ML20216F544

Provides Results of Risk Evaluation of 990630,flooding Event at Sequoyah 1 & 2 Reactor Facilities.Event Was Documented in Insp Rept 50-327/99-04 & 50-328/99-04 & Transmitted in Ltr,

ML20216F544
Person / Time
Site:	Sequoyah
Issue date:	09/07/1999
From:	Fredrickson P NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION II)
To:	Scalice J TENNESSEE VALLEY AUTHORITY
References
50-327-99-04, 50-328-99-04, NUDOCS 9909220086
Download: ML20216F544 (3)

See also: IR 05000327/1999004

Text

'

I

t

4

.

September 7, 1999

Tennessee Valley Authority

ATTN: Mr. J. A. Scalice

Chief Nuclear Officer and

Executive Vice President ]

6A Lookout Place J

1101 Market Street

Chattanooga, TN 37402-2801

SUBJECT: SEQUOYAH FLOODING EVENT RISK DETERMINATION (NRC INTEGRATED

INSPECTION REPORT NO. 50-327/99-04 AND 50-328/99-04)

,

Dear Mr. Scalice:

The purpose of this letter is to provide you with the results of our risk evaluation of a June 30,

1999, flooding event at your Sequoyah 1 & 2 reactor facilities. The evaluation was conducted

under the NRC's Significance Determination Process (SDP). This event is documented in

Inspection Report 50-327/99-04 and 50-328/99-04 and was transmitted to you in a letter dated

August 13,1999, in our letter, we informed you that one potentially safety significant issue had

been identified. Specifically, on June 30,1999, inadequate performance of the storm drain

system resulted in water from a heavy rainfall backing up and flooding the turbine building

railroad bay where the 6.9kv unit boards were located, creating the potential for a loss of offsite

power. At the completion of the inspection, the risk significance of this event had not been

determined. '

The NRC has completed a preliminary risk determination of this event using the SDP. Our

evaluation has determined that performance for this event is in the White regulatory response

band, which indicates issues with some increased rlsk to safety, and which may require

additional NRC inspections. The assumptions used by the NRC in determining the significance

of this event are provided in the enclosure.

Although we believe we have sufficient information to make a final risk determination on this

issue, we are providing you an opportunity to provide us with the following information within 14

days of the date of this letter:

-

The results of your analysis of the risk significance of this issue, if different than the

NRC's significance level

-

A description of the assumptions you used and consider applicable to the analysis

in addition, if you choose to provide us with this additional information and believe it would be ,

beneficial to discuss your risk significance evaluation with us, you should contact me within 14 1

days of the date of this letter to schedule a meeting. \

9909220086 990907

PDR ADOCK 05000327 ,

O PDR l

1

l

$0l j

.

TVA 2

in accordance with 10 CFR 2.790 of the NRC's " Rules of Practice," a copy of this letter and its ;

enclosure will be placed in the NRC Public Document Room.  !

Sincerely,

l

l

(Original signed by)  !

Paul E. Fredrickson, Chief

Reactor Projects Branch 6

Division of Reactor Projects

Docket Nos. 50-327, 50-328

License Nos. DPR.77, DPR-79

Enclosure: NRC Assumptions for Sequoyah

Flooding Event

cc w/enci:

Senior Vice President

Nuclear Operations

Tennessee Valley Authority

Electronic Mail Distribution

Jack A. Bailey, Vice President

Engineering and Technical Services

Tennessee Valley Authority >

Electronic Mail Distribution

Masoud Bajestani

Site Vice President

Sequoyah Nuclear Plant

Electronic Mail Distribution

General Counsel

Tennessee Valley Authority

Electronic Mail Distribution

N. C. Kazanas, General Manager

Nuclear Assurance

Tennessee Valley Authority

Electronic Mail Distribution

cc w/enci cont'd: See page 3

.

.

,

'TVA 3

cc w/ enc!: Continued

Mark J. Gurzynski, Manager

Nuclear Licensing

Tennessee Valley Authority l'

Electronic Mail Distribution

Pedro Salas, Manager

Licensing and Industry Affairs

Sequoyah Nuclear Plant

Tennessee Valley Authority i

Electronic Mail Distribution

D. L. Koehl, Plant Manager

Sequoyah Nuclear Plant

Tennessee Valley Authority

Electronic Mail Distribution

Debra Shults, Manager i

Technical Services I

Division of Radiological Health i

Electronic Mail Distribution l

County Executive

Hamilton County Courthouse i

Chattanooga, TN 37402-2801

Distribution w/ encl: j

R. W. Hernan, NRR

H. N. Berkow, NRR

l

PUBLIC l

l

OFFICE RII/pR_5 g ,RII g

51bMATURL yf gg 0 g []

NAML R54 [qnr0 N Mbr gnoW

DATE

f) ,

q W/ g /W9 W/ /vv v/ /WW 9/ /v3 W/ /vy w/ /uv

turYf 1 E5* ND YE5 NO YE5 NO YES NO YES NO I YE5 NO YE5 NO

Ws

OFFICIAL RECORD C PY DOCUMENT NAME; 1:\RPB6\5Q\ REPORTS \5q9904-8-r2.wpd

<

, NRC Assumptions for Sequoyah June 30,1999 Flooding Event

The NRC has determined through examination of Sequoyah's Individual Plant Examination for

External Events (IPEEE), submitted to the NRC in June of 1995, that the impact of rain induced

external flood on risk was screened out in the IPEEE as not an important contributor to plant

risk. /

The NRC used the site's Individual Plant Examination (IPE) to characterize the impact of Loss of

Offsite Power (LOOP) events caused by the flooding of the 6.9 kV switchgear. Revision 1 of the

Sequoyah IPE lists LOOP as contributing 10.5% of the plant's internal risk frequency with a

contribution to risk of 4E-6 per year. The LOOP initiator frequency is 4.85E-2 per year. This

results in a risk to frequency ratio for LOOP of 8E-5. An additionalinitiating event frequency

contributor to LOOP of greater than 1.23E-2 per year would result in a contribution to risk that

would exceed the Green threshold of the NRC Significance Determination Process (SDP) of 1E-

6 per year. Thus a LOOP-inducing rainfall event would be a significant contributor to core

damage frequency (CDF) if it were predicted to occur at a frequency greater than once every 80

,

years.

Flooding of the 6.9kV panels, to sufficient depth, could result in a dual unit LOOP event. The

6.9kV panels have experienced rainfall-induced water intrusion to a depth of approximately 2

inches, twice in the last five years. These rainfall events did not result in a LOOP event but were

estimated to be at the 2-year and 5-year storm intensities. The factors that determine how high

the intruding water level must peak to cause a LOOP event and the relationship between peak

water level and rainfall intensity are numerous. Rainfall studies for the area around Sequoyah

indicate that rainfall intensity increases by a factor of two when comparing a 50-year to a 5-year

rainfall.

The capacity and condition of the storm and internal turbine building drains, the volume and rate

of run-off into yard drains in the turbine building railroad bay (TBRB) area, and the elapsed time

between onset of flooding and its isolation through closure of the roll-up door and/or placement i

of other barriers have a significant impact on how high the resulting flood waters in the building

l

will peak. The impact of the June 1999 rainfall was limited by closure of the turbine building roll- '

up door. This was evidenced by the peak water depth outside the door (approximately 18

inches) compared to the peak level inside the TBRB (approximately 4-6 inchesy. Timely

isolation of the TBRB would not be certain for future rainfalls because the flooding is not alarmed

and the required response time can be very short.

A LOOP event caused by flood-damaged electrical panels would result in a longer expected

recovery time than that assumed in the LOOP initiating event recovery curves, which are based

on grid centered losses, and would require adjustments to the analysis recovery assumptions.

Longer expected recovery times would lead to a higher CDF contribution and would have the

effect of increasing the calculated CDF.

Given the information and assumptions presented above, that two rainfalls of less than 5-year

storm intensity have resulted in partial flooding of the 6.9 kV panels in the last five years, and

given the likelihood of rainfalls of higher intensity during the plant's lifetime, the NRC has

determined that Sequoyah's performance for this event is in the White regulatory response band

of the NRC SDP.

Enc!osure

_~