U-601167, Forwards Addl Info Relating to Questions on Pneumatic & Hydrostatic Testing of Containment Isolation Valves & Pressure Isolation Valves.Info on Stroke Times of Certain Valves to Be Evaluated by NRC Also Encl

Forwards Addl Info Relating to Questions on Pneumatic & Hydrostatic Testing of Containment Isolation Valves & Pressure Isolation Valves.Info on Stroke Times of Certain Valves to Be Evaluated by NRC Also Encl

ML20151Q932
Person / Time
Site:	Clinton
Issue date:	04/18/1988
From:	Spangenberg F ILLINOIS POWER CO.
To:	NRC OFFICE OF ADMINISTRATION & RESOURCES MANAGEMENT (ARM)
References
U-601167, NUDOCS 8804270178
Download: ML20151Q932 (19)

Text

,

nO ,o;

^

U-601167 L30-88 ( 04- 18)-L '

8E.100 1

/LL/NDIS POWER 00MPANY _

CLINToN POWER STATION, P.o. 80x 678 CLINTON ILLINOIS 61727

"' 10CFR50.34 April 18, 1988 Docket No. 50-461 Document Control Desk Nuclear Regulatory Commission Washington,.D.C. 20555

Subject:

Clinton Power Station Containment Design

Dear Sir:

On March 30, 1988, Illinois Power Company (IP) met with members of the Office of Nuclear Reactor Regulation and Region III to discuss the appropriateness _of the Clinton design for primary containment integrity as presented in the Final Safety Analysis Report and as accepted by the NRC Staff in the Safety Evaluation Report. Additionally, telephone discussions were held on April 11 and 12, 1988, to address the March 30, 1988 questions and NRC questions pertaining to the Inservice Testing (IST) program.

As a result of the meeting and subsequent discussions, the NRC

Clinton Project Manager requested IP to supply additional information relating to questions on pneumatic and hydrostatic testing of containment isolation valves and pressure isolation valves. This information is provided in Attachment 1. Included too, as Attachment 2,

' are the stroke times of certain valves to be evaluated by the NRC. The listing corresponds to a list of valves provided by your Mr. M. Huber.

If you have any questions or require additional information, please contact me.

Sincerely yours, F. A. Sp en erg, I .

l Manager - Licensing and Safety i

l RFP/cke Attachments cc: NRC Resident Office NRC Region III Regional Administrator NRC Clinton Licensing Project Manager M. Huber, Region III

\

l SB841865 P

S88$$$

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

U-601167 Attachment 1 In conversation during the March 30, 1988 meeting, certain questions were raised which required IP to provide additional information. These questions and responses are listed below.

1. Identify the testable check valves that are Pressure Isolation Valves (PIVs) but not Containment Isolation Valves (CIVs).

The_ applicable valves are~as follows:

IE12-F041A LPCI RHR A Testable Check 1E12-F041B LPCI RHR B Testable Check 1E12-F041C LPCI RHR C Testable Check 1E21-F006 LPCS Testable Check ,

1E21-F005 HPCS Testable Check 1E51-F066 RCIC Testable Check

2. Provide a calculation comparing a 1000 psi PIV test to an Appendix J Type "C" test. Discuss surface tension differences and leak rate j differences between the two tests. Also compare the calculation results to actual test results on the valves which have been tested by both methods.

The results of the PIV to CIV correlation are provided in the ,

enclosed Sargent & Lundy Calculation No. 01ME114, Rev. 2.

3a. Provide tolerances for the test equipment used for the closed loop ,

and pressure isolation valve leakage tests.

Closed Loop

- Pressure 2% full scale accuracy

- Flow 21% full scale accuracy l

Pressure Isolation Valve

- Pressure 2.5% full scale accuracy i

- Leakage estimated to be t.05 gpm due to leakage ,

is collected in barrel then measured in a i beaker or graduated cylinder 3b. Provide test results for all 1000 psig PIV testing.

The test results are provided on page 13 of the enclosed Sargent &

Lundy Calculation No. 01ME114, Rev. 2 (Exhibit A).

3c, Provide Appendix J results for PIV's that are tested.

The test results are provided in the enclosed Sargent & Lundy

. Calculation No. 01ME114, Rev. 2 (Exhibit A) .

h

- - - w, .- .- . . - - - - . . , , - , - , - , . - - - , , , - , - , ~ . - - . . . - . - _ . . . - . - - . , - . . . - , . - . - - - . . . - .

Enclosure 1

'

• to U-601167

• 'Co:pericnn of Proccuro Icalcticn Tocto Cole. No. 01ME114 with Typ] SCe Toating Rov. 2 Dnto 4-/hff ,

Safety Related Page / of Illinois Power Company Prepared by c. - e 4 - //- 8 6 {

Clinton Power Station Reviewed by 8187& #4-//88 Proj. Ho. 7685 Approved by A.WA L C 4.//-7F

/ Vf OBJECT:

This calculation will use published methods to correlate the amount of leakage found in Pressure Isolation Valve Tests (PIV) with yater at 1000 psi to Appendix J Type "C" Tests for Containment Isolation Valves (CIV) with air at 9. 0 psi. This correlation will then be compared to leakages measured by both methods on valves in the plant. If the correlation provides conservative predictions of the CIV test results, this correlation will be used to predict the CIV test results for the ECCS injection check valves and the RCIC injection check valve based on the PIV testing which has already been performed. These calculated leakages will be compared to the remaining margin between the sum of the measured Type B and C tests and the allowed sum of type B and C tests. We v111 determine whether or not the addition of these leakages would raise the sum of the Type B and C above the allowable.

METHOD:

The standard correlation between velocity squared and pressure drop is assumed. The flow coefficient "K" however is not assumed to be constant under all flow conditionc. Instead it is assumed to be a function of both Reynolds number and the flow area to passage length ratio. It is found experimentally and presented in "Lyon's Valve Designer's Handbook", reference 1. By assuming that "K" varies with geometry and Reynolds number, the differences in fluid, fluid velocity, passage size, and passage chape are considered.

In determining a correlation, we first consider the geometry of the valve seats. We can model a leakage path in one of two ways. The first is to consider a scratch in the seat and model it as an orifice as long as the seat is wide. The diameter of the orifice would be as necessary to produce the leakage seen in the PIV test.

The second method would be to model it as a thin clearance between the seat and the disk. The length would be the seat width. The width of the clearance would be the circumference of the seat. The height of the clearance would be as required to produce the leakage measured during the PIV testing.

In determining the model to use we need to consider the past l history of the valves. If the valves had scratched seats, we l would expect the leakage to remain at the same level or increase j over time. If the leakage were occurring due to lack of tight l seating, we would expect that the leakage would be subject to

, erratic change from one test to the next. From the attached

) summary of PIV testing, Exhibit A, we can see that the valves j show low and orratic leakage during the PIV testing. Because of l

this we v111 model the valves as clearance between the seat and the dick.

f .. ..

i 1

Cccp3rican of Procouro Icciotien Toato Colc. No. 01ME114 with Typ2 'C' Tocting R:v. 1 Dato Safety Related Page 1 of Illinois Power Company Clinton Pnver Station Proj. No. 7685 After determining the shere of the passage we must determine its size. The seat diameter is assumed to be equal to the nominal pipe size for these full ported gate and check valves. The seat l width ic given by the vendor. The clearance between the disk and l the seat is calculated from the results of the PIV tests.

After the geometry and size of the passage is determined, we use  !

the same methodology for predicting the leakage that we would obtain in the CIV tests. We use the formula for compressible 1sentropic flow to determine the air leakage during the CIV test. These results are then compared to those found during actual tests on valves in the plant. This comparison will show that these methods are either accurate, conservative or -

nonconservative. In the event that the method is accurate or conservative, we can use the method to conservatively predict the leakage expected during a CIV test from the results of the current PIV test.

The method described above is based on testing with both air and water and is sensitive to the difference in the viscosity between air and water. The method above is similar to the Kozeny-Carmen methods used to predict flow in mixed bedc in that the flow is a function of flow passage geometry, viscosity and pressure drop.

Surface tension is not considered in either method. This is due to the fact that the same fluid is assumed to occupy the area leading to the passage, the passage itself and the area beyond the passage.

The assumption that surface tension does not afiect the flow rate during a leak test is then only valid where there is the same media on both sides of the valve. It would not be valid where a PIV test was performed using water upstream of the valve and venting the piping downstream of the valve. A PIV test performed in this manner may show no lenkoge of water due to the fact that there any be insufficient pressure to force the surface of the water to expand from the small area of the passage to wet a larger area of the valve dosnatream of *.he seating area. A prediction based on this type of test may conclude that there would not be any air leakage from a CIV test performed on the same valve. This might prove to be incorrect because an air test would not have an air to water interface, no pressure would be required to stretch the surface area of the interface and air leakage might result.

At Clinton, the PIV test is conducted with water on both sides of the seating area. Leakage is collected by overflow on the downstream side of the valve. Therefore, there is no air to

) water interface and no pressure is required to overcome surface tension. The CIV test is performed with air on both sides of the valves. Therefore, surfcce tension censiderations need not be included in the correlation between air and water tests at Clinton.

Cocpsricon of Proacuro Icolotion Tooto Ccic. No. 01ME114 with Typ#t 'C' Tooting Rry. 3 D7tn Safety Related Page 3 of __

Illinois Power Company Clinton Power Station Proj. No. 7685 r

REFERENCES:

1) Lyon's Valve Designer's Handbook. Pages 165 through 170 Van Nostrand Reinhold Co. ISBN: 0-442-24963-2

2) Crane Technical Paper No. 410, 17th. printing

3) XTP-OO-07 Clinton initial ILRT/LLRT (Type A end Type 0) test procedure and results

4) Clinton Procedure CPS 9861.02 results for Type C testing

5) Clinton Procedure CPS 9843.01 results for PIV testing

6) AIR 1909 CALCULATION :

We must first find the relation between passage size and the leakage rate measured during PIV testing. Such a relation is defined in reference 1. Since PIV testing is performed with water we will use the equations for incompressible flow. This section will rearrange the incompressible flow equations into a convenient form for our further use.

CALCULATE THE CLEARANCE REQUIRED TO PROVIDE THE LEAKAGE RATES MEASURED DURING THE PIV TESTING FROM REFERENCE 1, PAGE 167 FOR INCOMPRESSIBLE FLUID WITH D >> L (seat vidth)

V = KI(2geA P)/g 3 5 W = @ VnDC = n DCK C 2gef d P3

• Equation (1)

Re =

(2WC)/(A,Nge) = (2WC) / Uf Dguge ) = ( 2W ) / U7 D,bge ) where W is known Equation (2)

Re =

( 2CK I 2 (Do P 3 5, ) / ( // ( ge 3 * ) where W is unknown but C is known Equation (3)

C = Clearance between surfaces (ft.) P = Test pressure (psf)

D = Seat diameter (ft.) //= Viscosity (ib.-sec/ft')

f = Dennity (ib./ft8) W= Mass flow rate (lb/sec)

SUBSTITUTE THE VALUES FOR () , P, juand ge INTO EQUATIONS 1, 2, & 3 ABOVE.

'Cc p3rican of Prcccuro Ic21ction Tocto Cole. No. 01ME114 with Typa DC'.Tocting Rsv. 1 Dato Safety Related Page 4 of Illinois Power Company Clinton Power Station Proj. No. 7685

!P = 1000 psi = 144000 psf

= 62.4 lb/ft8

)au = 1. 00 cer.tipoise (Crane T.P. 410 A-3) or 6.72 x 10-

• 1b-sec/f t*

ge = 32.2 ft/sec*

W = lb./sec q = gpm W = .1390q W =gDCKI2ge pa P3 5 = DCXg [64.4(62.4)(1000)(144)3 * = 75573DCK Re= ( 2W) / (nj Duge ) =2W/ I (7r ) ( 6. 72 x 10-

• V32. 2 ) ( D ) 3 = 29.4W/D = 4.087q/D CK = W/75,573(D) = 1. 839 x 10- S q/D FOR A SEATING SURFACE OF 1 = 1/8' L= 1/[(8)(12)3 = 1/96 ft C/L = 96C (C/L)K e 96(1.839 x 10-5)(q/D)

(C/L)K = 1. 765 x 10- * (q/D)

THE FOLLOWING SUMMARIZES THE RESULTS OF THE PIV TESTS:

VALVE DIA. q(gpm) Re (C/L)K 1E12-F042C 12' .0044 .0180 7.78 x 10 '

1E21-FOO5 12" .00018 .000737 3.18 x 10-8 1E22-FOO4 10' .000079 .00039 1.67 x 10-8 1E51-F013 6' .00026 .00213 9.18 x 10-*

1E12-FOO9 18' .0121 .0330 1.42 x 10 '

1E12-F053A 10" .095 .466 2.01 x 10-8 1E12-F053B 10' .0095 .0466 2.01 x 10-'

THE VALUES FOR Re ARE BELOW THOSE SHOWN IN REFERENCE 1, PAGE 168, CHART I WE WILL ASSUME THAT THESE CHARTS CAN BE INTERPOLATED BETWEEN THE Re=5 LINE AND THE Re=0, K=O LINE USING LINEAR REGRESSION OF THE VALUES FOR K AT VARIOUS C/L VALUES AND Re LINES 5, 10, 20, 50, 100, & 200.

) -

f ?

Cacporiccn of Proccuro Ico10tien Tocto Ccic. Nn. 01ME114 cith Typ2 CC8 Tocting Rov. 1 D2to Safety Related Page di of Illinois Power Company Clinton Power Station Proj. No. 7685 THE FOLLOWING VALUES OF M WERE EXTRACTED FROM REFERENCE 1, PAGE 168 FOR VARIOUS VALUES OF C/L AND Re:

Re IC/L .01 .05 .10 .20 .30 .03 5 1 .020 .066 .093 .135 .150 .049 10 I .031 .100 .143 .200 .227 .071 20 1 .O75 .150 .200 .259 .302 .122 50 I .136 .231 .303 .372 .439 .193 1001 .175 .330 .407 .492 .539 .271 2001 .270 .428 .508 .580 .625 .360 USING LINEAR REGRESSION TO FIT K AS A FUNCTION OF Re FOR SACH OF THE ABOVE C/L VALUES WE FIND THE FOLLOWING:

K = b(Re)" with the square of the correlation (r* )

C/L b a r'

.01 .00691 .7155 .973

.03 .0212 .5499 991

.05 .0308 .5084 .995

.1 .0482 .4584 .991

.2 .0772 .3940 .990

.3 .0898 .3874 .978 USING THE ABOVE EQUATIONS FOR VARIOUS Re's WE HAVE:

Re IC/L O .05 .1 .2 .3 .01 .03 1

2 i O .0438 .0662 .101 .117 .0113 .0310 1 1 O .0308 .0482 .0772 .0898 .0069 .0212

.5 i O .0217 .0351 .0588 .0687 .0042 .0145

.2 i O .0136 .0230 .0409 .0481 .0022 .0087

.1 1 O .010 .0168 .0312 .0368 .0013 .0060

.05 i O .0067 .0122 .0237 .0281 .0008 .0041

.02 i O .0042 .0080 .0165 .0197 .0004 .0025

.01 1 O .0030 .0058 .0126 .0151 .0003 .0017 These are plotted on pages 6 and 7 of this calculation.

SOLVING FOR C AND X IS DONE ITERATIVELY. A VALUE OF C/L IS SELECTED, THE CORRESPONDING VALUE OF M FOR THE GIVEN Re IS FOUND. C/L AND K ARE MULTIPLIED TOGETHER TO FIND (C/L)K AND THIS IS COMPARED TO

SUMMARY

OF THE PIV TESTS. VALVE 1E12-F053A IS BEING SHOWH AS AN EXAMPLE.

M.

g,

< VS C/_ _OW Re  :

a;;;;4

' G 8 G -. "*

n  !"E= g,."lj

.04  ! E

/pe'! in $%

El 7:

"a ";

.03 I ^ $

Pc'g  ;

4 8

at 02 ' # d

/

i

} ' .I 5 $

01 ' '

7 m

, / e...os n, qe Re o2 g R*

,, g . - __

o ne . -

0_- .

• O .01 .02 .03 .04 .05 C/L c - - - - - - - - - - - - - -

1

< VS C/ _ _OW Re ,e-..e

.15 ., - - . - o o--***

? ". ; *.? ?

= 8 E" #1

- P , * "18

.125 4: 2 7 ."

@4i[M E

.1 3 "a *:

/ M-t '

/ s f-  :

8 f f 3?,  ?

os .z

=

1. de ' ' = ,: : :.

ozs l/ y.

J '

Re=*' ,,

~ '

a.=.az yl.-e

.f- - Rc = 0 '  ?

o Os ,

O .05 .1 .15 .2 .25 .3 C/L .

n Cncpariccn of Proccuro Icolatien Tocto Ccic. Ms. 01ME114 with Type 8C8 Tooting R2v. 1 Dato Safety Related Page __fk_ of Illinois Power Company -

) Clinton Power Station Prod. No. 7685 Re = .466 (C/L)K = 2.01 x 10-8 Try K (C/L)K C/L = .01 .004 4. 0 x 10-5 C/L = .006 .0025 1. 5 x 10- 8 C/L = .0071 .0028 1. 99 X 10- 5 C= (C/L)(1/96) = 7.4 x 10-8 HOTE THAT LINEAR INTERPOLATION OF THE TABLE WAS USED IN MANY CASES INSTEAD OF READING FROM THE GRAPH. FOR SMALL VALVES THIS ALLOWED FOR MORE ACCURATE RESULTS.

THE SAME PROCESS IS USED FOR ALL OTHER VALVES. THE RESULTS ARE SUMMARIZED BELOW:

VALVE Re (C/L)K C/L M C 1E12-FO42C .0180 7.78 x 10-' .0045 1.7x10-* 4.7 x 10-5 1E12-FOO9 .0330 1.42 x 10-5 .0049 2.9x10-* 5.1 x 10 5 1E12-FO53A .466 2.01 x 10-5 .0071 2.8210-8 '* * **-5 1E12-FO53B .0466 2.01 x 10 ' .0051 3.9x10-* 5.3 x 10-5 THIS FIXES THE SEAT GEOMETRY. WE WILL NOW DEVELOP THE FORMULA NEEDED FOR PREDICTING THE AIR LEAKAGE FROM THE CIV TEST BASED ON THE GEOMETRY. FROM THE EAME LOCATION IN REFERENCE 1, WE HAVE EQUATIONS FOR COMPRESSIBLE FLOW. SINCE THE VALVE SEAT IS RELATIVELY SHORT, WE WILL ASSUME ISENTROPIC FLOW. CIV TESTING IL PERFORMED WITH AIR AT AMBIENT TEMPERATURE USING AN UPSTREAM PRESSURE OF 9.0 PSI AND A DOWHSTREAM PRESSURE EQUAL TO ATMOSPHERIC. REFERENCE 1, PAGE 166 PROVIDES A FORMULA FOR CALCULATING THE AIR LEAKAGE.

W = KrtDC P. (2aeJC.) a - 'a+'a R Ts gps; qPs, P. = Upstream pressure - 23.7 psi (3412.8 psf)

P. = Downstream pressure - 14.7 psi J = Mechanical equivalent of heat - 778 ft-lb,/ BTU C. = Constant pressure heat capacity - 0.24 BTU /lb.

T. = Inlet temperature *R = 70+459=529'R R = Gas constant for air -

53.3 ft-lb lb

• F n = 1. 4

g Csup3rican of Proccura Icolction Tocto Colc. No. 01ME114 cith Typa 'C' Tocting Rov. 2 Dato Safety Related Page 9 of

~

Illinois Power Company Clinton Power Station Proj. No. 7685 5

W= (3412.8) f 2(32.2)(778;(0.24) f(14.738 * -

(14.7 3 i

e *'8 *9l 53.3 529 423.7j (23.7j (XDC) g W= 243.5KDC Re = 2W = 2(243.5)MDC = 481,000KC

1. D,vg. RD ( 1x10- 5 ) ( 32. 2 )

FOR THOSE VALVES THAT WE HAVE CALCULATED GEOMETRIES FOR:

VALVE D C Re W C/L 1E12-F042C ift. 4.7 x 10-5 22.6K .011K .0045 1E12-FOO9 1.5ft. 5.1 x 10-8 24.5K .019K .0049 1E12-F053A .833ft. 7. 4 x 10-5 35.6K .015K .0071 1E12-F053B .833ft. 5.3 x 10-8 25.5K .011X .0051 The solution for K 1s.again iterative. C/L is known, a reynolds number is guessed. K is found from the chart, reynolds number is then calculated and compared to the original guess. This iterative process is shown below for 1E12-F053A. The chart on page 6 and the table on page 5 are interpolated to arrive at these numbers.

We know C/L = .0071 Re = 35.6K Re(GUESS) K .Re(CALCULATED)

.05 .00013 .005

.01 <.0001 The chart is unreadable below this point but K is less than .0001.

The upper limit on mass flow rate can then be found. This is also true for valves 1E12-FOO9, 1E12-F042C, and 1E12-F053B.

Q= 60 sec. 13.2 ft8 1728 in8 16.39 cc W= ( 2. 24 x10- ' ) W min. Ib. ft8 in8 VALVE W lb./sec K W lb./sec Q(sccm) 1E12-F042C .011 <.0001 <1.1 x 10-' <25 1E12-FOO9 .019 <.0001 < 1. 9 x 10- ' <43 1E12-F053A .015 <.0001 <1.5 x 10-' <34 1E12-F053B .311 <.0001 <1.1 x 10-' <25 Using the above equations we vill predict the CIV leakage from the ECCS injection check valves based on the latest PIV test results.

l '

Cncp3ricon of Proccuro Icolotten Toc.tc Ccic. No. 01ME114 cith Type 'C' Testing Rev.'2 Date ,

Safety Related Page /0 of

_, Illinois Power Company Clinton Power Station Proj. No. 7685 Valve 1E51-FOO6 was PIV tested with 1.27 gpm leakage.

First find the geometries required to give the above leakages.

From page 4, Re = 4.087(q/D), C/LK = 1.766 x 10- * ( q / D ) , and C = 1/96(C/L)

For the 10" and 12" ECCS checks 1E12-F041A, B & C, 1E21-FOO6 and 1E22-FOO5 use D= 1 ft., q= .05 gpm. Since the lowest measurable leakage is ;OS gpm and these valves showed no leakage, use q = .05 gpm. Since the equations predict higher air leakage for a given

. water leakage in larger velves, use 12' rather than 10' for diameter.

Re=4.087 (.051=.204 (C/L)K = 1.766 x 10 * (.05) = 8. 83 x 10- '

1 1 TRY K = (C/L)K =

C/L = .01 2.22 x 10-2 2.22 x 10-5

.006 1.33 x 10-8 8x 10-5

.0063 1.4 x 10 3 g, e x 10.s C = 6.6 x 10-8 C/L = .0063 For RCIC valve 1E51-F066, use D= (4/12) = .333 and q = 1.27 gpm Re = 4.087 1.27 = 15.6

.333 (C/L)K = 1.766 1.22 x 10 * = 6.73 x 10 *

.333 Try C/L = K= (C/L)M=

.01 .056 5. 6 x 10- *

.011 .058 6.39 x 10 *

.012 .060 7.2 x 10

• C = 1.15 x 10-* C/L = .011 Predict the air leakage for these valves under CIV conditions, using equations from page 8:

I W = 243.5 KDC Re = 481.000 KC For ECCS valvest W = 243.5 KC = 243.5(6.6 x 10- 8 ) K = .016K

) ,

Re = 481,000(6.6 x 10-5)K = 31.7K

-,-s,--

. Cccp2rictn of Proccuro Iccicticn Tosto Ccic. N2. 01ME114 with Typs CC8 Tooting Rov. 2 Dato Safety Related Page // of

~

Illinois Power Company Clinton Power Station Proj. No. 7685 C/L = .0063 TRY Re = .02 K= .0003 Re = .008 Re = .01 K= .0002 Re = .006 The chart becomes unreadable with X < .0001 Therefore X < .0001 Use K = .0001 to find upper lenkage limit.

W= .016K = .016 x 10-* lb./sec q(sccm) = (1.6 x 10- ' ) ( 2. 24 X 10? ) = 36 seem ,

For the RCIC valva

Re = 481,000(1.15 x 10-*)K = 55.3X

, W= 243.5(.33)(1.15 x 10-*) = (9.32 x 10-8)K C/L = .011 TRY Re = .1 K= .0015 Re = .083

, Re = .05 K = .00097 Re = .054 l Re = .075 K= .00124 Re = .068

_)

Re = .07 K= .00118 Re = .065 Re = .06 K= .00108 Re = .0595 l

l W= (9.32 x 10-8)K = (9.32 x 10-8)(.00108) = 1.007 x 10-5

! q(sccm) = (2.24 x 10' ) W = L2.24 x 10' ) ( 1. 007 x 10-5) = 226 seem l

CONCLUSION:

l The method used is reasonable to provide an expected upper limit on the air leakage from a CIV test based on the results from a PIV test on the same valve. A statistically valid correlation between the predicted valve of CIV leakege (based on measured PIV i leakage) and the measured CIV leakage could not be shown for the l following reasons:

l 1) Current tight sealing of the PIV's with less than l measurement threshold leakage (0.05 gpm) result in too low Re number and flow coefficient for accurate CIV prediciton.

2) PIV and CIV tests are not performed on these valves at the same time.

1

3) The measured leakages appear to flucuate randomly '

) around the minimal measurable leakage rate. <

i 4) PIV testing measures the leakage of the seats of the l

PIV valve where CIV testing measures the leakage across I all the valves in the penetration boundary.

1

l

'Cacp rican of Proccuro Icolation Tocto Cole. No. 01ME114 with Typ2 8C* Tacting Rov. 2 Dato Safety Related Page /2 of

^~

Illinois Power Company Clinton Power Station Proj. No. 7685 j

, 1 1

The lack of trendable test results does not invalidate the I calculation it merely provides no input Anto the validity of the calculation.

The minimum leakage which would be found by this method, even if no leakage was found in the ECCS Anjection checit valves, would be 40 secm. Since these valves have been found to seal v1th less than measurable leakage under the PIV test, the minimum leakage must be assumed for them. The predicted leakage for valves 1E12-F041A, B, & C, 1E21-FOO6 and 1E22-FOO5 would be less than 40 seem each. The leakage from 1E51-F066 which leaked 1.27 gpm under the PIV test would be less than 300 secm.

In comparing these predicted leakages with the sum of the measured Type B & C test and the total allowed sum of the Type B& C tests there are certain assumptions which could be made to lessen the impact of including these predicted leakages. These include addition of only the leakages from those ECCS check valven.which would see containment pressure as a result of the one single active failure of a divisional power supply to the outboard isolation valve; and the addition of only the difference between the .

predicted leakage from the check valve and the measured leakage from the outboard containment barrier. In our comparison we will not include these assumptions but will be extra conservative and heap the predicted leakages on top of those already measured for the outside penetration barrier.

.t

' the beginning of our April, 1988 outage the total of the measured Type R t C tests is 76.26' .98 scfh or 36,460 secm. The allowable total is 222,280 secm. This leaves a margin of 185,820 secm. The total additional leakage that we would expect is 500 cccm from the 5 ECCS injection checks at 40 seem each and the 1 RCIC check at 300 scem. This provides a safety margin between the amount of leakage which would cause failure of the Type B & C tests of over 370 times that which would be predicted for the ECCS and RCIC injection check valves.

) *

+w

. . - - . . - . . . . . .-- - - - - __ w j .

s- .

EXHIBIT A ~

CURRENT TEST DATA PIV Leakage in GPM+ CIV Leakage in SCCM (DATE) (DATE) 4owm4m npparo OePM&a

<. 01 <.05 403t2O <20t5 "3 **75 1E12-FO42A (12" Gate) (6-20-86) (10-19-87) (11-15-85) (8-26-86) *

$ $ x" g *

<.01 <.05 24901201 4000:201 N a **O 25 e 1E12-F042B (12" Gate) (7-11-86) (10-28-87) (10-22-95) (8-27-86)

• g

< . 01 <2025.14 1E12-F042C (12" Gate) (8-08-86)

<. 05 (4-02-88) (10-19-85)

<2025.14 (8-15-86) $$0*no mnee en-

<.01 u na

<20t5.14 300s22.5# m 4m 1E12-F023 (4* Gate) (6-25-86) (12-06-85) (10-21-86) e@

ea ne ea

<.01 <. 05 429220 50.222 <2015.14 Ft *a 1E21-FOO5 (12" Gate) (S-09-86) (10-18-87) (11-05-85) (3-07-86) (10-19-87) @* j$

$ .Q1 <. 05 36.8t2 11612 <20t5.14

• 1E22-FOO4 (10" Gate) (&-23586) (11-01-87) (1-11-86) (4-07-86) (8-19-86) w gMi$ F-n. a

< . 01 <2O25.14 300t22.5#

1E51-FO13 (6" Gate) (6-25-86) w (12-06-85) (10-21-86) e e

<. 01 <. 05 39.312 <20t5.14 <2015.14

• 1E12-FOO8 (18" Gate) (6-05-86) (11-05-87) (10-08-85) (6-24-86) (10-02-86) $

.012 <. 05 475120 <2015.14 180 5.14 4 1E12-FOO9 (18" Gate) (6-05-86) (J1-05-87) (4-02-86) (6-24-86) (10-02-86)

.095

{

<. 05 26015.14 <2O25.14 a 1E12-FO53A (10" Gate) (6-12-86) (10-18-87) (12-09-85) (9-12-86)

< . 01 <. 05 43.2t2 <2015.14 1E12-F053B (10" Gate) (6-13-86) (10-28-87) (12-22-85) (9-11-86)

< . 01 <. 05 1E12-FO41A (Check) (6-20-86) (10-19-87) mmn

<. 01 <. 05 S$0 e- n 1E12-F041B (Check) (7-11-86) (10-28-87)

• M

< . 01 <.05 1E12-FO41C (Check) (8-07-86) (4-02-88) so[

go-e 1.27 *O 1E51-FO66 (Check) (11-14-86) I

• o $

<. 01 <.05 a n e

1E22-FOO5 (Check) (7-12-86) (11-01-87) h "

% s

<. 01

~E21-FOO6- (Check) (7-10-86)

<. 05 (10-18-87) h o

• PIV test threshold of measurement le .01 gpm for tests perforned prior to 1987 and . O5 for testa performed during and after 1987 0 Shared penetration total = 600 secm

( .,

q U-601167 Attachment,_2 In the meeting with members of the Office of Nuclear Reactor Regulation and NRC Region III conducted on March 30, 1988, Mr. M. Huber provided a list of valves for which IP was requested to provide the associated stroke times. Enclosure 2 is a list of those valves with stroke times provided.

In terms of Clinton Power Plant operation, if the determined stroke time for a particular valve exceeded the most conservative of the stroke times listed in the system design specification, ISI program, FSAR or Technical Specifications, consideration must given to declaring the valve inoperable and taking appropriate corrective action. (Clinton Power Station Inservice Inspection Manual, Appendix 5, Item 3.c).

4 t

Enclosura'2 to U-601167 .

J Page 1 of 3

+

VALVE MAXIMUM ISOLATION TIMES i

j =

1 Technical Design

{ Valve Number *FSAR (Amend. 38) /USAR** Specifications *** ISI*=** Requirement j Open Closed i

d 1E12-F008 39 53 54 52 52 53 1E12-F009 39 53 54 53 53 53 IE12-F003A # # # 135 135 135 IE12-F003B # # # 127 127 135 ,

i ISF001 240 Same 68 68 68 68 ISF002 240 Same 68 68 68 68 4 ISF004 240 Same 84 84 84 .85 IVP004A, B 240 Same 74 NA 70,69 70,69 l

i IVP005A, B 240 Same 74 NA 70,70 70,70 I IVP014A, B 240 Same 74 NA 71,70 71,70 IVP015A, B 240 Same 74 NA 70,72 70,73 IVQ004A, B 6 Same 10 6,6 6,6 Open.30 Close 6 IVQ006A, B 10 Same 16 NA NA 10 l

l IVR002A, B 10 Same 16 NA NA- 10 l

! IWO001A, B 240 Same 44 NA 42,43 .42,43 IV0002A, B 240 Same 44 NA 41,43 .41,44

IWX019 240 Same 2 NA 2 2 j IWX020 240 Same 2 NA 2 2 OMC009 240 Same 35 NA 35 35-j OMC010 240 Same 35 NA 34 34

• I

! ORA 026 7 Same NA' NA 2 NA

+ ORA 027 7 Same NA NA 2 NA ORA 028 # # # NA 2 NA l

ORA 029 # # # NA- 2- NA 1

RFP2/TBE14

, + - - - -

, ,,-r, , . - , . . .,w., , , , . , - . _ . , ,

.s

Encloruro 2 to U-601167 .

Page 2 of 3 Technical Design Valve Number *FSAR (Amend. 38)/USAR** Specifications *** ISI**** Requirement Open Closed ICC054 240 Same 89 NA 81 81 1CC071 240 Same 35 32 32 32 1CC072 240 Same 35 30 30 34 1CC073 240 Same 35 33 33 34 1CC074 240 Same 35 36 34 34 1CY017 240 Same 44 NA 43 43 IFP050 240 Same 48 NA 44 44 IFP051 240 Same 66 NA 65 66 1FP052 240 Same 87 NA 87 87 1FP053 240 Same 68 NA 65 65 1FP054 240 Same 68 NA 66 66 1FF092 240 Same 48 NA 47 47 IFC036 240 Same 59 NA 57 57 1FC037 240 Same 59 NA 58 58 11A005 67 Same 20 NA 15 45 IIA 006 67 Same 20 NA 14 45 1RE021 240 Same 16 NA 9 45 1RE022 240 Same 16 NA 5 45 1RF021 240 Same 16 NA 9 45 1RF022 240 Same 16 NA 7 45 ISA029 67 Same 16 NA 15 45 ISA03C 67 Same 16 NA 10 45 1HC001 80 112 117 112 112 112 1HG004 80 88 117 88 88 88 1HC005 80 118 117 117 117 118 1HC008 80 95 117 95 95 95 RFP2/TBE14

Enclo:ure 2 to U-601167 ,

J Page 3 of 3 Technical Design Valve Number *FSAR (Amend. 38)/USAR** Specifications *** ISI**** ' Requirement Open Closed IE12-F004A, B NA fame NA 142,144 142,144 144 IE12-F006A B # # # 114,115 114,115 115 IE12-F014A, B # # # 117,116 117,116 117 IE12-F024A, B 240 Same 117 114,117 114,117 117 IE12-F026A, B # # # 33,34 33,34 34 IEi2-F027A, B NA Same NA 91,93 91,93 100 IE12-F042A, B, C NA Same NA 29,30,30 29,30,30 30 IE12-F047A, B # # # 115,117 115,117 117 IE12-F04dA, B # # # 130,133 130,133 133 IE12-F053A, B 240 Same 65 64,63 64,63 64 1E12-F064A, B, C NA Same NA 63,63,63 63>.63,63 64 IE12-F068 # # # 117.117 117,117 117 1E12-F094A, B # # # 35 NA 36 1E12-F096 # # # 36 NA 36 1E12-F105 NA Sace NA 137 137 137 EOTES:

• This column represents those stroke times listed in the latest amendment (No. 38) to the Final Safety Analysis Report (FSAR) issued prior to receiving an Operating License. These times were provided in accordance with the guidance of Regulatory Guide 1.70, subsection 6.2.4.2.

• • This column represents those stroke times which will be listed tu the Updated Safety Analysis Report (USAR).

Changes in these times are representative of changes in design which have been evaluated for both system operation and containment integrity.

• • • The BASES for these times is provided in the Clinton Power Station Technical Specifications B3/4.6.4.

• • • • The Inservice Inspection Time, determined in accordance with the ASME Code Section XI, IWV-3413(a).

1. The valve listed is not a Primary Containment Isolation Valve (CIV).

RFP2/TBE14