ML20212G003

Rev 2 to OPGP-04-ZA-0002, Condition Rept Engineering Evaluation Program

ML20212G003
Person / Time
Site:	South Texas
Issue date:	10/21/1997
From:	Kersey R, Starks V HOUSTON LIGHTING & POWER CO.
To:
Shared Package
ML20212F334	List: ML20212F328 ML20212G003
References
NUDOCS 9711050212
Download: ML20212G003 (7)

Text

_

P:ge i sf_~7

. CR 97-14434 6 Eoptr** tins R* Port OPGPO4-ZA-0002 Rev. 2 Page 17 of17 Condition Report Engineering Evaluation Prvgram Fonn1 Condition Report Engineering Evaluation Fonn (Sample)

PageIofI CR No.

97-14434 Page 1 of 7 l

l.

Identify the.yy. yde CR Category

[ Engineering Action

[Er.gtneering Report

.~'.g. kagineenna Changegoclaulflod 1

. e.3 a

2.

Reason for changr.

--m

.3H Nonconfonnance EEnhancement

7.0%Adminiserstive m

u :.a

.,m.

3.

NCR Disposition:

]Use-Apis NRepair MRework s$tInvalidate E.9'nReject(NFAA Only)

(

l l

4.

aan,e r-i:35ihtsjor MOD E. E..E..

Non-Design Change w

~

.a-e g33 Minor MOD 5:. Procedure Change yMinor Change gTesoporary MOD MP Per Change E Other

\\

l Aczionfrask/Sub Task /No.

Owner Dept Descripdon 97 14434 6 V. Starks DED Review RWST Vortex Breaker Design Acceptabdrty 5.

Disposition / Evaluation:

See Anached.

6.

V.J. Sanrks

(,

to-L t-1997 1.

?

m zt.py

/Deze Supervisor /Deze q r.t/97 f

a.

R.P.Kersey g Renewer /Date 9711050212 971023 yDR ADOCK 0500o49g PDR 904Td 96E8 E4A ETS D41SN331"l ht:f3"DrN es:ET /.E41-TE-L-)O

_~

W6-WM

.-.m.

Page 2 or 7 CR 9714434.c -

o Enghewkg Repet L

Description of Problem:-

Review RWST vortex breaker design a~=y Nity.

II.

Background:

(

The NRC requested ju.uTceurs, for the RWST suction nozzle operation at low water levels. Note that instrument uncertainty was critiotted by the NRC as not in socordance with industry standards so Reference 2 was lasued to demonstrate RWST level uncertainty in MC5037 was conservative.

IN.

Analysis:

INPUT instrument Span = 31.2 ft.

(Ref.1).

Tank radius = 27.0 ft.

(Ref. 7)

Swuoverinitiation setpoint = 11.0%

(Ref.1)

Level of tank outlet = 0.0% (11.375" a ve bottom of tank)(Ref.7)

(Note that scaling sheets, p' ant operating procedures, and the instrument read out sets the actual bottom of the tank as 0.0% level, and adjusts setpoints socordingly.)

ID of outlet pipe = 23.0"(24" X strong short radius elbow) (Ref. 6) instrument error of one train = 2.8g%

(Ref. 2)

CALCUl.ATION gal /ft = z(27.0). 7.4805 gal /ft' = 17132 gal /ft.

gal /% = 31.20 ft

• 17132 gal /ft /100% = 5345 gal /%

ft/% = 31.2 ft /1'0% = 0.312 ft/%

PER MC05037 Method, with updated instrument error:

Lowest point of initiating switchover = 11.0% - 2.89% = 8.11%

. Pumped volume during switchover = 11,100 gal.

(Ref.1) 11100 gal /5345 gal /% = 2.08%

Luwest point following switchover = 8.11%-2.08% = 8.03%

Top of voriax suppressor = 24" above bottom of tank (Ref. g) = 24-11.375 = 12.63" above outlet pipe and 12.63"above 0% level.

vortex suppressor level = (12.63/12) / 0.312 ft/% = 3.37%

Ma gin = 6.03% - 3.37% = 2.66%

The above analysis combines the single train instrument error with the switchover volume calculated using the three train flow rata, in addition, the flow rates and valve stroke times were cxmservatively

i

.high.

Maximum flow rate with 3 trains operabng:

HHSI pump = 1820 gpm (based on max.Ted Spec)

LHSI pump = 2800 gpm (based on max. Tech Spec)

Containment Spray = 1800 ppm (based on 4g71 gpm 3 train fkw from Westinghouse Calc.

o FSD/SS-TGX 842 described in Ref. 8, witn margin added to account for train differences, etc.)

Total = 6220 gpm per train

• 3 = 18660 gpm total 8048'd 8628 E24 ZTS D41Sh3D1"I kMrrrN BStEt 466(-TE.L30

~

m

Pep,,,3,M 7 cm sM44344 Note this is approximately 8 % higher than an informal best estimate analysis of 3 train flow rate.

The standard setpoird methodology is based on a 0.95 probability of actuation of a single instrument within the uncertainty band. Thus the standard method results in a 0.025 probability of actuating above the uncertainty band and a 0.025 probability of actuating below the uncertaltdy band. With 3 -

trains involved, there is a 50% chance each train will switch over below the nominal switch over setpoint. The probability of all three trains switeing over below the nominal selpoint (i.e. all 3 trains still drawing trorn the RWST st the nominal setpoint) = 0.8*0.5*0.5 = 0.125. We shalldetermine the uncertainty band corresponding to a probability of 0.025 of all 3 trains actuating below the uncertainty band.

If the probability of one train actuation below the uncertainty bond equals 0.2924, then the probability of all 3 trains actuating below the uncm1.kity band = 0.2924*0.2924*0.2924 = 0.025 Or, (0.025)" = 0.2924

~

From normal probability function tables in Ref. 3, page 585:

1-F(x) = area under the normal distribution curve to the left of x = 0.2912 @ x = 0.55 x =1 corresponds to 1 standard deviation.

l The one train instrument uncertainty of 2.89% (Ftef. 2) is based on 1.96 standard deviations, giving I

a 0.95 probability of one instrument actuating within the uncertainty band, or a 0.025 probability of one instrument actuating below the uncertainty band.

2.89%* (0.55 /1.96) = 0.81%

Thus an instrument uncertainty band of 0.81% provides a 2.5% probability that all 3 trains continue to drew from the RWST below the uncertainty band, or a 97.5% probab'lity that at least one train will initiate switchover above the bottom of the unwrtainty band, Bottom of uncertainty band for switchover point for three running trains = 11.0 - 0.81% = 10.19%

j Pumped vo!ume during 16 aeoond stroke time = 18660 gal / min *(16 sec)So sec/ min) = 4976 gal 4976 gal 15345 gal /% = 0.93%

Lowest level with all three trains running = 10.19% - 0.93% = 9.26%.

S = Submergence = 9.26%*0.312 ft/% = 2.89 ft.

V = Velocity = 0.4085*gpm/dia = 0.4035*18660 / 23.0' = 14.4 Weec l

Fr = Froude No. = V / (g*S)" = 14.4 / (32.2 ft/sec

• 2.89 ft)* = 1.5 8

Minimum switsover point for the last running train = 11.0% - 2.89% = 8.11%

Pumped volume during 16 second stroke time = 6220 gal / min *(16 secJ60 sec/ min) = 1659 gal

- 1659 gal / 5345 gal /% = 0.31%

Lowest level with one train running = 8.11% - 0.31% = T.80%

S = Submergence = 7.80%*0.312 ft/% = 2.43 ft.

8 8

V = Velocity = 0.4085*gpm/dia = 0.4085*6220 / 23.0 = 4.80 ft/sec L

Fr = Froude No. = V / (g*S)" = 4.80 / (32.2 ft/sec*

• 2.43 ft)" = 0.54 l

(.

l '

filPP9'd 9628 Z46 ZIS D415 NIDI"I 4MirtrN IS:ET 4661-TE.DO a t: -

.- m

=~

cm eM44w4 Scre 't of '7 Engheartne Report Postulate that one train does not switchover, but continues to draw from the RWST while the last effective train switches over, due to a single adive failure. In this es,se:

Minimum owitchover point for the inst running train = 11.0% - 2.8p% = 8.11%

' Pumped volume during 16 second stroke time = 12440 gal / min *(16 sec)S0 sec/ min) = 3317 gal 3317 gal / 5345 gal /% = 0.62%

Lowest level with two trains running a 8.11% - 0.62% = 7Ag%

8 = Submergence = 7.4p%*0.312 ft/% = 2.34 ft.

8 V = Velocity = 0.4085'opm/dia's 0.4085*t2440 / 23.0 = g.61 ft/sec Fr a Froude No. = V / (g*S)" = g.61/ (32.2 ft/sec ' 2.34 ft)" = 1.1 8

Refs. 4 & $ tested containment emergency sumps for air entrainment, including a wide range of vortex suppressors made of floor grate in all of the tests invoMng vortex suppressors, the presence of the vortex suppressor virtually eliminated air entrainment.

From page 23 of Ref. 4 "The strongest air drawing vortices (selected porturtied flow tests) were completely suppressed to a surface dimple by the cage type vortex suppressor tested (4 ft x 4 ft x 4 fl. cage in most cases) " The cage was made of standard 1" floor grate.

From page 22 of Ref. 5, section 3.1.a. Lge shaped vortex suppressors made of floor grating to form cubes 3 and 4 ft, of a side, and single layer horizontal floor grating over the entire sump area, worn both found to be effective in suppressing vortions and reducing sit-ingestion to zero. These suppressors were tested usih012 inch outlet pipes, and with water levels ranging from 0.5 ft. to 6.5 ft. above the top of the suppressors. _ Both the cage shaped grating suppressors as well as the horizontal floor grates were made of standard 1.5 inch floor grates."

From rege 22 of Ref. 5, section 3.1.b,

• Tests on a cage shaped suppressor less than 3 ft on a side indicated he existenes of air core vortices on the water surface for certain ranges of flows and submergence, even though air-withdrawals were found reduced to imignificant levels."

The switchover volumes calculated using the altemate calculation remain highly conservative. The flow would switch to the emergency sump line before the amergency sump isolation valve was fully cpen. In addition the Safety injection pump flow rates used are based on Tech Spec maximum pump flow rates.

The design of the RWST vortex suppressor is enveloped by the parameters included in the vortex suppressor tests described in References 4 & 5. Those references found negligible air entralnment in all of the configurations tested.

The Low Head Safety iriection (LHSI) pumps and Containment Spray pumps contain at Isast 5 stages plus an inducer stage, and the High Head Safety injection (HHSI) pumps contain 16 stages plus an inducer stage. Per figure 3 g of Ref.10, the head degnadation of a pump to a given air volume fraction is inversely rt!sted to the number of stages of the pump. With 1/No. of stages = 1/5 i

L

= 0.2, and accepting degradation to g3% of rated head (based on Westinghouse " degraded pump" curves), the LHS! & Containment Spray pumps would operate successfully with 5% to 6% sir volume fraction at the pump inlet. Since the pump elevation is more then 30 feet below the RWST outlet, the corresponding air volume fraction at the RWST outlet would be on the order of 10%. This demonstrates that air entrainment at the RWST outlet would have to be very high to effect pump performance.

i 88/S8'd OGES EJA ETS D41SND1"I WFDrN ts:ET 2461-TE-r0

{

~

~

. =

Pape_.Lof7 i

CR 97144344

+

Engineering Report l

N.

==

Conclusions:==

.The RWST. vortex supppressor design and existing switchover eetpoints provide an adequate level of water above the top of the vortex suppressor to ensure negligible air entrainment using the highly conservative flow rates and switchover times used in MC-5037 (Ref.1) combined with the instrument uncertainty values found in Ref. 2.

Additional margin can be demonstrated by using more reasonable, but still conservative, values for pump flow rates and switchover times. Use of instrument uncertainties consistent with the nember of operating trains provide additional margin in level over the top of the vortex suppressor and submergence of the tank outlet pipe.

The Safety injection and Containment Spray pumps would operate successfully with air volurre fractions on the order of the highest observed in any of the tests described in Refs.4 & 5.

V.

Actions:

No additions actions created by this CREE.

VI.

References:

1. MC05037 R7, Determination / Validation of RWST Setpoints

2. CREE 97-14434-6 Supplement 1

3. CRC STANDARD MATHEMATICAL TABLFS 22" Edition

4. NUREG / CR-2758, A Parametric Study of containment Ememenev Sumo Performance

5. NUREG / CR-2761, Results of Vorter Sunor= sear Tests. Sinole Outlet Sume Tanta and Miscellaneous Sensitivtty Testa

6. Vendor drawing 0149 01113-BM Submittal C

7. Vendor drawing 014S 01100-BM Submittal B

8. 6N10gMB1024 R1, Containment Spray System DBD

g. Vendor drawing FSC0023 00065 RU Submittal D

10. NUREG / CR-2792, An Atsanament of Rattdual Heat Removal and Containment Rnrav Pumo Perbrmance under Air and Dehns Ingesting. Conditions VII.

Figures:

1. Sketch of outlet configuration, instrument error, & switchover volume per MC-5037

2. Sketch of outlet configuration, instrument error, & switchover volume per Altemate Method t

2 SIV99'd 9628 ZuiZTS DNISNiDl"I nkrTDrN TT, '.1 466T-TE-130

=.

4

41- \\'14 5 'l~ lo CALC NO.

SHT b OF ~7 SOUTHTEXAS PA0KCT ELECTTUC GENERATWG STADON REV.

PREPARER /DATE REVIEWER /DATE HOUSTONUGHTING & POWER GENERAL COMPUTATION SHEET suSJECT UNITis McSD37 WT#d 14 II It

'E de Ms J hru SWsTLW6VG)l

$6ffonJT tl N

4 k

2.g1 % md. u.ncM.

,. 3 Q

10 b%

9

~a b

If 7o B

a b

Z.08 % SWIM *VN g

7 Vot.t4 M 8 e

3r 6

g 1

3 s

3 af

-r,r o r vet.f W s ufffE5

.I.

2.

I b

Fl444RF I alk/A *d 8628 224 ZIS ENISNiDI*1 mcTDrN TS ET JA61-TE.l.00 z.. -.

' '~, ::: ~~-

.--.7--

'~~~~~~-~~~~~'T-":~

~

... _ _ =

wa w.uu 47-iyy3 M

~

- m m n, 1

CALC NO.

SHT 7 OF_'7 SO'JTHTEXAS PROKCT ELECTRIO GENERATING STATION REY.

PREPARER /DATE REVIEWER /0 ATE HOUSTON UGKRNG A POWER j

GENERAL COMPUTATION SHEET SUBJECT UNIT /s ALTER t)hTG METAbb l4 i

il tt y

g, A* Mi J hu SW s'rt Wo VG)L $67forJT IN5'c EARoR F R 3 TRA'43 A.u Wiv4 l*

Sws rteldJed votuMr unn.5 Taks Ls RudWrokt N

~. a a waru s ruin b

i ofee ATiA/%

TA A '4 b

[ 2. sty, pgg.eR 6A I

h

.:r 8

N 7 swiTcHo#C VOLUME D

F-eA : rxA,.J oP#hrex4 g

~>

7 i8 4

6 wu 3

S'

!o>

4

--cat OT tot.YGY suPPRES e

1 I

O

.=

Ft&ltRE 2 BEV80*d 96E8 E46 ZIS CHISEDI'1 W3"DnN ES:ET 4661-TE.LCO gyneuta-w.--i

~ w ww > wg e we s-w m.,>

w

-um.--

v.%.

--v,...w..

-4w-san==

--,m w-,-e-

.