ML20034B660
| ML20034B660 | |
| Person / Time | |
|---|---|
| Site: | Hope Creek  |
| Issue date: | 04/19/1990 |
| From: | Labruna S Public Service Enterprise Group |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| NLR-N90073, NUDOCS 9004300151 | |
| Download: ML20034B660 (3) | |
Text
.
Pubhc Lefvice sectric and Gas Company Stanley LaBruna Pubhc Service Electric and Gas Company P.O. Box 236. Hancocks Bndpe, NJ 08038 609 339-4800
- n. n o.n,. uw m-e.
APR 10 1999 NLR-N90073 United States Nuclear Regulatory Commission Document Control Desk Washington, DC 20555 Gentlement RESPONSE TO REQUEST FOR ADDITIONAL INFORMATION REGARDING BAILEY SOLID STATE LOGIC MODULE RELIABILITY - 1988 AND 1989 DOCKET NO. 50-354 HOPE CREEK GENERATING STATION Public Service Electric and Gas Company (PSE&G) hereby provides material requested by Mr. Clyde Shiraki, NRC Licensing Project Manager for our Hope Creek Generating Station.
Attached are copies of monthly reports submitted to the Hope Creek Station Operations Review Committee (SORC) regarding Bailey Solid State Logic Module (88LM) failure rates subsequent to our March 31, 1988 report, " Reliability Summary Report For The Bailey 862 Solid State Logic Module".
That report was submitted to NRC pursuant to a previous commitment.
The subsequent Bailey 862 monitoring and monthly reports to SORC are part of a program to ensure continued SSLM reliability and to improve failure rates.
PSE&G has evaluated the performance of the Bailey 862 SSLMs with respect to module failure rates during 1988 and 1989.
The total number of failed modules for those two years were thirty four per year and the corresponding number of failures per million hours of operation is, therefore, 1.70 for 1988 and 1989.
When compared to the 1987 report failure rate of 1.8 failures per million hours of operation, a distinct improvement has been achieved, especially in light of the improved testing equipment /
methodology and more extensive testing in use after 1987.
That the reduction in failure rate has been observed since the implementation of design changes to the ventilation supply to the Control Equipment Room (which houses the Bailey Logic System) and humidity control modifications to the SSLMs, themselves, indicates that our concerns regarding the relationship between humidity control and SSLM failure rate, identified in our 1987 summary report, Were Valid and have been adequately resClved.
It should be noted that our peak failure periods are those related to unit outages which generally involve extensive testing due to A
Design Change implementation and trouble shooting efforts.
3 90043001D1 900419
\\k PDR ADOCK 05000354 P
j Document Control Desk 2
APR 19 1999 NLR-N90073 The failure rates reported by PSE&G include, conservatively, not only service-related failures, but also:
- 1) failures on unused functions on the SSLMs...while the modules have numerous circuits available, and only a few may actually be used in a particular application, all functions are tested and any failure is counted.
Dy comparison, it is highly unlikely that unused relay contacts are tested or considered failures in IEEE reliability data, and
- 2) storeroom spares that are tested prior to install., tion in the plant...these modules are counted as failures contrary to industry practice for relays and other electronic components.
F Although PSE&G does not believe that Bailey 862 SSLMs can, or should, be directly compared to relays, IEEE 500 Specification provides a relay reliability of 16.73 failures per thousand units por year.
This corresponds to 523,969 hours0.0112 days <br />0.269 hours <br />0.0016 weeks <br />3.687045e-4 months <br /> between failures.
If a standard SSLM were considered to be equivalent to a sinale relay, PSE&G's calculated Mean Time Between Failures (MTBF) for 1988 and 1989 would be 587,322 hours0.00373 days <br />0.0894 hours <br />5.324074e-4 weeks <br />1.22521e-4 months <br />.
This demonstrates a roughly 12% better level of performance for the SSLMs over relays.
If, more consistent with circuit design, a single SSLM de considered to be equivalent to seven relays, the MTBF becomes
>4.1 x 108 hours0.00125 days <br />0.03 hours <br />1.785714e-4 weeks <br />4.1094e-5 months <br /> or about 85% better performance than relays.
l Finally, in response to Mr. Shiraki's request, Attachment 1 tabulates Bailey 862 SSLM reliability data for the two year period after that which was provided to NRC in our summary report l
for 1987.
Should you have any further questions, we will be pleased to discuss them with you.
Sincerely,
<m Attachment I
C Mr. C. Y. Shiraki USNRC Licensing Project Manager Mr. T. P. Johnson USNRC Senior Resident Inspector Mr. T. T. Martin, Administrator l
USNRC Region I Mr. K. Tosch, Chief Bureau of Nuclear Engineering New Jersey Department of Environmental Protection
i 1
l ATTACENENT 1 l
i i
BAILEY 862 LOGIC MODULE RELIABILITY FOR THE PERIOD 1988-89
- OF FAILURES PER FAILURES FAILURES MILLION HOURS PER MILLION l
HONTH YEAR (NOTE 7&B)
(NOTE 9)
HOURS (NOTE 10)
J January 1988 1
0.590 0.084 i
February 1988 l'
O.631 0.090 March 1988 4
2.360 0.337 (note 1)
April 1988 3
1.829 0.261 (note 1) l May 1988 4
2.360 0.337 June 1988 1
0.610 0.087 (note 3)
July 1988 4
2.360 0.337 (note 3)
August 1988 1
0.590 0.084 (note.3) i September 1988 2
1.219 0.174 i
October 1988 6
3.540 0.506 (note 2)
November 1988 4
2.439 0.348 (note 2)
{
December 1988 3
1.770 0.253 l
January 1989 1
0.590 0.084 f
February 1989
'4 2.613 0.373 (note 4)
March 1989 6
3.540 0.506 (note 4) i April 1989 1
0.610 0.087 May 1989 1
0.590 0.084 June 1989 0
0.000 0.000'(note 3)
July 1989 2
1.180 0.169 (note 3)
August 1989 0
0.000 0.000 (note.3) l September 1989 5
3.048 0.435 (note 5) r October 1989 5
2.950 0.421 (note 5)
November 1989 9
5.487 0.783 (note 5)
I December 1989 0
0.000 0.000 NOTES:
- 1. First refueling outage (note 6)
- 2. Forced outages (note 6)
- 3. High humidity period
- 4. Second mid-cycle outage (note 6}
l
- 5. Second refueling outage (note 6) 1
- 6. Outage periods involve extensive design change implementation as well as troubleshooting work which is not normally done during power operations, t
As a result, we generally experience abnormally high failure rates during these times.
- 7. Failed modules include those from all causes. These include storeroom spares tested prior to installation.
modules which faulted as a result of troubleshooting and/or design implementation work, as well as modules which failed during operation.
- 8. Logic modules are considered failures even if the particular function which failed is not used in the specific application.
- 9. Failure rate considering one logic module as one relay.
- 10. Failure rate considering one logic module as seven relays. This rate is more consistent with circuit design wherein one module provides significantly more than one relay function.
b
_,. -, -,....,