ML20210C317

From kanterella
Jump to navigation Jump to search
Provides Results of Wind Vulnerability Study for Diesel Generator & Control Bldgs.Possibility Exists That Bldgs & Condensate Storage May Be Damaged by Missiles Created by Failure of Nearby Structures
ML20210C317
Person / Time
Site: 05000000, Indian Point
Issue date: 03/03/1986
From: Stello V
NRC OFFICE OF THE EXECUTIVE DIRECTOR FOR OPERATIONS (EDO)
To:
Shared Package
ML082401831 List:
References
FOIA-86-126, FOIA-86-127, FOIA-86-131, FOIA-86-80, TASK-PII, TASK-SE CLI-85-06, CLI-85-6, SECY-86-073, SECY-86-73, NUDOCS 8603240231
Download: ML20210C317 (12)


Text

pa arco E

n 5

I

%,...../

POLICY ISSUE March 3, 1986 SECY-86-73 (Informat. ion)

MEMORANDUM FOR:

The Commissioners FROM:

Victor Stello, Jr.

Acting Executive Director for Operations

SUBJECT:

RESPONSE TO COMMISSION'S INDIAN POINT SPECIAL PROCEEDING DECISION (CLI-85-06) (WIND VULNERABILITY STUDY)

PURPOSE:

To provide the Commission with the results of the wind vulnerability study for the Indian Point Unit No. 2 Diesel Generator and Control Buildings and the possibility that these buildings and the condensate storage tank could be damaged by missiles created by failure of nearby structures.

DISCUSSION:

The Commission, in its Decision dated May 7, 1985, directed the NRC staff to perform a study of the vulnerability of the Indian Point Unit No. 2 Diesel Generator and Control Buildings and the possibility that these buildings could be damaged by missiles created by failure of nearby structures.

l The staff's evaluation, details of which are included l

in Enclosure 1, concentrated on the review of Amend-l ment 2 to the Indian Point Probabilistic Safety Study (IPPSS) which was submitted by Consolidated Edison after the conclusion of the hearing testimony and was intended to address the questions raised during Sandia's review of the IPPSS.

The minimum structural capacity in terms of wind velocity is identified as 123 mph for the diesel generator building and 103 mph for the control building.

The probability of winds of this magnitude af Indian Point is estimated to be in the range of '10 / year.

The licensee's analysis also provides a reestimate of the frequencies of the sp..ctrum of wind velocities i

I associated with a tornado taking into consideration wind channelization and site topography. The inter-action of other structural failures with structures CONTACT:

M. Slosson 492-7090 t#x;Ica:D ORIGINAL

/

40231 860303 04puflad By b

h 4pF5 SECY

()

.86-073 V

c

. housing critical safety functions was also included in.the accident sequence. The net result is ~an estimated core-melt frequency of 3.6 X 10

  • for a tornado induced sequence..This value is within the range of other Indian Point ? accident sequences identified during the hearing process.

In addition, the staff requested Consolidated Edison to report on alternate capabilities in the event of the loss of the Diesel Generator Building,_the Control Building and the Condensate Storage Tank due to high winds and/or missiles created by failure of nearby structures. Consolidated Edison's report is included in Enclosure 2.. The licensee indicates that Indian Point 2 has extensive alternate power supplies, remote shutdown capability and alternate secondary heat sink water supply.

CONCLUSIONS:

Indian. Point 2 has certain safety-related buildings which are vulnerable to tornado level winds. However, the core-melt frequency associated with tornados is well within the range of frequencies of other Indian Point 2 accident sequences identified during the hearing process.

It should be noted that the core-melt frequency estimation gives no credit for the alternate shutdown provisions at the plant, nor does it credit possible restoration of damaged equipment before a point of no return. However, even in the unlikely event of complete failure, there would be i

approximately two hours available to restore core cooling before core melt and another eleven hours after melt to avert late overpressure failure of containment. As a result, the staff believes that the core-melt frequency estimate cited above is quite conservative. Thus the staff recomends no corrective actions.

[

Victor Stello, Jr.

Acting Executive Director for Operations

Enclosures:

As stated DISTRIBUTION:

b Commissioners OPA l

OGC REGION I OPE EDO OI ASLBP OCA ASLAP OIA SECY b

INDIAN POINT ? - CLI-85-06 STAFF RESPONSE The staff was recuested by memorandum from Samuel J. Chilk to William J. Dircks to' address concerns identified in CLI-85-06 pages 26, 34, ani'59.

The staff was directed to investigate the vulnerability of the s

Indian Point Unit 2 diesel generator and control buildings to high winds and to the possibility that these structures could be damaged by missiles created by failure of nearby structures. Also, the staff was to investigate the possibility of either the turbine building or superheater building,.or parts from these buildings, failing and falling on the control building, and the possibility of the superheater building failing and falling on the diesel generator building and the condensate storage tank:

The licensee performed a probabilistic assessment of tornado initiated events as part of the initial Indian Point Probabilistic Safety Study (IPPSS) and estimated a core melt frequency of 1.6x10-5/RY. The relative importance of tornado induced accident sequences was presented in a Comission briefing on September 5,1984, as shown in Figure 1.

This estimate of core' melt frequency was adopted in the staff's testimony based on the Sandia National Laboratory review reported in NilREG/CR-2934.

In the initial PRA study, the dominant contributors to this sequence were the loss of offsite power lines coupled with the direct failure of the control building due to. wind induced loads.

Subsequent to testifying at the hearing, the licensee submitted Amendment 2 to IPPSS which contained a reanalysis of the probabilistic assessment of wind and tornado accident sequences at Indian Point, Unit 2.

Amendment 2 provided a reestimate of the frequencies of the spectrum of wind velocities associated with a tornado, taking into consideration wind channelization and the topography of the site. A reexamination of critical structures was also -

provided. The structural reevaluation considered the effects of a more

. causes total loss of control and/or power at the outset, there is a substantial period for restoration of key safety functions before a containment failure would occur.

It is noted that there is a technical specification requiring protective actions in the event of approaching hurricanes.

The Sandia study indicated that the Hurricane sequence without the technical specification actions had a-probability of occurrtace one order of magnitude higher (5.4x10-4) than the toronado sequence. Hurricanes have potential flood-related hazards in addition to wind-induced problems.

The higher likelihood of hurricanes estimated by Sandia, taken in conjunction with readily available credictive capabilities for. hurricanes, suggest that protective steps can and should be taken for tnese hazards. No such predictive capability exists with regard to tornados. Thus, taking into account the lower probability of tornado occurrence, lack of reasonable predictive capability, and the fact that tornado initiated accidents are within the recurrence intervals of a number of other accident sequences at this level, no further corrective actions for tornado hazards are believed to be warranted.

SUMMARY

The staffs review indicates that structures including the control, diesel generator, turbine, and super heater building, and the condensate storage tank would be vulnerable to the wind speeds and missiles postulated for the tornado event.

However, the likelihood of a severe accident resulting from a tornado is well within the range, of frequencies of other accidents considered.

Further, alternative power supplies may be available, and at least 11 hours1.273148e-4 days <br />0.00306 hours <br />1.818783e-5 weeks <br />4.1855e-6 months <br /> are available to actuate other sources of containment cooling to further l

reduce the likelihood of containment failure from overpressurization.

Finally, no capability is available to provide early predictions and/or l

warning for this rare tornado event, and the plant meets all regulatory j

requirements. Thus, the staff recomends no further corrective actions with regard to tornado vulnerability.

l l

l l

t

FIGURE 2 LOWEST WIND VELOCITY CAPACITIES OF IP BilILDINGS Min. Wind Velocity Structure Capacity in mph Failure Mode / Wind Direction Diesel Generator B1dg.

123 Siding / East Control Bldg.

103 Roof Suction / North Turbine Bldg.

94

. Fracture of Glass Windows (north wall)/ West Superheater Bldg.

200 Anchor bolts / North f

I 6

l l

i I

L

Jonn D. O'Tstl)

, e : st ta*t s

o

  • Cceso:ca:e E son ccrreaay c' New ve < : :

8

- 4 Irvn; 2 a:e. New Yora. NY ?0003 i

' Te eoat e '212e 450-2533 February 18, 1986 Re:

Indian Point Unit No. 2 Docket No. 50-247 Mr. Steven A. Varga, Director

- PWR Project Directorate No. 3 Division of PWR Licensing - A U. S. Nuclear Regulatory Commission Washington, D. C. 20555

Dear Mr. Varga:

The ' Attachment to this letter confirms the information provided to you during telephor discussions with members of my staff on January 31, February 3, and February 13, 1986 and is also in response to your letter of February 6, 1986.

Although your February 6 letter was quite broadly phrased, based upon the telephone discussions, we understand that you are requesting our analysis of alternative measures and systems, which could be called upon in the event of the unavailability of the. Diesel Generator Building, the Control Building ard the Condensate Storage Tank due to damage caused by high winds.

As noted in the Attachment our current assessment of high wind risks for Indian Point Unit 2 (IP-2) is contained in Amendment 2 of the IPPSS submitted to NRC in April, 1984.

If you or your staff have any further questions do not hesitate to call us.

Ve truly yours,

/

fp

'h

^

cc:

Senior Resident Inspector U. S. Nuclear Regulatory Commission P. O. Box 38 Buchanan, New York 10511

~.

Background

g The - initial wind effects analysis published in the Indian Point Probabilistic Safety Study (IPPSS) as submiited to NRC in March,1982, based 'on a preliminary analysis of wind

hazards, structural was fragilities, and failure scenarios.

In'1983, soon after the IPPSS was published and during the Indian Point Hearings initiated by the NRC in

- response toL a petition by the Union of Concerned Scientists.(UCS), we.

commenced a revised analysis refining the earlier work, giving greater-attention to site specific wind _ hazard data, to wind fragilities of key and' adjacent structures, and the consequences of their potential failure.. In addition to Pickard, Lowe and Garrick, Inc., who were contracted for the IPPSS, _ and Research Triangle Institute (RTI), who had prepared ~ the initial wind and tornado. hazard analysis, the revised

- analysis used Structural Mechanics Associates, Inc. to provide a more detailed structure fragility analysis; Cermak/Peterka 4 Associates to e

analyze adjacent building and ground effects on building fragilities, and York Research Corporation for support to RT1 with local and regional wind information and effects.

The initial analysis determined that any potential increases in windspeed the Indian Point site due to wind channelization through the Hudson at River Valley would be more than offset by the conservative assessment of i

terrain roughness attenuation of offshore winds.

Peak winds were also assumed in the nitial s:uay to occur from any direction.

The revised '

study used site-specific wind data and accounted for wind channelization, hurricane wind attenuation with distance inland, terrain roughness in the vicinity of the site, and differences in wind speed in each of the four principal wind directions.

The resultant wind hazard curves for each of the four directions were significantly lower than those presented earlier.

The revised tornado / wind analyses were documented as part of Amendment 2 4

to the-IPPSS submitted to NRC in April, 1984.

Thus, while that amendment addressed NRC Staff /SiGLIA comments on the original IPPSS wind analysis, its submittal came after the closure of the Indian Point Hearing record.

.Therefore, the hearings record and decision do-not reflect the significantly lower IP2 high wind risk which resulted from the analyses contained in Amendment 2 of IPSSS.

At IP2, the control building, diesel generator building and Condensate Storage Tank referenced in the NkC's letter of Febniary 6,1986 are not specifically designed to withstand tornados.

Their fragilities and frequencies of unavailability were, however, determined in connection with the IPPSS Amendment 2 efforts referred to above.

Tornados are rare events in the vicinity of Indian Point. Due to the specific locations of these structures it is even more improbable that a single tornado could cause their simultaneous failure and at the same time cause failure of backup equipment on-site and off-site.

Tornado damage to safety related equipment within the control building and diesel generator building was not specifically analyzed in the IPPSS; rather, subsequent failure of all 1

equipment within those structures and their assoicated functions was conservatively assumed.

. _. _ _ _ _ _ _. ~.,,.. _

(.

Case 1:-

The scenario described above for-case 1,

a-loss of A.C.

station power, has been analyzed and symptom-based recovery

-procedures exist for the operators to shutdown IP-2 from the central control room.

For this case, the reactor would be -

tripped, and natural circulation would be established.

Decay heat would be removed by the steam turbine driven auxiliary feedwater pump and atmospheric steam relief.

The reactor coolant pump seals would. be cooled by either seal injection from a charging pump or. component cooling to the thermal barriers.

It should be noted that recent seal testing l

without cooling performed by the Westinghouse Owners Group t

has shown seal degradation. to be significantly less than that assumed in the IPPSS.

Thus, restoring seal cooling, which requires A.C.

power, is much less of concern and would thereby allow more time to restore A.C. station power.

If a natural' circulation cooldown was commenced,f reactor. coolant system makeup and boration would be provided by a charging pump and the refueling water storage tank.

1 i

480V A.C. power to the charging pump and/or component cooling pump (and service water pump) would be provided by one of several 13.8 KV or 6.9 KV power feeds to the station as shown in FSAR Figure 8.2-1.

These are described below:

1)

Gas Turbine No.1 (GT-1) is located on-site at elevation 15'.

This unit could be started and loaded in about 15 minutes, providing 13.8/6.9 KV power directly to IP-2.

Alternately, GT-1 can provide 13.8 KV power to Indian Point Unit 1 (IP-1) Light and Power (LGP) bus sections which could be transferred to IP-2 6.9 KV buses and 480 V safety buses or transferred to IP-1 switchgear that supplies the IP-2 Alternate Safe Shutdown System.

The i

13.8 KV connection (13W92) is underground from GT-1 to the LGP room inside IP-1.

Presently,- Gas Turbine No. 1 can be started remotely from the CCR but has to be loaded onto IP-l's L6P. bus sections or IP-2's 6.9 KV buses via breaker controls in the gas turbine building.

We are planning to restore these controls to the CCR in the near future.

2)

Gas Turbines No. 2 6 3 are located offsite at the Con l

Edison Buchanan sub-station, approximately 1/2 mile from the IP-2 diesel generator building.

13.8 KV (13492 and 13W93) are supplied underground

  • power feeds from these gas turbines into IP-1 LGP bus sections.

Either (or both) of these gas turbines would supply IP-2 480 V safety buses in much the same manner as described above for gas turbine No.1.

Gas turbines No. 2 or 3 would be used to supply IP-2 480 V safety buses IMTE:

  • The feeder is run in conduit and the conduit is run underground except for connections to transformers and/or to enter buildings.

n

Should the seal deteriorate rapidly due to the loss of seal

cooling, a

less likely condition, leakage rates could increase causing the pressurizer to empty faster.

With decay heat being removed through the steam generator safety valves, and without the ability to replenish water lost through the RCP

seals, this situation would eventually result in

~

saturation conditions in the RCS and a stabilization of temperature and pressure at values slightly above those in the steam generator.

Under these conditions, the symptom-based emergency operating procedures direct the operators to take actions to rapidly reduce reactor coolant system pressure and temperature.

This causes a significant corresponding reduction in reactor coolant pump (RCP) seal leakage.

Reducing seal leakage will extend the time

.i available to restore A.C. station power before a potential inadequate core cooling situation can develop.

In addition to reducing the amount of water lost from the RCS, reducing RCS pressure and temperature will also reduce the rate and potential magnitude of seal degradation.

Finally, decreasing RCS pressure via secondary cooling allows for the injection of the water in the passive low pressure ECCS accumulators to g -

replenish some of the lost RCS inventory and even further g

extend the time available to restore A.C. station power.

5 1

Case 2:

The scenario described above for case 2 is equivalent to those cases analyzed for 10 CFR 50, Appendix R where the CCP.,

the control building, IP-2 480 V switchgear, and the diesel generators become non-functional due to fire induced electrical failures.

In the case of a tornado induced high wind condition striking the control building, the assumed

. structural damage is the failure of the north wall, exposing the interior of the control building.

The operators would initiate plant shutdown from outside the CCR.

The procedure which directs the operator can achieve safe shutdown from outside the CCR in any of three different ways.

First, sr.fe shutdown control with normal station A.C. power available; second, safe shutdown control with emergency 480 V A.C. power available; third, safe shutdown control with alternate IP-1 power available.

For Case 2, since offsite power was lost, the preferred safe shutdown method would involve local control of IP-2 safe shutdown equipment powered from the 480 V emergency buses being supplied by the emergency diesel generators or other A.C. power sources described in case 1.

Upon reactor trip RCS decay heat will be removed by natural circulation with steam generator feedwater flow being i

supplied by either a motor driven auxiliary feedwater pump or steam turbine driven auxiliary feedwater pump and atmospheric relief f rom the steam generators.

RCP seal cooling, RCS makeup and beration would be provided by a charging pump and

=

the refueling water storage tank (RWST). A component cooling (C04) pump and service water pump could be used to water provide cooling water to the RCP thermal barrier and CGI heat exchangers.

J h

D and E is multicourse brick construction-as are the walls surrounding the kitchen and locker room in the south east corner.

In -order to incapacitate both the CCR and the ASSS cables in the southeast corner, _the north wall siding or

.other~ debris would have to penetrate a) the interior IP-2 and IP-1 control panels, b).the steel. panel wall or several-multicourse brick walls and c) the steel conduits surrounding the ASSS cables at the ceiling level.

Since such an eventuality is considered. to be highly unlikely, the' ASSS cables in the IP control building are not considered susceptible to tornado or high wind induced damage.

Thus, the ASSS will be available in the highly ~~ improbable event that. the. IP-2 control building were damaged by a tornado or-high wind condition.

Case 3:

The scenario described above for case 3 is similar to previous post-DfI evaluations for reliability of secondary heat sink.

The likely structural damage to the _ condensate storage tank (C5T) from a tornado induced high wind condition would be a missile impact rupturing the tank and causing loss of its contents.

The high wind condition by -itself cannot cause gross failure of the CST due to its design. Given 'a gross rupture of the CST the city

.t. ;e r systou provides adequate backup capability for secondary heat sink.

The redundant low water level alarm system will-alert the

- operators to manually switch over to the city water system.

The city. water tank is located at elevation 120' approximately 1/2 mile from 1the CST, near the~ Buchanan service center.

Sufficient secondary inventory exists to

- allow the operator time to accoinplish switchover from the CST to city water.

Upon receipt of the low CST level alarm the i

operator would trip the running auxiliary feedwater pumps, re-align suction and re-start the pumps.

Therefore, even tnough a tornado generated missile may momentarily interrupt secondary heat sink it is very unlikely to cause a long tera loss.

Emergency Plan:

Con Edison's emergency plan procedure IP-1032 (attached) describes the i

actions to be followed in the event of a tornado watch or warning at the Indian Point site.

4 6

i r-n..,

-,n

- - - - - - ~ - - - - - - - - - - - - -, - ~ ~ - - - - ~

-a

f

..[,' ' '

Implementation Procedure IP-1032 Operating Personnel

\\f Upon notification of-a tornado watch; 1.

Maintain a watch to' listen for and look for a.

tornado.

2.

If a tornado is sighted, notify Control Room Operators-immediately.

1 i

I i

O IP-1032-2

.'AR 2 S 1985

w j

'v#.

d 21ABF EDCS PUMr> -

\\

i IB S

23 CHRG 4

I 12FD3 lM PUMP I

(Unit i SRM S.H. Bldg.

TH AND TC el. 33')

2B L2 23CCw EDF9

]

g PUMP 43 IT 12 12 EDC3 EDGl EZG4 2R

\\

"Via Casualty Cable" "Via Casualty Cable"

\\

MCC-27 EZHi 4

I d

23 5W 12RW3 IM PUMP I

Auxiliary Feedwater Building, el.15' (Unit i Screenwell

@ Primary Auxiliary Building, el. 80' l

House Roof) 3M L3 EDC4 24 SW 3

Adjacent to IP-2 Intake Structure i

PUMP t

e 9

FIGURE - 2 ALTERNATE SAFE SHUTDOWN FEEDS FROM UNIT I

.