ML20058A327
| ML20058A327 | |
| Person / Time | |
|---|---|
| Site: | Quad Cities  |
| Issue date: | 11/10/1993 |
| From: | Falevits Z, Hausman G, Mendez R, Salehi K, Shafer W, Tella T NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION III) |
| To: | |
| Shared Package | |
| ML20058A316 | List: |
| References | |
| 50-254-93-24, 50-265-93-24, NUDOCS 9312010046 | |
| Download: ML20058A327 (16) | |
See also: IR 05000254/1993024
Text
.
'
.
U. S. NUCLEAR REGULATORY COMMISSION
REGION III
-
Reports No. 50-254/93024(DRS); No. 50-265/93024(DRS)
Docket Nos. 50-254; 50-265
License No. DPR-29; No. DPR-30
,
Licensee:
Commonwealth Edison Company
Executive Towers West III
1400 Opus Place, Suite 300
Downers Grove, IL 60515
Facility Name: Quad Cities Nuclear Power Station, Units 1 and 2
Inspection At: Quad Cities Site, Cordova, IL 61241
Inspection Conducted:
July 28-30, August 2-4, and August 9-20, 1993
Inspectors:
~
9MA ifb.
H/f/f 3
.
(IEaK llad Inspector
D4e/
li f fo f9 3
w
Z. Falevits
Dat'e l
'
f %& ,
/$/0f93
I
R. Mendez
8
Date
II off 3
f
~
'
~
,
K. Salehi
Date
%f'aMdv 26
IIhol93
' ~
T. Tellh
Date
l
/[!/
5
kt} hR 0A
b
Approved By:
W.D.Shafer,-Chje"f
Date
j
Maintenance and Outages Section
4
Inspection Summary
._
Inspection on July 28-30. Auoust 2-4. and August 9-20. 1993 (Report Nos.
50-254/265-93024(DRS)):
Areas Inspected: Special, announced team inspection of licensee activities
associated with an unusual event declared on July 27, 1993, due to a loss of
offsite power (LOOP).
In addition, selected portions of NRC inspection
module 62700 were used to ascertain whether electrical maintenance activities
9312010046 931112
ADOCK 05000254
G
,
. , .
.-
. - .
._
.
.
.
DETAILS
,
1.0
Principal Persons Contacted
Commonwealth Edison Company
- R. Pleniewicz, Site Vice President
- D. Bax, Station Manager
- J. Burkhead, Quality Verification Superintendent
- D. Craddick, Maintenance Superintendent
- R. Dralle, Electrical Maintenance
- H. Hentschel, Operations Manager
D. Kanakares, Regulatory Assurance NRC Coordinator
- J. Leider, Technical Service Superintendent
- A. Hisak, Regulatory Assurance Supervisor
,
- M. Pacilo, Master Electrical Maintenance
,
U. S. Nuclear Reaulatory Commission
- R. Gardner, Chief, Plant Systems Section
- P. Hiland, Chief, Reactor Projects Section IB
- T. Taylor, Senior Resident Inspector
- Denotes those present at the exit meeting on August 20, 1993.
Other persons were contacted as a matter of course during the inspection.
2.0
Licensee's Actions Reaardina Previously Identified NRC Findinas
(Closed) Unresolved Item (254/88027-01(DRS):265/88028-01(DRS)):
Neutron flux
.
monitoring instrumentation did not meet Regulatory Guide (RG) 1.97, Category 1
)
requirements.
The Office of Nuclear Reactor Regulation (NRR) completed an
_
evaluation of the boiling water reactor (BWR) owners group report, NED0-31558,
" Position on NRC Regulatory Guide 1.97, Revision 3, Requirements for
Post-Accident Neutron Monitoring System," and concluded that for current BWR
r
,
license holders the NED0-31558 criteria was an acceptable alternative to the
recommendations of RG 1.97.
NRR requested the licensee to review their
"
neutron flux monitoring instrumentation against the NED0 criteria and submit
i
the results of that review to NRR. On August 17, 1993, the licensee provided
somelof the requested information; however, the licensee stated that further
-
review would be required to assess the actions necessary to comply with those
NEDO recommendations not ' currently addressed by ' system design. The licensee
stated that this information would be submitted to NRR in 90 days
l
(November 15, 1993). No further Region III action is required.
Therefore,_._
this item is closed.
i
-
t
1
1
l
l
, _ _ _ .
.
. .-
. - - .
.-
.-
.
.
.
-
,
Inspection Summary
2
[
on high, medium and low voltage breakers were effectively accomplished and
assessed by the licensee.
Results: Three violations, some with multiple examples, and one' inspection
followup item were identified.
The inspectors concluded that the major
component failures for the LOOP event were the loss of transmission line 0403
,
"
(Section 3.1.1), failure of the nonsafety-related bus 22 main feed breaker to
close on demand (Section 3.1.2) and failure of oil circuit breakers. (0CBs)
9-10 and 10-11 (Section 3.1.3).
In addition, several built-in design features were not correctly modeled-on
the Quad Cities simulator, such as the reserve auxiliary transformer 22
(RAT) to unit auxiliary transformer 21 (UAT) undervoltage time delay (slow
transfer) as discussed in Sections 3.1.4 and 3.1.5.
Failure by plant.
'
personnel to recognize a RAT to UAT slow transfer existed per plant design was
,
seen as a weakness in personnel training.
Electrical maintenance activities
on the high, medium, and low voltage breakers were adequate with no
significant safety issues identified; however, a weakness was identified with
substation construction documentation related to the identification of
preventive, predictive and corrective electrical maintenance activities on
OCBs and associated 34SkV switchyard equipment. Weaknesses were also noted.
j
with evaluation of GE Service Information Letter 448 and with some system
engineers, as discussed in Sections 3.2.6 and 3.3, respectively.
.
_ _ _ - . _ _ _ _
4
.
.
3.0
Introduction
The purpose of the inspection was to review licensee activities associated
with an unusual event declared .on July 27, 1993, due to a loss of offsite
power (LOOP).
Both units were affected by the LOOP, declared an unusual
event, and entered the appropriate limiting conditions for operation (LCOs).
Attachment A contains a chronology of activities that occurred, prior to and
during the July 26-27 LOOP event.
The inspection also assessed / evaluated the quality and effectiveness of
electrical maintenance activities associated with high, medium and low voltage
breakers.
The inspectors conducted personnel interviews, performed walkdowns,
reviewed past operating experience, evaluated engineering and technical
support, reviewed licensee assessments of electrical breaker maintenance, and
reviewed related corrective, preventive, and predictive maintenance
activities.
3.1
Loss of Offsite Power Event
On July 27, 1993, a fault occurred on the offsite transmission line 0403.
This caused an auto trip of oil circuit breakers (OCBs) 7-8 and 8-9.
Operations personnel had previously tripped open OCBs 9-10 and 10-11 due to
smoke and boiling noises observed coming from these OCBs.
The combination of
the tripped OCBs resulted in a LOOP to the Unit 2 reserve auxiliary
transformer (RAT).
By design, bus 22 should have transferred from the RAT to
the unit auxiliary transformer (UAT); however, the nonsafety-related bus 22
main feed breaker did not close on demand due to a faulty position switch.
3.1.1
Loss of Transmission Line 0403
Loss of transmission line 0403 from the Nelson transmission substation (TSS)
resulted in the LOOP and de-energization of the Unit 2 PAT.
Prior to the LOOP
event, a fault was sensed on line 0403 near the TSS; however, a type G
reclosing relay failed to reclose the breakers after the fault was cleared.
Consequently, the Quad Cities relaying scheme sensed a loss of voltage and
tripped OCBs 7-8 and 8-9, which resulted in the de-energization of the RAT.
The licensee stated that there were approximately 600 type G relays throughout
the Commonwealth Edison Company system and that only 25 failures had been
experienced in the last 5 years.
3.1.2
Loss of B0P Bus 22
The loss of Bus 22 (feed to 28 and 2C reactor feed pumps and 2B recirculation-
pump) was directly attributable to a faulty position switch which prevented
closure of the main feed breaker. When the loss of voltage condition.
occurred, the bus should have transferred from the RAT to the UAT (after a 1.2 ;
second. time delay). Adequacy of maintenance of nonsafety-related breakers,
including the main feed breaker to bus 22, is discussed in Section 3.2.3.
2
.
_
-
q
.
.
3.1.3
Oil Circuit Breaker 9-10 and 10-11 Failures
In preparing to synchronize the Unit 2 generator to the grid, an equipment
operator (EO) performing a routine surveillance in the 345kV switchyard
noticed smoke and heard boiling noise coming from OCB 9-10, B phase. When-
OCB 9-10 was tripped open, the boiling noise stopped.
To synchronize the
Unit 2 generator to the grid with OCB 9-10 open, OCB 10-11 was tripped open.
This caused a partial loss of the station ring bus.
After Unit 2 was synchronized to the grid through OCB l-11, the control room
operators were unable to close OCB 10-11. Operations personnel called the
,
Operational Analysis Depal tment (OAD) to assist in closing OCB 10-11. 0AD
'
notified the operators that the HACR synchronization relay would not allow an
j
OCB to close on a dead bus. 0AD placed a jumper across the HACR relay
allowing closure of OCB 10-11. Operators later tripped OCB 10-11 open when a
j
substation construction (SSC) worker, performing corrective maintenance (CM)
i
on OCB 9-10, heard boiling noise coming from OCB 10-11, C phase.
The licensee disassembled and inspected all three phases of each OCB.
Internal arcing occurred in the B and C phase interrupters of OCBs 9-10 and
10-11, respectively.
In each instance, one of the four movable silver tipped
l
contacts had disintegrated. Damage was also visible to the interrupter
stationary assembly, where a portion of the stationary. contacts melted away.
,
for 0CB 9-10, the licensee discovered that the cotter pins used to secure the
I
movable contact assembly to the ladder assembly had fallen out on one side.
The two cotter pins had either been bent the same way or not bent at all
during installation. The absent cotter pins eventually caused an improper
movable contact to stationary contact alignment. This produced a high
resistance path, which- resulted in the internal arcing noticed on July 26,.
'
1993. The licensee concluded that the cotter pins had been incorrectly
installed during assembly (about 1969) at the factory.
For OCB 10-11, the licensee discovered two loose bolts which held the
r ossover arm in place.
The licensee stated that improper movable contact to
stationary contact alignment was again the cause.
One bolt was found to be
about a half flat loose and the other was loose by about one flat of the nut.
The loose bolts prevented proper penetration of the movable contacts into the
stationary contacts, which resulted in a high resistance path.
3.1.4
As a result of the LOOP event, the licensee determined that fast transfers
from the Unit 2 RAT to UAT were not possible due to an VV time delay (TD)
I
relay installed on the low side (4kV) of the RAT with a 1.2 second time-delay.
Under nominal conditions a RAT to VAT transfer (fast transfer) would occur in_
,
1
about six cycles (= 0.1 seconds). The TD was installed to prevent a fast
transfer from the RAT to the UAT during voltage fluctuations on the system
grid.
The transfer delay caused a very low voltage condition on 4kV safety-related
,
bus 23 and the associated feeds to 480V buses 25 and 28 resulting in the drop
3
1
'l
. - - .
-,
- -
- .
.-.
-.
.
. , ,
-
.-
.
._
-
4
.
s
out of some motor contactors and relays. The. following is a partial list of
equipment tripped from the normal power source because of the delay.
The 2B feedwater regulating valve locked up due to loss of the power to
O
hydraulic pump (normal feed bus 25).
- 0
The SPING terminal in the control room, which provided control room
indication of the stack discharge concentrations, was lost (fed from bus
,
28).
J
o
The essential service system uninterruptible power supply (UPS) switched
to its alternate power source (UPS fed from bus 28).
O
The operators received area radiation monitor (ARM) downscale alarms and
'
were unclear as to how to clear the alarms (ARMS fed from bus 28).
The reactor building sump pumps tripped (sumps fed from bus 28).
o
o
The drywell cooler tripped (fed from bus 28).
The emergency core cooling system fill pump (jockey pump) tripped (fed
O
from bus 28).
O
There was a momentary loss of reactor protection system (RPS) "A" (fed
>
from bus 28).
'
t
The diesel cooling water pump started due to loss of voltage on the
o
4kV bus.
In addition the diesel run light was lit due to pickup of the
associated control relays, although the diesel generator did not start.
>
Both of these events occurred per design but were not expected
'
occurrences by plant personnel.
The licensee stated that a similar TD relay was previously installed at the
Dresden station but was removed in 1973 at the recommendation of Sargent and
i
Lundy (S&L). A similar recommendation was made by S&L to remove the TD relay
at Quad Cities; however, the licensee could not explain why the recommendation
'
,
was not implemented. The licensee stated that this issue would be reviewed
further.
Therefore, this issue is considered- an inspection followup item
'
(254/265-93024-01(DRS)).
3.1.5
Simulator Modelina/ Operator Traininq
!
As a result of the LOOP event, the licensed operators became aware of certain
!
design features that were not covered in operator training and were not
i
modeled on the Quad Cities simulator. The following is a list of the design
features not modeled on the simulator.
_
o
Undervoltaae Time Delay Relav
The FSAR and the operations training manual stated that the RAT to VAT
transfer would occur with no inherent time delay.
Furthermore, the
4
i
,
,,
e
- . .
,
-
.
.
I
\\
l
.
.
normal.and abnormal operating procedures did not address loss of
equipment due to a time delay. The response observed during the event
was not expected.
O
HACR Relays
The operators were not aware that the HACR relays did- not allow the OCBs
I
to be closed in on a dead bus due to improper simulator modeling and a
lack of training.
These relays were installed as part of modifications
in 1990 and 1991 for Units 1 and 2, respectively.
In a memo dated
'
May 23, 1990, the cognizant engineer stated that no operator training
would be required as part of the HACR relay modification.
The simulator
allowed closure of the relays on a dead bus.
,
o
ARM Downscale Alarms
During the momentary loss of power, control room operators did not know
how to reset the alarms. The control room annunciator window procedure
did not address resetting the ARMS.
In addition, on a momentary loss of-
power, the simulator did not duplicate the ARM alarms.
O
Rod Permissive Liaht
.
1
The operators were not aware that a rod permissive light could be
received at less than 30% power even though a rod was not selected.
The
simulator did not model this event.
o
EDG Run Indicator Liaht and Coolina Pump Start
>
During the recent transfer of power from the RAT to the UAT, the
licensee noticed that the EDG run indicator light was illuminated but
,
that the EDG had not started.
In addition, the EDG cooling water pump
started. Pr. events occurred per plar,t design; however, the events
-
were unexpated since operators were trained not to expect such an
,
occurrence and the simulator was not modeled to duplicate the event.
i
The licensee found that due to the EDS control circuit-design, a
momentary loss of voltage on the 4kV buses can result in the
'
illumination of the EDG run light and the start of the cooling water
pumps. The licensee indicated that time delays may be placed in the
circuits to prevent _ inadvertent starting of components. A similar event
,
occurred in 1989 when the UAT transferred to.the_ RAT. Although the
licensee performed'an analysis of the event, no action was taken to
'
notify operations personnel.
3.2
Electrical Breaker Maintenance
'
The inspectors performed system walkdowns, reviewed corrective, preventive, ~~
I
~
and predictive maintenance activities, evaluated engineering and technical
4
support and. licensee ~ assessments of maintenance activities, reviewed' licensee
. i
maintenance improvement programs, and.rev_iewed past operating experience-
associated with high, medium and low voltage breakers.
Selection of licensee
j
documents were based on component / equipment safety significance.
5
l
. .
--
.
.
- .
.
_
.
.
_ . .
..
-
.
,
.
.
~
.
3.2.1
System Walkdowns
The high, medium, and low voltage breakers, as well as adjacent areas, were
observed for proper identification, accessibility, installed scaffolding,
radiological' controls, housekeeping and unusual conditions.
Unusual
conditions included, but were not limited to, water, oil or other liquids on
.
the floor or equipment; equipment in need of repair or out-of-service;
indications of leakage through the ceiling, walls, or floors; loose
insulation; corrosion; excessive noise or vibration; and abnormal
'
temperatures, ventilation or lighting. The inspectors verified that work
requests had been initiated for broken or defective equipment.
!
The material condition of the electrical areas, such as, 4kV and 480 Volt
switchgear, station batteries, battery chargers, main power transformers, unit
auxiliary and reserve auxiliary transformers was adequate.
system areas were generally clean. The out-of-service equipment was tagged.
Six cubicles in the IB motor control center (250 Volt DC) were tagged to
'
indicate that parts from these cubicles were taken out and used in other
cubicles.
Licensee personnel mentioned that the original equipment
manufacturer no longer supplied these parts and that alternate sources were
being investigated. The system engineers were knowledgeable of the areas
!
inspected.
3.2.2
SSC OCB Maintenance
l
l
The inspectors concluded that SSC Department CM and PM record documentation
I
was tiot adequate to accurately assess the licensee's OCB maintenance program.
i
Information documenting the scheduled dates for maintenance, the date the
maintenance activity was accomplished and the PM due dates were available;
l
however, specific details of the maintenance activities were absent. The
i
following concerns were noted by the inspectors:
i
o
CM and PM work history documentation was minimal to non-existent.
This
would make trending of important parameters, such as dimensional
tolerances, difficult.
o
Although the licensee stated that the latest revision of the vendor
manual was being used, there was no record of which revision was used
during maintenance.
1
o
The interface between the SSC crew and the 345kV switchyard system
,
engineer was poor. Whenever work was performed in the switchyard the
!
system engineer was normally not notified.
,
Additionally, the inspectors noted problems with the checklist currently being
,
used.
_
o
Development of the checklist did not require engineering review or
j
approval by an OCB specialist.
l
j
6
i
-
-
,
,-.
em
_
_
_
.
.
'
.
.
O
There was no information as to the revision of the vendor manual used to
develop the checklist.
O
There was no method for incorporating future recommendations into the
checkl i st.
o
There were no references in the checklist to a specific section in the
vendor manual.
. ;
.
O
There were no torque values or torquing of bolts included in the
checklist. This was significant since the licensee's investigative team
concluded that the probable cause of the 10-11 OCB failure was loose
- i
bolts on the crossover arm. Furthermore, the SSC crew failed to
document the discovery of loose bolts in OCB 10-11 on the checklist or
in the comment section.
O
During recent maintenance on OCB 9-10, work was started without the use
of the checklist.
As a result, some of the steps requiring as-found and
as-left values were left blank.
O
A number of dimensional measurements made during maintenance were found
l
out-of-tolerance without an immediate disposition.
The licensee's OCB
specialist later reviewed the out-of-tolerance values and found that
>
none would affect the operation of the OCBs.
The inspectors concluded that the licensee's 0CB PM program needed significant
,
attention and review by Quad Cities management. The licensee planned to
,
formalize history record keeping and develop a trending program for switchyard
OCB breaker problems.
, 3
3.2.3
4kV Breaker Maintenance
. 1
The inspectors reviewed the recent maintenance history of the main feed
!
breaker to bus 22 and the other nonsafety-related breakers on buses 11, 12, 21
and 22.
The inspectors found that maintenance performed during the last
,
Unit 1 outage on buses 11 and 12 was satisfactorily accomplished and
i
post-maintenance tests (PMTs) were specified. The inspectors noted, however,
that in several instances after maintenance, PMTs were not performed for the
vertical lift breakers on bus 22. Failure to specify post-maintenance testing
!
on breakers important to safety, as identified in the following nuclear work
requests (NWRs), is considered a violation of 10 CFR 50, Appendix B, Criterion
'
XI (254/265-93024-02A(DRS)).
o
NWR Q90581 was issued on February 24, 1992, to clean, inspect and repair
all cubicles on bus 22.
The licensee found that the position switch in
'
the cubicle to the bus 22 main feed breaker was bent or would not
operate. The position switch was repaired; however, no PMT was
,
7
'
specified for any of the breakers on .the bus.
j
t
-
,
,
7
!
.
i
. _ _
_
_
,
.
-
w
-
r-
-
f
. - - - _ - - -
-
,
-
.
,
O
NWR Q99959 was issued on April 26, 1993, to perform the remaining
overhaul work on bus 22. Step 22 of the NWR required that the position
switch be adjusted, repaired or replaced. Again, no PMT was specified
for a single breaker on bus 22, although required by QAP 1500-17, " Post-
Maintenance Testing / Verification Procedure." The licensee had failed to
specify the required tests on PMT verification matrix form QAP 1500-S38.
The inspectors found that the shift engineer signed the " post test
l
review section and denoted the test as " accept" although the PMT
section was left blank indicating no test or no test required. The
licensee stated that no tests were specified since normal plant
surveillances would prove the breakers functional; however, there was no
existing plant surveillance which would have tested the main feed
breaker to bus 22.
On July 27, the bus 22 main feed breaker failed to auto-transfer. The failure
of the breaker caused the loss of the 2B reactor feed pump and the 2B
recirculation pump which resulted in single loop operation for the unit. .The
loss of the two pumps was a significant plant transient in which reactor power
dropped from 28% to 17%.
It should be noted that GE had previously identified
problems with the position switch and documented the problem in a service
advice letter (SAL) dated May 23, 1978.
The SAL applied to vertical lift 4kV
breakers used at Quad Cities. The licensee had not taken any' corrective
action to address this vendor identified problem.
The inspectors observed maintenance work performed August 11, 1993, on a 4kV
breaker in cubicle 2 on 4kV bus 31, Safe Shutdown Feed to MCC 30 and concluded
that the maintenance work was satisfactorily performed and the technicians
were competent and knowledgeable. This work was performed according to
procedures, QEPH 200-6, " Inspection and Maintenance of 4kV Switchgear
Cubicles", Revision 2, and QCEPM 200-12, " Inspection and Maintenance of 4kV
Vertical Circuit Breakers Type 4.16-250-9", Revision 0.
The inspectors
observed that the maintenance work was adequately supervised and that
applicable procedures were followed.
3.2.4
Undervoltage Relavs
'
The inspectors reviewed calibration data sheets for the RAT safety-related 4kV
bus and the 4kV Technical Specification (TS) type IAV69A undervoltage (UV)
l
,
relays. The inspectors found that Unit I had a 1.7 second time delay for a
transfer from the RAT to the UAT and Unit 2 had a 1.2 second time delay.
By
station procedures the tolerances on these relays were plus or minus 10
percent.
Therefore, the RAT UV relays could be set with a time delay as high
as 1.87 seconds while the TS relays could be set as low as 1.8 seconds. This
has the potential to result in improper coordination between the RAT. and the
TS UV relays.
i
l
In addition, the inspectors identified the UV tolerance setpoint on
calibration data sheets, OADMP-B1, was 83 volts 15% (79 to 87 volts) for TS
relays on buses 23-1 and 24-1.
However, TS Table 3.2-2, an OAD relay setting
.*
orrier, and procedure QC0 ADS 100-1, " Rock River Division 0AD Undervoltage Relay
Calibration," Revision 0, required a setting of 87 volts i 5% (82.65 to 91.35
8
_ _ _ _ _ - _ _ - _ _ _ _ _ _ _ _ _ _ _ .
.
.
-
.
.
volts) for the 4kV emergency bus UV relays. Although these relays were
fortuitously set within the TS limits on March 18, 1993, the licensee had not
used the latest calibration procedure revision which was issued in February
1993. The latest procedure revision incorporated the appropriate design
requirements and acceptance limits associated with 4kV bus 23-1 and 24-1 UV
relays into the calibration data sheets.
Failure to implement ~ appropriate
design requirements and acceptance limits into procedures used during testing
is considered a violation of 10 CFR 50, Appendix B, Criterion XI
(254/26593024-02B(DRS)).
3.2.5
D.C. Grounds Program
As a result of previous inspections 254/88011 and 265/88012, a violation and
civil penalty was issued on the 1/2 Emergency Diesel Generator (EDG). The EDG
was determined inoperable in excess of five months due to a hard ground that
was installed in the EDG during modification.
The licensee implemented
several corrective actions, including the development of a corporate policy on
-DC grounds and the revision of four plant procedures.
The corporate policy
was issued as Nuclear Operations Directive, N00-0P.16, "DC Ground Action
Requirements," on October 1, 1989. The inspectors noted that several N0D
items were not adequately addressed in the implementing procedures as noted
below:
The N0D defined three levels of DC grounds (Levc1 I, II, and III), as
o
stated in the licensee's original commitment. However, the plant
procedures did not refer to the levels of DC grounds and which operator
actions were required.
O
Some items in N0D paragraphs 5.3.3, 5.3.4, ard 5.4 were not addressed in
the plant DC ground procedures.
These paragraphs referred to.
identification of new grounds masked by other grounds and preparation of
a "DC System Ground Report".
The N00 included a table indicating voltage / resistance correlations and
o
response actions at different Ceco nuclear power plants.
The contents.
of the table were not included in the plant procedures.
The N0D directed that a "0C System Ground Report", be prepared for each
I
o
i
level II or III ground, and that this report must be filed at the
'
station. The plant procedures did not include this requirement.
The
i
i
l
ground reports were not being issued or filed. As all .the grounds data
was not being compiled in one file, it was difficult to trend the DC
grounds and to evaluate the root causes,
The "DC System Ground Report" form in the N0D stated that, "If the
o
ground is a level III and cannot be corrected within three days of its
occurrence, a copy of this' data form must be sent to the general offic'e'
-(Nuclear Engineering and the appropriate Nuclear Operations General
i
Manager) within 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />." This requirement was not~ included in the-
i
plant procedures.
9
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
_ _ _ _ _ _ _ _ _ _ _ -
..
.
.
At the exit meeting, the licensee stated that N0D-0P.16 was canceled on
April 28, 1993, The cancellation letter stated that the N00 was canceled
since the station had implemented most of the N00 functions by station
procedures.
The licensee's failure to implement procedures to adequately
address the requirements of N0D-0P.16, a licensee's commitment to a previous
escalated enforcement action, is considered a violation of 10 CFR 50,
,
Appendix B, Criterion V (254/265-93024-03(DRS)).
3.2.6
Followun of Industry Experience on Breaker Maintenance
The inspectors determined that the licensee was not reviewing industry
initiatives for applicability on a timely basis.
Procedura QCAP 2300-6,
" Station Commitment and Action Item Tracking," Revision 0, required that
vendor items, such as, General Electric (GE) service information letters
(SILs) be reviewed within 90 days. The inspectors requested the licensee to
perform a review of SIls received during the past year and identify those SIls
that did not meet the QCAP 2300-6 requirement. The review showed that for
most of the SILs timely evaluations were not performed.
For example, three
SIls Nos. 548, 550, and 551 were not reviewed within the specified 90 day
period.
Reviews for SIls Nos. 550 and 551, which were received in March 1993,
were not completed until August 20, 1993.
The licensee's failure to perform
timely evaluations of SIls in accordance with procedure QCAP 2300-6 is
considered a violation of Criterion XVI of 10 CFR 50, Appendix B
(254/265-93024-04A(DRS)).
In addition, a problem involving 4kV breakers was revealed during the review
of GE service advice letter (SAL) 205, which concerned the failure of SBM
control switches manufactured during August through October 1982.
The
licensee added an inspection requirement to inspection procedure, QEPM 200-1,
revision 3, to identify breakers that contained SBM switches and effect
replacement of the defective SBM switch.
The licensee completed 35 of the 89
scheduled breaker inspections, but for unknown reasons removed the inspection
requirement from the next revision to the procedure.
Consequently, the
licensee failed to complete the corrective action inspections to identify and
replace, as required, defective SBM switches on 4kV breakers. The licensee
.
stated that the inspections would be reinstated for the remaining breakers.
The failure to complete corrective action inspections is considered a
violation of 10 CFR 50, Appendix B, Criterion XVI (254/265-93024-04B(DRS)).
On December 23, 1986, GE issued SIL No.448, which described the maintenance of
GE Type AK Circuit Breakers. The SIL recommendations included performance of
preventive maintenance (PM) and inspections at twelve month intervals,
complete disassembly / overhaul of the breakers at intervals not exceeding five.
years, and the use of a specific grease.
The licensee completed the SIL
evaluation on February 19, 1987, and took action regarding the grease
recommendations. However, the licensee's review did not address the
recommendations on frequency and type of maintenance recommended for the
~
breakers.
The inspectors considered this a weakness in the SIL evaluation.
The team noted that 4kV breakers were not being maintained in accordance with
vendor recommendations.
GE recommended inspection and lubrication of all
breakers at every scheduled refueling outage and a complete overhaul of
10
L
,
-
.
,
-
-
-
9
.
j
~
.
.
i
,
breakers every five years or earlier if problems were detected. During 1990,
the licensee completed an engineering study and decided that safety related
i
4kV breaker maintenance at Quad Cities would be performed at 36 month
'
intervals.
The licensee stated that this decision was based on Quad Cities
breaker failure history, which was bett( .- than the industry average (6.5
failures / unit versus 7.7 failures / unit). This decision was made with
consideration for a review of failure history after the program was in place
for at least one cycle on all breakers. A current review of NPRDS data (past
two years) indicated that Quad Cities breaker failure history was 4.38
i
failures / unit versus 3.11 failures / unit for the industry, placing Quad Cities
failure rate in the top 17% of the industry. The licensee stated that 4kV
breaker maintenance schedules would be reviewed based on current breaker
failure rates.
l
3.2.7
Repeat Breakaae of Secondary Disconnects
The inspectors noted that there were 11 broken secondary disconnects in 1992
and one in 1993.
In addition, there were eight missing components (such as
springs and washers) and eight broken components, other than disconnects. An
apparent reason for the repeated damaged to 4kV breaker components was
perceived to be the handling process during racking the breaker and its
transport to and from the busses.
There were two specific documented notifications of improper breaker handling.
One was an entry in a system engineer's log book and the other was in a work
'
request, which referenced excessive force was used during breaker handling.
The systr: engineer's log book rede a direct reference to improper handling of
the 4kV breakers by E0s during racking.
However, this notification was
neither reviewed by the group leaders and supervisors; nor was it elevated by
the system engineer to higher management for consideration and evaluation.
The failure to provide appropriate notification to plant management after
identification of improper breaker handling is considered a violation of 10 CFR 50, Appendix B, Criterion XVI (254/265-93-0024-04C(DRS)).
3.3
System Enqineerina
System engineers' knowledge about the impact of nonsafety-related high, medium
and low voltage breaker failure on safety systems revealed some weaknesses.
One system engineer was not aware of the content of the PM procedure for the
designated system and did not monitor various surveillance tests performed on
i
the breakers.
In addition, the system engineer was not aware of the impact of
4 kV breaker failure on the safety related systems. It appeared that, at
times, problems on the breakers could be identified and resolved without the
system engineer's awareness and knowledge. These instances of inadequate
,
communication, inadequate understanding of maintenance and testing, and
{
inadequate awareness of system interaction was considered a weakness.
4.0
Inspection Followuo Items
Inspection followup items are matters which have been discussed with the
licensee, which will be reviewed further by the inspectors, and which involve
some action on the part of the NRC or licensee or both. An inspection
followup items disclosed during the inspection is discussed in Section 3.1.4
1
of this report.
11
.
--
._
.
.
..
-..
-
a
.
9
5.0
Exit Meetina
The inspectors met with the licensee's representatives (denoted in Section 1)
during the inspection period and at the conclusion of the inspection on
August 20, 1993. The inspectors summarized the scope and results of the-
inspection findings. The inspectors discussed the likely content of the
inspection report with regard to documents or. processes reviewed by the
inspectors. The licensee acknowledged the information and did not indicate
that any of the information disclosed during the inspection could be
considered proprietary in nature.
,
,
s
f
-
r
i
I
i
?
!
f
I
h
. . .
l
i
l
12
W..
. . .
,
,.
. . , - .
-
F
.
e
'
...
ATTACHMENT A
PAGE 1 0F 2
CHRONOLOGY OF ACTIVITIES FOR
QUAD CITIES LOOP EVENT
o
JULY 26. 1993
Time Description of Activity
1500 Preparations were being made to synchronize the Unit 2 generator to the
grid.
Quad Cities Unit 2 mode switch was in the run mode.
1948 An E0 noticed smoke and heard boiling noise coming from OCB 9-10,
B phase, while performing a routine surveillance in the 345kV
1955 0CB 9-10 was tripped open. The boiling noise stopped.
2025 0CB 10-11 was tripped open in order to synchronize the Unit 2 generator
to the grid. This caused a partial loss of the station ring bus.
2043 Unit 2 was synchronized to the grid through OCB l-11; however, the
control room operators were unable to close OCB 10-11. Operations
personnel called 0AD to assist in closing OCB 10-11.
2150 The operators received a rod permissive light indication on the rod
select matrix although no rod had been selected (this was not related to
the LOOP event).
o
July 27.1993
Time Description of Activity
0025 Operators again attempted to close OCB 10-11 without success.
0AD
notified the operators that the HACR relays would not allow an OCB to
close on a dead bus.
0233 OAD placed a jumper across the HACR relay allowing closure of OCB 10-11.
0252 Operators tripped OCB 10-Il-open when a SSC worker performing CM on
OCB 9-10 heard boiling noise coming from 0:B 10-11, C phase.
0315 A fault was sensed on transmission line 040? and OCBs 7-8 and 8-9
tripped causing de-energization of RAT.
The control room operators expected a fast transfer (approximately
"~'
6 cycles) from the RAT to UAT. However, numerous indications of
momentary loss of power and alarms were received, such as, the
instrument bus and the essential service (ESS) bus alarms (loss of 480V
buses 25 and 28) and the associated 2A RPS MG.
In addition, the nonsafety-related bus 22 failed to transfer to the UAT.
The operators were unable to manually close the main feed breaker to
bus 22. The failure of the breaker to close resulted in the loss of the
2B recirculation pump and 2B reactor feed pump.
Consequently, reactor
L
.__
',
.
. .
9
ATTACHMENT A (CHRON0 LOGY)
PAGE 2 0F 2
July 27.1993 (Cont.)
'
Time Description of Activity
j
.
. !
level dropped to 25 inches and reactor power dropped from 28% to 17%.
Reactor level recovered shortly thereafter.
In addition, a Unit'2 LCO
1
was entered due to loss of the recirculation pump (reactor in single
loop operation).
-
0337 An unusual event was declared due to the LOOP. Both units entered into
!
LCOs.
,
0443 Operators attempted to energize bus 22 through the main feed breaker but
>
were not successful.
!
0508 Power was restored through OCB 8-9 and the RAT was re-energized.
This
restored offsite power and both units exited their LCOs.
0536 Bus 22 and safety-related bus 23 were loaded on to the RAT.
0547 Operators again attempted to close OCB 7-8 without success.
,
0618 The unusual event was terminated.
i
0700 Unit 2 was kept at less than 30 percent power until the rod celect
matrix was repaired.
1315 The 2B recirculation pump was started.
The single loop operation LCO
was exited.
1400 A problem identification form was issued to troubleshoot the rod select
matrix. The rod select matrix was later found to have operated in
accordance with the vendor instructions,
i
!
!
,
--%.
F
i
L.