ML20115B423

Forwards Response to NRC 920811 Telcon Request for Addl Info on RXE-91-002, Reactivity Anomaly Events Methodology

ML20115B423
Person / Time
Site:	Comanche Peak
Issue date:	10/12/1992
From:	William Cahill, Woodlan D TEXAS UTILITIES ELECTRIC CO. (TU ELECTRIC)
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
TXX-92484, NUDOCS 9210160051
Download: ML20115B423 (13)

Text

'

t Log # TXX-92484 File # 10010 F7 915 2

~

r

=

7UELECTRIC c:tober 12, 1992 William J. Cahill, Jr.

au r. v,a r,nue.,1 U. S. Nuclear Regulatory Commission Attn: Document Control Desk Washington, DC 20555

SUBJECT:

COMANCHE PEAK STEAM ELECTRIC STATION (CPSE!)

DOCKET NOS. 50-445 AND 50-446 RE0 VEST FOR ADDITIONAL INFORMATION ON RXE 91-002

• REACTIVITY ANOMALY EVENTS METHODOLOGY" REF:

Teleconference betweer 'Ir. Tony Attard, NRC and Mr. Steve Maier, TU Electric dated Auga' 11, 1992 Gentlemen:

In the above referenced conference call, ailditional analyses were discussed which would demonstrate the conservative niture of TV Electric's point kinetics analytical model of the Rod Ejec+, ion event.

The addit 1onal analyses have been completed and are summarized in the attachment to this letter.

If clarification regarding the additional analyses is required, please contact Mr. Steve Maier at (214) 512-8229.

i Sincerely, i

William J. Cahill Jr.

By:

D. R. Woodlan Docket Licensing Manager DNB/dnb Attachment c - Mr. J. L. Milhoan, Region IV Mr. T. A. Bergman, NrR Mr. B. E. Holian, NRR Resident Inspectors CP5ES (2) 140073 9210160051 921012 f

PDR ADOCK 0500044S

,'i I

P PDR 4m N. Olive Street L.fl. 81 Dallas, Tnas 75201 l.

t/

Attachm:nt to TXX-92484 Paga 1 of 12 TOPICAL REPORT RXE-91-002 REACTIVITY ANOMAL 7 EVENTS METHODOLOGY I

Note:

The references, figures, tables, and nomenclature quoted in this response correspond to those provided in Topical Report-RXE-91-002.

References, figures, and tables within the text'of this response that are not found within RXE-91-002 are identified by alphabetic-character, and are located'at the end of the response to this_ question.

Question Please demonstrate the conservative nature'of TU Electric's point kinetics analytical approach to the Control-Rod

~

Ejection (CRE) event by providing a comparison to the current-CPSES, Unit-1-CRE event analyses.

i Answer The TU Electric analytical methodology for the CRE event utilizes a point kinetics model to determine the core average power response.

In contrast, the Westinghouse analytical methodology for the CRE event employs.a one-dimensional (1-D) kinetics model to generate the core averige power response.. The CRE event results presented in the CPSES FSAR were gen. rated using the Westinghouse 1-D approach.

Westinghouse has previously demonstrated the conservatism of the 1-D approach with respect to a more detailed 3-D approech in-WCAP-7586, Revision 1-A (26).

In addition, comparisoris between the RETRAN-point = kinetics approach and the Westinghouse 1-D approach have concluded a

, Attachm:nt to TXX-92484 Paga 2 of 12 that the RETRAN point kinetics approach compares fasorably with the Westinghouse.1-D approach (A).

f Four CRE event scenarios are presented in'the CPSES FSAR.

A comparison analysis was performed fo..each of these cases.

The four cases analyzed are:

BOC HZP - Beginning-of-Cycle Hot Zero Power; BOC HFP - Beginning-of-Cycle Hot Full Power; EOC HZP - End-of-Cycle Hot Zero Power; and, EOC HFP - End-of-Cycle Hot Full Power, i

The important input parameters for each scenario are summarized in Table A-1.

Although the analytical results of all four cases are presented in the FSAP. plots of the results are available for only two cases - BOC HFP and EOC HZP.

The plots available for each of these two cases include the normalized p.

v._..= nuclear power history,- and - the-fuel pellet -centerline and 4

radially-averaged temperatures at the core hot spot.

It should be noted that the data points representing the FSAR curves are derived manually from the plots.

Tables A-2 and A-3 provide a comparison between the RETRAN point kinetics and FSAR 1-D event response times-for the BOC HFP and EOC HZP cases, respectively.

The calculated power histories for the BOC HFP scenario (Figure A-1) demonstrate that the peak core average fission power predicted by the 1-D and point kinetics models are nearly identical.

In eddition, the peak power for each model occurs at the same time, as shown in Table A-4.

However, two noteworthy differences are evident.

Attechm nt to TXX-92484 Pags-3 of 12 The more significant of the two differences relates to the predicted power history after the role se of the control rods following reactor trip signal (-0.56 second).

The use of different nathods for modelling the reactivity insertion caused by the control rods falling into the core after reactor trip r uses the power histories to deviate.

The methods employed by the 3-D kinetics model achieve greater accuracy for trip reactivity insertion by crediting the changes in the reactivity insertion rate as a function of

[

the axial flux shape.

The point kinetics model, on the other hand, must utilize a fixed table of reactivity versus.

time to conservatively represent the scram characteristics.

This table is based on an axial' flux distribution selected to provide a delayed insertion of the trip reactivity.

The less significant of the two differences pertains to the fact that the 1-D kinetics model tends to maintain the core power at a relatively high value longer than the point-kinetics model prior to the insertion of the scram ereactivity.w A, review of Figure A-2cindicateswthat,'ho power oM 7

deviation prior to the insertion of the scram reactivity does not substantially affect the total energy release for the BOC HFP case.

Because the fuel temperature and enthalpy responses are a direct result of the integrated energy release, it can be concluded that the difference in core power levels prior to scram reactivity insertion has little impact on the event results.

The calculated centerline and radially-averaged fuel temperature histories for the BOC HFP scenario (Figures A-3 and A-4, respectively) also exhibit the impact of the delayed trip reactivity insertion modelling employed by the e

point kinetics model.

The prolonged power generation predicted by the point kinetics model causes the fuel temperature to increase for a longer duration, and to I

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

Attachm nt to TXX-92484 Paga 4 Of 32 eventually peak at a much greator temperaturo.

This increase in fuel temperature also resalts in a greator peak radially-averaged fuel onthalpy, as indicated in Tablo A-4.

The different contarline fuol temperatures predicted by t!'a models can be attributed to two major diffaroncos in the hot spot modelling.

The first of thoso differoncos is the result of the fuel pollot model used for tho hot spot analysis.

The fuel pollet model is colected to be consistent with the computer codo used to generato the power distribution witain the fuel pollot.

The second of those differences is a consequence of the fuel pollet power d.'

tribution (PPD) chosen for uso in the hot spot analysis.

Westinghouse considers the calculated PPD to be propriotary information; thus, the specific values used by Westinghouse for the PSAR analysis are not available.

In view of the tact that the initial fuel average temperatures are nearly the same, and that the PPD selection process employed by TU Electric considers the propor controlling factors (fuol

-exp to-and-initial-onrichment), it can-bo concluded-that

>-+c--

the 1 contorlino temperaturo predicted by TU Electric is appro 'iato.

As with the BOC HF cenario, the impact of the trip reactivity insertion model is evident for the EOC HZP scenario.

Although the peak coro fission power occurs at approximately the same time (see Tablo A-3), the peak power preddnted by the point kinetics model is significantly greator than the response predicted by the 1-D kinetics model.

The difference in peak power is great enough to-reculc in a noticeable differenco in the total energy release, as shown in Figuro A-6.

However, once the initial peak subsides, the energy release for the two cases increases at about the same rato until the time of trip reactivity insertion; the point kinetics model then predicts

. Atta:hmont to TXX-92484 Pcgo 5 of 12 a greater amount of total energy release.

It can therefore be concludod that tho initial power peak 1.as the more significant impact on the HZP caso, and that the point kinetics model predicts a greater peak power than the 1-D kinetics model.

The calculated centerlino and radially-averaged fuel temperature histories for the EOC HZP sconario (Figuros A-7 and A-8, respectively) also exhibit the impact of the coro power history.

The greater initial p'ak power, as predicted by the point kinotics model, results in a more prominent jump in fuel temperaturo during the initial phaso of the event.

The impact of the delayed trip roactivity insertion modelling employed by the point kinotics model is evident as a shift in the timing of the peak temperatures.

As with the BOC HFP scenario, the increased fuel temperature also results in a greater peak radially-averaged fuel enthalpy, as indicated in Tablo A-a.

m....~.-..~ An'summarizedain* Table-A-4,-the'TU~ Electric' predicted ad responses to the FSAR ovent scenarios compare very well with the Westinghouse calculated responses.

For the two cases where plotted dats are available, the conservatism of the TU Electric methodology is quite evident.

Based on thouc results, it can bo concluded that the TU Electric Control Rod Ejection analytical methodology produces a more conservative ccre response than the Westinghouse analytical methodology.

REFERENCE A.

VEP-NFE-2-NP-A, "VEPCO 0 Valuation of the Control Rod Ejection Transient," Virginia Power Company,-Docomber 1984.

AttachnOnt to TXX-92484 Pcgs 6 of 12 Table A-1 FSAR CRE Analysis Input Parameters t

TIME IN LIFE PARAMETER BOC HEP BOC llFP EOC HEP EOC llFP Initial Power Level (t) 0 102 0

102 Ejected Control Rod Worth (pca) 700 200 900 265 Doppler Weighting Factor 2.07 1.2 3.55 1.3 Doppler Reactivity Defect (pcm)

( (1) )

( (1) )

( (1) )

(.(1)i Moderator Temperature

+5.0

+5.0

( (1) )

( (1) )

Coefficient (pcm) z.

Delayed Neutron Fraction, pgtr 0.0055 0.0055 0.0044 0.0044 Prompt Neutron Lif etime(II(10 6 seconds) 26.0 26.0 17.5 17.5 Trip Reactivity (pcm) 2000 4000 2000 4000 Time to Dashpot (seconds) 2.70 2.70 2.70 2.70 Pre-ejected Peak ~Fa'

'2.50 2.50 A

Post-ejected Peak Fo 13.0 4.43 20.0 5.41 Nur.ber of Operational RCPs(3) 2 4

2 4

(1) Westinghouse proprietary information (2) Assumed value for point kinetics input (3) Reactor Coolant Pumps

Attachm3nt to TXX-92484 Pago 7 of 12 Table A-2 BOC HFP Sequence of Events TIME, seconds EVES TU Electric FSAR Initiation of Control Rod Ejection 0.00 0.00 Power Range liigh Neutron Flux 0.05 0.05 High Setpoint Reached Peak Nuclear Power Occurs 0.

3 0.13 Control Rods Begin to Fall into Core 0.55 0.55 Peak Fuel Average Temperaturt Occurs 2.90 1.9P Table A-3 20C HEP Sequence of Events

+'

TIME, beconds EVENT TU Electric FSAR Initiatkon of Control Rod Ejection 0.00 0.00 Power Range High Neutron Flux 0.14 0.16 Low Setpoint Reached Peak Nuclear Power Occurs 0.17 0.18 Control Rods Begin to Fall into Core 0.64 0.66 Peak Fuel Average Temperature Occurs 1.46 1.51 j.

I.-.

iI-A

AttoChm3nt to TXX-92484 PCCJO 8 Of 12 Table A-4 Analysis Results Comparison e

CASE 10 PARAMETER TU ELECTRIC FSAR m-PEAK FISSION POWER 7.42 NOT REPORTED (FRACTION OF NOMINAL)

PEAK FUEL CENTERLINE 3725 3561 TEMPERATURE, F

BOC HZP PEAF. PUEL AVERAGE 3246 3056 TEMPERATURE, F PEAR FUEL AVERAGE 143 127 I

ENTilALPY, cal /gm

% PINS IN DNB 52.9 510

% CORE HELT 0

0 PEAK FISSION POWER 1.59 1.590)

(FF1.CTION OF NOMINAL)

PEAR FUEL CENTERLINE 4835 4570

^

BOC HFP PEAK FUEL AVERAGE 3643

'383 J

TEMPERATURE, F

PEAK FUEL AVERACE 169 143 ENTilALPY, cal /cm t PINS IN DNB

$0.01

$10

% CORE MELT 0

C PEAK FISSION POWER 64.7 30.003

~(FRACTION OF NOMINAL)

-i PEAK FUEL CENTERLINE 4238 3718 T

A RE, F

EOC !!ZP PEAK FUCL AVERACE 3807 3296 TEMPERATURE, F

PEAK FUEL AVERAGF, 174 139 ENTilALPY, cal /gm

% PINS IN DNB 57.3

$10

% CORE MELT 0

0

.s PEAK FISSION POWER 2.34 NOT REPORTED (FRACTION OF NOMINAL)

PEAR FUEL CENTERLINE 4844 4642 A

F EOC HFP PEAK FUEL AVERAGE 3851 3458 TEMPERATURE, F

PEAK FUEL AVERAGE 184 147 ENTilALPY, cal /gm

% PINS IN DNB

$0.02

$10

% CORE MELT 0.028 0

m Estimated from FSAR plot

AttcchOnt to TXX-92484 Pags 9 of 12 Figure A-1 BOC HFP Core Average Pission Power Response POWER (traction)

I'0 "

- RETRAN

- PSAR 1.0 - 7%,,,,,,,

1.4 -

1.2 -

I 1.0 -

e 0.8 -

0.0 -

0.4 -

0.2 -

.*..**......s,e,.,-.,,,

+

0.0 0

1 2

3 4

6 TIME (seconds) rigure A-2 BOC HrP Energy Release Response 4

INTEGRATED FUIL POWEft SECONDS 5.0

- RETRAN

- ISAR 4.0 -

"~

p................

.0 2.0 -

,/

1.0 -

/'

/

p'

/

0.0 0

1 2

3 4

5 TIME (seconds)

AttachOnt to TXX-92484 Pago 10 of 12 s

Figure A-3 BOC llFP Fuel Centerline Temperature Response TEMPERATURE (F)

MELTING FOINT 4500 -

e e

4000 -

3500-

- RETRAN

' TSAR

{

3000 0

1 2

3 4

5 TlWE (seconds)

Figure A-4 BOC llFP Average Puel Temperaturo Rosponse TEMPEllAn!RE (F)

MELTING POINT 4500 -

4000 -

C 3500 -

/ g, 3000 -

/

/

/

2500-

- RETRAN

• FSAR 2000 i

i 0

1 2

3 4

5 TIME (seconds)

Attachmont to TXX-92484 PacJo 11 of 12 Figure A-$

EOC HEP Coro Averago rission Power Responso POWEH (fraction) 100y

~

- LETRAN

• FSAR 10 3 i

OO 1 5 o

( v..,

0.1g i

~

0.01 I

0 1

2 3

4 TIME (seconds)

Figure A-6 EOC HZP Energy Release Response INTEGRATED FULL POWER SECONDS 2.0

- RETRAN FSAR 1.5 -

.s 1.0 -

,,,,...=*****

.e*

/

0.S -

t 0.0

='

i 0

1 2

3 4

TIME (seconcis)

ui Attachm:nt to TXX-92484 Pago 12 of 12 Figurie A-7 EOC llEP Fuel Centerline Temperature Response TEMPERATURE (F)

Ma maPomi 4500 -

4000 -

3000 -

B 2500 -

2000 -

1500 -

1000 -

600 M

- RLTRAN TSAR 0

0 1

2 3

4 6

TIMC (seconds) rigure A-0 EOC llEP Average Fuel Temperatura Response

,5 TEMPERATURE (F)

MamG Po!W 4500 -

4000 -

X 3000 -

2500 -

2000 -

1500-1000 -

500 --

- RETRAN

' FSAR 0

0 1

2 3

4 5

TIME (seconds) 6

-