ML19095A613

Submit Supplemental Report to Special Report No. SR-S1-75-08, Summary Report - Containment Leak Rate Testing

ML19095A613
Person / Time
Site:	Surry
Issue date:	07/01/1976
From:	Stallings C Virginia Electric & Power Co (VEPCO)
To:	Moseley N NRC/RGN-II
References
Serial No. 114 LER 1975-008-00
Download: ML19095A613 (10)

Text

,.

• ~ - -'.

,~-.,

--.**e ..

~.'

,I

~RIQHMO;D .y1im;1Nt.A:..2 3 ~ 61

.. Jul'y. '.*1976. 'r* *.** .

.*. rJJ~i{/. > <*:_.: : ... , ~

Mr. N?rman' c. *Moseley, *Di'rector .... * * :__

• office of Inspection and -Enforcem~t *, ~6;-, , .. * :* * . 4.

~s~-.* '\i15'?: :_,. . * -Serial No. 114

• P,O&M/AUI:jlf *.

I U. *s. Nu.clear_ Re'gulatory Commissi~_ <J,\:;Ji* *.1,\ ~ 00i-,o~* . ....._ -

• * * . * .. " e,\c<'<-

Region II - Suite *818 ..._. 0 * * .... '.~:c . .c' _* *. * ,':f_ '. Docke't No. ' ;50-280:

230 Peachtree*street,-'Northwest -_*. /' :-:,-. ~ _,('()_

• Licens.e No .. DPR.:..32 Atlanta,. Georgia *. 303.03 : .. :*.:* * * . -*_-

• Cb~~o-: ..

. . . *.' . *.* ~.

Dear' Mr. Moseley:

. . .. Th~::V!rg:lnia ~1~*~ir.ic:* and Power *G~pan~ hereby submit~ ~ c~py. of the

• 1.

. supplemerit_al_ report to Special Report :sR.:.si-75-08 entitled' "Summary* R,:port

.*. Conta:1.Ulllent**:~~ak'Rate. Testing 11 for.* Uni(* 1 dated .June *30, ..1976~ ..

. . . . *.. ' .. The. s~b~t;nce\,£ th':i,s*: rep~rt' h_a'{.,been .reviewed by the: .s~at,ion Nuclear

. Safety* and* Operating.'Conmittei-an"d :will ,b,./*pla:'ced,ori the age*nda .for* the riext_

meeting of* the Systei\1 *Nuc.ie:ar: S~fety,.an.d:Op<?rat:i.ng Comnit.tee*~ . . . .

Very- truly yours, ..

'. *. ~ '

Y! *.::")/>~ ;* ~L'?*q:-t2(:_*.. ~. :'*

,lj,~*~1.~~1~

' ..' ' ,, ' ~ , .

.* :,_, . ::, -: /-.::_ ._:c.j_.1'.'!**At'.'111ings ,* :,..

, :c.Vi~_e* President.:.Power Supply

. and Pr'oduction_* Operat:l.'oris:

Enclosure*

c_c: Mr* Rober~ W* Re:t°d, : *chief (46 *copies)(:.

_: Operating Reactors B;r.~nch'-.. 4. . . .

I *, .* *.*

I

. ~' ,, .

• . -j

SUPPLEMENTAL REPORT l

-t It SPECIAL REPORT l'

l i

SR-Sl-75-08 I

'iI i

i

.* I

SUMMARY

REPORT CONTAINMENT LEAK RATE I~.'.!1!!Q.

DOCKET NOo 50-280 LICENSE NO. DPR-32 JUNE 30, 1976 SURRY POWER STATION VIRGINIA ELECTRIC AND POWER COMPANY

e This report is submitted to provide supplemental information to SR-Sl-75-08, Summary Report of Containment Leak Rate Testing. The following topics will be discussed:

1. Errors inherent in correlating Type A leakage tests with Type B and C tests.

2. Instrument error analysis.

I. Correlation of Type A, Band C Tests Proposed Appendix J to 10CFR50 dated August 27, 1971 and specified in Technical Specifications states that if repairs are necessary to meet

• the acceptance criteria of the Type "A" Test, the test need not be repeated provided local measured leakage reductions achieved by repairs reduce the containment's overall measured leak rate to meet the acceptance criteria.

This*calculation is explained on pages 16-18 of SR-Sl-75-08. The results are repeated below with the addition of a correction for instrument error in the Type "B" and "C" tests (5.4%).

Correlation of data:

Integrated Leak Rate (1617 SCFH) *.I.,*

plus confidence band (60 SCFH) + 1677 SCFH Leakage reduction through SG tubes 1232 SCFH Leakage reduction through containment isolation valves in systems connecting the cont. atmosphere 38 SCFH with open systems outside the containment Leakage reduction thru containment isolation valves in systems connecting. the cont. :-atmosphere 796 SCFH with closed systems outside the containment.

Correction-for instrument error in Type Band + 112 SCFH c*Tests (5.4%)

277 ,SCFH

• * - * * *.~~,-..---.- .. ,-,* * * * . *** *****.** , * * , ... .;Q _ _ _ ~-,~* ..., * . .*, ' . ' * * * . * ~ ,~- - * * - . .---... ......... ~----**

~ .... ---.~ -~, ............... *-*~""" . ...------~~-,,.,. ....... .

- ' ,,*. -~ "**'<""*..- - - ,,,._'-~**, .. **I'

e e The apparent negative leakage is due to errors inherent in this type of correlation which tend to give different results in Type "B" and "C" tests than the leakage which existed at the time of the _Type "A" test. This error is due to the difficulty of duplicating during the Type "B" and "C" tests the exact conditions which existed at the time of the Type "A" test. Causes of this type of error include the following:

1. For leakages through valves which connect the containment with closed systems outside, a significant back pressure may have existed during the Type "A" test which was not present for the "B" and "C" tests. This would yield higher results in the Band C tests than existed during the Type A test.

2. During the Type "A" test, the containment sump is flooded

_to simulate accident conditions. As a:result,the leakage through_*certain '.Valves is water leakage rather than air leakage. -

Type "B" and "C" tests are air leakage tests and thus tend to give* higher leakage rates than existed during the Type "A" test. For this test, the outside recirculation spray pump suction valves (MOV-RS-155A & B) and the Low Head Safety Injection Pump Suction Valves (MOV-1860 A&B) which have water leakage during the Type "A" test, were major sources of leakage. The as found leakage test results for these valves were probably conserva*tive contributing to the negative corrected leakage.

The above items are the major sources of error in the correlation of leakages from Type A, Band C tests~ Other sources include changes in seating due to valve cycling betwe~n tests, seating variations due to different pressurization rates between tests, and instrument error.

.e .e II. INSTRUMENT ERROR Equipment Descriptions The differential pressure between the reference bulbs and the containment atmosphere was measured with a 0-60 inch U-Tube manometer, Meriam Instrument Model 20CB10. The fluid used was an oil with a specific gravity of 1.04. Accur-acy was to 0.01 inches with repeatibility to 0.0025 inches of oil.

The absolute pressure of the containment atmosphere was measured with a 0-120 inch mercury manometer, Exactel Model 500. Digital readout and accuracy were to 0.001 inches of mercury with repeatibility to .00025 inches of mercury.

Containment temperature was measured with 18 RTD's, Electric Thermometer Inc.

Model P-B2155. Individual RTD accuracy was to 0.1°F. An additional error of.*

.05°F is inherent in computer analog to digital conversion. Therefore accuracy is to 0.15°F with *repeatibi!itj 1* to 0.04°F.

Vapor pressure was determined using two dew cells, Foxboro Model 2711AG.

Accuracy~ including computer error was to 0.55%, repeatibility to .14%. This corresponds to .OS in H20 and .01 in H o, respectively.

2 The makeup air flow meter used for verification was an American Meter Controls, Inc. Model AL2300 with an accuracy.of 0.5%.

Instrument Error Analysis Both the reference and the absolute method of leakage determination require an hourly calculation of the mass of air in the containment. The leakage rate is then determined by a least squares fit of the calculated masses over a 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> period. The uncertainty in the mass calculation for each method is estimated as follows.

e Reference Method - uncertainty in mass calculation.,

The reference method mass calculation is by the formula:

WHERE: Q = quantity of air in containment at time, ti; lb mass.

Qo = quantity of air in containment at time, to; lb mass.

PD = manometer differential pressure at time, ti; in H2o.

Pno = manometer differential pressure at time, to; in n2o.

Pv = containment vapor pressure at time, ti; in H 0.

2 Pv

. 0

= containment vapor pressure at time, to; . in H o.

2 T = average containment temperature at time, ti; oR.

The change or uncertainty in Q due to uncertainties in the measured variables is given by:

• ~Q = 1.7558 X The formula assumes*that all errors are systematic rather than random

~PD = JpD = ep = epD = .01 in. H20 0 Do Jv0 = ~v\ = ev0 = ev = .os in. HzO JT =- eT = o.15/ {Ts = .035°R I

e e Combining equivelant terms:

eQ C 1.7558 e = 23.'4 lb mass i Q

Absolute Method - uncertainty in mass calculation.

The absolut.e method mass calculation is by the formula *.

~~... ...... Q = 1.7558 X 10 5

Pa = *1. 755.8 X 105 (P-Pv)

T T WHERE!: Q = quantity of air in con ta inmen t, lb mass.

Pa = containment ab.solute pressure, in. H20 P = containment pressure~ in H 0 2

Pv = containment vapor pressure, in. H20 0

T * = c*ontainment average temperature, R The change or uncertainity in Q due to uncertainties in the measured variables is given by:

. 5

[(

~ . ~p

,2.+(~. ~ Pv + d..Q. .~ _)2]

2 .( * .

I V~

-~Q = 1.7558 x 10 olP d.R, 1 d:f The formula assumes that all errors are systematic rather than random in character. The error in eQ = 1. 7558 X I/ II e

. p = .001 . Hg =.0136 H20 eV =

e = .15°R + .035°R T fl8

e Q

= 1.7558 x 10

- [f t

. **: -6...;.

.0136\

540 =-1 2

+ /.os, ts40J 2

. e

+ //1491.0 - 4.2) .035'2] 1/2

\l540 X 540 /

= 1. 7558 10 5 -10

+ 8.5

-9

+ ~-2

-81J 1/2 .

X [ 6.3 X 10 X 10_ X 10

= 1. 7558 X 10 5 [ 4.11 X 10- 8 Jl/ 2

.eQ = 35.6. lb *.*:mass

R'e

pea.':tibili t,y Error Analysis Using the same formulas as above, the repeatibity error for the two methods is estimated as follows:

Reference method:

Instrument repeatibility epD '

= .0025 *H2o 11 e = .0100 H 0

, V 2 eT = .007°R 112 eQ ~ 1.7558 X 10 5[2[.00:i~c,+,2/.0l\2 +ll28.423~2Q 02Q-4.244+4.l87).0Q7\il J

0

. *\*540 ,. ~4a°/ ' * ~ 540 X 540 )

sr -11 -10 * -14] l/2

~ 1$7558 x 10 L4.3 x 10 + 6.9 X 10 + 4

• 0 . X 10 . . .

= le7558 X 105 X 2.7{ X 10-S eQ = 4.76: lb. mass.

. . ... e e instrument repeatibility:

11 11 ep = .00025 Hg = .0034 H 0 0 2 I/

ev = .01 H0 2

eT = .04 OR= .009 f,18  ;*

• i~

eQ = l. 7558 X 10 [

5 f003~2 +

540 fo\)2 540

+(1491-4.2) 540x540

.oofJ1'2 5 10-ll 10-lO + 2 11 10-9] l/Z

= 1. 7558 X 10 ( 3.61 X + 3.00 X 0 X

5 -5

= 1.7558 X 10 X 4 95 0 X 10

= 8.69 lb. mass.

Leakage rate error due to instrument error.

As shown on pages 6 and 7 of report SR-Sl-75-08, the, leakage rate is,,

. determined by a least squares fit analysis of the hourly calculations of mass over the 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> test period. The effects of systematic instrument error are diminished by this method of analysis rendering repeatibility error the predom-inent source of instrument uncertainty. The effects of errors in individual mass calculations on the *overall leak rate may be estimated by the formula e

m

(}:X) 2 tr'"

-::1

.. )

Where e = error in slope in lbm/hr

• m -\I *--:_

0 For the reference method, the instrument repeatibility translates to an error in leakage rate as follows.

e m

= 4. 76 2

/ l:X - o:x) 2 ..,.

V . N ..:*,

• .}

e m =

• 132 lbm/hr.

em = .0007wt%/24Hr.

e For the absolute method.

• e m

= 8.69 2

~r.x -(EX)~

= .24 / lb/hr

= .0012 \wt%/24 hr.

Type Band C tests are performed using rotameters of various ranges manufactured by the Brooks Instrument Co. All are accurate to+ 5%. Flow pressure corrections are made based on a 0-50 psig Ashcroft pressure gage of+ 2% accuracy. Commutation of these uncertainties yields an overall error of 5.4%.

e =

• e = 0054 ... . -- . .

In conclusion, a negative corrected leakage rate could be expected due to the errors inherent in _correlating the Type A, B, and C tests as discussed herein. A correction for instrument error during the Type Band C tests still yields a negative corrected leakage after repairs. It is believed that the correlation between the Type A, B, and C tests remains valid and the measured leakage reductions achieved by repairs have reduced the containment's overall measured leak rate to acceptable limits.