ML20072R330
| ML20072R330 | |
| Person / Time | |
|---|---|
| Site: | Hope Creek  |
| Issue date: | 03/13/1991 |
| From: | Office of Nuclear Reactor Regulation |
| To: | |
| Shared Package | |
| ML20072R327 | List: |
| References | |
| NUDOCS 9103250097 | |
| Download: ML20072R330 (10) | |
Text
- - - - - - - -
i
- pA Rio
'~g UNITED STATES
[
3, g
NUCLEAR REGULATORY COMMISSION
=t WASHING TON, D. C. 20$$$
\\...**/
SAFETY EVALUATION BY THE OFFICE OF NUCLEAR REACTOR REGULATION SUPPORTING AMEN 0 MENT NO. 40 TO FAClllTY OPERATING 1.lCENSE NO. NPF-57 PUBLIC SERVICE ELECTRIC & GAS COMPANY ATLANTIC CITY ELECTRIC COMPANY HOPE CREEK GENERATING STATION DOCKET NO. 50-354
1.0 INTRODUCTION
By letter dated December 28, 1990, Public Service Electric & Gas Company requested an amendment to Facility Operating License No. NPF-57 for the Hope Creek Generating L t. ion. The proposed amendment would replace the Modules (SSLMs), with a new license condexisting license condition 2.C.(5), rega' ing
'on 2.C.(5.
The existing license condition 2.C.(5) requires tnat tne licensee implement a SSLM reliability program and submit the results of the reliability program prior to the end of the first refueling outage. The new license condition 2.C.(5) requi es that the SSLM reliability program be continued for the life of the plant.
2.0 BACKGR0diD During the review of the Final Safety Analysis Report (FSAR) issued for the Hope Creek Generating Station, Public Service Electric and Gas Company (PSE&G)
Wentified the Bailey 862 System as providing m interface between the engineered safety features (ESF) systems and the main control room. This system contained solid state logic modules (SSLMs) which until now had not been used in a safety-related system and had not been reviewed or approved by the r,taff for use in a safety related system. The staff expressed several concerns regarding the extensive use of the Bailey SSLMs within numerous safety systems.
These concerns were resolved with the exception of the at-power testability concern.
The testability concern centered on the fact that 1) with such an extensive use of the SSLMs, including a common SSLM actuation path for both automatic and manual initiation circuitry, a failure of a SSLM could render multiple safety systems inoperable, which would prevent a safe shutdown of the plant, 2) the logic for the manual initiation of a number of safety systems was interlocked with the logic for the automatic initiatien of the same systems with the interlocking occurring through the SSt
.nd 3) the reliability of the SSLMs had not been demonstrated to be acceptau m.
Because of these concerns, the staff regtested additional information from the licensee regarding the Bailey FA? SSLM design, their reliability and cvailability fLete s, and their EMI rejection capability.
910325oor 910313
~2 DR ADOCK 05000354 p
. \\
The licersea responded to the staff concerns by letters dated July)25 (Ref.1), October 17, 1985-(Ref. 2), and November 20, 1985 (Ref. 3.
In general, the letters described the seismic and environmental qualification tests, the EMI/RFI tests that were conducted, and the modifications made to the
$SLM's input circuit as a result of the EMI/RFI tests.
The letters further stated that one medule had failed a common mode test and a transverse mode test, and that two modules had failed the seismic and en"ironmental qualification tests by malfunctioning at a low relative humidity of 60%. The staff reviewed the tests and the supporting data submitted by the licensee and stated that by incorporating this new technology (Bailey SSLMs) into safety sysums at-Hope Creek, the licensee was introducing both the potential for improving system performance and the potential for introducing new failure modes within the safety systems. To further aid in resolving their concern with respect to the reliability of the SSLMs, the staff requested that the licensee provide:
(1) Adequate justification for the two test failures that were not*' in the above references.
(2) Verification that the single failure criterion can be satisfied for each case where the minimum number of operable channel requirements that will be proposed is less than the total number of channels provided for each safety-related system that uses the SSLMs.
(3) A qualitative assessment of the Bailey 862 SSLM channel behavict that was observed during the channel check procedure.
(4) Verification that the channel functional tests as discussed in the Hope Creek Technical Specifications will include a verification of the in-situ operability of each of the' safety-related SSLMs.
The licensee responded to these concerns by letters dated December 23, 1985 (Ref.4), December 26,1985 (Raf. 5) and February 3,1986 (Ref. 6). The staff reviewed this additional infor... ' ion provided by the licensee and concluded that the information submitted wi acceptable to resolve Items 1, 2, and 3.
However, the staff was not able tt esolve Item 4 - the verification, through testing, of-in-situ operability.
their letter dated February 3. -1986 (Ref.
6), the licensee indicated that motu.nly functional testing would not be performed on any of the safety related SSLM channels however, the licensee proposed to conduct logic functional tests on the SSLMs on an 18-month-frequency as is required by the Standard Technical Specifications. The licensee in an effort to further demonstrate the reliability of the SSLMs by a letter dated February 14,1986 (Ref. 7) as supplemented by a letter dated February 24, 1986 (Ref. 8), committed to implement a surveillance test program for a test-population of 557 SSLMs (the nmnber of modules performing a safety-related function). The staff stated an opinion that reliability data should be achieved by in-situ testing and not through physical removal of the SSLMs from the system as was being proposed by the licensee. The staff noted that the L
initial-review and acceptance of the SSLM design by the staff was based,-in i
part, on its understanding that the SSLM design contained enhanced-testability features which the staff concluded was a major design advantage of the SSLM t
y concept.
e
-- - - - - ' - - - " ' - " - - - - - ' - - - ' ' - ' - - - ~ ~ ~ - ^
l
. In view of the licensee's position, the staff recommended that the lice mee examine the existing test procedures which will be performed under cold conditions every 18 months, to determine if the logic functional tests, with o-without modifications, can be used at-power without challenging the plant safety systems. Additionally, the licensee was asked to investigate other methods by which an increase in the frequency of testing could be achieved.
In a letter dated April 8,1986 (Ref. 9), the licensee co;eitted to provide the Kaff with a description of the options aveilable to perform in-situ surveillance testing, at-power, of the SSLHs that come under the surveillance of the Technical Specificatiors, ie, the safety related modules.
The licensee followed up on this commitment by letter dated May 23, 1986 (Ref.10) which contained, among other items, a description of four options M perform in-situ, at-power surveillance testing of the safety related SSLMs that were covered by the Hope Creek Technical Specifications and a proposal te develop a program by which reliability data associated with the Bailey 862 SSLMs would be gathered, from other users, to demonstrate the reliability of the Bailey units.
Ir. reviewing the four options presented by the licensee, the staff had determined tMt celecting Options 1, 2 or 3 would result in disabling equipment required to be available to operate or would possibly inhibit safety functions or permissives on equipment already in operat" The staff concluded that tne licensee should adopt Option 4 - maintain the current Hope Creek configuration and testing in accordance with current Technical Specificatior requirements - and initiate a reliability program.
In letters dated June 13,1986 (Ref.11) and June 24,1986(Ref.12),the licensee provided details of the proposed reliability program intended to demonstrate the reliability of the Bailey 862 SSLM modules. The program would consist of three major tasks a follows:
1.
An in-plant reliability program to monitor the performance _of the Bailey 862 SSLMs installed at Hope Creek.
This task would obtain reliability data, failure characteristic.information -and the root cause of any failure of either a safety related or nonsafety related SSUt.
2.
Contract an accredited laboratory to perform physical testing of a statistical sample of the Bailey 862 SSLMs.
The tests would simulate plant service conditions and would include the effects of aging.
3.
Bailey Controls Company (BCCo) would collect reliability data over a period of at least 18 months from other industrial users of the Bailey 862 SSLMs.
The staff recommended that the reliabN!!c Fog #sm be made a license condition to be added to the Hope Creek full power M ense, with the_results of the program submitted to the staff prior to % u d of the first refueling outage.
3,
- .=
3.0 EVALVATION' By letter dated A
- Condition 2.C.(5)pril 4,1988 (Ref.13), the licensee satisfied License of the Hope Creek Operating Licensee NPF-57. The letter-closed out previous comitments made by the licensee as well-as transmitted the results of the program which was designed to demonstrate the reliability of the-
-8ailey 862 SSLM modules. Thelettercontainedeight(8)attachmentswhichare listed below:
1.
_ Reliability Summary Report for the Bailey 862 Solid State Logic Module, March 31, 1988, t
2a. _ Component Failure Analysis Report, 862 Logic Module, p/N 6631291A1, Report No. QR-5106-E93-75-ADDI, Rev "0", January 12, 1988.
2b.
Functional Verification Report. 862 Logic Module P/N 6631291A1, Report No. QR-5106-E93-75, Rev "A", March 14, 1988.
- 3.
Site Related Module Reliability Improvements.
4.
Bailey Reliability Data.
5.
Modification of Existing Test Equipment.
6.
Verification of Safety Related Functions, 7._
In-Situ Testing -Feasibility Study, Report 110. MPR-1056, " Hope Creek Huclear: Generating Station Feasibility Study-for in-Situ Testing of
-Bailey:862 Solid State Logic-System for Class 1E Equipment,"
March 1988.
g
'8._
Accelerated Aging and Cycling, Report No. a8815 "Reliab.ility Test Program," March 25, 1988
- Attachment No._1, the executive summary report, written by the-licensee, closes
'3
.out all outstanding commitments and addresses the results of the reliability
= program. This_ attachment will be add m ssed as it applies to the remaining individual attachments.
Attachment:
No. 2 reports-on the-testing of the logic modules which had failed in-service at the Hope Creek site.
PSE&G contracted BCCo~to perform a failure
. analysis on-the 34 SSLMs _which had ra1 functioned as of November 1986.. Bail.ey testad the failed. modules and compiled a list' of_45. failed components of which 138.were then subjected to individual-component analysis. Outside of identifying the failed components,_the Eailey module' tests were inconclusive.in that-the.
tests were not able-to determinc the root *cause of the module failure other.
than a component had failed.- The M components selected for detailed microscopy analysis consisted of 19 - 4H36' Optical Couplers, 7 - ULN2001A Buffer Outputs, and 12 - 40508 CMOS Gates. These three_ components comprised 8a% of the total component failures.
m.
-. ~.
o t.
In its sumary on the SSLM Failure Analysis Program, Attachment No. I concluded that BCCo was unsure of the root cause of the optical isclator failures, that the devices may not have been properly screened (QC) upon receipt, and that "many of the failures were externally induced." Overall, the data contained in this Attachment did not appear to support the reliability claim for the modules.
Attachment No. 3 details the results of a Hope Creek site inspection conducted by a team from the Bailey Controls Company which consisted of personnel from Quality Assurance, Engineering, and Field Service. The team inspected the licensee's techniqtes for the storage, shipping, and trouble-shooting of the 862 system at the Hope Creet site.
The BCCo team identified what they considered to be deficiencies in seven (7) areas of interest and forwarded the list to the licensee accompanied by a recommendation for correcting each deficiency. The licensee then either incorporated the recomendation or submitted supporting data for not incorporating the recommendation. This attachment satisfied a requirement to have BCCo review the Hope Creek handling of the Bailey 862 logic modules as documented in PSE&G 1etter dated October 5, 1986 (Ref. 14).
Attachment No. 4 addresses Task No. 3 of the Reliability Program and satisfied a comitment documented by References 11 and 12.
An. analysis of the data, forwarded by Bailey showed the 862 system to have a failure rate ranging from 3 'ow of 0.8 to a high of 1.4 failures per million hours of operation. This set of data was supplied to Bailey by other comercial users of tie 862 system.
The 862 System failure rate will be addressed again in this report.
- Attachment No. 5 documents the fulfillment of a comitment made by the licensee to modify the SSLH test equipment and procedures to permit testing of the modules without manipulating the voltage selection jumpers (staple jumpers).
The existing SSLM bench tester was modified and is now capable of testing the modules in their field configured state.
Station procedure IC-GP.ZZ-031Q,
" General Procedure-Bailey Logic Module, Type 862," was revised to allow functional tests of the modules without manipulation of the staple jumpers, it should be noted that the use of the SSLM 3ench tester still requires that the modules be removed from the equipment cabinets in W to be tested. The bench tester, as modified, does not support in-situ tast mg.
Attachment No. 6 documents the fulfillment of a comitment made by the licensee to retest and verify all safety related functions of a Bailey S*1.M by procedure whenever-a logic' change is made to a module.
Station procedures SA-AP.ZZ-050,
" Station Retest Program" and IC-GP.ZZ-031(Q), " General-Procedure - Bailey Logic Module, Type 862" require the retest of-the safety features designed into a logic module.
Attachment No. 7 reports on a comprehensiva study of the in-situ testing possibilities of the Bailey 862 system. The study was perfctmed by MPR Associates, Inc. (MPR) for PSE&G.
In ta study MPR identified three objectives that the in-situ testing of the modules should meet and they are:
1.
The in-situ tests snould ful'y exercise as much as possible the logic and the input /outpur compesents of the logic module.
L
, - - ~ -.., _
,---,-------.-~-n-n-
, 2.
The tr,.ts should be performed without decreasing the overall system re1%bility.
3.
Thetestingshouldbeaccomplishedon-line(at-power)withoutadversely affecting the plant equipment alignment.
In conjunction with the three objectives, MPR studied three possible locations for the insertion and extraction of the test signals. 1)nder consideration were the field wire termination cabinets, the module backplanes, and the logic modules: themselves.
MPR cxamined the field wire termination cabinets and the module backplanes and determined that the_ required rewiring would be too extensive. Also the modifications would not meet the three in-situ test o;jectives. The third alternative, modifying the logic cards and modules, was determined by MPR to meet all.of the in-si+u testing objectives and subsequently was given a detailed evaluation.
The evaluation concluded that all logic phases of the modules could be tested in-situ,:however, this-alternative would require extensive circuit board,
modifications as.well as extensive modifications to the front panel of the logic modules and also some modifications to the Bailey bench tester. The evaluation also showed thr.t in certain cases during the in-situ testing non-disabled logic signals could result in actuation of field equipment and-alarms.- A study showed that this spurious actuation would not adversely affect -
-plant. alignment._ MPR reasoned that appropriate guidance for the o)erators to-
~ deal with this problem during-in-situ testing could be developed w11ch would result ^1n.only minor added burden for the operators.
The summary' report discounted the recommendation made by MPR for several
-reasons, two of which are 1) the testable SSLMs (result of the proposed modifications)- would-have no previous operating histcry to justify them as an improvement to system reliability, and 2) PSE&G does not consider in-situ testing as a viable or cost effective method 'of improving plant reliability.
Attachment B contains the details of the Accelerated Aging Program conducted on the Bailey modules which simulated a 2-year, a 5-yer.r, and a 10-year plant life. The program contained a test sequence which ttarted with a baseline functional test at ambient conditions, followed by thermal aging, environmental-
- functional. testa,- and-ending with another ombient conditions baseline functional test. The environmental functional tests-included tests such as temperature
'(low humidity); logic cycling and humidity tests (high tenp, high humidity). The thermal aging consisted of soaking the devices at a high temperature for:a prolonged period of time. The 2-year plant. life was simul ted by naking the devices at 85 C (185 F) for a time period of 179 hours0.00207 days <br />0.0497 hours <br />2.959656e-4 weeks <br />6.81095e-5 months <br />, fcr the 5-year lifa simulation, a soak temperature of 95 C _(203 t) for a wriod of 210. hours was selected and th 10-year plant life was simulated by soaking the devices at95C'(203_F)for.-420 hours.
Iu I
L l
f r
1 1
-7 The tests, at an elevated humidity, performed af ter each thermal aging cycle, involved raising the devices' environment to 110 F 0 80% RH for 96 hours0.00111 days <br />0.0267 hours <br />1.587302e-4 weeks <br />3.6528e-5 months <br /> and-performing the functional tests under these conditions, then changing the environment to 140 F @ 90-95% RH for 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> and retesting the devices. These tests proved to be more than the SSLMs could handle as the majority of them suffered from an "out of spec" condition during the tests. Of the 29 modules tested in the accelerated aging program, anly six of them did not exhibit some sort of an anomaly.
The failure modes were determined by the testing organization to be non-time dependent or else indeterminate and no failures were credited to the aging process. Three modules were removed from the test program as a result of component failure.
In the Accelerated Aging Prograr:, modiles consistently failed during the tests which were performed-at an ele"-ted humidity and temperature of 110 F and 80%
RH and 140.F and 90% RH.
These +est limits were extracted from Bailey's Product Information for the SSLMs and i v:orporated into the Test Specification by the licensee.,The licensee discussed this test with Bailey and the Bailey representative stated that the upper limits of temperature and relative humidity were not intended to occur sinultaneously. The licensee then attempted to modify the test procedures before the aging program was completed, but was not successful.
Con n ently, tP modulas were not tested under the conditions of low temperature r
- ed with.'
re:iative humidity or with high temperature coupled with lot t h ive hud d ny.
As mentioned earlier, Attachment No.1 is the Reliability Summary Report in which the licensee provides a discussion on the Bailey 862 SSLM reliability program as described by the other 11 Attachments. Of particular interest is the apparent agre..ent of the failure modes noted in the licensee's "In-House
- Data Assessment Prcgram" and the Accelerated Aging-Program. The data assessment program included tracking in-service module failures on a monthly basis. The period of tine covered by this program was from March 1986 through February 1988.-
In the data assessment program, the licensee concluded that the " increase in failures experienced through the suniner months is a result of higher average Relative Humidity in the Lower Equipment Control Room (LECR) which houses the 862 system." The relative humidity would go from 35% RH in the winter mor.ths to 65% RH during the summer months (high temperature, high humidity). The results of this _ program conducted by the licensee is contradictory to the previous concluse s reached during the licensing review that the Bailey 862 n
SSLHs were qualified for_both high temperature and high humidity ranges.
Realizing that the Bailey modules would not operate under adverse environmental cc4" ions of high temperature and high humidity occurring simultanecusly, the 1 +r s-e then embarked upon a program to modify the HVAC system servicing the LEG the system was modified to maintain the relative humidity between 20% and 60% year round. The tests that were performed on the Bailey system indicate that the modules cannot operate within their published specificatiors as contained in Attachment No. 7.
. -~
'9 Finally, the licensee compared the latest Hope Creek failure rates of 1.15,1.20 and 1.25 (Referencer 15,16 and 17 respectively), failures per million hours of service with the recommended failure' rate contained in IEEE Std 500-1984, "lEEE Guide to the Co11 action and Presentation of Electrical Reliability Data for Nuclear Power Generating Stations." This standard states a recommended failure rate of 1.1? failures per million hours of-service for " Solid State Computation Moduks" and ::ith a composite failure rate of 1.15 for " Computation Modules." Also, of particular interest, the IEEE standard states a recommended failure rate (composite) rate for Protective Relays of 0.85 and for Control Relays of 0.07 failures per million hours of operation.
Thit, effort on the part of the licensee was intended to demonstrate that the Bailey SSLMs have an operational reliability as good as or better than mechanical relays performing the same safety related functions. The licensee has not been able to demonstrate that the SSLMs are as reliable or more reliable than relays performing the same safety function.
Because of this, we conclude that the licensee should make all reasonable and practical efforts to continue improving the reliability of the Bailey 862, System. As part of this effort, the licensee should continue tracking the performance of the Bailey 862 SSLMs and correcting the root causes of failures that occur. The licensee committed by letter dated September 10, 1990 (Ref.15) to continue (for the life of the plant) the reliability tracking program as originally implemented by HCGS.
This program will continue to obtain the reliability data, failure characteristic information, and note the cause and corrective ection of failure of both safety-related and non-safety-related Bailey 862 SSLMs failures on a t.ontinuous basis. This program will continue to indicate the actual in-plant performance of the Bailey 862 SSLMs. The results of this program will be made available to NRC upon request.
In addition, the licensec stated that the improvements made in temperature and humidity control have been incorporated as permanent changes to the appropriate HVAC systems. This should contribute towards improving SSLM performance.
The staff concludes that this approach is acceptable and that the current license condition' bas been satisfied.
However, the commitment to continue the SSLM reliability program for the life of the plant should be retc.ined as a condition in the Hope Creek operating license. Therefore, the staff concluded that the following license condition, proposed by the licensee, be added to the full-power license:
~
1 u
, (5) Solid State Logic Modules PSE&G shall continue for the life of the plant, a reliability program to monitor the performance of the Bailey 862 SSLMs insta11eo at Hope Creek Generating Station. */his program should obtain reliability data, failure characteristics information, and root cause of failure of both safety-related and non-safety-rehted Bailey 862 SSLMs. The results of the reliability program shall be maintained on-site ard made available to the NRC upon request.
As noted in Hope Creek SSER No. 6, the staff has categorized the on-line testability of protection systems at power as Generic Issue 120. Any requirements that emanate from the resolution of this generic issue will be applied to the protection systems at Hope Creek.
4.0 ENVIRONMENTAL CONSIDERATION
This amendment involves a change to a requirement with respect to the installation or use of a facility component located within the restricted,
area as defined in 10 CFR Part 20. The staff has determined that the amendment involves no significant increase in the amounts, and no significant change in the types, of any effluents that may be released offsite and that there is no significant increase in individual or cumulative occupational radiation exposure. The Comission has previously issued a proposed finding that this amendment involves no significant hazards consideration and there has been no public comment on such finding. Accordingly, this amendment meets the eligibility criteria for categoricalexclusionsetforthin10CFR51.22(c)(9).
Pursuant to 10 CFR 51.22(b), no environmental impact statement or environmental assessment need be prepared in connection with the issuance of this amendment.
5.0 CONCLUSION
The Comission made a proposed determination that the amendment involves no significant hazards consideration which was published in the Federal Register (56 FR 4870) on February 6,1991 and consulted with the State of New Jersey. No public coments were received and the State of New Jersey did not have any coments.
The staff has concluded, based on the considerations discussed above, that-(1) there is reasonable assurance that the health and safety of the public will not be endangered by operation in the proposed manner, and (2) such activities will be conducted in compliance with the Comission's regulations and the issuance of this amendment will not be inimical to the comon defense and security nor to the health and safety of the public.
Dated:
Mn un 13, 1991 Principal Contributor:
J. Mauck
o c..
11 -
+
REFERENCES 1.
Letter, R. L. Mitti (PSE&G) to W. Butler (NRC), "HCGS SER Outstand _ing i
Issue 5 - Solid State-Logic Modules," July 25, 1985.
2.
^
Issue 5 - Solid State Logic Modules.," October 17, 1985.
3.
Letter, R. L.:Hitti (PSE&G) to W. Butler (NRC), " Separation of Automatic and Manual Initiation _ Circuits," November 20, 1985.
4 '.
Letter,~ C. A. McNeill (PSE&G) to E. Adensam (NRC), " Manual / Automatic Initiation Design of Bailey Solid State Logic Module," December 23, 1985.
5.
~ Letter, C. A. McNeill (PSE&G) to E. Adensam (NRC),"Be.iley 862 SSLM elecon of December 18, 1985," December 26, 1985.
)
6.
Letter, C. A. McNeilli(PSE&G)- to E. Adensam (NRC), " Bailey 862 Solid State p
Logic Module," February 3, 1986.
7.
Letter, C. _ A.:McNeill:(PSE&G) to E. Adensam (NRC), " Testability of' Bailey Solid State Logi_c Modules," February 14, 1986.
8.
Letter, C. - A. McNeill (PSE&G) to E. Adensam (NRC), " Bailey Solid Stste Logic Module Telecon of February 19, 1986," February 24, 1986.
9.
- Letter, C. A. McNeill (PSE&G) to E. Adensam (NRC), " Bailey b62 Solid State Logic Module Concerns," April 8, 1986.
- 10. Letter, C. A. McNeill (PSE&G) to E. Adensam (NRC), " Bailey 862 Solid State Logic Modules," May 23, 1986. -
- 11. - Letter, C. A. McNeill (PSE&G) to E. Adensam (NRC), " Bailey 862 Solid State Logic Module Reliability Program," June 13, 1986.
12.
Letter, C. A. McNeill (PSE&G) to E. Adensam (NRC), " Bailey 362 Solid State Logic Module Reliability Program," June 24, 1980.
13.
Letter, S. E. Miltenberger (PSE&G) to USNRC, " Bailey Reliability Program
-Final Report," April 4, 1988.
4.
t etter, C. A. McNeill -(PSE&G) to T. A. Murley (NRC), "Sumary of Meeting on Bailey Solid State Logic Module " October 5, 1986. ~
- 15. Letter, S. LaBruna-(PSE&G) to U.S. NRC, " Bailey Solid State Logic Modules Reliability Program,"- September 10, 1990.
- 16.. Letter, S. LaBruna (PSE&G) to U.S. NRC, " Bailey Solid State Logic-Modules Reli:bility Program," October 26, 1990.
- 17. Letter, S. LaBruna-(PSE&G) to U.S. NRC, " Bailey Solid State Logic Modules Reliability Program," February 1, 1991.
..-...a.
.. -... - -.