ML20236M057
Text
)
UNITED STATES 8
'I NUCLEAR RESULATORY COMMISSION 1
wommoTON, D.C. SBHMEN j
r December 3, 1996 MEMORANDUM TO:
Joseph W. Shea, Project Manager Project Directorate 1-2 FROM:
Christopher I. Grimes, Chief h Technical Specifications Branch
SUBJECT:
REQUEST FOR WRITTEN INTERPRETATION OF TECHNICAL SPECIFICATIONS In your note of November 26,1996, you requested a written interpretation of the technical specifications (TS) for Peach Bottom Atomic Power Station, Units 2 and 3.
Specifically, you requested to know under what plant conditions, including reference to the diesel generator testing conditions, SR 3.8.1.17 must be met and how that explanation fit with the requirements of SR 3.0.1. You requested the interpretation following your receipt two days ago of a verbal position on the meaning of the TS, from members of this branch, which apparently differed from that taken by the NRC staff in an August 1996 telephone call with the licensee.
j in order to ensure that this response to your request adequately addresses the matter, a brief discussion of the sequence of events is warranted. in August 1996, the senbr resident inspector raised a number of questions about the licensee's evaluation of potential problems with the EDG control circuits (see inspection Report 50 277 and 278/96-06).
The apparent inability of the circuitry to satisfy SR 3.8.1.17 (specifically 3.8.1.17 b) raised
)
a potential TS compliance issue which resulted in the telephone call between the staff I
(Region I, the senior resident inspector, NRR projects, TSB, and EELB) and the licensee.
During that telephone call, it was decided that there was not a TS compliance issue but that the licensee should submit a TS change to clarify SR 3.8.1.17 and support its position that the loading of a LPCI pump on an EDG without a time delay is acceptable. As discussed in your request, while awaiting approval of the TS change the licensee committed to place the LPCI pump, associated with any EDG placed in test,into a.
condition where it would not automatically start and declare that LPCI pump inoperable for the duration of the test of the EDG.
While no written record was made to specifically document the basis for the position taken by the staff during the August telephone call, the inspection report referenced above indicates that the staff accepted the licensee's position that the SR was only applicable during EDG testing. I believe the staff position w:,s based on more than the licensee's position (which, by itself, would be inadequate justification, as discussed further below).
How ever, without en accurate written record of the staff decision, I am unable to reconstruct the basis for the decision. Therefore, rather than expend additional staff resources on a matter we all agree has no safety significance, I defer to the position CONTACT:
James Luehman, TSB/NRR 415-3150 9807130283 980624 PDR FOIA l
, o.,
i J.Shea December 3, 1996 outlined below which was formulated when members of TSB began to consider the licensees request for a TS change. I recognize that this position is fundamentally different in one respect from the one reach in August, as it concludes that there is a compliance issue. That conclusion will unfortunately require the staff to have to expend additional resources on this matter which, as I have stated above, has no safety significance and is best resolved by amending the TS to remove what is either an ambiguity or at most a literal compliance problem.
SR 3.8.1.17 contains a surveillance requirement for a feature of the emergency AC power system known as the Emergency Ciess! Generator (EDG) return-to-standby feature. SR 3.8.1.17 states:
Verify with a DG operating in test mode and connected to its bus, an actual l
or simulated ECCS initiation signal overrides the test mode by:
s.
Returning DG to ready-to-load operation; and b.
Automatically energizing the emergency load from offsite power.
With regard to item "b", the Bases for SR 3.8.1.17 state:
The requirement to automatically energize the emergency loads with offsite power is essentially identical to that of SR 3.8.1.12. The intent in the requirements associated with SR 3.8.1.17.b is to show that the emergency j
loading is not affectedby the DG operation in test mode in lieu of actual
]
demonstration of corsnection and loading of loads, testing that adequately shows the capability of the emergency loads to perform these functions is acceptable. This testing may include any series of sequential, overlapping, or total steps so that the entire connection and loading sequence is verified (emphasis added).
You stated that, in August 1996, the licensee and NRC staff became aware that implementation of the return-to-standby feature caused the LPCI pump to effectively sequence start onto its offsite power source in a manner that was inconsistent with the start sequence design basis description in the Bases for T'S 3.3.5.1, "ECCS instrumentation." There was a concern that operation of the LPCI pumps in conjunction with the operation of the retum to-standby feature was not consistent with the Bases for SR 3.8.1.17 in that emergency loading is affected (by virtue of the new LPCI start sequence) by operation of the EDGs in the test mode. Thus the staff was concerned that the licensee did not meet SR 3.8.1.17.
1 As you noted, SR 3.0.1 states:
SRs shall be met during the MODES or other specified conditions in the Applicability for individual LCOs, unless otherwise stated in the SR. Failure to meet a Surveillance, whether such failure is experienced during the i
performance of the Surveillance or between performances of the Surveillance, shall be failure to meet the LCO. Failure to perform 1
l i
?
J. Shou December 3, 1996 a Surveillance within the specified Frequency shall be failure to meet the LCO exc@t as provided in SR 3.0.3. Surveillance do not have to be performed on
'noperable equipment or variables outside specified limits.
Bec6use there are no notes associated with SR 3.8.1.17 that modify the applicability of the LCO for this surveillance requirement, the SR must be met at all times while in the applicability of LCO 3.8.1, i.e., Modes 1,2, and 3. Therefore, once it was determined that SR 3.8.1.17 could not be met due to the effect of the return to-standby feature on the emergency loading (LPCI start sequence), the EDGs should have been declared inoperable.
Until such time that the staff concludes that the licensee's analysis of the current LPCI start sequence with the EDG in the test mode is acceptable and the TS as well as the associated Bases are modified to reflect this analysis, the licensee remains in non-compliance with LCO 3.8.1.
Again, given that the only time the licensee cannot meet its licensing basis is when the EDG is in the test mode, and the licenses has taken steps to address this problem when the EDG is in the test mode, the issue is only one of literal non-compliance and is not safety-significant.
cc:
RZimmerman SVarge JZwolinski 9
l l
l l
l