
-
the damper repeat problems.
j 10 CFR 50, Appendix B, Criterion XVI, Corrective Actions, states in part,
" Measures shall be established to assure that conditions adverse to quality, such as failures, malfunctions, deficiencies, deviations, defective material and equipment, and nonconformances are promptly identified and corrected.
In the case of significant conditions adverse to quality, the measures shall
'
assure that the cause of the conditions is determined and corrective action taken to preclude repetition." The licensee's failure to promptly identify the root causes of the failure of the VR dampers, and to take measures to prevent recurrence as discussed above is considered an apparent violation (373/93036-01(DRS); 374/93036-01(DRS)).
.
4.0 Review of Events Relatina To RPS Bus 1rios When A Larae Pumo Was Started On November 29, 1993, Unit 2 "B" RPS bus was de-energized due to an Electrical Power Monitoring (EPM) assembly sensing undervoltage condition caused by the start of the "A" Reactor Reci culation (RR) pump while the MG RPS bus was being powered from its alternate power supply. This resulted in a "B" half _
scram and several isolations.
During recovery, while attempting to reset the
,
half scram by transferring the switch from " Alternate B" to " normal" position,
- '
the operator apparently over traveled to the " Alternate A" position, which
resulted in a full scram (LER-93-009).
l The inspectors determined that two similar events occurred at LaSalle in 1983 and 1984 where the alternate RPS bus tripped during the start of a large pump.
These are documented in DVR-1-1-83-0453 and DVR-1-1-84-0365 (LER-84-076). To
r-m y
--
--
l.
.
.
correct this design problem, modification 1-1-82-279 dated July 1983 was issued to install and test a regulating transformer to preclude spurious trips of the alternate RPS feed. This modification did not correct the problem.
Work Request L54217, dated November 1985 was then initiated to increase th'e RPS EPH assembly overvoltage (0V), undervoltage (UV), and underfrequency (UF)
relay settings to a 3 second time delay before tripping.
WR L54217 further required that subsequent to resetting the 3 second time delay, a test be performed on the RPS feed to verify that the RPS alternate feed does not trip
'
when a large load is started (i.e., RR pump). The inspectors determined that the 3 second time delay settings were accomplished on Unit 1 per WR L54217 on June 3, 1986. However, the test requirements specified in the WR were subsequently cancelled by the operating engineer and not performed.
No further action was taken to test the 3 second time delay change and correct the design problem.
On November 29, 1993, the RPS bus tripped when a RR pump was started with the RPS bus fed from its alternate feed.
This confirmed that the problem had not been corrected by past design changes.
The licensee could
'
not explain why the test requirements were cancelled by the operating engineer, t
Failure to assure that all testing required to demonstrate that structures, systems, and components will perform satisfactorily in service is identified and performed in accordance with written test procedures which incorporate the requirements and acceptable limits contained in applicable design documents and test results are documented and evaluated to assure that test requirements a
have been satisfied is an apparent violation of 10 CFR 50, Appendix B,.
Criterion XI (373/93036-02(DRS); 374/93036-02(DRS)).
'
During review of the circumstances related to this event; examination of engineering design documents, tests, maintenance records; and interviews with System, Site and Operational Analysis Department (OAD) engineers, supervisors and operations staff the inspectors identified the following concerns associated with the 3 second time delay settings on the EPH assemblies.
'
(a)
0AD EPM assembly calibratien surveillance procedures LES-GM-300 and 400 were inadequate in that they did not require monitoring or recording the 3.0 second relay time delay settings for the EPM assemblies. The Relay Setting Order (RS0) sheet specified a 3.0 second time delay setting.
0AD engineers and the supervisors were aware of this difference between the RSO and the procedure, but failed to initiate corrective action for an extended period of time.
There appeared to be a need to review OAD activities to assure familiarity with safety requirements.
The failure to require monitoring and recording as found and as left time delay settings is considered an example of an apparent violation of 10 CFR 50, Appendix B, Criterion V (373/93036-03A(DRS); 374/93036-03A(DRS)).
(b)
The inspectors determined that as of December 21, 1993, Operating i
'
Procedures LOP-RR-04, " Preparation and startup of Reactor Recirculation Pumps in Slow Speed", dated June 10, 1993; LOP-RR-05, " Changing Reactor Recirc Pump Speed from Slow to Fast Speed",
,
l
-
---
,
.
V dated May 28, 1993, and LOP-RR-06, " Restart of Tripped Reactor Recirc", dated March 4, 1992, failed to have a warning to the operators that whenever the RPS bus was fed from its-alternate (dirty) Power supply'and a large pump such as a Reactor Recirculation pump was started, the RPS bus would trip on undervoltage and result in a half scram. This scenario occurred at LaSalle in 1983, 1984 and again on November 29, 1993.
In 1984 and 1993 additional problems resulted in a full scram.
The licensee failed to take interim measures while this problem continued.
The failure to caution the operators of the' potential trip is considered an example of an apparent violation of 10 CFR 50, Appendix B, Criterion V (373/93036-03B(DRS); 374/93036-03B(DRS)).
(c)
On November 29, 1993, control room operators failed to adhere to the requirements delineated in procedure LAP-220-5, " Equipment Operability Determination", dated July 8, 1993, when the secondary containment isolation damper IVR04Y,A failed to fully close on demand. The procedure required that the operator determine if a previous evaluation existed in the Equipment Operability Binder (s), check if the failure mode is the same and judge whether the evaluation applies. The inspectors concluded that a more conservative approach was warranted to determine if a failed condition existed.
In addition, procedure LAP-220-4, " Degraded Equipment Log", dated May 11, 1993, was inadequate in its requirements to document / record degraded components and was weak in defining performance standards and as such was' considered
.
,
inadequate.
For this event neither a Degraded Equipment Log entry nor a Work Request was written. Additionally, a specific PIF was
not issued for this failure. The inspectors. considered failure to follow procedures and inadequate procedures to be an example of an apparent violation of 10 CFR 50, Appendix B, Criterion V (373/93036-03C(DRS); 374/93036-03C(DRS)).
(d)
The instructions provided by corporate engineering in the RSO sheets used by 0AD to set the RPS time delay trip settings were inadequate in that they specified 3.0 seconds with no tolerance
'
allowed. A value of exactly three seconds was not practically-obtainable. Consequently, the as found trip settings were observed to be as high as 4.69 seconds and most were above 3.0 seconds. This occurred because the technician applied his own tolerances and used a wrist watch.to produce the trip settings.
The RSO sheet was inadequate in that it failed to specify a tolerance and is considered to be an. example of an apparent violation of 10 CFR 50, Appendix B, Criterion V (373/93036-03D(DRS); 374/93036-03D(DRS)).
(e)
The OAD engineer had been using a wrist watch to record the time delay. When asked why he did not use a calibrated and certified stop watch, he stated that he did not know that one existed..This practice was also known to 0AD supervisors and had been going on
..
_
_
.
,
t i
w
'
L since 1986..The inspectors determined that a calibrated stop watch was available from the shift supervisor.
'
Failure to assure that measuring and testing devices used in activities affecting quality are properly controlled is considered an apparent violation of'10 CFR 50, Appendix B, Criterion XII (373/93036-04(DRS); 374/93036-04(DRS)).
(f)
Generic General Electric (GE) spec 21A3120, Revision.2, stated in
>
paragraph 4.1.4-that the EPM trip components, over-voltage, under-voltage, and under-frequency, shall activate after a time delay 'of between 0.1 and 3.0 seconds after any trip conditions.
In addition, G.E. Spec 22A3771AJ, Revision 2, stated in paragraph
6.1.4, all time delays, if any, associated with the OV, UV and/or UF trip logic circuitry should be adjusted for minimum time delay of from 0.1 to 0.3 seconds..., in no case should the time delay exceed four (4) seconds.
When the inspectors requested that the licensee verify 'the 3 second time delay setting on the Unit 2 alternate EPH assembly, the OAD engineer performed the test using (1) a wrist watch and i
(2) a certified, calibrated stop watch obtained from the shift supervisor. The data obtained in both tests was as high as 4.69 seconds for EPM assembly IC71-S003E.
FSAR Question 031.278 specified the trip settings nominal range
(operating) for the time delay adjustment for the EPM assembly OV, UV, and UF components as 0.1 to 3.0 seconds.
The inspectors determined that the licensee has been setting-the time delay above 3.0 seconds and as high as 4.69 seconds. This item is a deviation from FSAR requirements (373/93036-05(DRS); 374/93036-05(DRS)).
t (g)
The inspectors determined that the OAD test engineer and supervisor had not contacted the OAD engineers performing the same time delay testing activities at Quad Cities and Dresden to obtain the best test method and equipment used a: c.ortipered to the ones used at LaSalle.
Subsequently, the OAD engineer found that Quad
,
Cities has been using a state of the art measuring test equipment which could have been used at LaSalle in the interim.
(LaSalle has this same equipment on order.)
(h)
The inspectors requested that the licensee review a sample of Relay Setting Orders (RS0) data sheets that were sent to 0AD from corporate engineering to perform field relay i
setting changes. The purpose of the review was to determine if additional cases existed where the work specified in the RSO sheets had not been properly implemented ~ in the field.
Subsequently, the licensee informed the inspectors that in May 1992, new RMS-9 (solid state) trip units were installed on the main feed breakers for 480V switchgear 136X,.136Y, 235X and 236Y.
However, the' settings specified in RSO data sheets for the transformer ground,overcurrent-relays (IAC-
.
i
,-
.
-
..
-
.
.
__
_
e
'
.
.
60) and 4.16kV side phase overcurrent relays (C0-4) were inadvertently not reset to coordinate with the newly installed RMS-9 trip units. The present settings result in an inadequate coordination time ~ interval between the main 4.16kV feed breakers and the transformer ground overcurrent relay.
On a ground fault on the 480V bus, the ground overcurrent relay would operate and trip the upstream 4.16kV breaker and power to both 480V buses on one division would be lost.
This was not per design intent.
On December 29, 1993, the licensee performed an operability
-
evaluation and concluded that on a bus fault only one division will be lost. Although this is true, the inspectors were concerned that the same design deficiency exists on division 2 of Unit 1 and on both divisions of the other unit.
The licensee committed in ENC-QE-40.1, exhibit D, dated December'29, 1993, to adjust the IAC-60 relay settings and the C0-4 overcurrent relay settings by April 1, 1994.
This item is considered unresolved pending licensee action and NRC review (373/93036-06(DRS); 374/93036-06(DRS)).
(i)
The EPM assembly's function is to monitor the output voltage and frequency of the RPS MG sets. When an overvoltage, undervoltage or underfrequency condition is detected, the output circuit breaker of the EPM assembly will trip thus interrupting power to the RPS system. This ensures that no damage occurs to the RPS loads.
The trip setpoints for overvoltage, undervoltage and
underfrequency are specified in Technical Specification 4.8.3.4:
however, the time delay trip values are not specified in
'
Technical Specifications.
.
Th'e concern raised by the NRC was whether RPS equipment operatinij with voltage and frequency values outside of the acceptable range
for more than the 4 seconds maximum value specified by GE could be damaged, thus affect,ag their ability to perform their RPS safety
. function.
The licensee stated-that based on GE documentation provided to Susquehanna Station, their RPS loads can tolerate EPH assembly trip delay times of up to 7.2 seconds and due to the similarity of GE BWR RPS equipment, it was LaSalle's engineering-judgement that EPM assembly trip delay time of up to 6 seconds are justified.
Licensee proposed corrective action was to' test the EPM trip delay times in a timely manner and realign all~ trip delay settings to 3.0 seconds or less.
In December 1993, the licensee tested and adjusted the time delays on the Unit -2 and alternate Unit 1 EPM assembly units. The licensee informed the NRC that they intend to test and set the normal unit 1 EPM assemblies when the unit comes down.
In the interim, the licensee at the request of the NRC,
,
-., - ~
.
_
.
.
.
contacted GE'to evaluate and confirm that LaSalle engineering evaluations and test.results using 6 second time delays were acceptable for the LaSalle design.
General Electric, in a letter to LaSalle dated December 22, 1993, concluded that an RPS EPM assembly time delay of up to 6 seconds,
'
for undervoltage, overvoltage, and underfrequency conditions, will
_
have negligible affect on the performance.of the LaSalle
.
components listed in Reference 3 of the letter.
However, GE took.
exception to Item E in the letter.
For Item E, which relates to
'
an HMA relay that provides the 135# reactor pressure _ interlock.for RHR shutdown cooling, LaSalle will put in place compensatory-operator actions to ensure that any such effects do not affect the
-
RHR system operation. The licensee performed an engineering evaluation of the worst case effects of a RPS bus power
disturbance on the RHR SDC pressure interlock and concluded that the potential effects on the interlock relay are not significant.
The compensatory measures will be in effect until GE evaluates and accepts the test data.
This item is considered open pending GE.
results and NRC review (373/93036-07(DRS); 374/93036-07(DRS)).
(j)
Review of engineering audits showed that licensee self assessment and problem identification needs increased management attention.
At the conclusion of the inspection, the licensee informed the' inspectors that a res,ew of the design'of the alternate RPS MG set feeds would be performed to determine how EPM assembly tripping on large load.startup can be prevented.
Also, the inspectors requested that the licensee consider a design change to eliminate the potential loss-of RPS bus power from over travel of the RPS transfer switch which typically results in a full scram.
The licensee has committed to review this issue.
,
5.0 Review of. Electrical and I & C Modification Packaaes-The inspectors evaluated the licensee's performance and programs relating to design changes, temporary systems changes and modifications.
The areas examined included design inputs, review and approval process, implementation, post modification testing, 10CFR50.59, training and satisfactory completion of design requirements.
The following modifications were reviewed:
(1)
Modification P01-2-93-503
,
-1 (2)
Modification M-1-2-88-052 (3)
Exempt Change Requests E01-93-902A, 902B, 902C (4)
Exempt Change Request EC-92-916A (5)
Exempt Change Request EC-93-954 (6)
Exempt Change Request EC-93-976A, 976B, 976C, 977 and 978
l
,
_.
. _.
.
.
.
..
"
.
t
'
(7)
Temporary System Change No. 1-1374-91 (8)
Temporary System Change No. 1-645-91 (9)
Temporary System Change No.- 1-203-91 l
.
(10)
Temporary System Change No. 1-489-93 The inspectors concluded that the licensee was adequately controll'ing the design control program for these modifications.
However, the licensee was not adequately implementing the design control process for some electrical modifications.
During recent field implementation of two electrical modifications, several significant wiring errors were made.
The errors
,
resulted in the application of 277Vac on a 120Vac. relay and inadvertent-start of a pump when a breaker was closed. This problem was investigated under the
..
- 373-200-93-0065PIF Problem Identification Form and exhibited good
{
investigative followup review.
The inspectors determined that NOD Directive N0D-TS.5, issued in 1993, provided improvements in the form of focused responsibilities, streamlined
checklists, simplified project plan format and engineering involvement from initial planning to completion.
.
In addition, the inspectors noted that the number of Temporary Design Changes.
was high with about 100 in place (which was above the licensee's goal 'of 60).
Some of the temporary changes were instituted as_long ago as 1986.
-
6.0 Review of 10CFR50.59 Safety Evaluations During the review of the modification packages the inspectors noted weakness in the 10CFR50.59 safety evaluation process.
(1)
WR #L22715, dated May 17, 1993, requested that bracket support be installed on damper IVR04YA to relieve the damper blade sag.
The safety evaluation screening performed on LAP-1200-13, attachment B, dated November 20, 1992 stated for question F.5.a "will change alter the design or function of any system, structure or component as described in the SAR." The licensee's answer was "no" that the
" dampers function will not be altered and its design is not described in the SAR."
The inspectors determined that although the design details of the damper were not described in the FSAR,
,
the closing function of the VR system dampers was.
(2)
In a response to question 031.278 as stated in amendment 61 of the
-
FSAR, dated December 1982, the licensee specified 0.1 - 3.0 second time delay adjustment for.RPS EPM assembly rip units for overvoltage, undervoltage and under frequency. Also, the Relay Setting Order (RS0) sheets, dated November 4, 1985, specified 3.0 second time delay settings for the RPS EPH trip units; however,
,
the OAD engineer has been adjusting the time delays to as high as 4.69 seconds. As of this inspection, GE allowed up to 4.0 second
>
.
,
- -. _,.
-, _ -
_.
.
-
-
-
.
..
.
- -
,
d time delay on the EPM assemblies.
Consequently, the licensee had been inadvertently operating with a 4.0 second time delay. A deviation from an FSAR commitment is discussed in section 4.0.(f)
of this report.
7.0 Review of Desian Control and Enaineerina Audits The inspectors concluded that some of the engineering related audits reviewed did not appear comprehensive in scope and depth.
However, the Technical Services audit performed in 1992 was a comprehensive audit and identified several concerns in the engineering area. The inspectors noted that the related Corrective Action Reports (CAR) remained open.
The current Technical
,
Services audit has been delayed until March / April 1994 to accommodate other organizations.
The inspectors concluded that management was not adequately supporting the audit effort in order to resolve these CARS.
'
The inspectors deterained that a recent Audit (QAA-01-93-06) evaluated QC independence by re'aewing inprogress work in several areas such as welding, concrete anchor installations, etc., but did not review the electrical area where recent problems had occurred.
8.0 Enaineerina and Technical Support The inspectors evaluated LaSalle's electrical engineering and technical
'
support capability and involvement in the modification process.
Site and system engineers were interviewed and related activities were examined.
Based on the review the following concerns were identified:
System engineers did not have a clear understanding of their
duties and responsibilities.
System Engineers were usually not involved in the design process
for modifications or in the determination of post modification testing.
No formal program existed for crosstraining or designating a
backup engineer for each system.
I System engineers had insufficient interface with 0AD engineers.
>
Improved communication would have identified the EPM assembly time delay issue prior to the NRC discovery.
Operations did not get the system engineer involved in the
>
problems identified during the recent IVR04YA damper failure to close on demand.
System engineers did not use an established method to prioritize
work.
.-
.
.
.
.
. -..-
.
...
.
..
System engineers did not receive specific training on their
systems' components (such as vendor ~ training, etc.)
Continuing.
training has been weak for system engineers. The licensee has
,
designated a dedicated engineering training coordinator to address this issue.
Different Engineers have received several different root cause
analysis courses.
There appeared to be a lack of consistent
~i approach to root cause analysis among the engineers.
There was a relatively high turnover of system engineers because
of factors such as lengthened time for first promotion for new
'
system engineers and perceived lack of career paths.
-
Engineers showed a reluctance to write PIFs, particularly if the
resolution would be assigned back at them. Also, training on the PIF process to plant staff needed to be improved. A previous NRC report identified inadequate PIF implementation.
Equipment aging and hence failure was identified as being above
'
the industry average rate in the latest Component Failure Analysis Report (CFAR) and.needed closer management attention.
l
.
The licensee indicated that they were working to resolve these problems.
9.0 Unresolved items
'
Unresolved items are matters about which more information is required in order to ascertain whether they are acceptable items, violations, or deviations.
Unresolved items disclosed during this inspection are included in Section 4.0.
-
10.0 Open Items Open items are matters which have been discussed with the licensee, which will be reviewed further by the team, and which involve some action on the part of the NRC or licensee or both. Open items disclosed during this inspection are included in Section 4.0.
11.0 Exit Interview
The team conducted an exit meeting on December 21, 1993, at the LaSalle site
.
to discuss the major areas reviewed during the inspection, the weaknesses
'
observed, and the inspection findings.
NRC personnel-and licensee representatives who attended this meeting are documented'in Section 1.0 of this report.
The licensee did not identify any documents or processes as proprietary.
"
,
16