B16062, Submits Changes to & Errors in Emergency Core Cooling Sys Evaluation Model or Application of Model

Submits Changes to & Errors in Emergency Core Cooling Sys Evaluation Model or Application of Model

ML20134D178
Person / Time
Site:	Millstone
Issue date:	01/29/1997
From:	Bowling M NORTHEAST NUCLEAR ENERGY CO., NORTHEAST UTILITIES SERVICE CO.
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
B16062, NUDOCS 9702050014
Download: ML20134D178 (15)

Text

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

107 Selden Stract, Berlin, Cr 06037 M

Utilities System Northeast Utilities Smice Company P.o. Box 270 i

Hartford, CT 06141-0270 (860) 665-5000 January 29,1997 Docket No. 50-336 B16062 l

Re:10CFR50.4S(a)

U.S. Nuclear Regulatory Commission Attention: Document Control Desk '

Washington, DC20555 i

,! Gentlemen:

Millstone Nuclear Power Station, Unit No. 2 Reporting of Changes to, and Errors in, Emeroency Core Coolina System Models or Applicat,ons i

[ 1

~

in accordance with 10CFR50.46(a)(3)(ii), Northeast Nuclear Energy Company (NNECO) hereby submits changes to, and errors in, the emergency core cooling systen, (ECCS) i evaluation model or application of the model for MilNone Unit No. 2.

The last update was submitted to the NRC Stoff on March 22, 1996W. Based on notifications received from Siemens Power Corporation (SPC) dated December 22,1996 and January 6,1997, this report covers changes to, or errors in the large break loss of coolant accident (LBLOCA) and small break loss of coolant accident (SBLOCA) analyses  :

performed for Millstone Unit No. 2, since March 22,1996. The following is a synopsis of the  !

information provided in the Attachments.

1.

An error has been identified in the SPC 1986 EXEM PWR LBLOCA model modifie in 1991 to eliminate a non-physical reflood heat transfer trend in the TOODEE2 code of the model. SPC observed a non-physical trend in the approved 1986 EXEM PWR LBLOCA evaluation model heat transfer coefficients computed by the Fuel Cooling Test Facility (FCTF) correlations in the reflood rate range of approximately 1.0 in/sec to 1.77 in/sec. In this range, as the reflood rate decreases, the calculated heat transfer coefficient can under certain conditions increase. In 1991, SPC made W

F. R. Dacimo letter to U.S. Nuclear Regulatory Commission, " Millstone Nuclear Power Station, Unit No. 2-Reporting of Changes to, and Errors in, Err.ergency Core Cooling gl (i

System Models or Applications," dated March 22,1996.

3 9702050014 970129 PDR P

ADOCK 05000336 PDR ,

lf

U. S. Nucisar Regulatory Commission B16062/Page 2

j .

i changes to the FCTF reflood heat transfer correlations in the TOODEE2 code in the 3 approved 1986 EXEM PWR LBLOCA evaluation model. These changes were made j to eliminate the non-physical trend in the heat transfer coefficients as a function of

reflood rate. The NRC rejected those changes and also concluded, based on

information available to the Staff, that the FCTF correlation model includes an 2

unacceptable error not only because of the non-physical heat transfer trend but also j because of non-conservative heat transfer model predictions.

} 2. To correct the unacceptable error, SPC implemented an interim modification to the .

FCTF reflood heat transfer correlations in the TOODEE2 code in the 1986 EXEM

!- PWR LBLOCA evaluation model to eliminate the non-physical trend. The j modification restricts the reflood heat transfer coefficients to increase linearly j between the values calculated by the FCTF reflood heat transfer correlations for i

]

reflood rates of 1.0 in/sec and 1.77 in/sec. The interpolated heat transfer coefficient

{ is also restricted to be no greater than the value corresponding to the retlood rate of i 1.77 in/sec. Using this linear interpolation approach, SPC analyzed the 0.4, 0.6, 0.8,

and 1.0 double ended cold leg guillotine (DECLG) breaks with both a middle and an 3

end of cycle axial power profile. Single failures were reviewed and a bounding j analysis was performed. The resulting peak clad temperature (PCT) was 2111 F, an

l increase of 189 F from the previously calculated PCT of 1922 F. This increase in j PCT is considered significant with respect to 10CFR50.46(a)(3)(i) and, as such, is

~

l 2'

being reported as required by 10CFR50.46(a)(3)(ii). -

3. By copy of letter dated December 3,1996*, the NRC requested additional information regarding the LBLOCA model error described above and further -

requested that NNECO provide this information to the Staff at least one month prior i i to the restart of Unit 2. The requested information is as follows

,

c i

j a. Submit data to justify an LBLOCA evaluation model that is conservative for  :

heat transfer coefficients over the entire range of reflood rates during an

LBLOCA, including 1"/sec to 1.77"/sec,  ;

4

! b. Reanalyze the LBLOCA for the unit using the evaluation model proposed in j item a., and, 1

4

c. Provide the following information:

( >

l (1) Maximum linear heat generation rate ,

(2) Reflood rate as a function of time  ;

j (3) Core collapsed level as a function of time

]

• U.S. Nuclear Regulatory Commission letter to B. D. Kenyon, "10CFR50.46 Large Break

. Loss-of-Coolant Accident Evaluation Model for Millstone Nuclear Power Station, Unit No.

l 2, Request for Additional Information (TAC No. M96355)," dated December 3,1996.

i

l

U. S. Nucl:ar Regulatory Commission

B16062/Page 3 1

1 (4) Quench time as a function of core height '

(5) Core pressure as a function of time (6) Core subcooling as a function of time (7) Heat transfer coefficient for PCT location as a function of time 4

(8) Clad temperature for PCT location as a function of time in response to items a. and b., SPC reanalyzed the Millstone Unit No. 2 LBLOCA

. using an interim modification to the FCTF reflood heat transfer correlations as j described above. In response to item c., the maximum linear heat generation rate '

4 used in the reanalysis was 15.1 kw/ft. Core inlet subcooling is 132.1 F. This value of subcooling is used in the FCTF correlations throughod the reflood period. The

remaining requested information is provided in Attachmen. i in the form of plots for

' l the limiting LBLOCA. '

4. In late 1992, an error was discovered in the z-equivalent model in the TOODEE2 code. This model uses equivalent energy deposition to allow TOODEE2 to adjust between axial power shapes. The z-equivalent modei was coded with the axial shape for the FLECHT test rod instead of the FCTF test rod. At the time of error i discovery, test cases analyzing the impact of the correct coding showed that the l PCT using the incorrect FLECHT shape as the basis provided either conservative or l

insignificant (+15 F) changes in calculated PCT. The NRC was notsied of this error  !

in early 1994 as part of SPC's annual reporting for 1993. SPC reevaluated the limiting LBLOCA case using the interim analysis model, described above, with the correct z-equivalent model. The result was estimated to be a decrease in PCT of about 2 F. NNECO chooses to conservatively report this effect as a PCT benefit of 0.0 F.

5. SPC submitted a new reflood heat transfer model for NRC review on December 20, 1996. This model will correct the non-physical trend of the heat transfer coefficient as a function of reflood rate and the z-equivalent model error. It is expected that NRC review and approval will be completed within three months following submittal. 1 Subsequent to NRC approval, SPC will reanalyze the Millstone Unit 2 LBLOCA event. SPC will provide NNECO with a new LBLOCA analysis report and the changes in PCT for 10 CFR 50.46 reporting. SPC will set the completion date for this work as soon as the NRC review / approval are completed.

6. SPC identified three input errors in the SBLOCA analysis performed for Millstone Unit 2. The errors in the SBLOCA analysis are as follows:

a. Analyst failed to replace the steady-state steam generator pressure control system with the transient pressure controls when the SBLOCA transient was initiated.

i

b. The 1% uncertainty applied to the main steam safety valve (MSSV) setpoints does not support the current Technical Specification value of 3%.

1 U. S. Nucitar Regulatory Commission i

,B16062/Page 4 j i

j c. The time of scram in the decay heat tables for TOODEE2 was not updated i

for break sizes other than the limiting small break size.

The errors in the steady-state pressure control and safety valve setpoint uncertainty were discovered during SPC's ongoing activities supporting NNECO's .10 CFR l l 50.54 review efforts. The error in the scram time used for the TOODEE2 analysis

'~

was discovered when the analysis was reviewed in response to the first two' errors. j SPC reanalyzed the Millstone Unit 2 SBLOCA spectrum with the correct inpt'ts to i verify that the limiting break size was the same. The resulting PCT for the limiting  ;

small break was calculated to be 2010 F, an increase of 303 F from the previously calculated PCT of 1707 F. This increase in PCT is considered significant with respect to 10CFR50.46(a)(3)(i) and, as such, is being reported as required by .

10CFR50.46(a)(3)(ii).

7. Attachment 2 provides both the small and large break ECCS Evaluation Model  !

Margin Utilization sheets wnich account for (A) the analysis of record, (B) prior permanent 10CFR50.46 LOCA model assessments, (C) 199610CFR50.46 LOCA ,

model assessments including those for which this notification is being performed, j and (D) margin utilization attributable to 10CFR50.59 evaluations up to the present. ,

8. Considering the changes summarized in Attachment 2, the corrected PCTs for the  ;

limiting LBLOCA and SBLOCA remain below the 2200 F limit as defined by 10CFR50.46(b)(1). ,

We believe that this information satisfies the 30 day reporting requirements of '

10CFR50.46(a)(3)(ii) and fully meets the requirements for information requested in reference (2) to be supplied prior to unit restart. Based on the above information, the i Millstone Unit No. 2 FSAR will be updated as necessary. If you have any questions, please  !

contact contact Mr. Richard T. Laudenat at (860) 444-5248.

Very truly yours, NORTHEAST NUCLEAR ENERGY COMPANY i

kb M. L. Bowling d '

Recovery Officer >

Millstone Unit No. 2 ,

l l

i

U. S. Nucicer Regul tory Commission B16062/Page 5  !

1 l

l cc: W. D. Travers, Dr., Director, Special Projects Office l H. J. Miller, Region I Administrator P. F. McKee, Deputy Director of Licensing, Special Projects Office W. D. Lanning, Director, Millstone Oversight Team D. G. Mcdonald, Jr., NRC Project Manager, Millstone Unit No. 2 D. Beaulieu, Acting Senior Resident inspector, Millstone Unit No. 2 I

l I

i i

j f

l 2

i l

Docket No. 50-336

B16062 ,

l 1

I l

Attachment 1

]

^

1 4

1 l

Millstone Nuclear Power Station, Unit No. 2

Request for AdditionalInformation 4  !

1

10 CFR 50.46 Large Break Loss-of-Coolant Accident (LBLOCA) Evaluation Model with 1 1

l 3 Linear InterpolaSon of Reflood Heat Transfer Correlation j i

, Plots of: l r Figure 1: Reflood rate as a function of time l i Figure 2: Core collapsed level as a function of time i Figure 3: Quench time as a function of core height Figure 4: Core pressure as a function of time Figure 5: Heat transfer coefficient for PCT location as a function of time ,

4 Figure 6: Clad temperature for PCT location as a function of time l

I i

January 1997 l

4

1 FIGURE 1 1

Reflood Rate As A Function Of Time g

{

'a kv a 5 l

- _ g$ o 1

8 b 1 Re sm

'E W

- _ g~

- _ g

- _ s-1 1 1 I i e d 9 I I ( l [

3A (S/NI) 31'sJ 0001338 3800 3AI103333

d FIGURE 2 Core Collapsed Level As A Function Of Time

. . . . . . I.

_ _ g

_ _ g

_ _ g. a E

- - g$ o 8

Re m

- ~

m

_ _ g~

_ _ g

. _ s t I I t i I 4 9 S t C Z t I OH (.L.:1) 73A31 OlnOn 035dVT100 3800

FIGURE 3  :

Quench Time As A Function Of Core Height i

1

. . . , , , I.

d

_ _ g

]

kv l E 8

=

o d

b gg M

_ _ s3

'E 1'

_ _ g~

_ _ g

_ _ g I I t i I t vi et et e s , e B Oz (.L.d) 13A37 HON 300 3800

FIGURE 4

Core Pressure As A Function Of Time I , , , ,

i i - _ g l

i lv I

ho ai b i la<  ;

F-M

- _ BB i

'E W

l

- _ E N

- - E f

t I f I t t 99 ZG 9t tt Ot 90 Et df ZN*d (VISd) 380SS38d 1371no 3800

15.0 .... . ... ....,,...,

t m x  ;

u. - -

e '

l 12.5 - -

BL N

d

, - - B

• a w

. S l 10.0 - -

o u - .

g r

_c - - g; N -

- 9.

3 -

g) 7.5 - -

n ,

v - -

R y T 5 .

O c w - . A x ,

0 5.0 - -

E m  ;

o -

. g o, t O -

. e O

_ a H - - >

m 2.5 - -

y ,

y -

.n

( hte 5

g.

~ -

0 0.0 . ...i.... ....i....i... 1....i....i....i....:.... a 0.0 50.0 150.0 200.0 o ,

100.0 250.0 300.0 350.0 400.0 450.0 500.0 -

-i  ;

Time /s/

\

T D' e

.j. 'i : ) ,

~ao .

. C al

. d T

( lo .

e 3

) .

p m

a .

e r

a .

g n '

t u

o . r i.

. e s .

.- F d

ao -

., . o r

w .

.s -

P t

C IF 6 t, T G u . L U p

l.ll j'.;: o p c R o .

a E o eo .

.l,,l,;

t oi n

6

.  ;, . A

, s

A y/

F u

n c

ao ... ... ... . - ~. . .... . . .... .... .... ti o

6o $o go $o ko nRo gco n$o .8o ,So e n

".E9 vmou mm O f

T mi e

e-

t Docket No. 50-336 B16062 i

Attachment 2 Millstone Nuclear Power Station, Unit No. 2

.l Reporting of 10CFR50.46 Margin Utilization '

l I

l h

i 1

January 1997 4

l l

'. I U. S. Nucirr Regulatory Commission B16062/ Attachment 2/Page 1 Reporting of 10CFR50.46 Margin Utilization Small Break LOCA PLANT NAME: Millstone Unit No. 2 Clad Temperature Notes A. Analysis of Record (4/95) PCT = 1707*F (1)

Eval. Model: EXEM/PWR Small Break Vendor. Siemens Peak Linear Power 15.1 kw/ft 2

Break- 0.1 ft B. Prior Permanent LOCA Model Assessments-Thru 12/1995

1. None APCT= 0*F C. Current Permanent LOCA Model Assessments-Thru 12/1996 (Permanent Assessment of PCT Margin) i

1. SBLOCA Input Errors APCT= 303*F  !

-SG steady-state pressure control system error

-MSSV setpoint uncertainty error

-Scram time error D. 10 CFR 50.59 Safety Evaluations APCT= 0*F ANALYSIS OF RECORD PCT + MARGIN ALLOCATIONS PCT = 2010*F Notes:

(1) New Analysis of Record.

~

.I' U. S. Nucl:ar Regulatory Commission B16062/ Attachment 2/Page 2

)

i Reporting of 10CFR50.46 Margin Utilization Large Break LOCA PLANT NAME: Millstone Unit No. 2 Clad Temperature Notes l A. Analysis of Record (4/94) PCT = 1811*F Eval. Model: EXEM/PWR Large Break Vendor. Siemens Peak Linear Power.15.1 kw/ft Break: 0.6 DECLG B. Prior Permanent LOCA Model Assessments-Thn 12/1995

1. None APCT= 0*F C. Current Permanent LOCA Model Assessments-Thru 12/1996 (Permanent Assessment of PCT Margin)

1. Hydraulic Diameter Input Error APCT= 111*F

2. Non-physical Trend in Reflood Heat Transfer Correlations APCT= 189'F

3. Z-Equivalent Model Error APCT= 0*F (1)

D. 10 CFR 50.59 Safety Evaluations APCT= 0*F ANALYSIS OF RECORD PCT + MARGIN ALLOCATIONS PCT = 2111*F Notes:

(1) SPC reevaluated the limiting LBLOCA case using the interim analysis model with the correct z-equivalent model. The result was estimated to be a decrease in PCT of about 2 F. NNECO chooses to conservatively report this effect as a PCT benefit of 0.0 F.

1