IR 05000272/1996013
| ML20135D201 | |
| Person / Time | |
|---|---|
| Site: | Salem  |
| Issue date: | 12/03/1996 |
| From: | NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I) |
| To: | |
| Shared Package | |
| ML18102A646 | List: |
| References | |
| 50-272-96-13, 50-311-96-13, NUDOCS 9612090289 | |
| Download: ML20135D201 (34) | |
Text
- -
.
- - -
.- -
--
.
.
.
.
.
--
.
'
U, S. NUCLEAR REGULATORY COMMISSION
REGION I
!
Docket Nos:
50-272, 50-311 License Nos:
.
Report Nos:
50-272/96-13, 50-311/96-13 i
Licensee:
Public Service Electric and Gas Company
Facility:
Salem Nuclear Generating Station, Units 1 and 2
,
Location:
Hancocks Bridge, NJ
]
l Dates:
August 12 - September 20,1996 Inspector:
A. Della Greca, Sr. Reactor Engineer, EEB, DRS R. Bhatia, Reactor Engineer, EEB, DRS B. Smith, NRC Contract Engineer M. Gareri, Controls Engineer, NRR M. Shuaibi, Systems Engineer, NRR Approved by:
William H. Ruland, Chief Electrical Engineering Branch Division of Reactor Safety l
I i
-
9612090289 961203 PDR ADOCK 05000272 G
.
d TABLE OF CONTENTS PAGE EX EC U TIV E S U M M A R Y............................................. iv R e p o rt D e t a il s.................................................... 1 Ill. Engineering
...................................................
E2 Engineering Support of Facilities and Equipment....................... 1 E2.1 Introduction
...........................................
E2.2 ' NRC Restart issue 11.25 - Inadvertent Auxiliary Spray (OPEN)......... 1 a.
Inspection Scope................................... 1 b.
Observations and Findings............................ 1 c.
Conclusions
.....................................
E2.3 NRC Restart issue 11.40 - 0 /erhead Annunciator Failures (CLOSED)
....
a.
I nspection Scope................................... 3 b.
Observations and Findings............................ 4 c.
Conclusions
......................................
E2.4 NRC Restart issue 11.7 - Emergency Diesel Generator has Minimal Load Margin (OPEN)........
.................................
a.
Inspection Scope.
.................................
b.
Observations and Findings............................ 6 c.
Conclusions
......................................
E2.5 NRC Restart item 11.11, EDG Load Fluctuations (CLOSED)........... 7 a.
I nspectio n Scope................................... 7 b.
Observations and Findings............................ 7 c.
Conclusions
'
......................................
E2.6 EDG Governor Configuration Control.......................... 9 E2.7 NRC Restart item lli.11, Engineering Problems Resolution, including Safety Evaluations (OPEN)
.................................
a.
Inspection Scope (37001)
............................
b.
Observations and Findings........................... 10 c.
Conclusions
.....................................
E2.8 Offsite Safety Review (OSR) Group.......................... 13 a.
Inspection Scope.................................. 13 b.
Observations and Findings........................... 13 c.
Conclusions
.....................................
E8 Miscellaneous Engineering issues................................ 15 E8.1 (Closed) Deviation 50-272 and 50-311/93-82-07................ 15 a.
Inspection Scope.................................. 15 b.
Observations and Findings........................... 15 c.
Conciasions
.....................................
,
I ii I
i
,
l
.
- -.
..-_
_
.
.
__
-
.
.
. _ -._.
. -.
,
.
.
l Table of Contents i
Paae
,
j E8.2 (Updated) Unresolved item 50-31 1/9 3-8 2-04................... 18 a.
inspection Sc ope.................................. 18 l
b.
Observations and Findings........................... 19.
c.
Conclusions
.....................................
E8.3 (Closed) Unresolved item 50-272; 50-311/93-82-01.............. 20 a.
inspection Scope.................................. 20 b.
Observations and Findings........................... 20 c.
Conclusions
.....................................
E8.4 (Closed) Unresolved item 50-272/93-82-16 EDG
................
a.
Inspection Scope.................................. 21 l
b.
Observations and Findings........................... 21 l
c.
Conclusions
j
.....................................
l-E8.5 (Closed) Violation 50-272/93-82-14
.........................
,
E8.6 (Closed) Violation 50-272; 311/94-07-01
i
.....................
\\
,
a.
Inspection Scope.................................. 22
l b.
Observation and Findings............................ 22 c.
Conclusions
.....................................
l E8.7 (Closed) Unresolved item 50-271 ; 311/96-01 -08................ 23 l
E8.8 (Closed) Unresolved item 50-27 2/94-18-02.................... 24 E8.9 General Conclusions
....................................
l E8.10 Review of UFSAR Commitments............................ 25 l
V. M a na g e m e nt M e e ti ng s........................................... 25 t
XI.
Exit Meeting Summ ary........................................ 25 PARTI AL LIST O F ATTEN D EES....................................... 26 Attachment 1
...................................................
LIST O F ACRO NYM S USED......................................... 29 l
l
I i
l i
1 e
lii i
l l
=.-
.-
.
_ -.. - -
. -. _
..
_
_ _
1.
.
!
EXECUTIVE SUMMARY Salem Inspection Reports 50-272/96-13; 50 311/96-13 August 12,1996-September 20,1996
,
This inspection included aspects of licensee engineering. The report covers a 6-week l
period of inspection related to equipment and engineering performance issues that require resolution prior to Salem restart. These issues are included in Checklists ll and Ill.a of the i
NRC restart action plan, t
t Based on their review of five closure packages and eight unresolved items and violations, the inspectors concluded that:
i e
Quality of engineering issues continued to show inconsistencies.
'
t Technical issues were in most cases acceptably resolved, with acceptable i
e engineering evaluations. Some issues lacked the required depth of review.
e Lack of attention to detail resulted in two procedural violations involving failures to
,
document an unqualified cable termination and to review and approve changes to a safety evaluation.
e High visibility and safety significant root cause analyses received the required level of engineering and management attention and resulted in good products.
e The revised 50.59 process was acceptable, but too new to evaluate its l
effectiveness.
!
i l
'
l i
,
)
iv
- _. -
-
-
--_
-
-
. ~-
--
- -_
i
.
Report Details Ill. Enoineerina
E2 Engineering Support of Facilities and Equipment l
E2.1 Introduction l
On February 23,1996, the NRC issued the restart action plan for Salem Units 1 l
'
and 2. Restart Issue Checklists ll and Ill.a include 43 technical and 22 programmatic issues, respectively, that require resolution. These issues, related to NRC concerns regarding equipment performance problems and plant personnel
,
I issues, involved previously identified unresolved items and violations as well as generic concerns. The purpose of the current inspection was to review the closure l
packages prepared by the licensee to address these issues. Except as noted, the l
review was conducted in accordance with inspection procedure 92903.
!
E2.2 NRC Restart Issue ll.25 - Inadvertent Auxiliary Sorav IOPEN)
a.
Insoection Scoce On June 7,1995, during a Technical Specification (TS) 3.0.3 required shutdown of Salem Unit 2, complications resulted in a reactor trip and a loss of reactor coolant pumps (RCPs) 23 and 24. With RCPs 23 and 24 not in operation, auxiliary spray l
was used to control pressurizer pressure. Accordin0 to control room logs, auxiliary spray was in service for approximately 12 minutes during which the temperature difference between the spray fluid and the pressurizer reached a maximum of 520 F. TS 3.4.10.2 limits this temperature to a maximum of 320*F. The purpose of this inspection was to review the actions PSE&G took to address the potential effects of the event on the pressurizer and spray nozzle and to prevent its recurrence.
l
!
b.
Observations and Findinas l
l PSE&G's evaluation of the June 7,1995, event determined that emergency
,
operating procedure 2-EOP-Trip-2, Revision 13, Step 12 allowed the operators to
!
initiate auxiliary spray with letdown isolated in conflict with the Westinghouse Owners Group (WOG) Emergency Response Guidelines (ERG). The licensee performed a review of their EOPs and found other conflicts between their EOPs and the WOG's ERGS. They determined that these conflicts resulted from an inadequate EOP review and was taking steps to address this issue through programmatic issue P-15, Adequacy of Operating Procedures, (P-15 is item Ill.a.15 of the NRC restart i
plan for Salem). Two of the conflicts identified, one in Procedure 2-EOP-Trip-2, Revision 13, and the other in Procedure 2-EOP-Trip-3, Revision 10, allowed operators to initiate auxiliary spray with letdown isolated. By the end of the inspection period, revisions to these procedures were in their final approval phase.
,
!
!,
.
.
PSE&G's evaluation of the effects of the auxiliary spray on the pressurizer concluded that the event was within the design boundaries of the pressurizer. This conclusion was based on Westinghouse equipment specifications No. 676440, Revision 4, " General Pressurizer Vessel Assembly," and No. 677230, Revision 4,
" Addendum to Equipment Specification 67440 Rev. 4," both of which stated that the pressurizer was designed to withstand ten inadvertent auxiliary spray occurrences over the life of the plant, with a spray nozzle delta-T of 560 F. The licensee's evaluation noted that this was the first inadvertent auxiliary spray occurrence at Salem Unit 2 and that the maximum delta-T reached was 520*F.
In reviewing these specifications, the inspector noted that the event described in the above specifications as inadvertent auxiliary spray had a duration of 300 seconds. The actual event of June 7,1995, lasted approximately 12 minutes (720 seconds), in addition, the 520*F delta-T was based on control room logs which do not account for instrument accuracy. The inspector discussed these discrepancies with PSE&G Engineering. They stated that their conclusion that the Westinghouse analyses bounded the Salem Unit 2 event was based on Figure 18 of equipment specification 676440, which showed that the normalized pressurizer insurge rate stabilized to a value of approximately 210 gallons per minute at 300 seconds.
After further discussions about potential effects of the event on the spray nozzle, PSE&G contacted Westinghouse to obtain additional documentation in support of their conclusions. The licensee was not able to provide such documentation during this inspection, but indicated that they would pursue this aspect of the event. In addition, PSE&G stated that they would perform a visual inspection of the spray nozzle and associated piping within the pressurizer prior to restart of Unit 2.
The inspector asked the licensee if they had reviewed previous inservice inspection (ISI) data on the pressurizer spray nozzle (e.g., inner radius and safe-end welds) to ensure that any previously identified flaws and evaluation of such flaws were not affected by the event. The licensee stated that they had not performed such a review. A subsequent PSE&G search of the ISI database determined that no flaws had been previously identMied and recorded.
During their search of the ISI database, PSE&G was unable to fmd data for the inner
)
radius examination of the 4-inch pressurizer spray nozzle (Component ID 4-PSN-1231-IRS). Per TS 3.4.11.1, Surveillance Requirement (SR) 4.0.5, and ASME Section XI,1974 through the 1975 Addenda (Unit 2's first ISI interval code), this examination was required to have been completed during the first ISI interval which ended May 1992. Apparently, the examination was recorded as having been completed in the Fall of 1988, but PSE&G's review determined that it had not been performed. PSE&G attributed the incorrect data entry to personnel error and initiated Action Request 960815249 to track resolution of the issue. They plan to inspect the 4-inch spray nozzle inner radius in accordance with ASME Section XI, 1986 Edition (the current Unit 2 ISIinterval code) prior to restart of Unit 2. In addition, they plan to perform a search of Salem Units 1 and 2 and Hope Creek's first ISI interval records for similar occurrence !
.
.
c.
Conclusions
,
Based on the above review, the inspector concluded that PSE&G's actions to
,
address the potential effects of the event on the pressurizer and spray nozzle were
'
insufficient to close the issue in that they had not:
evaluated the effect of the longer than analyzed duration of the inadvertent auxiliary spray on the pressurizer and spray nozzle, J
verified the actual maximum delta-T reached during the event;
conducted a visualinspection of the pressurizer spray nozzle and associated piping; performed an ISI inspection of the pressurizer spray nozzle inner radius in
accordance with the ASME Section XI,1986 Edition;
and finalized the necessary emergency procedures revisions.
This item remains open pending the licensee's completion of these activities and the NRC review of their acceptability.
The licensee's failure to perform the required ISI inspection is a violation of the plant Technical Specification requirements. (Violation No. 50-311/96-13-01)
E2.3 NRC Restart issue 11.40 - Overhead Annunciator Failures (CLOSED)
a.
Inspection Scope This item was opened to track the licensee's followup and resolution of issues raised from the December 12,1992; October 4,1995; May 8,1996; and
,
May 24,1996, events involving overhead annunciator failures, j
i The original Salem 1 and 2 control room annunciator system was replaced with a digital computer-based annunciator system. This new system has failed on multiple occasions leading to plant alerts and loss of confidence by the operators. The most serious failure mode involves system lockup in which new alarms are not indicated, and the operators are not warned that the system has failed. The purpose of this inspection was to review PSE&G's actions to address the overhead annunciator failures.
_
_
...~ -
- -- -
. -_ -.- -
-
-- -
-
.
.
i
,
b.
Observations and Findinas-l The new overhead annunciator (OHA) system is a high speed distributed data acquisition and display system. Nonredundant input scanners collect alarm contact state data for redundant main and auxiliary controllers. The combination of scanners, main and auxiliary controllers, referred to as a sequential events recorder (SER), provides alarm data to redundant distributed logic controllers. If the primary main controller fails, a watchdog timer switches the function to the backup main controller.
In the December 12,1992, event, an operator inadvertently placed the SER software in a mode that disabled the OHA unbeknown to the operator, in the subsequent events, similar occurrences involving a loss of alarm processing capability were not annunciated in the control room.
The licensee's root cause analysis of the failures, design reviews, and other failure analyses uncovered several design flaws in the hardware and software of the new system. The design flaws were identified by examining system architecture and software code, and running numerous tests at the vendor site with the vendor's system developers. PSE&G determined that the design flaws resulted from ineffective system development methods and the lack of rigorous consumer examination in the past.
Presently, the digital system group reviews and performs failure analyses of significant digital devices before they are incorporated into the plant. At the time this system was installed, however, PSE&G did not review digital instrument and control (l&C) upgrades to the current level. As a result, the original software quality
.
assurance procurement requirements and failure analysis for this system were less than adequate.
For Unit 2, PSE&G addressed the identified failure mechanisms in design change package (DCP) 2EC-3508. To the extent practical the DCP corrected the system design flaws through firmware and software replacement. in addition, several out-of-specification power supplies were replaced and an independent external testing circuit was added to the system to detect its failure to process and to independently alarm the failure in the control room. Finally, PSE&G revised the applicable procedure to simplify the steps required to determine the cause of the failure and restore the ' annunciator to operation in a timely manner.
Through detailed discussions with the engineering staff as well as document and equipment reviews, the inspector verified that the modifications of the Unit 2 overhead annunciator had been completed. The licensee also stated that they had verified the vendor validation testing of the software changes and performed the Critical Digital Review of the software, firmware, hardware, and system.
l l
l i
f I.
,
!*
!
To evaluate the adequacy of the licensee actions, the inspector reviewed a variety of documents, including the design change package and applicable sections of the FSAR and technical specifications. The inspector identified no areas of concern and considered the addition of the external alarm circuit an acceptable means to detect system failures in a timely manner. Discussions with the licensee indicated their intent to install a completely different system to resolve apparent design problems
.
that are inherent to the system and to prevent a recurrence of the failures. The licensee stated that a design change would be pursued to replace the present OHA system with a new system following plant restart.
The inspector also determined that applicable procedures had been revised te provide additional guidance for alarm response and that the training program had been upgraded to reflect the procedure changes and to strengthen the identified weaknesses, i
c.
Conclusions Based on the above review, the inspector concluded that PSE&G had taken sufficient steps to correct known Unit 2 overhead annunciator system failures. The inspector also recognized that future OHA system failures, similar to those experienced in the past, cannot be precluded or discounted. The inspector, nonetheless, concluded that PSE&G, through physical and procedural changes, had developed acceptable means to detect those failures and respond to them in a timely manner. This item is closed for Unit 2 only. For Unit 1, the item remains open pending NRC verification that the developed design changes have been implemented.
E 2.4 NRC Restart issue 11.7 - Emeraency Diesel Generator has Minimal Load Marain (OPEN)
a.
Insoection Scope The NRC originally evaluated the load carrying ability of the Salem emergency diesel generators (EDGs) during the 1993 electrical distribution system functional inspection (Inspection Report Nos. 50-272; 311/93-82). At that time, the inspection team concluded that the EDGs had no spare capacity for future load increases. To address NRC observations and discrepancies identified during this and subsequent inspections, the licensee reevaluated the EDG loads and issues a revised calculation.
In February 1996 (Inspection Report No. 50-272: 311/96-01), the NRC reviewed Revision 2 of EDG loading calculation No. ES-9.002. The NRC observed that the previously identified discrepancies had been resolved, but that the margins were still limited. Further, the method used to calculate the 230 Vac loads was not correct.
Therefore, the impact of the identified discrepancy on the EDG loading profiles was unknown. The objective of this inspection was to evaluate the licensee's resolution of this issu _ _ _ _
_ _ _
. - _ - - _ _ _
_ -_
_ _ _ __._. - _.. -
_ _. ~. _.. _
.
.
b.
Observations and Findinas l
l To evaluate the load imposed by small 230 Vac motors on the EDG, the licensee
'
had conducted field measurements of selected motors and used the results to calculate an " average" percent loading. The averaging method, however, was incorrect because it averaged the calculated loading percentages (service factors)
rather than the measured loads. The averages calculated with the two methods were 79.02% (adjusted to 80%) and 91.71 %, respectively.
To address this discrepancy, the licensee used the same field measurements and manufacturer motor data to calculate the motor powerfactor (pf) and watts (load).
,
l The average power factor (.78) was then calculated. This calculated average pf showed that the 80% (service factor) used in the EDG loading calculation was i
acceptable.
Although a minor discrepancy was evident in the powerfactor calculation and the
!
use of an average powerfactor is not the preferred method to develop the load i
profile when margins are low, the inspector found the method acceptable for the I
following reasons: (1) 80% pf is reasonable for small motors; (2) the load from the 230 Vac motors is only a fraction (approximately 10%) of the total EDG loads; (3) a nonconservative powerfactor error of 10% (for example) for all the 230 Vac motors represents approximately only 1 % of the EDG load carrying capacity; and (3) not all motors are started during the first few minutes following an accident, when the EDG load is at its maximum.
In the closure package, the licensee provided an interim list of loads added and removed from the vital bus as a result of implemented and scheduled design changes since the last EDG loading calculation revision. This list indicates a total load reduction ranging from 70 kW on EDG 1C to 169 kW on EDG 2B. Most of the load reduction resulted from changes to the service water pumps. Because the irnpact of the load reduction on the EDG load carrying ability is not known until new load profiles are developed the inspector had no comments regarding the interim load change list. The licensee did not plan to revise the load calculation until after i
l fuel load, when the list of pending DCPs becomes final.
c.
Conclusions Based on the above review, the inspector concluded that the licensee had properly resolved the 230 Vac motor load discrepar.cy. While some improvement in EDG spare capacity appeared to be derived from the load reductions, particularly from the service water pump changes, this spare capacity continues to be small and the fullimpact of the load reduction will not be known until the EDG load calculation is i
revised again. This issue remains open pending revision of the load calculation.
.
i
-
-
.
.
E2.5 NRC Restart item II.11. EDG Load Fluctuations (CLOSED)
a.
Inspection Scope
,
This item originated from the August 16,1994, load fluctuations that were observed on EDG 1 A during monthly surveillance testing. On March 12,1996, during an NRC review of the root cause analysis of the EDG 1 A load fluctuations, EDG 2A experienced load fluctuations.
The EDG 2A load fluctuations occurred while the licensee was returning the diesel from an 18 month scheduled outage test. At that time, the licensee believed the
fluctuations were caused by inadequate venting of the governor oil chamber
~
istlowing an oil change. To ensure that all factors were considered, however, the
,
licensee decided to perform a level 1 root cause analysis. Further, the licensee decided to re-perform the root cause analysis for EDG 1 A to address concerns expressed by the inspector associated with the original analysis (NRC inspection
>
l report 50-272; 311-96-01).
On August 5,1996, at the completion of a 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> endurance run, the EDG 2A frequency was observed to be oscillating between 59.5 to 60.1 Hz. This oscillation occurred after the generator was unloaded and the generator output breaker was
opened. The frequency osciliatens continued for 50 minutes and then ceased
,
!
without operator action.
(
l The inspector reviewed the restart clowre package, root cause analyses, l
procedures, and related corrective actior documents to determine the adequacy of l
the licensee's actions to address the EDG 1 A and 2A load fluctuations, and the EDG 2A frequency oscillations issues.
l b.
Observations and Findinas Root Cause Analysis Results
,
l For EDG 2A the licensee attributed the load fluctuations to loose fuel rack linkage l
and problems with governor tuning. The excessive play in the fuel rack linkage was caused by improper inspection and adjustment of the fuel rack and was attributed to insufficient training and procedural guidance. The governor tuning problems, which resulted in governor instability, were similarly attributed to insufficient procedural guidance and less than adequate training.
For the EDG 2A frequency oscillations, the licensee determined that the cause was an insufficient warmup time of the diesel engine governor oil, prior to performing the last governor alignment.
For the EDG 1 A load fluctuations, the most probable causes were determined to be a leaking capacitor in the electrical governor (EGA) and/or postulated oil contamination or misalignment of the mechanical governor (EGB).
I
!
!
.
.
NRC Reviow
The inspector reviewed the root cause analysis for the EDG load fluctuations and frequency oscillations and concluded that the identified causes, causal factors, and corrective actions were reasonable. The inspector also found that the 1996 root cause analysis of the EDG 2A load fluctuations was more thorough and detailed than the 1994 analysis for the EDG 1 A load fluctuations and that the new root cause analysis of the 1994 load fluctuations had resulted in different conclusions and additional corrective actions.
The Salem staff's review of the load fluctuations issue identified several concerns related to the EDG governors and fuel racks that required corrective action.
Performance improvement Request (PIR) #960318088 was issued to track the various corrective actions identified by the Salem staff. The inspector reviewed the completed corrective actions to verify their implementation. No concerns were identified during this review.
For those corrective actions that had not been completed, the inspector reviewed the items, verified that they could be delayed until after the plant restart and l
confirmed that they were being tracked. These longer term items included the development of a Woodward governor training program; revision of the repetitive tasks for governor oil changes; establishing a repetitive task to periodically energize l
,
the EGA modules; addressing EGA refurbishment plans; addressing parts availability l
problems; evaluating root cause analyses for impact on main and auxiliary feedwater pumps; and evaluating the effectiveness of the Woodward governor corrective actions.
During the review, the inspector identified a potential configuration control issue with the EDG governors. The inspector determined that the configuration control issue was unrelated to the load fluctuations and frequency oscillation issues. The governor configuration control issue is discussed in Section E2-7 of this report.
c.
Conclusions The inspector concluded that 2A EDG load fluctuations root cause analysis, which included a reanalysis of the 1 A EDG load fluctuations, was thorough and the corrective action plan aggressive. The root cause analysis for the 2A EDG frequency oscillations was acceptable. These analyses uncovered several deficiencies in the licensee program for the testing of the EDGs.
!
Although the effectiveness of the corrective actions can be determined by future l
performance monitoring, the inspector concluded that the EDG load fluctuations and j
frequency oscillations issues had received sufficient engineering and management
attention and that the completed corrective actions provided reasonable assurance that the EDG operating reliability would be improved. This restart issue is closed.
In addition, the licensee identified and corrected the violation of the EDG testing program and is being treated as a non-cited violation, consistent with Section Vll.B.1 of the NRC Enforcement Policy.
l l
l
.
l
'.
l The inspector also concluded that the remaining open items had been either l
adequately addressed for the near term or they were long term tasks that could be completed after plant restart. These items will be reviewed, after restart, during a future inspection of the licensee programs. Unresolved item 50-272;311/96-13-03 E2.6 EDG Governor Confiauration Control
,
!
,
During the review of the EDG load fluctuations issue, the inspector determined that configuration changes had been made to the Salem EDG governor systems during refurbishment and/or repair activities conducted at the vendor's site. These activities, involving modifications to the (',
.ric governors (EGA) as well as the l
mechanical governors (EGB), were conducted without PSE&G being aware of the l
'
'
exact nature of the changes. The repair records, while containing information regarding the changes made, failed to provide sufficient detail to support an adequate evaluation and acceptance by PSE&G.
,
Although the vendor documentation process and PSE&G's documentation review were less than acceptable, the inspector identified no technical concerns with the l
l equipment in that the repaired / refurbished governors were required to pass post-l repair test specifications and the vendor was required to furnish a Certificate of Conformance before the device was returned to the PSE&G spare parts inventory.
Finally, when installed on an EDG, the governors were required to pass alignment i
calibration and surveillance testing requirements, prior to the EDG being declared operable.
Considering the vendor testing requirements described above, the periodic TS surveillance testing required to maintain operability, and the lack of documented j
EDG failures attributable to configuration changes, the inspector concluded that the significance of the issue was minor. However, since vendor-related configuration control issues were identified previously (e.g., Hagen modules and auxiliary feedwater pump governors) the extent and safety significance of this issue as applied to other plant components has not been determined. This issue will be l
evaluated in detail during the NRC review of restart item Ill.2, Configuration Control.
E2.7 NRC Restart item Ill.11, Enaineerina Problem Resolution, includina Safety Evaluations (OPEN)
a.
Insoection Scone (37001)
The inspectors conducted a review of the 10 CFR 50.59 Safety Evaluation Program to assess the effectiveness of the licensee's corrective actions in addressing programmatic deficiencies identified by the NRC and the licensee.
,
i
.
.
. _ _ _
.
.
.
!
!
.
'
l b.
Observations and Findinos Licensee's Corrective Actions:
l PSE&G took the following steps to address deficiencies identified with the 10 CFR
,
!
50.59 Safety Evaluation Program-l I
e From January 5,1996, to February 17,1996, PSE&G performed a self assessment to evaluate the 50.59 program. This assessment included an evaluation of procedures and controls, program implementation, training and i
qualification, and a review of the 50.59 program at other licensed facilities.
i
'
PSE&G concluded that the program contained weaknesses, but that the program could support restart and safe uneventful operation, o
Consistent with assessment results and recommendations, PSE&G:
1) revised Procedure NC.NA-AP.ZZ-OO59(O), "10 CFR 50.59 Applicability
'
Reviews and Safety Evaluations," to include some of the good practices observed in other licensees' programs, and to establish formal training and l
qualification requirements for 10 CFR 50.59 safety evaluators; 2) issued a lessons learned document, detailing various 50.59 safety evaluation j
problems, to all managers for distribution to preparers and peer reviewers of safety evaluations; and 3) conducted training to a " core group" from each department to ensure proper training for those involved in the safety
,
evaluation applicability review process.
!
In addition, the licensee had issued Action Request # 951114244 to require
periodic program sampling by the sponsoring organization (Licensing) to
,
'
provide feedback on program effectiveness. This action had not been implemented as of the end of the inspection period.
l e
From November 6,1995, to April 30,1996, the licensee hired a contractor to provide a team of experienced engineering personnel to evaluate the design change package (DCP) process. As part of that effort, a member was
!
added to the team to perform 10 CFR 50.59 safety evaluation reviews for DCPs and to work with PSE&G project engineers to help enhance the quality l
of their safety evaluations.
l 10 CFR 50.59 Proaram Review:
The inspector reviewed procedure NC.NA-AP.ZZ-0059(Q), Rev. 4, "10 CFR 50.59 Applicability Reviews and Safety Evaluations," for compliance with 10 CFR 50.59
)
and NRC Inspection Manual Part 9900, " Interim Guidance on the Requirements Related to Changes to Facilities, Procedures and Tests (or Experiment)."
The inspector noted that the procedure guidance was largely based on NUMARC's NSAC-125, " Guidelines for 10 CFR 50.59 Safety Evaluations." Consequently, the guidance indicated that a smallincrease in the probability or consequences of an accident or malfunction previously evaluated in the safety analysis report does not i
l l
--
.
_
-. _.
-. --.
.
- _ -
-.
.
l l
.
l l
involve an unreviewed safety question (USQ). As stated in NRC Inspection Manual l
Part 9900, this guidance conflicts with 10 CFR 50.59 which states that a USQ exists if the probability of occurrence or consequences of such an accident "may be increased."
The licensee had evaluated the differences between the NRC Inspection Manual
!
Chapter 9900 and the 10 CFR 50.59 procedure (AR # 960517138). Prior to the completion of the inspection, the licensee issued Revision 5 to NC.NA-AP.ZZ-0059(Q) which removed the existing 10 CFR 50.59 guidance from the procedure I
and issued Administrative Standard NC.NA-AS.ZZ-0059(Q) on August 30,1996.
l This administrative standard also incorporated the NRC's interim guidance on
!
10 CFR 50.59 safety evaluations.
l 10 CFR 50.59 Imolementation Review:
i l
The inspector conducted a review of safety evaluation applicability reviews and i
!
safety evaluations that were completed to support permanent plant modifications, temporary plant modifications, and procedure changes. Attachment 1 to this report lists the documents reviewed by the inspector.
l In general, the inspector determined that the licensee's implementation of the 10 CFR 50.59 safety evaluation process was acceptable. The safety evaluation applicability reviews and safety evaluations reviewed by the inspector were determined to adequately document that a technical specification change or an unreviewed safety question were not involved. The inspector, however, noted
'
l l
some examples of procedural non-compliance, as documented below:
Section 3 of safety evaluation No. S96-069 described the changes to be made to the boron injection tank (BIT) discharge pressure transmitter (2PT942) sensing line. Included in this description was a paragraph stating that, "The transmitter output cable is terminated to a Buchanan terminal block inside Panel 215-2 terminal box.... The BIT Room is a harsh environment area subject to HELB [high energy line break] effects, and i
installation of 2PT942 must be qualified for this environment. Terminal
'
blocks may not be used in environmentally qualified installations. Therefore, the terminal block termination will be replaced with a qualified splice."
!
Discussions with the licensee revealed that the need to replace the Buchanan terminal block was discovered during the DCP field walkdown. This condition, however, was not documented on an appropriate Action Request (AR) as required by procedures NC.NA-AP.ZZ-0006(Q), Revision 13,
" Corrective Action Program," and NC.NA-AP.ZZ-0000(O), Revision 0,
)
j
" Action Request Process." Therefore, the licensee missed an opportunity to
evaluate the existence of other Buchanan terminal blocks subject to HELBs.
This failure to follow plant procedures is a violation of 10 CFR 50, Appendix B, Criterion V, instructions, Procedures, and Drawings. (Violation No. 50-l 311/96-13-02)
'
l l
I As a result of this finding the licensee issued two corrective action l
l documents. The first, CRCA #3, was added to the original AR #960507302 to document the corrective av. ions for 2PT942. The second, AR
- 960820102, was issued to address corrective actions associated with the Salem Unit 1 BIT discharge pressure transmitter and to initiate an overall evaluation of similar Unit 1 and Unit 2 applications.
- The SORC accepted and the GM approved Revision 0 to Safety Evaluation No. S96-037 on May 24,1996. This safety evaluation was written to
,
'
support DCP-2EC-3559, Package 1. The safety evaluation stated that a TS
change would be required. Subsequently, during a closeout review of the DCP, a Salem staff member revised the safety evaluation to change the TS
'
change determination from a "YES" to a "NO". This determination was based on the fact that 10 CFR 50.36 excluded the TS Bases from the TS and thus did not require a TS change. The revision to the safety evaluation was incorporated without SORC ac.ceptance and GM approval, which is contrary to procedure NC.NA-AP.ZZ-0050(O), Revision 4, "10 CFR 50.59 j
Applicability Reviews and Safety &aluations." This is another example of procedure adherence failure (refer to the previous example in this section).
10 CFR 50.59 Trainina Review:
l
The inspector reviewed the safety evaluation training lesson plan, training and qualification requirements, training attendance records, and conducted interviews with licensee personnel to evaluate the effectiveness of the training.
The inspector concluded that the licensee's revised training lesson plan,0905-300.20-5059ZZ-03, Revision 3, "10 CFR 50.59 Reviews and Safety Evaluations,"
provided an acceptable vehicle to train Salem staff personnel in the requirements of 10 CFR 50.59.
The inspector reviewed training records and determined that ninety-seven (97)
individuals had successfully completed the 10 CFR 50.59 training between
-
January 4 and July 17,1996. The inspector interviewed ten (10) of these individuals. The inspector concluded that the training was effective. Further, the majority of those interviewed believed that, as a result of the additional training and program changes, the 50.59 process had improved over the last several months.
c.
Conclusions The inspector concluded that the Salem staff had improved the quality and effectiveness of the 10 CFR 50.59 program through self assessment, effective use l
of contract personnel, improved training, lessons learned feedback, and program
'
revision. Further, the NRC interim guidance on 10 CFR 50.59 had been incorporated in the licensee's program. However, based on the NRC limited review of recent 10 CFR 50.59 safety evaluations, recently implemented changes to the
'
l l
.-
.
.
=-
. - -
_
-.
-.
.
-
--.
-
-
i
.
program, and the long history of 10 CFR 50.59 related problems, this issue will remain open pending further NRC evaluation of the program implementation during the NRC review of other programs included under this restart item.
One violation was also identified. This violation involved two examples where the Salem staff failed to follow procedural requirements.
E2.8 Offsite Safety Review (OSR) Group a.
Inspection Scope As part of the 10 CFR 50.59 program review (Section E2.4), the inspector evaluated the activities of the OSR Group to ascertain compliance with TS 6.5.2.
The inspection included a reviaw of the program procedure, OSR document review guidelines, OSR staff members technical expertise, interviews with OSR staff
members and supervision, and a sampling of OSR safety reviews (refer to Attachment 1),
b.
Observations and Findinas Procedure ND.SN-AP.ZZ-0001(Q), " Independent Safety Review Program, Revision 4," delineated the review responsibilities of the OSR Group. The OSR l
Group also had issued " Document Review Guideline No.1," Revision 0, to provide l
general guidance to an OSR engineer reviewing change documents, including 10 CFR 50.59 safety evaluations.
j During his review of the latter OSR document, the inspector identified a potential concern. Specifically, Section 3.2, Review, Step "a" stated: "For change documents which do not involve safety significant SSCs [ structures, systems and components), document the conclusion that no USQ or SHC is deemed to be involved." A flowchart included as Attachment 1 to the guideline, indicated that, if the change was determined to not involve a safety significant SSC, then "NO FURTHER REVIEW [was] REQUIRED." The inspector was concerned that the statement might cause the OSR Group not to evaluate some 10 CFR 50.59 safety evaluations for USQ determination, as required by TS 6.5.2.4 and procedure ND.SN-AP.ZZ-0001 (Q).
The licensee acknowledged the inspector's concern, but stated that: 1) the OSR staff did not interpret the guideline in the same manner as the inspector; 2) the OSR staff performed an independent review of all 10 CFR 50.59 safety evaluations for USO determination, regardless of safety significance; and 3) the OSR Document Review Guideline #1 was not a procedure and did not provide specific review instructions for performing USQ and SHC determination instead, the guideline provided direction to OSR staff engineers to ensure that greater attention was
"
placed on the more safety significant items, while preserving the independence of the OSR engineer to perform further review of any document regardless of the safety significance.
I
-
-
.
. _ - _.
.-
-
... --
_
-_
.
.
.
.
.
.
Through interviews of OSR staff members the inspector concluded that the OSR
,
staff performed USQ determinations for all 10 CFR 50.59 safety evaluations reviewed by the OSR. Based on the inspector's concerns, however, the licensee J
issued revision 1 to the OSR Review Guideline No.1 to clearly indicate that all safety evaluations required a review for USO determination. The inspector reviewed the revised document and had no further concerns.
In his review of the OSR procedure and guideline implementation, the inspector observed that, for DCP 1EC-3349, the OSR had concluded that: "Since the design l
change does not have the potential to impact a risk significant function in the service water system or the containment building ventilation system, the change is deemed not to involve an unreviewed safety question (USO)."
Because the above statement implied that risk significance was used as the basis
,
for an USQ determination and, therefore, did not meet the intent of TS 6.5.2.4, the
!
inspector discussed the issue with the responsible OSR engineer. The OSR engineer
)
agreed that the conclusion statement was improperly worded and on i
August 9,1996, he revised the statement to indicate that the USQ was made independent of the risk significant function. The inspector had no further concerns l
with this item.
With regard to the technical expertise of OSR staff members, the inspector determined that documentation was available to substantiate the technical expertise i
of the OSR staff members and that review assignments were made with due
'
consideration of the technical ability of the reviewer. During their interview, the l
l OSR staff members individually stated that they felt obligated to seek additional
!
technical expertise, if the technical content of the material being reviewed was l
beyond their level of expertise.
i c.
Conclusions The OSR Document Review Guideline #1 required clarification to be in agreement with actual OSR practice. The inspector concluded that the OSR staff members l
(
technical expertise was acceptable and appropriately utilized. The inspector further l
concluded that for the sampled 10 CFR 50.59 safety evaluations, the OSR had met i
the independent review criteria required by TS 6.5.2.4.
.
,
.
.
1 E8 Miscellaneous Engineering issues i
E8.1 (Closed) Deviation 50-272 and 50-311/93-82-07. Onsite Fuel Oil Requirement for Seven-Day Emergency Diesel Generator Operation (Part of NRC Restart issue 11.8).
l a.
Inspection Scope Previous inspections documented in NRC inspection reports 50-272; 50-311/93-82, 94-33, and 96-06 identified several discrepancies between the diesel fuel oil storage tanks (DFOST) as-built configuration and the description in design and licensing documents. Nine specific issues were left open as documented in NRC Inspection Report No. 96-06. The purpose of this inspection was to review PSE&G's resolution of the nine issues.
b.
Observations and Findinas i
To evaluate the adequacy of the licensee actions to resolve the identified
'
discrepancies the inspector reviewed Calculation S-C-DF-MDC-1316, " Salem 1 & 2 EDG Fuel Oil Storage Basis," Revision 2, and other applicable documents. The specific issues and the licensee's actions are described below.
Suction Pioina Protrusion The NRC had expressed a concern that the actual suction piping protrusion into the DFOSTs might be greater than the value assumed in the fuel storage capacity calculation and, therefore, affect the amount of available fuel for the EDGs. To address this issue, the licensee emptied the tanks one at a time and measured the protrusions of the 4-inch suction piping. They found actual protrusions of 0.75-inches, 0.75-inches, 2.5-inches, and 2.25-inches for DFOST 11,12, 21 and 22, respectively. Since, before the measurements were taken, the calculation of record assumed actual protrusions of 2-inches for all tanks, the calculation was non-conservative. The licensee revised the calculation and con wryativelv used protrusions of 3-inches for all tanks. The inspector concleod that thc licensee's use of 3-inches for the protrusions was appropriate. The inspector also reviewed the quantity of fuel maintained in DFOSTs and concluded that the revised technical specification requirements for fuel oil had not been violated as a result of the nonconservative calculation assumption.
Fuel Consumotion Rate IR 50-272; 50-311/96-06 documented the NRC review and acceptance of the licensee's bases for using the EDG-2C consumption rate. In revising the calculation, PSE&G revisited the issue and used fuel consumption rates established by the diesel generator manufacturer (GE/Alco) for each EDG under Salem-specific design basis loading conditions. The loading conditions were based on profiles generated in the EDG load calculation. A 3% margin was added to the consumption rate to account for ambient and engine condition differences between the test and installed conditions. Consistent with ANSI Standard N195-1976, the calculation also included a 10% margin to account for load profile variations. The inspector concluded that the licensee's use of actual consumption rates and added margins was appropriat _
_
.
l l
'
l
.
Hose Connection Fittinas l
To determine the compatibility of the hose connection fittings on the DFOSTs emergency connection with those typically found on tanker trucks, the inspector conducted a field inspection of the emergency filllines. He found that the licensee had modified the existing fill lines and installed 2-inch fittings and caps. The inspector also reviewed Detail Specification No. S-C-M255-NDS-0181-1,
" Specification Requirements for the Purchase of Fuel Oil, Units 1 and 2, Salem Generating Station," and supplier contracts. He determined that these documents required the tanker trucks to be equipped with compatible hose and fittings for direct hook-up to the 2-inch emergency fill connections of the DFOSTs. The inspector concluded that the specified requirements should be sufficient to ensure that the tanker trucks will be able to connect to the DFOSTs' emergency fill connections.
Seismic Capability of DFOST Emeraency Fill Pipina The emergency fill connections were originally classified as Seismic Class Ill. To address NRC concerns in this area, the licensee revised stress calculation 2671402C to include an analysis of the Unit 1 emergency fill piping. Based on this calculation the licensee concluded that the piping met the applicable stress analysis
,
codes including the Seismic Class I criteria. The licensee also concluded that, since j
Unit 2 was a mirror image of Unit 1, the same analysis was valid for both units and
.
that appropriate revisions to the Unit 2 calculation should be made.
The inspector reviewed revision 3 of the stress calculation and noted that it described the connection as a 2-inch brass hose connection with a clastic cap.
During the walkdown, the inspector observed that the cap material was also brass,
'
therefore, he asked the licensee about the discrepancy and questioned the validity of the calculation results. The licensee again revised the calculation and confirmed their original conclusions regarding the emergency fill pipe.
Procedural Control of Hoses and Pumos EPiP 501, " EOF - Integrated Engineering Response," Revision 6D, Appendix 6,
" Maintenance of Fuel Oil to Salem Emergency Diesel Generators," Step 5, item B stated "... assemble temporary hoses and pumps to transfer fuel from the FOST to the DFOST Emergency Fill Connections." IR 50-272; 311/96-06 documentea a NRC concern that no hoses or pumps had been dedicated for this service and that no instructions were available on how to obtain such equipment. During this inspection the inspector found that the licensee had revised the emergency procedure to avail themselves primarily of offsite fuel oil and to use the onsite storage facility as a backup source of fuel. This issue is closed.
l
i
.
-
l
'
l l
Availability of Emeroency Fill Pioina l
The NRC had found that, although the licensee relied on the emergency fill connection to replenish the DFOSTs following an accident, they had never i
operationally tested tt.e piping. During the current review the inspector determined j
that the licensee had tested the emergency connections by draining the DFOSTs, l
flushing the emergency lines, and then refilling the DFOSTs using the emergency fill
'
)
connections. The inspector reviewed the completed work orders and concluded l
that the test adequately demonstrated the licensee's ability to fill the DFOSTs using
!
the emergency fill connections.
i Technical Review of Surveillance Procedure The issuance of the DFOST volume calculation had resulted in changes to j
Surveillance Procedure SC.OP-DD.ZZ-OD26(O), Revision 16, that included a reduction of the DFOST administrative level limit from 97-inches to 85-inches, the level corresponding to the TS limit of 23,000 gallons. As documented in IR 96-06, the inspector expressed a concern that the lowering of *.ne procedural limit eliminated all margin above the TS limit and effectively increased the probabilities for an inadvertent violation of the TS limit. Discussir ns with design engineering at the time of the previous inspection indicated that tha procedural changes had not been reviewed by them, apparently against existing procedural requirements.
During the current inspection, the inspector reviewed Procedure NC.NA-AP.ZZ-0001(O), " Nuclear Procedure System," Revision 7, and discussed the issue with responsible licensee personnel. He determined that the technical review required by the above procedure had been performed by technical personnel within the Operations Department. Therefore, the procedural requirements for technical review had not 'oeen violated. Nonetheless, the failure to provide some margin on level to allow for diesel operation during regularly scheduled diesel generator tests, indicated an inadequate understanding of the DFOST volume calculation results and of the TS limit bases. Based on his determination that the TS limit had not been violated during the period of interest and that the surveillance had been subsequently revised to provide a level margin above the TS limit (see the section below), the inspector concluded that no adverse impact had resulted by the less than thorough review of the procedure and that the concern regarding potential inadvertent violation of the TS limit no longer existed.
Administrative Limits on Minimum DFOST Level I
The NRC had observed that the DFOST surveillance procedure had not imposed an administrative limit for minimum fuel oil level to ensure that the TS limit was not l
exceeded. During the current review the inspector evaluated the licensee's i
I l
.__
--....
- -
_ __~ - - - - -
-.
-
_- -
- -
_.
'
.
d d
i
i resolution of the issue. He determined that, based on the revised DFOST capacity calculation, the TS limit of 23,000 gallons per tank, minimum, is achieved when the indicated levelis at 88-inches. The licensee had established an administrative limit of 92-inches indicated level, corresponding to 24,101 gallons per tank. This limit
,
considered early warning and response time and the risk of fuel oil spillage. The
inspector concluded that the licensee's administrative limit of 92-inches was e
reasonable.
,
Fuel Oil Soecific Gravity
The inspector had questioned the validity of PSE&G using a maximum fuel oil specific gravity of 0.86 in the DFOST volume calculation when ASTM standard D975, referenced in surveillance requirement 3.8.3.4.b of the previous TS version, a
provided a range of 0.83 to 0.89.
l In the revising their calculation, the licensee also revised the specific gravity value and used 0.876. They considered this value limiting because: 1) Detail
'
Specification No. S-C-M255-NDS-0181-1 requires that the fuel have a gravity of 30
minimum and 38 maximum corresponding to specific gravities of 0.8762 and
.
0.8348, respectively; and 2) a maximum specific gravity results in maximum, non-I conservative error of fuel level indication (i.e., highest level indication). The a
inspector reviewed the detailed specification and the licensee's calculation and
?
concluded that the licensee's use of 0.876 as the limiting specific gravity was appropriate.
c.
Conclusions
j PSE&G had taken comprehensive actions to ensure that sufficient fuel oil will be i
available for the EDGs to meet operational and regulatory requirements. PSE&G's I
attention to detail for resolving this issue was improved since the inspector's prior i
review. However, the mis-identified oil cap suggests that continued licensee attention may be needed for field verification of equipment.
E8.2 (Uodated) Unresolved item 50-311/93-82-04. Overvoltage Effects on 230V Rated Motors (Part of NRC Restart Issue ll.8).
i
'
a.
Insoection Scoce
.
Walkdown records enclosed in the EDG loading calculation indicated that the
operating voltage of two 230V rated motors was 248V and that the motors operated 6t 115% of the rated horsepower (HP). The concern was that, if the 4160V bus was at its upper limit, these motors could be exposed to a voltage
,
higher than 248V and the load service f actor could be higher than 115%. The licensee had not addressed the high voltage effect on motors and control logic relays for the specified upper operating voltage limit of the 4160V vital bus.
,
- __
.
.
As delineated in NRC combined Inspection Report 50-272; 311/94-33 this issue was closed for Unit 1 based on PSE&G implementing a major power distribution system modification (DCP No. ISC-2269) that included the installation of new station and auxiliary power transformers equipped with new load tap changers (LTC). The purpose of this inspection is to review the licensee's actions to evaluate and correct overvoltage conditions at Unit 2.
b.
Observations and Findinas The licensee performed calculation No. ES-15.004 (Q), Revision 0, " Load Flow and Motor Starting Calculation," to analyze the system voltage profile during transient and steady-state loading. The analysis evaluated the voltage across motor terminals, under worst-case motor starting and running conditions and under light load conditions. The results of this calculation showed that the maximum voltage on 230V rated motors would be 109.6%. This corresponded to a maximum voltage of 254.28V at the load side of the auxiliary transformer and 252V at the motor terminals.
The inspector reviewed the above calculation, including the assumptions and methodology used. He found the assumptions appropriate and conservative and the maximum voltage of 252V for light load plant operation within the 110% motor-rated voltage criteria stated in National Electrical Manufacturers Association Standard MG-1, " Motors and Generators."
Further review of other licensee documents, including Calculation ES-8.007 (Q),
Revision 1, " Transformer Tap Change Setting Calculation," revealed that the plant practices and procedures might not be consistent with the limits established in the load flow analysis. Calculation ES-15.004 established the minimum and maximum voltage for the 4KV group and vital buses to be 4200V and 4380V, respectively.
Therefore, operation of the buses outside these limits should not be allowed. The inspector, however, determined that, with the transformer load tap changer (LTC) in the automatic mode, operation could be allowed outside these limits.
The inspector reached the above conclusion based on the control room log and operating procedures which require operator action when the voltage, as read in the control room, is outside the 4220V-4360V band. When the inaccuracies of the control room voltmeters are also considered, i40V as determined in calculations SC-4KV001-01 and SC-DG003-01, the operating voltage could be as low as 4180V and as high as 4400V and, hence outside the limits of the load flow analysis.
Consistent with calculation ES-8-004, calculation ES-8.007 indicated that, with the LTC in the manual mode, the bus voltages should be maintained between 4240 and 4340 volts.
In response to the inspector's concern regarding operation outside the analyzed voltage limits, the licensee stated that the operator would have the option of using i
more precise local digital meters (i20V error) scheduled to be installed prior to restart. The inspector, nonetheless, was concerned that, when the voltage on the i
control room meter had reached one of the limits (4220V or 4360V), the bus
.
- - -
-.
- -
-- -.-- - -. - -
-
-
- - -. -.
!
-
,.
i
voltage could have already been outside the limits. Furthermore, no procedural
requirement existed for an operator to consider control room metering inaccuracies
- and switch to more accurate meters when the voltage on the control room meter fell outside the 4240-4340V band.
'
The licensee indicated that they would address the inspector's concerns and
,
j determine appropriate actions to ensure adequacy of the bus voltage.
c.
Conclusions j
!
PSE&G's calculation indicated that, with a bus voltage within the range analyzed in j
ES-8.004, an overvoltage at the motor terminals was not a concern. Plant practices and procedures, however, did not assure that the bus voltage would not exceed the
analyzed limits. This item remains open pending appropriate analysis and action by the licensee and review of its acceptability by the NRC.
,'
i i
E8.3 (Closed) Unresolved Item 50-272: 50-311/93-82-01. Class 1E Transformers Subject to Voltage 'Nrges Higher than Basic impulse insulation Level (Part of NRC
,
!
Restart issue 11.8).
j j
l a.
inspection Scope
'
Class 1E 4160/480 and 4160/240 Volt transformers are fed from the same 4kV j
breaker. Therefore, they are not electrically independent. The NRC inspector was
concerned that this lack of electricalindependence could render the transformers susceptible to unnecessarily high switching voltage surges generated by the other
,
i transformer on the same circuit. The purpose of the inspection was to review the l
action taken by the licensee to ensure that Class 1E transformers would not be affected by switching voltage surges.
l b.
Observations and Findinas a
The licensee replaced all the vital bus 4160V/480V transformers of both units with
,
other transformers having higher Basic Impulse Insulation Level (BIL) ratings. The
'
inspector verified that the new transformers had been installed during a plant
,
walkdown.
The licensee did not plan to replace the Class 1E 4160V/240V transformers for the j
following reasons:
j
A Power Technologies Inc. (PTI) analysis determined that switching surges were not a problem for the existing circuit. The licensee confirmed this with a test conducted on August 7,1996. During the test, the licensee monitored the transient voltage on the transformers while opening and closing 4kV feeder breaker 2A4D. The licensee did not observe any
transient voltage on the transformers. The inspector reviewed the test results and found them acceptable,
'
j
-
I l
.
I
e The licensee planned to have a spare 4160/240V transformer available on site in November 1996, with upgraded insulation and BIL requirements.
l e
The licensee had installed protective screens in the transformer enclosure ventilation to prevent foreign materialintrusion. The inspector performed a walkdown of the screens and verified that the material conditions and cleanliness of the transformers were acceptable. Tags had been posted in some cases where the material condition and cleanliness were not adequate.
c.
Conclusions Based on his review of the licensee's closure package and test results and on the results of the plant walkdown, the inspector concluded that the corrective actions
)
and the transformer voltage monitoring performed by the licensee were acceptable.
This item is closed.
E8.4 (Closed) Unresolved item 50-272/93-82-16 EDG. Transient Load Test (Part of NRC Restart issue 11.8)
i
a.
Inspection Scope The electrical distribution system functional inspection team questioned the response of the emergency diesel generators (EDGs) to transient loads. Specific items of interest were the EDG voltage and frequency response to the addition of
,
the service water pump (SWP) while the EDG is loaded to the worst-case maximum load prior to SWP loading. PSE&G provided the required data for Unit 2 and
'
indicated they would provide the Unit 1 transient voltage and frequency results following the auto-sequence EDG tests performed during regularly scheduled refueling outage surveillance. The purpose of this inspection was to review the results of the Unit 1 auto sequence EDG tests.
b.
Observations and Findinas The inspector confirmed that the licensee had recorded the Unit 1 EDGs voltage and frequency transient response during the EDG automatic sequence test. The response data had been obtained during tests performed in accordance with procedures S1.OP-ST. SSP-0002(O), Revision 7; S1.OP-ST. SSP-0003(O),
Revision 9; and S1.OP-ST. SSP-0004(O), Revision 7. The inspector reviewed the test result and confirmed that the voltage and frequency had not decayed
)
excessively during load addition (less than 20% voltage and 1Hz) and that the j
generators recovered quickly (within approximately 2 seconds) foliowing the load addition. Test results showed similar responses for all Unit 1 EDG _
_._
__
_
_ __
__
__
t
-
t I
.
c.
Conclusions The inspector found toe EDG auto-sequence test results acceptable. In particular, l
the test results showed that all three Unit 1 EDGs were capable of accepting and j
accelerating the design basis accident loads in support of accident mitigation. This item is closed, l
F8.5 (Closed) Violation 50-272/93-82-14. Failure to implement the requirements of surveillance procedures for the station safety-related batteries (Part of NRC Restart issue 11.8).
The inspector verified the adequacy and implementation of the corrective actions to address the finding.
In their letter to the NRC, dated January 6,1994, the licensee concurred with the violation and, to prevent recurrence, stated they would revise procedure SC.MD-PT.ZZ-0013(Q), " Inservice Single Cell Battery Charging," to include requirements that: 1) the specific gravity measurements be greater than or equal to 1.215 to prove a cell was " fully charged"; and 2) after the cell equalizing completion, the cell float voltage measurements be taken until stable to verify TS compliance.
These changes were to be completed by January 31,1994. A revision to procedure SC.MD-PT.ZZ-0013(Q) incorporating the first item was approved on January 28,1994. On January 22,1996, during an on-going commitment verification program review, the licensee discovered that they had not incorporated the second item and issued AR #960120140. On March 28,1996, procedure SC.MD-CM.ZZ-0024(Q) (formerly SC.MD-PT.ZZ-0013(Q)), including the second item, was approved and issued.
Based on the inspector's verification that the procedure changes had been incorporated in the revised procedure, this item is closed.
E8.6 (Closed) Violation 50-272: 311/94-07-01. Failure to perform and document 10 CFR 50.59 safety evaluation of a modification.
a.
Insoection Scope The inspectors reviewed the licensee's corrective actions to address their failure to perform a safety evaluation in accordance with 10 CFR 50.59 in conjunction with plant design modification 2EC-3219, involving the replacement of Salem 2A 460V vital bus transformer. This deficiency was documented in the NRC inspection report Nos. 50-272; 311/94-07 for Salem.
b.
Observation and Findinas j
I l
In their letter to the NRC, dated August 5,1994, the licensee concurred with the violation and identified the corrective actions that had been taken to prevent
!
recurrence. The inspector verified that the stated actions had been completed. In
.
.
particular, he verified that licensee had prepared a 10 CFR 50.59 safety evaluation of the above design change and determined that there was no unreviewed safety question associated with the implemented design change. The inspector also noted that the licensee had revised the appropriate figure of the FSAR to reflect the accurate transformer kVA ratings, consistent with the design change documentation.
j The inspector's review of five additional randomly selected design change packages associated with the Salem Units 1 and 2 determined that the PSE&G had appropriately prepared the required safety evaluations as per the established licensee procedures. Since the 10 CFR 50.59 safety evaluation program effectiveness was evaluated by the NRC under the Salem restart item 111.11, Section 2.7 of this report, the inspector did not review the programmatic aspects of this issue.
c.
Conclusions Based on the above review, the inspectors concluded that the licensee had satisfactorily completed the appropriate corrective actions to address the above modification concern. This item is closed.
E8.7 (Closed) Unresolved item 50-271:311/96-01-08 Interaction Between Fire Suppression and DG Room Ventilation System.
This item was opened to ensure that design concerns raised by the NRC during a review of an event at Hope Creek, were evaluated against the Salem design.
The inspector reviewed the Salem design and determined that:
1.
During a 198710 CFR 50 Appendix R review of Salem's safe shutdown systems, a single failure concern was identified that could have resulted in CO, flooding in more than one EDG.
2.
The EDG CO, system was modified to address spurious operation and interface concerns with the EDG ventilation and control systems. In addition, the EDG ventilation system was modified to address single failure concerns. (DCPs 1SC/2SC-1609)
3.
The thermal detectors near the EDG exhaust manifolds were relocated to prevent inadvertent overheating and actuation during EDG operation.
4.
The thermal detectors are Pyrotronics, Thermal Plug-In Type, Model DTC-200P, with a setpoint of 200* F. The detectors are not adjustable and operate based on different coefficient of expansion of the tubular shell and the struts. The detectors are designed to actuate upon a rapid heat rise as well as a slow heat ris.
.
5.
Nuclear Fire Protection procedures S1/S2.FP-ST.FD-0029(Q), Revision 2,
" Functional Test of Class 1 Smoke and Thermal Detectors," directs the performance of functional testing of the EDG thermal detectors every six months (one detector per CO, zone every six months) on a rotating basis.
6.
The Salem EDG ventilation system was designed with air intake in the ceiling at one end of the room and exhaust in the ceiling at the opposite end of the room.
7.
Thermography readings taken on August 15,1996, during the extended operation of the #2A EDG indicated a relatively large margin between the actual temperatures near the detectors and the actuation point of detectors (49' F or greater).
8.
The Salem EDG room design did not use fire dampers.
The inspector concluded that the design concerns associated with the Hope Creek i
event were either already addressed by previously installed design changes at Salem or were not a concern due to differences in design.
During the review of the thermography data recorded for #2A EDG Room during an extended run of 2A EDG on August 15,1996, the inspector noted that the temperature of detector #4 had two recorded readings in excess of 120' F (141 and 130* F). This detector, located several feet from the EDG exhaust manifold, had been previously relocated further away from the exhaust manifold due to overheating concerns (see item 3 above). Salem UFSAR, Revision 14, Section 9.4.5, Diesel Generator Area Ventilation, Paragraph 9.4.5.1, Design Bases, states that, "The ventilation systems are designed to limit the temperature of each diesel generator compartment to 120' F and each control room to 110* F in the summer with equipment in the room operating."
Since the UFSAR did not discuss localized high temperature areas, the inspector questioned the licensee about the acceptability of the high ternperature readings.
Based on the inspector's concern, the licensee issued AR #960820220 to evaluate the localized high temperature conditions. The results of this evaluation will be reviewed in a future NRC inspection. Based on his observation that no safety related equipment was located in the vicinity of detector #4, the inspector did not consider this a restart issue. (Unresolved item 50-272: 311/96-13-03)
E8.8 (Closed) Unresolved item 50-272/94-18-02. Emergency Diesel Generator Load Fluctuations Root Cause.
This issue pertains to the load fluctuations identified as item 11.11 of the NRC Restart Action Plan for Salem.
The NRC conducted a review of the licensee's analysis and actions to address this issue and found them acceptable. The details of the NRC review are contained in Section E2-6 of this inspection report. This item is close.
.
E8.9 General Conclusions During the current inspection period, the quality of engineering continued to be inconsistent. The licensee took acceptable actions to resolve the issues reviewed.
The root cause for the EDG fluctuations was thorough and detailed and with aggressive corrective actions. Resolution of some issues, however, lacked the required depth of review. For instance, in the case of the inadvertent auxiliary spray, engineering accepted without documented basis a pressurizer spray nozzle thermal transient in excess of that analyzed by Westinghouse.
In the case of the motors subject to overvoltage, engineering properly addressed the voltage effects, but failed to address meter accuracy in the surveillance procedure thereby allowing the voltage to potentially exceed the value analyzed.
The NRC evaluation of the current 10 CFR 50.59 program indicated an acceptable process that, if properly implemented, should provide good results. The revised program, however, was too new to evaluate its effectiveness. Two procedural adherence violations were discovered during this review, indicating lack of attention to detail. Another example of insufficient attention to detail was provided by the
,
'
licensee resolution of the DFOST issues.
E8.10 Review of UFSAR Commitments 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 discussed in Section E of this report, the inspector reviewed the applicable portions of the UFSAR that related to the areas inspected. This included portions of sections 9.5.1 pertaining to the fire protection program. The inspector verified that the UFSAR wording was consistent with the observed plant practices, procedures and/or parameters.
V. Manaaement Meetinas XI.
Exit Meeting Summary The inspectors presented the inspection results to members of licensee management at the conclusion of the inspection on September 20,1996. The licensee acknowledged the findings presented.
The inspector asked the licensee whether any material examined during the inspection should be considered proprietary. No proprietary information was identifie. -..
..... -.. - _... _
.. -.. - _..
_. - - - - -
_ -.__.
. _ -. - -,
.
,-
.
i
.
'
.
PARTIAL LIST OF ATTENDEES
,
!
j Public Service Electric and Gas Comoany
>
G. Boerschig, Manager, Nuclear Electrical Engineering
'
A. Giardino, Project Manager i
R. Hoffman, Licensing M. Maradeo, Principal Engineer G. Phelps, Supervising Engineer i
J. Ranalli, Manager, Salem Restart Engineering Plan l
M. Stephens, DE&P J. Schubert, Lead Engineer E. Villar, Licensing i
,
U. S. Nuclear Reaulatorv Commission B. Smith, Engineering Contractor l
i
l
.
.
Attachment 1 The following safety evaluations applicability reviews were reviewed:
Temporary procedure TS1.OP-SO.CH-0003(Q), Revision 3, Chilled Water System e
Flush Procedure S2.OP-AR.DG-0001(Q), Revision 6,2A DG Alarm Response e
Procedure S2.OP-SO.FHV-0001(Q), Revision 10, Fuel Handling Building Ventilation e
Operation Procedure SC.OP-AP.ZZ-0030(O), Revision 0, Operator Work Around Program e
e Procedure S2.OP-AR.ZZ-0015(O), Revision 6,2RP1 Radiation Monitor Alarm Annunciator Temporary Modification 96-016, Revision 0, Installation of Unit 1 Lightning Mast
onto Unit 2 e
Temporary Modification 96-017, Revision 1, install 2 Westronics Series 3000 Recorders for Use The following safety reviews and evaluations were reviewed:
Safety Evaluation No. S96-001, Revision 0 & 1, SG Primary Manway Bolting e
Replacement Safety Evaluation No. S96-014, Revision 0, Alternate Shutdown System Inverter
Replacement Safety Evaluation No. S96-016, Revision 0, Debris Mitigation on Charging / Safety e
injection Cold Legs e
Safety Evaluation No. S96-021, Revision 0, Modify Diesel Generator Crankcase Piping e
Safety Evaluation No. S96-031, Revision 0, Temporary Blank Installation for Control Area SPAV System Supply Fans Safety Evaluation No. S96-032, Revision 0, ABVS HEPA/ Charcoal Filter Damper Fail e
Position Modifications e
Safety Evaluation No. S96-037, Revision 0 & 1, Fuel Handling Area Ventilation Exhaust Fans Interlock with Radiation Monitors
Safety Evaluation No. S96-046, Revision 0, Non-Routine Unit 2 SG Fill and Chemical Feed (procedure S2.CH-AD.CN-1144(Q))
e Safety Evaluation No. S96-050, Revision 0, Emergency Lighting Inverter Logic Card Modification e
Safety Evaluation No. S96-065, Revision 0, Temporary Operating Procedure for Auxiliary Building Ventilation Operation e
Safety Evaluation No. S96-069, Revision 0, Rework BIT Discharge Pressure Transmitter Sensing Line e
Temporary Modification 96-002, Revision 0, Temporary Relocation of Operations Support Center (OSC) from Control Room Corridor to Service Building Corridor e
Temporary Modification 96-010, Revision 0, Temporary Ventilation for Switchgear and Penetration Area e
Temporary Modification 93-042, Revision 0, Temporary Power
_. _
_
_
.
.
__ _ _. _.. _. _ _. _ _ _ _.. - -_
~ _ _
... _ _ _
..
_ _. _
i k
i
,.
,
- !
l
l
The inspector reviewed the following OSR safety reviews:
'
e Record No. 58, Safety Review for T-Mod 93-042, Temporary Power j
e Record No. 845 - Safety Review for DCR 1EC-3349, Relocation of 12 CFCU Return
.i Flow Control Valve i
Record No. 847 - Safety Review for DCR 1EC-3476, Annular Steel Framing e
Modifications
e Record No. 851 - Safety Review for DCR 1EC-3481, MSR Vent and Drain System
,
!
Modifications e
Record No. 852, Safety Review for DCR 1EC-3428, Main Turbine Lube Oil Evactor i
Drain Piping Rework I
e Record No.1017, Safety Review for DCR 2EC-3306, Turbine Runback /SGFP I
Control Circuit Modification a
I
.
!
,
I l
l
i i
.-
-.
-.
.
.
LIST OF ACRONYMS USED AFW Auxiliary Feedwater AR Action Request CAG Corrective Action Group CAP Corrective Action Program CA/OS Corrective Action and Quality Services CCHX Component Cooling Heat Exchanger CRDM Control Rod Drive Mechanisms CRs Condition Reports CVC Centrifugal Charging ECAC Emergency Control Air Compressor EDG Emergency Diesel Generator EOPs Emergency Operating Procedures ERG Emergency Response Guideline FME Foreign Material Exclusion HDI Hilti Drop-In l&C Instrumentation and Controls INPO Institute of Nuclear Power Operations ISI Inservice inspection j
LER Licensee Event Report MRC Management Review Committee i
MSIVs Main Steam Isolation Valves N/A Not Applicable NBU Nuclear Business Unit NRC Nuclear Regulatory Commission NTOC Nuclear Training Oversight Committee OD Operability Determinations OEF Operating Experience Feedback OTSC On-The-Spot Change PDR Public Document Room PMT Post-Maintenance Testing PSE&G Public Service Electric and Gas PWSCC Primary Water Stress Corrosion Cracking RCP Reactor Ccolant Pump RCS Reactor Coolant System RHR Residual Heat Removal RVLIS Reactor Vessel Level Indicating System SERT Significant Event Response Team SI Safety injection SIRA Salem Integrated Readiness Assessment SNSS Senior Nuclear Shift Supervisor SORC Station Operations Review Committee SRG Safety Review Group SRO Senior Reactor Operator SW Service Water TDR Technical Document Room TRGs Training Review Group TRIS Tagging Request inquiry System TS Technical Specification UFSAR Updated Final Safety Analyses Report