ML20070M671

Application for Amends to Licenses NPF-68 & NPF-81,changing TS 3/3.8.1.1, AC Sources Operating, Based on GL 93-05,RG 1.9 & NUMARC 87-00,Rev 1

ML20070M671
Person / Time
Site:	Vogtle
Issue date:	04/28/1994
From:	Mccoy C GEORGIA POWER CO.
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
Shared Package
ML20070M674	List: ML20070M671 ML20070M676
References
RTR-REGGD-01.009, RTR-REGGD-1.009 GL-93-05, GL-93-5, LCV-0212, LCV-212, NUDOCS 9405020375
Download: ML20070M671 (9)

Text

-

bCO!!yl.5 bOWC? CU % D 40 i i errti: r, Center Pmkw3y r t V CY('O Daw 1;1.15 4

B m m .;u oj, R '3 w.a T m 1

]t v; d f.r.1: e 's Hil11:?

JEmme.t C. K. McCoy Georgia Power vt,4 n a.m rm .a von rw e April 28, 1994 W*o~" emc a *m LCV-0212 Docket Nos. 50-424 50-425 U. S. Nuclear Regulatory Commission ATTN: Document Control Desk Washington, D. C. 20555 Gentlemen:

VOGTLE ELECTRIC GENERATING PLANT REQUEST TO REVISE TECilNICAL SPECIFICATIONS RELATED TO AC SOURCES OPERATING In accordance with the provisions of 10 CFR 50.90, Georgia Power Company (GPC) proposes to amend the Vogtle Electric Generating Plant (VEGP) Unit 1 and Unit 2 Technical Specifications (TS), Appendix A to Operating Licenses NPF-68 and NPF-81. The proposed changes would afTect TS 3/4.8.1.1, AC Sources Operating, and are based on Generic Letter 93-0$, "Line-Item Technical Specifications Improvements to Reduce Surveillance Requirements for Testing During Power Operation," Regulatory Guide 1.9, Revision 3, " Selection, Design, Qualification, and Testing of Emergency Diesel Generator Units Used as Class lE Onsite Electric Power Systems at Nuclear Power Plants," and NUM ARC 87-00, Revision 1, " Guidelines and Technical Bases for NUMARC Initiatives Addressing Station Blackout at Light Water Reactors."

The proposed changes and their bases are provided in enclosure 1. An evaluation pursuant to 10 CFR 50.92 showing that the proposed changes do not involve significant hazards considerations is provided as enclosure 2, and the marked-up pages are provided as enclosure 3.

In accordance with 10 CFR 50.91, the designated state official will be sent a copy of this letter and all enclosures.

Georgia Power Company requests approval of the proposed amendment by September 1994 which corresponds to the next scheduled refueling outage for Unit 1.

000000 9405020375 940428 -w1 i PDR ADOCK 05000424 "

'\

P PDR  ;

l U. S. Nuclear Regulatory Commission LCV-0212' Page 2

i Mr. C. K. McCoy states that he is a Vice President of Georgia Power Company and is authorized to execute this oath on behalf of Georgia Power Company and that, to the best of his knowledge and belief the facts set forth in this letter and enclosures are tnie.

GEORGIA POWER COMPANY BY: .

C. K. McCoy Sworn to and subscribed before me this c2 f day of [24t .[ ,1994.

/

4 G(A 4./

N6]ary Public ,

NYCOMIS20N EFIRES1/ERf 12.1997 CKM/NJS

Enclosures:

1. Basis for Proposed Change 2.10 CFR 50.92 Evaluation

3. Marked-Up Pages xc: Georgiafower Comp _any Mr. J. B. Beasley, Jr.

Mr. M. Sheibani NORMS U. S. Nuclear Regulatory Commission Mr. S. D. Ebneter, Regional Administrator Mr, D. S. Hood, Licensing Project Manager, NRR Mr. B. R. Bonser, Senior Resident Inspector, Vogtle State of Georgia Mr. J. D. Tanner, Commissioner, Department ofNatural Resources

.a

ENCLOSURE 1 VOGTLE ELECTRIC GENERATING PLANT REQUEST TO REVISE TECIINICAL SPECIFICATIONS RELATED TO AC SOURCES OPERATING IIASIS FOR PROPOSED CII ANGES hop 2sni Chajigts The proposed change would revise Technical SpeciGcation (TS) 3/4.8.1.1, AC Sources Operating to:

• Incorporate the recommendations of Generic Letter 93-05, "Line-Item Technical Specifications Improvements to Reduce Surveillance Requirements for Testing During Power Operation,"

e incorporate provisions of Regulatory Guide 19, Revision 3, " Selection, Design, Qualification, and Testing of Emergency Diesel Generator Units Used as Class lE Onsite Electric Power Systems at Nuclear Power Plants," as they relate to reducing DG stress and wear due to testing and the elimination of unnecessary reporting requirements, and

. Incorporate the recommendations of NUM ARC 87-00, Revision 1, " Guidelines and Technical Bases for NUMARC Initiatives Addressing Station Blackout at Light Water Reactors," as they relate to the test frequency for a problem diesel generator (DG).

SpeciGcally, the changes are as follows:

1. Per Generic Letter (GL) 93-05, action statement a. would be revised to delete the requirement to test the DGs in the event of an inoperable offsite circuit.

2. Per GL 93-05, action statement b. would be revised to climinate the need to test the operable DG in the event that the other DG became inoperable due to an inoperable support system or an independently testable component in addition to the existing .

provision excluding preplanned preventive maintenance or testing. Furthermore, if the operable DG must be tested, it would have to be tested within 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> (as opposed to 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />) unless the absence of any potential common mode failure for the remaining diesel generator is demonstrated.

3. Per GL 93-05, action statement c. would also be revised to eliminate the need to test the i operable DG in the event that the other DG became inoperable due to an inoperable  !

support system or an independently testable component in addition to the existing-provision excluding preplanned preventive maintenance or testing. In addition, the operable DG would not have to be tested if the absence of any potential common mode failure for the remaining diesel generator is demonstrated A reference to action statement i

a. would also be deleted because of the change to action statement a. described above.

El-1

.. - . --- .- . . - . . . . - . . -- - - .~ _ .- ..

3 ,

f ,

ENCLOSURE 1 (CONTINUED)

VOGTLE ELECTRIC GENERATING PLANT .

REQUEST TO REVISE TECIINICAL SPECIFICATIONS

• RELATED TO AC SOURCES OPERATING BASIS FOR PROPOSED CIIANGES

4. Per GL 93-05, action statement c. would be revised to eliminate the need to test the DGs in the event that two offsite circuits are inoperable.

5. Per Regulatory Guide (RG) 1.9, Revision 3, surveillance requirement (SR) 4.8.1.1.2.a.4 ,,

would be revised to incorporate the provision to perform routine monthly testing by  :

gradually accelerating the DG to operating speed as opposed to requiring the DG to attain y rated voltage and frequency within 11.4 seconds. As a direct result of this proposed change, action statements b., c., and f. would be revised to reference SR 4.8.1.1.2.g.1 instead of SR 4.8.1.1.2.a.4 in the event that an operable DG must be tested when the other DG is inoperable. This has the effect of requiring the operable DG to be fast-started far -

testing pursuant to the action statement.

6. Consistent with the recommendations of GL 93-05, SR 4.8.1.1.2.g.2 would be revised to allow the DG to be gradually loaded as opposed to a fast loading. .

7. Per RG 1.9, Revision 3, SR 4.8.1.1.2.h.7 would be revised to separate the 24-hour endurance run from the hot restart test. As a result, new SR 4.8.1.1.2.h.8 would be created to require the hot restart test. The DG would be operated for a minimum of 2 ,

hours at a load of 6800-7000 kw, and the DG would be shut down. - Within 5 minutes of shutdown, the DG would be restarted and required to attain rated voltage.and frequency -

within 11.4 seconds. Footnote ## (which provides for reperforming the hot restart test without repeating the 24-hour endurance test) would be deleted since it would no longer-be required. Existing SRs 4.8.1.1.2.h.8,9,10,11, and 12 would be renumbered to >

accommodate the addition of new SR 4.8.1.1.2.h.8.

,j

8. Per RG 1.9, Revision 3, and the Improved Technical Specifications, Revision 0, dated -

September 28,1992, SR 4.8.1.1.3, Reports, would be deleted.-

- 9. Per NUMARC 87-00, Revision 1, table 4.8-1, Diesel Generator Test Schedule, would be:

revised to incorporate the test schedule of section D.2.4.'4, Appendix D to NUMARC 87- l 00, Revision 1. Specifically, under the proposed schedule, testing pursuant to SR 4.8.1.1.2.a would be conducted monthly provided the number of valid failures in the last 7

' 25 demands for a given DG is less than or equal to 3. If the number of valid failures 'is 41 ,

or more, testing would be conducted at least once per 7 days (but at intervals of no less than 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />) until seven consecutive failure free starts from standby conditions and load-- A run demands have been performed. Note that both NUMARC 87-00, Revision 1, and RG 1.9, Revision 3, use and defme the terms " start demand, start failure, load-run demand, .

and load-run failure" rather than the old RG 1.108 terminology of valid tests. In fact,-

. section D.2.4.4, Appendix D to NUMARC 87-00 refers to the last 25 " demands" rather-El-2 w # . - . .s- . ~ . ,,-o, , * , , . ,- ,

~

ENCLOSURE 1 (CONTINUED)

VOGTLE ELECTRIC GENERATING PLANT REQUEST TO REVISE TECHNICAL SPECIFICATIONS RELATED TO AC SOURCES OPERATING BASIS FOR PROPOSED CH ANGES than tests. Therefore, our proposed markup of table 4.8-1 will count valid failures in terms of demands rather than valid tests. The criteria for determining the number of valid failures and demands would be in accordance with RG 1.9, Revision 3.

10. As a result of the deletion of SR 4.8.1.1.3, SR 4.8.1.2 of TS 3/4.8.1.2, A. C. Sources Shutdown, would be revised to delete the reference to SR 4.8.1.1.3.

13. The associated Bases would be revised to be consistent with the above changes.

Basis The proposed changes to actions statements a., b., c., and e. are changes that are recommended by Generic Letter 93-05. In turn, these recommendations are based on NUREG-1366,

" Improvements to Technical Specifications Surveillance Requirements," December 1992. In section 10.1 of NUREG-1366, the NRC concluded that DG surveillance testing could be improved by eliminating unnecessary testing that contributed to aging of the DGs. Specifically, when one DG is inoperable due to causes other than preplanned preventative maintenance, an j

inoperable support system, or an independently testable component, testing of the remaining DGs l

should be limited to those cases where the cause for inoperability has not been conclusively demonstrated to preclude the potential for a common mode failure. Therefore, testing of the remaining DGs occurs only when the cause for inoperability could affect the remaining DGs also.

(However, in those cases testing should occur within 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> rather than 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />.)

The next proposed change involves revising SR 4.8.1.1.2.a.4 to allow routine monthly testing of the DGs by gradually accelerating the DGs to operating speed as opposed to requiring the DG to attain rated voltage and frequency within 11.4 seconds. At VEGP, the DG governors are being replaced with a model that will accelerate the engines to operating speed within a range of approximately 20 to 40 seconds in the test mode. -When the DGs are in standby for. emergency operation, they will continue to fast start in accordance with their emergency function. This is consistent with the Improved Technical Specifications (ITS) and RG 1.9, Revision 3. Paragraph 2.2.1, " Start Test," of RG 1.9, Revision 3, states that for these tests the emergency diesel generator can be slow started and recch rated speed on a prescribed schedule that is selected to minimize stress and wear. NRC sponsored aging research work on nuclear service diesel generators, documented in NUREG/CR-5057, " Aging Mitigation and Improved Programs for Nuclear Service Diesel Generators," has identified fast sta'.1ing and fast loadhg tests of DGs as an aging stressor. Section 2.4.1 of this report states that a correctly managect monthly testing program involving slow starting and loading would induce little aging effects in the emergency DG. By contrast, a fast starting and loading test program can produce substantial harm and significant aging effects through the production oflarge mechanical and thermal stresses, El-3 m-

ENCLOSURE 1 (CONTINUED)

VOGTLE ELECTRIC GENERATING PLANT REQUEST TO REVISE TECIINICAL SPECIFICATIONS RELATED TO AC SOURCES OPERATING ,

BASIS FOR PROPOSED CIIANGES inadequate lubrication during initial acceleration high rotating and sliding pressures overspeeding, etc. The slow start is also consistent with recommendations of ASME/ ANSI Operations and Maintenence Standards and Guides Part 16. (Note that NUREG-1366 states a.

concern that this feature may be incorporated at the cost of reducing DG reliability by eliminating a redundant overspeed protection in the form of the backup mechanical governor. At VEGP, the backup mechanical governor will remain in service during the slow start.)

One additional aspect of the slow start mode of the new governors warrants mention here.

Presently, if the DG is started manually for testing and an actual undervoltage or loss of voltage - ,

occurs, the DG will automatically switch to the emergency mode of operation. When the new  !

governors are installed and the DG is started in the slow start mode, an emergency start signal will .

not override the slow start for the brief period that the governor is in the slow start mode (approximately 20 to 40 seconds). The effect would be that the availibility of the DG in test to supply emergency power would be delayed by approximately 8 to 28 seconds beyond the approximately 12 seconds that is presently allowed for the DG to start and begin loading. The- .

likelihood of a loss of offsite power occurring during a 40 second slow start is small (approximately 1.55X10-6 per unit)l, and only one DG will be in the test mode at a time, so the remaining DG will be available to supply emergency power. Therefore, the expected benefits due to redticed aging of the DGs by performing the slow start offsets the negative aspect of the potential delay in the diesel start time2 However, when one DG is inoperable and the remaining; operable DG must be tested pursuant to the action statement, the operable DG will be fast-started ,

so that it remains capable of responding to an emergency start signal.

9 The governors for the Unit,1 DGs will be replaced during the Fall 1994 refueling' outage scheduled to begin in September 1994. The governors for the Unit 2 DGs will be replaced during the Spring 1995 refueling outage. Assuming that these proposed changes are approved by -

September 1994, routine monthly testing on Unit 2 will consist of fast starts until such time as the governors can be replaced in 1995.

In addition, SR 4.8.1.1.2.g.2 would be revised to allow the DG to be loaded gradually rather than.-

requiring the DG to be loaded to an indicated value of 6100-7000 kw in less than or equal to 60

, . seconds, This is consistent with the recommendations of GL 93-05 which recommends this' change for the routine _ monthly surveillance. The routine monthly surveillance at VEGP presently l does not require fast loading, so we are proposing to extend this concept to the six-month .

1 The VEGP IPE assumes a frequency for a loss of offsite power of.051 events per year. Assuming two DGs tested ~ ,

12 times per year using a 40 second slow start yields a value of approximately 1.55X104 2

NSAC 130, "The EITect of Diesel Start Time Delay on Westinghouse PWRs," cvalua'tes the effect on safety margins of relaxing the diesel start time requirements for Westinghouse PWRs. That study concluded that a diesel start time delay of approximately 30 seconds was feasible with currently acceptable licensing calculations.

El-4 2

w, m ,e--- 's- -----l---c.---N-----.._'.

ENCLOSURE I (CONTINUED)

VOGTLE ELECTRIC GENERATING PLANT ,

REQUEST TO REVISE TECIINICAL SPECIFICATIONS I RELATED TO AC SOURCES OPERATING BASIS FOR PROPOSED CIIANGES surveillance. In fact, NUREG-1366 states that fast loading during surveillance testing is the most significant cause of accelerated degradation of diesel generators. In NUREG-1366, the NRC staff recommends that all testing of the DGs, with the exception of the loss of offsite power tests that are performed with and without an ESF signal once each refueling, be performed by gradual loading in accordance with the manufacturer's recommendations. This is also consistent with the Improved Technical Specifications, Revision 0, September 28,1992.

The next two paragraphs discuss proposed changes based on the recommendations of RG 1.9, Revision 3.

. The 24-hour endurance run would be separated from the hot restart test. The existing VEGP SR 4.8.1.1.2.h.7 requires a hot restart test be performed within 5 minutes aller completing the 24-hour endurance run. According to NUREG-1366, further improvement in DG surveillance requirements could be realized by separating these two tests, therby giving the outage planners added flexibility in scheduling these tests and preventing their becoming the source of critical path complications. The purpose of the hot restart test is to ensure that the DG does not in any way have impaired performance following operation at full load or equilibrium temperature. In order to ensure that the DG is at full load or equilibrium temperature for the hot restart test, the new SR would require that the DG be operated for a minimum of 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> at a load of 6800-7000 kw. Within 5 minutes of shutdown, the DG would be restarted and required to attain rated voltage and frequency within 11.4 seconds. The proposed new SR would continue to ensure the hot restart capability of the DGs.

. The next proposed change would delete the DG failure reporting requirements of SR 4.8.1.1.3. In addition, SR 4.8.1.2 of TS 3/4.8.1.2, A. C. Sources Shutdown, references SR 4.8.1.1.3. This reference to SR 4.8.1.1.3 would be deleted from SR 4.8.1.2. With respect to ,

reporting criteria, Regulatory Guide 1.9, Revision 3, states that licensees must conform to the requirements of 10 CFR 21,50.72, and 50.73. These requirements must be complied with regardless of TS requirements. This change is administrative in nature since the reporting of DG failures has no direct bearing on the ability of the DGs to perform their safety function.

Furthermore, the requirements of 10 CFR 21, 50.72, and 50.73 should capture those events and conditions that are of the most interest to the NRC. In the case of VEGP, the guidance of

~

RG 1.108 resulted in special reports of questionable safety significance. For example, GPC has issued special reports resulting from start attempts following major refueling outage maintenance activities. These start attempts were for the purpose of adjusting the governor or priming the fuel lines, and by a strict interpretation of the guidance of RG 1.108 they were l reportable as invalid failures. However, these " invalid failures" were of no safety significance l since they were a normal part of the post maintenance activities and were not failures in any  ;

El-5

ENCLOSURE I (CONTINUED)

VOGTLE ELECTRIC GENERATING PLANT l REQUEST TO REVISE TECIINICAL SPECIFICATIONS RELATED TO AC SOURCES OPERATING i

BASIS FOR PROPOSED CIIANGES sense of the word. This is also consistent with the Improved Technical Specifications, Revision 0, dated September 28,1992. H l

Finally, the next proposed change would revise table 4.8-1, Diesel Generator Test Schedule, to  !

incorporate the test schedule of section D.2.4.4, Appendix D to NUMARC 87-00, Revision 1. j The existing TS table 4.8-1 specifies DG test frequency based on the number of valid failures in - .l the last 20 valid tests and the last 100 valid tests. If the number of failures in the last 20 valid tests  ;

is less than or equal to 1 and the number of failures in the last 100 valid tests is less than or equal  !

to 4, the test frequency is once per 31 days. If the number of failures exceeds either of the above - I criteria, the test frequency is increased to once per 7 days and continues at that frequency until both of the above criteria are satisfied. Under the proposed new schedule, testing pursuant to SR 4.8. L 1.2.a would be conducted monthly provided the number of valid failures in the last 25 j demands for a given DG is less than or equal to 3. If the number of valid failures is 4 or more,  !

testing would be conducted at least once per 7 days (but at intervals of no less than 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />) until seven consecutive failure free starts from standby conditions and load-run demands have been performed. Note that both NUMARC 87-00, Revision 1, and RG 1.9, Revision 3, use and define the terms " start demand, stan failure, load-run demand, and load-run failure" rather than  ;

the old RG 1.108 terminology of valid tests. In fact, section D.2.4.4, Appendix D to NUMARC 87-00 refers to the last 25 " demands" rather than tests. Therefore, our proposed markup of table  ;

4.8-1 will count valid failures in terms of demands rather than valid tests. The criteria for determining the number of valid failures and demands would be in accordance with RG 1.9, .1 Revision 3.

The recommendations of NUMARC 87-00, Revision 1 have. been recognized by the NRC in i Regulatory Guide 1.155, " Station Blackout," as providing guidance that is in large part identical to RG 1.155 guidance and is acceptable to the NRC staff for meeting the requirements of 10 CFR 50.63 regarding station blackout. With respect to a DG reliability program, VEGP committed by letter dated March 26,1992, (ELV-03551) to develop a procedure that would implement the guidelines of Appendix D to NUMARC 87-00, Revision 1. This has been accomplished to the - l extent possible considering that our existing licensing basis commitments and TS requirements are - l based on RG 1.108. Upon approval of the proposed change, VEGP can fully implement the emergency DG reliability program of Appendix D to NUMARC 87-00, Revision 1, including the corrective actions in response to exceeding the trigger values. At that point VEGP will be in a position to monitor, maintain and improve DG reliability in accordance with the recommendations -

of NUMARC 87-00, Revisien 1.

El-6

ENCLOSURE 2 VOGTLE ELECTRIC GENERATING PLANT  !

REQUEST TO REVISE TECIINICAL SPECIFICATIONS l RELATED TO AC SOURCES OPERATING 10 CFR 50.92 EVALUATION l l

Georgia Power Company has reviewed the requirements of 10 CFR 50.92 as they relate to the proposed changes and has made the following determination: i

1. The proposed changes do not involve a significant increase in the probability or consequences 1 of an accident previously evaluated. The proposed changes affect the required actions in response to inoperable offsite and onsite ac sources, sutveillance requirements for the j emergency diesel generators, and reporting requirements for diesel generator failures. The proposed changes are based on the recommendations of Regulatory Guide 1.9, Revision 3, NUMARC 87-00, Revision 1, and Generic Letter 93-05. They are expected to result in improvements in diesel generator testing and failure reporting and reduce diesel generator aging due to excessive testing. As such, the proposed changes should result in improved diesel generator reliability, thereby providing additional assurance that the diesel generators will be capable of performing their safety function. Therefore, the proposed changes will not significantly increase the probability or consequences of any accident previously evaluated.

2. The proposed changes do not create the possibility of a new or different kind of accident from any accident previously evaluated. The proposed changes afTect the action and surveillance requirements for the onsite and offsite ac sources. Accordingly, the proposed changes do not involve any change to the configuration or method of operation of any plant equipment, and no new failure modes have been dermed for any plant system or component nor has any new limiting failure been identified as a result of the proposed changes. Therefore, the proposed changes do not create the possibility of a new or different kind of accident from any previously evaluated.

3. The proposed chh..Jes do not involve a significant reduction in a margin of safety. The proposed changes are based on existing regulatory guidance. Under the proposed changes, ,

the emergency diesel generators will remain capable of performing their safety function, and  ;

the effects of aging on the diesel generators will be reduced by eliminating unnecessary testing. Therefore, the proposed changes do not involve a significant reduction in a margin of safety.

I Conclusion Based on the preceding analysis, Georgia Power Company has determined that the proposed '  ;

changes to the TS will not significantly increase the probability or consequences of an accident  ;

previously evaluated, create the possibility of a new or difTerent kind of accident from any H accident previously evaluated, or involve a significant reduction in a margin of safety. Therefore, the proposed changes meet the requirements of 10 CFR 50.92 (c) and do not involve a significant hazards consideration.

-i E2-1 j