ML20008E664
| ML20008E664 | |
| Person / Time | |
|---|---|
| Site: | Arkansas Nuclear  |
| Issue date: | 02/28/1981 |
| From: | Trimble D ARKANSAS POWER & LIGHT CO. |
| To: | Clark R Office of Nuclear Reactor Regulation |
| References | |
| 2R028112, 2R28112, NUDOCS 8103090379 | |
| Download: ML20008E664 (4) | |
Text
_
7 ARKANSAS POWER & LIGHT COMPANY POST OFFICE BOX 551 LITTLE ROCK. ARKANSAS 72203 (501) 371-4000 February 28, 1981 e. DJp:/,,
U N
b(60 w l L}
'9 d
- }
,.E O G 1981" !.12 b
hh u.s. uuan nun *C" COWW I
Director of Nuclear Reactor Regulation 4,k
,6 ATTN:
Ftr. R. A. Clark, Chief 4'7 Operating t(cactor Branch #4 Division of Licensing U. S. Nuclear Regulatory Commission Nashington, D.C.
20555
SUBJECT:
Arkansas Nuclear One - Unit 2 Docket No. 50-368 License No. NPF-6 CPC/CEAC-Plant Computer Datalink (File: 2-0520.2)
REFERENCE:
(1)
-090-06, same subject, dated September 3, 1980 (2) 2-090-26, Startup Report, Supplement 4, dated September 19, 1980, Trimble to Seyfrit Gentlemen:
AP6L's September 3, 1980 letter (Reference 1) concerning the CPC/CEAC-plant computer datalinks requested NRC staff concurrence regarding our desire to leave the datalinks in service permanently.
That letter also included a description of the datalinks, their use and APGL's reasons for making the request.
In Supplement 4 to the ANO-2 Startup Report (Reference 2) AP6L identified the end of the power ascension test program regarding the CPC system as having occurred on Septebmer 11, 1980, based on meeting the final acceptance criteria of all CPC related tests.
Tnese acceptance criteria required review of the CPC related test results by Combustion Engineering.
The CPC-plant computer datalinks had been left in service through this period to support any repeat testing which might have been required based on the CE review.
During this period, APGL reactor engineers continued to use the CPC-plant computer datalink information to perform surveillance on the CPC instrumentation and system function. Fluch of this information was transmitted to CE for review as part of our core-follow program.
Upon completion of the test program, APSL decided to leave the datalinks in service based on our September 3, 1980 request and our belief that plant safety was enLanced by the surveillance which the datalink allowed.
APSL hoped that NRC review would proceed rapidly and that NRC concurrence f
8103 09037h '
MEMDEA MtDOLE SOUTH UTILITIES SYSTEM L
n i
Mr. R. A. Clark February 28, 1981 would be forthcoming. APSL felt confident that NRC review would also conclude that safety was enhanced via use of the datalinks when the facts were reviewed.
During the recent (February 24, 1981) ACRS review of the Core Protection Calculator System,- the subj ect of CPC-plant computer datalinks came up
[
and APSL learned that NRC (NRR) was unaware of AP6L's decision to leave
'l the datalinks installed.
It was further revealed that NRC review of APEL's request had been postponed indefinitely due'to other priorities.
p APSL feels that it is now necessary to reiterate our request and to further explain our belief that the CPC-plant computer datalinks are a f
necessary feature to ensure proper CPC function during core operation.
The datalink was found to be essential during the startup testing period because of the large quantity of time coordinated data that was required
[
7 as part of the startup test program.
Without the datalink, this data could not have been obtained.
l
[
Each reload fuel cycle the CPC database constants utili ed by the CPC system for power distribution synthesis must be verified during the reload startup tests to ensure that the design calculations are consistent with the actual com measurements.
If any constants c e significantly j
different than the measurements, then the CPC penalty factors must be j
adjusted to account for the differences.
The following calculated CPC i
database constants as a minimum are verified during this test program:
i Shape annealing factor matrix constants Boundary point power correlation constants t
CEA shadowing factors Planar radial peaking factors
[
4 The data requirements to support the test program demand the fast, time i
j coordinated, reliable data logging capability provided by the datalink.
?
Consequently, it is essential that the datalink be operable to support l
future reload t 'st program.
}
In addition, since the CPC system has a significant amount of dependence I
i upon predicted core behavior, and particularly because the ANO-2 CPC is a first of a kind system, confirmation that core parameters and CPC response continue to behave as expected throughout the cycle is necessary to ensure continued proper operation.
For this reason APSL has entered into a core-follow program involving Combustion Engineering. APSL has found that the only practical way to obtain the necessary information for this surveillance is via the CPC-plant computer datalinks.
Removal of the datalinks would have resulted in a c'ecrease in the overall confidence that l
the CPCs would perform the required safety function.
j Another important consideration in assuring the proper operation of the CPC system involves periodic surveillance of the sensors used by the l
CPCs.
As discussed in the February 24, 1981 ACRS meeting, the only CPC t
J Mr. R. A. Clark February 28, 1981 failure mechanism which has been identified to have a significant proba-bility of causing " unsafe" failures (failures which do not produce a trip) involve the long term degradation of sensors (drift) in the non-conservative direction. The CPC-plant computer datalinks have been found to be the only existing practical method for accurate trending of sensor performance.
Should the datalinks be removed, sensor failures may remain undetected and the overall capability of the CPCS to perform its safety function may be decreased.
Use of the information contained on the reports obtained from the plant computer has helped AP6L reactor engineering personnel to identify problems and have sensors repaired on several occasions.
These instrument surveillances compliment the periodic power level and flow rate calibrations performed under the Technical Specifications.
Our decision to leave the CPC-plant computer datalinks in service was based on our judgment that their use, described previously, enhanced safety by ensuring proper CPCS operation throughout the core cycle.
Since NRC was unable to review the APSL letter (Reference 1) due to lack of time, APSL concluded that the most prudent course of action was to proceed with the datalinks installed.
Had the datalinks been remnved during this time, it is APSL's belief that it would have had an adverse impact 2pon the overall safety of the plant.
APSL has observed no detrimental effects related to the datalink during the startup test program or subsequent power operation either with respect to the CPC/CEAC system reliability or impairment of the system safety functions.
There has been some concern expressed that the CPC-plant computer datalinks may have contributed to certain CPC channel failures documented on Licensee Event Reports since there have been several CPC failures which resulted in Watchdog Timer initiated channel trips. This is not believed to be the case since the datalinks to the plant computer are not monitored by the watchdog timer.
Further, the software is designed such that higher priority tasks (all protection related software) will override the datalink software even if hung up in a loop.
The usefulness of the datalinks for preventing CPCS failures is probably limited to those cases described previously (identifying instrument drift and verifying adequacy of power distribution synthesis).
Other than during startup testing, AP6L's practice with regard to use of the datalinks precludes taking action to alter or adjust CPC software or hardware based purely upon information obtained from the datalinks.
Ti* datalinks are used to help identify problems, but are not used to make corrective action nor as the sole source for assessing the adequacy of the corrective action.
In summary, it is AP6L's opinion that the CPC/CEAC-plant computer datalinks should remain in use and that their removal would have a detrimental effect on the safe operation of ANO. Your expeditious review of our September 3, 1980 request to leave the datalinks in service permanently is requested.
Should the NRC reach a decision contrary to AP6L's position on this matter, lk
Mr. R. A. Clark February 28, 1981 please provide detailed bases for that decision in your response. AP6L will continue to operate with the datalinks in service pending your reply.
Very truly yours, b Y$-
David C. Trinble Manager, Licensing DCT:TilC:1p cc: ANO-DCC
,