IR 05000010/1996013
| ML20133C810 | |
| Person / Time | |
|---|---|
| Site: | Dresden  |
| Issue date: | 12/31/1996 |
| From: | NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION III) |
| To: | |
| Shared Package | |
| ML20133C791 | List: |
| References | |
| 50-010-96-13, 50-10-96-13, 50-237-96-13, 50-249-96-13, NUDOCS 9701080102 | |
| Download: ML20133C810 (24) | |
Text
- .e- -
42 w m_ = A- --
A+r-5 -
- - - 4 ES &- A*wJ'r aA-aA+4 4 -M-M +45 A *i-MS---' A#--- 44 3--~>-4 4 J u d& +% M 4as * +-@@
,
1-
.
4
,
,
i
'
U.S. NUCLEAR REGULATORY COMMISSION REGION lil i,
- ,
.
,
Report No: 50-010/96013; 50-237/96013; 50-249/96013 i
j Licensee: Commonwealth Edison Company
l Facility: Dresden Nuclear Station Units 1,2 and 3 j Location: Opus West lll
- 1400 Opus Place - Suite 300 2 Downers Grove, IL 60515
<
Dates: September 1 through October 18,1996
)
t i
Inspectors: C. Vanderniet, Senior Resident Inspector T. Kozak, acting Chief, Plant Support Branch 2 4 J. Hopkins, Project Engineer l J. Hansen, Resident inspector D. Roth, Resident inspector j C. Settles, inspector, Illinois Department of d Nuclear Safety I Approved By: P. L. Hiland, Chief l Reactor Projects Branch 1 i
9701080102 961231 PDR ADOCK 05000010 G PDR
_ _ _ _ _ _ __ _ _ _ _ _ . . ~ . . _ _ _ . -. . _ .
l
i
.
.
EXECUTIVE SUMMARY
! I l Dresden Nuclear Station Units 1,2 and 3 '
NRC Inspection Report 50-10/96013; 50-237/96013; 50-249/96013 l This routine resident inspection included assessments of licensee operational, maintenance, {
engineering, and plant support activities. These inspections were conducted through direct l j- observations in the field, documentation reviews, and discussions with station personne !
l Operations
- The facility was operated in a safe manner with good communication. Minor discrepancies continue to occur and operations personnel need to remain focused on attention-to-detail type issues (Section 01.1).
- Unit 3 Drywell closeout inspection. Two items identified by the inspector delayed Unit 3 startup (Section 02.1).
- Operator properly respond to the annual "Grizzard Shad" fish migration which challenged the facility's service water system (Section 04.1).
Maintenance 1
)
- Restorative work for leaking control rod drive valve was well planned and execute i However, original job planning and procurement weaknesses were identified l (Section M2.1). I l
- Unit 3 high pressure coolant injection system material condition and configuration ,
control problems effect surveillance performance (Section M2.2). !
- Station's backlog of outstanding work and action requests recoded to remove most ;
of the planned corrective work (Section M3.1). l l
Enaineerina e Unit 3 local leak rate testing of inboard and outboard isolation valves resulted in the inability to maintain containment leakage within allowable limits (Section E2.1).
- Design Engineering response to isolation condenser support and feedwater anchor l issues was good (Section E2.2) l
- Emergency diesel generator fuel oil transfer pump discharge gauge found over-ranged for the third time. Poor root cause identification contributed to repetitive maintenance (Section E2.3).
- Unit 2/3 emergency diesel generator ventilation fan Unit 2 power supply breaker ,
troubleshooting was poor (Section E2.4). I
l l
_
_ _ __ ._.m_ .____ _ _ _ _ . . . _ - . _ _ _ . . _ . _ . _ _ _ _ _ _ _ . _ . . _ .__
I
- i l
'
.
'
!
i .. *
The control room ventilation system was unable to maintain the required 1/8 inch l
?
water positive pressure inside the control room envelop (Section E2.5).
I
- Plant Suocort J
The contaminated material control program improved since the discovery of
-
approximately 450 uncontrolled, contaminated items in April 1995. However, the corrective acuans for this problem have not all been fully implemented and i continued improvement in this area is needed
- (Section R1.1).
I j *
Repairs to the station's east fire-main header did not recognize slip joint piping fit which rr,sulted in flooding and increased work scope (Section F2.1).
.
e
!
i !
! l
J
1 i
i
!
l
_
'
l
.
REPORT DETAILS Summary of Plant Status Unit 2 was in a power ascension at the beginning of this report period and achieved effective full power (98 percent) on September 23. The unit continued at full power for the remainder of the report perio l Unit 3 began this period in Cold Shutdown as the licensee completed work on 4kV breakers and cubicles. Criticality was achieved on September 22, and effective full power (83 percent) was reached shortly thereafter. The unit was coasting down in power in preparation for the 1997 refueling outage (D3R14).
l. OPER ATIONS 01 Conduct of Operations
0 Routine Observations (Units 2 and 3) ' Inspection Scone (71707)
j Routine day-to-day facility operations were observed and the inspectors attended various licensee planning and scheduling meetings. Additionally, the Unit 3 Startup Plan was reviewed and direct observations of control room and plant activities were j performed on a shiftly basis during the ensuing start u ' Observations and Findinas Generally, the facility was operated in a conservative manner maintaining the reactors in a safe condition consistent with the licenses. The Operations Department did a good job controlling the start of station work by using the work j execution center (WEC) to screen all work before authorization by the unit '
supervisor (US). This minimized the number of non-operating personnel in the control room and decreased the administrative burden on the US. However, some attention-to-detail discrepancies were identified in the control room during this perio On a few occasions the inspectors noted a decline in the decorum and use of 3-way communications in the control room. This observation was discussed with licensee management. The licensee had also noted the decline in performance and were in the process of addressing it. In addition, several annunciators were in the alarm condition which was not consistent with current plant conditions. Further inspection identified that the on-shift nuclear station operator (NSO) had not reset the alarms to clear the annunciators. On all occasions the NSO knew why the alarm had come in and cleared it immediatel On several occasions, the inspectors discovered one or more average power range monitor high level alarms illuminated on the main control board backpanel. These
. -. - -- - _ - - - -._ ~- . .. -
i
. i
. altrm conditions were brought to the attention of the NSO and were also immediately reset. Note that the presence of the backpanel alarm would not have l prevented the main control board frontpanel annunciator from alarming. After a review of documentation and discussions with cognizant personnel, it was
,
determined that the alarm conditions were the result of a "Ln-in" function of new local power range monitors (LPRMs). The alarm conditions were also of such short :
duration that the signal spikes from LPRMs caused the more responsive back-panel ;
instrumentation to alarm without the corresponding audible and visible front-panel alarm. The NSOs indicated that there was no formal guidance on what to do when a backpanel alarm was discovered without a corresponding front-panel alarm. This was discussed with licensee management and was being reviewed by the licensee for corrective actions at the end of the report period. Since discussing these issues with the licensee, the inspectors have noted an increased awareness of these alarms by the NSO {
1 Conclusions The facility was operated in a safe manner with good communications. Minor discrepancies and attention-to-detail type issues were observe O2 Operational Status of Facilities and Equipment 02.1 Drvwell Closecut (Unit 3) insnection Scone (71707)
On September 20, the inspectors performed a "closecut" inspection of the Unit 3 drywell. This included a detailed tour of the drywell inspecting for operability, material condition, insulation, and housekeeping deficiencie Observation and Findinos Preparations and briefings for the drywell entry were detailed and well controlled by )
the Radiation Protection (RP) Department. Specific confined space training was '
performed and documented to ensure personnel entering the drywell were properly qualified. No deficiencies were noted with the radiation work permit or any other activities involving drywell entr l During the tour, the inspectors identified two condensing chambers for the isolation condenser high steam flow isolation signal that were covered with insulation. The chambers had been inadvertently insulated during the outage. Condensing chambers need to be exposed to ambient condition so steam can readily condense and fill the lines of the attached instrumentation. By insulating the condensing chambers, the condensation of the steam was inhibited and could have effected the proper operation of the instrumentation. The insulation that was used was not the loose fibrous material that had oeen found in the past but was of a blanket type, appropriate for use in the drywell. The identified insulation was removed and the inspectors had no further concerns with this issu rp p
.
.
The inspectors identified a disconnected spring hanger during the tour. The hanger supported a reactor building closed cooling water (RBCCW)line going to the 3B reactor recirculation pump. The licensee made the necessary repairs to the hanger
_ prior to startup, correcting all concerns vwith this issu There were several smaller housekeeping items such as tie-wraps, screws, and pieces of tape that were found by the inspectors which the licensee promptly retrieved and disposed of completing the closecut inspectio Conclusions The RP Department did a good job of controlling entry to the Unit 3 drywell. The drywell was generally clean, orderly, and contained no significant loose materia Although the actions necessary to correct the concerns noted above delayed startup, the drywell was in better condition than on previous " closeout" irsoection Operator Knowledge and Performance 04.1 Service Water Strainer Hiah Pressure Due To Fish Miaration (Units 2 and 3) Inspection Scope (71707)
Operator actions were observed in the control room and the inspeceirs toured the cribhouse and portions of the affected systems. A review of the licensee ~c prompt investigation of the event was also performe Observations and Findinas Beginning September 27, the annual week-long "Grizzard Shad" run occurred at the station. The shad are small sardine-like 1 5 that migrate in large schools up the cooling channels in the fall. Last year's shaC run resulted in a manual reactor trip of Unit 3 due to an anticipated loss of circulating water. This year the traveling screens were able to handle most of the shad; however, some carry-over of fish occurred due to an inadequate screen wash spray patter On September 30, the NSO noticed a decrease in the service water system pressure. A non-licensed operator (NLO) dispatched to the cribhouse found al! three service water stainer differential pressure gauges pegged upscale (greater than 5 psid). The shad carry-over had entered the service water system, plugging the strainers and causing a reduction in system pressure. The strainers automatic backwash function was in the manual mode instead of the automatic mode and was not able to perform an automatic backwash of the strainers. At about 0600 the licensee initiated a power reduction on both units while simultaneously commencing manual backwash operations. By 1100 the strainers differential pressure and the service water system pressure were restored to normal.
6
_
,
.
.
The licensee initiated a prompt investigation into the event and identified a number of deficiencies. The inspectors reviewed the licensee's investigation and concluded !
it was thorough. Several immediate and long-term corrective action were )
established which appeared adequate to mitigate future occurrences of the annual l'
shad run Conclusion Work completed on the travelling screens to enhance the performance after last year's shad run was successful. With the travelling screens working properly other deficiencies were now evident including:
Service water strainers backwash function in manual instead of automatic mode (an operator workaround).
t Inadequate travelling screen wash nozzle spray pattern allowing for some fish carry-over which entered the service water syste * The lack of an abnormal operating procedure for dealing with the annual y shad run, i I
Operators reacted in a controlled and conservative manner which minimized the transient on the facility. The prompt investigation was thorough and corrective
'
action appeared adequate to reduce future vulnerability to the annual shad ru Miscellaneous Operations issues (92701)
08.1 (Closed) IFl 50-237:249/96009-02: Posted Procedures. On August 2,1996, the inspectors identified a " posted procedure" in the power block which was not the most current revision. The procedure had been revised on July 29, and the copies in the main control room, WEC, and other required areas were the correct revisio Subsequently, the inspectors reviewed how posted procedures were controlled, it i was noted that Dresden Administrative Procedure (DAP) 09-02, " Procedure Revision and Processing," Revisions 41 and 44, had not required posted procedures to be immediately updated. However, DAP 09-02 had required immediate update of main control room, WEC, and other non-posted procedures. Procedure DAP 09-01,
" Station Procedures," Revisions 28 and 29, described how posted procedures were updated, but had not specified timeliness requirements. The licensee's corrective actions included changing the distribution list in procedure DAP 09-01 to add
" posted procedures."
The inspectors reviewed the parformance improvement form (PlF) data base for the past year and only identified a couple of other examples of incorrect revisions of
" posted procedures." The licensee's corre::tive actions appeared adequate to prevent recurrence. This item is close .
II. Maintenance M1 Conduct of Surveillance M 1,1 Operators Performance Durina Execution of Surveillance (Units 2 and 3) Inspection Scope (61726/71707)
Operator performance was observed and documentation reviewed for the following surveillance tests:
Dresden Operating Surveillance (DOS) 1400-01, " Core Spray System Pump Test with Torus Available," Revision 2 * DOS 1400-02, " Core Spray System Valve Operability Check," Revision 1 *
DOS 1400-04, " Core Spray Check Valves Inservice Test (IST) and Piping Flush," Revision 1 * DOS 1500-02," Containment Cooling Service Water Pump Test and Inservice Test (IST)," Revision 2 *
DOS 1500-05, " Low Pressure Coolant injection (LPCI) System Quarterly Flow Rate Test," Revision 1 * DOS 150010, "LPCI System Pump Operability Test with Torus Available and in-Service Test (IST) Program," Revision 2 * DOS 6600-01, " Diesel Generator Surveillance Tests," Revision 4 The inspectors also reviewed the diesel generator surveillance tests performed in using DOS 6600-01 that were completed due to a minor fire in the motor control center for the Unit 2/3 diesel generator ventilation fan (see Paragraph E4.1), Observations and Findinos For the surveillances observed, operators performing the surveillances followed the appropriate procedures, completed all tasks, and were knowledgeable of Technical Specification requirements. Good communication practices were used throughout testing and operators responded correctly to discrepant conditions which were encountered. For. example, during the performance of DOS 6600-01, " Diesel Generator Surveillance Tests," on October 15, the operators correctly declared Unit 3 emergency diesel generator (EDG) inoperable due to excessive coolant discovered in the EDG's air bo Conclusions The operators correctly followed the surveillance procedures. The operators were knowledgeable of the systems and took appropriate actions when unexpected results occurred. No significant deficiencies were identifie _ . . _ . __ .. _ . _ _ _ _ _ _ -- _ _ _ _ - _ _ _ _ __ __ . _ _ _ _ _
.
t
'
. .
!
'
i
-
M2 Maintenance Material Condition of Facliity and Equipment l M2.1 Reolacement and Reoair of the 3A Control Rod Drive (CRD) Pumo Discharoe Valve l (Unit 3)
I Insoection Scope (62707)
The inspectors observed in-field activities associated with the rep'acement of the 3A CRD discharge valve and subsequent rework. A review of ti e work package and associated documentation was also performe l l Observations and Findinos )
l The 3A CRD pump inboard seal exhibited leakage at an increasing rate and a l
decision was made to isolate the pump and repair the seal. The 3A CRD pump discharge valve (3-301-1 A) would not provide an isolation boundary and a decision was made to replace the non-safety related valve. On September 25, about 1-week after the repairs, an operator noticed a pin-hole leak in the body of the replaced valv Inspection of the defective valve body showed a sand inclusion in the body which communicated with the outer valve body wall through a single p'n-hole fla Radiographs showed that the sand inclusion was fed by severe; tiny fissures extending from the inner valve body wall to the inclusion. The radiographs also showed additional fishhook shaped fissures on either side of the sand inclusion which made several repair options impossibl The initial' corrective actions involved an attempt to establish a freeze-seal boundary l to allow for valve replacement. This failed and the licensee decided to encapsulate the entire discharge valve. The encapsulation vessel exceeded the pressure rating of the system and after testing allowed continued facility operation. Radiographs were performed on all welds and a separate support bracket assembly was designed and installed to carry the extra weight. There were seven PlFs written to document the breakdowns in engineering, maintenance, work control, procurement, and !
quality control for the original replacemen Conclusions The licensee's actions following the identification of the pin-hole leak were conservative and well planned. Those actions recognized the work processes which failed to prevent the use of the defective valve body. The inspectors determined that the problems were limited to non-safety related work items. The interim repair was well engineered and a permanent repair will be performed at the earliest opportunit .-. . - . _ - . - .
i
.
M2.2 Hiah Pressure Coolant Iniection (HPCI) System Problems (Unit 3) Inspection Scoce (61726/71707)
Portions of the HPCI system operability surveillance wera N ad from the control room and a review of the licensee's troub! shooting efforts were completed by the inspector Observations and Findinas During the performance of the monthly Unit 3 HPCI system operability surveillance on October 2, one of the four HPCI room temperature monitors alarmed in the control room. Loca room temperature readings for that monitor indicated a temperature of 122 F had been reached. The monitor that alarmed was located near the turbine steam chest and the instrument probe was discovered to be in contact with the turbine steam chest insulation. The other three room temperature sensors were reading about 95 F which made the general room ambient temperature well below the 120 F alarm setting. Because the procedure required the HPCI system be declared inoperable when any one of these sensors alarmed, the licensee declared HPCI inoperabl The licensee's evaluation determined that the intent of the room temperature monitors was to determine the general room ambient temperature not a lochized temperature. Also, the location of the alarming monitor, in contact with t insulation, was not a true measure of ambie:it room air temperature. Therefore, the procedure was revised to use the average of all four temperature detector reading as the basis for determining operability. Additionally, the licensee initiated a work request (WR) to have the alarming probe moved away from the turbine chest insulation. On October 4, the Unit 3 HPCI was declared operabl During the test other problems were identified with the operation of the HPCI auxiliary and emergency oil pumps (AOP & EOP). The AOP automatically secured after the attached main oil pump achieved a pressure greater than 75 psig. This was unanticipated because the NSOs previous experience had been that the AOP continued running throughout the surveillance. The licensee's investigation determined that a 1992 AOP oil pressure trip set-point change, which raised the pressure switch setting from 75 psig to 95 psig, was completed on Unit 2 but not on Unit 3. This change was to prevent the AOP (a 40 Hp dc motor) from drawing repetitive starting current loads from the station's safety related 250 Vdc batter The investigation also determined that the Unit 3 switch was last calibrated on September 13 with an as-found setting of 86 psig. The switch was reset to 75 psig. While reviewing the calibration data it was determined that the as-found value of 86 psig was apparently high enough to preclude the AOP from tripping during previous surveillance tests. Until this year, neither unit's AOP oil trip
! pressure switch had
l
!
-. . -- - _ - . , - - . - . - - . - . _
.- .
l
.
! . been in the calibration program. For corrective action the licensee made the appropriate changes to the calibration procedure and recalibrated the switch to 95 psi Unrelated to the AOP pressure switch problems, the EOP failed to start when the
- operator turned the manual EOP start switch in the control room after the AOP unexpectedly stopped during the test. Later the EOP also failed to start locally in
- . the HPCI pump room. Troubleshooting determined that the secondary contacts in the circuit breaker were dirty and had caused the failure. A WR was initiated to
,
have the breaker's contacts cleaned and the breaker was successfully tested and i
returned to service. At no time during the performance of this testing did the Unit 3
, HPCI turbine lose lubricating or control oil because the main oil pump operated properly. Additionally, the AOP was available if pressure had fallen below 75 psig.
) Conclusions
'
Operations personnel properly responded to the HPCI system performance and correctly declared the system inoperable based on the guidance in place at the time l of the test. System engineering's review of the problems identified during the test was good and all difficulties were corrected in an expeditious manner. Although the i specific problems were of minor safety consequence, the problems were indicators
! of longstanding issues such as material condition, coordination of setpoints between
, units, configuration control, electrical maintenance practices, and procedural
- adequacy.
. Maintenance Procedures and Documentation M3.1 Status of Maintenance Work Backloa (Units 2 and 3)
a. Insoection Scope (62707)
.
The inspectors reviewed the past and current data tables for action requests (ARs)
, and work requests (WRs) both for outage and non-outage, planned and corrective backlogs. Discussions were conducted with cognizant licensee personnel j associated with managing these backlog Observations and Findinas
,
f While reviewing tabular data from March 27, through September 13,1996, the
! inspectors noted significant changes in some of the data. These changes occurred j in June and included the following:
- Non-outage planned WRs dropped from a previously consistent level of 966 to a new level of 406, a decrease of 560 WR * Outage planned WRs dropped from a previously consistent level of 1065 to a new level of 49, a decrease of 1016 WR ; * Minor maintenance ARs awaiting approval dropped from a previously decreasing level of 43 to a new level of 1, a decrease of 42 AR .. . . _ . . . . . . . - . . - . - - . - - - . . - . ..
i
. ,
' - * Facility WRs increased from a previously consistent level of 528 to a new level of 650, an increase of 122 WR l The inspectors were told that the changes were intended to reduce the number of planned WRs. The WRs with the job code " planned" (PL) were recoded to
" periodic" (PE) or facility maintenance depending on the tasks involved. The licensee stated that there was a directed effort not to recode jobs as corrective because that was a sensitive number but to leave them coded as PL and work them off. The final plan was to work off all WRs coded as PL and start coding new items as corrective from June forwar c. Conclusions Recoding of items in the maintenance work backlog has occurred several times )
during the past 2 years. Each time this has been done the true backlog picture gets a little clearer. This latest recoding may establish a better standard for categorizing backlog item Ill. Enoineerina ;
!
E2 Engineering Support of Facilities and Equipment E2.1 ' Containment Leak Rate Limit Exceeded (Unit 3)
! Insoection Scoce (37551)
l Follow-up of the postulated radiological dose consequences due to an as-found '
leakage pathway from the Unit 3 inboard and outboard main steam line (MSL) drain primary containment isolation valves. License Event Report (LER) 50-249/95007, Revisions 0,1, and 2, " Leakage Limit Exceeded Due to Valve Internal Damage Caused by Manual Operation of Motor Operated Valve," were reviewed during the l inspection, Observations and Findinos in June 1995, the licensee determined that the Unit 3 inboard and outboard MSL drain primary containment isolation valves were leaking greater than the licensee's local leak rate test (LLRT) test equipment could measure. License Event Report 50-249/95007 was issued because when the as-found leakage was added to the existing maximum and minimum pathway leakage rates, the total containment leakage limit for Type B and C testing exceeded 0.6 La (488 standard cubic feet per hour (SCFH)). The significance of this event was previously considered an Unrosolved item (URI 50-237;249/95010-02) pending the licensee's assessment of the potential radiological consequences of the as-found leakage pathwa __ _ _ ~
i
.
'
i
,
'
-
Brief Backoround Information During Unit 3 refueling outage 13 (March - November 1994), the existing MSL drain valves were replaced with 2 inch Anchor Darling double disk gate valves as part of j the licensee's response to Generic Letter 89-10. In January 1995, the inboard j valve was declared inoperable due to valve position indication problems. Both the
'
inboard and outboard valves were shut from the control room to comply with
, Technical Specifications. On May 28,1995, Unit 3 entered a forced outage, in June 1995 the inboard MSL drain valve was inspected, and the valve's position
!. limit switches were engaged and the valve appeared closed. A LLRT was
, performed on the MSL drain valves and the "as-found" leakage for each of ths valves was greater than test equipment could measur The cause for the inboard valve leakage was low spots on the valve's seat due to poor alignment (i.e., fit up) of the disk to seat. The root cause appeared to be poor
- maintenance instructions for valve assembly and lack of licensee experience with l Anchor Darling double disk gate valves. The cause for the outboard valve leakage
. was the valve disk's lower wedge was missing and the stem was bent. The
] licensee determined that the failure mechanism of the outboard valve was excessive thrust applied during manual handwheel operation during Unit 3 refueling outage 1 The root cause was attributed to an inadequate design modification review in
- June 1994 that failed to identify that low torque values (about 33 f>lbs) would
. damage the valve during normal handwheel operation. A contributing cause was informal controls for manual handwheel operation of motor operated valves.
] The outboard valve was replaced with a "like-for-like" valve and the iaboerd was
- repaired. The as-left LLRT results were acceptable (see Inspection Report 50-
)
, 237;249/95010, dated November 2,1995). l
Radiolooical Consecuences I
Based on the leakage from the MSL drain valves and from other primary containment isolation valves (described in the LER 50-249/95007), the licensee assessed the potential radiological consequences of the leakage in the event of the design basis accident. Assuming the worst case break of the Unit 3 MSL drain piping in the turbine building, preliminary calculations were performed using the
" minimum pathway" leakage for the Unit 3 primary containment. The licensee concluded that during the design bases accident:
- The dose to the control room operators established by General Design Criteria (GDC) 19 of 10 CFR Part 50, Appendix A, would have been exceede * The Exclusion Area Boundary and Low Population Zone dose limits established by 10 CFR Part 100 would have been exceede The licensee performed a " realistic" assessment of the radiological consequences of a loss of coolant accident (LOCA) and concluded that the control room and offsite
l
,
l
.
doses were lower than those calculated using regulatory guidance. The licensee l
stated in LER 50-249/95007, Revision 2, that when the realistic Unit 3 MSL drain i leakage values were combined with the Unit 3 realistic containment leakage component of the post-LOCA releases, the thyroid and whole body doses were less than the applicable limits in 10 CFR Part 100 and Part 50, Appendix A, GDC-1 The licensee assessed the potential radiological consequences of the leakage from the other primary containment isolation valves and concluded that limits in 10 J:FR l
Part 100 would not have been exceede i i
c. Conclusions '
I The licensee's root cause evaluation for the failures of the MSL drain valves was l methodical and thorough, and the corrective actions were comprehensive (see inspection Report 50-237:249/95010). However, based on the "as-found" condition of the MSL drain valves in June 1995, the valves were unable to maintain primary containment integrity since at least January 199 ,
Technical Specification 3.7.A.2 required that primary containment integrity be maintained at all times when the reactor is critical. Technical Specification 3.7.A.2.b (2)(a) stated that when primary containment integrity was required, primary containment leakage shall be limited to a combined leakage rate of less than or equal to 0.6 La for all testable penetrations and isolation valves subject to Type B and C test From at least January 1995 to May 28,1995, with the U-3 reactor critical, primary containment leakage was greater than 0.6 La due to leakage past the inboard and outboard MSL drain primary containment isolation valves. The inboard and ,
outboard MSL drain isolation valves are subject to Type B and C tests. Failure to l maintain primary containment leakage less than or equal 0.6 La is an Apparent )
Violation of Technical Specification 3.7.A.2 (50-249/96013-01)
The apparent violation is being considered for escalated enforcement action in I accordance with the " General Statement of Policy and Procedure for NRC l Enforcement Actions" (Enforcement Policy), NUREG-1600. Accordingly, no Notice of Violation is presently being issued for this inspection finding. Since the circumstances surrounding the apparent violation, the significance of the issue, and corrective actions were discussed with the licensee in the exit meeting for this (
inspection period on October 17,1996, a pre-decisionel enforcement conference may not be necessary in order to enable the NRC to make an enforcement decisio Before the NRC makes its enforcement decision, the licensee will be provided an opportunity to either respond to the apparent violation or request a pre-decisional enforcement conference, as described in the cover letter to this repor ,
. . - - -. . . . . - - . . - -.- - - - .. --
-
t i
.
E2.2 Isolation Condenser Suooorts (Units 2 and 3) 1
! Insoection Scope (37551)
The inspectors performed a visual verification of the structures supporting the Unit 2 and 3 isolation condensers and several feedwater pump recirculation line Licensee responses and documents were also reviewed included:
Mechanical Drawing M609C, sht 1 of 4, Condensate Booster Pipin *
Mechanical Drawing M609C, sht 2 of 4, Condensate Booster Pipin *-
Mechanical Drawing M609C, sht 3 of 4, Condensate Booster Pipin *
Mechanical Drawing M609C, sht 4 of 4, Condensate Booster Pipin *
Mechanical Drawing M609R, sht 10 of 14, Condensate Booster Pipin l
- Mechanical Drawing M1163D-553, Hanger Mark No. M-1163D-55 ~
- Mechanical Drawing M1163D-554, Hanger Mark No. M-1163D-55 * Mechanical Drawing M1163D-555, Hanger Mark No. M-1163D-55 * Phillips Drill Company Anchoring System Technical Manua * letter No. 0005201396, " Resolution of Isolation Condonser 2-1302 and 3-1302 Anchorage issues," dated 10/08/96 Observations and Findinos .
The inspectors identified differences in the anchorage for the Unit 2 and Unit 3 isolation condensers. Design engineering evaluated these and determined that the differences were acceptable and would not have effected the safety function of the i isolation Condenser. The licensee initiated changes to two drawings to reflect the i as-built plan i While performing a routine plant tour, the irispectors noticed a whip restraint on the i condensate booster piping going to the Unit 2 Reactor Feedwater Pump room that .
had anchor bolts which were tack welded to the base plates. This condition was l immediately brought to the attention of the condensate system engineer and a request for further information was mad During resolution of the identified issues, the licensee found additional discrepancies including some related to calculation of loads and seismic accelerations. The licensee had calculations performed to assess the discrepancies and determined that the existing isolation condenser supports were acceptabl Conclusions The drawings did not reflect the actual support configuration of the isolation condensers or the feedwater restraints. However, the licensee's response to the inspectors' concerns was thorough, as demonstrated by their identification and resolution of additional discrepancie . - . __ .. . - .. - - - - ._ _ - - . . - -
<
.
E2.3 Emeraency Diesel Generator (EDG) Fuel Oil Transfer Pomo (Unit 3)
- Insoection Scope (37551)
The inspectors toured the Unit 2,3 and 2/3 EDG rooms observing general i housekeeping and material conditions.
<
. Observations and Findinas On September 1 the inspectors noted that the Unit 3 EDG fuel oil transfer pump discharge gauge, Pl 3-5241-1, was over-ranged. Normal range on the gauge with '
the pump secured was 4 psig. The O to 15 psig gauge was found reading greater i than 15 psig. This condition had been identified by the inspectors on three occasions over the previous 8 months. The failed gauge was again reported to the '
WEC for corrective action.
.
'
The gauge was used to verify pressure of the EDG fuel oil transfer pump during EDG day tank and diesel fire pump day tank filling operations. The gauge was not i required for emergency operation of the EDG, A review of work documents-indicated that the gauge had been replaced at least three times since March 1995 without a root cause for the failure having been performed. The system engineer believed that the latest occurrence of the over-ranging of the gaugo was due to early closing of a solenoid valve in the pump discharge pipe which occurred during
'
Unit 2/3 diesel fire pump day tank filling operation. At the close of the report period, the licensee was in the process of performing additional testing to determine the root cause of this failure.
l Conclusions Corrective actions to repair the Unit 3 EDG fuel oli transfer pump discharge pressure i gauge have been inadequate. Although this problem only delayed surveillance
-
testing of the EDGs, it impacted the operator's ability to ensure the pump was functioning properly when operated. The repetitive nature of the gauge failure demonstrated the difficulty the licensee has in resolving seemingly minor problems (i.e., replace the gauge vice identify and fix the root cause).
E3 Engineering Procedures and Documentation E Updated Final Safety Analysis Report (UFSAR) Discrepancies Inspection Scope (71707)
The inspectors used the UFSAR as a technical reference for section M1.1,
" Operators Performance During Execution of Surveillance," of this report. A comparison of actual plant cperation and configuration was made to the descriptions contained in the UFSA _
. . . . - -
'
,
- Observations and Findinos Plant operations described in the Dresden Operations Surveillance (DOS) were compared to the UFSAR. One example was identified where the description in the .
UFSAR was not clear. Section 9.5.6, " Diesel Generator Air Starting System," l
^
stated, "Two air compressors maintain the air start receiver pressure at 250 psi j The inspector noted that DOS 6600-01, " Diesel Generator Surveillance Tests,"
'
j verified that the compressors start at 220 psig, not 250 psig. The licensee indicated that the UFSAR would be revised to read that the receiver pressure was maintained greater than or equal to 220 psig.
, Conclusion
'
The above example showed that some discrepancies between plant operation and the UFSAR exist. However, the licensee was taking prompt actions to resolve the i identified discrepanc ,
- - I E4 Engineering Staff Knowledge and Performance ,
!
E Unit 2/3 EDG Ventilation Fan Breaker (Unit 2) I l
a. Inspection Scope (37551)
J t
The inspectors observed the licensee's response to a small fire on September 15 in the 480 volt breaker for the Unit 2 supply breaker to the Unit 2/3 EDG Ventilation i Fan. Additional follow up, troubleshooting, and maintenance actions were also reviewed and observed.
.
- b. Observations and Findinos
!
.
On September 15 a small fire occurred in the 480 volt breaker for the Unit 2 supply to the Unit 2/3 EDG Ventilation Fan. The source of the fire was the control power transformer in the breaker. During a subsequent start of the Unit 2/3 EDG, after repairs to the breaker, the NSO observed that the ventilation fan was running from the Unit 3 feed. The Unit 2/3 breaker logic normally seeks power for the EDG auxiliaries from Unit 2. Further investigation identified that the relay controlling which unit supplies power to the ventilation fan was also burned ou The " power seeking" relay, located in a control panel in the Unit 2/3 EDG room, was replaced. Engineering completed a point-to-point wiring verification of both the Unit 2 and Unit 3 feed breakers and the Unit 2/3 EDG control panel. An unauthorized electrical jumper was discovered in the Unit 3 breaker that connected 120 Vac to 125 Vdc alarm power. The electrical jumper was old and had, most likely, been installed during initial construction. The electrical jumper in the Unit 3 breaker had no effect on the correct operation of the breaker and was not related to the fire in the Unit 2 breaker. The remainder of the logic and electrical wiring was satisfactor l
.
- , , , .
. __-_ . _ _ _
.
. C_onclusions
!
The engineering department's root cause evaluation was weak. The initial root l cause was considered to be simply a control transformer failure. Subsequently, the l licer.see determined that the failed power seeking relay overloaded the circuit and resulted in the control transformer failure. A thorough, point-to-point check of wiring and logic initially following the fire was not performed. Had that level of troubleshooting been performed, the additional problems could have been identified and corrected prior to retesting the EDG. This event demonstrated the licensee's difficulty in identifying root cause of equipment failure E2.4 Control Room Ventilation System Onerability (Units 2 and 3) ! Insnection Scope (37551)
The inspector observed the repair and sealing of the control room ventilation leakage and attended plant operations review committee (PORC) meetings where corrective actions were evaluated. The operability determination and other i associated documentation were also reviewe Observations and Findinas in September 1996 the licensee was closing out an open 1989 modification for the Unit 1 to Unit 2/3 control room fire-wall. Upon entering the control room, the cognizant system engineer noted that there was no differential pressure between the control room and the outside hallway. This observation prompted the licensee to perform a test to determine the actual control room to outside air differential pressur On October 7 the licensee performed testing and determined that the system was unable to maintain the required + 1/8 inch water-gauge pressure in the control room with respect to all adjacent areas. The control room ventilation system was declared inoperable and Dresden Administrative Technical Requirement (DATR)
3/4.6.1 was entered. The DATR required control room ventilation to be returned to operable within 14 days or be in a condition where secondary containment was not required (i.e., Cold Shutdown).
The control room ventilation system was declared " operable but degraded" on October 21 after significant on-site and corporate resources were expended. Many of the walls making up the control room envelop were stripped and leaks were identified and sealed. Numerous cable penetrations were identified as leaking and were also sealed. The auxiliary computer room was removed from the control room ventilation envelop with Temporary Alteration lll-53-96. The inspectors will monitor the actions taken by the licensee as this issue continues to evolve. Due to the on-going inspection of this is:ue it is considered an Unresolved item (URI 50-l 237;249/96013-02).
i l
. _ . _ . . ._.- _ _ _ _ . _ - _ _ _ _ _ _ _ _ . ~ _ _ _ _ _ _ - _
l
.
t
,
. Conclusions
'
i initial licensee reactions were aggressive and comprehensive when the issue was I finally understood. This issue was directly tied to a 1989 modification which
,
changed the fire-walls between the Unit 1 and Unit 2/3 control rooms. Further l
,
evaluation will be necessary before final conclusions can be mad I i
I IV. Plant SuDDort
- R1 Radiation Protection and Chemistry (RP&C) Controls I
R 1.1 Review of Radioactive Material Control Proaram i Inspection Scope (83750)
! The inspectors observed RP technicians performing unconditional release surveys of
- material from radiologically protected areas (RPAs) and interviewed RP technicians
and management regarding contaminated material control practices and policies.
- Observations and Findinas l Dresden Station identified approximately 450 uncontrolled contaminated items j i
outside of radiologically posted areas (RPAs)in April 1995 during a site-wide
-
survey. As documented in a June 2,1995 letter, enforcement discretion was i exercised for this issue by the NRC. This was based, in part, on immediate actions, I
such as the survey, to identify the full scope of the problem, and planned corrective i
'
actions to improve performance in this area. The inspector's review indicated l mixed results from the corrective actions which were implemented subsequent to '
'
the discovery of the item l Twenty seven corrective actions were developed to improve contaminated material control and improve radiation protection technician performance in general. On January 26,1996, an effectiveness review of the corrective actions was completed and presented to the PORC. It essentially concluded that most of the corrective actions were ineffective or were not implemented as intended. A radioactive material committee was then established and the following new, primary corrective actions were developed: 1) eliminate the temptation to remove tools and materials into and/or out of the RPAs; 2) heighten worker's level of awareness; and 3)
implement program accountability. A number of sub-items were specified for completion under each of the main action i Progress in implementing these corrective actions had been mixed. Color coding and accountability of tools to enhance workers' awareness of the contamination status of the items was well underway and was a positive step. A coritaminated tool shop located in the Unit 1 turbine area provided a storage location for contaminated tools so that the number of clean tools needed to be brought into the RPA was limited. A hands-on material release training session for radiation protection technicians (RPTs) was conducted using .slightly contaminated item _ . . _ __ _ . _ . _ _ _ _ _ . _ _ . _ . _ . . _ _ _ . . _ _.
.
t
'
!
!-
! . This was well received by the RPTs and good practices were noted while RPTs
{ were observed surveying potentially contaminated material for unconditional release j at RPA exit points. A tool monitor was used for items that required unconditional release from the RPA.
i Licensee management had intended to reduce the number of RPAs as part of the 1
- initial corrective actions for the material control problem. However, tne nurnber of l J
"outside" RPAs remained essentially the same as in 1995. Although most were
under lock and key, the inspectors noted that an RPA established in the electrical shop for 4 kV breaker work had evolved from an area with continuous RP coverage
- to one with virtually none. TMs had not provided much of a deterrent for moving materialin and out of this RPA.
.
- Approval of a RP supervisor was needed for items to be released from the RPA.
- However, all the current first-line supervisors were new to their positions within the i past year. Although all had appropriate backgrounds from previous experience
outside of Comed, none had received Dresden-specific systems training. Given the
, significant amount of past cross contamination of normally uncontaminated systems I
at the station, this lack of knowledge could effect their ability to make a proper i decision on whether or not an item should be released. During interviews, three l different RPTs indicated that one of the supervisors recommended that an
) instrument air cooler, which would normally not be expected to be contaminated, be
- released even though all areas of the cooler could not be surveyed. Due to prior l knowledge of the system, a RPT conducted thorcugh surveys subsequent to this
- recommendation and determined that the item was actually contaminated.
4 A further problem with this mechanism for releasing items was that only limited transfers of materialin and out of the RPA was intended during the initial corrective i actions. Although the restriction allowed for items to be unconditionally released l from the RPA with supervisory app. oval, it was identified that numerous items were
- released at the exit point on a daily basis. Management acknowledged that this i was not the intention of the original corrective action and that this policy would be 4 reevaluated.
!
l Conclusions l The contaminsted material control program has improved since April 1995.
i However, corrective actions such as minimizing the number of "outside" RPAs and i
'
effectively controlling all outside RPAs have not been fully implemented. Continued improvement in the control of contaminated material was needed.
i
$
$
i
)
e i
, 20 l
i a
, - - - . , -- .
- - - - -
. _ _ _ _ _ _ __ _ _ . _ . _ . _ . _ _ _ . _ __ . _ _ . . _ _.. _ .
!
l
.
I
. F3 Fire Protection Facilities and Equipment F East Fire-main Header Break
' Inspection Scope (71707) (Units 1,2 and 3)
l The inspectors toured the area of excavation, monitored work activities, and l
, observed licensee discussions on potential causes and corrective action '
' Observations and Findinas i
On October 15 at 5:35 p.m. the Unit 2/3 diesel fire pump (DFP) started due to low 1 system header pressure. At 5:40 p.m. a member of the station's Loss Prevention Group noted that the recently dug east fire-main trench was flooded. The licensee j
responded to the flooding by securing the Unit 1 DFP and extending the fire-main
. system out-of-service (OOS) isolation. The original OOS isolated only a portion of j the east header to facilitate the removal of a short section of piping suspected of j leakage. The extended boundary placed the Unit 1 crib house and site blackout diesel generator building inside the OOS. This new boundary made the Unit 1 DFP
!
and the Unit 1 Screen Wash pump unavailable to serve as sources of firewater.
. Therefore, the site fire-main system was supplied by the Unit 2 and 3 service water
'
system backed up by the Unit 2/3 DFP.
.
, Conclusion a
Poor job and OOS planning resulted in flooding and the ultimate extension of the east fire main out-of-service boundar'/. The planners did not take in to account the fact that slip joint piping was installed on the portion of the system that was being
,
worked on. After the OOS was hung and the intended section of piping removed,
the slip joint upstream of the isolation point gave way and resulted in the flooding.
i This event emphasized the continued need for focused attention on the preparation
'
of OOS and work plannin ,
V. Manaaement Meetinas
X1 Exit Meeting Summary The inspectors ,> resented the inspection results to members of licensee management
- at the conclusion of the inspection on October 17,1996. The licensee
,
acknowledged the findings presente The inspectors asked the licensee whether any materials examined during the i inspection should be considered proprietary. No proprietary information was ]
identifie l i
j 21
I I
!
. ~ . - - - .. . -
w
PARTIAL LIST OF PERSONS CONTACTED Licensee S. Perry, Site Vice President, Dresden E. Connell, Design Engineering Superintendent T. Foster, Work Control and Outage Manager R. Freeman, Plant Engineering Superintendent J. Heffley, Units 2 and 3 Station Manager C. Howland, Radiation Protection Manager R. Kundalkar, Sito Engineering Manager-T. Nauman, Unit 1 Station Manager T. O'Connor, Operations Manager F. Spangenburg, Regulatory Assurance Manager P. Swafford, Unit 2/3 Maintenance Superintendent P. Tzomes, Support Services Director D. Winchester, Safety Quality Verification Director l
,
,
f
.
INSPECTION PROCEDURES USED IP 37551: On-site Engineering IP 62707: Maintenance Observations-IP 64704: Fire Protection Program IP 71707: Plant Operations IP 92901: Followup - Plant Operations ITEMS OPENED, CLOSED, AND DISCUSSED Opened 50-249/96013-01 .VIO Failed LLRT (Apparent Violation)
50-237;249/96013-02 URI Control Room Ventilation
' Closed 50-237:249/96009-02 IFl Posted Procedures Discussed 50-237:249/95010-02 URI Assessment of the Potential Radiological Consequences of Failed LLRTs
!
-I
,
. ,,.e , l
. . - - - . . ~ . - - = - _ . - - - - . - . - . . - . - - - . . .. . . - . . . - - . . . . . . - - . . . . .
-
.:
s
LIST OF ACRONYMS USED
, AOP Auxiliary Oil Pump
, AR Action Request CRD Control Rod Drive l DAP Dresden Administrative Procedure
- DATR Dresden Administrative Technical Requirements DFP Diesel Fire Pump
'
dc Direct Current
,
DOS Dresden Operations Surveillance
- EDG Emergency Diesel Generator EOP Emergency Oil Pump.
IST Inservice Test
, Kw Kilowatt
- LER Licensee Event Report
- LLRT Local Leak Rate Test
- LOCA Loss Of Coolant Accident
- LPCI Low Pressure Coolant injection 4 LPRM Local Power Range Monitor MSL Main Steam Line
,
MW Megawatt
- NLO Non-Licensed Operator NSO Nuclear Station Operator
- OOS Out Of Service
{ PE z iodic
- PlF Performance Improvement Form PL Planned
'
PORC Plant Operations Review Committee l psig Pounds Square Inch Gauge
, RBCCW Reactor Building Closed Cooling Water
-
RP Radiation Protection I I
l RPT Radiation Protection Technicians
!
RPA Radiologically Protected Areas SCFH Standard Cubic Feet per Hour UFSAR Updated Final Safety Analysis Report i URI Unresolved item l
- US Unit Supervisor
~
WEC Work Execution Center WR Work Request i
i i
I l
24 !
\
!
,
.-. - .-,----..-e y