ML20046C612

Informs That Addl Sentence Included W/Util 930122 Proposed Amend to TS for Load Sequencer Should Be Added as Footnote Applicable to Table 3.3-2 Functional Units 1.b,6.d & 8 Instead of Change to Action Statements,Per NRC 930607 RAI

ML20046C612
Person / Time
Site:	Vogtle
Issue date:	08/06/1993
From:	Mccoy C GEORGIA POWER CO.
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
Shared Package
ML20046C613	List: ML20046C612 ML20046C620
References
TAC-M85700, TAC-M85701, NUDOCS 9308110226
Download: ML20046C612 (7)

Text

o

.. .. Georga Power Company 40 hverw. Center Paa wmf

, Past Ote Der I?95

. Erwngt;am Abutma 35?01 Terenbone PD5 877 7122 m

c. x. uccoy Georgia Power te era a rn,< v

/w Poed August 6, 1993 """#"""*"*""

LCV-0063 Docket Nos. 50-424 50-425 TAC Nos. M85700 M85701 U. S. Nuclear Regulatory Commission ATTN: Document ControlDesk Washington, D. C. 20555 Gentlemen:

VOGTLE ELECTRIC GENERATING PLANT REQUEST FOR ADDITIONAL INFORMATION TECHNICAL SPECIFICATION CHANGES FOR LOAD SEOUENCER By letter ELV-03945 dated January 22,1993, Georgia Power Company (GPC) proposed to amend the Technical Specifications for the Vogtle Electric Generating Plant by adding specific requirements for the load sequencer. The NRC's letter dated June 7,1993, requested additional information in support of the requested change. The responses to the NRC questions are attached to his letter.

The proposed changes included an additional sentence to be added to Action statements 14,20, and 23. Based on the NRC's question, GPC has determined that the change should be added as a footnote applicable to table 3.3-2 functional units 1.b,6.d, and 8 instead of a change to the Action statements. This results in a less confusing implementation of the proposed change.

Enclosure 3 ofletter ELV-03945 contained marked up pages and instructions for incorporation. A revised enclosure 3 is attached to this letter and should replace the one that was attached to ELV-03945. Since this does not affect the consequences of the proposed Technical Specification revision, the conclusions of the significant hazards evaluation that was attached to ELV-03945 are not changed.

Sincerel ,

100053 0.9/[

C. K. McCoy l

CKM/HWM/gmb , )

I l

9308110226 930806 f~"

PDR ADOCK 05000424 E P ppg j j

I

.' l l

l i

Georgia Powerkn  ;

U. S. Nuclear Regulatory Commission Page 2 Attachment 1 - Response to NRC Questions Attachment 2 - Revised Enclosure 3 (Instructions for Incorporation) cc: Georgia Power Company Mr. J. B. Beasley, Jr.

Mr. M. Sheibani NORMS U. S. Nuclear Retrulatory Commission Mr. S. D. Ebneter, Regional Administrator Mr. D. S. Hood, Licensing Project Manager, NRR Mr. B. R. Bonser, Senior Resident Inspector, Vogtle LCV-0063 rwm

4 ATTACHMENT 1 l l

VOGTLE ELECTRIC GENERATING PLANT [

RESPONSE TO NRC OUESTIONS ,

i Ouestion:

1. The NRC staff understands that the sequencer design at Vogtle is such that an inoperable sequencer will prevent automatic actuation of the diesel generator (DG) i and Engineered Safety Features (ESF) loads in response to a design basis accident or  !

event, regardless of the availability of offsite or onsite power. With respect to this aspect of the Vogtle design, provide the following:  ;

1.1 State whether the staffs understanding is correct and briefly describe the results of l an inoperable sequencer at Vogtle. (If the staffs understanding is not correct, l discuss how the load sequencer functions and justify why load sequencer inoperability should be treated the same as the condition of one offsite power source  :

and one DG inoperable.) ,

Response: l 1.1 The staffs understanding is correct for major loads which are sequenced with the  ;

following clarification: The sequencer is required for each train for automatic >

loading of major loads for loss of offsite power (LOSP) or safety injection (SI). The -

solid state protection system (SSPS) actuates some smaller loads independent of the sequencer. Also, the diesel generator will auto start on an SI signal from SSPS i independently of the sequencer and is available for manual loading when the ,

sequencer is not available. The loss of a sequencer is not as severe as a loss ofboth  :

onsite and offsite power to the 4.16-kV engineered safety features (ESF) bus,~ l because the sequencer does not affect power source availability.  ;

1 Ouestion:

1.2 You propose an allowed outage time (AOT) of 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> when one automatic load i

sequencer is inoperable. Given Vogtle's load sequencer design and the result of an inoperable load sequencer, on what basis did you conclude that the condition of an ,

inoperable load sequencer is not equivalent to a " dead bus" such that an 8-hour AOT corresponding to TS 3.8.3.1 would be appropriate? Ifyour basis credits manual starting of the DG and manual actuation of ESF loads, state whether procedures are in place for these actions and briefly describe these procedures and their compatibility with Vogtle accident analyses.

l

I j

ATTACIBENT 1 (CONTINUED)  ;

1 RESPONSE TO NRC QUESTIONS Resoong 1.2 A dead bus, corresponding to TS 3.8.3.1, would result from a loss of both the -

preferred offsite power source as well as the diesel. With an inoperable sequencer, the bus would remain energized, and manual loading of emergency loads would be ,

possible. Therefore, and inoperable sequencer is not equivalent to a dead bus and  !

i a longer allowable action time for an inoperable sequencer is justified.

In the event of a LOSP during the limited time during which continued operation with an inoperable sequencer is allowed, the unaffected train would start and provide power for the loads necessary to stabilize the plant. Under these circumstances priority would be given to restoration of the affected train's ac power using the respective diesel generator. Procedural controls exist for manual sequencing ofloads in the event of a LOSP for one train. Emergency Operating Procedure 19100 responds to total loss of ac power. It includes steps for the manual starting of the DG, including the manual loading of the necessary DG support loads.

Question:

2. Your proposed TS change would add a statement to Actions 14,20, and 23 of Table 3.3-2 to instruct plant personnel to " apply the Action statement of Specification 3.8.1.1 for an inoperable load sequencer" if the instrumentation channels are inoperable due to an inoperable load sequencer. In this respect, provide the following:

2.1 Discuss how an inoperable load sequencer causes instrumentation channels to become inoperable.

Response

2.1 An inoperable sequencer would affect the ability of the instrumentation listed in functional units 1.b,6.d, and 8 of TS table 3.3-2 to complete its safety function.

The effect of functional unit 1.b is as described in the response to question 1, above.

"he primary effect would be the failure to automatically load large safety injection ,

loads such as pump motors. It should be noted that the allowable outage time for a train of emergency core cooling system (ECCS) pumps is 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br />. That portion of the safety injection actuation circuitry that does not rely on the sequencer for .

automatic loading would be unaffected by the inoperable sequencer. Therefore, l

ATTACIBiENT 1 (CONTINUED)

RESPONSE TO NRC QUESTIONS an inoperable sequencer would only affect a portion of one train of safety injection automatic actuation circuitry.

Functional unit 6.d is for instrumentation that starts the auxiliary feedwater (AFW) system. An inoperable sequencer would result in a loss of the AFW start signal to the motor driven AFW pump of the affected train, as a result oflow voltage on the 4.16-kV bus. (The other signals for starting the pumps would not be affected). It would also affect the train related signal for starting the turbine driven auxiliary ,

feedwater (TDAFW) pumps . However, the TDAFW also receives a start signal from the other train. Therefore, the TDAFW would remain available to start and supply all four steam generators in the event of a LOSP. This is equivalent to a loss of redundancy to one train, but AFW would continue to be available to both trains in the event of a LOSP during the time that one sequencer is inoperable. t Functional unit 8 applies to instmmentation to sense the loss of voltage or degraded voltage on the 4.16-kV bus. With a sequencer out of service, the affected train would not automatically disconnect from the offsite power source and start the DG on an undervoltage or degraded voltage condition. Should such a condition occur ,

during the limited time that a sequencer is allowed to be inoperable, these functions I could be manually initiated.

Question:

2.2 State the number ofinstrumentation channels made inoperable by each inoperable load sequencer.

Response

2.2 See the response to question 2.1.

QuestioJn 2.3 Regarding Action 14, justify why the action of Specification 3.8.1.1 should be followed rather than the action (s) for an inoperable instrumentation channel with its associated completion time.

ATTACHMENT 1 (CONTINUED)

RESPONSE TO NRC QUESTIONS

Response

2.3 Actions 14,23, and 20 are applied to functional units 1.b,6.d, and 8, respectively.

These actions are also applied to numerous other functional units in table 3.3-2. In order to apply to the diverse functional units, these action statements are overly conservative for the effects of an inoperable sequencer. For example, the effects of an inoperable sequencer coupled with Action 23 for functional 6.d results in an entry into specification 3.0.3 even though AFW remains available to all four steam generators and one train of AFW is completely unaffected. The effect of an inoperable sequencer on functional unit 8 is that the undervoltage and degraded voltage signals for one train would be unavailable. However, the associated action statement only allows entry into 3.0.3. Specification 3.0.3 requires a plant shutdown within I hour.

The introduction of a specific action statement for an inoperable sequencer provides a reasonable allowable outage time based on the limited effect of the inoperable sequencer.

In order to clarify the proposed specification, it is suggested that a note be added to functional units 1.b,6.d, and 8 stating "with channels inoperable due to an inoperable load sequencer, apply the ACTION statement of specification 3.8.1.1 for an inoperable load sequencer." This would allow the change to the Technical Specification to be made without affecting the wording of Actions 14,23, and 20.

'Ib proposed marked up pages are attached and should replace pages 3/4 3-25 and 3/4 3-26 which were included with our letter ELV-03945 dated January 22,1993. It does not affect the conclusion of the significant hazards evaluation that accompanied ELV-03945.

Quntion:

3. On page E2-2 ofyour request of January 22,1993, you indicate that no specific LCO for sequencers is required in Modes 5 and 6. It is not clear from this discussion what impact an inoperable sequencer would have upon DG functions in these modes. Discuss the need for automatic starting and some automatic loading of the DGs in Modes 5 and 6. Include in this discussion how the sequencer operation impacts DG starting and automatic loading of DG support systems (e.g., service water, ventilation)in Modes 5 and 6.

l 4

1 l

ATTACHAENT 1 (CONTINUED)

RESPONSE TO NRC QUESTIONS

Response

3. The engineered safety features actuation system (ESFAS) instrumentation for safety injection, auxiliary feedwater, and loss of voltage for the 4.16-kV bus is not required to be operable in hIodes 5 and 6. However, Technical Specification , 3.8.1.2 requires one diesel generator to be operable in these modes. The DG function in Modes 5 and 6 is independent of sequencer availability. An inoperable sequencer will not prevent the manual starting of a DG and the subsequent loading of ESF equipment to support Modes 5 and 6 requirements. Automatic starting and auto loading of the DGs is not required in Modes 5 and 6.

i l

1