ML20057C714

Forwards Suppl Info Related to PWR Reload Safety Analysis Methods to Document Salient Points of 930629 & 0921 Telcons, in Response to NRC 921008 RAI to Facilitate Review of Util Topical Rept NFSR-0069

ML20057C714
Person / Time
Site:	Byron, Braidwood, Zion  File:ZionSolutions icon.png
Issue date:	09/23/1993
From:	Simpkin T COMMONWEALTH EDISON CO.
To:	Murley T NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM), Office of Nuclear Reactor Regulation
References
TAC-M76660, TAC-M76661, NUDOCS 9309290293
Download: ML20057C714 (39)

Text

_ _.

Commonwsalth Edison C

1400 Opus Place Downers Grove, Illinois 60515 September 23,1993 Dr. Thomas E. Murley, Director Office of Nuclear Reactor Regulation U.S. Nuclear Regulatory Commission Washington, D.C. 20555 Attn:

Document Control Desk

Subject:

Zion Nuclear Power Station Units 1 and 2 Byron Nuclear Power Station Units 1 and 2 Braidwood Nuclear Power Station Units 1 and 2 Commonwealth Edison PWR Reload Safety Analysis Methods l

(TAC Nos. M76660 and M76661) l NRC Docket Numbers 50-295.304.454.455.456.457 l

References:

a)

Letter from C.P. Patel to T.J. Kovach dated October 8,1992.

b)

Letter from S. F. Stimac to Dr. T.E. Murley dated April 23,1993 l

c)

Conference call between CECO (T. Simpkin, J. Silady, et al.), NRR l

(C. Shiraki, F. Orr), and ITS (H. Komoriya) on June 29,1993.

d)

Conference call between CECO (T. Simpkin, J. Silady, et al.), NRR (C. Shiraki, F. Orr), and ITS (H. Komoriya) on September 21,1993.

Dear Dr. Murley With reference a), the NRC provided Commonwealth Edison Company (CECO) with a Request for Additional Information (RAI) to facilitate the completion of the NRC staff review of CECO Topical Report NFSR-0069. CECO responded to the RAI i reference b). As indicated in reference c) and d), subsequent conversations were held for the purpose of elaborating on several areas of the reference b) response.

The purpose of this letter is to document the salient points of the reference c) and d) conversations, and to submit for staff review the information which was verbally provided. Enclosure 1 to this letter is related to the reference c) phone call, j

which discussed the steam generator modelling and the SI delay time for the Main Steamline Break Analysis (MSLB) Enclosure 2 provides additional MSLB information pertinent to the conversations held in reference d).

i j

j k:nla: zion:raiwp:1

/

9309290293 930923 I[I i 3\\

PDR ADDCK 05000295 g, P

PDR j

i Dr. Murley September 23,1993 CECO believes that these enclosures are responsive to the Staffs inquiries and slnt ld support timely issuance of an SER approving CECO's Safety Analysis Methods fe WVRs. CECO desires to begin using the RETRAN and VIPRE methodologies in i

yport oflicensing actions at the earliest opportunity in order to reduce our reliance n the fuel and NSSS vendor's services. This will contribute both to cost reduction l

u.d improving in-house capabilities as encouraged by NRC Generic Letter 83-11.

Please direct any questions to this office Sincerely, QlsmE-d ~'

l Terrence W. Simpkin 1

1 Nuclear Licensing Administrator Enclo ures: 1 - CECO Supplemental Information 2 - CECO Supplemental Information for MSLB cc:

J. B. Martin - Regional Administrator - RIII (w/o enclosures)

C. Y. Shiraki, Senior Project Manager - NRR (2 copies)

J. B. Hickman, Senior Project Manager - NRR (w/o enclosure)

R.A. Assa, Project Manager - NRR (w/o enclosure)

J. D. Smith, Senior Resident Inspector - Zion (w/o er3 closures)

H. Peterson, Senior Resident Inspector - Byron (w/o enclosures) i l

S. Dupont, Senior Resident Inspector - Braidwood (w/o enclosures) l l

l k:nla: zion:raiwp:2

l 1

1 I

i e

~

t 1

] : Supplemental Information Zion PWR Topical RAI 2 i

i i

l i

i i

1 I

i J

Question 14 Supplemental Information REQUEST:

Provide the mass distribution for the multi volume steam generator model, including the void distnbution for the boiling region.

RESPONSE

Per your request, Figure 1 provides a plot of the total fluid mass,'FMAS', for each steam generator volume on a single plot. This figure provides a relative mass distribution and represents a significant amount of data. As a result, some detailis sacrificed due to the scaling required to include all 8 steam generator volumes. For this reamn. additional figures are being provided for clarification. Definitions of the various parameters that have been plotted are provided in the attached definition list for reference.

Figure 2 provides the steam dome (Volume 101) total fluid mass and quality. These parameters were selected to best describe a mostly steam filled volume.

I Figures 3 through 9 provide the total fluid mass and liquid mass for Volumes 102 l

through 108, respectively. By providing the total fluid mass and liquid mass for each of these volumes, the amount of vapor mass can be inferred. The reactor trip on 10-lo steam generator level occurs at approximately 28 seconds. Shortly thereafter the steam generator begins to pressurize until the main ^ steam safety valve setpoint is l

reached (t - 58 seconds). The mass in the boiling region increases shortly after the time of reactor trip as a result of the steam generator pressurization. The mass drops off after this point and then recovers as a result of Auxiliary Feedwater (AF) initiation. AF is initiated 60 seconds after reactor trip. Cold AF begins entering the steam generator at t - 586 seconds and AF flow is increased manually by the operator at t = 1828 seconds.

Figure 10 provides the total fluid mass for the three boiling region volumes. This figure allows for an easier assessment of the amount of fluid in the boiling region itself. The volume corresponding to each boiling region RETRAN volume is also provided on this figure for information. Note that the boiling region volumes are not equal.

Figure 11 provides the void fraction in each of the boiling regions. From this figure, the increase in void fraction with respect to tube bundle region elevation can be seen. This figure is also important because the heat transfer correlation associated with each boiling region volume throughout the transient can be inferred. As discussed in the response to Question 14, when using stacked homogeneous volumes to represent the tube bundle region, a lowering liquid levelis represented by an increase in void. With an increase in void, the heat transfer regime makes a transition from one heat transfer correlation to another. For example, nucleate boiling is assumed to occur until the homogeneous void fraction reaches the value of a = 0.8.

The fully developed subcooled boiling correlation of Thom is used in RETRAN for

forced convection boiling where the volume average thermodynamic properties are used to calculated the amount of heat transferred. Between a = 0.8 and a = 0.9, a linear interpolation of the heat flux between the Thom correlation and the forced convection correlation is used,if the critical heat flux is not exceeded. The forced I

convection vaporization heat transfer regime is assumed to occur if the vapor void fraction satisfies a 2 0.9 and the forced convection boiling curve is used. In contrast, the single volume steam generator model utilizes the local conditions heat transfer option. This option allows the heat conductors attached to a single volume region to respond to the appropriate fluid conditions above and below the mixture level. For heat conductors above the mixture level, heat transfer to or from saturated vapor occurs. For heat conductors below the mixture level, heat transfer is to a two phase mixture where the quality and void fraction at the conductor midpoint elevation are determined from the bubble rise model and used in the heat transfer correlations.

REQUEST:

Indicate when the tubes become uncovered for the single volume steam generator model.

l

RESPONSE

b The tubes were uncovered at the initiation of the event. The results of the single volume steam generator presented in response to Question 14 for the comparison of the single vs multi volume steam generator models were for an initial mixture level of 30.45 ft, where the top of the tubes is 34.381 ft. Mixture level sensitivity studies were performed with the mixture level above and below the tube bundle which concluded that the single volume steam generator model always remained limiting when compared to the 8 volume steam generator model.

c

DEFINITION LIST Total Fluid Mass,'FMAS',is defined as the sum of the liquid mass and the vapor mass.

Liquid Mass,'LIQM',is defined as the mass of saturated or subcooled liquid mass in a RETRAN control volume.

Average Quality,'AVEX',is defined as the static quality corresponding to the average thermodynamic state within the control volume. The quality x is determined from the following equation:

x = (v - vi)/(vg-vi)

Where:

v - average specific volume vi - saturated liquid specific volume g - saturated vapor specific volume v

Void Fraction,'VOlV', is the vapor volume fraction. The void fraction is a function of quality, vapor density and fluid density.

l 1

I l

l i

i 1

i 50000 i

.s s

E 40000 -

N

.o C

\\

c

.9 3

T g

.$ 30000 --

FMAS101 a-g

- -- - MAS 102

\\

..... FMAS103 l

m y

g

- - -. FMAS104

._.. _ FuAsim m

g FMAS12 l

g 20000 --

FMAS107 5

FMAS108 c

o

\\

O

\\

. -.. -.. ~.. -.. -...

E r 'i ;

h Y.

\\\\

N-

~~-

10000

-I~

' \\.j, y

(

-~

"*}..~.. X e

jy

[_.

- M...%.* Q. ify

,,,,.r e, -

0

,,,,,,................a.

i l

1 10 100 1000 10000 1

Time [sec)

Figure 1: Mutli Volume Steam Generator Model Mass Distribution for LONF Transient I

I I

l

'FMAS' = Total Fluid Mass (IDm) i i

l

~

r,

i 1

5 000 - - - - - - - - - - - -.

~-- -

- 1 l

0.99 E 4000

-- 0.98 <

FMAS101 m

b I:.

..... AVEX101 e

c r

m m

-- 0.97 3 m

e

E i

o

- 0.96 -'

e 3000 --

l E

M o

l o

o E

-- 0.95 m 3

m o

1 7 2000 --

j

-- 0.94 g o

i e

e

-- 0.93 oe E

.s.

E i.:.

-- 0.92 *

$ 1000 -

-- 0.91 0

'l

'"l 0.9 1

10 100 1000 10000 Time [sec]

Figure 2: Mutti Volume Steam Generator Model Steam Dome Volume Mass and Quality (Volume 101)

'FMAS' = Total Fluid Mass (Ibm) i

'AVEX' = Volume Average Quality

1 l

15000 --

E 12500 --

lI

.o c.

m

((

l:I %

,..s j:

10000 --

S E

l.i m

- i Q

g I

rj$

7500 --

FMAS102 E

uouio2 g

m E

5000 --

o t

C l

E i

l

.2 f

O

.W 2500 --

1, q&

y; 0

"l

"l' 9

l 1

10 100 1000 10000 Time (sec)

Figure 3: Mutti Volume Steam Generator Model l

Steam Separator Volume Mass (Volume 102) l

'FMAS' = Total Fluid Mass (Ibm)

'LIQM' = Liquid Mass (Ibm) l I

l I

1 l

l 5000 --

T 4000 --

l

.om t

i e

m m

E 3 3000 --

u E

i m

E ruAsios m

- - - - UQM103

?

m T 2000 --

o s,

e A*..

E

s

$ 1000 --

l f.,

• g i

3 1

i

e...

V O

t if f f 1 t il t

i 1 1 t iett t

t t t I fitI t

t t I f lit!

i 1

10 100 1000 10000 i

Time (sec)

Figure 4: Mutli Volume Steam Generator Model Steam Barrel Volume Mass (Volume 103) l l

'FMAS* = Total Fluid Mass (Ibm)

'LIQM' = Liquid Mass (Ibm) i

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

i

(

50000 --

p

$ 40000 --

m g

111E i

I g

l30000--

=c 3

-~

FMAS104 u

..... uouto4 e

$ 20000 --

a eo e

e E3 10000 --

O>

.. v%di. --- -

v r r en.

-++:h.

0 1

10 100 1000 10000 Time (sec) l Figure 5: Mutti Volume Steam Generator Model Upper Downcomer Volume Mass (Volume 104)

'FMAS' = Total Fluid Mass (ibm)

'LIQM' = Liquid Mass (Ibm) l 1

)

.]

h

i l

l i

l l

9 15000 --

f i

3 E 12500 - -

.o t

w m

E L. 10000 -..

E ou c

g Q

(;

o 7500 --

m Ast

..... uouim i

,a I

3

4, 3

, :.j '.

5000 --

m Q

\\

I r

Q

/

E L!

s x\\

o 2500 -

5' '

h,

g.'

{

-..E/

0

"l

"l

"l

"l l

1 10 100 1000 10000

)

Time (sec]

i

\\

Figure 6: Mutti Volume Steam Generator Model Lower Downcomer Volume Mass (Volume 105)

'FMAS' = Total Fluid Mass (Ibm)

'LIQM' = Liquid Mass (ibm) i

d 1

i s

t 1

5 4

i 15000 --

i i

E e 12500 --

m m

s E

i c 10000 -

.9 d

a i

?

g 2:

1 e

i c

4 5

7500 -

Fwsim O-uouim m

1 e

3 0

J 5000 --

4 j

E,-

........H..

J

....... n..-

1 g

.2 2500 --

O

.\\g%..._.. M 0

"l

"l

"l

"l 1

10 100 1000 10000 Time [sec)

Figure 7: Mutti Volume Steam Generator Model Lower Boiling Region Volume Mass (Volume 106)

'FMAS' = Total Fluid Mass (Ibm)

'LIQM' = Liquid Mass (Ibm) d M

e acw---

e

.s w w< 3sy,

-,*g-es'-.c ne my--

- - -w--

r--*eg.

w-

i 1

15000 --

)

E a 12500 - -

m m

t 3

C 10000 --

O eW co c-

.I 't FMAS107

{

l:

7500 - -

O uoutor m

.e_

z 3!

l E

5000 --

l l

5

?

c3 r

e e

j i

E.

.4

.2 2500 -

..\\

l

.' )

l 0

g......j t

e e e e e eif a

,et e

e i e etei t

,,e e e f aet1 0

e a e f.

i 1

10 100 1000 10000 Time [sec) 1 i

Figure 8: Mutti Volume Steam Generator Model Middle Boiling Region Volume Mass (Volume 107)-

'FMAS' = Total Fluid Mass (Ibm)

'LIQM' = Liquid Mass (tbm) i g,

y

,yn

-=

+

1 g

.iw.+

l

\\

15000 --

[

h FuAsia E

l.

I e 12500 -

t.

a us Q3

. s.

6 llE l

j c 10000 -

.O

,e a

m 0:

m l

c D-E 7500 --

g}

t O

.i u

s c.

l e

s Q,

4

\\

5 i

3 5000 - -

N 9

', ' t ".

,o

.v

>..e a

~

E

\\.

a 2500 -

3 O

O t

t t i e t r il i

e e i etest t

i e s etest i

e t i etset 1

10 100 1000 10000 Time (sec]

Figure 9: Mutli Volume Steam Generator Model Upper Boiling Region Volume Mass (Volume 108)

'FMAS' = Total Fluid Mass (Ibm)

'LIQM' = Liquid Mass (Ibm)

t 15000 --

l-t'-

l E

l ' ;.

ruAstos

- - - FMASW p.

.o

..... FuAstos i

m m

l I

g 12500 --

s; y

I s-s l

I:\\

s

g. \\\\

I:

C t

0

': i j

g810000 --

3 t 1 j,

i i a:

fl

\\ l ei c

1/\\

.E

=

s

}

l 0

f s

i i

c3 7500 --

s 1

/

co i

\\

N t.

\\

s

,a f.:

.. ';. ?

i e

s

.(

n

.o r

\\\\ \\,;

,/

c l

5000 -- - -

a,-

-.................3

., i t

4

~~

w

./

a l

l 2500 -

E

.E O

0 i

1 10 100 1000 10000 Time (sec) i Figure 10: Mutti Volume Steam Generator Model Boiling Region Volume Masses (Volumes 106,107 and 108)

'FMAS' = Total Fluid Mass (Ibm)

Volume 106 = Lower Boiling Region (225.460 cubic ft) l Volume 107 = Middle Boiling Region (686.580 cubic ft)

Volume 108 = Upper Boiling Region (1037.90 cubic ft)

\\

3 I

i 1

\\

l f

g --

/

,./,i 0.9 -=. ---

/..

i s.,

\\.

,f

..> y 3.

0.8 -

.. i.. ? '

t x--

'/Y I

}

0.7 -

I g,,J 1.,

4 x

w s i,

e 0.6 - -

O i

n I

c O,

u'. O 5 -.

Og

> 0.4 --

0.3 -

0.2 --

vowim

- -- vowtor 0.1,_

- vowim 0

i i

i i

l 1

10 100 1000 10000 1

Time (sec]

Figure 11: Mutli Volume Steam Generator Model Void Fractions in Boiling Region (Volumes 106,107 and 108)

NOlV' = Void Fraction i

Question 29 j

Supplemental information l

REQUEST:

Provide the basis behind the 14.3 seconds assumed in the Safety injection (SI) pwge

~

time for the Main Steam Line Break (MSLB) analysis. Compare the CECO assur sed value to the purge time assumed by Westinghouse.

RESPONSE

CECO assumed a purge time of 14.3 seconds in the MSLB analysis, as stat ad in the response to Question 29. The purge time is define'd as the time for unb, rated water l

in the Safety injection (SI) line to travel from the Baron injection Tank (3:0 to the RCS l

cold leg. The time was calculated based on Si line dimensions from Zios specific l

drawings. The total pipe length from the BIT to the cold leg consisted c f 57.52 feet of 4 inch diameter pipe and 44.8 inch of 3 inch diameter pipe, resulting n a purge volume of 7.22 cubic feet. An average Si line flowrate of approximately 31 lb/sec was used, which was obtained from RETRAN calculations. The density of the fluid is representative of the conditions in the Siline at the time ofinjection. Tl e purge time can then be calculated as follows: '

7.22 Ft' lb Purge Time - 31 lb / sec(61.73 Ft') - 14.3 seconds l

=,,,

-i A&

X RWr$T No uncertainty has been explicitly incorporated into the calculation. This is valid, since the pipe dimensions are taken from as-built station drawings, ai d conservatism already exists in the conditions used for the fluid parameters.

Note : The total pipe length used is from the BIT to the cold leg and not fror i the Refueling Water Storage Tank (RWST) to the cold leg as stated in the response. This is illustrated in the sketch of the system above. However the BIT was removed subsequent to the original analyses presented in the opical report, and the charging pump now discharges directly to the cold leg (n a SI-actuation signal. This means the purge volume must now include the st ction line of the charging pump back to the RWST isolation valve, as illustratet,

This will result in an increased 51 delay time, which will be properly accounted for in all subsequent MSLB analyses.

l Westinghouse does not specifically account for a purge time in the 51 delay.

l The Si line is modeled in LOFTRAN and the code calculates the purge time internally.

The volume of the Siline downstream of the BIT used by Westinghouse is comparable to the CECO volume and check cases performed by CECO with LOFTRAN have shown that the delay time calculated also appears to be comparable.

l l

l w

w r

a

4 i

J I : Supplemental Information on Main Steam Line Break Methodology

1 Question 1.

What is the value for the boron purge time as the plant is presently constructed?

As I understand it, with the removal of the BIT, the old purge time is no longer j

valid, and you go all the way back to the RWST. What ever the future value will be for the MSLB analysis is the one that we want to use now. With this value, how long does it take the borated water to reach the RCS? When does the single failure occur, before or after the purge volume has passed? If yes, has CECO performed parametrics with respect to this ? If yes, what where the l

results ? What temperature is assumed for the unborated water ( assuming

" cold"is more conservative ) ?

i

Response

This question contains six separate issues with each respective response f

l detailed below :

I 1a)

The calculated value of the boron purge time as the plant is presently constructed is 75 seconds.

ib)

With the removal of the Boron Injection Tank (BIT) the old purge time is no longer valid. The new purge time of 75 seconds is calculated based on the swept fluid volume in the Safety injection piping system downstream of the RWST isolation valves.

i i

ic)

Introduction:

]

The CECO original topical report (Reference 1) describing the MSLB was completed in 1989.

Since then, the Boron injection Tank (BIT) has been,

removed. Therefore, to address this issue and to adequately respond to the NRC reviewer's questions, a MSLB case was run incorporating the analysis parameters that presently exist in the plant. This new MSLB case is referred to as the CECO updated MSLB case.

The changes made to the original Reference 1 topical case for the updated case are identified below. All other parameters and models are identical to that presented in the topical report.

Purge Time The BIT removal necessitated the calculation of a new purge volume and consequently an increased purge delay time for the Safety injection (SI) water entering the RCS cold legs. The new purge time is 75 seconds. This 75 seconds is the time it takes for the cold unborated water to purge the Safety injection flow paths from the check valve downstream of the SI/ Charging pumps 1',

i.

l-

(upstream of the Volume Control Tank (VCT) check valves) to the RCS cold legs. In addition to the 75 seconds,25 seconds is added to account for the 10 second Diesel Generator start time and 15 seconds for realignment of the VCT and RWST isolation valves on the Si signal. This results in a minimum total i

delay time of 100 seconds for the cold borated water to enter the RCS cold legs.

For additional conservatism, the updated MSLB case assumed a total safety injection delay time of 140 seconds which includes the following times:

i) 75 seconds for the safety injection purge delay time ii) 10 seconds for the diesel generator start time lii) 15 seconds for the RWST and VCT isolation valve realignment time iv) 40 seconds additional delay time for conservatism SI Enthalpy The enthalpy of the Safety injection water assumed is 10 Btullb, which is conservatively colder than the topical case and increases the RCS cool down.

SI Flowrates The Safety injection flow rate versus RCS pressure assumed is listed in Table 1.

This flow rate is the same as documented in the Updated Zion FSAR Figure 15.1-13B.

Si initiation and Reactivity due to Boron in order to conservatively increase the RCS cooldown, the Si flow was initiated at 1 second while no credit for the negative reactivity addition due to boron was taken until after the purge delay time.

I l.

2 l

ic Results Table 2 lists several important updated MSLB case output parameter values, while Figures 1 through 9 plot key output parameter results. This data will become the new reference topical case for the CECO MSLB methodology. For comparison purposes available FSAR data has also been plotted in these figures.

Figure 1 provides a plot of the Core Heat Flux which shows that the CECO updated MSLB case predicts a heat flux of 31.68% at about 144 seconds compared to about 15 % for the FSAR analysis. This demonstrates the conservatism of the CECO MSLB analysis which is due primarily to the kinetics feedback modeling. The initial differences in the core heat flux can be attributed to the modeling differences between the FSAR and the CECO model. These difierences include the mixing model (discussed in detail in Reference 3),

auxiliary feedwater and the kinetics feedback modeling. In the CECO model after overcoming the shutdown margin of 1.6%, the reactor becomes critical.

The initial (before 50 seconds) difference in the heat flux plot is not significant in the transient, since the time of minimum DNBR occurs at 144 seconds which is 1

the time of peak heat flux. At about 184 seconds the faulted steam generator dries out. This occurs due to the higher heat flux and consequently higher l

steam flow predicted in the CECO MSLB model.

Figure 2 provides the boron and doppler reactivity feedback for the CECO updated MSLB analysis. There are no comparable figures in the FSAR. At l

about 52 seconds, the reactor goes prompt critical until the negative doppler reactivity mitigates the power increase since there is no boron in the RCS at this time. After the 140 second delay, the SI system injects boron into the RCS and at about 144 seconds small amounts of boron reach the core. The peak heat flux occurs at 144 seconds at which time the small negative reactivity contribution due to boron is negligible compared to the doppler reactivity feedback as shown in Table 2 (4.448 pcm for boron compared to 425.04 pcm for the doppler).

Figure 3 compares the steam flow for the faulted steam generator between the CECO updated and FSAR MSLB cases. As expected (due to the higher heat flux) the steam flow from the faulted steam generator for the CECO model remains significantly higher than that shown in the FSAR case. As stated earlier, in the CECO case, the faulted steam generator dries out at about 184 seconds. This occurs since the plant auxiliary feedwater system is not capable of maintaining such a high steam generator steam flow rate. However, the overall characteristic trend of the steam flow results for the CECO MSLB model is comparable to that of the FSAR analysis.

3

Figure 4 provides the core boron concentration as a function of time. In the l

FSAR analysis (FSAR Figure 15.1-19C), boron enters the core at about 100 seconds compared to about 144 seconds for the updated CECO case. The CECO boron transport model incorporates a variable lag that calculates the core '

boron concentration as a direct function of the RCS mass flow and density.

Although some MSLB models incorporate a boron delay transport model, CECO has determined this variation in boron models is unimportant since the negative reactivity feedback due to boron is insignificant compared to that contributed by

+

l the doppler.

Figure 4 shows that when the minimum DNBR occurs at 144 seconds, the core boron concentration for the CECO updated MSLB case is insignificant compared to the 2 ppm for the FSAR analysis.

Figure 5 compares the Pressurizer pressure response for both the FSAR and the CECO updated MSLB transient. The plots follow the same trend except for i

the period between 35 to 50 seconds when the FSAR case predicts a slightly i

higher pressurizer pressure compared to the updated CECO MSLB model. This is due to the fact that the FSAR model predicts higher initial core heat flux.

Figure 6 provides the CECO updated MSLB nuclear power for which there is no comparable plot in the FSAR. The heat flux (Figure 1) is used to calculate the minimum DNBR, but this plot is provided to illustrate that the doppler reactivity l

mitigates the peak power while the boron does not have a significant effect. In j

the Reference 1 topical report submittal, the boron injection was initiated at l

about 40 seconds and did affect the peak nuclear power. However, in the updated MSLB case, the boron only has the long term effect of maintaining the core subcritical.

Figure 7 provides the faulted steam generator liquid level. Although the steam generator level is conservatively initialized at a 52 feet liquid level, the steam generator has still dried out by about 184 seconds due to the large heat flux during the event. However, this dry out does not affect the results, since it occurs after the minimum DNBR.

Figure 8 provides the intact and faulted loop core inlet temperatures. As discussed previously, the faulted steam generator dries out at about 184 seconds. at which time the faulted loop temperature increases significantly.

Figure 9 compares the reactivity between the FSAR and the updated CECO MSLB case. The trend of the reactivity for both the FSAR and the CECO case is

^

similiar. The differences in the reactivity between the FSAR and the CECO model are due to the ef'ects as discussed earlier for the core heat flux response (Figure 1). These differences were mixing, auxiliary feedwater and the kinetics models.

4

s 1

Table 1 Safety injection Flowrate from the Updated FSAR MSLB analysis i

SI Flowrate assumed from FSAR Fig 15.1-13B Pressure Flow rate

( psia)

(Ib/Sec) 15.0 58.664 215.0 55.957 615.0 51.444 1015.0 45.126 1415.0 37.906 1615.0 35.650 l

1815.0 31.588 2015.0 27.076 2215.0 21.661 2315.0 18.953 2415.0 11.7332 2515.0 0.0 e

l l

5 i

Table 2 Key Parameter Values for the Updated MSLB Case Maximum heat flux 31.64 %

Time of maximum heat flux 144 seconds Minimum DNBR

> 1.45 Time of Minimum DNBR 144 seconds Minimum RCS Pressure 729 psia l

Negative Reactivity due to Boron at the time

-$0.0096 l

of Peak Heat Flux l

Negative Reactivity due to Doppler at the time

-50.924 l

of Peak Heat Flux l

l 1

o I

1 l

6 l

l i

Figure 1 Updated MSLB Core Heat Flux i

0.35 l

i i

i 0.3 I

I

<e C

i

'5 025--

l Updated MSLB l

o z

Case O

0.2 c

.o.

- - - - FSAR Case U

tC a- 0.15

,,... ~

t X

e 2

O.1 w, - -

i E

e X

II l

0.05

{

0-i 0

50 100 150 200 l

l Time (Seconds) t l _.

7 e

i s

i Figure 2: Updated MSLB Doppler & Boron Reactivity 0.2 l

-7 0

l

'l I

1

-0.2 1

i G

t l

t I

i Doppler

~~

E. -0.4

\\

U Boron

/

E

\\

I e

\\

I-

-0.6

\\

I

\\ s f

-0.8 N.

I s

~,~~_.____s l

-1 i

1 0

50 100 150 200 Time Seconds l

I 8

i i

Figure 3 Updated MSLB Steam Flow i

8000 i

7000 i

i 6000 -

t

-0 5000--

m i

CECO Case l

A c 4000

- - - - FSAR Fig 15.1-19B E 3000 E

i O

a S 2000

. - - - - -...l.......l....

1000 0

l

-1000 0

50 100 150 200 Time (Seconds) 9

l 1

I i

l Figure 4: Updated MSLB Core Boron Concentration 25 j

i t

20 t i

I E

i S 15 3 CECO Case t

C e

- - - - FSAR Fig 15.1-e j

19C E 10 O

o U

r 9

5 j q

e p

e

_..'I g

O 50 100 150 200 Time (Seconds) i 10

Figure 5: Updated MSLB Presurizer Pressure 1

2500 P

s 2000 CECO MSLB s

q

1500 -

~

- - - - FSAR Fig 15.1-1 3

20C e

o an

@1000 l

~~.._ _

..._e 500 o

0 50 100 150 200 Time (Seconds) 11 L

l

9 Figure 6: Updated MSLB Nuclear Power n

0.45 l

^i 0.4

.c C

i o 0.35 z

+

O l

0.3 c

.9 i

M 0.25 b

0.2 l

3 O

@0.15 N

7 0.1 5

b l

E 0.05 0

0 50 100 150 200 Time Seconds t

I l

12 j

o.

o Figure 7: MSLB Faulted SG Liquid Level 60 i

50 -

J 2

• 40

• _S u

82 30,

w C

4 0o E 20 re UM 10 0

0 50 100 150 200 Time Seconds 13 i

4 Figure 8: Updated MSLB Core inlet Temperature 6.00 E+02 -

l i ~'s i

5.00E+02 --

s\\.

)

1 n

$ 4.00E+02 p

ab E

3 3.00E+02 l

E oa.

E o 2.00E+02 F-'

Intact Loops 1.00E+02 Faulted Loop i

0.00E+00 l

o 50 100 150 200 Time Seconds I

1 i

14 t

r-

t

.i I

P l

Figure 9 Updated MSLB Reactivity f

500

+

0

-e

' ' ~ ~ ~ ~ -

)

E g -500 b

=

5 e

i CECO Case l

r M -1000 -

- - - - FSAR Fig 15.1 -

l E

e 208

-1500 -

i 1

-2000 0

50 100 150 200 Time (Seconds) i I

i Note: Initially the reactivity for both the cases is zero.

4 15

t J

id) As discussed in the CECO updated MSLB analysis results, the boron l

reaches the core at about 144 seconds. The reactivity due to boron and the doppler is plotted in Figure 2.

]

i 1e)The CECO Main Steam Line Break (MSLB) analysis presented in the Transient Analysis Envelope assumes that the most limiting single active failure i

is the loss of power to one ECCS injection train. This failure is assumed to occur when the Si signal is generated. CECO has determined that this single failure.

)

5 assumption is consistent with the current Zion licensing basis and_the applicable j

i regulatory standards.

ANSI /ANS 51.1 (Reference 5) which establishes the nuclear safety l

criteria for PWRs, specifies in Section 3.2.1 that the single failure criterion be applied in demonstrating acceptable mitigation for analyzed events. In particular, j

Section 3.2.1.b requires that electrical nuclear safety related equipment conform i

to the single failure criterion as provided in ANSI /IEEE 379 (Reference 6) and l

ANSI /IEEE 279 (Reference 7). However, the only reference to single failure timing is found in ANSI /IEEE 279 Section 4.2 which states Any single failure within the protection system shall not prevent proper protective action at the system level when required.".

l l

CECO has always interpreted this design basis requirement as the single i

4 failure occurring simultaneous with the demand for system or component action.-

}

Examples include failure of a valve to open or close when required, failure of a i

protection channel to actuate when required, and in the MSLB event, failure of-l an ECCS injection train to initiate flow when required. Conversely, a component or system single failure occurring randomly at some time other than when it is i

required to function is considered a physically different phenomenon than the single active failure discussed in the ANSI /IEEE standards.

i

?

q in summary, the current CECO MSLB event assumption of one ECCS injection i

train failing to function upon the SI signal demand is considered consistent with the required ANSI /IEEE standards single failure criterion. CECO has not evaluated this single failuro occurrence with respect to the SI purge delay timing

.since this is considered outside the reload safety analysis design basis methodology.

if) The unborated SI water was assumed to have an enthalpy of 10 Btullb.

l 16 2

a.-..

,.,,. ~.,

> r O

Question 2 The boron transport model must be qualified.

Response

The CECO boron transport model was described in detail in References 2 and 3.

As stated in those references, the CECO boron transport model incorporates a l

variable lag based on the RCS mass flow and density. The overall conservatism j

of the model was substantiated by performing a mass balance as shown in Table 4-2 of Reference 2. As discussed previously, the boron does not significantly affect the transient results. As can be seen from the response to Question 1, the boron negative reactivity is a factor of 100 less than the doppler reactivity at the time of peak heat flux (the minimum DNBR occurs at 144 seconds which is the time of pea < heat flux). The insignificant boron effects on the updated MSLB case results verify that the CECO boron transport model is adequately conservative e

17

u.

o l

l l

Question 3 The use of the 75/25 split of Doppler reactivity weighting factors must be justified. What is used by Westinghouse ?

Response

The Doppler weighting factor of 75 % for the reactivity feedback from the faulted loop and 25 % from the remaining three intact loops was used in the CECO l

MSLB analysis. A constant Doppler Temperature Coefficient (DTC) salue of -1.0 pcm/ F was used which corresponds to the applicable Nuclear Design least negative DTC limit.

The Westinghouse MSLB analysis assumes a reactivity weighting factor of 50 %

from the faulted loop and 50 % from the remaining intact loops.

The reactivity weighting factor does not determine the overall MSLB reactivity feedback effects and must be evaluated in combination with the other kinetics feedback parameters. If all other kinetics feedback parameters are identical, then a 50% doppler reactivity weighting from the faulted loop will be conservative compared to a 75 % weighting factor assumption. However, the l

overall kinetics feedback model used in the CECO updated MSLB analysis is very conservative compared to the FSAR analysis. This is clearly illustrated in the plot of heat fluxes in response to Question 1. The CECO model predicts a heat flux of about 31 % compared to a value of only about 15 % for the FSAR analysis.

As stated in Reference 2 (page 245) the reactivity calculated by the RETRAN point kinetics model is compared to that calculated by the CECO PWR Nuclear Design MSLB model (Reference 4). In all of the CECO MSLB cases the reactivity calculated by the Nuclear Design model is significantly lower than that calculated by the RETRAN point kinetics model. It is CECO's understanding that this methodology is consistent with Westinghouse and other utilities who perform similar checks to ensure the overall transient reactivity feedback is conservative.

In summary, the CECO MSLB analysis will ensura the overall transient reactivity

~

feedback is conservative based upon comparison to the more detailed CECO Nuclear Design calculations.

18

hI

References:

i 1.

Document Number NFSR-0069, Dated November 8,1989, " Transient Analysis Envelope for Zion Units 1 and 2" 2.

Document Number RAl-1, Dated September 1991, " Responses to Nuclear Regulatory Commission Staff Request for Additional Information on CECO Transient Analysis Methods for PWRs" 3.

Document Number RAl-2, Dated April 1993, " Responses to Nuclear Regulatory Commission Request for Additional Information on CECO PWR Transient Analysis Methodology" 4.

"NRC SER on CECO Topical Report in PWR Nuclear Design Methods for Zion and Byron /Braidwood" Letter from NRC (S.A.Varga) to CECO (D.L.Farrar),

December 2,1983.

5.

ANSI /ANS-51.1-1983, " Nuclear Safety Criteria for the Design of Stationary Pressurized Water Reactor Plants" 1

)

6.

IEEE-379-1977, " Application of the Single Failure Criterion to Nuclear Power Generating Station Protection Systems" i

7.

IEE-279-1971, " Criteria for Protection Systems for Nuclear Generating Stations" l

l l

t 9

19