ML13330B387
| ML13330B387 | |
| Person / Time | |
|---|---|
| Site: | San Onofre  |
| Issue date: | 09/16/1988 |
| From: | Office of Nuclear Reactor Regulation |
| To: | |
| Shared Package | |
| ML13330B384 | List: |
| References | |
| GL-83-28, TAC-53148, NUDOCS 8809270034 | |
| Download: ML13330B387 (4) | |
Text
SUPPLEMENTAL SAFETY EVALUATION SAN ONOFRE NUCLEAR GENERATING STATION, UNIT 1 REACTOR TRIP SYSTEM RELIABILITY ITEM 4.2.1 AND 4.2.2 OF GENERIC LETTER 83-28 INTRODUCTION Item 4.2, Parts 1 and 2 of Generic Letter 83-28 addresses actions to be taken by licensees and applicants aimed at assuring an acceptable program of surveillance testing and preventive maintenance of reactor trip breakers (RTBs) in pressurized water reactors,. A safety evaluation, dated March 6, 1987; based on the licensees responses dated November 28, 1983; April 27, 1984; October 1, 1984; and May 14, 1985; found the licensee's program acceptable except for the omission of under voltage trip response time measurement and trending of the resulting data.
The licensee first responded to staff concerns by letter dated June 3, 1987.
The staff requested further information by letters dated July 20, 1987 and March 15, 1988; to which the licensee responded by letters dated November 13, 1987 and July 6, 1988.
EVALUATION San Onofre Unit 1 has two Westinghouse DB-50 breakers in series. It has no bypass breakers. Thus, on-line testing of the RTBs is not possible since tripping of either series breaker results in a reactor trip. Due to space considerations, installation of'bypass breakers would require relocation of several large non safety related equipment cabinets, and modification of the RTB power and control circuitry.
The scram breaker control scheme for San Onofre Unit 1 differs from that of most other Westinghou'se PWR plants. See Attachment A for an illustration of the auto matic portion of the scram control. The scram signals simultaneously deenergize the undervoltage trip device and energize the shunt trip device of each RTB.
Also, the 125 volt D.C. used for control of the RTBs is used to power the gripper P
ioc:K
-5OC, jj02 P
-2 and lift coils of the control rod drive mechanisms. Thus, the circuit arrange ment is fail-safe, i.e., loss of 125 volt D.C. power will drop the rods even if the RTBs fail to open.
The NRC staff's primary concern with RTBs is failure on demand of the undervoltage trip device. Following the Salem ATWS event, features were added to most West inghouse PWR RTBs to automatically initiate or demand the shunt trip as well as the undervoltage trip. The shunt trip attachment exerts much greater force and is considered to have an order of magnitude higher reliability than the under voltage trip attachment. However, on most Westinghouse PWRs, the shunt trip de vice is not fail-safe. Loss of control power to the shunt trip device on these PWRs would prevent the shunt trip device from functioning on demand and would not drop the control rods. Under this scenario, an automatic reactor scram would depend on satisfactory functioning of the undervoltage trip device.
In addition to the fail-safe feature of the San Onofre Unit 1 scram design, historical breaker performance data has demonstrated a high degree of reliability in that the breakers have never failed to open on demand. Changes to the RTB power and control circuitry would complicate the present design, and it is judged that the marginal reliability to be gained by response time testing of the under voltage trip device for the San Onofre 1 design would be offset by the reliability loss resulting from the circuit complication. Thus, we conclude that on-line response time testing of the undervoltage trip device is not necessary.
CONCLUSION The reactor scram circuitry at San Onofre Unit 1 differs from that at most Westinghouse PWR plants in that loss of voltage to the shunt trip attachment of the reactor trip breakers results in a reactor scram. Thus, when the reactor is operating, there is assurance that power is always available to the shunt trip attachment. Since the shunt trip attachment exerts a greater
-3 force than the undervoltage trip attachment and therefore is considerably more reliable in tripping the RTB than the undervoltage trip attachment, we find that design changes to permit on-line response time testing of the undervoltage trip attachment will not improve overall reliability, and such on-line testing is not required.
SRAA4 C.fSA
~a~a I
SCRA F
1I A
C, -
C.
04
,WFF Avl~
OCOF co.i,~ tw MWArr PmW SCRAMXWrPWA COAKR OAr~L fNt "CC) rri