LD-85-029, Forwards Results of CESSAR Analysis of Increased Blowdown for Pressurizer Safety Valves,Per Reactor Sys Branch Request.Analysis Consistent W/Analysis Submitted by Arizona Public Svc

Forwards Results of CESSAR Analysis of Increased Blowdown for Pressurizer Safety Valves,Per Reactor Sys Branch Request.Analysis Consistent W/Analysis Submitted by Arizona Public Svc

ML20126H042
Person / Time
Site:	05000470
Issue date:	06/04/1985
From:	Scherer A ABB COMBUSTION ENGINEERING NUCLEAR FUEL (FORMERLY
To:	Thompson H Office of Nuclear Reactor Regulation
References
LD-85-029, LD-85-29, NUDOCS 8506100253
Download: ML20126H042 (11)

Text

'

. w -- ..

C-E Power Systems Tel. 203/688-1911 Cornbustion Engineenng. Inc. Telex: 99297 1000 Prospect Hill Road Windsor, Connecticut 06095 M POWER SYSTEMS-STN 50-470F June 4, 1985 LD-85-029 Hugh L. Thompson, Director Division of Licensing U.S. Nuclear Regulatory Commission Washington, D.C. 20555

Subject:

System;80" Safety Valves - Increased Blowdown

References:

(1) NRC Letter, G. W. Knighton to E. E. Van Brunt, Jr. (Arizona Public Service), dated March 12, 1985 (2) Arizona Public Service Letter ANPP-32381, E. E. Van Brunt, Jr., to G. W. Knighton, dated April 12, 1985 (3) Arizona Public Service Letter ANPP-32401, E. E. Van Brunt, Jr., to G. W. Knighton, dated April 15, 1985

Dear Mr. Thompson:

The NRC Staff recently requested additional information from Arizona Public Service [ Reference (1)] on the effect of increased blowdown settings for pressurizer safety valves (PSVs) which resulted from the EPRI safety valve test program in 1981. The Staff's primary concern was that increased blowdown (delayed closure). could result in high pressurizer water levels during transients and two-phase discharge through the PSVs. Responses to NRC questions [ Reference (2)] showed that the pressurizer water level would not increase to the elevation of the PSV nozzles.

In subsequent conversations with the Reactor Systems Branch reviewer, C-E was asked to address the issue of increased PSV blowdown the on CESSAR docket. In addition, C-E was requested to provide information on the effect of increased blowdown for the secondary-side main steam safety valves (MSSVs).

The attachment to this letter provides results of the CESSAR analysis of increased PSV blowdown. This analysis is consistent with the analysis submitted by Arizona Public Service for the Palo Verde Nuclear Ge.erating Station (PVNGS). The CESSAR analysis demonstrated that the maximum water level for the most limiting transient is well below the elevation of the PSV nozzles. There is no concern, therefore, regarding two-phase discharge through the PSVs.

8506100253 850604 6 PDR ADOCK 05000470 A PDR i[!

/

1

Mr. Hugh L. Thompson LD-85-029 June 4, 1985 Page 2 C-E has also reviewed the issue of increased blowdown for the CESSAR MSSVs.

Due to the fact that increasing MSSV blowdown changes neither the valve's capacity nor opening setpoint, the maximum secondary pressure of any CESSAR event will remain unchanged. Increasing the MSSV blowdown can, however, affect secondary steam releases and the resulting radiological doses. For the safety analyses in CESSAR, the effect of increased MSSV blowdown on radiological doses was evaluated and it was found that only the steam generator tube ruptu re (SGTR) events would have measurable changes in the consequences. For the SGTR with loss of power and with a stuck open atmospheric dump valve, the increased MSSV blowdown setting (18.5%) was used for the analysis already in CESSAR and, therefore, the doses remain unchanged. The safety analysis conclusions in CESSAR, therefore, remain valid.

If you have any questions or comments concerning the above information, please contact me or Mr. S. E. Ritterbusch of my staff at (203) 285-5206.

Very truly yours, COMBUSTION ENGINEERING, INC.

in A. 6 Scherer Director Nuclear Licensing AES:las Attachment

m - -

o 1

i LD-85-029-Attachment Page 1 of 9

]

. ANALYSIS OF INCREASED BLOWDOWN FOR' PRESSURIZER SAFETY VALVES I

Fulliscale', full -pressure prototypical _ testing of pressurizer safety valves was performed (by ~ EPRI Lin -1981 - (C-E Owner's Group Report CEN-227, December _1982).

Results showed that the blowdown settings required to insure stable valve

. operation during iblowdown = from - the set pressure were above the 5% setting

. specified Ein the ASME Code. In ' order to insure that the extended blowdown would ~not adversely, affect . plant performance, anal ses were performed to.

evaluate Lthe 'NSSS responses. _The analyses demonstrated that a blowdown setting

of ~ 18.5% . is. acceptable.

These analyses _ included a review of overpressure;

~

protection ~(reported 'in CESSAR-F appendix SA)'and a review of the: safety

' analysis in Chapter 15 of.CESSAR-F.

in CESSA'R~-Section 15.2 provides results for those events that fall in the-Decreased -Heat --Removal category and result in the greatest increase 'in

. pressurizer pressure, pressurizer water level, and RCS temperature. Of these

, , (transients,. the Feedwater Line Break (FWLB) event with the loss: of offsite

power is the . most adverse when maximizing pressurizer _ level. This f act 'is confirmed by close examination' of the pressurizer water volume response of the

~.C hapter 15 transients and by reviewing the response to the NRC's request _for additional. information on the increased blowdown for the Palo Verde pressurizer o

. safety _ . val ves ' (PSVs ).

t IS f

LD 029 Attachment Page 2 of 9 To determine. the maximum pressurizer water level of the CESSAR FWLB, the initial conditions of the event were adjusted in the following manner:

-(a) The initial pressurizer pressure was raised from 2250 psia to the maximum Technical Specification pressurizer pressure of 2370 psia. Although this

' change causes an earlier reactor trip on high pressurizer pressure, and therefore, less pressurizer insurge, the maximum pressurizer level is potentially greater due to the larger volume of flashed steam as pressurizer pressure decreases during tne extended blowdown.

-(b) The initial assumed pressurizer water volume is increased from 900 ft 3 to

- 998 ' ft3 , which is the maximum CESSAR Technical Specification value. This assures that the maximum level of the transient is conservatively predicted.

.(c) The blowdown setting of the PSVs is increased from the nominal value of 13.5% (closure at pressures below 2160 psia) to 18.5% (closure at pressures below 2040 psia). The maximum ' pressurizer level of the transient is increased primarily due to the fact that a larger volume of flashed steam will be generated during the increased blowdown.

(d) The initial RCS cold leg temperature is increased from 565 F to 570*F, which is the maximum Technical Specification value. While this change does not significantly increase the pressurizer water volume, it does reduce the RCS subcooling margin, which is of interest when concerned with RCS voiding.

The other significant initial conditions for the CESSAR FWLB case to maximizethe pressurizer water volume are summarized in Table 1.

u

, .c LD-85 029 Attachment-Page 3 of 9 The . limiting FWLB case was simulated with the CESEC III code and the results are discussed below. Figu re 1 provides the pressurizer water volume for the transient. The solid line indicates the CESEC-predicted pressurizer level response and shows a maximum value of 1213 ft 3. Figure 2 provides the RCS pressure response as a function of time for this transient. Figure 3 provides

-the RCS subcooled margin - and Figure 4 provides the minimum DNBR for the transient. As demonstrated by these figures, there is adequate subcooled margin available to prevent RCS voiding and the minimum DNBR for the transient remains well above 1.19.

The CESEC code assumes that the subcooled insurge is mixed homogeneously with the water volume in the pressurizer. The pressurizer water volume is assumed to be saturated liquid at the initiation of the event. This mixing potentially underpredicts the volume of water that will flash due to the additional depressurization caused by the increased PSV blowdown. Secondly, the code assumes instantaneous phase separation of any voids formed during the depressurization and consequently may underpredict the pressurizer level response.

To conservatively account for these two phenomena, the CESEC pressurizer level results presented in Figure 1 were adjusted. The maximum adjusted value is also shown in Figure 1. Although some mixing will occur, the adjustment assumed that none of the subcooled insurge entering the pressurizer during the

~

transient mixed with the ~ 998 ft3 of saturated liquid (at 2370 psia) in the pressurizer at the initiation of the event. This maximized the potential for this initial volume to flash during the depressurization caused by the increased blowdown, maximizing the pressurizer level response.

g LD 029 Attachment Page 4 of 9 The additional' volume added to the CESEC predicted response to account for phase separation is calculated in the following manner:

As . the pressurizer pressure drops below its initial value of 2370 psia, the volume initially in the pressurizer is assumed to begin to flash. Thereafter, in discrete time intervals during the depressurization to 2040 psia (18.5% blowdown), an average void fraction is calculated. Once this average void fraction is calculated, a linear bubble gradient is assumed over the volume

initially present in the pressurizer. This is reasonable since the static pressure will be highest near the bottom and the bubbles tend to coalesce ~ as they- rise through the mixture. (This method is analogous to what the RELAP and RETRAN simulation codes utilize.)

Coupled with this assumed void distribution, a conservatively low bubble rise velocity of 0.5 ft/sec is assumed. This velocity is based on the Wilson bubble velocity model (Wilson, et al., Velocity of Rising Steam in a Bubbling Two-Phase Mixture, ANS Transactions , 5, 151, 1962) using an average void fraction which is small ' compared to the void fraction expected in the pressurizer. A more realistic value of the bubble rise velocity is approximately 2.0 ft/sec.

This conservative adjustment to the limiting FWLB increases the CESEC predicted maximum level response by 306 ft3 (from 1213 ft3 to 1519 ft 3). Even so, the pressurizerf level is 12% below the elevation of the bottom of the PSV nozzle for the limiting CESSAR case. (This corresponds to a distance greater than 4 ft. below the nozzle.) Because of this large margin to the PSV nozzle, water carryover will not occur for this or any other event and, therefore, no concern exists regarding two-phase safety valve discharge.

)

_- a

LD 029 Attachment Page 5 of 9 TABLE'1 INITIAL CONDITIONS FOR THE

. LIMITING FWLB WITH RESPECT TO MAXIMIZING PRESSURIZER WATER LEVEL Nominal Plant Operating Conditions Analysis Value Power, MWT - 3817 '3893 Cold Leg ~

565 570 Temp eratu re, . 'F Pressurizer- 2250 2370 Pressure, psia PSV Opening' 2500 2525 Setpoint, psia High PZR~ Pressure 2400 2475-

Trip'Setpoint, psia PSV Closing 2160 2040

?Setpoint, psia Pre 900 998

~Ft.gsurizerLevel,

~

PZR Pressure Automatic Manual Control System PZR Level ~ Automatic Manual Control System Steam Generator 173000 107500 Inventory, lbm Feedwater. Pipe- ----

0.2 L Break Area, Ft.2 Moderator Temper- ---- Most Positive Technical ature Coefficient Specification Value

i. . Steam. Bypass Automatic Manual

. Control System

.0ffsite Elect. Power Available Unavailable After Turbine Trip

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

LD 029 I Attachment Page 6 of 9 FIGURE 1 PRESSURIZER LEVEL RESPONSE FOR THE LIMITIflG FWLB TRANSIENT WITH LOP 1800 i i BOTTOM OF PSV N0ZZLE - 1741 FT3 (99.4%)

1600 1519 FT3 (87%)

(ADJUSTED) m 1400 C-1213 FT3 (70%)

y 1200 -

5 G

/

FWLB + LOP

$ 1000 E

E m

u a- -

800 600

=

400 O 10 20 30 40 50 TINE, SECONDS

LD 029

. Attachment Page 7 of 9 FIGURE 2 RCS PRESSURE RESPONSE FOR THE LIMITING FWLB TRANSIENT WITH LOP 3100 i i i i

.2900 -

2700 -

E -

, 2500 d

E E -

v3 2300 M

2100 1900 1700 0 10 20 30 40 50 TIME, SECONDS

...e LD-85-029 Attachment Page 8 of 9 FIGURE 3 RCS SUBC00 LING MARGIN FOR THE LIMITIflG FWLB TRAflSIENT WITf! LOP I I I I 60 -

g. 50 -

T--

52 n= -

i 40 -

Li w

C e

5 30 s

8 sw -

y 20 -

10 0

O 10 20 30 40 50 TIME, SEC0iiDS l

1 i_ 4

LD-85-029 Attachment Page 9 of 9 FIGURE 4 MINIMUM DNBR RESPONSE  ;

FOR THE LIMITING FWLB + LOP TRANSIENT 1

i 4.0 i i i i

3.0 -

E E

E m 2.

5 x

N = 1,31 _ .

1.0 -

0,0 O 10 20 30 40 50 TIME, SECONDS l