ML17188A049
| ML17188A049 | |
| Person / Time | |
|---|---|
| Site: | Dresden  |
| Issue date: | 02/06/1998 |
| From: | NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION III) |
| To: | |
| Shared Package | |
| ML17188A047 | List: |
| References | |
| 50-237-97-28, 50-249-97-28, NUDOCS 9802200080 | |
| Download: ML17188A049 (30) | |
See also: IR 05000237/1997028
Text
-
.
- -- --* - ----
.
U.S. NUCLEAR REGULATORY COMMISSION
REGION Ill
Docket Nos:
50-237, 5~249
License Nos: -OPR-19, DPR-25
Report No: *
50-237/97028(DRP); 50-249/97028(DRP)
- Licensee:
Facility:
Location:
Dates:
.:*
. ;: ;
- Commonwealth Edison Company
Dresden Nuclear Station, Units 2 and 3
6500 N. Dresden Road
Morris, IL 60450
November 23, 1997, to January 12, 1998
Inspectors:
K. Riemer, Senior Resident Inspector
8. Dickson, Resid~nt Inspector
D. Roth, Resident Inspector
Approved by: M. Ring, Chief
Reactor Projects Branch 1
9802200080 980206
ADOCK 05000237
Q
II l \\
EXECUTIVE SUMMARY
Dresden Generating Station, Units 2 and 3
NRC Inspection Report No. 50-237/97028(DRP); 50-249/97028(DRP)
.
.
This inspection included routine resident inspection from November 23, 1997, to January 12,.
1998.
Operations
The material condition of the HPCI system impacted system availability and required
operator work-arounds to assure HPCI system operability. Repetitive equipment
problems with the gland s~al*condenser level switch caused Unit 3 HPCI to be declared
. inoperable, and the alignment of.the condensate storage tank once caused both HPCI *
systems to be declared inoperable. (Section 02.1) .
The licensee's response to identified errors in the setpoints for system oil temperatures
was poor. The licensee's original explanation of setpoint tolerances was incorrect, and
the situation was not addressed until operators wrote. a second problem identification
form. (Section 02.1)
The operators' response to the automatic reactor trip was good. The inspectors
concluded that the actual safety consequences of this event were low. Operator and
plant equipment response were generally'as expected for an automatic reactor trip from
full power.* The exception involved the response of the feedwater level control (FWLC)
system which over filled the vector vessel; however, in this case there were no adverse
- consequences from the level overshoot since the HPCI system was already isolated for
troubleshooting efforts. (Section 04.1)
Operations personnel exhibited safe operating practices during the startup of Unit 2 .that
commenced on December 26. Crew briefs and heightened level of awareness briefs
were informative, contingency actions were discussed, and peer checks were performed.
(Section 04.2)
The operations staff was slow to declare the HPCI system inoperable following the gland
seal leak off condenser low level switch failure on December 29, 1997. More than
17 hours1.967593e-4 days <br />0.00472 hours <br />2.810847e-5 weeks <br />6.4685e-6 months <br /> passed from the first symptom until operations recognized that the system was *
inoperable. Even after recognition, the limiting conditions for operation were not
retroactively entered. (Section 04.3)
Operators generally were knowledgeable .of the HPCI system parameters, settings, and
requirements. The inspectors identified one instance involving turbine lube oil
temperature where the requirements were not known. (Section 04.4)
The licensee failed to follow the procedural requirements to provide feedback to the
problem identifieation form (PIF) originator. Subsequent to the inspectors' review of the
. process, the licensee independently identified this procedural adherence concern and
entered it into the corrective action program. The licensee met the procedural
requirements before the end of the inspection period. (Section 07.1)
2
.,
- -* - -*-
- --
-*
,-----------*-* -
Maintenance
-*
Collectively, equipment failures and material condition issues involving a control rod drive,
a torus cooling test valve, an HPCI isolation, erratic operation of a. recirc motor generating
set, the offgas system, reactor water cleanup and feedwater level control, represented
challenges and distractions for operators and other plant staff. The issues especially
represented a burden to operato~s who had either to respond to the original event, or to
take additional compensatory actions. (Section M2.1)
The inspectors were also concerned with the licensee's ability to resolve the issues
effectively and eliminate deficiencies. (Section M2.1)
On December 13,. work**performed without referencing a required procedure, combined
with material condition, resulted in a trip of the Unit 3 125 V battery charger and placed
both units unexpectedly into*a two-hour limiting condition for operations. The subsequent
follow up work was not performed in accordance with station administrative procedures.
Specifically, the "condition" met sign off was used when a condition had not been met.
(Section M4.1)
Not all rework was captured into the rework trending program. (Section MB.1)
Engineering
The inspectors concluded that the licensee's root cause team performed a thorough
investigation of the reactor trip, the root cause, and equipment response following the
reactor scram. The licensee's team concluded that the root cause of the event was the
failure to perform the actions identified in GE SIL 500 regarding local power range
monitor spiking. The inspectors' review reached the same conclusion.
The FWLC system response presented a potential challenge to the operators following
the reactor scram. The compensatory actions that the operators were required to take
following a scram on Unit 2 were operator work-arounds. Pending permanent resolution
of the Unit 2 FWLC system issues, the station was relying on operator intervention
following a scram to prevent water intrusion into HPCI steam lines. (Section E1 .2)
Operability evaluations appeared to meet the licensee's requirements. The evaluations
were reasonable* and provided adequate bases for the conclusions. (Section E2. 1)
Plant Support
The s~tup and control of contaminated areas and work in contaminated areas were
usually correct. (Section R4.1)
3
1=--- -.-':..---** -------- -*-----------
The inspectors identified one example of an improper1y secured hose ~hat crossed a
contamination boundary. The hose had been staged in response to a poor1y performing
sump pump. (Section R4.1)
The inspectors were concerned with the lack of attention to detail exhibited by the plant
staff to radiation controls, and that the PIF record showed this to be an emergent trend.
(Section R4.1)
, .
4
Report Details
Summary of Plant Status
Unit 2 started the reporting period in a load recovery from the power reduction required for single-
loop operations and 2A feedwater regulating valve (FWRV) work. On December 3, power was
reduced to about 600 MWe for repair of the 2E condensate demineralizer service unit. Recovery
started on December 8, but oscillations on the 2A FWRV delayed full-power operations until
December 12. On December 23, Unit 2 automatically scrammed from full power due to a local
power range monitor (LPRM) spike, and a forced outage was entered (D2F30). Unit 2 was
placed back on the grid by December 27.
Unit 3 maintained full power throughout most of the period. On December 6, load was reduced
to mitigate a fire in the off-gas piping. Several times during the period, load was reduced to
attempt to address problems with the 38 reactor water cleanup demineralizer bed. Maximum
power on Unit 3 was slightly limited to maintain the average turbine control valve positions less
than 85 percent.
Maximum power on both units was limited by feedwater flow. Feedwater flows were limited to
9.735 Mlbm/h to remain within the anticipated-transient-without-scram (ATWS) analysis. The
licensee was pursuing additional analysis to remove this restriction.
I. Operations
01
Conduct of Operations
01.1
General Comments
Using Inspection Procedure 71707, the inspectors conducted frequent reviews of ongoing
plant operations. Overall, the conduct of operations was safe and according to
procedures.
During the inspection period, several events occurred for wtiich the licensee was required
by 10 CFR 50.72 to notify the NRC. The events and the notification dates are listed
below:
11/26/97
12/01/97
12/06/97
(Units 2, 3) Units 2 and 3 HPCI systems declared inoperable after
engineering determined that HPCI system operation could result in air
intrusion into HPCI .system.-
(Unit 3) Failure of torus cooling outboard test valve caused a potential for*
diversion of cooling flow from the reactor during a design basis loss of
coolant accident.
(Unit 3) Control rod inserted into the core during a surveillance test. The
event report was retracted on 12/18/97 after the licensee concluded that
the *event was not an engineered safety feature (ESF) actuation.
5
. -
- J
- =*.::..:.: .. .....::=..-:-.:..=..._-=-::=._ .. , ----**--*-___ _.::=....::__::. ___ -=.._ _____ ~
__ .:. ____ . ____ _
12/12/97
12/22/97
12/23/97
12/29/97
1/07/98 .
{Units 2, 3) Loss of 138 kV line that fed the lift station required operating
the units in lake-bypass mode violating the national pollution discharge
elimination system permit'. Offsite notification made to the Illinois
emergency management agency.
{Unit 2) ESF Actuation - HPCI Isolation during a routine surveillance for
unknown. reasons.
{Unit 2) Full reactor scram from 100 percent power due to a spurious
APRM signal during an unrelated surveillance of reactor vessel high
.pressure scram signals.
{Unit 3)* HPCI system .declared inoperable due to failure of gland seal .
condenser low level switch ..
{Units 2,. .3) Unanalyzed condition that may significantly compromise plant
- safety identified when calculations showed the. post-LOCA reactor building
temperatures to be significantly higher than the limitihg value stated in the
02
Operational Status of Facilities and Equipment
- 02.1 (Units 2. 3) Engineered Safety Feature System
a.
. Inspection Scope (71707)
The inspectors conducted a detailed review of the Unit 2 and Unit 3 High Pressure
Coolant Injection {HPCI) systems to verify operability, assess the performance, and
assess material condition of the systems. The inspectors also performed a cursory
walkdown of the HPCI system to ensure that the alignment procedures, piping and
instrumentation diagrams {P&IO), and the as-built configurations were current. The
inspectors reviewed the following operating procedures, schematics, and the results of
quarter1y operability surveillances against information established in the Updated Final
Safety Analysis Report (UFSAR), Technical Specifications (TS), and the licensee's
operation training manual:
DOP 2300-01 Unit 2(3) HPCI System Standby Operation, Rev. 15
DOP 2300-02 Unit 2(3) HPCI System Turning Gear Operation, Rev. 06
DOP 2300-03 Unit 2(3) HPCI System Manual Startup and Operation, Rev. 24
DOP 2300-M1/E1 Unit 2(3) HPCI System Checklist, Rev. 15
DOS 2300-03 Unit 2(3) HPCI System Operability Verification, Rev 49
DOS 2300-07 Unit 2(3) HPCI Fast Initiation Test, Rev. 18
P&ID M-51, HPCI Piping Unit 2,
P&ID M-374, HPCI System Piping Unit 3.
WR No. 970093784-01 Unit 2 Quarter1y TS HPCI Pump Test (IST Program) dated
Nov. 19, 1997
WR No. 970094986-01 Unit 3 Quarter1y TS HPCI Pump Test {IST Program) dated
Nov.26, 1997
6
b.
Observations and Findings
During walkdown of the HPCI systems, the inspectors determined that the alignment of
the systems was in accordance with the operating procedures. The inspectors also noted
that housekeeping for both Unit 2 and Unit 3 HPCI system rooms was good.
The inspector noted several oil l~aks throughout both the Unit 2 and Unit 3 HPCI
systems. The licensee also noted-these leaks and had written action requests (ARs) to
address the deficiencies.
- . ~--:
The Unit 2 HPCI System room was at elevated temperature due to steam leaking past the
HPCI steam supply shutoff valve (2-2301-3) and into the HPCI floor drain sump through
the* HPCI *stop valve above seat drain :line.- ' *
FrontStandard Temperatures
On December 19, the inspectors noted that the alarm setpoints for four temperature dial
switches (Unit 3 HPCI bearing oil cooler outlet temperature, Units 2 and 3 low pressure
bearing drain oil temperature, and Unit 2 thrust bearing oil drain temperature) shown on
the HPCI system front standards appeared to be at different settings than listed in the
Dresden Annunciator Procedures (DANs) associated with the alarms. The inspectors
informed the Unit 3 Unit Supervisor (US), who contacted the system engineer and wrote a
problem identification form (PIF) to document the concern. The PIF was subsequently
canceled by the PIF screening committee:*
The system engineer reviewed the alarm setpoints and concluded that the settings were
within the allowed tolerances based on a review of the instrument maintenance
department (IMO) data cards.
The IMO data card showed that the allowed tolerance for temperature indicator face value
was +/-5°F and the tolerance for the dial switch was+/- 2°F. Using this information, the
system engineer thought that the two tolerances could be added to give an acceptable *
tolerance of+/- 7°F. The inspectors questioned this conclusion, and determined that the
appropriate tolerance for the dial switch was only +/- 2 ° F.
The inspectors also reviewed the instrument maintenance department (IMO) data cards
for the instruments and concluded that the instruments were originally set correctly, but
had subsequently drifted out of tolerance. In one case the temperature dial alarm setting
was greater than 5°F outside tolerance, upscale. In another case,- the setting was 16°F
outside tolerance, downscale. The inspectors informed operations that a concern still
. existed, and operations wrote a new.PIF to document the concern. The licensee
eventually wrote an Action Request (AR) tag to correct temperature switch settings.
The inspectors determined that the out-of-tolerance-temperature switches did not make
the HPCI system inoperable.
Exhaust Drain Pot Alarms
Jhe control room logbooks documented that the Unit 2 HPCI system exhaust drain pot
high level alarms were annunciating at least once a day. The US explained that
7
condensation resulting from' seat leakage of the Steam Supply Shutoff Valve (2-2301-3)
caused an abnormal input to the exhaust drain pot via the* stop valve above seat drain
line.
The licensee concluded in Engineering Operational Problem Response/Troubleshooting
Plan (EOPR) 98-02-23-318, Rev. 0, that operations personnel needed to take
compensatory actions to ensure the operability of the HPCI system. The compensatory
actions included the non-licensecfoperators (NLOs) manually draining the exhaust drain
pot once per week. The EOPR suggested that the quantity' drained be evaluated and the
draining frequency adjusted as needed. The inspectors concluded that the EOPR
appeared reasonable.
The Operations*Department declared the HPCI drain pot.~evel high-alarm as a control *
room distraction, and considered the draining to ~e an operator work-around.
HPCI System Availability
As stated in Section 01, 1, three issues resulted in one or both HPCI systems being
declared inoperable during this inspection period. On November 26, both units' HPCI
systems were declared inoperable after engineering determined that air intrusion into
HPCI system could occur due to condensate storage tank (CST) alignment. On
December 22, during a routine surveillance test, the Unit 2 HPCI system unexpectedly
isolated for unknown reasons. On December 29, the Unit 3 HPCI system was declared
inoperable due to failure of the gland seal condenser low level switch. Similar problems
with gland seal condenser level were discussed in Licensee Event Report (LER)
50-249/97-09-00 and LER 50-237/97-013-00, and in Inspection Reports (IR) 97012,
97019, 97024. Additional follow.,.up for all three issues will be tracked through the LER.
c.
Conclusion
The HPCI systems were property aligned in accordance with procedures.
The material condition of the HPCI system impacted system availability and required
operator work-arounds to assure HPCI system operability. Repetitive equipment
problems with the gland seal condenser level switch caused Unit 3 HPCI to be declared .
inoperable, and the alignment of the condensate storage tank once caused both HPCI
systems to be declared inoperable.
The licensee's response to identified errors in the setpoints for system oil temperatures
was poor. The licensee's original explanation of setpoint tolerances was incorrect, and
the situation was not addressed until operators wrote a second PIF .
8
03
Operations Procedures and Documentation
03.1
(Units 2. 3) Control Room Rounds
a.
lnspection'sc0pe cr1707)
The inspectors reviewed operator use of panel monitoring sheets.
b.
dbs~rvations and Findings
The use of these sheets helped the.licensee in detecting abnormal trends (e.g., the
valved-out fuel pool cooling discussed in IR No. 50-237/97019;.50-249/97019
Section 03.1). The rounds sheets included a column of normal operating parameters.
The inspectors noted that some values being recorded were not within the normal
operating bands. For example, the torus temperature was 5°F above the listed value .
. Discussions with the operating staff indicated that the bands on the rounds sheets were
not always ttie actual normal parameters, and that some operators did not routinely verify
the parameters against the bands. The values were instead compared to values the
operators knew from TS and operations procedures. The inspectors also noted some
confusion about who performed the primary review of the parameters (US or nuclear
station operators).
The inspectors considered the acceptance of rounds sheets with "normal" operating
parameters that were not normal to be a poor practice. All operators interviewed were
aware that the bands listed on the rounds sheets were not always the actual normal
operating band. The operators explained that when the rounds sheets were created, the
intent was to heighten panel monitoring and trending. The full "normal" bands for all
equipment had not been entirely listed, and equipment outside the "normal" bands may
still be within required bounds.
At the end of the inspection period, the licensee was evaluating improvements to the
rounds sheets.
c.
Conclusions
The use of panel monitoring rounds sheets helped operators identify trends and maintain
panel awareness. However, the operators did not .ensure that the rounds sheets
contained the correct normal operating parameters for all equipment. The inspectors
were concerned that acceptance of the incorrect bands reduced the rounds sheets'
effectiveness.
04
Operator Knowledge and Performance *
04.1
Unit 2 Automatic Reactor Trip
a.
Inspection Scope (71707. 93702)
The inspectors reported to the main control room and observed operator performance
following a Unit 2 automatic reactor trip (scram) that occurred December 23, 1997. The
inspectors reviewed the significance of the event, performance of safety systems, and
9
i
b.
actions taken by the licensee. The inspectors also reviewed station logs, control room
recorder indications, the scram investigation team's results, and assessed the functioning
of the Plant Operations Review Committee {PORC) meeting held to approve a unit
restart.
Observations and Findings
Before the reactor scram, the HPCI system was out of service and isolated due to a
spurious full isolation that occurred on the previous day. All other emergency core
cooling systems {ECCS) were in normal alignments.
Instrument maintenance department personnel were performing Dresden Instrument
Surveillance (DIS) 0500-01, "Reactor Vessel High Pressure Scram Pressure Switch
Calibration." Part of the surveillanee,resulted in actuation of the reactor protection system
{RPS) Channel "B" half scram. This was expected. *While the planned half-scram was
actuated, an unexpecte~ average power range monitor {APRM) high-high signal
occurred, and actuated Channel A of RPS. This resulted in a full scram signal *
Operator Response
Operator response to the transient and performance during scram recovery were good.
The inspectors observed good procedural usage by the operators, formal
communications throughout the event, and effective command and control by the US.
Unit 3 activities were reduced to lirriit distractions to the Unit 2 operators.
Equipment Response
All rods inserted, and the reactor automa~ically shut down. However, not all equipment
responded ideally. The reactor feedwater level control system caused the reactor
pressure vessel {RPV) level to increase above +48." This would have flooded the HPCI
system's steam lines, but the HPCI system was already isolated. This item, and required
operator compensatory actio!'ls to prevent a repeat occurrence, are discussed further in *
Section E1 .2 of this report. The inspectors concluc;ted that the actual safety
consequences of this event were low.
Prompt Root Cause
.
.
The licensee formed a prompt root cause team to de_termine the immediate causes of the
reactor trip. A formal root causes investigation was also assigned,* but was not
completed during this inspection period.
The immediate cause of the event was that LPRM 20-24-41 spiked high, causing
APRM 2 to spike high, which in tum generated a trip on RPS Channel "A." Since a trip on
RPS Channel "B" half scram was already actuated due to testing, the RPS scram logic
was satisfied and a full automatic reactor scram occurred.
General Electric {GE) service information letter {SIL) 500, issued October 23, 1989,
discussed the phenomenon of LPRM spiking. The SIL stated that a "whisker" {buildup of
uranium oxide) arcing in the detector caused spikes. The arcing typically eliminated the
whisker. To prevent spikes, the GE SIL recommended that detector breakdown
10
- - -- *-* *---* ---
- ~----- -*--
--*-***--
- -*--------~**
_: __ ,. _ _: -- . . __ :_ -- . *-----* -
--*
(current-voltage) tests be performed at specified intervals to look for whiskers in the
detectors. The SIL also provided guidance on how to bum-off. the whiskers to prevent
spikes.
The licensee found that ComEd's Nuclear Fuel Services had reviewed the SIL, but did not
require performance of all ofthe recommendations in the SIL. Instead, bum-offs were
only performed on problem detectors. The licensee's investigation team concluded that
the failure to perform all of the GE SIL recommendations directly contributed to the spike.
The inspectors assessed the PORC meeting held to review the prompt root causes and
corrective actions prior to plant restart.* The PORC thoroughly discussed the event, the
investigation team's cpnclusions, and the reconimende'd actions prior to restart. The
PORC maintained an appropriate safety-focus during review of the prompt root cause.
c.
Conclusion
The operator response to *the automatic reactor trip was good. The inspectors concluded
that the actual safety consequences of this event were low. Operator and plant
equipment response were generally as expected for an automatic reactor trip from full
power. The exception involved the response of the feedwater level control (FWLC)
system; however, in this case there were no adverse consequences from the level
overshoot since the HPCI system was already isolated for troubleshooting efforts.
The prompt root cause investigation team formed by the licensee performed a thorough
review of the event and subsequent equipment problems. The inspectors concluded that
the prompt root cause of the event was the failure to perform the actions identified in GE
SIL 500 regarding LPRM spiking.
04.2
(Unit 2) Operations Performance During Startup
a.
Inspection Scope (71707)
The inspectors conducted observations of startup activities from forced outage D2F30.
b.
Observations and Findings
During the Unit 2 startup, operations observed were performed in a careful and controlled
manner. Good communications were evident, and the operators were knowledgeable of
the plant conditions and issues. The crew performed correctly and maintained
awareness of the plant status. The shift manager and US maintained correct command
and control during the startup. The US held crew briefs as necessary, and directed entry
into the correct abnormal procedures in response to several instances of double-notching
The startup and power ascension of Unit 2 were hampered by some minor equipment
problems such as double-notching control rods. The inspectors concluded that the
actions taken were appropriate to the symptoms and in accordance with plant
procedures.
11
- ..:. __ . __ . ____ _:_;:_: ___ :::__:,_ -*---
i
c.
Conclusions
Operations personnel exhibited safe operating practices during the startup of Unit 2 that
commenced on December 26. Crew briefs and heightened level of awareness briefs
were informative, contingency actions were discussed, and peer checks were performed.
04.3
Unit 3 HPCI lnoperabilitv
a.
Inspection Scope CT1707)
The inspectors reviewed the operators' initial operability call for the HPCI system on
December 29, 1997~ after the low level switch in the gland seal*leakoff (GSLO) condenser
failed to stop the drain pump.
b.
Observations and Findings
At 0234 on December29, 1997, while performing DOP 2300-01, "Unit 2(3) HP~I System
Standby Operation," Rev. 15, for the Unit 3 HPCI System, the operators started the GSLO
pump to pump down the GSLO condenser hotwell as required by the procedure. The
pump did not stop at the low level switch as designed but continued to pump down the
GSLO condenser. The Unit 3 nuclear station operator (NSO) manua.lly secured the
pump. The US control room logs stated that operability was immediately discussed by
the US, and a previous event was also considered. However, the US concluded that; for
this event, the Unit 3 HPCI system was operable.
The operability issue was revisited later in the shift. An entry made at 0447 in the US log
book showed that the US reviewed the surveillance tests, the UFSAR, emergency
notification system (ENS) requirements, previous ENS calls made for the HPCI system,
the historical limiting conditions for operation (LCO) logs, and the design* basis
documents to determine operability requirements. The US also discussed the issue with
the on-call system engineer. The US again concluded that the HPCI system was
. operable.
The previous event occurred on September 5, 1997, and was discussed in
LER 97-009-00/50-249. The Unit 3 HPCI system was declared inoperable during a
surveillance test due to the failure of the GSLO condenser drain pump low level switch to
shut off the pump at the required low level. This led to cavitation and air entrainment in
the pump suction and air accumulation in the discharge pressure regulating valve sensing
line":.
The licensee continued review of HPCI system operability. Subsequently, the HPCI
system was declared inoperable at 1945, December 29, 17 hours1.967593e-4 days <br />0.00472 hours <br />2.810847e-5 weeks <br />6.4685e-6 months <br /> after the first
symptoms. The inspectors concluded that, due to the similarity between this failure and
the failure of September 5, the length of time was excessive .
12
c.
Conclusion
Operator performance with respect to the initial operability call was weak. While the
operators followed the station's administrative procedures for determining operability,
sufficient data was available to support a more timely declaration of HPCI inoperability.
When operations declared the HPCI system inoperable, the staff entered the appropriate
TS limiting conditions for operation from the time of the operability determination, but not
from the actual first symptom . .The inspector concluded that not entering the limiting
condition retroactively was not the most conservative decision. Information was available
to link the start of the HPCI system problem to the original failure of the GSLO pump to
stop automatically at the GSLO condenser hotwell low level.
This was of concern to the* inspectors because of its similarity to a recent issue regarding
the standby liquid control (SBLC) system. In IR 50-237/97024; 50-249/97024, the NRC
documented an instance in which operations failed to declare the SBLC system
inoperable upon receipt of valid control room alarms.
04.4
(Units 2. 3) Operator Knowledge of HPCI System Parameters
a.
Inspection Scope U1707)
The inspectors questioned .operators about HPCI system indications. The questions were
limited to operator knowledge of the actual HPCI parameters present on the units.
b.
Observations and Findings
The Nuclear Station Operators (NSOs) were generally aware of the HPCI initiation and
isolation signals, the valve interlocks and the operating parameters (temperature and
pressure limitations) of the HPCI turbine lube oil system.
One exception to this occurred during questioning of one Unit 2 NSO. During the
walkdown the inspectors noted that all HPCI system turbine lube oil temperature
computer monitor points on Control Panel 902-3 indicated between 92 to 95°F. The
inspectors then questioned the NSO on temperature limits while the HPCI system was in
standby operation and the expected temperature bandwidth for HPCI turbine lube oil
temperature when he performed his control board walkdown. The NSO answered by
stating that he was not aware of any temperature limitations of the HPCI turbine lube oil
system while in standby operations. Both DOS 2300-03 and DOS 2300-07 required HPCI
oil c0oler discharge temperature be greater than 96°F. The manufacturer's manuals
cautioned that bearing inlet temperatures outside of limits should be avoided to prevent
- damaging the bearings due to poor oil circulation.
After a discussion of concerns with the inspectors, the US ordered the NSO to place the
HPCI turbine on the turning gear since the oil pump adds a significant amount of heat to
the oil. The US also initiated an AR to calibrate the HPCI sump heater to a higher
temperature to ensure that HPCI oil temperatures are maintained within the design
temperatures. Instrument Maintenance Department (IMO) *personnel adjusted the
controller for the HPCI sump heater on December 23, 1997.
13
- .
-
...
--
--
- *--
The inspectors noted that control room hourly rounds sheets did* not include sump oil
temperatures. The licensee was considering adding these parameters to the rounds
sheets.
c.
Conclusions
Operators generally were knowledgeable of the HPCI system parameters, settings, and
requirements. The inspectors identified one instance where the requirements were not
known.
07
- Quality Assurance in Operations
07.1
Failure to*Follow Integrated Reporting Program Procedure *
a.
Inspection Scope*(71707. 40500)
The inspectors reviewed the licensee's corrective actions procedure, Nuclear Station
Work Procedure (NSWP) A-15, "Integrated Reporting Program," to determine the
station's compliance with the stated requirements. * The inspectors reviewed whether the
PIF process was property followed after PIFs were submitted.
b.
Observations and -Findings *
The licensee documented problems and non-conforming conditions via use of PIFs. *
Licensee Procedure NSWP-A-15, "ComEd Nuclear Division Integrated Reporting
Program," Rev. 1, required a feedback form be provided to the originator of the PIF. The
inspectors identified that the licensee was not sending the required feedback forms to the
PIF originators in accordance with NSWP-A-15.
The licensee was required by TS 6.8.A to implement the applicable procedures
recommended in Appendix A of RG 1.33, Rev. 2, February 1978. Procedures for
administrative controls were recommended in RG 1.33. Contrary to the above, the
licensee failed to.implement the requirement of NSWP-A-15 to provide feedback to PIF
originators. This was a violation of TS 6.8.A (VIO 50-237/97028-01 (DRP);
50-249/97028-01 (DRP)).
Subsequent to the inspectors identification of the failure to initiate the required feedback
forms, but prior to the inspectors' presentation of the finding to licensee management, the
licensee independently identified that the PIF process was not being followed with respect
to the feedback forms. The licensee's Quality and Safety Assessment (Q&SA)
organization identified that the procedure was not being followed across the board with
respect to the feedback issue. The Q&SA department concluded that when the new
requirement was put into the NSWP, the Dresden Staff did not realize the new
requirement and did not change local practices to comply. On December 3, 1997, the
Q&SA department initiated i:>IF 01997-08096 to document that the required feedback
forms were not being distributed to the originators as required by NSWP A-15.
14
I
The licensee commenced generating the required feedback forms following identification
of the issue and provided several examples to the inspectors for review. The licensee
entered the procedure adherence concern, via PIF D1997-08096, into the corrective
action process for long term resolution.
c.
Conclusion *
The licensee failed to follow the procedural requirements to provide feedback to PIF
originators. After the inspectors' review of the process, the licensee independently
identified this procedural adherence concern and entered it into the corrective action
program. The licensee met the procedural requirements before the end of the inspection
period.
..
II. Maintenance
M1
Conduct of Maintenance
M2
Maintenance and Material Condition of Facilities and Equipment
M2.1
Plant Material Condition
a.
- inspection Scope {62707. 37551)
The inspectors noted that several material condition issues and self-revealing equipment
failures during the inspection period required plant personnel to take prompt action. The
inspectors reviewed the failures to determine the effect on plant safety.
b.
Observations and Findings
Control Rod Drive (CRD) F1 Unexpected Insertion:
On December 6, 1997, during the performance of DOS 0500-03, "APRM Rod
Block and Scram," functional test, CRD 22-03 (F-1) unexpectedly scrammed to
position 00. The single-rod scram occurred approximately 15 seconds after
receiving an RPS Channel 'B' half scram signal. The licensee performed detailed
troubleshooting of the scram solenoid pilot valve (SSPV) and the RPS/SSPV
logic. The licensee concluded in Prompt Investigation Report No. 0005576052
that rod F-1 had not exhibited any specific characteristics of a degraded or failed
SSPV. Despite this, the licensee concluded that based on previous licensee and
industry experience the resultant single rod scram was consistent with that of a
degraded or failed SSPV. The licensee replaced the SSPV, successfully
performed scram functional testing, and placed rod F-1 back in service.
Unit 3 Torus Cooling Test Valve Failure to Operate:
On December 1, 1997, the torus cooling outboard test valve failed to operate on
demand, resulting in a potential for diversion of cooling flow from the reactor in the
event of a design basis accident. The licensee determined that the cause of the
15
.,
valve failure was due to a faulty auxiliary contact on the open contactor for motor
operated valve MOV 3-1501-388. The licensee replaced the auxiliary contact.
The faulty component was relatively new, and the licensee could not immediately
determine the reason for the. failure.
Unit 2 HPCI Full Isolation:
During performance of DIS 2300-03; "HPCI Low Pressure Isolation Channel
Functio_nal Test,n the HPCI system received a Group IV isolation and HPCI steam
line inboard and outboard primary containment isolation valves (2-2301-4 and 5)
went closed. The licensee could not determine the cause of isolation during
.inspection and. troubleshooting efforts, but replaced the relays that completed the
isolation logi~*(2391~02C and 2391.,.02D). The licensee sent the relays to the*
vendors for analysis.
2A Reactor Re.circulation Motor G.enerator Erratic Operation:
On December 6, 1997, the Unit 2 control room operators received a Recirc Pump
Mismatch alarm. The 2A recirculation pump ran back from approximately
49 percent pump speed to approximately 39 percent pump speed and appeared to
be running back up to the original speed when the unit operator locked-up the
scoop tube. Despite detailed testing and troubleshooting efforts, the licensee was
unable to determine the cause of the event. The licensee installed a temporary
recorder on the pump speed control circuitry to monitor for any future* possible
abnormalities.
3A Offgas Fire:
The 3A steam jet air ejector (SJAE) train did not work properly and was
susceptible to offgas fires. In the past, the licensee repaired and replaced
portions of the Unit 3 offgas system to restore the 3A train. However, during this
inspection period, the licensee experienced an offgas fire on December 6, 1997,
while the operating crew was adjusting the 3A recombiner booster/dilution steam
pressure using the manual bypass valve around the inoperative, isolated pressure
- control valve. The licensee fixed various leaks and attempted to fine tune the
booster jet pressure control valve (which was identified as oscillating). However,
on January 13, 1998, when the licensee again tried to bring the 3A train in
service, another offgas fire occurred. At the end of the inspection period, the
licensee was evaluating if modifications to the system were.necessary.
38 Reactor Water Clean Up (RWCU) Demineralizer High DIP:
The station experienced repeated difficulties in placing the 38 RWCU
demineralizer in service. The operators eonsistently received indication of a high
differential pressure across the bed when the unit was placed on line. The
operators had to perform multiple power reductions/ascensions to support taking
the demineralizer out of service for maintenance and returning it to service once
the maintenance efforts were complete. At the end of the inspection period, the
licensee was still engaged in troubleshooting efforts to restore the 38
demineralizer.
16
-- ----....:....:=--:::::.::::.:.::-=-- _--.:; _____ ....:.: ___ . _..:.:..:..._--=.. --:: __
--
-* --
.
- --* --* --- ------
- -
Unit 3 _ HPCI lnoperability:
On December 29, 1997, while placjng the HPCI system in standby, operators
started the GSLO pump to pump down the GSLO condenser hotwell. _The pump
failed to stop when it should have. This caused air entrainment in the pump
suction and air accumulation in the discharge pressure regulating valve sensing
line, Which then caused a reduction in the pump capacity. This event was similar
to a September 5 event Where the HPCI system GSLO condenser pump failed to
stop on the low level switch. Troubleshooting and inspection of the low level
switch in the GSLO condenser revealed no-abnormalities with the-low level
switch. .Further inspection efforts of the HPCI GSLO condenser revealed that
linkages' on the_ high level switch were not intact. This event w~s the third event
where_ the li~nsee's HPCI Sy$tem ,GSLO pump failed to operate properly due to
level switches malfunctioning (Ref. LERs 97-009/50-249 and 97-013/50-237).
Unit 2 Feedwater Level Control System:
The response of the feedwater level control (FWLC) system caused.reactor level
to overshoot following the automatic trip of the reactor on December 23, 1997.
The HPCI steam line would have flooded with water had not HPCI already been
isolated for troubleshooting. The FWLC system presented an additional operator
work-around due to the required operator compensatory actions following a
reactor trip on Unit 2.
c.
Conclusion
Collectively, the above issues represented challenges and distractions for- operators and
other plant staff. The issues especially represented a burden to operators who had either
to respond to the original event, or to take additional compensatory actions.
The inspectors were also concerned with the licensee's ability to resolve the issues
effectively and eliminate the deficiency. In some cases, the plant staff knew what
happened (and took appropriate short term corrective action) but did not know the root
cause of why the problem occurred.
M4
Maintenance Staff Knowledge and .Perfonnance
M4.1
(Unit 3) 125 V Battery Work
a.
Inspection Scope (62707)
The inspectors reviewed an error performed during battery testing that caused the station
to enter a two-hour dual shutdown LCO. The inspectors also reviewed the
documentation of the subsequent battery work.
17
L ------- --- ------- -----*-
__ :........: ___ -=::__..:__ __ -**- --
-
b.
Observation and Findings
Equipment Operation Outside of Procedure
On December 13, workers from the electrical maintenance department (EMO) were
authorized by operators to place the Unit 3 125 VDC charger on equalize-per
WR 970131667. At 0818, shortly after work started, the control room unexpectedly
received the 125 voe battery undervoltage and charger trouble alarms, followed by
battery high discharge alarms; The control room immediately dispatched a high-voltage
operator (HVO) to investigate. The HVO reported that the No. 3 125 voe charger had
tripped. The voltage on the batteries had rapidly dropped to 117 voe, then more slowly
to 115 VDC. The HVO aligned the 3A charger and restored voltage to normal* by 0822.
-*.
-
' * f ~
.
' *
~ :
. :
The electrical maintenance department (EMO) supervisor who was present at the .
batteries reported that he had manipulated the battery charger without a procedure in
hand.
The inspectors discussed the actions with the EMO supervisor. The supervisor stated his
woikers were having difficulty getting readings, so the supef'Visor commenced
troubleshooting. He went to the charger and noted that the charger's toggle switch was
in float. The supervisor believed it should have been in equalize, so he immediately
changed it. He did so without any procedures, without ariy other workers present, and
without verifying the response to his actions~ The toggle switch broke and caused the
charger to trip off, but since the supervisor did not wait to observe the expected
-
response, he did not realize this.
The* supervisor told his worker what he had done, and the workers in the room informed
the supervisor that his actions were incorrect. The HVO responding to the charger
trouble alarms then arrived at the seene, and the supervisor informed the HVO and
operations of the incorrect actions.
The inspector questioned the EMO supervisor about what, if any, distractions were
present. The supervisor stated that no distractions were present, and that he had just
been concentrating on troubleshooting so much that he had failed to maintain his role as
a supervisor, and that he had failed to adhere to the requirements for procedural use.
The EMO supervisor did not perform correctly when he manipulated equipment. The WR
used a "Category 1" procedure, and OAP 09-13 defined the level of use for a Category 1
procedure as "Continuous." Also, the supervisor should not have touched the equipment
in his over~ight role.
Dresden TS 6.8.A.1. required that written procedures be implemented covering the
activities recommended in Appendix A of RG 1.33, Revision 2, February 1978. The guide
recommended procedures covering maintenance work. Dresden Administrative
Procedure (OAP) 09-13, "Procedural Adherence," Rev. 06, required continuous use of a
Category 1 procedure. Contrary to this, on December 13, licensee personnel performed
work without continuously using the appropriate Category 1 procedure. As a
consequence, licensee personnel incorrectly manipulated equipment. The switch
failed, causing the 125 voe battery charger to trip. This was a violation
(VIO 50-237/97028-02 (DRP); 50-249/97028-02 (DRP)).
.
18
Immediate Licensee Response
The licensee immediately removed.the supervisor from duty and took other actions in
accordance with established management principles. The supervisor was tasked with
performing an Apparent Cause Evaluation" and performing other training and
remediation.
The licensee assigned another EMD supervisor to complete the initial work request..
Documentation Errors in Rework
The inspectors reviewed the completed work request (WR 970131.667) and noted several .
errors in how the work request was filled out.
First, it did not refer to the incorrect manipulation, and the work progression sign-off' had
been marked as "N/A.7 The use of the *sign-off was ;at the discretion of the supervisor, so
it was not a vi*o1ation to "NIA" the sign-off. When the battery was tripped and the work
was stopped on December 13, no one chose to fill in the work progression sign-off, so the
WR did not record why work was stopped, nor who stopped it.
Second, the inspectors noted that the Steps 1.1 and 1.2 of the procedure were initialed by
the second supervisor as "CM 12/13/97" for conditions met on December 13. The
inspector noted that the second supervisor did not have the assignment on December 13.
Also, Step 1.2. stated, "Document that Prerequisites are completed on Data Sheet 1, "'but
the second supervisor had marked Data Sheet 1 as "NIA 12/18/97." For.Step 1.2, the
conditions clearly were not met, and the "NIA" should not have been used.
The inspectors discussed the concerns with the second supervisor. The supervisor
stated that he had used "CM" for the first two steps after verifying that the prerequisites
were complete, and that the intent date of "12113/97" was to indicate the day the
conditions were met. The second supervisor stated that he did not review Data Sheet 1
. before marking Step 1.2 as "CM."
Procedure 09-13 also stated that "condition met (C/M) should be entered if an individual
finds that the requirements of a procedure step are already satisfied." The inspector
noted that the requirements of Procedure Step 1.2 were not met, in that the datil sheet
was not filled out.
The use of initials by each line in the procedures was required by the EMO
superintendent, but not by Dresden administrative procedures. Incorrectly using "CM"
was therefore not a violation. The use of "CM" led to a failure to document completion of
prerequisites on Data Sheet 1, and therefore the use of "CM" led to a failure to follow
procedures. No violation is being issued for this example of failure to follow procedures
because the circumstances would be expected to be enveloped by the violation being
issued for the initial failure to follow procedures on the battery .
19
- ---* __;::_____:.:....;_-=-------* ----- - -.--::.:.*-===----==:=::..:_-.::.:..-=.. ____ :__;.....:_
- -
.
- --
*- - *-- ... -- ---
- -** --
c.
Conclusions -
On December 13, work performed without referencing a required procedure, combined
with material condition, tripped off the Unit 3 charger and placed both units unexpectedly.
into a two-hour.LCOi *'Fhe subsequent follow up work was not performed in accordance
with station administrative procedures. Specifically, the "CM" sign off was used when a
condition had not been met.
Section 08.1 of Dresden IRS0-237/97024; 50-249/97024 documented a case where the
incorrect. use of "CM" led to the high pressure. coolant injection (HPCI} system being
unavailable. The inspectors were concerned that the second example of "CM" being
used incorrectly occurred. ,
MB
Miscellaneous Maintenance Issues.
MB.1
Tracking of Rework (62707)
In Section EB.2 .of Inspection Report 97019, the inspectors discussed rework tracking.
The inspectors noted that the licensee's methodology was recently changed. To perform
an assessment of rework, the inspectors kept a list of seleCted items that were known to
be rework, then checked if the items were captured into rework tracking. The results
revealed that one of five items was not captured. The licensee was investigating why it
wasn't tracked. Based on the large number of items that were captured, though not .
specifically tracked by the inspectors, the inspectors concluded that one identified
anomaly did not invalidate the total tracking of rework, but was cause for the inspectors to
expand the sample size.
The inspectors also noted that the 125 VDC battery work discussed in Section M4.1 of
this report was not originally identified as rework. The inspectors noted that assigning a
second supervisor to become familiar with and execute work was unnecessary repetition
of work caused by inadequate performance of the original task, and was therefore rework.
The inspectors discussed this observation with the licensee's rework coordinator.
Ill. Engineering
E1
Conduct of Engineering
E1 .1
Root Cause Investigation of Unit 2 Automatic Reactor Trip
a.
Inspection Scope (37551)
The inspectors discussed the event's root cause with the licensee's investigation team,
interviewed individual engineers associated with the investigation, and attended the
PORC meeting where the team presented their results to licensee management.
b.
Observations and Findings
The event response team's charter directed the team to determine the root cause of the
event, evaluate the plant's response, and evaluate the suit~bility for restart .
20
The inspectors determined that the team was self-critical in evaluating the root cause and
did not hesitate to state that the utility's past actions in response to the GE SIL were not
consistent with current policies and practices. After the scram, the station experienced
several more LPRM half scrams. The half scrams occurred on Unit 3 and involved
detectors that had not previously exhibited spiking. For example, Unit 3 operators
received an unexpected Channel "8" half seram on January 5, 1998, due to a spike on
. LPRM 10-24-49. On January 7, 1998, a Channel "B" half scram occurred due to spiking
on LPRM 32-57 A. . The station re~reviewed the SIL and implemented new actions to *
address LPRM spiking. Before the end of the inspection period, the station worked
around the clock to perform the current-voltage curves for all LPRMs that fed the APRMs.
Results of the tests were used to determine the* need for *capacitance discharge (!'whisker
bums") tests on LPRMs, that exhibited the potential for spiking. At-the end of the
inspection period, the licensee was working on a schedule to complete testing on the
remaining LPRMs, as well as future long term actions to address the issue of LPRM
spiking.
The inspectors also reviewed. the team's assessment of the plant equipment response
following the scram. Except for the FWLC system response (discussed in Section E1.2),
plant equipment functioned as expected with only minor anomalies noted. The team
appropriately dispositioned these items.
c.
Conclusion
The.inspectors concluded that the licensee's root cause team performed a thorough: *
investigation of the event, the root cause, and equipment response following the reactor
scram. The licensee's team concluded that the root cause of the event was the failure to
perform the actions identified in GE SIL 500 regarding LPRM spiking. The inspectors
agreed with this conclusion.
E1 .2
Response of the FWLC System Following Unit 2 Automatic Reactor Trip
a.
Inspection Scope (37551)
The response of the FWLC system caused reactor vessel water l~vel to overshoot
following the scram. The overshoot would have flooded the HPCI steam line had HPCI *
not been already isolated for troubleshooting purposes. The inspectors reviewed the
licensee's investigation into the issue and proposed resolution prior to restart.
b.
Observations and Findings
The inspectors noted expected reactor vessel water level response immediately following
the reactor trip. Vessel level dropped to approximately -4" (normal level is +30") due to
shrink; this response was normal and expected. However, upon level recovery reactor
water level rose above +60" and was about 3" above the lower level of the HPCI steam
lines. Per design, at this level the HPCI system would have tripped, but not physically
isolated. The HPCI steam line would have flooded had the system not already been
isolated for maintenance purposes. The configuration of the steam supply piping was
such that the water would have remained trapped in the line after vessel level dropped
back below the trip setpoint.
21
--.:--. ___ __:. ... _:..:::.: __ -*. -------*--* **---
The response of the FWLC system was as designed. However, in *this case, appropriate
equipment response and appropriate post-trip operator response did not prevent a
reactor water level overshoot and potential filling of the HPCI steam lines. Upon receipt
of reactor scram and level drop from 30" (normal) to +2", "setpoint setdown" occurs.
During setpoint setdown, the FWLC system changes the required level setpoint and ramp
rate of the feedwater regulating valve (FRV) so that upon level recovery, water level does
not overshoot and raise the vessel level too high. However, the receipt of "bad quality"
data introduces a time delay in the initiation of setpoint setdown. Bad quality data was
defined as data that was outside the range of the narrow range level indicator scale; on
Unit 2 the narrow range level indicators read from O" to +60". Therefore, when reactor
- water level dropped to approximately -4", the FWLC system logic received "bad quality"
data and a time delay (between three to seven.seconds) occurred before setpoint
setdown was initiated. This time delay was sufficient to prevent th*e FWLC system from
responding quickly enough to prevent a level overshoot. Since the FWLC. system and
post scram water level drop responses were normal and as .expected, all Unit 2 scrams
from full power have the potential to result in a level.overshoot with the corresponding
effect of putting water.in the HPCI steam line.
Several differences existed between the Unit 2 and Unit 3 FWLC systems that would
have prevented a level overshoot on Unit 3. On Unit 3, the FWLC system was set up in
three element control, whereas on Unit 2 the system was set up in single element control.
The Unit 3 narrow. range level transmitters are scaled from -60" to +60" which would .
- prevent the receipt of "bad quality" level data following a reactor trip. Also on Unit 3,
when a reactor scram signal is received, the system immediately goes into setpoint *
setdown that aids in vessel level recovery efforts. The licensee's root cause investigation
team concluded that these factors would prevent a similar level overshoot from occurring
- on Unit 3. The inspectors reviewed the team's results and did not disagree with this
conclusion.
The licensee's resolution to address the response of the FWLC system included both
long term and short term corrective actions. The licensee.plans to modify the Unit 2
FWLC system during the upcoming refueling outage to make it similar to the system on
Unit 3. The licensee planned to rely on operator compensatory actions for the immediate
short term corrective actions. Upon receipt of a scram signal, operators would be
required to analyze feedwater system response and trip off one running feedwater pump
once level turned and started to rise following a reactor.trip. In addition to tripping a
reactor feedwater pump, operators would also be required to take manual control of the
feedwater regulating valves to ensure that level would not overshoot and flood steam
lines. The inspectors reviewed post event charts and concluded that these actions would
be required within the first ten seconds following the reactor scram. The inspectors were
concerned that the station was relying on operator action to prevent water intrusion into
HPCI steam lines following a reactor scram.
c.
Conclusion
The FWLC system response presented a potential challenge to the operators following
the reactor scram. The compensatory actions that operators are required to take
following a scram on Unit 2 constitute an additional operator work-around. Pending
permanent resolution of the Unit 2 FWLC system issues, the station was relying on
operator intervention following a scram to prevent water intrusion into HPCI steam lines.
22
q
_.,_ __ ::_;_ ___ ;.:. -* -- --* . .:
E2
Engineering Support of Facilities and Equipment
E2.1
(Units 2. 3) Operability Evaluations
a.
Inspection Scope (71707. 37551)
The inspectors reviewed recent operability evaluations. Compliance with the
requirements of OAP 07-31, Rev: 07, "Operability Determinations," and the impact on
plant operations were considered. A detailed evaluation of any engineering calculations
used to make the final determinations was not performed.
Operability determinations revi.ewed included: 97-105
97-107
97.,.108
97-'109 97-110
"Personnel Air Lock Equalizing Valve - Concern Identified with Material of
Seals"
"Concern of Potential Vortexing in CSTs"
- "Failure of the Unit 2 125 VDC Alternate Battery to Maintain 105 Voe at
Bus under D.B.A. Condition"
"HPCI Small Bore Lines Do Not Meet UFSAR Design Criteria" ,
"Reactor Building Superstructure Seismic Requirements"
"Post-LOCA Reactor Building Temperatures Beyond UFSAR Limit of
.104°F"
- * *
b.
Observations and *Findings
- The operability .determinations reached the following conclusions: 97-105 -
97-107 -
97-108 -
Operable, but degraded. The concern was that if exposed to design basis
accident (OBA) conditions *of 334°F and 63 psia, the airlock equalizing
valves' seats may soften and cause a leak. Based on calculations, the
licensee determined that the outer door would probably not be affected.
Corrective action required was to schedule replacement of equalizing
valves on all airlock doors during D2R15 and 03R15.
Operable, but degraded. The concern was that vortexing may occur and
cause air entrainment in HPCI system. Calculations showed that aligning
the HPCI system suctions to both condensate storage tanks (CSTs) would
be sufficient to prevent vortexing. Also, *postulated breaks inside
containment resulted in a swap to torus suction before reaching CST
vortexing. Postulated breaks outside of containment challenged the CST*
supply. Corrective actions included realigning the CST supplies and
evaluating changing the CST low level switches.
No concern exists. The concern was that the 1000-foot cable-run from the
- Unit 2 125 voe alternate battery to the *distribution panel resulted in a
- previously unconsidered voltage drop. Engineering concluded that based
on latest actual battery surveillance tests and calculations, no concern
exists. Although the battery currently met its requirements, engineering
added a request to add additional cells to address expected margin
- reduction due to aging.
23
97-109-
97-110 -
97-112 -
Operable but degraded. The concern was that several Unit 2 and Unit 3
HPCI system lines were inadequately supported to meet design basis
requirements. The lines included drain pot discharge lines and the gland
seal condenser discharge line to the torus. The licensee concluded that
the lines were operable because the lines met the requirements of ComEd
Nuclear Engineering Standard (NES) No. NES-MS-03.2, "Evaluation of
Discrepant Piping and Support Systems," Rev. 0, and the stresses were
less than twice yield stress. Corrective action planned was to reevaluate
the piping and determine how to restore it to design limits.
Operable but degraded. The concern was that calculations for the reactor
building crane being loaded during ,a safe shutdown earthquake (SSE}
were never performed. Engineering concluded that the 125-ton crane* *
must be limited to 12.5 tons based. on determining that the additional
12.5 tons was insignificant compared with the mass of the crane and
trolley ... For corrective actions, the licensee planned to determine what
- commitments existed regarding the required analysis, and, if ne.cessary,
perform a detailed analysis for SSE while the maximum crane load was
lifted.*
Operable but degraded. The concern was that recent calculations showed
- that the post.;.accident temperatures .of. the* reactor building were generally
from 120 to 160°F, whereas the UF~AR limit was 104°F. Engineering
concluded that as long as the outside air 30-day average temperature *
remained below 44 ° F or 68 ° F (the temperature depended on other
compensatory actions), then the safety-related equipment will function.
Corrective actions included direction to shut down the non-accident unit
and other actions to increa*se reactor building cooling, and re-calculation of
temperatures.
Operability determinations were governed by OAP 07-31, Rev. 07, "Operability
Determinations." The performance of the reviewed operability determinations met
DAP 07-31. The engineering used to make operability conciusion.s appeared reasonable.
The operability determinations reflect the status of the design basis at Dresden. The
licensee did not have supporting calculations for some systems. As the calculations are
performed, some calculations raise operability concerns. T1:'1e licensee was dealing with
these issues appropriately, even when the compensatory or corrective actions were
significant.
The licensee used the Dresden Engineering Assurance Group (DEAG} to review the
operability determinations. No PIFs or.rework resulted from the DEAG's reviews of
determinations97-107, 97-109, and 97-110-. The DEAG did comment on determination 97-110 regarding the need to provide a complete basis for corrective actions. The DAEG
review identified "major problems" with the basis for operability and the ci:>rrective actions
related to determination 97-106, "Lack of vent valves on CCSW suction headers,
24
J *
.
-
-
- -*- .. - *- *-*-*-*-*
-
repriming, and leakage criteria,"*and generated PIF No. 01997-08537. Note that 97-106
was not' discussed in this report because it was reviewed in Dresden Inspection Report
No. 50-237/97021; 59-249/97021. The DEAG did not review 97-108 because no concern
was identified.
c.
Conclusions
The operability evaluations.appeared to meet licensee requirements. The evaluations
were reasonable and provided adequate bases for the conclusions.
The scope of the evaluations reflected the current state of the design of Dresden Station,
in that several evaluations resulted from .performing calculations to replace missing or
never-performed calculations. * ;
-
E4
Engineering Staff Knowledge and Performance
E4.1
(Unit 3) Engineering Evaluation of Batteries
a.
Inspection Scope (37551)
.
b.
The inspectors assessed the engineering response to the December 13 event in which
the Unit 3 125 V battery charger was inadvertently tripped.
Observations and Findings
As discussed in Section M4.1, the Unit 3 battery charger was tripped inadvertently during
maintenance. The control room logs and the PIFs written about the event recorded that
the voltage on the batteries had dropped to 115 Voe. The inspectors asked the licensee
if 115 voe was the expected value for the 125 voe batteries.
The licensee had not previously considered the battery's response, because the battery
remained above.105 voe and passed its last surveillance test.
However, a few days after the initial question, the licensee compared the performance
with a vendor-supplied graph of the discharge characteristics of the battery and
concluded that 115 VDC was the expected voltage given the loads on the battery.
c.
Conclusions
The licensee had not performed a detailed reviewed the battery's performance. However:
the licensee eventually was able to show that the battery behaved as it should have.
25
I
..
---*-----=*-** - --~--'
IV. Plant Support Areas
Radiological Protection and Chemistry (RP&C) ..
R4
Staff Knowledge and Performance In RP&C
R4.1
(Units 2. 3) Treatment of Contam_inated Area Boundaries
-a.
lnspe~ion Scope (71750)
The inspectors observed worker performance around contamination area boundaries with
. respect to. the. licensee's. RP&C procedures.
b.
Observations and Findings
Workers generally respected the boundaries placed to control contaminated areas. The
placement of step-off pads and boundaries provided sufficient room for worker:s to work.
Equipment was generally staged or stored correctly and did not present a contamination
control hazard.
On December 19, the inspectors identified that a white hose that:crossed a step-off*pad
in the Unit 2 east torus basement area was not secured at the contamination boundary.
The inspectors informed the licensee and the licensee secured the hose. The licensee
issued PIF 01977-08731 to track the issue, and determined that the hose had originally
been placed to support work on the east low pressure coolant injection "(LPCI) system
comer room sumps. The work was completed, but the 2A east comer room sump was
still not working property since repairs, so the hose was staged as part of a contingency in
case the 28 east comer room sump failed. The licensee also found that the equipment in
use tag on the hose was not up to date.
A search of PIFs revealed four other examples of untaped or improperty marked hoses
(ref. PIFs No. 01997-07363, -07396, -08390, 08607). This was a marked increase from
the number of radiation control PIFs written during the previous months.
Procedure OAP 03-07, Rev. 09, "Control of the serviee air and domestic water systems
and hoses for general station use," Step F.3.e. stated, "IF a RED, WHITE, OR CLEAR
hose must cross the boundary between a contaminated area AND a non-contaminated
area, THEN the hose must be secured at the boundary using Radioactive Materials
Tape." The failure to secure the hose was a violation of OAP 03-07
(VIO 50-237/97028-03 (DRP); 5~-249/97028-03 (DRP)).
The inspectors also saw other examples of poor control of contamination boundaries.
For example, a worker mopping a contaminated area allowed the bucket to cross the
contamination boundary, and other workers placed equipment and work instructions on
contamination boundaries .
26
- ,
l "
c.
-
--~
- --*-*---* ---- ---
. . ' ~-
Conclusions
__ :*_::_ ------ - *--- ... :.: - . --
The setup and control of contaminated areas and work in contaminated areas were
usually correct.
The inspectors identified one example of an improperly secured hose that crossed a
contamination boundary.* The hose had been staged in response to a poorly performing
sump pump.
The safety significance of these' issues was minor. However, the inspectors were
concerned with the lack of attention to detail exhibited by the plant staff to radiation
controls, and that the PIF record showed this to b~ an emergent trend.
V. Management Meetings *
X 1
Exit Meeting Summary
The inspectors presented the inspection results to members of licensee management at
the conclusion of the inspection on January 12, 1998. The licensee acknowledged the
findings presented. The inspectors asked the licensee whether any materials examined
during the inspection should be considered proprietary. No proprietary information was
identified.
X3
Management Meeting Summary
On December 22, the NRC Region Ill Regional Administrator and the Director of the
Division of Reactor Projects visited the site, met with senior licensee management, and
discussed current licensee performance. *
27
' ..
\\j .
! * '
.
' * *
- ~'
'PARTIAL LISTOF PERSONS CONTACTED
G. Abrel, ComEd NRC Coordinator
D. Ambler, Regulatory Assurance Supervisor (Actirig)
- P. Bernice, Ops Staff *
T. Bezouska, Site Vice President Staff Assistant
- .
R. Fisher, Maintenance Manager *
,
R. Freeman, Site Engineering Manager.
M. Heffley, Site Vice President
C. Howland, Radiation Protection Manager
L Jordan, Training *Supervisor (Actirig) *
w. Liscomb, Site Vi-de President'Staff * ;
- * *
P. Stafford, Station Manager (Former Outage/Work Control Manager)
D. Willis, EMO s*uperintendent
' *
i
- *
D. Winchester, Q&SA Manager
28
L_ ______ _ _:_ __ --------=--- _* _ _:.. -::.::. ___ *.:.._::__:_:__.*-*::.....:....:.
LIST OF INSPECTION .PROCEDURES USED
Inspection Module: 71707.
Operational Safety Verification
Inspection Module: 83822
Radiation Protection
Inspection Module: 62707
Maintenari6e
Inspection Module: 61726 * Surveillance Observations
Inspection Module: 40500
Effectiveness of Licensee Controls in Identifying, Resolving, and
Preventing Problems
Inspection Module: 93702
Prompt On-Site Response t9 Events at Operating Power Reactors
Inspection Module: 37551
On-Site Engineering
LIST OF ITEMS OPENED, CLOSED, AND DISCUSSED
Opened
50-237; 249/97028-01
50-237; 249/97028-02
50-237; 249/97028-03
Closed
None
Discussed*
None
failure to follow PIF process
failure to use procedures
failure to follow radiation protection procedures
29
)
DAN
OAP
DATR
. DEOP
DOA
EOPR
IFI
IR
ISEG
ISi
LCO
LER
LPRM.
NRC
NSO
NSWP
01 oos
RUF SAR
Q&SA
TS
us
.
. .
- *---------
---. ... -- **--------- ----*
LIST OF ACRONYMS USED
Dresden Annunciator Procedure
Dresden Administrative Procedure
Dresden Administrative Technical Requirement
Dresden Emergency Operating Procedure
Dre.sden General Abnormal Procedure
Dresden System Operating Abnormal Procedure
Dresden System Operating Procedure
Engineering Operational Planning/Troubleshooting Report.
Feedwater Level Control*
High Pressure Coolant Injection
Inspection Followup Item
Individual Plant Evaluation
Inspection Report
Independent Site Engineering Group
lnservice Inspection
Limiting Condition for Operation
- Licensee Event Report
Low Pressure Coolant Injection
Local Power Range Monitor
Non-Cited Violation
Non-licensed Operator
Nuclear Regulatory .Commission .
Office of Nuclear Reactor Regulation
Nuclear Station Operator
Nuclear Station Work Procedure
- Office of Enforcement
Office of Investigations
Out-of-Service
Problem Identification Form
Plant Operations Review Committee
Revised Updated Final Safety Analysis Report
Quality and Safety Assessment
Quality Control *
Technical Specification
Unit Supervisor
Violation
Work Execution Center