ML20236D612

Advises That Util 881223 Response to Generic Ltr 88-17, Loss of DHR Appears to Meet Ltr Requirements for Expeditious Actions.Several Items to Be Considered to Assure That Actions Adequately Addressed Listed

ML20236D612
Person / Time
Site:	Waterford
Issue date:	03/08/1989
From:	Calvo J Office of Nuclear Reactor Regulation
To:	Dewease J LOUISIANA POWER & LIGHT CO.
References
GL-88-17, TAC-69791, NUDOCS 8903230187
Download: ML20236D612 (8)

Text

_ _ _ _ _ - - - - - ______ __ _ _ - _ _ - _ _ - _ _ _ - _ _ _ __ _ _ -

n l' # o osag'o g UNITED STATES 8 n . NUCLEAR REGULATORY COMMISSION l

7; E .

WASHINGTON, D. C. 20555

/ March 8, 1989 Docket No. 50-382 Mr. J. G. Dewease Senior Vice President - Nuclear Operations l Louisiana Power and Light l 317 Baronne Street, Mail Unit 17

New Orleans, Louisiana 70160

SUBJECT:

COMMENTS ON THE LOUISIANA POWER AND LIGHT RESPONSE TO GENERIC LETTER 88-17 FOR THE WATERFORD UNIT 3 FOR EXPEDITIOUS ACTIONS FOR LOSS OF DECAY HEAT REMOVAL (TAC NO. 69791)

The NRC staff has reviewed the Louisiana Power and Light (LP&L) letter of December 23, 1988 to response to Generic Letter (GL) 88-17. We find that LP&L appears to meet the intent of the generic letter with respect to expeditious actions.

The LP&L response includes some indication on scheduling, however, all expeditions actions addressed by GL 88-17 are to be implemented in advance of any requirement for part loop operation. We note that the next anticipated requirement for part loop operation at Waterford 3 should not occur before the refueling outage in October 1989.

Your overall response is generally conplete and more detailed than the average response we have reviewed. However, in a few areas, your response is sufficiently vague that we cannot fully understand your actions taken in response to GL 88-17. You may wish to consider several observations in order to assure yourselves that the actions are adequately addressed:

1. You mention discussion of the Diablo Canyon event with operations personnel and training for specific mid-loop operation and cooldown/draindown with your staff. It is not specifically stated that maintenance personnel are also included. The item was intended to include all personnel who can affect reduced inventory operation and preventive as well as mitigative training.

2. Your training program includes a review of the previous Waterford mid-loop events of 7/86 and 5/88. There are also a number of other mid-loop events that occurred at other plants that would be beneficial for review such as Arkansas Nuclear One, Unit 1 (October 26,1988),

Surry Unit 2 (September 19,1988), and Sequoyah (May 23,1988).

h e

a4o32g$

8 di 1

P

~.

Mr. J. G. Dewease l l

3. You have identified penetrations of concern in Section 3.1 as those "providing direct access from the containment atmosphere to the outside atmosphere...." However, we are concerned with all containment penetrations I that could cause a release (e.g., penetrations from the containment into a I fuel handling or auxiliary building). l

4. You have not addressed problems associated with remaining in containment j once boiling initiates with large openings in the RCS pressure boundary. l Loss of shutdown cooling (SDC) could lead to boiling in 30 minutes and core uncovery in 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br />. ~ All personnel may have to. leave containment or don special gear 30 minutes after the loss because of steam inside of containment. If you are assuming 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br /> in which to work inside containment, then this may require special equipment.

5. In your addressing of containment closure, no information is provided regarding how you will keep track of and control the many potential j openings which rey have tu be closed simultaneously. Your procedures and administrative controls should address this topic.

6. In Section 3.1.1 you have stated that'"if the RCS is open on the cold leg side due to a major disassembly or removal of an RCP, the pressure I increase in the upper plenum (hot side) will depress the water level in (

the core and the steam generator outlet pipe. This will force water out i of the RCS through the RCP opening." You further state that " vent paths '

are available but not credited to equalize pressure between the hot and cold sides of the RCS - e.g., leakage paths around the het leg nozzles and other in-vessel leakage paths." Although these paths exist, they provide minimal pressure equalization.

7. In Section 3.1.1 you state that if the RCS is closed and a steam generator  !

(SG), initially at 70% wide range level for secondary side inventory, is  !

available it should be able to cool for more than 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> before boiling away the secondary side inventory (assuming no credit for makeup). You have not identified a requirement that SGs be maintained filled during mid- J loop nor is provision for SG steaming addressed. You do state that one SG .j will be available if SG maintenance is not being performed.. The dynamics j of steam boiling in the SG and the effects on the RCS temperature, pressure, ,

and level are not addressed although some assumption is made that the RCS pressure shall not exceed 35 psig.

The methods to be used for low pressure secondary side steaming to maintain RCS pressure less than 35 psig have not been discussed. Analyses and '

procedures should be in place which justify any methods you choose.

(

i I

1

+

=,

Mr. J. G. Dewease -

8. In Section 3.1.1 you also stated that steam exiting the SG tubes.would be vented thrcugh the loop seal and condensed by water in the vertical' suction leg between the RCP and loop seal. We assume that you have l performed analyses which demonstrates this condensation process as well '

as determining the amount of water which would remain under various allowed system configurations.

9. In Section 3.3.1 you state that "The critical path to closing containment is closure of the equipment hatch." .You say that " Conservatively 2 ,

hours is needed to adequately close the hatch (closure can actually be l effected in 1-1.5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br />)." This is based on Figure 2 which shows the time after initiation of a loss of SDC by which containment' closure activities must begin, as a function of time after reactor shutdown.. Has the possibility of large quantities of steam in the containnient, from boiling l

~

in the reactor with large openings in the RCS pressure boundary, been i factored into the time to close the equipment hatch? See also number 4 l above. '

10. In Section 5.1.1 you discuss the features of the refueling water level indicating system (RWLIS) for which a diagram is given in Figure 3. The l RWLIS system uses stainless steel piping and has narrow and wide range i pressure transinitters. As shown in Figure 3 these pressure transmitters l show a comion top on the drain line below hot leg #1 in the vicinity of l the shutdown cooling suction line. Also, the pressure transmitters share  !

the common reference high point tap near the top of the pressurizer.

Therefore, if a comparison is made for a region where the wide and narrow range readings overlap, precautions must be taken in justifying the accuracy because of the dependence on common taps.

11. In Section 5.1.1, Figure 3, the RWLIS system shows downward slopes for the lower top horizontal runs but no slope is indicated for the upper top i

horizontal runs. It may not be practical for slopes.in the upper horizontal runs. However, there should be means addressed for insuring that there is no water in these upper reference legs. Also, Figure 3 a' shows closed valves RC-215A and RC-215 leading to the boron management

, system to which the lower reference leg for the RWLIS is connected to.

What would be the effect of opening these valves on the accuracy of the RWLIS readings?

12. In Section 5.1.2 the refueling level indication system (RLIS) is described as consisting of one inch rubber (ortac) tubing including a length of hardened tygon tube as a sight glass. The schematic is given in Figure 4. You state that the tubing runs from the bottom of hot leg

1. 1 (from the same tap-off as the.RWLIS), through the sight glass, and into the top of the pressurizer. No mention is made of cross checks between the RLIS and RWLIS level systems. This would be desirable. However, since both have the same bottom tap, care would be needed to avoid a common error. You do mention the ability to check the RLIS level measurement where it overlaps a pressurizer level indication prior to draining below the pressurizer. There is no indication of a potential level problem if the pressurizer surge line is not empty and pressurizer

r , ,

.Mr. J. G. Dewease -

air pressure is not equilibrated with the remainder of the RCS. Also, you later indicate in Section 5.1.3 a check against the HJTC system at several discrete elevations which is desirable as a cross check.

13. Figure 4 of Section 5.1.2 shows clear hose below the level of the hot

. leg. You indicate that there is a wide range of measurement from the bottom of the hot leg to'an elevation overlapping the bottom of the pressurizer. It is not clear what you consider the actual lower end of the range. A value below the bottom of. hot leg would be misleading.

14. In Section 5.1.2 you mention that the rubber hose for RLIS is rated for a l working pressure of 300 psig at.180*F and the. sight glass is rated for a working pressure of 15 psig at 180'F. If the system were to reach 1 boiling, it would appear that these. properties would not be sufficient. 1

15. Walking the rubber /tygon tube following installation to verify lack of kinks or loop seals is necessary.. Experience shows that periodic 'l walkdowns are needed after installation. We recontend daily walkdowns )

when the rubber /tygon tube is in use, with an additional walkdown immediately prior to its being placed in use. -Your preventive training for all affected personnel should address this, as well as procedures, to avoid the reoccurrence of the May 1988 events.

16. The HJTC method for level measurement is described.in Section 5.1.3 with a schematic showing the eight discrete elevations for a probe in Figure 1

5. You state that two independent and redundant probe assemblies are installed. Three of the elevations are located at heights approximately equivalent to the bottom, middle and top of the hot leg. This system is only effective for reduced inventory _ conditions'when the reactor vessel head is in place and the HJTC instrumentation is connected. You state.

that the HJTC system can be utilized for cross checks with the RWLIS and RLIS systems within the resolution capability of the HJTC system and that the sensor located at the hot leg centerline provides some indication of potential LPSI pump cavitation. Also, you indicate that the HJTC can be used as a backup in the event of the loss of the other level measurement l

capability. The HJTC should never be used as the primary level detection i system. The HJTC uses discrete points which are too widely separated to be useful for lowered inventory operation in any capacity other than for cross checking of other instruments.

17. In Section 5.2 you discuss the November 1988 draindown in which you state that the RLIS and HJTC systems-were designated as the primary level measurement instruments. The RWLIS system was available but undergoing further acceptance tests. The operators were instructed to predicate any l actions based on the lowest level indicated by the RWLIS/RLIS/HJTC embination which is a good conservative approach. This instruction H ould be part of the procedures and/or-training.

m i

Mr. J. G. Dewease

18. In Section 5.3 you state that you are evaluating the degree of independence of the RWLIS and RLIS systems sir ce they have a common tap-off from the RCS hot-leg. We will accept a common tap for instrument systems which have alreaoy been installed. You also state that you are making an evaluation to de:termine the extent to which the RLIS system, a rubber.

hose /tygen tube sight glass combination, can be crdited in the long term.

Rubber hose /tygon tube will not be acceptable for the long tenn instru-mentation needs. For the long term enhancements and because you have common taps for the RLIS and RWLIS, provision should be made for testing the taps before each use to assure they are open and free of restrictions.

19. In Section 7.3 'ycu state that when in a reduced inventory condition that 2 of the 3 HPSI pumps will bE operable and available for hot and Cold leg injection in addition to the SDC (LPSI) pumps. Also you state that "While the SDC pump may be unavailable for SDC purposes (e.g., pump cavitation, inadvertent SDC valve isolation, etc.) the pump will.still function to restore RCS level through an alternate valve lineup and appropriate sthrtup procedure." You indicate that OP-901-046 will be amended to credit a LPSI (SDC) pump as a second alternate means of RCS addition..

This is not responsive to GL 88-17 ff this action is taken for alternate lineup after one of the SDC pumps fails since a failure due to vortexing of one LPSI pump could easily affect the other LPSI pump. However, if the second LPSI pump is aligned up as a second available means before entering a reduced inventory condition it would be acceptable. Alternately it  !

would be desirable to have other sources of'RCS water make-up pumps lined i up in preparation for SDC purposes if effective. .  ;

i

20. In Section 8.2.1 (page 24) related to nozzle dams you stcte that "if a l steam generator was not blocked or did not have a manway onn it was assumed to be available as a heat sink for steam condensation." There is no follow-up in your discussion to indicate that there would be provisions for making sure water was in the SG and that steam could be relieved on the secondary side (see Item 7). A steaming path is mentioned later on page 25 but the means are not previoed.

21. In Section 8.2.1 you mention a review in which a pressurizer manway (a i relief area of 1.40 ft2) is open. We note that relatively large hot side I openings in the RCS, such as a presurizer manway, can still lead to a I pressure of several psi. The large steam flow rate in combination with I flow restrictions in the surge line and lower pressurizer hardware may j lead to pressurization. A: curate calculations should be performed to l verify the effectiveness of the opening.

22. It is noted that in many of your responses you have a section called "Long-Range Expeditious Actions." The Generic Letter 88-17 does not have such a category in expeditious actions. .As we not9d above, all expeditions actions should be impleniented in advance of arty requirement for part loop operation.

Mr. J. G. Dewease- 23. You appear to be attempting to work within existing technical specifications in meeting the. generic letter recommendations. We note that technical specification changes will be considered if existing specifications are f overly restrictive. I There is no need to respond to the NRC on the above observations at this-time.

As you are aware, the expeditious actions you have described are an interim  !

measure to achieve an imediate reduction in risk associated with reduced .

inventory operation, and these will be' supplemented and in some cases replaced -l by programmed enhancements. We intend to audit both your resp')nse to the i expeditious actions and your programmed enhancement program. The areas where we do not fully understand your responses as indicated above may be covered in the audit of expeditious actions.

If there are any questions on the above observations or the intent of GL 88-17 and expected actions, please let us know.  ;

Sincerely, ORIGINAL SIGilED BY JOSE A. CALVO Jose A. Calvo, Director i Project Directorate - IV Division of Reactor Projects - III, IV, V and Special Projects Office of Nuclear Reactor Regulation cc: See next page DIST'tIBUTION DE ket File NRT. PDR Local PDR PD4 Reading G. Holahan ,

L. Rubenstein l J. Calvo l P. Noonan  ;

D. Wigginton J OGC-Rockville i E. Jordan i B. Grimes H. Balkujian, SRXB W. Hod , SRXB ACRS (

PD4 Plant File  !

{

MEMO CALV,0 WATERFORD GL8817 PD4/L/$ Q PD4/PM SRXB PD4/D PNo Wi DWi inton:bj WHod es. JCalvo  !

03/ /89 03 /89 03/ /89 03/8/89

4 Mr. J. G. Dewease . i

23. You appear to be attemrtfvi to work within existing technical specifications l in meeting the generic 1+b,er recommendations. We note that technical i specification changes will be considered if existing specifications are overly restrictive.

There is no need to respond to the NRC on the above observations at this tirne.

As you are aware, the expeditious actions you have described are an interim measure to achieve an imediate reduction in risk associated with reduced i inventory operation, and these will be supplemented and in some cases replaced i by programmed enhancements. We intend to aucit both your response tc the i expeditious actions and your programed enhancement program. The areas where we do not fully understand your responses as indicated above may be covered in l the audit of expeditious actions.

, If there are any questions on the above observations or the intent of GL 88-17 l and expected actions, please let us know.

Sincerely, l 4 6- laka

'f Jose A. Calvo, Director Project Directorate - IV i

i Division of Reactor Projects - III, I IV, V and Special Projects Office of Nuclear Reactor Regulation cc: See next page

___ _m___ _ . _ _ . _ _ _ _ - _ _ _ _ . _ . _ _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ - _

c . , ,

Mr. Jerrold G. Deweas~e Waterford 3 Louisiana Power & Light Company cc:

W. Malcolm Stevenson, Esq. Regional Administrator, Region IV Monroe & Leman U.S. Nuclear Regulatory Comission  !

1432 Whitney Building Office of Executive Director'for New Orleans, Louisiana 70103 Operations 611 Ryan Plaza Drive, Suite 1000 Mr. c. Blake Arlington, Texas 76011 e She.., Pittman, Potts & Trowbridge l 2300 N Street, NW Mr. William H. Spell, Administrator, Washington, D.C. 20037 Nuclear Energy Division Office of Environmental Affairs Resident Inspector /Waterford NPS Post Office Box 14690 Post Office Box 822 Baton Rouge, Louisiana 70898 Killcna, Louisiana 70066 .j Mr. Ralph T. Lally President, Police Jury l Manager of Quality Assurance St. Charles Parish Middle South Services, Inc. Hahnville, Louisiana 70057 Post Office Box 61000 j New Orleans, Louisiana 70161 l Chairman William A. Cross Lcuisiana Public Service Comission Bethesda Licensing Office One American Place, Suite 1630 3 Metro Center Baton Rouge, Lcuisiana 70825-1697 Suite 610 Bethesda, Maryland 20814  :

Mr. R. F. Burski j Nuclear Safety and Regulatory Affairs Manager Louisiana Power & Light Company 317 Baronne Street ..

New Orleans, Louisiana 7011c j l

i i

l i

1

.