ML20236E519
| ML20236E519 | |
| Person / Time | |
|---|---|
| Site: | Millstone  |
| Issue date: | 05/31/1989 |
| From: | Boyle M Office of Nuclear Reactor Regulation |
| To: | Mroczka E NORTHEAST NUCLEAR ENERGY CO. |
| References | |
| TAC-60207, NUDOCS 8906060266 | |
| Download: ML20236E519 (4) | |
Text
-___
n May31[1989 Docket No. 50-245 DISTRIBUTION
- SVarga,
~le.
NRC & Local PDRs NWDocket@14/E/4.$Norris Mr. Edward J. Mroczka Senior Vice President BBoger,14/A/2 MBoyle Nuclear Engineering and Operations OGC EJordan, 3302 MNBB JLazevnick, 8020 BGrimes,)9/A/2 Northeast Nuclear Energy Company ACRS (10 Plant Fife P. O. Box 270 Hartford, Connecticut 06141-0270
Dear Mr. Mroczka:
SUBJECT:
UNDERVOLTAGE DETECTION - MILLSTONE NUCLEAR POWER STATION, UNIT NO 1 (TAC NO. 60207)
The staff has reviewed your August 15 and December 2, 1988, submittals regarding a revised undervoltage detection scheme at Millstone Unit 1.
- Also, we have received your May 25, 1989, letter stating that the undervoltage protection modifications have been installed and made operational during the current refueling outage.
The revised undervoltage detection logic, as described in your submittals, does not provide for an automatic transfer of a safety division to its emergency onsite power source when a loss of normal power (LHP) occurs on only that one safety division.
It is designed to only transfer the safety divisions to the emergency standby power supplies when normal power is lost to both safety divisiens.
The staff has detennined that this design does not meet the intent of GDC 17
'as defined in the review criteria for this topic in the Systematic Evaluation Program. NNECO should therefore develop an undervoltage detection scheme at Millstone 1 that automatically transfers a single safety division to its emergency onsite power source when a LNP occurs on only that division.
Sufficient analysis, integrated testing, and operator training should be p ovided relative to the new design to provide a good level of confidence that it can be utilized successfully.
This is acceptable because this design s ffers improvements (y be utiliz In the interim, the revised undervoltate detection scheme masuch as location of' existing undervoltage protection the relays to the Class IE buses) over tt $
logic. The enclosed staff position provk '.s further details on the findings and recommendations of the staff's review,f the Millstone Unit I undervoltage l
0Foi f
eiPof oD I
pp
.g
- Mr. Edward J. Mroczka.
protection scheme.
In addition, since the staff will not be acting upon the proposed Technical Specifications (TS) changes related to undervoltage protection until this issue is resolved, MNECO should continue to use the current TS augmented, as necessary, by administrative procedures based on the proposed TS.
Sincerely,
/s/
Michael L. Boyle.. Project Manager Project Directorate I-4 Division of Reactor Projects I/I!
Office of Nuclear Reactor Regulation
Enclosure:
As steted cc w/ enclosure:
See next page g-l l
l
[ TAC 60207]
MBoyle:1m)$
LA:PDI-A '
PM:PDI-4g PD: DI SNoPfTs Jsto 15/31/89 05/$l /89 05/{/.89
y.
NRC POSITION ON MILLSTONE 1 UNDERVOLTAGE DETECTION SCHEME In letters dated August 15, 1988 and December 2, 1988, Northeast Nuclear Energy Company (NNECO) described a revised undervoltage detection scheme at Millstone Unit I and provided revised technical specifications for the new scheme.
In a letter. dated April 25, 1989, NNECO provided responses to staff questions and additional information on the revised undervoltage detection logic that included schematic wiring diagrams of the logic.
The revised undervoltage detection scheme for Millstone 1 does not provide for an automatic start and connection of the emergency standby power source to its Class 1E division when nomal power is lost to only that one division. The under-voltage detection scheme will only signal the emergency standby sources to start 1
and connect to.their respective divisions when nomal power is lost to both divisions. Loss of normal power (LNP) is defined as loss of power to the safety buses from the nomal station service transformer that is supplied from the out-put of the main generator, together with loss of power from the reserve station 53rvice transformer (offsite source) that is supplied from the Millstone switchyard.
The staff finds that the omission of the capability to automatically transfer to emergency power on a LNP to a single division does not meet the intent of GDC'17.
GDC 17 specifies that the offsite and onsite power systems shall each perform their safety function assuming the other system is not available., GDC 17 also speci-fies that the onsite distribution system shall have sufficient independence to perfom its safety function assuming a single failure. The undervoltage protection provisions were reviewed against GDC 17 in the Systematic Evaluation Prgoram. The Millstone 1 design does not have the required independence within its onsite distribution, system because the automatic transfer of a division to its onsite emergency pcwer supply is dependent not only on a loss of power to its own division but also on a loss of power to the redundant division. Millstone 1 does not meet the single failure requirement because a LNP (same as a loss of offsite power) to a single safety division leaves only one division energized which is then vulnerable to a single failure in any system in that division needed to cope with the occurrence or accident.
This is especially true if the LNP to a single division occurs concurrently with a large break LOCA since there will likely be insufficient time to manually reenergize the division.
The fast transfer scheme that is utilized at Millstone 1 could also contribute to the likelihood of a LNP occurring on a single safety division. This scheme, which is used to transfer the safety buses from the turbine generator supply to the offsite power supply upon loss of the turbine generator and is initiated by non-Class 1E circuits, requires that the running supply breakers be opened and the incoming supply breakers be closed within.1 second of each other.
If the transfer is not completed within.1 second it is automatically blocked.
Fast transfer failures have been reported in LERs.
Because a turbine generator trip occurs in accident scenarios followed by a fast transfer, the LNP on a single safety division occurring as a result of a subsequent failed fast transfer is not considered unlikely during accidents.
1 L
r
-)
4 During a meeting with the staff on May 4,1989, and in their previous correspond-ence, NNECO expressed reservations regarding the effects of powering only one division from its emergency standby power source on a LNP to only that division.
The licensee indicated concern with the ability of the operators to identify and assess the develeping scenario due to the asymmetrical configuration of the two divisions. They are also unsure of the system response if only a While we recognize these single division is transferred to emergency power.
concerns we have not heard convincing arguments to warrant the vulnerability of.
the plant to the described single failure which remains with the revised under-voltage detection scheme.
It is the NRC staff's position that NNECO should develop an undervoltage detection scheme that automatically transfers a single safety division to its emergency onsite power source when a LNP ocet:rs on only that division to provide the independent operation of emergency sources required by GDC-17.
Sufficient analysis, integrated testing, and operator training should be provided relative to the new design to provide a good level of confidence that it can be utilized successfully.
In the interim, the revised undervoltage detection scheme described'in NNECO letters dated August 15, 1988, December 2, 1988, and April 25, 1989 ma be utilized. This is acceptable because this design offers improvements (y such as location of the relays to the Class IE buses) nvr.' the existing undervoltage In addition, since the ste' will not be acting upon the protection logic.
proposed Technical Specification changes relate > to undervoltage protection, until this issue is resolved, NNECO should continue to use the current Specifications augmented, as necessary, by administrative procedures based on the proposed specifications.
1 l
i 1
)
L
. I