ML20086E386

Pressurizer Surge Line Transient Justification

ML20086E386
Person / Time
Site:	Comanche Peak
Issue date:	10/31/1991
From:	Tilda Liu, Strauch P, Tandon S WESTINGHOUSE ELECTRIC COMPANY, DIV OF CBS CORP.
To:
Shared Package
ML20086E382	List: ML20086E373 ML20086E386
References
NUDOCS 9112020058
Download: ML20086E386 (9)

Text

_ _ . _ _ . _ _ - .

l COMANCHE PEAK UNIT 1 PREESURIZER SURGE LINE TRANSIENT JUSTIFIC ATION October 1991 S.Tandon Ventied by: -

P. L. Strauch Venfied by: T h.

T.H.Liu WESTINGHOUSE ELECTRIC CORPORATION Nuclear and Ach/anced Technology Division P.O. Box 2728 Pittsburgh, Pennsylvania 15230 2728

@1991 Westinghouse Electric Corp.

FM073710/8SI:101

?,hf'hlIh{k?Skbb04;

<< m

EXECUTIVE

SUMMARY

In Apnl 1980, Westinghouse completed WCAP 12248 to resolve NRC Bulletin 8811 which addressed the issue of thermal stratification in the Pressurizer Surgo Line for Comancho Peak Unit 1. Subsequent to this report, the NRC required Texas Utilities (T.U.) Electnc to continue monitonng of the surge line for one additional fuel cycle to ensure that the analysis was conservative. T.U. E:ectnc contracted Westinghouse to review the data from the fuel cycle and eva!uate whether the onginal analysis was conservative.

Based on this review it is concluded that the onginal analysis was very conservative. It is estimated that approximately one seventh as many stratification cycles are expected to occur dunng the life of the plant as assumed in the original analysis Consequently, no further monitoring of this line is necessary.

FM073710491:10-2

^

1.0 INTRODUCTION

in Apnl 1989, Westinghouse completed WCAP 12248 (1) to resolve NRC Bulletin 8811 for Comanche Peak Unit 1. This bulletin addressed the issue of thermal stratification in the pressurizer surge line. It required that the surge line meet the stress and fatigue critena defined by Section lli of the ASME Boiler and Pressure Vessel Code considering the effects of thermal stratification. Per the WCAP 12248 analysis, all Section ill criteria

{

was met for Comanche Peak Unit 1 (1). J h

When the original fatigue analysis was performed in WCAP 12248, stratification j transients were developed to simulate thermal cycling effects on the line. These .

transients were developed for normal, upset and heatup/cooldown conditions using:

(1) - The most recent systems standard design transients.

(2) Heatup and cooldown curves, as developed from review of procedures, operational data, and operator experience.

(3) Historical records of characteristic heatup and cooldown temperatures.

Heatups and cooldowns are characterized by the maximum temperature difference between the pressurizer liquid space and the reactor coolar;t loop hot leg during the heatup/cooldown period.

(4) Transients as monitored at five plants (excluding Comanche Peak),

Subsequent to completing this analysis, Bulletin 8811 was addressed on a genenc basis through a program with the Westinghouse Owners Group (WOG)[2]. Monitoring and operational data from 11 plants was utilized in this effort.

Per the reference (3) letter the NRC required T.U. Electric to confirm that the existing design transients utilized in the plant specific evaluation are conservative.

The purpose of this report is to review Comanche Peak monitoring data received for the period of March 1990 to July 1991 to ensure that the WCAP 12248 transients are conservative with respect to: 1. fatigue cycling and 2. tempprature profile loadings in

, the surge line. Based on this analysis, a conclusion will be drawn regarding the l necessity for further monitoring of the line.

FM073710/8/91:10-3

i

' 2.0 METHODOLOGY i'

in order to objectively compare the expected cyclic activity based on plant monitoring data to the WCAP 12248 transients, a transient set is created based solely on the Comanche Peak monitoring data using the WOG program [2] methodology. This Comanche Peak transient set is then compared to the WCAP 12248 analyzed transients to ensure conservatism.

Using the Reference [2] methodology in creating this transient set is beneficial as the analysis approach used in this report has been accepted by the NRC [4).

2.1 Normal and Uoset Transients The normal and upset transient set was created in WCAP 12248 using the latest Westinghouse systems design standard. As this standard had not changed between the -

time that the Comanche Peak 1 report was issued and the time that the WOG Generic Detailed analysis report was issued, this transient set remained unchanged in the Reference [2] analysis.

2.2 Heatuo and Cooldown Transients As desenbed in Reference [2), heatup/cooldown transients are developed using a ,

conservative historical system delta T distnbution (system delta T = pressurizer liquid space temperature reactor coolant loop (rc!) temperature) along with a Relative Strength of Stratification (RSS) distribution per plant operational mode over one heatup where:

TTp r Pipe ~ IBottom of Pipe RSS =

System Delta T (See Figure 1) l The heatup used in the analysis occurred between November 12 to 15,1990. This particular heatup was chosen because it contained the most stratification cycles. ' This is the case because after entering into mode 3, the plant returned to mode 4 during this heatup. As indicated in Reference [2), in general, the most cyclic activity is expected in the higher modes of operation. By comparison to other heatups, this was the case for Comanche Peak. Consequently, using this hectup results in a conservative assumption for the expected number of cycles per heatup in the analysis.

In this comparative analysis, a RSS distribution is created using Comanche Peak ,

monitoring data at the critical location along the line. The critical location is the FM073710/8,91:104

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

4 monitoring location with the greatest cyclic activity, and was determined to be location T 1. T 1 is located approximately halfway up the angled riser adjacent to the hot leg noule (see Figure 1). This t distnbution is combined with the historical system delta T distnbution developed in the WOG l analysis. The WOG system delta T distnbution is used because it best predicts the expected system delta T distnbution over the life of the plant, as it encompasses a number of operating practices from l within the industry.

t 3,0 RESULTS in reviewing the Comanche Peak monitored heatup, only seven stratification cycles were discovereo at the criticallocation. This is much less than the 36.5 cycles per heatup assumed in the WCAP.

12248 analysis. The RSS distnbution for the Comanche Peak heatup is as follows:

RSS Ranges (Max Values)

Mode 1 M M M Q10 Q15 Q10 M Q&Q 015 010 QA5 QAQ Q.15 (

5 0 0 0 2 0 0 0 0 0 0 1 0 0 0 4 0 0 0 0 0 0 0 0 0 0 1 0 0 0 3 0 0 0 1 0 0 0 0 0 0 0 0 0 0

~

2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 ,

(Recall from Reference [2] that the RSS listed is the maximum value of the range. For example. a cycle with an RSS value of 0.67 would be listed under 0.70),

Also, as indicated in Reference [2), it was determined that cooldowns contained less than half tre number of cycles as did heatups. This was confirmed in the review of the Comanche Peak monitoring data. Therefore, the number of cycles in all transients developed for heatups were multiplied by 1.5 to reflect the total for both heatup and cooldown, in order to compare the transients created here with those used in the WCAP 12248 analysis, seven delta T groups are used. All analyzed heatup/cooldown cycles are added into the higher delta T value group. The extrapolated number of cycles for the life of the plant (based on monitoring data) compared to those used in the Reference [1] analysis is depicted on Figure 2 and listed below:

l

FM073710/8/91

10 5

Do!!a 3 Analyzed Predicted 320'F 54 0 300'F 213 0 275'F 569 243 250'F 1262 360 200'F 741 0 175'F 5960 120 150'F 5857 1377 Total 14656 2100 There are approximately (14656/2100) = 7 times as many cycles used in the analysis as predicted based on monitoring data.

Based on the geometry of the surge line, this high level of conservatism is expected.

This line has a 45' entrance angle from the hot leg. Since stratification occurs as a result of insurges or outsurges to the line (see references [1] and [2)), a relatively high amount of fluid mixing is expected in the Comanche Peak surge line. This is due to the change in fluid direction induced by the angled portion of the line. This high amount of mixing tends to reduce the number of stratification cycles in the line. Since the original WCAP 12248 analysis used data from five different lines, the amount of cycling, on average, was greater than expected for Comanche Peak. This results in very conservative transients for Comanche Peak.

Additionally, the axial stratification profile was evaluated at severalinstances at which the high pipe delta T occurred during heatup and cooldown. All observed axial stratification profiles were determined to be less severe ttan the profile used in the original analysis.

4.0 CONCLUSION

S Based on a review of transients developed from monitoring data, the Reference (1)

WCAP 12248 analysis used a very conservative set of transients in calculating the fatigue usage factor for the Comanche Peak surge line. This high level of conservatism is depicted in Figure 2. Approximately seven times as many transients were assumed in the original analysis. Additionally, the axial stratification profile observed in the monitoring data was determined to be less severe than the profile used in the original l analysis.

FM073710,11SI:104 l

l _ __ _

All evaluations in the original WCAP 12248 analysis and conclusions reached remain valid. Based on the results of this comparative analysis, no additional surge line monitoring is necessary for Comanche Peak. This is consistent with the Reference (3) statement that the temperature monitoring program can be terminated at the first refueling outage.

5.0 REFERENCES

1. Westinghouse WCAP 12248," Evaluation of Thermal Stratification for the Comanche Peak Unit 1 Surge Line," Dated Apnl 1989.

2.

Westinghouse WCAP 12639, ' Westinghouse Owners Group Pressurizer Surgo Line Thormal Stratification Generic Detailed Analysis Program MUHP 1092 Summary Repor1," Dated June 1990.

3.

T.U. Electric Letter on NRCB 8811 " Comanche Peak Steam Electric Station (CPSES) Docket No s 50 445 and 50 446 NRC Bulletin 8811: Pressurizer Surge Line Thermal Stratification and Leak Before Break AdditionalInformation," Dated 12/12/89.

4. Nuclear Regulatory Commission Letter " Audit of Westinghouse on Generic Detailed Analysis of Pressunzer Surgo Line Thermal Stratification (NRC Bulletin 8811),' Shea to Walsh, Dated 8/6/91.

FM073710,1SI:10-7 1

)

'M FRE55URI:EF

• i WQ A

/ .,%

w ame . M :  !

se f 'f P

-rT  : ",

s g

Tl i

/~*'

.: e

( 77 -

s j~,/

/,

's ..s t,

i c' e' s,.

n.

M40 r,,,;

i Lint..nr_ <

t',,,a,L .L,... ) .sc

/ / l I

/ [h ,-

.Th. , '

[

?

/

180' '/ !E * / r/w\ *. ,

/ \, \

\

,\

N#/ OV. '

-r j

II,I4 I2,I3 FIGURE 1 COMANCHE PEAK UNIT 1 PRESSURIZER SURGE LINE MONITORING LOCATIONS l

FM073710/8SI:10-8

t 4

Comanche Peak Unit 1 Analyzed vs Predicted Heatup/Cooldown Transients 7,000 -

m m ~

! O 6,000 -

Gy ,

m0 :0:-

u.. F+C (D O F , V>

O o 5,000 i>9e; Oyl

, >C\1 .,

v,e,,.

vf -

O g 4,000 e.v4 ....).

sv

~w v:A

O v.

0,!

ve4 ve, u

(D y_) 3,000 :9:

M; ;O::

KL O 'jg l:::[

Eo3 O '"!

~

C 2 000 -

$2 4 l=$I l Z '\1 vec >v.s y w w v.;

.v -

f (D *g*s w e, l > 1,000 -

l- -

v4 ;O l O G7 - z  ;::p

.:1

~

0 '---f

/

1. "43! t

^I 93 y~# ,

I g~} P'k".

"i us . -

"~"' ; " * '

m:::3:s

~

1. ^'

320 300 275 250 200 175 150 l-l Stratification Delta Temperature (deg F)

~~

):p: Analyzed Predicted V8m (per WCAP-12248) (based on monitoring data)

Figure 2

<. ,