ML19284A406

in Response to NRC 790208 Request Forwards Addl Info for Review of Proposed long-term ECCS Mod of Small Break LOCA for Facility

ML19284A406
Person / Time
Site:	Crystal River
Issue date:	02/28/1979
From:	Stewart W FLORIDA POWER CORP.
To:	Reid R Office of Nuclear Reactor Regulation
References
3--3-A-3, 3-0-3-A-3, NUDOCS 7903060246
Download: ML19284A406 (16)

Text

.

oco' QC '

e O (,M f Q *O @ *

• o_oM W,ee.h.. #

Florida Power eoaeoaarin=

w e.siew^ar.oiaectoa POWER PRODUCTION February 28, 1979 File: 3-0-3-a-3 Mr. Robert W. Reid , Chief Operating Reactors Branch #4 Division of Operating Reactors U.S. Nuclear Regulatory Commission Washington, D.C. 20555

Subject:

Crystal River Unit 3 Docket No. 50-302 Operating License No. DPR-72 ECCS Small Breaks Analysis

Dear Mr. Reid:

In your letter of February 8,1979, you requested additional information in order to complete your review of our proposed Long Term ECCS Modification for the Small Break LOCA for Crystal River Unit 3. That additional information is enclosed.

If you have any further questions, please do not hesitate to contact me.

Sincerely, FLORIDA POWER CORPORATION b

4 . LC A W. P. Stewart Enclosure WPSemh Bright (ECCS)D50 7903 000 g tjf, General Office 3201 inirty-fourtn street soutn . P.O Box 14042, St. Petersburg. Fiorida 33733 e 813-866-5151

STATE OF FLORIDA COUNTY OF PINELLAS W.P. Stewart states that he is the Director, Power Production, of Florida Power Corporation; that he is authorized on the part of said company to sign and file with the Nuclear Regulatory Commission the information attached hereto; and that all such statements made and matters set forth therein are true and correct to the best of his knowledge, information and belief.

r I

I -(* f1Lueb W.P. Stewart Subscribed and sworn to before me, a Notary Public in and for the State and County above named, this 28th day of February,1979.

~

~

AJ 4 Notary Public b M- (

Notary Public , State of Florida at Large, My Commission Expires: August 24, 1979 (Notary 2 D12)

REQUEST #1: Please document the qualifications to which your proposed modification will be designed and tested. You should cite codes and standards where applicable. Specific reference to sections of your FSAR is acceptable.

RESPONSE #1: Our proposed modification will be designed and tested to meet the applicable parts of Sections 6.1 and 8 ot the Crystal River Unit 3 FSAR.

REQUEST #2: Active Component Failures 1, 2 and 4 require a different operator action than 3. Will the operator take the scme action if any of these four failures occur or must he evaluate the cause of loss of high pressure injection (HPI) flow before taking action? If the operator takes the same action in all cases describe what that is and demonstrate that correcting for an Active Component Failure that may not exist does not cause problems. What indications are available for the operator to assess the Active Component Failure?

RESPONSE #2: The operator will not take the same action if any of the Active Component Failures ( ACF) 1, 2, 3 or 4 occurrs.

Section 3.4.1 of Emergency Procedure 106 ((EP-106), " Loss of RC/RC Pressure", tells the operator to verify proper f unctions of Engineered Safeguards Channels "A" and "B",

High Pressure Injection, RB isolation and cooling and Low Pressure Injection.

Upon initiation of the Engineered Safeguards Actuation System, the operator can readily tell if both actuation signals have been automatically actuated. If either signal has not been actuated (ACF3), the operator will attempt to manually operate the equipment that the unactuated signal should have operated. The HPI valves are manually operated (opened) by remote manual switches on the Engineered Safeguards Panel in the Control Room.

If both actuation signals have been automatically actuated ,

the operator will verify the proper functions of the HPI system (and others). If either HPI train is not injecting water into the reactor coolant system (ACF 1, 2 or 4), the operator will turn the proposed transfer switch for that train which will transfer the operating power for the injection valves from its normal bus to the other ES 480v bus. This will supply operating power to the injection valves, which already have an "open" signal from the Engineered Safeguards System, and the valves will open.

If an operator does correct for an ACF that may not exist, this will not affect the mitigation of the accident as long as the operator does perform the proper corrective action also.

Bright (ECCS)

D50 i

RESPONSE #2: (Cont'd)

If the operator attempts to open the HPI valves for ACF 1, 2 and 4, this will be redundant to the Engineered Safeguards Actuation open signal and will af fect that signal. The operator still has to perform the proper corrective for ACF 1, 2 and 4 for adequate HPI flow.

If the operator switches the operating power from a live bus to a live bus on loss of an Engineered Safeguards Actuation Signal, the HPI valves will still have operating power and the operator will still have to perform the proper corrective action for ACF 3 for adequate HPI flow.

For ACF 1, the operator has the following indications at the Engineered Safeguards Panel (annunciators provide both visible and audible alarms):

Annunciator A/D-1-1 HP INJ ES ACT "A"/"B" Annunciat'or A/D-1-2 HP INJ Flow High/ Low Annunciator A/D-8-6 Diesel 3 A/B Fail to Start No Auto Start Annunciator B/E-4-1 Diesel 3 A/B Trouble Annunciator B/E-4-4 Diesel A/B Diff Relay Eng No Auto Start Annunciator B/E-4-6 Diesel Generator 3 A/B No Cont Volt No Auto Start In addition, the HPI valves would not open, the HPI and cooling water pumps would not start and there would be no HPI flow indication. On the electrical control room panel, the diesel irdications would not indicate megawatts, kilovolts, amperes or closed breaker.

For ACF 2, the indications would be the same as ACF 1 since ACF 2 results in the loss of an emergency diesel generator (ACF 1).

For ACF 3, the operator would receive an annunciator alarm (A/D-5-6 Trouble ES "A"/"B") as well as being able to visually compare the Engineered Safeguards Actuation signal status lights. For automatic Engineered Safeguards Actuation, the Trip Circuit Actuation status light windows become red. These status light windows are three across and fourteen high. It will be obvious if one Engineered Safeguards Actuation signal does not become automatically actuated.

For ACF 4, the operator would receive an annunciator alarm (A/D-1-2 HP INJ Flow High/ Low) as well as local indication that the hPI valves were closed, local indication that there was no HPI flow and the HPI valve ES status light would not change from amber (closed) to blue (open).

Bright (ECCS)

D50

REQUEST #3: Is it necessary that any flow paths in the makeup and puri-fication system be isolated af ter a LOCA to insure adequate HPI flow? If so, provide assurance that this isolation will occur for the postulated small break LOCA, concurrent loss of of fsite power and a single failure.

RESPONSE #3: There are two additional paths in the Makeup and Purification System that must be analyzed in order to assure adequate HPI flow. These paths are the RC Pump Seal Injection (through Penetration #338) and the normal Makeup and Purification pathway (through Penetration #435 in conjunction with one of the HPI flow paths).

Upon an Engineered Safeguards Actuation Signal, the normal Makeup and Purification pathway is isolated by MUV-27 while the RC Pump Seal Injection continues to operate. If the Active Component Failure was either the loss of an emergency diesel generator (ACF 1) or the loss of an Engineered Safe-guards 4160v bus ( ACF 2), operating power to MUV-27 could be lost and it would not isolate. An HPI pump would also be lost and one HPI pump would remain to mitigate the accident through the Seal' Injection line, the normal Makeup and Puri-fication line and the four HPI lines (af ter the two HPI valves that lost operating power were switched to the line 480v bus).

This scenario is considered worst case as one HPI pump must supply all HPI to the RC System and no more than 30% of it may be lost through the small break. (Note: The Seal Injection flow at each pump is actually into the RC System, it is adding to the coolant inventory at that point and is not being lost.)

The attached Gilbert Associates Report addresses this case and the mos. conservative set of potential losses is the flow through F etration #435 using HPI Pump A and one-fourth the flow through Penetration #338 (since the line goes to four pumps, one-fourth of it will go to the leg with the small break and subsequently lost).

This totals 29.3% of the flow; therefore, as required, more than 70% of the flow from one pump is available as injection into the Reactor Coolant System to mitigate the small break LOCA.

REQUEST #4: Address the possibility of the HPI pumps exceeding runout conditions when the system is in its minimum resistance configuration (apparently 4 RPI lines with 1 pump).

RESPONSE #4: Each of the three HPI pumps has been throttled on its discharge side to 500 gpm by a stop-check valve (MUV-10, -6 and -2 for pumps MUP-1A, B and C, respectively). The maximum flow calculated for any of the cases analyzed is Bright (ECCS)

D50

/

RESPONSE #4: (Cont'd) less that 510 gpm. The minimum NPSH available is forty-five (45) feet of water when the Borated Water Storage Tank is empty. The NPSH required is less than thirty (30) feet for flow rate less than 575 gpm. Thus, pump runout at the flow rates calculated is not considered credible.

REQUEST #5: You are requested to provide a commitment to verify, by meas urement , HPI flow rates prior to plant startup or within seven days after the modifications have been completed, whichever occurs first.

RESPONSE #5: In performing the calculations for our proposed modifica-tion, the Gilbert Associates Topical Report, GAI-TR-105-NP-1, was employed. This computer code has been verified by the Borated Water Storage Tank drawddown test at CR3 and it has also received NRC approval (J.F. Stolz, NRC to F.W. Symons, GAI dated March 29, 1978). The computer runs were based on actual pre-operational test data and the results show that the modification, as proposed , will miti-gate the small break LOCA in question. Therefore, we do not propose to commit to verify, by measurement, the HPI flow rates.

REQUEST #6: You indicated an installation schedule of '31 weeks from NRC acceptance' in your letter of November 17, 1978 and again on January 12, 1979. Your current exemption from 10 CFR 50.46(a) regarding this matter terminates at your next re-f ueling outage. There fore , you should make every ef fort to install an approved modification by that time.

RESPONSE #6: Florida Power Corporation has and is making every reasonable effort to install an approved modification by the end of our next ref ueling outage. If we find that this installation cannot be completed by that time, FPC will request an exten-sion to the exemption from 10 CFR 50.46(a) in order to com-plete the installation of an approved modificiation.

Bright (ECCS) 050

NOVDIBER 22, 1978 GAI REPORT No. 1998 REVISED: FEBRUARY 16, 1979 FLORIDA P0k'ER CORPORATION HIGH PRESSURE INJECTION SYSTDI ANALYSIS CRYSTAL RIVER UNIT 3 NUCLEAR GENERATING PLANT GILBERT ASSOCIATES, INC.

P. O. BOX 1498 READING, PENNSYLVANIA 19603 G@ertlL . :2

TABLE OF CONTENTS Page INTRODUCTION 1 SYSTEM DESCRIPTION 1 MATHEMATICAL METHOD 2 RESULTS 2 REFERENCES 3 TABLES:

1. Branch Data 4

2. Equivalent Length in Pipe Diameters (L/D) of Valves and Fittings 6

3. Pump Head Data 7

4. Control Run Results 8

5. HPI Flow Distribution 9 FIGURES:

1. Pump A Control Set Model 10

2. Pump B Control Set Model 11

3. Pump C Control Set Model 12

4. Schematic of High Pressure Injection System 13 APPENDIX A Rev.

Effects of Reactor Coolant Pump Seal Injection 3ystem and Reactor Coolant Makeup System on the High Pressure Injection Flow Distribution Gitert /Commomnealth

APPENDIX A Geert /Canunwealth

APPENDIX A EFFECTS OF REACTOR COOLANT PUMP SEAL INJECTION SYSTEM AND REACTOR COOLANT MAKE-UP SYSTEM ON THE HIGH PRESSURE INJECTION FLOW DISTRIBUTION INTRODUCTION SYSTEM DESCRIPTION MATHE!!ATICAL MODEL RESULTS TABLES:

A-1 Branch Data A-2 HPI Flow Distribution FIGURES:

A-1 Schematic of Reactor Coolant Pump Seal Injection System and Reactor Coolant Make-up System A-2 Schematic of the High Pressure Injection System, the Seal Injection System, and the Makeup System Giibert /Commamessith

INTRODUCTION Review of the main report indicated two potential flow paths were not considered in the original analysis. They are the reactor coolant pump seal injection system and the reactor coolant makeup system. This appendix reports the methods and results of the analysis performed to determine the effect of these flow paths on the high pressure injection flow distribution.

SYSTEM DESCRIPTION The reactor coolant pump seal injection system and the reactor coolant make-up system are designed to operate during normal plant operction (Figure A- 1). The normal seal injection flow is 32 gpm manually controlled by valve MUV-16. The normal make-up ficw is 45 gpm. This is controlled automatically based on the pressurizer level by valve MUV-31. One make-up pump is used to provide these flows.

In the event of a small pipe break, the seal injection system is designed to

..ontinue to operate while the makeup system is intended to be isolated automatically by valve MUV-27. In the event of loss of electrical power to MUV-27, the makeup system "ill continue to operate (Figure A-2). The seal injection valve MUV-16 will remain at the same operating oposition as before the break. Makeup valve MUV-31 will open fully because of the loss of system prestare.

MAT 5dMATICAL MODEL Two runs were needed to determine the flow in the high pressure injection system.

The first run modeled the normal system operation in order to determine the pressure loss in the seal injection line required to linit the flow to 32 gpm.

(Figure A-1) A reactor coolant pressure of 2200 psia was assumed. Due to the GibertICommomneelth

inability to model the pressure lose in the seals themselves, the contial valve model represented the whole seal injection system from MUV-16 to the reactor coolant system instead of only MUV-16. A make-up flow rate of 45 gpm was also assumed. The new branch data used to model the system is given in Table A-1.

Data unchanged from the original analysis is given in Table 1.

The second run modeled the high pressure injection system (Figure A-2). The seal injection system pressure loss determined in the first run was used. In addition, the stated pressure loss in MUV-31, 95 psi at 70 gpm, was assumed in the make-up system. The rest of the model was the same as in the original model.

RESULTS The pressure loss in the seal injection system required to limit the flow to 32 gpm was determined using each of the make-up pumps individually. A pressure loss of 658.5 1 0.5 psi at 31.3 gpm was calculated for each case.

The results of the high pressure injection flow distribution assuming a seal injection pressure loss of 658 psi at 31 gpm are summarized in Table A-2. The seal injection flow is presented separately. Operating experience indicates there is little leakage, on the order of 2 gallons per hour, from this system.

The make-up flow, although presented separately, is included in the flow through penetration 435.

Geert/Commonweesta

TABLE A-1 BRANCll DATA I Stop Check Pipe Straight Swing or Entr. Total Elev From To 1.D. Pipe 90 45 Cate Check Globe Tee Tee or Equiv Change lkt Junction Junction (inch) (feet) elbow elbow Valve Valve Valve Run Branch Reducer Exit L/D (feet) 8 7' 8 3.438 1.0 1/2 2 1 40 0.0 9 8 40 3.438 18.0 3 2 2 126 7.25 11 9 10 3.438 15.83 2 2 1 2

122 8.5 18 8 16 3.438 26.7 3 1 2

110 8.5 2

32 33 24 3.438 4.8 1 1 1 100 0.0 2

35 34 9 3.438 1.0 1/2 1 1 100 0.0 37 40 24 3.438 0.0 1 20 0.0 2

38 40 41 3.438 0.0 1 2 120 1.25 39 41 17 2.125 44.24 10 1 3 3!i 2 1 45 8 3

3.50 40 41 13 3.438 19.00 3 1 1!5 103" 25.30 1 - Data for branches 8 through 35 shown below are revised from Table 1. All other data on Table 1 was used without revision. Branches 37 through 40 are new branches.

2 - Nodes are located at these tees. Tee branches are modeled with an L/D of 50. The remaining L/D of 10 is modeled as 1/2 Tee run on the branch exiting the Tee.

3 - This branch represents the make-up line. Additional equipment loss of 85 psi at 70 gpm was used.

4 - This branch represents the seal injection system. Additional equipment loss of 658 gpm at 3 gpm was used.

TABLE A-2 IIPI FLOW DISTRIBUTION Total Penetration Flow Make-Up Seal Injection Flow #434 #435 #336 #337 Flow Flow Pen #338 Pump A - -

Flow - gpm 506.5 112.2 146.6 131.4 109.8 20.8 6.5 Percentage 22.2 28.9 25.9 21.6 1.3 Pump B Flow - gpm 507.4 108.6 143.1 135.6 113.4 23.5 6.7 Percentage 21.4 28.2 26.7 22.3 1.3 Pump C Flow - gpm 506.0 105.5 139.0 138.9 116.1 22.9 6.5 Percentage 20.8 27.5 27.5 22.9 1.3

• Make-up flow is included in the flow for penetration 435.

kN

' (9 . N D~-~~

k

yr k' u i C X N V V l '. n

@y\-

(n

'n (N X

' s O in[ M X

(,,#0 g (

y X

o e N- h-W ri)

"' 4) b -

q Z -()'\ N ,

,'N w a r FIGURE A-1 Schematic of Reactor Coolant Pump Seal Injection System and Reactor Coolant Make-up System

.M

%- -C

. wm W '$

_ X y

,n

\ >i i

/%

\

\ (0 x

x

\.

A-kN (q . N r.1

~ _. --

> F-r:1 O k .Dd i b \0 C >d h g'E h z7

[.

U h

@ ni r.4

) (q W V! ' s O (n

17

'(g}h'% C D<1- Nm K

X .

tp 0) to N Q S C [M "' -

C\

\N -

k*,W W

$ R p _ _ _ FIGURE A-2 Schematic of the High Pressure 7

[ . y Injection System, the Seal Injection System, and the Make-up System

• [q

q 'd by e n n~s yw &-m 5 X x X 5

~ .

\. .\.. Q \- -

sN x - .-

% \ / ,, #

% \ g/p '

~

Q

.&