IR 05000354/1996009

Insp Rept 50-354/96-09 on 960922-1109.Violations Noted.Major Areas Inspected:Licensee Operations,Engineering,Maint & Plant Support

ML20135D217
Person / Time
Site:	Hope Creek
Issue date:	12/05/1996
From:	NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I)
To:
Shared Package
ML20135D179	List: IR 05000354/1996009 ML20135D175 ML20135D207
References
50-354-96-09, 50-354-96-9, NUDOCS 9612090295
Download: ML20135D217 (57)

Text

l

.

.

U. S. NUCLEAR REGULATORY COMMISSION

REGION I

Docket No:

50-354

,

License Nos:

NPF-57

.

!

Report No.

50-354/96-09

<

Licensee:

Public Service Electric and Gas Company Facility:

Hope Creek Nuclear Generating Station Location:

P.O. Box 236 Hancocks Bridge, New Jersey 08038 Dates:

September 22,1996 - November 9,1996 Inspectors:

R. J. Summers, Senior Resident inspector S. A. Morris, Resident inspector J. D. Noggle, Senior Radiation Specialist G. K. Hunegs, Senior Resident inspector, Fitzpatrick G. S. Barber, Project Engineer J. F. Harold, Reactor Engineer D. H. Jaffe, Senior Project Manager H. K. Lathrop, Resident inspector, Calvert Cliffs A. L. Della Greca, Senior Reactor Engineer T. J. Kenny, Senior Operations Engineer S. M. Klein, Reactor Engineer K. Young, Reactor Engineer Approved by:

Larry E. Nicholson, Chief, Projects Branch 3 Division of Reactor Projects 9612090295 961205 l

gDR ADOCK 05000354 PDR g

I

.

- -

-.

--

- _- - -. _.

._ _

.

.

j

.

EXECUTIVE SUMMARY Hope Creek Generating Station

NRC inspection Remt 50-354/96-09

This integrated inspection included aspects of licensee operations, engineering, j

main.tenance, and plant support. The report covers a 7-week period of resident inspection; in addition, it includes the results of announced inspections by regional inspectors in the

'

areas of radiation protection and engineering support, and includes the results of an NRC team inspection of the licensee's corrective action program effectiveness.

,

Operations Operators responded properly to all of the operational events during the period. All i

i required reports to the NRC were made. Immediate corrective actions, when necessary,

.

J were determined to be acceptable. Prompt identification of the events and use of the

,

corrective action program were considered appropriate (Section 01).

,

Responses to Generic Letter 96-06 and Bulletin 96-03 wera responsive to the concerns.

i Further NRC technical review of the licensee's planned actions, when finalized, was j

deemed necessary to close these generic issues. However, the licensee's action to

'

examine accessible portions of the downcomers during the mid-cycle maintenance outage was considered a good action (Section 02).

'

J

The inspectors concluded that the licensee root cause analysis for the safety tagging violations in early 1996 was comprehensive. Corrective actions described in licensee letters dated August 6 and October 31,1996, were reasonable. The expected timetable to complete changes to the station tagging program to provide for better implementation by station personnel seemed appropriate considering the complexity of the changes and the need to ensure that both Hope Creek and Salem stations can support the changes. In addition, the inspectors noted that the frequency of tagging errors has decreased at Hope i

Creek in recent months, indicating that some improvement has already occurred even though all corrective actions have not yet been implemented (Section 03).

A weakness in turnover between operating shifts during the conduct of an inservice test was identified. Once the specific concern was resolved, subsequent conduct of the test was successful and completed in accordance with all applicable requirements (Section 04).

While two successive safety auxiliaries cooling system (SACS) on-line maintenance outages were effectively implemented per established work plans, the inspectors concluded that conducting the outages in series vice parallelincreased the overall time that the affected system was in a degraded condition (Section 04).

The corrective action program procedures and process appear to be generally effective in the identification of conditions adverse to quality. A significant effort to resolve some long term problems has been noted. For example, the technical specification surveillance improvement proc w. and the CBD validation and verification program have resulted in a cohesive effort to identify and resolve issues. However, violations of technical ii

.

.

cohesive effort to identify and resolve issues. However, violations of technical specification requirements for the Offsite Safety Review organization were identified during the corrective action program inspection, one of which regarding staffing, was a repeat violation (Section 07).

The overall effectiveness of the corrective action prograrn has been mixed. in some cases corrective action has not been fully effective as indicatad by repetitive and continuing problems with some systems and programs. In addition, the size of the corrective maintenance backlog including the number of control room deficiencies indicate that the program is not fully effective (Section 07).

Notwithstanding the missed opportunities to identify this problem earlier, the inspectors considered the fact that the independent oversight group identified the potential SACS pump runout issue a significant contribution toward safety and an excellent example of independent oversight review of plant operations for safety concerns. The licensee interim resolution of the technical concern was found to be acceptable; however, the NRC considers this matter unresolved and will continue to follow up on this and other SSW and SACS concerns during our assessment of the ongoing, licensee-led Service Water System Operational Performance inspection (Section 07).

Maintenance PSE&G adequately planned and controlled maintenance activities associated with a risk significant on-line "A" service water loop outage. However, the inspectors identified several non-conservative errors in the probabilistic safety assessment used to justify performing the outage with the plant at power. A poor replacement valve configuration verification prior to conducting the work led to a significant delay in outage completion.

An inspector discovery that operators intentionally defeated the automatic safety function of two trains of low pressure Emergency Core Cooling System (ECCS) solely for long term protection of pump motors indicated a non-conservative operational safety posture. A quality assurance finding that the safety evaluation associated with the service water system loop fill procedure failed to assess whether an unreviewed safety question was involved led to non-cited violation of 10 CFR 50.59 (Section M1.2).

Following discovery of a failed cell in a vital 125-volt battery, PSE&G promptly developed, reviewed and implemented an effective and comprehensive plan which restored the battery to a fully operable condition. Good coordination between the operations, maintenance, and engineering departments was evident throughout this activity, including active participation by QA department personnel (Section M1.3).

Enaineerina The NRC observed, in general, acceptable engineering processes and practices in the development and installation of design changes and in the support of plant related issues (Section E1).

Good coordination of troubleshooting and evaluation efforts between engineering and maintenance departments was evident following recent automatic shutdowns of the iii

_ _. -

_

_ __..

_ _. _. _ _ _ _.

_

__

._

_ _. _.

.

control room ventilation system. PSE&G's inability to identify specific root causes for these system shutdowns was inconsistent with other recent examples of unplanned system failures (Section E2.1).

Oversight was good and engineering personnel were knowledgeable. Root cause analyses were also good, detailed, and well focused (Section E7).

The inspector identified a violation of NRC requirements, due to_the licensee's failure to

,

evaluate and satisfactorily address the interaction between the diesel generator ventilation l

and the fire suppression system. This violation is of concern because our review indicated l

a reluctance to address the underlying technical concerns about the interaction (Section

!

E8.1 ).

I Plant Suonort l

The Radiation Protection (RP) program demonstrated mixed effectiveness during response

!

to a 6-day elevated radioactive gas leak in September 1996. While significant attention was addressed to a relatively low safety significant event and consideration for offsite release of radioactive material was thoroughly evaluated, in-plant measurements failed to successfully identify the source of the leak (Section R1.2).

The RP QA oversight, training, and respiratory protection programs were found to be strong program areas (Section R5.1).

The inspector concluded that the security alarm station operation was good and that the security equipment used was in good working condition (Section S1).

'

l

!

.

iv

,

l

. _ _.

.

.

s TABLE OF CONTENTS EX ECUTIV E S U M M ARY............................................. ii

i TABLE O F C O NTE NTS.............................................. y 1. O pe ra t io n s..................................................... 1 11. M a i n t e n a n c e.................................................. 17 111. Engineering

..................................................

I V. Pl a nt S u p p o n................................................. 39

.

V. M a n age m e nt Meeting s........................................... 48 j

,

i

l i

i V

i

--

..e--

.

.

Report Details Summarv of Plant Status Hope Creek began the inspection period at 100 percent power. Full power operations were maintained until November 2,1996, when the unit was shutdown for a planned mid-cycle maintenance outage. During the shutdown, the reactor was manually scrammed from about 29 percent power due to problems encountered with the Rod Sequence Control System. The unit entered cold shutdown and maintenance activities were initiated to repair a reactor core recirculating water pump seal and inspect associated motor bearing oil reservoir conditions. After completing the major maintenance activities in the primary containment, the unit was restarted on November 6,~1996 and power operations recommenced on November 8,1996.

l. Operations

Conduct of Operations 01.1 General Comments (71707)

Using inspection Procedure 71707, the inspectors conducted frequent reviews of ongoing plant operations, in general, the conduct of operations was professional and safety-conscious; specific events and noteworthy observations are detailed in the sections below.

01.2 Operator Response to Events a.

Inspection Scope (71707)

The inspectors reviewed the operator response to a number of events and transient conditions during the inspection period.

b.

Observations and Findinas On October 9,1996, plant operators entered their tampering response procedure as a result of being notified of a possible equipment tampering ever'. at the Salem Nuclear Generating Station. The senior nuclear shift supervisor briefed all on-shift personnel of the event. The nuclear control operators walked down all control boards and the radiation monitor computer. Nuclear equipment operators toured the plant facilities to look for unusual conditions and signs of possible tampering.

Engineering and controls technician supported operator review of Hope Creek equipment similar to the Salem equipment that was affected. Additional briefings were held at shift turnover to enhance the understanding of the personnel regarding tampering significance and the need to be observant of plant conditions that indicate unauthorized equipment use.

On October 18,1996, operators notified the NRC of a possible degraded condition for the Standby Liquid Control (SLC) System. The concern was related to the loss i

i

__.

-

__

. -

.

.

'

of net positive suction head for the SLC pumps due to operation of the SLC solution storage tank at a higher temperature than currently assumed in the design calculations. During past tank filling evolutions for the SLC solution, operators

,

sometimes raised the tank solution temperature above the normal operating band in order to assure that the solution would not go below the technical specification

minimum allowable temperature. While the associated operating procedure did

specify the minimum temperature limit, no maximum temperature limit was given.

Upon review of this concern, the licensee determir.ed that a common cause failure of the SLC pumps could occur during operations at excessia temperatures (greater than 122 degrees F). The operating procedures we's immediately corrected to keep operating temperatures within the analyzed operating band.

'

On October 25,1996, operators notified the NRC of a possible condition where facility operations were outside the design basis of the plant. The concern involved operation of the Safety Auxiliaries Cooling System (SACS)in a manner that could lead to a loss of one of two SACS loops; specifically during a LOCA with LOP condition and a single failure of one of two pumps in the loop. The remaining

,

operating pump in the affected loop would also fail due to excessive flow conditions causing a pump runout protective circuit to trip. The result of this condition would lead to the loss of two (of four) emergency diesel generators (EDG) that are cooled by the SACS loop, and the equipment powered by these EDGs. Since this sequence would occur during the early injection phase of the LOCA with LOP accident response, and since the design basis of the plant during this phase

,

assumes a minimum of three EDGs operating, this concern results in operation outside of the facility design basis. Operators immediately declared the affected SACS loop inoperable and revised the associated operating procedures to remove one of the non-emergency loads on the SACS loop. This load, the RHR heat

,

exchanger, was needed to support operations of the shutdown cooling system, or torus / containment cooling system. None of these systems were required at the time of this event. Additional details regarding this event are described in paragraph a

'

07.6 of this report.

On November 1,1996, operators notified the NRC of a condition that could have resulted in the loss of the "A" service water loop and subsequent loss of the associated two EDGs due to insufficient cooling. This concern was due to coincident failures of two Service Water System (SWS) vacuum breakers discovered during inspection on October 29,1996. The vacuum breakers are designed to reduce the pressure surge effects and possible failure in the SWS piping in response

,

to a LOCA with LOP accident condition. Loss of the SWS loop from these I

conditions would result in the affected SACS loop not being able to reject the operating heat from the associated EDGs and the equipment they power, similar to the event described above. The operators immediately declared the "A" SWS loop

,

inoperable upon discovery of the vacuum breaker failures. The vacuum breakers would not open as required due to internal rust build-up. One of the vacuum breakers was repaired and returned to service later on October 29, and the second repaired and returned to service on October 31,1996. During a subsequent review of the condition on November 1, the licensee determined that the failure could have possibly resulted in the described failure of two (of four) EDGs and determined that

,

-,

.

.

to be a reportable event. The licensee's engineering organization continues to evaluate whether the vacuum breaker failures would cause a non-mitigated pressure spike during LOCA with LOP conditions that would fail the associated SWS loop.

On November 2,1996, operators notified the NRC of an unplanned manual scram from 29 percent power at 4:41 a.m. during a planned reactor shutdown. A plant shutdown was in progress when operators discovered that the Rod Sequence Control System (RSCS) would not successfully pass its associated technical specification surveillance requirement. With the RSCS inoperable, normal control rod movement is prohibited below 20 percent power, except for scram. Immediate troubleshooting efforts were unable to identify and correct the RSCS problem. As a result, operators could not proceed to shut the plant down as planned. Operators then manually scrammed the plant from a power level above where RSCS is required to be operable. The plant response to the manual scram was as expected, and a planned mid-cycle maintenance outage commenced.

On November 7,1996, operators notified the NRC of an automatic isolation of the Reactor Core isolation Cooling (RCIC) system that occurred during warming of the RCIC steam line prior to placing the system in service. The isolation signal is

designed to isolate the steam supply on an indication of a steam line break on high steam flow. Operators verified that no steam line break occurred, reset the isolation signal, and successfully placed the RCIC system in service with no further difficulty.

The operating procedure for the RCIC system notes that this condition can occur during system warming due to instrument false response to the pressure change in the associated piping when opening the steam line warmup valve. The licensee continues to review the operating procedures and equipment to identify possible corrective actions to prevent future similar events.

On November 8,1996, operators notified the NRC of a single rod scram that occurred during main turbine valve testing at about 85 percent power. The valve testing did result in a half scram as would be expected. However, control rod 42-07 scrammed at that time, which is not an expected result. Operators initially suspected that a fuse had blown on the redundant reactor protection system (RPS)

channel scram pilot solenoid for this control rod. The main turbine valve testing was suspended allowing operators to reset the planned half scram and then actions were taken to recover the control rod to its normal position. Subsequent troubleshooting determined that the scram solenoid pilot valve had actually failed and not the associated fuse as suspected. The licensee continues to review this event, however, the initial results indicate that no core thermal limits were exceeded.

,

c.

Conclusion Operators responded properly to all of the operational events during the period. All required reports to the NRC were made. Immediate corrective actions, when necessary, were determined to be acceptable. Prompt identification of the events and use of the corrective action program were considered appropriate.

.

.

02 Operational Status of Facilities and Equipment The inspectors reviewed the licensee responses dated October 28,1996 and

.

November 4,1996 respectively, to NRC Generic Letter (GL) 96-06, Assurance of Equipment Operability and Containment, and NRC Bulletin 96-03, Potential Plugging of ECCS Suction Strainers by Debris. Regarding the GL response, the licensee stated that all requested actions would be completed within the required 120 day time frame, including a description of their actions; a determination of the susceptibility to the waterhammer, two phase flow and overpressurization concerns described in the GL; an operability actermination for any affected systems; and, corrective actions, as necessary.

Regarding the Bulletin response, the licensee stated that they intend to comply with the requested actions of Bulletin 96-03; however, a description of the planned actions and mitigative strategies could not be provioad at this time due to lack of an approved technical basis. Further, interim measures and discussion were provided, however final technical basis were being sought from the NRC via a BWR Owners Group utility resolution. Final actions are not required at Hope Creek until the end of the seventh refueling outege scheduled to begin in September 1997. The inspector also noted that based on a review of operating experience feedback for recent events involving identified debris in BWR Mark ll containment downcomers, the licensee visually inspected accessible portions of the downcomers in the Hope Creek drywell during the recent mid-cycle maintenance outage. No signs of debris were found during this examination.

The inspector concluded that the licensee responses to Generic Letter 96-06 and Bulletin 96-03 were responsive to the concerns. Further NRC technical review of the licensee's planned actions, when finalized, was deemed necessary to close these generic issues. However, the licensee's action to examine accessible portions of the downcomers was considered a good action.

Operations Procedures and Documentation in response to a request for information regarding planned corrective actions for identified violations of the station tagging program documented in NRC IR 50-354/96-04, the licensee provided information on August 6,1996. On October 31, 1996, the licensee provided additional information regarding the status of one of the planned corrective actions to revise the station tagging procedure. The licensee has determined that additional procedure changes to subordinate procedures are required and that personnel training on the new program elements are also necessary prior to fullimplementation. It is now expected that all necessary elements will be in place to implement the corrective actions by December 31, 1996. In addition to the response letters provided, the inspectors also reviewed the licensee's completed root cause analysis for the tagging error that occurred May 9, 1996, in the Hope Creek switchyard and events similar in nature since early 199 _

_

_

i

.

l e

Operations has implemented some of the corrective actions including a streamlined j

list of tag approvers; new qualification requirements for any new personnel added to

)

the list; dependent upon a job task analysis, individual improved training elements i

have been identified that all personnel on the list must receive within six months of implementation in August 1996: and, implementation of a trend program by operations to identify the effectiveness of the actions taken.

The inspectors concluded that the licensee root cause analysis for the safety tagging violations was comprehensive. Corrective actions described in licensee letters dated August 6 and October 31,1996, were reasonable. The expected timetable to complete changes to the station tagging program to provide for better implementation by station personnel seemed appropriate considering the complexity of the changes and the need to ensure that both Hope Creek and Salem stations can support the changes. In addition, the inspectors noted that the frequency of tagging errors has decreased at Hope Creek in recent months, indicating that some improvement has already occurred even though all corrective actions have not yet been implemented.

Operator Knowledge and Performance 04.1 Reactor Core Isolation Coolina System inservice Test i

a.

Insoection Scope (61726)

The inspectors observed Hope Creek operators perform a quarterly inservice test (IST) of the reactor core isolation cootng (RCIC) system. Additionally, the inspectors reviewed technical specifi:ations, procedure, and Updated Final Safety Analysis Report (UFSAR) requirements regarding the operability and testing of the RCIC system.

b.

Observations and Findinos On the morning of October 4,1996, the inspectors observed station operators preparing to conduct a RCIC system IST in accordance with HC.OP-IS.BD-0001(O),

revision 20. The inspectors noted that operators on the previous shift had commenced implementation of the IST procedure nearly ten hours prior, at 12:17 a.m., and had completed most of the preparatory steps for operating the system, including verification of allIST prerequisites and precautions. The inspectors determined that one of these precautions (step 3.1.10) mandated that all personnel in the reactor building torus room be evacuated prior to a start of the RCIC turbine as a personnel safety measure. While true that no personnel were in the torus room at the time night shift operators verified the precautionary step, the inspectors were aware that maintenance technicians had subsequently entered the torus room to perform troubleshooting on a suppression chamber purge valve.

Just prior to the reactor operator (RO) starting RCIC turbine, the inspectors questioned the RO about whether precautionary step 3.1.10 was still valid. Unsure ubout the presence of personnel in the torus room, the RO deferred to the shift

-

X

,

.

supervisor who stated that personnel were in fact present in that area performing work. Upon review of the specific procedural requirement and prior to starting the RCIC turbine, the shift supervisor concluded that personnel had to be removed from the torus room and ordered that this action be taken.

Once all previously completed steps of the IST procedure were reverified, operators recommenced the test and completed it without incident. The inspectors witnessed good control of the evolution and noted good "back up" by the other on-shift reactor operator. Except for the precaution step noted above, use of the " STAR" technique, recently emphasized by station management, was also evident.

Operators drafted and approved an action request in accordance with PSE&G's corrective action program to document the "near miss" on personnel safety.

However, the inspectors judged that, even though the noted precaution had been completed earlier by the night shift, day shift operators failed to reverify that all mandatory precautions were stillin effect prior to operating the RCIC system. This failure indicated a weakness in the manner that surveillance procedures are typically conducted; i.e., one shift " setting up" for procedures that will ultimately be conducted by relieving shift operators. The inspectors were concerned that, without care in ensuring that plant conditions have not been altered during a shift change or after a significant period of time has elapsed, similar circumstances to those observed during the October 4,1996 RCIC test could result.

The inspectors reviewed PSE&G documents governing procedure usage and station operating practices; i.e. NC.NA-AP.ZZ-0001 (NAP-1) and NC.NA-AP.ZZ-0005 (NAP-5), respectively. NAP-1 section 5.3.7.B states in part that " prior to beginning an implementing procedure activity, the procedure user controlling the activity should ensure that all appropriate procedure prerequisites have been met..." NAP-5 section 5.12.3 states in part that individuals performing test activities are responsible for being " cognizant of all the limitations, precautions, and requirements" of the evolution. In the case of the RCIC IST issue noted above, the specified NAP-1 and NAP-5 requirements were not satisfied. However, the inspectors judged that this failure constitutes a violation of minor significance and therefore is being treated as a Non-Cited Violation, consistent with Section IV of the j

NRC Enforcement Policy."

c.

Conclusions Following a shift turnover, operators did not adequately verify that all precautions mandated by a RCIC IST procedure were still satisfied prior to operating the system.

This indicated a weakness resulting from turnover between operating shifts durireg the conduct of surveillance tests. Once identified and resolved, subsequent conduct of the test was successful and completed in accordance with all applicable requirements.

- - -

-

.

.

l

.

04.2 Safety Auxiliaries Coolina System On-1:ne Maintenance Outaae lmolementation a.

Insoection Scoce (62707. 71707)

The inspectors reviewed the planning and implementation of recent safety auxiliaries cooling system (SACS) on-line planned maintenance outages.

b.

Observations and Findinas On October 14,1996, the inspectors reviewed the " technical specification action statement" tracking log maintained by Hope Creek operators and noted that, on that j

particular day, two consecutive (conducted in series) TS 3.7.1.1 action statement entries were made for planned SACS subsystem maintenance outages. The first planned subsystem outage, completed in just over 5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br /> of a 30 day allowed

,

outage time, was conducted in order to perform minor work on the "B" SACS pump. The subsequent outage was necessary in order to drain, open and inspect the service water side of the "B1" SACS heat exchanger; this activity was completed using approximately 17 hours1.967593e-4 days <br />0.00472 hours <br />2.810847e-5 weeks <br />6.4685e-6 months <br /> of a 72 hour8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> allowed outage time.

Operators reset the action statement tracking " clock" to zero upon commencement j

of the second outage, which began only 5 minutes after the conclusion of the first j

outage.

The inspectors questioned the responsible " work week" manager (charged with planning and scheduling weekly work activities) regarding the basis for conducting the two SACS subsystem outages in series. While the noted outage activities were i

mutually exclusive from work interference standpoint, the manager stated that his original plan to conduct the outages in parallel (to reduce the overall time the SACS system was degraded) was rejected by operations department personnel because it

,

would result in a condition in which both a SACS pump and heat exchanger would be inoperable simultaneously, and that this was not a condition explicitly defined by technical specifications. As a result, the outages were scheduled and conducted in series, j

The inspectors independently evaluated the SACS TS action statement requirements should both a SACS pump and heat exchanger be simultaneously inoperable, and concluded that not only was this configuration acceptable (for 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br />), but that conducting the planned maintenance outages in paral!el would have reduced the overall time the SACS system was degraded. Inspector discussions with Hope Creek licensing and engineering management indicated concurrence with the inspectors assessment.

c.

Conclusions While two successive safety auxiliaries cooling system on-line maintenance outages were effectively implemented per established work plans, the inspectors concluded that conducting the outages in series vice parallel increased the overall time that the affected system was in a degraded conditio.

06 Operations Organization and Administration On October 24,1996, Mr. Lawrence Wagner was named the Hope Creek Operations Manager. Mr. Wagner has worked for PSE&G since 1984, and has held Hope Creek positions of system engineer, senior nuclear shift supervisor (SRO), and maintenance superintendent. Mr. Wagner's SRO license is being reactivated. The inspectors determined that Mr. Wagner meets the TS requirements for operations manager qualification.

Quality Assurance in Operations 07.1 Overview (40500)

i Through the review of several licensee programs, performance indicators, material condition deficiencies and discussions with licensee personnel, the inspectors evaluated the effectiveness of the licensee's controls for identifying, resolving, and preventing issues that degrade the quality of plant operations and safety. The objective of the inspection was to determine if the licensee's corrective action program resulted in problems getting resolved.

The corrective action program procedures and process appear to be generally effective in the identification of conditions adverse to quality. A significant effort to resolve some long term problems has been noted. For example, the technical specification surveillance improvement program and the configuration baseline document validation and verification program have resulted in a cohesive effort to identify and resolve issues. However, the overall effectiveness of the corrective action program has been mixed. In some cases corrective action has not been fully effective as indicated by repetitive and continuing probleme, with some systems and programs. In addition, the size of the corrective maintenance backlog including the number of control room deficiencies indicate that the program is not fully effective.

07.2 Corrective Action Proaram a.

Inspection Scone The inspectors reviewed PSE&G's corrective action program as implemented at Hope Creek. The review included implementing procedures, program implementation, interviews with plant management and line personnel, support documentation, and several self assessments performed during the past year. The acceptance by and knowledge of the corrective action program by the Hope Creek staff was evaluated by conducting interviews with plant management and line personnel and review of documentation. In addition, Quality Assurance (QA) audit no.96-190 concerning the corrective action program and the licensee's performance indicators were reviewe.

.

b.

Observations and Findinas The corrective action program was implemented through the use of two procedures; NC.NA-AP.ZZ-0000(O), Rev.1, " Action Request Process" and NC.NA-AP.ZZ-0006(Q),Rev 14, " Corrective Action Program".

Ira general, the inspectors found that the two implementing procedures provided a process for effective identification and resolution of conditions adverse to quality.

The action request (AR) process was primarily an electronic one, although a paper form could be used. The use of the paper AR form was in response to a recent QA audit finding which indicated that some personnel were reluctant to use the AR system due to their unfamiliarity with computers. The licensee had also recently

<

streamlined the AR process, making it more user-friendly. The inspectors considered these enhancements to be positive initiatives to increase the usage of ARs to identify problems, and were indicative of strong management support for the corrective action program and its processes.

Through interviews with both management and line personnel, however, the inspectors determined that in several areas PSE&G management's expectations regarding the use of ARs were inconsistently understood or open to interpretation.

For example:

Many of the persons interviewed stated that the threshold for initiating an AR was either too low, indeterminate, or unknown, while management personnel felt the threshold was about right. Severalindividuals responded that because of the economic cost of processing an AR, they preferred to get the issue resolved without an AR.

• In almost every case, the person interviewed stated tnat he did not receive feedback on the status, corrective actions developed and the effectiveness of ARs initiated. [ Personnel familiar with the station's managed maintenance information system (MMIS) software were less concerned since they could track the AR resolution electronically].

Training on the use of the AR process appeared inconsistent. Several individuals indicated they had received very little or informal (tailgate) training on AR usage, while others had been given formal presentations.

The inspectors noted several positive factors with regard to program effectiveness.

PSE&G supervision at alllevels actively encouraged the initiation of ARs for problem identification and resolution. Also, none of the personnel interviewed felt any external pressure to limit the number of ARs they should initiate, nor was there a perception that such initiations could lead to reprisal or discrimination. However, several PSE&G personnel indiceted that they felt some reluctance to verite ARs for human performance issues wheie they felt some action might be taken against the person or group involve,.

. - -..- - -

.

.

- - -

.

.-

.. -. - -

. - -. - - - - -

-

!

-

a

?

'

i QA audit report 96-151/152 focused on the effectiveness of the corrective action

.

j program. One of the audit conclusions indicated that additional management p

attention was warranted to ensure that the program remains effective.' The QA

audit report detailed several observations which were judged to be indicative of the i

overall effectiveness of the corrective action program. For example, there have

!'

been nine identified instances of compensatory sampling errors involving either the i

Chemistry or Radiation Protection departments at the station between January 1, 1995 through August 31,1996. Each of these occurrences has previously been 2-subject to a root cause or apparent cause investigation. The audit documented that

.

many issues continue to challenge the day-to-day operation of the radiation

monitoring system and that a single focus to correct radiation monitor deficiencies was not apparent.

I l

l The licensee has developed'an extensive performance indicator system. For

example, the operations department tracks operator workarounds, control room j

deficiencies, number of hours in unplanned LCO's, outstanding operability

determinations and followup assessments. The inspector reviewed the licensee's

performance indicators and noted that the total non outage corrective maintenance j-backlog is greater than 1000 items (and has shown slow progress in reducing the

'

backlog) and that there were 58 control room deficiencies. The large number of corrective maintenance and control room deficiencies was considered to be one I

measure of the overall effectiveness of the corrective action program.

,

i c.

Conclusions

!

l The inspectors concluded that the licensee's corrective action program procedures and process as described appear to be effective in the identification and resolution

of conditions adverse to quality. With several minor exceptions, the process was in l

use site-wide and received strong management support. However, inconsistent

{

understanding of AR initiation threshold, lack of feedback on ARs submitted, and

potentially inadequate training on the AR process indicated that management's

expectations regarding AR usage were not always being met or not clearly j

comrnunicated.

The performance indicators provide a solid method for management to remain aware of the status of areas monitored.

The corrective action program results have been mixed based on performance indicators, QA audit results and continued and repetitive deficiencies which challenge operators.

07.3 Root Cause Analvsig a.

Inspect;on Scope The 'acensee's corrective action program requires a root cause investigation and the development of a root cause analysis (RCA)in accordance with DTG-CAP-003, Root Cause Manual, for all significance level one condition reports. The inspector

.

-

-

.

.

.

.

reviewed the adequacy of the level one evaluations and the qualification and training for the individuals that perform the RCAs. The RCAs for the reactor manual control system (RMCS) issues were reviewed in detail and condition reports (CRs)

for the RMCS related to transponder card failures, rod control pushbuttons and rod sequence control system were sampled. The team assessed the effectiveness of the corrective action review board (CARB) by reviewing the CARB charter and attending a CARB meeting.

b.

Observations and Findinas The inspector reviewed several significance level one RCAs and found that they were completed in accordance with plant procedures and sufficiently addressed the root causes. The training and qualification for individuals who perform the root cause analyses were also reviewed. The inspector noted that there was no formal station procedure requirements for the training and qualification of individuals who perform level one RCAs. However, memoranda were issued on November 30 and December 27,1995, which required department managers to identify departmental

'

root cause specialists. These individuals were then provided Root Cause training, Human Error Analysis / Reduction, and Organizational and Process Failure Analysis.

The names of the root cause specialists for each department were incorporated into a matrix that cross referenced the individuals and the types of training received.

i The inspector sampled CRs for the RMCS related to transponder card failures, rod i

control pushbuttons and rod sequence control system. The RMCS has had several unrelated problems with hardware since the original startup. When the new corrective action program was implemerted, the hardware problems were documented and a level one RCA was conducted. To correct the deficiencies, the licensee has proposed several design change requests (DCR) as part of an RMCS action plan.

The CARB provides an additional level of technical review and oversight. The CARB reviews completed root cause analyses, planned corrective actions, corrective action schedules and effectiveness review plans and dispositions evaluations as necessary. The inspector attended a CARB meeting and noted thorough and probing questions.

c.

Conclusions The root cause training was judged to be appropriate for individuals to perform RCAs. The level one RCAs reviewed were thorough and reasonably addressed the root causes of the deficiencies.

The problems with the RMCS have existed since the original unit startup and continue to exist. The NRC concluded that development of an RMCS action plan to resolve the longstanding issues was good. Although the corrective actions program had appropriately identified the RMCS problems, the corrective actions to date have not been fully effective, which presents challenges to the operations and maintenance staf !

.

.

The CARB oversight and review function has resulted in improved evaluation quality by the assigned evaluation manager.

07.4 Self Assessment a.

Inspection Scope j

The irspector reviewed a sample of prior self assessments (SA) to assess their

<

effectiveness. Licensee self assessments are govemed by two procedures: HC.SA-

'

AP.ZZ-0034(Z), Rev. O, Self Assessment Program, and HC.SA-AP.ZZ-0005(Z), Rev.

1, Operational Readiness Self Assessment Program.

b.

Observations and Findinas The Hope Creek maintenance department (HMD) performed an SA from January 2 to 12,1996 to determine their readiness to restart from the sixth refueling outage (RFO 6). During the SA, the HMD identified that there were 944 preventative maintenance (PM) items that were past their due date. The licensee indicated that they had scoped the backlog to identify the critical PMs that were important to restart and had performed them prior to restart. The inspector independently assessed the deferred PMs and concluded that the justification to defer the PMs was not always sufficient. For example, 97 environmental qualification (EO) PMs were deferred based on calculated vice actual temperature data. The licensee verified actual temperatures at three suspect locations and identified that two locations exceeded temperatures used in Design, inspection, and Testing System (DITs) 7.5 calculations (AR 961009304). The DITs 7.5 calculates room temperatures based on heating ventilation and air conditioning (HVAC) flows.

These elevated temperature conditions existed with mild ambient temperatures (58 F) and would be less conservative with summer-time high ambient temperatures (90

- 100 F). The inspector reviewed the licensee's followup engineering evaluation response that addressed operability for this specific matter and concluded that it was acceptable. To expand the sample, the licensee took 36 additional temperature readings and identified that 7 locations had readings above those expected from the DITs 7.5 design calculations. Subsequent evaluation by the licensee determined that there were no operability issues created by this concern.

This SA also noted that HMD was not analyzing and trending Level 3 CRs. A lack of resources was identified as the cause for the problem. During discussion, HMD management indicated that they have made some progress in this area, but was not fully successful.

An Operation's SA (HTE-PA-96-0001) was performed to evaluate the overall quality of Follow-up Assessments of Operability (FAQs). The inspector noted that the SA initiated three ARs to document potential concerns with FAQs for battery temperature limits, service water deicing valves, and weir flow instrumentation.

The SA also noted that requirements for when to perform a 10 CFR 50.59 review were unclear and the two governing procedures (NAP-6 & OPAP-108), Operability

,

l Assessment and Equipment Control Program, provided conflicting guidanc I

'

.

\\

e

Corrective action was provided to address this concern. In a related discussion regarding the need for a 50.59 review for a MCPR Limit /GE analysis issue that might result in an Unresolved Safety Question (USO), the SA inappropriately concluded that one was unnecessary because the NRC was already aware of the issue. Except for this one important point, the inspector found that this SA was a candid, self critical evaluation with good conclusions.

i An Operations SA on operator workarounds was conducted in September 1996. It identified that there was a great deal of emphasis on eliminating workarounds during the sixth refueling outage (RFO-6). Of the eight that remained after RFO-6,

'

three will be closed throughout the operating cycle while the remaining five will be closed during RFO-7. This appears to be good progress. The SA also notes that some of the workarounds are being eliminated by the undesirable practice of incorporating them in procedures without addressing the underlying problem. No ARs were written to address this matter.

c.

Conclusions The licensee has performed numerous self assessments which were self critical and generally included appropriate recommendations. However, several problems which were identified were not fully addressed.

The maintenance self assessment was particularly effective. However, many of the findings had not been addressed by the line organization ten months after the assessment. In one case, the deferral of EQ PMs was based on inaccurate design information.

An Operation's SA on the overall quality of FAQs was a candid, self critical evaluation with good conclusions. One major omission related to the intended use of a 50.59 review.

An Operations SA on operator workarounds was also found to be a concise, candid report that identified that many workarounds are being eliminated by the undesirable practice of incorporating them in procedures without addressing the underlying problem. There were no recommendations made or otherwise documented to assure that this problem is addressed.

07.5 Routine Safety Meetinas a.

Insoection Scoce

'

The inspector reviewed the activities of, attended meetings and held discussions with various members of the onsite and offsite safety review organizations including the OSR, the onsite Safety Review Group (SRG) and the Station Operations Review

Committee (SORC). To determine the value added by the OSR, the inspector reviewed records associated with OSR review of LERS, NRC violations and SORC minutes.

,

.

.

The inspectors also attended several General Manager, Plan of the Day, and Corrective Action Group meetings to determine senior management awareness and involvement with day to day operations.

b.

Observations and Findmas i

Offsite %fety Review: The inspector reviewed the most recent thirty nine LERs and found that the OSR had made very few comments and generated only two ARs as a result of these LER reviews. The inspector noted that several significant LERs, involving repetitive failure of important safety equipment (e.g. LER 95-021, Inoperability of High Pressure Safety injection Due to Water in the Lubrication Oil)

were reviewed by OSR without comment. It should be noted, however, that the licensee did a thorough review of LER 95-026-01 (failure of the reactor manual control system) and issued an AR regarding the root cause assessment, in addition, one hundred records of OSR review of SORC meeting minutes contained only three comments which were judged to be minor, and six records of OSR review of NRC violations contained no comments. During a review of license change requests (LCRs) by the OSR, the inspector identified that LCR No. 95-23, which requested a change to the plant technical specifications for the emergency diesel generator surveillance requirements, was not reviewed by the OSR. A drait j

of this LCR was initially reviewed by OSR on about October 5,1995. However, the actual submittal to the NRC, dated October 7,1995 and a subsequent revision, dated October 27,1995 were not reviewed by OSR. Both of these latter documents had substantive changes to the acceptance criteria and ultimately were incorporated into the technical specifications via License Amendment No. 92.

Technical Specification 6.5.2 and site procedure, ND.SN-AP.ZZ-0001(Q),

" Independent Safety Review Program" both require, in part, that the OSR review proposed changes to the technical specifications to evaluate technical rnerit and verify the adequacy of the significant hazards analysis. Failure to perform a review of LCR 95-23 was considered the first example of a violation of technical specification requirements for the OSR. (VIO 50-354/96-09-01)

The licensee has sought to replace the required OSR function with the Nuclear Review Board (NRB) and currently has an application for license amendment under review by the NRC staff for this purpose.

The inspector reviewed technical specification required OSR staffing and noted that the Hope Creek OSR had only three (3) assigned engineers instead of the required four and that this condition had existed since September 26,1996. The same violation was also identified in an NRC inspection 50-272/96-05, dated April 22, 1996. The failure to adequately staff the OSR is the second example of a violation of TS 6.5.2. (VIO 50-354/96-09-01)

Onsite Safety Review Group: The inspector reviewed a number of SRG reports and found them to be insightful and well written. A good example was Report HQA 96-390/HSR 96-19 which was prepared to review post Refueling Outage 6. Another good example of an SRG contribution was HSR 95-12 which was a Maintenance

l i

~

l l

=

l l

Rule Pre-assessment. Tne report provided a candid assessment of organizational I

strengths and weaknesses regarding implementation of the NRC's Maintenance Rule.

The inspector found that files for the SRG areas of review responsibilities were not readily available to audit. Since TS 6.5.2.5.2 only requires a review of " selected" elements, the inspector was able to determine that the SRG was meeting its review responsibilities by reviewing selected SRG assessment reports.

Site Operating Review Committee: The inspector reviewed a collection of recent meeting minutes and found that many of the required review area responsibilities were represented. In addition, the inspector attended two SORC meetings and found that SORC members asked probing questions, showed concern for safety and noted that discussions were candid and demonstrated a good focus on outstanding plant issues, c.

Conclusions The OSR performance in the areas of LER review, SORC meeting minutes review, and NRC Violation review, was found to be a weakness in that OSR did not contribute to root cause assessment and corrective action programs. An example of inadequate OSR reviews was identified regarding license change request submittals that were not reviewed. Further, an OSR staffing inadequacy was found by the NRC that is a repeat violation of TS requirements, la contrast with the OSR, the SRG and SORC were fulfilling their requirements with regard to staffing and required review responsibilities. The SRG reviews were well documented and insightful and appeared to make a positive contribution to safety oversight. Also, SORC made positive contributions to safety via the conduct of their meetings and the questioning attitude displayed at these meetings.

i 07.6 Safety Auxiliary Coolina Svstem (SACS) Problem identification As described in paragraph 01.2 of this report, the licensee identified a concern regarding the ability of the SACS to meet its design basis for postulated LOCA with

'

LOP accident conditions. The concern was identified by station QA while assisting with a station review of an earlier concern regarding the technical bases used to justify changes to the Technical Specifications to increase the allowed outage time

'

(AOT) for a number of components at the plant, as described in Amendment 75 to the Operating License. QA postulated a sequence of events that could lead to a failure of an entire loop of SACS. This failure mechanism was determined to be an

'

original design error, in that the flow calculation model used to establish a SACS -

pump runout protective trip set point did not consider the affects of non-essential loads, such as the RHR heat exchanger, in addition to the essential or accident loads on the system for single pump operation.

i l

!

O Q

While this design error resulted in an unanalyzed condition for SACS accident response for LOCA with LOP, the only time that the system would have been vulnerable to this failure would be when the RHR system was in service for various cooling modes of operation. However, sometime in 1987, the licensee changed the system operating procedures for SACS. The affect of the change was to maintain open one of the RHR heat exchanger SACS valves for either one of the two SACS loops, whichever loop was not aligned to provide flow to the Turbine Auxiliary Cooling System (TACS). The reason for this change appeared to be due to SACS motor / pump heatup during low flow operating conditions experienced on the SACS loop not aligned to the TACS. Adding the non-essential load to the SACS loop was successful in mitigating low flow operating concerns; however, the supporting safety evaluation was no of sufficient rigor to recognize that the additional load may #fect the accident function of the system. Further, the licensee apparently mi.nN this error during the engineering review conducted to support the v.atmentioned license change implemented in Amendment 75; during the development of the failure modes analysis for the system as part of the IPE; and again, during a recently completed design basis verification for the SACS system.

The licensee interim corrective actions restored the system to an operable condition.

Further, after additional information regarding the pump runout performance characteristics were ascertained from the pump supplier, the flow set point on the trip device was increased to a point where both LOCA with LOP essential and non-essential operating loads would not result in an inadvertent loss of a SACS loop.

The licensee then revised the operating procedure to again permit the " normal" alignment of the SACS system.

In conclusion, notwithstanding the missed opportunities to identify this problem earlier, the inspectors considered the fact that the independent oversight group identified the potential SACS pump runout issue a significant contribution toward safety and an excellent example of independent oversight review of plant operations for safety concerns. The licensee interim resolution of the technical concern was found to be acceptable; however, the NRC considers this matter unresolved and will continue to follow up on this and other SSW and SACS concerns during our assessment of the ongoing, licensee led Service Water System Operational Performance Inspection. (URI 50-354/96-09-02)

Miscellaneous Operations issues 08.1 (Closed) Violation 50-354/95-01-01: failure to implement technical specification action statement requirements when limiting condition for operation not satisfied.

The inspectors verified the corrective actions described in the licensee's response letter, dated April 28,1995, to be reasonable and complete. An additional concern involving the use of improperly derived technical specification " interpretations," has been effectively resolved subsequent to issuance of the violation. No similar problems have been identified.

I l

l

O O

08.2 (Closed) Violation 50-354/95-01-02: noncompliance with the condition requirements of technical specifications. The inspectors verified the corrective actions described in the licensee's response letter, dated April 28,1995, to be reasonable and complete. Similar problems were subsequently identified and additional corrective actions, specifically the formation of the Technical Specification Surveillance Improvement Project (TSSIP), were implemented to identify all similar instances of improper surveillance test development and implementation. The TSSIP has been effective in identifying and correcting numerous additional technical specification surveillance deficiencies.

08.3 (Closed) Unresolved item 50-354/94-09-02: LERs 50-354/94004, 94005, 94006 and 94007: multiple engineered safety feature (ESF) actuations. The inspectors reviewed the LERs for the four ESF actuations. The events involved: an main steam isolation valve (MSIV) closure due to a procedural deficiency; an RHR system isolation due to a design deficiency; an NSSSS outboard isolation due to a loss of power; and, a reactor scram due to design and training deficiencies associated with the digital feedwater control system modification. All of these events occurred at the end of refueling outage 5 (RFO-5), or in the last case, shortly after restart from RFO-5. The LERs accurately described the events and the corrective actions were considered reasonable and complete.

11. Maintenance M1 Conduct of Maintenance M 1.1 General Comments a.

Insoection Scope (62707 and 61726)

The inspectors observed all or portions of the following work activities:

o RCIC system functional surveillance test e

SACS subsystem on-line maintenance outage o

"A" SWS loop on-line maintenance outage o

"C" SWS subsystem on-line maintenance outage o

TACS Valve EGHV-2522F Accumulator replacement e

SWS Valve 1EAHV-2198A replacement e

Safety Tagging verification for the "A" SWS loop outage b.

Observations, Findinas, and Conclusions in general, the inspectors found that the work performed during the conduct of the above noted maintenance and surveillance activities were in accordance with approved station procedures and work control programs. Specific exceptions to this general statement are described in other sections of this report where appropriat i i

.

M1.2

"A" Service Water System Looo On-line Maintenance Outaae a.

Inspection Scoce (62707,37551,71707)

The inspectors observed and reviewed various portions of the planning, risk assessment, implementation and recovery from a service water system (SWS) loop outage. This significant on-line maintenance activity, conducted between September 30 and October 1,1996, affected the entire "A" SWS loop and cooled components, and required close coordination between the operations, maintenance and engineering departments. This assessment included management and staff interviews, work order and procedure reviews, and maintenance and operations

,

activity observations.

b.1 Outaae Preparation, Plannina and Risk Assessment Observations and Findinas Hope Creek management elected to perform an on-line maintenance outage of the SWS "A" loop (which includes both "A" and "C" subsystems) primarily to replace the "A" service water pump discharge isolation valve (1EAHV-2198A), which earlier was determined to have significant seat leakage. Because of this leakage, a planned inspection of "A" pump discharge strainer could not be performed at the frequency prescribed by a previously established strainer element failure corrective action recommendation. In order to conduct the valve replacement, the entire "A" SWS loop had to be drained, affecting the operability of 11 other supported systems, including the "A" safety auxiliaries cooling system (SACS), the "A"

'

emergency diesel generator, and several trains of ECCS and accident mitigation systems. Recent Hope Creek engineering department discoveries involving SACS and SWS design and licensing bases discrepancies compounded the complexity of this loop outage (see NRC Inspection Report 50-354/96-06 section E2.1).

The inspectors reviewed the SWS loop outage plan and noted that it was sufficiently comprehensive and provided generally adequate outage justification.

Additionally, the inspectors observed that the SORC meetings, at which members evaluated the plan's adequacy and net safety gain analysis, demonstrated thorough questioning and assessment. An operations department initiative to practice equipment casualty scenarios at the plant simulator with an "A" SWS loop outage as an initial condition demonstrated a proactive and conservative operations philosophy.

During the outage, the inspectors independently discovered multiple errors in the quantitative Probabilistic Safety Assessment (PSA) performed to evaluate the risk significance of the "A" SWS loop outage. Specifically, the PSA failed to account for several safety-related plant components / systems that were out of service during the outage, in part because the request for the PSA from the planning department was not sufficiently specific. A re-analysis confirmed that though there was a slight increase in overall outage risk as a result of the errors, it was not enough of an increase to cause it to exceed the criteria used by station management to reject on-line maintenance plan o a

b.2 Outaae imolementation Observations and Findinas During the actual work, the inspectors noted that the use of an outage " straw boss" improved the efficiency of the outage and minimized the time needed to resolve

emergent issues. Operators properly entered all applicable TS action statements for affected plant systems, and adequately protected safety-related redundant systems to ensure the effects of potential plant transients could be effectively mitigated. No other work was permitted in the station that had the potential to result in a plant transient.

The inspectors observed that, during the outage, operators placed both the "A" and

"C" core spray pumps in " pull-to-lock," effectively defeating their automatic design basis accident mitigation function. This action was directed by the standard operating procedure appropriate for the plant condition at that time in order to protect the associated pump motors from damage that could result from high room temperatures. Engineers stated that high room temperatures would be likely following extended core spray pump operation due to the fact that the respective room coolers were made inoperable by the SWS loop outage. However, the inspectors judged that, though permitted by technical specifications, intentionally defeating the safety function of two trains of low pressure ECCS solely for long term equipment protection was not indicative of a conservative safety posture.

The total outage duration was approximately 41 hours4.74537e-4 days <br />0.0114 hours <br />6.779101e-5 weeks <br />1.56005e-5 months <br />, which exceeded the scheduled goal of 36 hours4.166667e-4 days <br />0.01 hours <br />5.952381e-5 weeks <br />1.3698e-5 months <br />, primarily because the replacement "A" service water pump discharge isolation valve had to be modified in order to properly install it in the SWS. The inspectors concluded that a poor valve configuration verification prior to beginning the loop outage was responsible for the noted delay.

b.3 S_WS Looo Restoration Observations and Findinas After the scheduled outage work was completed, operators re-filled the drained "A" service water loop with the operating loop using a manually opened system cross connect valve. Hope Creek quality assurance (QA) personnel questioned the acceptability of this practice in terms of its impact on the technical specification requirement for what constitutes an operable service water loop flow path. Though an approved operating procedure governed the use of this loop fill method, follow up evaluation by engineering personnel in response to the QA concern determined that procedural enhancements were necessary to positively verify that adequate flow would be available to the operable SACS heat exchangers while conducting this evolution. The inspectors reviewed the 10 CFR 50.59 safety evaluation that justified the original procedure revision (HC.OP-SO.EA-OO')1Q Revision 8, "SWS Operation") to allow this practice and determined that it failed to address its effect on the operable loop.

The inspectors judged that, though operators adequately controlled the loop fill evolution using the applicable guidance (without incident), the subsequent QA finding led to the discovery that 10 CFR 50.59 requirements were not met when the guidance was originally established. Specifically, the safety evaluation failed to

.

.

20 i

consider whether an unreviewed safety question would be created by cross-connecting an operable SWS loop to an inoperable loop. Since this issue was I

ider+d the operations department enhanced the noted procedure and conducted an appropriate safety evaluation. This licensee-identified and corrected violation is being treated as a Non-Cited Violation, consistent with Section Vll.B.1 of the NRC Enforcement Poliev.

c.

Conclusions PSE&G adequately planned and controlled maintenance activities associated with a risk significant on-line "A" service water loop outage. However, the inspectors identified several non-conservative errors in the Probabilistic Safety Assessment used to justify performing the outage with the plant at power. A poor replacement valve configuration verification prior to conducting the work led to significant delays in outage completion. An inspector discovery that operators intentionally defeated the safety function of two trains of low pressure ECCS solely for long term protection of pump motors indicated a non-conservative safety posture. A quality assurance finding that the safety evaluation associated with the service water system loop fill procedure failed to assess whether an unreviewed safety question was involved ied to non-cited violation of 10 CFR 50.59.

M1.3 1BD411 125-Volt Batterv Cell Replacement a.

Insoection Scoce (62707. 37551)

The inspectors observed an " emergency" SORC meeting in which a safety evaluation for a temporary modification involving the removal of a Class 1E battery cell was discussed. Additionally, the inspectors reviewed the maintenance procedure used to conduct a battery cell replacement, as well as observed portions of the actual in-process maintenance activity.

b.

Observations and Findinas On September 16,1996, during a surveillance test of the 1BD411 125-Volt battery, a vital power source for the "B" channel required to be operable in operational condition 1 per TS 3.8.2.1, maintenance technicians determined that cell #19 (60 cells in all) did not meet Category A or B requirernents per the noted TS's surveillance acceptance criteria. Specifically, the cell f ailed to maintain a (level corrected) float voltage 2: 2.13 volts. Operators were promptly notified, an action request was written, and appropriate TS action requirements were implemented for increased battery monitoring.

A battery outage was planned in order to conduct the necessary cell replacement with the plant on line, which would result in a voluntary TS action statement entry with an allowed outage time of only 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br />. The work plan projected that 1 %

hours would be required to perform the activity. Based on replacement (,Wi concerns just days before the battery outage was to occur, engineering personnel devised a temporary modification (TMOD) to simply " jumper out" the affected cell

-

.

.

'

should the new replacement cell fail to meet acceptance criteria. An " emergency" SORC meeting was held to review the associated TMOD safety evaluation. The inspectors witnessed the SORC discussions and noted an excellent level of questioning and review. Several TMOD package and maintenance plan enhancements were made as a result of the meeting, which included active participation by QA department personnel.

The inspectors observed portions of the actual cell replacement activity and judged that it was well coordinated and controlled by all departments involved. The maintenance procedure and work package adequately guided the cell replacement process; the total battery outage time was limited to just over one hour, well ahead of the plan. Post-maintenance retests on the battery were satisfactory and the TMOD contingency was not needed.

c.

Conclusions

,

Following discovery of a failed cellin a vital 125-Volt battery, PSE&G promptly

'

developed, reviewed and implemented an effective and comprehensive plan which restored the battery to a fully operable condition. Good coordination between the operations, maintenance, and engineering departments was evident throughout this activity, including active participation by QA department personnel.

M8 Miscellaneous Maintenance issues M8.1 (Closed) Unresolved item 50-354/94-19-01: reactor protection system and engineered safety system actuation (reactor scram). One of the root causes involved in this event resulted in the identification of a violation of NRC requirements, specifically the application of fuses installed in safety related equipment control circuitry (Violation 50-354/94-19-03). Additionally, a simulator fidelity issue was identified pertaining to plant response following a loss of main turbine generator cooling systems. A review of the Licensee Event Report associated with this event (LER 50-354/94-12) and the licensee's violation response indicated that PSE&G adequately addressed the concerns raised by the inspectors.

Corrective actions implemented as described in these documents have been effective in precluding recurrence of similar problems.

M8.2 (Closed) Violation 50-354/94-19-03: failure to maintain fuse configuration control.

The inspector verified the corrective actions described in the licensee's response letter, dated November 14,1994 to be reasonable and adequate to prevent reoccurrence. However, similar problems have occurred since the violation was issued. To evaluate the occurrences the licensee performed a level 1 root cause analysis (RCA). The root cause was a failure to follow the procedure for the installation of a replacement fuse. The inspector determined that the RCA was sufficient and the occurrences were isolated. To ensure that the fuses currently installed are the correct fuses the licensee implemented a fuse configuration control program. The program was reviewed and the inspector found that the licensee has made adequate progress in the implementation and plans to complete the program by RFO __

.

M8.3 (Closed) Violation 50-354/95-10-02: missed technical specification required surveillance tests for TIP explosive squib valves. The inspectors verified the corrective actions described in the licensee's response letter, dated September 11, 1995, to be reasonable and complete. In response to this issue as well as several other documented instances of improperly established surveillance requirements, the Technical Specification Surveillance Improvement Project was initiated; this dedicated review effort, scheduled to complete in December 1996, has been effective in identifying and correcting numerous additional technical specification surveillance deficiencies.

M8.4 (Closed) LER 50-354/94012: reactor protection system and engineered safety system actuation (reactor scram). The inspectors reviewed the corrective actions prescribed by this LER and concluded that they have been effective at precluding similar occurrences. Additionally, this LER is discussed in the closure of Unresolved item 50-354/94-19-01 above.

M8.5 (Closed) LER 50-354/94008: noncompliance with techracal specification 3.3.7.11 actiori for out of service radiation monitoring instrumentation. This LER described a minor issue and is considered closed.

M8.6 (Closed) LER 50-354/94009: noncompliance with technical specification 3.3.7.4, remote shutdown system controls. This LER describes an event that resulted in an NRC violation (50-354/94-13-02). The corrective actions for this violation were reviewed in NRC IR 50-354/96-07 and the violation was closed at that time. This M8.7 (Closed) Unresolved item 50-354/96-03-01: surveillance testing of automatic depressurization system valves. Based upon review of the licensee's LER (50-354/96011), the NRC considers that the licensee made a conservative decision regarding the valve testing requirements. This item is considered closed.

M8.8 (Closed) LER 50-354/96-011: technical specification 3.0.3 entry. As described above, this LER provided the basis for the NRC assessment that the surveillance testing operations and operator use of plant technical specifications were conservative. This LER is closed.

Ill. Enaineerina E1 Conduct of Engineering E1.1 Plant Modifications - Confiauration Chances a.

Insoection Scooe (37550)

The inspectors reviewed design changes 4EC-3368 (removal of 125 Vdc undervoltage relay and associated wiring in the 4160 Vac Switchgear 10A401),

4EC--3411 (installation of a cross-tie between the "B" and the "D" pump discharge headers in the residual heat removal (RHR) system), and 4EC-3586 (installation of

.- - - -. - -,

--._ -

..

... -. - -... -. = -. -. - -

-..

,

,

'

f

4 uninterruptible power supplies with battery-backed inverters and bypass switches

,

for the feedwater system control cabinets). The inspection focused on the design

i change program implementation and the licensee's review of the plant j

configurations changes to ensure that the criteria delineated in 10 CFR 50.59 and

}

10 CFR 50, Appendix B, " Design Control," were met.

!

b.

Observations and Findinas

,

i

!

The inspectors determined that the procedures governing the implementation of the design changes were well written. The installation of the design changes were

]

.

.

planned, controlled, and implemented according to the station procedures. The -

)

l inspectors also confirmed that the changes had been reviewed and approved by the l

safety review committees as directed by the procedures.

'

The installation processes were in good order and contained directions for the

removal and installation of appropriate equipment. All components had been

!

ordered and received using approved procurement methods. The safety evaluations

'

properly addressed the 10 CFR 50.59 requirements. In the case of design change package 4EC-3411, welding procedures and weld travelers were all properly

,

j completed. Testing was in accordance with the ASME Section ill, Class 2,1994

edition, winter 1994 Addenda through 1986 Edition, and category l(Q) The l

welders were certified to ASME,Section IX. The certification of the weld wire and components were in proper order and in conformance with the purchase orders.

!

No anomalies or concerns were identified during the review of design change i

packages (DCP) 4EC-3368 and 4EC-3586. Regarding DCP No. 4EC-3411, the

]

inspectors noted that the licensee used code case N416-1 for pressure testing the

installed 18 inch piping and that the water for the testing had been taken from the

~

j suppression poc'. This water was approximately 75 degrees F, which is less than i

the system operating temperature. The NRC approved the use of the code case via

,

[

a letter, dated January 20,1995. The permission was granted provided that l

additional surface examinations would be performed on the root pass of the butt

and socket welds. The inspectors confirmed that radiography and dye penetrant examinations had been performed on these areas.

l The inspectors further confirmed that the UFSAR and design basis documents had been updated after the design changes were completed; the operators received the

{

necessary training; the operation procedures had been updated to reflect the design

~

i changes; and that the operational drawings were also updated for use by the

operators.

'

c.

Conclusions i

i i

I i

.

The inspectors concluded that the program for designing and installing configuration j

changes to plant systems was acceptable. The inspectors further concluded that j

the 10 CFR 50.59 program had been properly applied.

i i

i

a v

-

.-,

.m.y.f

--r-o--

e-,.-

-.-

.,

,v..

,-

- --

,

-

. - _ _ _

.

'

,

E1.2 Temocrary Modifications i

a.

Inspection Scone (37550)

The inspector reviewed a listing for all temporary modifications (TM) and PSE&G's procedure for their control. The inspectors reviewed the temporary modification log in the control room and selected five changes to walk down. The inspector also assessed the temporary modifications program for conformance with 10 CFR 50.59

,

i and 10 CFR 50, Appendix B, Criterion lil, Design Control.

b.

Observations and Findinas i

The inspector made the following observations: Twenty-seven TMs were installed, four were safety-related, all had been assessed by qualified technical reviewers for continued installation, and safety evaluations in accordance with 10 CFR 50.59 had been prepared, when applicable, to ensure that the change did not constitute an l

unreviewed safety question. The inspector also noted that the general manager and the Safety Operation Review Committee (SORC) had approved the changes prior to installation.

The drawings, in the control room, for all of the reviewed TMs had been red lined to show the changes made to the system. All of the installed modifications were clearly marked in the field to alert operators of the TM. Testing had been performed after installation to ensure that the modification did riot interfere with safe operations. The TMs had been verified installed correctly by a second qualified person, c.

Conclusions The inspectors concluded that the temporary modification process was well managed. The installed modifications were easily identifiable in the plant and were i

not permitted to languish in the plant for a long time.

E1.3 Desian Basis (37550)

a.

Inspection Scoce During the review of the design change for the RHR system the inspector also I

reviewed PSE&G's use of the design basis document.

b.

Observations and Findinas PSE&G's review of the safety auxiliary cooling system design bases identified several weaknesses. As a result of these weaknesses, PSE&G decided to expanded the design verification process to six risk significant systems. The verification includes an in-deptF aview of calculations associated with the design basis. As of the completion of the inspection the licensee had completed their review of two systems (station service water and the safety auxiliary cooling systems) and was

. _

_

_ _ _ _. _ _ - _.

_

_

__

__

_.

_ _ _.

.. _ -. _..

'

.

,

j

reviewing a third one (emergency diesel generator system). The other three

systems (high pressure coolant injection, reactor core isolation cooling, and standby liquid control) had not been started.

'

The inspectors also determined that the configuration baseline documents (CBD),

that contain good guidelines for engineers to plan design changes, had not been validated. Therefore, a letter was issued by the director of design engineering and

projects restricting the use of the CBDs. Currently CBDs were being updated when design changes were completed. This process was ongoing and, as of the end of the inspection, the licensee was validating the SLC CBD for the purpose of turning it into a design basis document (DBD). PSE&G plans to validate all of the CBD's.

c.

Conclusions

'

PSE&G had recognized that some weaknesses existed in their design bases and was i

implementing a program to validate them. PSE&G was also upgrading the

configuration baseline document to be used by engineering.

,

E2 Engineering Support of Facilities and Equipment l

l E2.1

"A" Control Room Ventilation System Trio - Unotanned TS Action Statement Entries

a.

Insoection Scope (37551)

The inspectors observed troubleshooting and root cause evaluation activities i

following unplanned outages of the "A" control room ventilation system (CRVS).

b.

Observations and Findinas

'

On September 18,1996, the "A" CRVS train automatically shut down following a

chilled water circulating pump trip. Operators appropriately declared the system inoperable and entered the action statement for TS 3.7.2. Subsequent troubleshooting did not identify a specific root cause, but several individual system

components were replaced and the system was successfully restarted on September 23,1996 and declared operable. On October 7,1996, the system again tripped for unknown reasons, and the system was declared inoperable.

In an effort to identify the cause(s) of this unplanned TS action statement entry, maintenance technicians and system engineers jointly developed and implemented a

'

!

focused troubleshooting plan that included inspections and testing of individual ventilation dampers and other system components. Additionally, technicians installed data chart recorders in the system's instrument panels (via temporary modification) to capture any transient or unusual process signals after the system was restored to operation. However, after several days of normal system operation, the CRVS failed to indicate any unusual condition. The inspectors noted that station management placed a high priority on resolution of this issue.

J l

.

,

On October 11,1996, primarily based on a recommendation by the responsible j

system engineer, operators declared the "A" CRVS train operable. The root cause evaluation thoroughly described the actions taken to identify potential causal factors, but also indicated that these efforts proved inconclusive. Because the system failed to repeat the trip initially experienced, and because all major system components had been verified to be operating correctly, the operators judged that the system trip was spurious and restored it to an operable status. The inspectors judged that PSE&G's inability to identify specific root causes for the two CRVS

,

system trips was somewhat inconsistent with other recent equipment problem examples in which high priority management attention had been focused.

c.

Conclusions Good coordination of troubleshooting and evaluation efforts between engineering arid maintenance departments was evident following recent automatic shutdowns of the control room ventilation system. PSE&G's inability to identify specific root i

causes for these system shutdowns was inconsistent with other recent examples of unplanned system failures in which high priority management focus had been placed.

E2.2 Seismic Qualification of " Hear-Here" Booths a.

Inspection Scoce (37551)

The inspectors conducted a seismic qualification assessment of various equipment installed in the reactor, auxiliary, service water, and control buildings. The focus of this evaluation was " Hear-Here" booth installation.

b.

Observations and Findinas The installation of the booths began in 1991 under work order #901029171, activity #1. The Hitti Kwik Bolt and the Hilti HDI Drop In Anchor were used for their installation. Both anchors were acceptable for use as seismic restraining points for non-permanent plant equipment when installed in accordance with Station Procedure HC.MD-GP.ZZ-010 (Q), Rev. 3, and Engineering Procedure DE-TS.ZZ-4007(Q). However, some bolts were not installed in accordance with the procedure as there was indication of damaged identification stamps, cut rebar, and inadequate torquing. Further, no concrete excavation record was not kept, contrary to a requirement of the engineering procedure. In August 1991, a Deficiency Report DR# HTE-91-164, was written identifying several anchor studs that were damaged during initial hammering of the anchors. To correct the anchor problems, remounting, new torquing and verification of stud stamp identification was completed under work order #901029171 activity #2. The rework was signed off as complete in May 199.

.

The licensee performed a walkdown in May 1996 to verify the rework activities (CR# 960527079) and determined that the booths were properly installed.

Additionally, a 10 CFR 50.59 review was completed which concluded that no unreviewed safety question was created by this booth installation.

c.

Conclusions The inspector noted that the installation of " hear-here" booths were completed per a work order rather than as a " permanent" plant modification; as a result, the installation was not evaluated per 10 CFR 50.59 to determine if an unreviewed safety question existed. This failure to perform a 10 CFR 50.59 review constitutes a violation of minor safety significance and is being treated as a Non-Cited Violation, consistent with Section IV of the NRC Enforcement Poliev.

l

E2.3 Seismic Qualification of the 480 Volt Electrical System

a.

Inspection Scope The inspector conducted an assessment of the seismic qualification of various components within the 480 Volt electrical system. Specifically the Motor Control

,

Center (MCC) panels, transformers, and switchgear panels.

'

b.

Observations and Findinos

,

The inspector performed a walkdown of various 480 Volt system components to j

assess their material condition. During the walkdown the inspector identified j

deficiencies involving lose or missing screws and bolts on MCCs, transformers and switchgear panels. Further, the inspector noted that the licensee had documented these and other deficiencies found in the class 1E system in Action Request #

960117071. The inspector reviewed the AR and noted that the licenseo performed j

an assessment of the impact of missing / loose fasteners and screws on the seismic qualification and fire protection capability of the equipment. The assessment concluded that there was no seismic or fire protection impact. Corrective maintenance was performed to correct identified deficiencies. As an additional preventative measure, the system manager conducts monthly walkdowns of one-

'

third of the system to document noted deficiencies and inputs the information into j

the corrective actions system for corrective maintenance.

<

c.

Conclusion Fastener deficiencies were noted in the 480 Volt system. The licensee identified the deficiencies and assessed the significance. The corrective action were appropriate for resolution of the deficiencies and the addition of the system manager monthly walkdowns is sufficient action to continually identify deficiencies when they arise.

)

_

_

_ _ _ _

. _ _ _ _

_ _ _ _

_ _. _. _ _ _ _ _

.

__

_

>

-

.t

.

E2.4 ASME Code System Pressure Test Review

a.

Insoection Scooe (37551)

The inspector reviewed ASME code system pressure test results.

b.

Observations and Findinos The inspector performed a review of ASME Code system pressure test. A system pressure test is performed on certain ASME Class 3 components following various work activities. The licensee's OA inspectors perform these tests using visua! test examination techniques to inspect for leakage while system / component i'; at operating pressure. These examinations are performed in accordance with written instructions in maintenance procedure HC.MD-GP.ZZ-040. The inspector reviewed the test results for vacuum breaker solenoids which normally operate in a vacuum i

condition. There is no written instruction for performing a system pressure test for j

a system that operate in a vacuum condition. When reviewing the system pressure

'

test results for the Service Water to SACS vacuum breakers the inspector noted that three smoke tests were performed. It was determined that these smoke tests were performed as an " enhancement" in accordance with the maintenance procedure that had been changed "on-the-spot" to include the smoke test in addition to the VT examination. However, the individuals performing the smoke tests had not been trained to perform the test. In addition, the system engineer did not obtain the appropriate level of review prior to adding the requ'rement to the maintenance procedure as required per station administrative pacedure NC.NA-AP.ZZ-0001, Nuclear Procedure System.

c.

Conclusions During an inspector review of ASME Code system pressure test results, the inspector concluded that the use of the "on the spot change" process for the smoke test requirements for the vacuum breaker solenoid valves was considered a violation of NC.NA-AP.ZZ-0001 and Technical Specification 6.8.3. This is an example of a licensee-identified and corrected violation, and as such, is being treated as a Non-Cited Violation, consistent with Section Vll.B.1 of the NRC Enforcement Poliev.

E2.5 Enaineerina involvement in Site Activities a.

Insoection Scoce (37550)

The inspectors reviewed licensee event reports (LERs) and root cause analysis reports in the area of engineering. The inspectors evaluated the effectiveness of the engineering staff in supporting plant needs through a review of these reports, interviews of responsible engineering personnel, and an assessment of the quality of the analyses performed to resolve the reviewed issue a

.

b.

Observations and Findinas Engineering personnel used a well written procedure to perform root cause analysis of plant incidents. Interviews of responsible personnel indicated that they had received training in root cause determination methods. These discussions also indicated that PSE&G engineering was actively involved in the day-to-day operation of the facility and that communications between system and design engineering and the operation staff was good.

The inspectors' review of the selected documents also determined that the causal i

analyses were detailed, chronologies of the incidents representative of the time lines related to the actual event, safety significance of the events assessed, recommended corrective actions offered to other departments, and engineering assessments, in the most part, independently done. The inspectors *. ? led the status of recommended corrective actions identified in the evaluation t & rts and found them to be completed in a timely manner. One example was iden6 :.ed where PSE&G was less than effective in the issue resolution.

Loss of isolator Cabinet Power On January 31,1996, the licensee experienced a loss of power to a Bailey optical isolator cabinet that affected alarm functions in the control room. This event was reviewed previously by the NRC, but at the time of the review a root cause analysis had not been completed.

The inspectors' review of the completed analysis determined that the responsible engineers had evaluated all anomalies observed during troubleshooting and the failure modes of the associated components. The engineers, however, were not able to identify the root cause of the event, only probable causes, the same as those theorized during troubleshooting. These causes (e.g., loose connection, excessive voltage drop across diodes), while reasonable, did not appear to be individually responsible for the loss of power to the cabinet. The licensee recognized the limitations of the root cause analysis, but appeared to partly rely on the comprehensiveness of the corrective actions taken during troubleshooting and the knowledge and expertise of the engineering staff in having identified the faulty components (see also Section E-8.2).

c.

Conclusions The inspector found the root cause analyses to be detailed and, in general, well focused. The engineering evaluations contained in the root cause analysis were effective in identifying required safety improvements. One exception was the evaluation of the loss of isolator power event, where only probable causes were identified and the resolution of the issue indicated some reliance on the comprehensiveness of the corrective actions and the knowledge and experience of the engineers. The inspectors further concluded that engineering was involved in the day-to-day operation of the plant, but that the implementation of the license change could have been done more efficientl.

.

E4 Engineering Staff Knowledge and Performance E4.1 Oraanization and Staffina a.

Inspection Scope (37550)

The inspectors evaluated the engineering organization with regard to staffing level and experience to determine their ability to support the operation organization.

b.

Findinas and Observations During recent months the engineering department has undergone consolidation and reorganization in an effort to improve the quality of engineering products. PSE&G evaluated the engineering staff for experience, capabilities, and aptitude. As a result, many engineering and supervisory positions were vacated and replaced with new personnel better meeting the standards imposed on the staff by senior management; engineers were reassigned to other duties; training was conducted to strengthen reccgnized weaknesses. In addition, some contract personnel were dismissed.

Because of the ongoing activities in personnel management, the inspectors were not able to perform an assessment of the engineering organization. The documents reviewed, however, indicated acceptable performance. The inspectors also found the engineering personnel interviewed to be knowledgeable of the issues discussed.

Backlogs appeared to be at an acceptable level.

In November 1995, PSE&G contracted an external agency to perform a quantitative analysis of the engineering quality. The analysis conducted through an engineering survey evaluated engineering judgement, problem solving skills, and engineering quality. This analysis was repeated in June 1996, with improvements noted in all three areas. The inspectors did not review the evaluation process used by PSE&G.

The results of the latter survey, however, indicated that current ongoing training of the engineering staff was ef'ective in improving the skills and capability of the engineering staff to support the operation of the plant.

c.

Conclusions The inspectors concluded that the engineering staff was knowledgeable of the assigned tasks and capable to fulfill their responsibilities. Further, the inspectors concluded that the ongoing training of engineers had positively impacted their performance. However, because PSE&G was undergoing a period of staff and management consolidation, the effectiveness of the reorganization could not be fully evaluate,

.

E7 Quality Assurance in Engineering Activities E7.1 Audits and Assessments a.

Inspection Scooe (37550)

The inspectors reviewed internal audits and assessments to determine the ability of the licensee oversight organizations to conduct independent evaluations of the licensee staff and the quality of the technical reviews.

b.

Observations and Findinas The Quality Assurance (QA) organization conducted an audit of the Nuclear Engineering Department between March 6 and April 6,1995. A tea m of 17 auditors and technical support personnel from both PSE&G and external organizations evaluated a number of engineering programs, including configuration control, design changes, corrective actions and environmentai qualification of safety-related equipment. The team also evaluated the training, certification and qualifications of engineering personnel.

A review of the QA report indicated a good audit program with good plans, appropriate interviews, and substantive findings and observations. The report also included a list of recommendations for areas requiring additional management attention. The findings were clearly stated, directed to the appropriate management personnel, assigned a tracking number, and followed up to resolution.

The inspectors also reviewed an engineering assurance audit of the design change process performed by an external organization. The audit concentrated on rnechanical and civil / structural design and testing of the design changes. The report indicated a detailed review of portions of several design packages, including safety evaluations, applicable licensing documentation, analyses and calculations and other design documents. The inspector noted quality observations and appropriate recommendations.

In January 1996, the nuclear engineering management issued a self-assessment report, No. 96-04, in which they provided an overall evaluation of the Hope Creek engineering improvement plan. This document, addressing areas like engineering quality, problem identification, and communications, indicated that the licensee had evaluated past performance of the engineering organization and developed a plan to resolve the identified deficiencies. The evaluation had been done through a review of performance observations documented in recent internal and external reports and surveys. The inspectors noted that the document provided a good insight in the status of ongoing activities to irnprove engineering quality and that PSE&G was evaluating past performance deficiencies and was attempting to resolve the _ _.

.

.

c.

Conclusions

]

reight of engineering was good. The use of self-assessment was an effective r

ans to improve performance.

E8 Miscellaneous Engineering issues E8.1 (Closed) Unresolved item 96-03-04: EDG Ventilation / Fire Suppression Systems Interface.

i a.

Insoection Scope The NRC review in February 1996 of the fire suppression system (FSS) in the emergency diesel generator (EDG) rooms identified two issues that required review and action by the licensee: one issue pertained to the margin between the design temperature of the EDG room and the setting of the FSS temperature sensors; the i

other pertained to the potential interaction between the nonsafety-related FSS and the safety-related EDG room ventilation system. Both issues were based primarily on a NRC concern that, during a desian basis accident (loss of coolant accident (LOCA] concurrent with loss of offsite power (LOOP]), the inadvertent actuation of the FSS might shutdown the ventilation in the affected EDG room, debilitate the EDG, and reduce the accident mitigating equipment to less than the quantity required by the design analysis. These concerns were described in the subject inspection report and discussed with PSE&G during and after the inspection.

On October 4,1996, the licensee issued letter No. LR-N96296 to the NRC. The attachment to this letter described the results of PSE&G's evaluation and the conclusions regarding the concerns expressed by the NRC. The purpose of this inspection was to review the licensee's actions to address the issues and the bases for their conclusions.

b.

Observations and Findinas EDG Room Temperature During the original inspection the NRC observed that the maximum EDG room temperature (120 'F) was specified as bulk temperature and that the setting of the temperature sensors (160 'F) was less than the range (175 - 249 'F) recommended by the National Fire Protection Association (NFPA). The inspector expressed a concern that radiated heat during operation of the EDGs at design basis conditions, combined with nonconservative manufacturing nensor tolerances and air stratification, might raise the temperature of any one of 28 ceiling sensors sufficiently to actuate the FSS and shutdown the ventilation to the affected room.

To address the NRC concern, on April 1,1996, during a scheduled monthly run of the "C" EDG, the licensee took infrared temperature measurements of the seven sensors in the "C" EDG room. Based on the results of these measurements, extrapolated to the design basis temperature of the safety auxiliary cooling system

.

.

(SACS), the licensee concluded that acceptable margin existed between the maximum room temperature and the sensor minimum setting.

The inspector's review of the measurements taken by the licensee observed that:

(1) no procedure had been prepared to conduct the measurements; (2) the work order had minimal information regarding required actions; (3) only three measurements had been taken of each sensor -- before, during, and after the diesel run; (4) no information was available, either in the work order or on the data recorded, as to when, during the diesel run, had the measurement been taken to assess their reasonableness; (5) the difference between the sensors temperature was sufficiently high (approximately 20 *F) to reject the possibility of extending those results to other diesels.

The inspector communicated his observations to the licensee engineering. As a result, during the test of the "D" EDG, PSE&G decided to take air temperature measurements, every five minutes, from a platform high in a corner of the room.

This test was also witnessed by the inspector who noted that the technician assigned to take the thermography of the sensors was not aware and did not take the measurements near the end of the surveillance run, as specified in the work order.

Based on the inspector's observation that a correlation between the air and the sensor temperature measurements was not immediately evident, for the remaining two diesels the licensee took seven measurements, five of which were taken while the diesel was running, j

in the above noted letter to the NRC, PSE&G provided a detailed summary of the results obtained during the EDG surveillance runs and their conclusion that the calculated margin of 28 'F between the maximum extrapolated sensor temperature (124 'F) and the minimum sensor setting (152 'F) was within the industry practice i

for sensor settings in fire protection applications.

The inspector reviewed the results of the measurements taken, the NFPA recommendations, and the licensee's basis for their resolution of the issue and i

concluded that sufficient margin existed between the maximum expected j

temperature in the EDG room and the minimum setting of the sensors. However, PSE&G needed prodding before they were able to take meaningful measurements and develop useful results.

The licensee's failure to develop appropriate tests for the EDG room terr:>erature-constitutes a violation of minor significance and is being treated as a Non-Cited Violation, consistent with Section IV of the NRC Enforcement Poliev.

Fire Suooression System /EDG Room Ventilation Interaction During the original review, the inspector determined that, in the event of an actuation of the EDG room FSS, the fire dampers of the affected room would automatically close and the associated ventilation shutdown. The power for the

.

i

34 FSS equipment and the fire dampers was derived from nonsafety-related (non-Q in the Hope Creek definition) uninterruptible power supplies (UPS). The FSS equipment was seismically qualified, but considered non-Q.

The inspector also determined that signals from the non-Q FSS were used to trip the

,

safety-related recirculation fans, powered by Class 1E sources. Apparently, to

isolate the nonsafety-related FSS contacts from the safety-related fan circuits, the

'

licensee used interposing relays (3ZZ). The power to these relays, however, was derived from nonsafety-related and non-UPS sources.

Because approximately three hours would be required to reopen the fire dampers, following an advertent actuation of the FSS, and an indeterminate time would be required to troubleshoot and reestablish ventilation lost due to a fault in the nonsafety-related portion of the fan circuit, the inspector concluded that the inadvertent loss of ventilation to any EDG would render that EDG ineffective for accident mitigation purposes. Therefore, the adequacy of the design and the acceptability of the observed interactions between safety and nonsafety-related components and circuits were questioned.

in the above noted letter to the NRC, PSE&G stated that the design of the EDG fire suppression system was adequate and no modifications to the system were necessary. In the basis for their position, the licensee did not address the design concerns stated in the NRC inspection report and in meetings with them. Instead, they referred to a number of licensing documents and the NRC acceptance of the design in the October 1984 Safety Evaluation Report (SER). The inspector disagreed with PSE&G's conclusions for the following reasons:

The NRC conclusion, in Section 8.3.3.1.3 of the Hope Creek SER, that "the e

design met the requirements of GDC 17 and was acceptable," was based on PSE&G's response to Open item 243 in the draft SER (DSER). Open item 243 questioned the capability and qualification of the Class 1E system and components to perform their function when subject to the effects of the line orotection system operation design basis event.

Addressing the same event, in their response to the open item PSE&G described an inadvertent discharge of carbon dioxide due to a seismic event in the diesel generator rooms and stated that shutdown of the plant required only two diesel generators in the same mechanical division. Two diesels would not suffice to mitigate the consequences of LOCA/ LOOP event.

Further, the response did not address potential multiple failures of the nonsafety-related 3ZZ relay circuit under any design basis event requiring extended use of the EDGs.

• The NRC acceptance of the " Fire Hazard Analysis" in Section 9.5.1.1 of the SER was based on PSE&G response to DSER Open item No. 218. In their response to NRC concerns regarding compliance with General Design Criterion 3, PSE&G stated that each system had been reviewed for spurious actuation either by seismic-induced error or operator error and that such

_

o

.

actuation would not prevent safe shutdown with the remaining diesels generators.

The licensee response, as above, did not address the LOCA/ LOOP concerns questioned during the subject inspection, nor the failure of the 3ZZ relay circuits. Instead, the response addressed plant shutdown during normal operation or following a seismic event as a result of an inadvertent actuation of the fire suppression system in only one EDG room.

e A correlation between Section 9.5.1.6.30 of the UFSAR, dated April 1988, and section 9.5.1.6 of the SER, " Fire Protection of Emergency Diesel

Generator Rooms", dated October 1984, is difficult to establish, because the i

latter document was not clear regarding the fire suppression design.

Therefore, it is similarly difficult to assume that the NRC acceptance of the SER description also implied an acceptance of the FSAR description.

Further, the NRC acceptance of the design, even in the 1988 description, did not signify their acceptance of design deficiencies not specifically expressed in the submittal; only a detailed review of wiring diagrams would have identified the use of a nonsafety-related source to power auxiliary relays, as found by the inspector, e

Section 8.1.4.14.1.3 of the UFSAR states that "Non-Class 1E circuits or equipment may be connected to Class 1E equipment or circuits by the use of approved qualified isolation devices to assure that in the event of failure of the non-Class 1E equipment, the Class 1E equipment will continue to perform its function." The inspector did not evaluate allinterfaces between safety-related EDGs and auxiliary systems and the nonsafety-related fire suppression system. However, in the case of the Class 1E EDG room ventilation system, although qualified isolation devices (3ZZ relays) may have been used to electrically isolate the safety-related from the nonsafety-related circuits, their application was incorrect (reversed). Therefore, a failure of the non-Class 1E relays 3ZZ could prevent the ventilation system from performing its function. Therefore, the 3ZZ relay circuits did not comply with the above UFSAR statement. This concern was clearly stated in inspection report 96-03.

e PSE&G, in Sections 1.8.1.53 and 8.1.4.10 of the UFSAR, states compliance with IEEE Standard 379-1972 and Sections 5.2 through 5.5 of IEEE Standard 379-1977, respectively. IEEE 379-1972 provided guidence in the application of the single failure criterion for the design and evaluatian of protection systems. IEEE 379-1977 extended the guidance to all Class 1E systems.

Sections 5.2 through 5.5 apply to failures in Class 1E systems.

Section 6.3 of IEEE 379-1977 states, in part, that, "If they (non-Class 1E systems coupled in some manner to Class 1E systems) can degrade any portion of the Class 1E system to the point of failure, the single failure analysis of the Class 1E system shall be oreconditioned by the f ailures that non-Class 1E systems may cause." This statement, although not found in

. _

_

_

_ _

_ _ _

.-

_ _.

_

_. ~

.

IEEE 379-1972, is implied in the analysis of Type 2 and Type 3 single failures of Section 6.5. This statement is also implied by Section

8.1.4.14.1.3 of the UFSAR mentioned above.

Within the context of the above IEEE Standard description, both the FSS and the 32Z relays, coupled mechanically and electrically to the EDG ventilation system, fail to meet the requirements of the standard and, hence fail to comply with the above FSAR statement.

c.

Conclusions Based on the above the inspector concluded that the licensee had adequately addressed the EDG room temperature issue, although several discussions and meetings between the NRC and PSE&G technical personnel were required before j

meaningful measurements were taken. Regarding the interaction between the EDG and the fire suppression system the inspector concluded that the licensee had failed to properly correct a design deficiency identified by the NRC. The identified deviations from the UFSAR constitutes a de facto change to the facility without a safety evaluation to show that the change was not an unreviewed safety question.

Therefore, the deviation is a violation of the 10 CFR 50.59 requirements.

l (VIO 50-354/96-09-03)

E8.2 (Closed) Unresolved item 50-354/96-03-06: Loss of Optical isolator power.

Bailey Opticalisolator Cabinet 10C633 houses the components necessary to separate and isolate safety-related and nonsafety-related circuits. Power to these components is provided by two internal 24 Vdc power supplies. On January 31, 1996, a loss of power to this cabinet affected many nonsafety-related functions in the control room. The licensee's troubleshooting and corrective actions were observed by the NRC during the original inspection. The issue, however, was unresolved pending their completion of a root cause analysis of the event..The purpose of this inspection was to review the results of that analysis and determine the adequacy of the actions taken by the licensee to prevent recurrence.

The root cause analysis was completed on April 25,1996. As also stated in section E-2.1, the licensee evaluated the anomalies observed during troubleshooting and failure modes of affected components. The analysis, was unable to identify the root cause of the event, only probable causes. As a result, potentially defective components were replaced. For instance, a circuit card housing an undervoltage relay was evaluated for defective components and bad solder connections. No anomalies were found. Nonetheless, the circuit card was replaced because a failure of the card, "could not be ruled out with a 100% confidence factor."

Discussions with the licensee revealed their belief that the most probable causes of the event were a loose connection and an excessive voltage drop across some diodes, both of which could have impacted the bus voltage and caused a undervoltage relay to actuate and trip the power supply breakers. They also recognized that the loose connection was not sufficient to cause a voltage drop to

-

A

.

the affected bus and that the voltage drop across the diode was similarly insufficient, even during lamp tests, to reduce the bus voltage below the relay setpoint.

The inspectors' discussions with responsible engineers indicated good understanding of the system and that the potential sources of the problem had been addressed. To prevent recurrence, the licensee acted conservatively and replaced all potentially defective components, even when no deficiencies were identified. in addition, the licensee reevaluated the negative bus undervoltage relay setting and dropped it from 22 V to 19 V. This new setting allowed a larger bus voltage drop without impacting the function of the bus. A safety evaluation of the modified setpoint had been made.

The inspectors evaluated the bast._.or the changes made and the applicable change documentation. They also rev!ewed the revised maintenance plans for the cabinet j

and the equipment and the saf sty function of the cabinet and circuits. They l

-

concluded that the actions ta!.en and those planned for the next refueling outage were acceptable. This item is closed.

E8.3 LQl.osed) Violation 50-354/94-18-01: Inadequate 125 Vdc battery acceptance criteria.

The inspector's review of station procedure for the 18-month service and surveillance testing of the 125 Vdc batteries, identified that the minimum voltage of 105 Vdc specified in the test acceptance criteria did not appropriately include the voltage drop between the battery terminals and the safety-related loads it served.

In their October 13,1994, response letter to the NRC, PSE&G denied the violation.

Nonetheless, they did formulate the corrective actions they would take to prevent recurrence of the identified issue. The NRC reviewed the licensee's response and on January 13,1995, they reaffirmed the violation stating that PSE&G had not provided an adequate basis for withdrawing the violation. The purpose of this inspection was to determine if the licensee completed the corrective actions stated in their October 13,1994, letter and to determine the adequacy of those actions, i

During the current review, the inspectors determined that the licensee had:

o revised Procedure HC.lC-ST.PK-0002(Q), Revision 1, "18 month Surveillance and Service Test of the 125 Volt Batteries Using BCT-2000," dated March 18,1994. This procedure had changed the minimum acceptable voltage to 108 Vde, as found in the applicable battery sizing calculation.

e revised Procedure NC.DE-AP.ZZ-0002(O), Revision 4, " Design Calculations and Analysis," dated December 14,1994. This procedure ensured that revisions to calculations were communicated to the station so that the applicable procedure could be revised.

e revised the UFSAR to reflect the revised minimum voltage of 108 Vd a

.

)

in addition the inspector verified that the findings from Hope Creek's EDSFl had been adequately evaluated for applicability to Salem and that the licensee had

completed a department roll-down of information to reinforce the responsibility of the design and system engineering to communicate design and system changes to appropriate personnel and to ensure that required changes were tracked to completion.

Based on the above review and verification that the actions had been completed and acceptable, this issue is closed.

E8.4 UFSAR Review A recent discovery of a licensee operating their facility in a manner contrary to the UFSAR description highlighted the need for a special focused review that compares plant practices, procedures and/or parameters to the UFSAR descriptions.

While performing the inspections in this report, the inspectors reviewed the applicable portions of the 'JFSAR that related to the areas inspected.

Inconsistencies were noted between the wording of the UFSAR and the plant practices, procedures and/or parameters observed by the inspector, as documented in Section E8.1 of this report.

E8.5 (Closed) Unresolved Item 50-354/94-22-01: reactor protection system actuation -

invalid main turbine trip signal results in reactor scram due to design deficiency.

This issue is further described in LER 50-354/94-14-01 and in NRC Inspection

,

Report 50-354/95-02. The root cause of this event was attributed to inadequate development and implementation of a digital feed watu control system modification in that the original design concept to employ a redundant set of main turbine trip signals was not incorporated correctly. This error allowed a single failure in the logic to trip the main turbine. Additionally, the post-modification testing to prove the function of the design modification failed to verify the design feature was properly installed, contrary to the requirements of 10 CFR 50 Appendix B Criterion XI, " Test Control." The inspectors verified that PSE&G has subsequently employed measures to aid in prevention of future occurrences of this problem (described in the noted LER). This licensee-identified and corrected violation is being treated as a Non Cited Violation, consistent whh Section Vll.B.1 of the NRC Enforcement Poliev."

E8.6 (Closed) Violation 50-354/95-10-03: Hiller-actuated safety auxiliaries cooling system valve reliability problems. The inspectors verified that the corrective actions described in the licensee's response letter, dated March 13,1996, later supplemented by a letter dated September 25,1996, to be reasonable. Specifically, the inspectors confirmed that the licensee plans to replace permanently the subject valves and actuators with a combination better suited for their respective applications. This design change implementation was scheduled to begin in November 1996 and be completed within about one year. Until the time that all the changes are complete, PSE&G has implemented effective ccmpensatory measures

.

_

. _ _. -.

.

_._._ __ __

_

_

_

_

. _ _..

_

_

_._

o

.

l l

I to ensure that the valves and/or the function they provide remain in an operable condition.

l E8.7 (Closed) LER 50-354/94-014: reactor protection system actuation -invalid main l

turbine trip signal results in reactor scram due to design deficiency. This event was discussed in detail in NRC Inspection Report 50-354/94-22 and 50-354/95-02, and further addressed in the closure of Unresolved item 50-354/94 22-01 above. No new issues were revealed by this LER.

IV. Plant Supogri

'

R1 Radiological Protection and Chemistry (RP&C) Controls

R 1.1 Exoosure Performance By reference to NRC annual exposure data, the inspector determined that Hope Creek ranked 10th lowest out of 26 BWR sites for 1995 (with an annual exposure of 196 person-rem) and achieved the same ranking over the 5-year period from 1991-1995. Licensee data indicated that the 1995/1996 refueling outage lasted 136 days (versus 30 days) and resulted in 262 person-rem versus an original estimate of 165 person-rem.

R1.2 RP Resoonse to an Elevated Radic. active Gas Event a.

Scope The inspector reviewed the licensee's response to a September 6,1996, Radiation Monitoring System (RMS) alarm indicating elevated airborne radioactivity in the p! ant's turbine building compartments. Tours of the facility, review of docurnentation, and interviews with applicable RP & C staff were conducted.

b.

Observations and Findinas On September 6,1996, a control room RMS alarm indicated that the turbine building compartment exhaust had reached the alert level. The RP organization was notified and began an investigation. Over the next 6 days, an aggressive investigation was conducted that included surveys inside the plant and outside of plant buildings to determine the cause of the continued elevated turbine building compartment exhaust readings. The RP organization provided a broad scope review and evaluation of possible causes, which included the possibility of offsite release

that was evaluated at the outset of the event. The results of the RP surveys did not

,

determine any elevated airborne radioactivity either offsite or in any of the plant

j turbine building compartments.

l There had been an ongoing leak in the B steam jet air ejector (SJAE) room since

'

j early summer of 1996 that was isolated on September 12,1996, by back-seating a l

l i

i l

I

- - _

._.

_ -.

_

.

.

_

_

.

-_

-

---

-

-

.

.

_

-

l

,

l l

l valve. After isolating the leak, the turbine building compartment exhaust l

radioactivity level trended back down to normal.

l

!

The inspector's review indicated that the licensee initiated a prompt and detailed I

investigation of elevated turbine building airborne radioactivity readings. Although a very broad evaluation was performed, which included early evaluations of potential offsite releases, the licensee did not identify through in plant measurements, the source of the radioactive gas leakage. Three days into the event, a noble gas grab

!

sample was obtained from the B SJAE room, however, no radioactive gas was detected. The inspector observed that all of the continuous air monitors (CAMS)

located in the plant are particulate CAMS with an available charcoal grab sample that c:n be replaced and counted on an as-needed basis. After the event was

,

tarminated, it was determined that noble gas and steam were being released into the B SJAE room. The station CAMS were not capable of detecting this radioactivity. The licensee did not identify this limitation in the event evaluation.

During operations, system leaks may more easily be detected by multi-channel CAMS that indicate relative levels of particulate, iodine, and noble gas. The RP Lead Supervisor indicated that the need for multi-channel CAMS would be

,

'

evaluated.

The licensee's evaluation of the event included several findings. After the control room alarm occurred, an operator switched the turbine building compartment exhaust to the alternate exhaust train. The operations procedure required this

,

action indicating that the alternate exhaust train was a filtered pathway. Several operators that were interviewed by the licensee also believed that the alternate

'

exhaust train contained filters. In fact, the alternate exhaust train did not contain any filters, which was consistent with the UFSAR and P&lDs. Although the alternate turbine building compartment exhaust train was originally designed to -

,

include both charcoal and HEPA filters, they were not installed.

I c.

Conclusions The inspector's assessment of the licensee's response to the event was mixed.

There was a broad scope investigation of the cause of the event and the event itself j

was a fairly low increase in airborne radioactivity that did not represent a i

measurable increase in effluent readings exiting the plant through the south plant vent. The RP organization conducted a thorough investigation for a relatively low i

safety significant event. However, the RP organization was not able to measure the noble gas within the confines of the station and determine its source during the 6 days of their investigation due to equipment limitations. Further, an operator knowledge issue was discovered regarding the design of the alternate exhaust train,

.

in that no filters are available even though procedures require the action.

l i

!

i

.

.

R1.3 Internal Exoosure Measurements a.

Scope The inspector reviewed the adequacy of air sample analysis and direct bioassay measurements through a review of analytical data and by conducting interviews

with cognizant licensee personnel.

b.

Observations and Findinas i

The inspector reviewed the licensee's radiochemical analysis results for the 10 CFR 61 dry active waste (DAW) waste-stream that was sampled on February 7,1996, j

utilizing smear samples collected throughout Hope Creek S.ation during the previous j

refueling outage. This was the first time that DAW was established as a separate waste stream. Previously, samples of waste sludge were utilized to charactenze J

Hope Creek's contamination radionuclide constituents. The results of the radiochemical analysis indicated that only 10% of the radioactivity was measurable by Hope Creek RP instrumentation. Ninety percent of the activity resulted from tron-55. This corresponds to 11% of the derived air concentration (DAC) as not being accounted for using current gamma spectral analysis and DAC determination calculations used for air sample counting. This could have the potential of underestimating internal exposure assignments by 11 %. The Lead RP Supervisor agreed that the air sample analyses could be more accurate by including the unmeasured radioactivity component and committed to making the applicable program changes.

The inspector reviewed with the RP staff the current status of onsite direct bioassay measurement capability for making internal exposure assessments. During the last inspection in this area (Inspection Report No. 50-354/96-05), the licensee had established a memorandum of understanding with Brookhaven National Laboratory in Long Island, New York for providing investigational whole body counts, if required. At that time, the licensee had also indicated that several other options were being considered to allow for more expedient bioassay measurements to be

,

performed. During this inspection, the Manager of Technical Support indicated that

'

the existing germanium whole body counter would be calibrated within the next 45 days in order to support the Salem steam generator replacement project, as necessary. Additionally, options to upgrade the equipment and/or the analysis software were also being pursued.

c.

Conclusions The inspector determined that the licensee has committed to taKO actions neCessary to more accurately provide the necessary internal exposure measurements in both l

air sample analysis and in direct bioassay to improve the adequacy of the internal l

exposure assessment program.

I

,

.

R2 Status of RP&C Facilities and Equipment During this inspection, the inspector conducted numerous tours of the plant during operating conditions and noted that all required radiological postings and locked i

areas met regulatory requirements, and that the areas were clear of unnecessary equipment, wellilluminated and free of safety hazards.

R2.1 Turbine Buildina Comoartment Exhaust Filtration System a.

Scone Due to the licensee's evaluation of an elevated radioactive gas release event, it was determined that an operations procedure directed the operator to switch exhaust trains to the filtered train. The plant configuration and UFSAR indicate that the alternate turbine building compartment exhaust train has the provision for charcoal and HEPA filters, however, they do not currently exist. The inspector reviewed the

plant design basis to determine the safety significance of the current plant configuration.

b.

Observations and Findinas The inspector found that the "as-built" drawing for the alternate turbine building exhaust system as described in the UFSAR clearly indicates that the filters are "to

be installed later." In addition, the inspectors verified that the filters are, in fact, not present in the filter bed compartment of the exhaust system. While the operating procedure and knowledge of some operators were inconsistent with this fact, the

inspectors found that the actual configuration agreed with the UFSAR.

The inspector reviewed a construction installation specification dated February 17,

,

1976, that indicated plans for incorporating prefilters, heating coil, HEPA filters, charcoal absorber bank, and postfilters into the alternate turbine building compartment exhaust train. The inspector also reviewed an evaluation conducted by the licensee during the licensing stage of Hope Creek for compliance with 10 CFR 50, Appendix l that was submitted to the NRC on June 1,1976, that indicated without the subject alternate filter train, Hope Creek Generating Station would produce iodine and particulate effluent estimated to result in 0.17 mrem per year to the nearest milk cow offsite, which represents 1.1 % of the 10 CFR 50, Appendix I limit of 15 mrem per year. The inspector reviewed the Hope Creek 1995 Annual Radioactive Effluent Release Report, which indicated iodine and particulate releases representing 0.7% of the Appendix I limits.

c.

Conclusions The licensee's effluent release design basis does not take into account the provision t

l for a filtered turbine building compartment exhaust and plant release calculations, and operating experience indicates very low particulate and iodine release levels.

'

No discrepancies were noted.

I

!

l l

.

. _

.

..-

-

-. -

-. _... -.

.

-

.

o

,

.

l R2.2 Dosimetry Calibrations l

a.

Scone l

The inspector reviewed the licensee's program for calibration of electronic pocket dosimeters (EPD). The inspector toured the licensee's calibration laboratory,

'

reviewed calibration documentation, and conducted interviews with the instrument calibration staff, b.

Observations and Findinas I

The inspector observed the calibration methodology of Ainor EPDs. The licensee utilized a 3 curie cesium-137 source and has established fixed geometries to allow for timed exposures of 70 mrem and 1500 mrem for calibration purposes. The inspector determined that the source had been calibrated on July 1,1996, with

'

appropriate NIST transfer instruments and utilizing an acceptable methodology.

Licensee procedure calls for 6-month calibration periodicity for EPDs, which is met through automatic computer review of calibration dates as each EPD is being activated by a user in the plant.

c.

Conclusions The licensee has implemented a good program for ensuring EPDs are accurately calibrated and maintained.

R2.3 Respiratorv Protection a.

Scope The inspector toured the respirator processing facility, examined documentation of air quality analyses, and conducted interviews with cognizant licensee personnel.

i b.

Observations and Findinas The inspector observed a well maintained respirator processing facility that provides for the necessary contamination controls and inspection / testing required prior to authorizing respiratory protection equipment for reissue. Features of the program include: respirator quantitative leak testing after every washing and respirator particulate canister leak and flow testing after every use. Air supply bottles are filled using an oil-lubricated air compressor. The air quality is tested prior to every use using draeger tubes and quantitative air analysis is provided semi-annually through a vendor. The inspector reviewed documentation demonstrating that the semi-annual air quality tests had been performed.

c.

Conclusions l

The inspector determined that the respirator processing facility was an excellent

>

facility that was well maintained.

I

e

.

R5 Staff Training and Qualification in RP&C R 5.1 B_P Technician Trainina a.

Scope The inspector reviewed the licensee's 1996 continuing training program for RP technicians through a review of documents and the conduct of interviews with the training and RP staff.

)

i b.

Observations and Findinas Since January 1996, the RP training program was redirected to replace the training laboratory components of training with onsite on the-job-training (OJT) components.

This change of focus began in March 1996 with a pilot of the new OJT program and provided approximately 4 days of continuing training on the following topics:

operation of the IPM-9; operation of the SAM; respirator test unit operatiors management expectations related to RMS and effluent release levels; and, electronic aids for RWPs and survey documentation. During the fall of 19bS, approximately 3 additional days of training are planned to cover the following subjet.?s: radwaste shipping regulation update; liquid radwaste effluent discharge process; high risk job i

coverage industry events; and RMS skid operations. Eight cycles of continuing training are scheduled to ensure all shifts of RP technicians attend the continuing training classes /OJT. The selection of curriculum results from monthly meetings of the training review group, which is chaired by the Hope Creek Radiation Protection Manager. This group is made up of training instructors, RP supervision, and RP technicians from both Salem and Hape Creek Stations.

The training review group identified the need for refresher training prior to rotating into certain RP specialist positions. In response to this suggestion, the RP training instructors developed an OJT module for shift RP technicians and for emergency drill RP response to meet the identified needs.

The inspector observed that the continuing training curriculum did not typically contain any review of RP fundamental subject matter and that since initial RP technician qualification, there has been no evaluations made to determine the continued level of knowledge and skill pertaining to the qualifications.

c.

Conclusions The inspector determined that the licensee has a strong and viable RP training program. The OJT aspect of the program is viewed as a good enhancement to the program. Also, the development of several refresher training modules for rotating l

RP staff was considered a good initiative. One area of enhancement was identified l

by the inspector. The RP fundamental knowledge and skill level of the staff is not (

periodically evaluated, and fundamental RP subjects are not typically included in the

'

continuing training curriculum. The licensee indicated that this subject would be i

included and evaluated in the next training review group meeting.

!

._

_ _ _.

_...

.

__ _ _ _ _

_

_. _ _ _ _ _

__

.. __

e

.

R7 Quality Assurance in RP&C Activities R7.1 Quality Oversicht of RP Activities a.

Scope The inspector reviewed the licensea's audits and assessments of the RP program during 1995 and 1996.

b.

Observations and Findinas The inspector reviewed the latest RP audit (No.95-150, conducted in June 1995).

This biennial scheduled RP audit included outside technical participation and was sufficiently broad and detailed. In general, the audit determined that the RP program was effectively implemented and there were no significant findings.

,

The inspector reviewed the QA surveillance observations conducted during 1995-1996 and found occasional surveillance observations were conducted through January 1996, with a renewed emphasis with monthly QA assessments provided since April 1996. These more recent QA assessments have been very effective.

Two surveillance observations [a review of an emergent maintenance task and a review of the Radiological Occurrence Report (ROR) program] were of high quality, in addition to the independent QA audits and surveillance observations, the RP group has begun implementing a self-assessment program since April 1996. A

'

review of these self-assessments indicated that few results for improvement have yet been obtained, c.

Conclusions The inspector determined that the RP group has a very effective combination of QA audits and surveillance activities. Implementation of a program of RP self-assessments has been underway since April 1996 that has provided limited i

additional value to improving the quality of the RP program.

'

R7.2 Radioloaical Occurrence Reports l

a.

Scoce The inspector reviewed the licensee's documentation and resolution of radiological

!

problems experienced since the fall of 1995 through September 1996. Interviews with cognizant licensee personnel were also conducted.

,

l l

b.

Observations and Findinas

,

During the 4-month refueling outage that ended in late March 1996, there were 178 RORs recorded. These consisted of 77 personnel contaminations, 58 dosimetry

,

problems, and 43 other issues. These resulted in 14 action requests requiring l

-_

_

-

-

-

...

-.

.

.

.

resolutions. The action request system is a station-wide system that provides a mechanism for corrective action assignment and resolution tracking to be completed. Two action requests included the uncoordinated actions of different

,

scaffolding groups that resulted in rebuilding scaffolding which incurred additional

exposure and a radioactive liquid spill in the drywell. The action request actions for these issues was comprehensive and effectively resolved in a timely fashion. Most

'

of the RORs represented issues of low safety significance. Since the outage, there l

have been 30 RORs. There were several that represented opportunities for saving

some unnecessary personnel exposure and a few captured EPD hardware problems.

!

The licensee indicated that Salem and Hope Creek stations were tasked with development of a common ROR program. Non-significant personnel contaminations

'

are to be stripped from the ROR program and would be trended separately.

c.

Conclusions

!

The inspector determined that the RP organization has a very good, low threshold for capturing radiological occurrences and there have been very few safety

significant radiological occurrences. The use of the station-wide Action Request program to ensure significant RORs are resolved and actions completed was considered an effective vehicle, implementation of the ROR program has been effective and received very good management involvement.

R8 Miscellaneous RP&C issues R8,1 (Closed) Violation 50-354/96-03-02:

This violation involved multiple RCA entry procedure violations due to the improper use of electronic pocket dosimeters. The licensee's response to the violation

included: a thorough investigation of the procedure violation instances; effective root cause determinations; and proposed several corrective actions that were verified during this inspection. These corrective actions include installation of positive-locking personnel tuinstiles at the entrance to the RCA that requires possession of an activated electronic personnel dosimeter for entry. Several modifications were made to the RCA access control software to activate the EPD alarm if the EPD is left in the EPD reader too long on either entry or exit from the RCA. In addition, several human factor enhancements have been made to the RCA entry / dress out area to facilitate RCA entry procedure adherence. The major enhancement was the relocation of the RCA entry point located after the electronic turnstile and after exiting from the dress out area. Based on verification of the above corrective actions, this violation is closed.

R8.2 (Closed) Unresolved 50-354/96-05-03:

This unresolved item involved the adequacy of the licensee's limiting the expo:,ure monitored population of radiation workers. The licensee had established that Category 1 radiation workers were not likely to receive 10% of the annual exposure limit and were not issued TLDs. They were required to wear EPDs in the RCA,

. _.

_

_ - -,

.

..

. _ -

,

-

.

,

l

.

however, and the tracking of the EPD dose demonstrated compliance with the regulations. Previous procedure violations of workers at both Salem and Hope Creek stations involving RCA entries without functioning EPD and with accessibility to radiation areas in the station, questioned the adequacy of personnel monitoring l

for Category 1 workers based solely on EPDs.

Due to the installation of positive-controlled electronic turnstiles at both Hope Creek and Salem stations, the inspector determined that the current risk of unmonitored radiation worker exposure was very low and this unresolved item is closed.

Additional review by the licensee has determined that all radiation workers that routinely enter the RCA should be issued record TLDs and by the end of 1996,

'

Category 1 and Category 2 classifications will be abolished. All radiation workers will be considered monitored and will be issued TLDs.

R8.3 Uodated Final Safety Analysis Report

,

a.

Scoce (83750)

The inspector reviewed current Hope Creek Station practices with respect to selected portions of Sections 12.3 and 12.5 of the UFSAR.

b.

Observations _pnd Findinas The inspector toured the plant and verified that the CAM locations specified in Section 12.3 of the UFSAR were accurately represented in the plant. The insp ector noted that they were particulate monitors only and that they have limitations for the monitoring of noble gases as was described in Section R1.2 of this report, c.

Conclusions All sections of the UFSAR reviewed were accurately reflected in the plant.

S1 Conduct of Security and Safeguards Activities During the course of a CAS extended observation, the inspector observed the conduct of a perimeter alarm surveillance test; personnel and equipment response to actual and test alarm conditions; and the conduct of operations with radio communication systems degraded due to the use of one of the radio frequencies in support of a Salem emergency response drill. The inspector concluded that CAS operations were good and that the security equipment used was in good working

.

condition.

I

o

.

V. Manaaement Meetinas X1 Exit Meeting Summary A recent discovery of a licensee operating their facility in a manner contrary to the UFSAR description highlighted the need for a special focused review that compares plant practices, i

procedures and/or parameters to the UFSAR descriptions. While performing the l

inspections discussed in this report, the inspectors reviewed the applicable portions of the UFSAR that related to the areas inspected. Except as noted in Sections E8.1 and E8.4 of this report, the inspectors verified that the UFSAR wording was consistent with the observed plant practices, procedures and/or parameters. The exceptions resulted in a violation of NRC requirements pertaining to the design of safety related circuits associated with the emergency diesel generators, as described in the Notice of Violation that accompanies this report.

The inspectors presented the inspection results to members of licensee management at the conclusion of the inspection on November 19,1996. 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.

i X3 Management Meeting Summary On October 16,1996, William Kane, Deputy Regional Administrator for Region I and Richard Keimig, Chief of the Emergency Preparedness and Safeguards Branch, Region I, met with representatives of both Salem and Hope Creek and the Licensee's Security Organization. Facilities at both stations were toured and interviews with select staff and management were conducted to ascertain the seriousness of corrective actions for a series i

of security related events.

i I

l i

-.

. - - -.

.

.

.

-

-

-

- -

. -.

-

o

.

e i

INSPECTION PROCEDURES USED IP 37550:

Engineering IP 37551:

Onsite Engineering (

IP 40500:

Effectiveness of Licensee Controls in identifying, Resolving, and Preventing l

Problems j

IP 61726:

Surveillance Observations

'

i IP 62707:

Maintenance Observations IP 71707:

Plant Operations IP 83750:

Occupational Radiation Exposure ITEMS OPENED, CLOSED, AND DISCUSSED Opened i

50-354/96-09-01 VIO failure to adequately staff the OSR and inadequate OSR reviews of LCRs 50-354/96-09-02 URI SACS design basis concern regarding pump runout 50 354/96-09-03 VIO failure to conduct 50,59 review for de facto design change for fire suppression systems interaction with Class 1 instrumentation Closed 50-354/94-18-01 VIO inadequate 125 Vdc battery acceptance criteria 50-354/94-09-02 URI engineered safety features actuations 50-354/94-19-01 URI reactor protection system and engineered safety system actuation 50-354/94-19-03 VIO fuse configuration control deficiency 50-354/94-22-01 URI reactor protection system actuation -invalid main turbine trip signal results in reactor scram 50-354/95-01-01 VIO failure to implement TS action statement requirements 50-354/95-01-02 VIO noncompliance with the condition requirements of TSs 50-354/95-10-02 VIO missed TS required surveillance tests for TIP explosive squib valves 50-354/95-10-03 VIO Hiller-actuated safety auxiliaries cooling system valve reliability problems 50-354/96-03-01 URI surveillance testing of automatic depressurization system valves 50-354/96-03-02 VIO failure to use electronic pocket dosimeters on RCA entries 50-354/96-03-04 URI EDG ventilation / fire suppression systems interface 50-354/96-03-06 URI loss of optical isolator power 50-354/96-05-03 URI evaluate the adequacy of limiting the exposure monitoring population of radiation workers.-

50-354/94-012 LER reactor protection system and engineered safety system actuation 50-354/94-014 LER reactor protection system actuation -invalid main l

turbine trip signal results in reactor scram t

50-354/94-004 LER MSIV actuation l

50-354/94-005 LER RHR system isolation 50-354/94-006 LER engineered safety feature actuation

'

50-354/94-007 LER reactor scram

.

....... -. -,... - -

.

. - -. - ~. -. -

. -. - - - - - _

.

. -... -.. ~.

. - - _

.

i o

. *

,-

. I

l

50-354/94-008 LER condition prohibited by technical specifications-50-354/94-009 LER remote shutdown system control deficiency

50-354/96-011 LER technical specification 3.0.3 entry

i f

i i

i t

,

i

,

'

s i

i I

i

-!

'

f j

l

,

-

t r

,

l

!

I t

i l

.-

i (-

l i

!

i t

!

'

.

.

..

-

-

-

._ -

.

=

. _ _.

.

.

---

- -.

.

a

.

'

PARTIAL LIST OF ACRONYMS USED l

AOT Allowed Outage Time l

AR Action Request CAMS Continuous Air Monitors

!

l CARB Corrective Action Review Board CBD Configuration Baseline Document i

l CRs Condition Reports CRVS Control Room Ventilation System DAC Derived Air Concentration

,

l DAW Dry Active Waste DBD Design Basis Document DCR Design Change Request DlTs Design, inspection and Testing System EPD Electronic Pocket Dosimeters EQ Environmental Qualification ESF Engineered Safety Feature l

FAOS Follow-up Assessments of Operability

!

FSS Fire Suppression System GL Generic Letter HMD Hope Creek Maintenance Department HVAC Heating Ventilation and Air Conditioning l

lPM-9 Installed Personnel Monitor - 9 l

lST Inservice Test LCR License Change Request MCC Motor Control Center MMIS Managed Maintenance Information System MSIV Main Steam Isolation Valve NFPA National Fire Protection Association l

NRB Nuclear Review Board i

NRC Nuclear Regulatory Commission OJT On The Job Training OSR Offsite Safety Review l

P&lD Piping and instrumentation diagram PDR Public Document Room PM Preventative Maintenance PSA Probabilistic Safety Assessment l

l QA Quality Assurance l

RCA Root Cause Analysis l

RCIC Reactor Core Isolation Cooling RMCS Reactor Manual Control System RMS Radiation Monitoring System RO Reactor Operator ROR Radiological Occurrence Report RP&C Radiological Protection and Chemistry Controls RSCS Rod Sequence Control System SA Self Assessment

.

SACS Safety Auxiliaries Cooling System

,

SER Safety Evaluation Report x

'

SJAE Steam Jet Air Ejector

_--

>

l

i

.

.

e

SLC Standby Liquid Control SORC Station Operations Review Committee SRG Safety Review Group i

SWS Service Water System TACS Turbine Auxiliary Cooling System TLD Thermoluminescent Dosimeter TMOD Temporary Modification TS Technical Specifications TSSIP Technical Specification Surveillance improvement Project UFSAR Updated final safety analysis report USO Unresolved Safety Question

,

!

)

l i

l

,