0CAN028907, Forwards Summary of SPDS Isolation Device Test Results

Forwards Summary of SPDS Isolation Device Test Results

ML20235W166
Person / Time
Site:	Arkansas Nuclear
Issue date:	02/28/1989
From:	Howard D ARKANSAS POWER & LIGHT CO.
To:	Craig Harbuck Office of Nuclear Reactor Regulation
References
0CAN028907, CAN28907, TAC-51219, TAC-51220, NUDOCS 8903100525
Download: ML20235W166 (3)

Text

_

~ '

ARKANSAS POWER & LIGHT COMPANY  ;

February 28, 1989 1

BCAN028907 l U. S. Nuclear Regulatory Commission Document Control Desk Mail Station P1-137 Washington, DC 20555 ATTN: Mr. Craig Harbuck, NRR Project Manager NRR Mail Stop 13-D-18  ;

SUBJECT:

Arkansas Nuclear One - Units 1 & 2 Docket Nos. 50-313 and 50-368 '

License Nos. DPR-51 and NPF-6 SPDS Isolation Device Test Results (TAC Numbers S1219 and 51220)

Dear Mr. Harbuck:

1he remaining open issue from the NRC staff's review of the ANO-1 and 2 Safety Parameter Display System (SPDS) involved the adequacy of the testing  :

of certain isolation devices installed for the SPDS to assure separation of i Class 1E from non-Class 1E circuitry and equipment. AP&L responded to an NRC request for additional information by our letter dated June 30, 1988

($CAN068812) and specifically addressed the adequacy of testing of certain isolation devices used for the installation of the SPDS. At that time we committed to evaluate the manufacturers' testing for applicability to ANO, and perform additional testing if necessary and report the results to the NRC.

This issue was subsequently discussed in the October _13, 1988 AP&L/NRC meeting on longstanding safety issues, and AP&L committed to perform and submit the results of testi7g of the certain model of isolator in question. This testing is now complete, and AP&L has reviewed the results to verify that the adeq acy of the isolation devices has been demonstrated. Attached is the summary of the test results which we committed to provide. AP&L expects this information to allow the NRC to effectively close this issue for ANO.

Very truly yours, f)

Dan R. Howard Manager, Licensing DRH:RBT 3 Attachment g l

y~ [ 3$hh

[

P

_ _ _ _ . _ _ _ _ _ _ _ l

\  %

. 2 ATTACHMENT TO OCAN028907 The following is a summary of test results for the' Rochester Instruments' single output signal isolators Model SC-1302 used in the SPDS input circuits.

The purpose of the test was to verify electrical isolation capability of the isolator.

IEEE 384-1981 Section 7.2.2.1. defines an isolating device as follows: "A device is considered an electrical isolation device for instrumentation and contro1' circuits'if it is applied so that (a) the maximum credible voltage or current transient applied to the device's non-Class 1E side will not degrade the operation of the circuit connected to the device Class 1E or associated side below an acceptable level; and (b) shorts, grounds,.or open circuits occurring in the non-Class-1E side will not degrade the circuit. 1 connected to the device Class 1E or associated side below an acceptable i level ..."

The isolator testing was performed at National Technical Systems Laboratories, Acton, Massachusetts, in accordance.with Test Procedure 25647-89N. The tests included Baseline Functional Test (calibration), Dielectric Withstand Voltage Test, fault testing (120 VAC 20 amp), shorts, grounds, and open circuit tests.

A total of four isolators were tested. These isolators represented the types and vintage of isolators employed at ANO. Internal fuses were jumpered.

Shorts, grounds, and open circuit on the output of the isolator had no effect on the 1E input signal.

The 120 VAC 20 anp maximum credible fault voltage and current applied between each output terminal and ground had no effect on the 1E input signal. The application of maximum credible voltage across the output terminals of the isolator provided destructive testing and damage to all four isolators. Three out of.four isolators let energy through the barrier at the time the fault was applied, but the let through disappeared as the destruction of the isolator began. The measured voltage excursion on the 1E side was 0.5 Vdc and 0.8 Vdc for the devices with 1-5 Vdc input, and 4.85 mA for the device with 4-20 mA input. In no case did destruction of the isolator result in a hard breach of the barrier in such a way that the output was connected to the input and the fault was then directly connected to the safety system input.

Following the fault voltage test, a hi pot (resistance) test between the output and input for each isolator confirmed isolation integrity of the devices. )

An evaluation of the impact of the postulated input signal excursion due to application of the fault voltage on the isolator output was conducted for ANO Units 1 and 2 SPOS input signal instrumentation loops incorporating the j RIS SC-1302 isolators. It has been determined that the input signal excursion l of the magnitude experienced during the fault voltage test should have no I noticeable effect on the operability and function of the class 1E input circuit I devites which, at most, include an indicator in addition to a transmitter and )

loop po'ter supply. There are no control or trip devices in the SPDS isolator input circuits. The SPDS inputs come from isolated redundant Class IE instrumentation channels.

1 J

l I

l l

I'nthree cases, the isolator input signal originates from a multi-output  !

l transmitter; however, in each case, the signal to the SPDS isolator is isolated from other outputs internally to the transmitter.

During the test, isolator power input voltage and current was monitored in addition to the signal input and output. At the time the fault voltage was applied to the isolator output terminals, the power input current increased on four units then returned to initial value on two units and stayed higher than initial value for the other two units. The maximum recorded power input current at 120 VAC was 555 mA, 338, 147, and 145 mA. The initial values were 71 mA, 53 mA, 51 mA, and 45 mA correspondingly. The unit which initially drew 71 mA appeared to be suspect since it drew significantly higher initial current than the next unit. AP&L decided to investigate further by extending the post test monitoring for three hours. Three of the units had no significant change in current; however, the suspect unit had a significant current change which blew a two amp fuse after one and a half hours. It is therefore believed that this is an anomalous result; further investigation is being performed to determine the' root cause of the anomalous behavior of the unit.

Since results of the destructive fault testing cannot be fully reproduced, it is very unlikely that results obtained for the suspect test unit can be obtained again. A review of available reports on fault testing of RIS SC-1302 signal isolators, including NUREG/CR-3453, indicates that in none of the previous test units was the power input current draw of the magnitude observed on the suspect test unit. However, AP&L has evaluated the impact of an increased power input current as well as 1E input signal excursion on the ANO Units 1 and 2 instrumentation loops employing RIS SC-1302 signal isolators.

Evaluation of the other circuits revealed that no control or protective function is affected by worst case postulated failure of the isolator. The extent of the impact would be limited to the possible loss of indication only. This failure would be readily detectable and applies only to approximately four instrument loops in Unit-2 and two loops in Unit-1.

Based on the tests and the follow-up evaluation (with consideration that the excessive current draw of the suspect unit constitutes anomalous behavior thought to be related to random failure), it has been determined that the Rochester Systems SC-1302 signal isolators employed at ANO meet the intent of IEEE-384 1981 for electrical isolation between Class 1E and non-Class 1E SPDS ,

circuits. This specific testing further substantiates AP&L's position that '

the isolation devices used for the SPDS installation at ANO are " suitable",

per the requirements of NUREG-0737 Supplement 1, Item 4.1.c.

_ _ - _ _ - _ - _ _ _ _ _ _ _ _ _ _ - _ _ _ - _