U-600319, Forwards Response to Request for Addl Info Re SER (NUREG-0854) Confirmatory Issue 71 on Humphrey Concerns. Responses Complete & Should Resolve Outstanding Concerns

Forwards Response to Request for Addl Info Re SER (NUREG-0854) Confirmatory Issue 71 on Humphrey Concerns. Responses Complete & Should Resolve Outstanding Concerns

ML20137G877
Person / Time
Site:	Clinton
Issue date:	11/25/1985
From:	Spangenberg F ILLINOIS POWER CO.
To:	Butler W Office of Nuclear Reactor Regulation
References
RTR-NUREG-0854, RTR-NUREG-854 U-600319, NUDOCS 8512020267
Download: ML20137G877 (22)

Text

U-600319 L30 P 2 3-85 ( 11- 2 5)-L lA.120

/LLINOIS POWER COMPANY IP CLINtoN PC%tR ST AlloN. P.O Scu 678. CLINfoN tL LINols 61727 November 25, 1985 Docket No. 50-461 Director of Nuclear Reactor Regulation Attention: Mr. W. R. Butler, Chief Licensing Branch No. 2 Division of Licensing U.S. Nuclear Regulatory Commission Washington, DC 20555

Subject:

Clinton Power Station Safety Evaluation Report Confirmatory Issue No. 71 - llumphrey Concerns

Dear Mr. Butler:

The purpose of this letter is to respond to a request for additional information concerning the Clinton Power Station (CPS) Safety Evaluation Report (NUREG-0854), Confirmatory Issue No. 71 - llumphrey Concerns. Enclosed are CPS responses to the information that was requested for NRC Staff review. We believe that these responses will resolve the particular concerns involved. With these responses, Illinois Power considers the information requested by the Staff to close Confirmatory Issue No. 71 to be complete.

If you have any questions concerning this information, please contact me.

Sincerely yours, f>

F. A. Sp. zenb4 g Manager-L ensi ,; and Safety LRH/cke Attachment cc B. L. Siegel, NRC Clinton Licensing Project Manager NRC Resident Office Regional Administrator, Region III, USNRC Illinois Departenent of Nuclear Safety 0512020267 051125 1 PDR ADOCK 0300 Bei

'l e o

i Attechment to U-600319 r L30 P 23-85 (11-25)-L Additional Information Regarding the Humphrey Concerns This attachment provides responses to informal questions and comments i provided by the NRC Staff and Brookhaven National Laboratory (BNL) regarding CPS SER Confirmatory Issue No. 71 - Humphrey Concerns. The following is a list of the Humphroy Concerns and the associated ,

Containment Issues owners Group Action Plan (AP) numbers which are addressed in this attachment:

Humphrey Concern 1.0 (AP 1) - Encroachment Issue (Generic)

L l Humphrey Concern 1.4 (AP 2) - Encroachment Issue (Plant Specific) i 1

Humphrey Concern 2.1 (AP $) - Safety Relief Valve (SRV) l q

Resonance (Plant Specific) 4 Humphrey Concern 2.2 & 2.3 (AP 5) - SRV Discharge Line (DL) Loads t (Plant Specific)

Humphrey Concern 3.1 (AP 6) - RHR SRVDL Lateral load -

(Plant Specific) ,

t Humphrey Concern 3.4 (AP 8) - RHR Steam Condensing Mode Use (Plant Specific) )

t l

Humphrey Concern 3.6 (AP 9) - RHR Steam Condensing Mode Termination i (Plant Specific) i Humphrey Concern 3.7 (AP 6) - RCIC Turbine Discharge Line Load i (Generic)

Humphrey Concern 8.4 (AP 27) - Containment Ultimate Negative Pressure l Capability (Plant Specific) i

Additional Humphrey Related Concern - SRVDL Thrust Loads  ;

^

i (Plant Specific) 1 i

i i

1 1 l

U-600319 L30 i P23-85 ( 11- 25)-L NRC/BNL Comments Regarding Humphrey Concern 1.0 (AP 1) - Encroachment Issue This concern was primarily related to the effect of encroachments on pool swell loads. As a result of the 1/10-scale test program, it has '

now been established that, for the most part, the effect of encroachments is to decrease pool swell velocities relative to an unencroached or clean suppression pool. A detailed report describing the basis for this conclusion has aircady been provided to the NRC by i BNL. Thus, so long as a given Mark III containment is designed to accommodate the loads associated with the pool swell velocities encountered in e clean pool, this concern is resolved. Unfortunately, it would now appear that the containments may not have this capability.

This is because pool swell velocities significantly in excess of design values (50 fps) can be inferred in certain regions of the wetwell in the unencroached version of these 1/10-scale tests. Again, a detailed discussion of this finding is provided by BNL.

Response

This comment has been addressed generically in a letter from G. W.

Smith, Chairman of the Containment Issues Owners Group, to 11. R. Denton, NRC, Reference CWS-0G-157, dated September 10, 1985. These results apply to CPS.

i l

1 q

U-600319 L30 P23-85 ( 1 k- 25)-L NRC/BNL Comment Regarding Humphrey Concern No. 1.4 (AP 2) - Encroachment Issue Film of the 1/10-acs c tests of the Clinton Encroachment 11 indicate the presence of a 4-5 foot wide water ligament near the containment wall i which achieves velocities greater than those without encroachment.

Provide information to demonstrate that structures located near this encroached region below the HCU floor are capable of withstanding these increased impact and drag loads.

i Response i To address this comment, the applicability of the 1/10-scale tests to obtain absolute full scale velocities and the test specific phenomena of the Clinton Encroachment II tests are discussed below. This discussion i will show that the water column velocity observed in the Clinton Encroachment II tests is not representative of the expected pool response for a postulated Design Basis Accident (DBA), and the present ,

Mark 111 design velocity of 50 fps bounds the expected encroached water i column velocity.

It was determined that the 1/10-scale pool swell velocities were at least 30% higher than the water column velocity in the cican pool. Vent capacitance effects were identified to be the main contributor to the high velocities. Therefore, the 1/10 scale absolute velocity values are not appropriate for design. However, the relative 1/10-scale water column velocities between the clean and encroached pool are represencative of full-scale velocities and can be used to determine the relative response for the Clinton Encroachment II. I The design velocity for the pool swell of 50 fps (NUREC-0978) was  !

developed from an envelope of the peak surface velocities measured during the Mark III 1/3 area scale Pool Swell Test Facility (PSTF) tests converted to full scale. The peak velocity occurred near the center of the pool over the top of the rising bubble. The 1/3-scale velocities i above the bubble were 24% higher than the velocities measured at the containment wall. Therefore, the maximum cican pool containment wall i velocity is 24% lower than the Mark III design basis velocity.

In the 1/10-scale tests, the velocities measured immediately next to the wall during Clinton Encroachment 11 tests were 21% higher than the clean l pool containment wall velocities, as illustrated in Figure 1,4-1.  !

However, as noted above the Mark 111 design velocity is more than 24% ,

higher than the value of the clean pool surface velocities adjacent to  !

the containment wall. This comparison shows that the relatively high wall velocities measured during the Clinton Encroachment II tests are still lower than the design velocity of 50 fps.

Two further items should be considered in evaluating the 1/10-scale ,

Clinton Encroachment 11 test data. As pointed out, the apparent i measured velocity of the encroached pool water column at the containment  :

vall is higher than that measured in the unencroached or clean pool  ;

region. However, as shown in Figure 1.4-1, this is only true next to  :

the containment wall. Additionally, the averaged velocity of the water j column in the clean pool is greater than the average velocity of the Clinton Encroachment II.

3 1

w_

- - . __- _ .. . ._ .. - _ - _ . - - _ - . . _ - - _ - . = _ - _ - -

U-600319 L30 i P23-85 ( 11- 25)-L  !

Finally, the apparent water column velocity next to the containment wall [

is not a true measure of the water partial vertical velocity. The I

change in the shape of the water column surface profile with elevation  !

observed on film indicates that a major contributor to the surface velocity next to the containment wall is runoff from the crest of the water column as the surface of the water column rises and flattens.

Therefore, the true vertical velocity component of the water next to the containment wall is less than the measured velocity. t Based on the above discussion, it is concluded that it would be inappropriate to use data from the Clinton Encroachment 11 testa j directly to obtain absolute pool surface velocities for design evaluations. l l

i I

L I

f i

h t

4 l

U-600319 L30 l'2 3-85 ( l L. 2 5)-L N

N pool 20 -- N CI'"

l 1

C N

Q N

1. oi, t%

,0,i4c

- N

$ city s

~

D D

t 3 10 "

0 l

i 1

0 Thickness of w.1ter column j

l (4-5 feet)

Distinen from containment wall l

Figure 1.4-1 Water Column Railin! Velocity I'rof tien for the CI's i Encroachment II Tents anil the Clenn l'ool Tent.

l l

l

! b i

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

U-600319 .

L30  !

l P23-85( 1 L. 25) L l l

I NRC/BNL Comment Regarding j Humphrev Concern No. 2.1 (AP 5) - SRV Resonance l

, l The approach described does not account for possible resonance between l:

sleeve annulus condensation oscillation (CO) frequency and sleeve acoustic frequency as suggested by Mr. Humphrey. These could Icad to I pressure loads in excess of those developed via the CE scaling laws. '

Additional justification needs to be supplied to demonstrate the adequacy of the proposed loads. l l

Response ,

i The expressed comment snus that the methodology used to address this j concern (Reference 1); did not account for a possible resonance between  ;

the sleeve annulus C0' frequency and the niceve acoustic frequency. An l l alternative approach for estimating the safety reitef valve discharge l

! line (SRVDL) sleeve CO load has been used, which conservatively utilizes i Mark 1 Full Scale Test Facility (FSTF) CO data in which significant I excitation of the vent acoustic modes was observed (References 2 and 3). [

l An estimate of the CO load in the SRVDL sleeve for CPS has been made

• using Mark 1 FSTF data (Table 2.1-1) which include significant  ;

excitation of the acoustic modes in the vent upstream of the discharge.  ;

This was done to address concerns regarding resonant amplification '

resulting from coupling of the sleeve acoustic frequency and the CO j frequency. The amplified response of the SRVDL sleeve CO load for the j dryvell vall and the containment vall shown in Table 2.1-1 can be compared to the response spectrum values in Figures 2.1-1 and 2.1-2. l The FSTF dets have been conservatively applied without any amplitude  !

reduction. The Amplified Response Spectra of the resulting SRVDL sleeve C0 load combined with the main vent CO load are lower than other DBA I LOCA design loads (chugging and pool owell), as shown in Figures 2.1-1 ,

and 2.1-2. Therefore, these loads are bounded by the current design. i References

1. Mississippi Power and Light (MP&L) letter No. AECM 82/574 dated ,

Dcctukir 3.1982 f rom L. F. Dale (MP&L) to li. R. Denton (NRC).

2. " Mark 1 Containment Program Full-Scale Test Program Final Report".

NFDE-24539-P, April 1979.

3. " Hark 1 Containment Program Load Definition Report". NEDO-21888  !

Revision 2. November 1981.  !

i s ,

i I

6 i

i

i i

TABLE 2.1-1 4

SAFETT RELIEF VALVE DISCHARCE LINE SLEEYE CONDENSATION OSCILLATION LOADS FOR CPS BASED ON MARK I IDAD DEFINITION REPORT l

i Pressure Harmonic Amplitude (PSI) Amplified Response (PSI)

Frequency Frequency SRYDL Sleeve Drywell Containment Drywell Containment Mode Range (Hz) Annulus Vall Wall Wall Wall 1 24 to 48 3.6 1.73 0.22 43.2 5.4 2 48 to 96 1.3 0.62 0.08 15.6 2.0 i l 3 72 to 144 0.3 0.14 0.02 3.6 0.5 i

w j r

i

• c r* c: i Y

N L3 9 4 p O c* I O

l M

. - e f

", i

'r e-i l

1 I

I l  !

! , I '

?" 8 - .

CC 3U s _~y $ L.

8 0

Ol 1 Cl 1

t n1 e

V1 _

n _

iI ,

a M

^

O _

C

- l .

e l v a e e e o W

) ' l i c ~

0 l S S l e 1 1

l -

v e R + l e Z w e y

[A i

V l H r g M l S D M n '.

I i

n R

~

g ~ o T g ~

T C _

u C n h a/ N o E

P C /

a/r ~ E U i s

S ll ~ r

% D a E / 7 f E p R

S / m N }

/

F o O ( C .

y P /

S S ,

E R R  :

A -

D 0

(

E 1

1 I

F 1 l_

t 2

P '

e r

M f \)

u g

\ i F

l e -

0 1

, 0 O -

R I E -

y q[% g u Bt- ,

ii l l  !.  ; i! l,  ;

i4

l l

\

RWLIFIED MSPONSE SPECTRUM (ARS) 12 r

i,

'O n  % , , s Pool 9 Swell 9 '

2 7

My + Sleeve CO 6

I E y s

-~ ~~ ~~

, sleeve ':0 L J

' B _

t

_ \

3 (

Main vent to .

o I 1111! 3C?

uom 10

• 10 8 10

• 10 '*k 8 FREQUENCY ( HZ ) x S r

i Figure ,1 7 - ARS Cor parison on Containment Wall b

6 4

4 U-600319 -

4 L30 P23-85(11-25)-L l

l NRC/BNL Comments Regarding i

]

Humphrev Concern No. 2.2 and 2.3 (AP 5) - SRVDL Loads

] (a) Provide a detailed description of how the 28 kip sleeve and SRV  !

j discharge line lateral load during chugging is derived.

2 i (b) Chugging loads tend to be random in direction. Is the lateral

)j chugging load applied in a worst caso direction in terms of the i

i

! stress levels induced on the discharge line, sleeve and any L supporting structural members?

i (c) If the lateral chugging load derives from the Mark 11 methodology,  ;

,' the region of application (1 to 4 feet f rom the vent end) is not

]

appropriate. This region of application needs to be scaled with

sleeve diameter.

i (d) There is also a certain ambiguity associated with the region of I i application. Thus, one interpretation is that the load on the SRV i discharge line is applied 1 to 4 feet below the vent end while for [

the sleeve it is applied 1 to 4 feet above the vent end. In this t case the centroids of the distributed loads would be 5 feet apart  ;

which is unrealistic. The only other possible interpretation is )

that the load is applied to both structures 1 to 4 feet above the  ;

J vent end implying that the chug load on the SRV discharge line occurs well inside the sleeve annulus - again not very realistic, i i Clarify just where the loads are applied and how sensitive the l structural response is to the point of application. j

]

i (e) The submittal does not indicate specificatinn of a lateral load l t during the C0 phase of the blowdown. Justify this omission or i

! provide an appropriate load specification for review.  !

i 1

j Responne f a) The analysis to provide the lateral loads for the Sat.".et. and nieeve for CPS is performed by using Mark 11 data and scaling to the outer  ;

diamater of the SRVDL sleeve. No credit is taken for the presence j of the SRVD1. in the bubble, providing an extremely conservativo  :

loading. Iha scaling base is the Mark 11 chugging lateral load  :

j specified in NUREG-0808 and given in Equation 1 below. l E

I 1

! F = 65000 sin tf t for 0 t .003 (1)  !

! .003  !

I f I i i where t = time (sec)  !

t F = lateral load (1bf) 1 f

t >

10 I

l U-600319 i L30 P 23-85(11 -25 )-L l l \

This load specification was based on low mass flow and cold pool  !

condition. Crosskraftwerko Mannholm (CKM) 11 test data was fitted by I the NRC as in Equation 2. l i

Ni (2) pgg) , ,-F/a , N2xN3 where P(f) = exceedance probability of the load F F = lateral load c' = empirical constant  ;

N1 = # of exceedances N2 = # of pool chugs

! N3 = # of downcomers N2xN3 = total # of individual chugs.

The Mark 11 load was based on 100 downcomers to the pool. Since there  ;

are only 16 SRV's in the CPS design, there is expected to be only 16% of the total number of individual chuge for CPS. When the reduced number of chuge is figured into the Mark 11 load, the peak force reduces from 65.000 lbf to 55,000 lbf.

The revised Mark 11 load is scaled from Mark 11 24" vents to the CPS SRVDL sleeve outside diameter of 16". The scaling is performed using the scaling rotation of Equation 3.

f_l, , '!)Q' (3) r2 ,0 2, where F1, F2 = lateral load Dt. D2 = pipe diameter n a empirical factor t

l A compilation of the 4T statistical average data from the Mark 11 Pressure Suppression Test Program (references 1, 2, and 3) results in an exponent of n = 1.7. Using the exponent with the outer diameter of the sleeve in Equation 3, the peak forco decreased to 28 kips.

The resultant loading equation in Equation 4 F = 28,000 min _wt (1bf) for 04t <.003 (4)

.00J l

11 l_ _ _ _ _ _ - - - _ - _ _ _ _ _ _ _ - -

U-600319 L30 P23-85 ( 11- 25)-L References

1. McIntyre. T. R., et al. " Mark 11 Pressure Suppression Test Program". Ceneral Electric Proprietary Report # NEDE-13442-P-01 May, 1976.

2. McIntyre. T. R., " Mark 11 Pressure Suppression Test Program Phase 11 and III Tests". General Electric Proprietary Report NEDE-13468-P. October,1976.

3. Byrd. P. K. et al. "4T Condensation Oscillation Test Program Final l Test Report". General Electric Proprietary Report # NEDE-24811-P.

May, 1980.

i t

I i

1 l

?

r i

i i

i V

b i

12 ,

i U-600319 i L30 P23-85( 11- 25)-L [

b) Chugging loads are applied perpendicular to the axis of the pipe  !

and sleeve in the direction that in the most critical for the  !

member under investigation.  ;

c) The CPS SRVDL sleeve chugging latern1 load application length han been redefined. The magnitude of the load previously specified is i unchanged. f The application length has been reduced from the Mark 11 values (1 [

to 4 feet from the end of the downcomer) previously specified. The [

revised application length was determined by reducing the Mark 11 l values by the ratio of the SRVDL sleeve diameter to the Mark 11 l downconer diameter. This scaling approach results in peak  !

pressures on the SRVDL sleeve (determined from the lateral load and j application length) which are comparabic to those obtained from the j reference Mark 11 lateral load and application length.  ;

)

The chugging lateral load is distributed uniformly on the SRVDL and  ;

the SRVDL sleeve over the application length of 0.7 to 2.6 feet for i CPS. This length is from the end of the SRVDL alcove in the [

wetvell.  !

d) The chugging load is applied perpendicular to the member axis. The l load on the SRV sleeve is applied over 0.7 to 2.6 feet from the end j of the SRV sleeve. The load on the SRV line is applied over 0.7 to  !

2.6 feet of the SRV line below the SRV sleeve (Figure 2.2-1). l

)

e) A sketch of the SRVDL aleeve is shown in Figure 2.2-2 for Grand l Culf and in Figure 2.2-3 for CPS. For the Grand Culf SRVDL sleeve l configuration, there is an unbalanced loaded area at the discharge  !

end of the sleeve, which may introduce a dynamic lateral loading.  !

For the CPS sleeve geometry this unbalanced area is not present, i therefore there is no CO lateral load. f i

1 I

i l

l l

[

1 l

l t

(' 13 j,

U-600319 L30 P23-85( 11- 2 9-L I

l

/

/

/

/ .,

( DRWELL WALL

,/ s

/

/ '

/

/s s s

s \

/ \ \

l / \ \

g \

\ g) SRV SLEEVE s s

/[/

/ \

s -

% c APPL.Y CllUCC!NG 1,0AD

[// <[

ON Tile OUTSIDE OF SLEEVE IN Til!S AREA

/

/

/ APPL.Y CllVCGING !.0AD

/ ON Tile OUTS!DE OF Tile

/ SRV I.!NE IN Til!S AREA NOT TO SCALE Figure 2.2 1 Chugging 1.nad Applicalion Area for the CPS SRVDL and !Ilcovo 14

r

. U-600319 L30 P23-85 (11 -25 )-L l

e

. . . I

~

h. . ' . s .g . . ~

. ; ?. ,f '~

q., * -

.ip, N 'E a ,,

N . . . t1

.N s*. .. g _.- .

.N s-

' ** . N

' . . ,, a . g g N

n..# .

c

',**,N NN N i s'-

'O,,.',  :' NNs

,,s g sj. . .

L r.

s't ,

s i A

! Figure 2.2-2 Sketch of the Grand Gulf SRVDL Sleeve l

l o.

N . , $s. ' '..'k

, e=' ,i.

's

, N, , a , ,. .. f

% . e' -l ,

,. ty

\ ' '/

j- s

- .. { -* .

.. . . s ,

$d # *f i ,\ \ s

.t #.,,..;,,g .

s 'e ' . ; ., ,,, r g\ N

... . * * . , N

'L ', ,

n

),

Figure 2.2-3 Sketch of the CPS SRVDL Sleevo 15

U-600319 L30 P23-85 ( ll- 25)-L NEC/BNL Comments Regarding Humphrey Concern No. 3.1 (AP 6) - RHR SRVDL Lateral Load (a) Provide a more detailed description of how the lateral load on the RHR SRVDL is derived. In particular, justify the non-exceedance probability level employed relative to the number of chug impulses that will be experienced by this structure during the life of the plant.

(b) The region of application of the lateral load (along the final 1 to 4 feet) is not appropriate since it has not been scaled f or pipe diameter. We expect that such scaling will result in a loading that increases bending moment. Comment on whether the structure can accommodate such an increase.

Response

a) The RHR SRV discharge pipe chugging load is based on the Mark II single downcomer desi2 n load given in NUREG-0808. This load has a magnitude of 65 kips and a duration of 0.003 seconds. 5 * *

• accepted as having a non-exceedance probability of 10- for the 265 chugs expected for the Mark II LOCA event.

The magnitude of this load is reduced to account for the loading area difference between the^24-inch downcomer for which the load is defined and the 12-inch RHR discharge line diameter. This reduction is based on the ratio of the two diameters as described in MP&L letter AECM-82/574 from L. F. Dale to H. R. Denton dated December 3, 1982. Therefore, the NUREG-0808 single downcomer load, as applied to the RHR SRV discharge line is:

F = (12/24) 65,000 sin ?r t (lbf) for 0 (t (0.003 0.003 F = 32,500 sin Tft (lbf) for 0 4t 40.003 0.003 where F is the time history of the force applied to the RHR SRVDL and t is time in seconds, b) The stresses in the RHR piping and supports were assessed with the load applied uniformly over the final 0.5.to 2 feet of the pipe.

Should the area of application be changed due to scaling, the margin in the current design would allow the accommodation of this load.

4 16 T - - - - - - - - - - - - - - - -

U-600319 L30 P23-85 ( 11-25)-L NRC/BNL Comments Regarding Humphrey Concern No. 3.4 (AP 8) - RHR Steam Condensing Mode Use d

In item 8.1 of AP 8.0, it is stated that "The steam condensing mode (SCM) may also be used for heat rejection in the post-LOCA period". It is BNL's understanding that this SCM is not permitted pcst-LOCA.

Applicant needs to clarify the stated position.

, Response For post-LOCA, the RHR SCM is isolated and the associated RHR logic precludes the initiation of the SCM. Therefore, the SCM is not permitted in the post-LOCA period.

l.

l l

l 17

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

U-600319 L30 P23-85 (ll-25)-L NRC/BNL Comment Regarding.

Humphrey Concern No. 3.6 (AP 9) - RHR Steam Condensing Mode Termination IPC's response to this concern differs from all the other plants in that they do not adopt the generic approach. This involved use of a thermal

deposition model which shows that, if steaming begins at normal l suppression pool temperatures (95'F), they remain well below values that i would induce steam condensation instabilities (130'F) for an extended

time period (~ 20 minutes) . This would appear to allow sufficient time l to terminate operation in the SCM. IPC claims that SCM can be  !

! terminated within two minutes following a failure of the RHR heat exchanger SRV. In BNL's judgment, there is no need to terminate that quickl) if SCM begin at normal pool temperattire. However, it may be

, necessary to terminate that rapidly if SCM is used post-LOCA, which

+

brings us back to the question we have posed for Humphrey Concern No. 3.4.

Response

For post-LOCA, the RHR SCM is isolated and the associated RHR logic I precludes the initiation of the SCM. Therefore, the SCM is not permitted in the post-LOCA period.

i I

a 4

4 4

i i

18

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

, U-600319

L30 P23-85(11-25)-L NRC/BNL Comment Regarding j Humphrey Concern No. 3.7 (AP 6) - RCIC Turbine Discharge Line Load Action Plan 6.0 does not address this concern. Provide appropriate response to this concern as it relates to the RCIC turbine exhaust

] discharge line.

! Response i Results have been submitted generically in a letter from G. W. Smith, Chairman of the Containment Issues Owners Group, to H. R. Denton, NRC, Reference GWS-0G-143, dated April 23, 1985. These results are applicable to CPS.

4 i

1 2

R g

D b

d z

.i 19

. . _ . _ . . ._ . - - _ - _ _ . _ _ _ . . .~ _ _ __ ._ _ _ . _

U-600319 L30 P23-85 ( 11- 25)-L NRC/BNL Comment Regarding Humphrey Cor.cern No. 8.4 (AP 27) -

Containment Ultimate Negative Pressure Capability Define the negative pressure capability of the Clinton Power Station

- containment.

Response

The CPS containment was evaluated for both the design and ultimate negative pressures. The negative design pressure is -3 psid. The ultimate negative pressure is -11 psid. This ultimate negative pressure is the pressure at which the steel liner would separate from the concrete containment shell. The containment shell would still maintain

- its structural integrity.

4 f

i i

l, 1

(

s 1

20 4

- .+ , - , , , , , , - - - ,. -.,-.--,-,n ...,~n .

T U-600319 L30 P23-85 (11-25 )-L NRC/BNL Comment Regarding An Additional Humphrev Related Concern - SRVDL Thrust Loads Upper pool dump increases the length of the water column within the main steam SRVDL. This will tend to increase pipe thrust loads during SRV actuation. Provide an analysis of this effect.

Response

The upper pool dump has the effect of increasing the suppression pool depth approximately by two feet. The actuation of the upper pool dump is designed to take place after the simultaneous occurrence of a LOCA signal (high drywell pressure or low reactor water level) and a low-low suppression pool water level. Therefore, during the normal accident sequence the dump of the upper pool has the effect of restoring the suppression pool water level and is not a concern.

Inadvertent actuation of the upper pool dump would lead to an increased level of the suppression pool and SRV discharge line submergence of approximately 2 feet. (The water leg would increase by approximately 2.83 feet due to the slope of the line). All SRV discharge lines have been designed for a subsequent actuation water level of 6 feet above the normal suppression pool water level. Therefore, the upper pool dump level increase is within the design consideration for the SRV lines.

l 21 l

l