ML18026B073

Forwards Description of Scram Discharge Vol long-term Mods, Implemented Per BWR Owners Group GDC & Criteria of Generic Ser,Bwr Scram Discharge Sys

ML18026B073
Person / Time
Site:	Browns Ferry
Issue date:	06/27/1984
From:	Mills L TENNESSEE VALLEY AUTHORITY
To:	Harold Denton Office of Nuclear Reactor Regulation
References
NUDOCS 8407050001
Download: ML18026B073 (11)

Text

1 REGULATORY NFORMATION DISTRIBUTION S

'TEM (RIDS)

ACCESSION NBR:8407050001 DOC.DATE.: 84/06/27 NOTARIZED: YES DOCKET '0 FACIL:50 259 Browns Ferry Nuclear Power Stations Unit l~ Tennessee 05000259 50 260 Browns Ferry Nuclear Power Stations Unit 2s Tennessee 05000260 50"296 Browns Ferry Nuclear Power Stations Unit 3E Tennessee 05000296 AUTH'PME AUTHOR AFFILIATION MILLSI L"s M ~

Tennessee Valley Authority REC IP, NAME RECIPIENT AFFILIATION DENTONeHOR ~

Office of Nuclear Reactor Regulationi Director SUBJECT$

Forwar ds description of scram discharge vol long,"term modsi implemented per BAR Owners Group GDC II, criteria of "Gener ic SER p BNR Scr am Dischar ge Sys ~"

4 DISTRIBUTION CODE:

A001S COPIES RECEI)EDtLTR g ENCL/

SIZE1 TITLE$

OR Submittal$

General Distribution NOTES$ NMSS/FCAF,icy. icy NMSS/FCAF/PM'L!06/26/73 NMSS/FCAF icy, icy NMSS/FCAF/PM, OL$ 06/28/74 NMSS/FCAF icy, icy NMSS/FCAF/PM,'L:07/02/76 05000259 05000260 05000296

-RECIPIENT ID CODE/NAME NRR ORB2 BC Ol INTERNAL: ELD/HDS4 NRR/DL DIR NRR~S IAIETB 04 COPIES LTTR ENCL 7

7 1

0 1

1 1

1 1

1 RECIPIENT ID CODE/NAME NRR/DE/MTEB NRR/DL/ORAB NRR/DSI/RAB RGN2 COPIES LTTR ENCL 1

1 0

1 1

1 EXTFRNAL$ ACRS NRC PDR NTIS NOTES!

09 02 6

6 1

1 1

1 2

2 I.P DR NSIC 03 05 1

1 1

'TOTAI. NUMBER OF COPIES REQUIRED $

LTTR 27 ENCL 25

1

'(g[ "0)i<<(',

) g g<<

'p <<<< 'll P(>>c ())<<",$ ~ >>

1 '>>

f}P PCI e il t

,I f r(s r r(<<

>> (1(

m>><<q I)pr g <<plrg)>r fp atro

}0 )( ~

<< I(>>>>

f,

$ <

I f) '.!8 g V C, ~,0 i%i ~ ~ ( pi ~ I) ) ~T( r Q C.J )f<< I f (<< ( t. ~" p (p <<>h r r ' r'1 Ic'~(t' l(><<i'<< '1 JP > r <<~(l'( '1<<P(r>>>><<9 t('9( g f (1 Ik>><~8>> r gy'f,, I, r >>e>> e ~ ~>> ~ II) 3>>T j>> f( r ~ $ /1( (J>>f,;1 (lt(: l ll I I P (,') I)i) Q ii<<>>f ( i >> p f>>') << II II ) f >>)'lp+w >(ply> ) ~ l$ $,) <,pl

gal, 3A 7 %f4'> X

,~>l (rC X5r]X ~o (,"I'i )( p),c>>("" (ling J)$ 'i i)tfy I 4 (I l J >>li il"r p ('7'<< q(tx TENNESSEE YALLEYAUTHORITY CHATTANOOGA. TENNESSEE 37401 400 Chestnut Street Tower II June 27, 1984 Mr. Harold R. Denton, Director Office of Nuclear React;or Regulation U.S. Nuclear Regulatory Commission Washington, D.C. 20555

Dear Mr. Denton:

In the Matter of the Tennessee Valley Authority Docket Nos. 50-259 50-260 50-296 By my letter to you dated October 6,

1982, we submitted a description of the long-term scram discharge volume modifications to be performed at the Browns Ferry Nuclear Plant.

As stated in that letter, consideration was given to the BWR Owners'roup design criteria and the criteria of the "Generic Safety Evaluation Report, BWR Scram Discharge System" transmitted in D. G. Eisenhut's letter To All BWR Licensees dated December 9,

1980.

Our review indicated that we complied with the intent of the various crit;eria above.

In addition, by letter dat;ed June 24,

1983, TVA received from NRC Confirmatory Orders requiring that; our modifications be completed and in conformance with the criteria discussed above on units 1 and 3 before reactor operation in cycle 6.

During implementation of our original design, we encountered various problems and performed additional analyses in support of the modifications.

In response to these problems and analyses, we made several design changes.

These design changes were necessary in order to meet the intent of the above criteria.

In accordance with section 4.3 of the Generic SER, these modifications require no preimplementation approval by the NRC staff.

We believe that, although we meet the intent of the criteria, we should inform

" you of these changes.

Therefore, enclosed is a description of the problems encountered, status of the modifications, and our expected final configurations.

If you have any questions, please get in touch with us thr ough the Browns Ferry Project Manager.

Very truly yours, 840705000i 840627 PDR ADOCK 05000259 P

PDR

.. 'ubscribe sworn to efore me this day o NNESSEE VALLEY AUTHORITY L. M. Mills, Ma ager Nuclear Licensing 1984.

Notary Public My Commission Expires Enclosure cc:

See page 2

An Equal Opportunity Employer

4)", '

(4

~V, Ihtt Jl =

I 44

~

thh 4

44 4

I 4

4 I

~"

I I tt I

'4 4

4

Mr. Harold R. Denton June 27, 1984 cc (Enclosure):

U.S. Nuclear Regulatory Commission Region II ATTN:

James P. O'Reilly, Regional Administrator 101 Marietta Street, NM, Suite 2900 Atlanta, Georgia 30323 Mr. R. J. Clark Browns Ferry Prospect Manager U.S. Nuclear Regulatory Commission 7920 Norfolk Avenue

Bethesda, Mar yland 20814

1 S

J f

n

'4

ENCLOSURE BRO>1NS FERRY NUCLEAR PLANT SCRAM DISCHARGE VOLUME (SDV)

LONG-TERM MODIFICATIONS Except as described in the following, the SDV long-term modification configuration is as described in TVA letter from L. H. Mills to H. R. Denton dated October 6, 1982.

Piping Configuration Changes During the initial installation of the long-term modifications on unit 2, calculations were being performed as required by the safety evaluation report (SER) to show that the scram discharge instrument volume (SDIV) would initiate a scram while sufficient volume existed in the SDV to accept the scram water.

Two scenarios were postulated for introduction of water into the SDV during power operations, namely excess control rod drive (CRD) leakage (approximately 10 gpm/SDV) and fast-fill leakage (approximately 465 gpm/SDV) caused by a degraded control air event.

The calculations indicated that the modified SDV would adequately respond to the normal leakage event but that, even with the 6-inch SDV-to-SDIV connecting pipe, the modified system might not be able to adequately respond to the fast-fillevent because of air binding in the connecting pipe.

Since these calculations were completed late in,the outage and there were no means to perform a test to verify the computer model, it, was decided to leave the installed CRD control air header pressure scram switches in place rather than removing them as originally intended.

These switches were required by IE Bulletin 80-17 specifically to prevent the fast-fill event.

Therefore on unit 2, the normal leakage event was addressed by piping configuration, and the fastfill event was addressed by the air header pressure switches.

The intended final configuration for the long-term modifications included deletion of the air header pressure switches with reliance for event mitigation to be only on piping design.

Thus, on units 1 and 3 an additional 2-inch vent pipe between the SDIV and SDV was designed to prevent the air binding.

Updated calculations then indicated that the modified configur ation would adequately respond to both postulated scenarios.

It was then decided that, based on the calculations, unit 2 would be modified during a future outage to include the 2-inch vent

'J pipe to allow removal of the pressure switches.

Instrumentation Configuration Changes As required by the SER and confirmatory orders, the SDIV high-level scram was designed to include redundant and diverse level switches.

The four original Robertshaw float switches were to be replaced with two Magnetrol float switches and two Rosemount sealed differential pressure (dp) transmitters on each SDV.

This was the configuration installed on unit 2; r ~ during the initial instrument calibration and installat io process~

a problem appeared with the dp transmitters't was discovered

.'.<-:.Ii--,;-~y~

that, because of their sealed capillary design, the transmitters could have ';<<>,'"p~

an actuation time to a step change in SDIV level of as much as 71 seconds..~','"<,:,-~-,"~~"..'$

'<<e

I

The original design calculations for event mitigation had assumed a prompt actuation.

Again, calculations were performed which indicated that for the normal leakage event the actuation time was acceptable;

however, foe the fast-fill event, the air header pressure switches were required to prevent the occurrence of the event.

It was determined that, although the dp transmitters provided acceptable results with the pressure switches installed, the final configuration should have instrumentation with a more reasonable actuation time.

Con-sequently, the designs for units 1 and 3 were changed to replace the dp transmitters with Fluid Components, Inc. (FCI), heated reference resistive temperature devices (RTDs).

Since we had no experience with the RTDs and our postmodification tests were not conclusive for validating the modeling of the fast-fill event, the air header pressure switches were left installed to provide an additional degree of safety.

Based upon the excellent experience of the RTDs on unit 1, the design for unit 2 was changed to replace the dp transmitters with the RTDs.

This work will'be performed during a future outage on unit 2.

In addition to the above, another problem with the instrumentation has recently been discovered.

Based upon analyses of scrams on units 1 and 2

since modifications, it has been determined that under scram conditions the t4agnetrol float switches have.a time delay of approximately 20 seconds compared to the RTDs.

We believe that the RTDs provide a more realistic representation of SDIV level based upon previous scram data and analyses.

The situation does not, however, present a safety concern.

Since these switches are required to function to initiate a scram only for those events that fillthe SDV prior to a scr am, they are required only for the earlier discussed normal leakage and fast-fill events.

For the normal leakage

events, the float switches are expected to operate without a time delay.

Even assuming a time delay for the normal leakage

events, the response time remains acceptable as previously discussed concerning the dp transmitters.

For the fast-fill event, the air header pressure switches are still installed to prevent the scenario.

Because of the above

problems, we have now decided that we will leave the air header pressure scram switches permanently installed, in addition to the 2-inch vent pipe, to prevent and/or mitigate the fash-fill scenario.

The pressure switches installed were well in excess of the original requirements of IE Bulletin 80-17.

However, additional qualification may be necessary consistent with incorporation as a permanent feature.

In summary, the final configuration for the SDV long-term modifications will be as described in the TVA letter dated October 6, 1982, with the following changes:

(1)

A 2-inch vent pipe will be installed between the SDV and SDIV;

~

r

~

a~a I

~

~

(2)

The dp instruments will be replaced with heated reference

=RTDs; and, (3)

The air header pressure scram switches will remain permanently installed and upgr aded as required.

ln addition, we will submit proposed technical specifications covering these switches.

Status We have corn are in the current ref pleted the long-term modifications on Browns Ferry unit l.

process of completing the modifications on unit 3 during the ueling outage.

We '

~"

~

C