ML20033G650
| ML20033G650 | |
| Person / Time | |
|---|---|
| Site: | Sequoyah  |
| Issue date: | 04/03/1990 |
| From: | Office of Nuclear Reactor Regulation |
| To: | |
| Shared Package | |
| ML20033G649 | List: |
| References | |
| NUDOCS 9004100462 | |
| Download: ML20033G650 (2) | |
Text
e*
II Al0 4
- o UNITED STATES 8,
- r p,
NUCLEAR REGULATORY COMMISSION
.t WASHING TON, D. C. 20666 ENCLOSURE I
SAFETY EVALUATION BY THE OFFICE OF NUCLEAR REACTOR REGULATION SOLID STATE PROTECTION SYSTEM (SSPS) INPUT AND MASTER RELAY RESPONSE TIME TESTING REQUIREMENTS TENNESSEE VALLEY AUTHORITY SEQUOYAH NUCLEAR PLANT, UNITS 1 AND 2 DOCKET NOS. 50-327 AND 50-328
1.0 INTRODUCTION
In its letter dated November 28, 1989, theTennesseeValleyAuthority(TVA) explained that in the upcoming Cycle 4 refueling outages for Sequoyah Units 1 and 2, it proposes to implement an upgrade to the reactor protection system (RPS) by replacing the present Foxboro H.Line analog process measurement equipment with Westinghouse Eagle-21 digital process measurement equipment.
TVA stated that the Eagle-21 system has built-in automated features for performing functional, calibration and response time testing.
These features will thoroughly test the Eagle-21 system including both its reactor trip (RTS) and engineered safety feature (ESF) functions.
This would not include response time testing of the SSPS input and master relays.
The current SSPS semi-automatic tester also does not response time test these relays.
In lieu of developing procedures and test methods to response time test the SSPS input and master relays with the Eagle-21 RPS configuration. TVA requested that channel functional testing in the Technical Specifications be considered a sufficient verification of the relays' ability to perform their safety function because the additional information gained from the response time testing of the SSPS input and master relays does not warrant the resource expenditures needed to develop the new methodology and procedure to test the relays in the Eagle-21 i
RPS configuration.
Therefore, TVA is proposing the elimination of response time testing requirements for the SSPS input and master relays in the Eagle-21 RPS configuration.
Sequoyah currently performs response time testing on all components of the l
analog process instrumentation system and the SSPS, including the input and master relays.
This is done for these relays by measuring the time from l
application of a step change in instrument loop current to the time when the relay contact changes to its trip state.
2.0 DISCUSSION AND EVALUATION The TVA proposal to eliminate the response time testing requirements for the SSPS input and master relays means that not only would these relays not be tested for response time by any of the automatic functions of either the Eagle-21 system or the SSPS tester, but also that response time testing done at refueling outages would not include these relays.
9004.100462 9004o3 PDR ADOCK 0500o327 l
P PDC
)
i 2
Currently, the response time tests are performed only at refueling outages when response time testing could be performed on the complete instrument channel l
from the sensor output through the logic to the actuated device (RTS breaker.
l ESF valve operator, contactor, motor control center, etc.) when the unit was shut down.
This is done then because an entire train of the RTS/ESFAS could be taken out of service for such testing.
This becomes possible because the current Sequoyah Technical Specifications require that, for the RTS and ESFAS, the response time for each trip or actuation function be demonstrated to be within its limit only at least once per 18 months on a staggered test basis.
l Thus, for a function having four channels, each channel is tested at least l
once every 72 months and the SSPS logic trains tested only once every 36 months.
l The response times of these relays are an important part of the total response t
l time of the RTS and ESF functions because the SSPS input relays contribute directly to the response times of both the RTS and ESF functions while the SSPS master relays are in the output circuits of the ESF actuation channels and contribute only to the total response times of the ESF functions.
Even though these relays would be functionally tested by the various testing schemes of the Eagle-21 system and the SSPS, the testing would not detect slow or delayed operation of the relays.
Review of the relay response time data for.Sequoyah provided in the enclosure to TVA's November 28, 1989 submittal shows that the relay response times vary from a low of 4 milliseconds to a high of 3.26 seconds with variatians in response time measurements on individual relays of as much as 1.4 seconds. The staff believes this is too wide a variation in response times to be ignored and not tested.
In addition, for a previous similar case involving the General Electric solid state Nuclear System Protection System (NSPS) for the Clinton Nuclear Power i
Station, the staff, in Section 7.2.3.3 of their SSER 6 (NUREG 0853) dated July 1986, required the licensee to develop methods and procedures to test areas of the protection system that are not tested for response times by the automatic self-test system.
3.0 CONCLUSION
S Based on our review of the information submitted by TVA, supplemental information obtained in our December 11, 1989 telephone conference call with TVA and information from referenced NRC documents, we have concluded that, to maintain consistency with previous staff decisions and because the variations in response times of the SSPS input and master relays tested at Sequoyah are sufficiently large, TVA should be required to develop means for testing the total response times of the trip and actuation channels for each RTS and ESF function.
These tests should include all equipment in the functional channel from the sensor output through the octuated device (contactor, RTS breaker, motor control center, etc.), including the SSPS input and master relays.
These tests should be performed on a schedule consistent with the current requirements in the Sequoyah Technical Specifications.
Principal Contributor:
D. Lasher Dated: April 3, 1990
.