ML20003G671

Testimony on Behalf of Util Re Doherty Contentions 3,39 & 20(a) Re Fuel Specific Enthalpy,Fuel Swelling & Gap Conductance,Respectively.Prof Qualifications Encl

ML20003G671
Person / Time
Site:	Allens Creek File:Houston Lighting and Power Company icon.png
Issue date:	04/20/1981
From:	Holtzclaw K, Robert Williams GENERAL ELECTRIC CO.
To:
Shared Package
ML20003G672	List: ML20003G671 ML20003G673 ML20003G674 ML20003G675 ML20003G676 ML20003G677 ML20003G678
References
NUDOCS 8104300472
Download: ML20003G671 (23)

Text

N 4 --2 s

(

cKETED

\\\\

1 UNITED STATES OF AMERICA I

NUCLEAR REGULATORY COMMISSION 9-APR 2 31981>

-12 h

BEFORE THE ATOMIC SAFETY AND LICENSING A

3 Branch N

0 4

In the Matter of

).

y

)

'D 5

HOUSTON LIGHTING & POWER COMPANY

)

Docket No. 50-

)

g

/

6 (Allens Creek Nuclear Generating

)

i I

Station, Unit No. 1)

)

/;

)

l APR 5

7 I

vi u,,,

2DSA TESTIMONY OF KEVIN HOLTZCLAW AND RICHARD WILLII %

k %*g 7g

[-

8!

i ON BEHALF OF HOUSTON LIGHTING & POWER CO. ON N

9 i

DOHERTY CONTENTION 3 - FUEL SPECIFIC ENTHALPY;#s

[

DOHERTY CONTENTION 39-FUEL SWELLING; AND DOHERTY g

'g 10 CONTENTION 20(a)-GAP CONDUCTANCE 11 Q.

Mr. Holtzclaw, please state your name and place of 12 employment.

13 A.

My name is Kevin Holtzclaw and I am employed as a Senior 14 Licensing Engineer with the General Electric Co.

My business 15 address is 175 Curtner Avenue, San Jose, California.

16 Q.

Would you describe your professional qualifications?

i 17 A.

A copy of my professional qualifications is set forth 18 in Exhibit KH-1, 19 Q.

Dr. Williams, please state your name and place of 20 employment.

l 21 A.

My name is Richard Williams and I am employed as an l

l 22 engineer with the General Electric Co.

My business address 23 is 175 Curtner Avenue, San Jose, California.

24 Q.

Would you describe your professional qualifications?

25 A.

A copy of my professional qualifications is set forth i

l 26 in Exhibit RW-l.

27 Q.

Mr. Holtzclaw, what is the purpose of this testimony?

28 A.

The purpose of this testiaany is to address Mr. Doherty's h

r l

I

$*ft i 8104 3 0 047A

1 2

contentions 3 on fuel specific enthalpy, 39 on fuel swelling 3

and 20 (a) on gap conductance.

4 Q.

Mr. Holtzclaw, who prepared this testimony?

5 A.

Dr. Williams and 2 jointly prepared this testimony 6

responding to Mr. Doherty's three contentions.

Since 1979, Dr. Williams has worked in the Fuel Rod Thermal and Mechanical 7

Analysis Unit, and I worked in this same unit from 1971-1980.

g!

our combined expertise over this time period covers the in-g 10 formation required to respond to Mr. Doherty's contentions.

11 Although this testimony was jointly prepared, I will be the 12 chief spokesman on Mr. Doherty's contentions 3 and 20(a),

13 and Dr. Williams will be the chief spokesman on Mr. Doherty's 14 contention 39.

15 Q.

Mr. Holtzclaw, what is your understanding of the concern 16 raised in Mr. Doherty's contention 3?

17 A.

In contention 3, Mr. Doherty asserts that tests on 18 General Electric fuel rods show that the cladding will 19 rupture at an energy deposition of between 147 cal / gram and 20 175 cal / gram.

Mr. Doherty goes on to assert that rupture of 21 the fuel cladding leads to the following:

22 (a)

Fuel fragments being released into the coolant.

23 (b)

Pressure pulses due to fuel contacting the coolant 24 water.

25 (c)

Further degradation of cladding strength.

26 (d)

Jamming of control rods.

27 Q.

Describe the energy deposition in the fuel specifically 28 with respect to a power excursion.

=

1.

2 A.

The temperature of a fuel pellet is a measure of its 3

energy content.

The higher the temperature, the higher the 4

energy content of the fuel pellear.

As the neutron population 1

5 in the core grows, the tempe:-a.ture -of the fuel increases due 6:

to the energy generated from the fissioning process.

This 7

increase in neutron population is a consequence of a reactivity g

increase.

The amount of temperature increase depends on how il 9

rapidly the reactivity increases and on how much of the I

10 fission energy is transferred trom the fuel pellet in the 11 form of heat across the pellet-clad gap, across the clad, 12 and into the reactor coolant.

For a very rapid increase in 13 reactivity, the fuel pellet temperature increases rapidly.

14 The result of this rapid energy deposition in the fuel is 15 that the fuel and cladding temperature will increase.

16 The event resul'cing in the most rapid energy deposition 17 in a Boiling Water Reactor is a postulated control rod drop 18 accident.

The reactivity increase in a rod drop accident is 19 terminated by a combination of the inherent neutronics of 20 the fuel (Doppler Effect) and by insertion of the control 21 rods.

22 Q.

Please describe the course of events of a postulated 23 control rod drop accident.

24 A.

It is assumed that the core is in the optimum state 25-which results in the highest incremental rod worth and 26 ensures that withdrawal of a control rod results in the 27 maximum increase in reactivity.

The following events are 28 then postulated:

1.

2 (a)

The maximum worth control blade becomes decoupled 3

from the control rod drive.

4 (b)

The operator selects and withdraws the control I

S rod drive of the decoupled blade.

6 (c)

The decoupled blade sticks in the fully inserted 7

position.

3 (d)

The control blade becomes unstuck and drops to 9l the drive position.

10 (e)

The reactor goes supercritical, and in less than 11 one second the initial power increase is terminated 12 by the Doppler Reactivity Feedback.

13 (f)

The signal at 120% power from the average power 14 range monitor initiates a reactor scram.

15 (g)

The reactor scrams, terminating the accident 16 in less than five seconds.

It should be noted 17 that the blade which is assumed to have become 18 dislodged would be driven back into the core 19 during the scram.

20 Q.

What are the results of analyses carried out for the 21 Rod Drop Accident?

22 A.

A conservative generic rod drop accident analysis has 23 been performed by General Electric and is described in NEDO-24 21231, " Banked Position Withdrawal Sequences."

The calculated 25 maximum total energy deposition is less than 135 cal / gram of 26 UO.

The generic analysis is applicable to ACNGS and is 2

27 appropriate for the PSAR licensing stage since ACNGS will 28 use the same hardware, procedures and similar fuel design

1.

2 included in the generic analysis.

The ACNGS specific rod 3

drop accident analysis calculation will be performed at the 4

Final Saftey Analysis Report stage when the final bundle 5

enrichments and core configuration are established.

6 The General Electric analysis makes use of worst case 7

input assumptions to maximize the reactivity increase.

It 8

also employs an adiabatic model which takes no credit for 9

moderator feedback.

An analysis carried out by Brookhaven 10 National Laboratory (BNL-NUREG-28109, July, 1980) indicates 11 that inclusion of moderator feedback can reduce the erargy 12 deposition significantly depending on core conditions at the 13 time of the accident.

Reductions in the range of 28% to 50%

14 were identified in the Brookhaven report.

15 Q.

What are the limits used by General Electric in the 16 design of ACNGS regarding the Rod Drop Accident?

17 A.

To minimize damage to the reactor coolant pressure 18 boundary and core internals, and to ensure the maintenance 19 of both short term and long term core cooling capability, a 20 limit of 280 cal / gram total energy deposition is employed.

21 Q.

What is tho basis for this limit?

22 A.

This limit was based on a review of data available from 23 SPERT and TREAT projects.

SPERT and TREAT were two test 24 series funded by the US Atomic Energy Commission to assess 25 fuel integrity under reactivity-initiated accident conditions.

26 The findings of this review indicated that energy deposition 27 in the test fuel was the single most important vari.ble.

28 Fuel failure was relatively insensitive to cladding material, L

1 - '

y heat treatment, fuel form, fuel material and gap width.

2 Pressure pulses and fuel fragmentation resulting from rod g

failure were found to be insignificant below 240 cal / gram 4

total energy deposition for both irradiated and unirradiated 5

fuel rods.

These tests, therefore, demonstrated that a 280 6

cal / gram total energy deposition limit is conservative in 7

terms of preventing pressure pulses, which could adversely g

affect the reactor coolant pressure boundary, and in pre-9 venting fuel fragmentation which could impair the maintenance 10, gi of a coolable geometry.

The 280 cal / gram total energy deposition limit has been 12 further substantiated by recent independent tests carried 13 14 out in Japan.

Between 1975 and 1980 over 350 separate RIA tests have been carried out in the Japanese Nuclear Safety 15 Research Reactor (NS'AR).

Results from these tests confirmed ig the earlier SPERT/ TREAT results as they indicated no detectable 17 18 pressure pulses or fuel fragmentation below 380 cal / gram.

Additional tests in single rod and multi-rod geometries 19 have been carred out at Idaho National Engineering Laboratory 20 by EG&G Idaho, Inc., which further support the 280 cal / gram l

21 22 total energy deposition limit.

It should be noted that the NRC Regulatory Guide 1.77 23 24 expresses the limit of 280 cal / gram in terms of radial l

25 average peak fuel enthalpy, whereas the SPERT and TREAT data and General Electric calculation results are expressed in 26 27 terms of total energy deposition.

Total energy deposition i

28 is greater than the associated radial average peak fuel l

L

1 -

2 enthalpy because of heat transfer from the fuel to the 3

cladding and coolant during the power transient, and the 4

relatively large fraction of the total energy which is due 5

to delayed fissions.

The conversion from total energy 6

deposition to radial average peak fuel enthalpy is reactor 7

design dependent.

For the SPERT and TREAT reactors, the 3

delayed neutrons account for approximately 25% of the total g

energy deposition; therefore, 280 cal / gram total energy 10 deposition is equivalent torv230 cal / gram radial peak fuel 11 enthalpy.

For a BWR 6 such as Allens Creek, the delayed 12 neutrons account for approximately 40% of the total energy 13 deposition; therefore, 280 cal / gram total energy deposition 14 is equivalent torv200 cal / gram radial average peak fuel 15 enthalpy.

l 16 Q.

Is the current limit of 280 cal / gram radial average i

17 peak fuel enthalpy given in Regulatory Guide 1.77 being 18 re-evaluated?

l 19 A.

The current limit is being questioned in terms of l

20 maintaining a coolable geometry.

There is no question of 21 the appropriateness of the limit in preventing damaging 22 pressure pulses.

Both the NRC and EG&G have suggested that l

23 a radial average peak fuel enthalpy limit of 230 cal / gram l

24 might be more appropriate for ensuring a coolable geometry.

25 This new limit for a BWR 6 would be equivalent to a total 26 energy deposition limit of 320 cal / gram.

Even if the limit 27 were changed from the current Reg. Guide 1.77 value to that 28 suggested by the NRC Staff, the current limit of 280 cal / gram

~-

1 -

2 total energy deposition employed by GE would be conservative.

3 More importantly, the bounding results of the Rod Drop 4

Accident would still be well below the limit.

5 Q.

Explain the fuel failures that Mr. Doherty cites in his 6

contention.

I 7-A.

Mr. Doherty refers to Rod #568 of the SPERT tests.

Rod 8

1. 568 was reported to have experienced cladding perforation g'

at an energy deposition of 147 cal / gram.

This rod was 10 subjected to a total energy deposition of 199 cal / gram.

11 There was no prompt fuel dispersal from this failure nor was 12 there any indication of resulting large pressure pulses.

13 Rod #859 of the SPERT test series, although not cited 14 by Mr. Doherty, was also reported to have experienced a low 15 energy cladding perforation of 85 cal / gram while undergoing 16 a total energy deposition of 190 cal / gram.

Again, there was 17 no prompt fuel dispersal nor resulting large pressure pulses.

18 These two tests are consistent with the 280 cal / gram 19 limit as well as with results from other SPERT and TREAT 20 tests.

They indicate that, while cladding perforation can 21 occur at less than 280 cal / gram, the failures will not 22 result in the generation of pressure pulses or loss of 23 coolable geometry which could adversely impact the reactor 24 coolant pressure boundary and core internals.

25 Q.

What are your conclusions?

26 A.

Rapid deposition of energy in a fuel rod can result in 27 fuel failure.

For a boiling water reactor, the worst event 28 relative to rapid energy deposition is a postulated rod drop

. 1 2

accident.

Current conservative GE calculations indicate 3

that the maximum total energy deposition from a rod drop 4

accident is less than 135 cal / gram.

Numerous test results 5

indicate that the 280 cal / gram limit on total energy deposi-6 tion is conservative and adequately prote. cts the reactor 7

core from prompt fuel dispersal and the attendant pressure g

pulses and loss of coolable geometry that Mr. Doherty suggests.

9 Although the appropriateness of the Regulatory Guide 1.77 10 limit for maintenance of coolable geometry is in question, 11 GE's application is consistent relative to all the data 12 sources, and the ACNGS rod drop accident analysis result 13 will be well below the limit value.

14 Q.

Dr. Williams, turning to Mr. Doherty's contention 39, 15 what is the purpose of your testimony?

16 A.

The purpose of this testimony is to address Mr. Doherty's 17 Contention 39, as reworded by the Board in its March 10, 18 1980 order, which asserts that the Applicant has not pro-19 vided an adequate showing that the degree of fuel swelling 20 and incidence of rupture are not underestimated.

The wording 21 of the contention comes directly from Section 1.B of Appendix 22 K to 10 CFR 50.

23 Q.

Has General Electric estimated the degree of potential 24 fuel swelling and rupture for ACNGS?

25 A.

Yes.

The methodology which General Electric uses to l

26 address Appendix K requirements is contained in NEDO 20566*/

27

• / General Electric Company Model for Loss-of-Coolant l

28 Analysis in accordance with 10 CFR 50, Appendix K.

l l

1 2 and is based on extensive testing.

The methodology has also 3

been subject to extensive review by the NRC.

As the state 4

of the art develops and more testing is done, the methods 5

used in NEDO 20566 will be updated as appropriate.

The 6

final review of Allens Creek will come at the FSAR stage 7

with the plant being reviewed against Appendix K require-8 ments using the latest methodology and test results as 9

available which have been approved for use by the NRC.

10-O.

What is the most severe

,ent in terms of swelling and 11 rupture of the fuel rods?

12 A.

The Loss of Coolant Accident (LOCA) event is the most 13 severe in terms of swelling and rupture because the LOCA 14 event causes the largest differential pressures across the 15 clad, and the highest clad temperatures.

Demonstrating 16 compliance with Appendix K for the LOCA event will bound all 17 other design basis conditions.

18 Q.

Please describe the LOCA event with respect to swelling i

19 and rupture of the fuel.

l 20 A.

Normally, the Reactor Pressure Vessel (RPV) internal i

21 pressure exceeds the fuel rod internal pressure, causing an 22 inward force on the cladding which prevents swelling.

23 During a postulated LOCA event, a rapid depressurization of

(

24 the RPV occurs and the fuel and cladding heat-up due to the 25 stored energy of the fuel at the onset of the accident and 26 the decay heat generation.

This depressurization of the RPV 27 can result in the fuel rod internal pressure exceeding the 28 RPV pressure.

As soon as the fuel rod internal pressure l

i a

I 1 2 exceeds the RPV pressure, fuel cladding begins to deform 3

outward.

This deformation initially will be elastic, i.e.,

4 no permanent deformation will occur as long as the stress on 5

the cladding remains below the yield strength of the clad.

6 As cladding yield strength decreases with increasing tempera-7 ture, a point in the LOCA transient can be reached when the 8

stress in the fuel cladding exceeds the yield strength.

At 9

this point, the cladding deforns plastically (permanent 10 deformation) at a rate that increases with increasing tempera-11 ture.

This rapid permanent outward deformation of the 12 cladding is termed fuel cladding swelling.

swelling continues 13 until termination of the transient or until the ultimate 14 tensile strength is exceeded.

When the ultimate tensile 15 strength of the cladding is exceeded, the cladding fails and 16 a perforation of the clad is formed.

The driving force, for 17 expansion, which is the internal gas pressure is relieved 18 upon perforation and the expansion stops.

19 Q.

You state that the fuel rod internal pressure is the 20 driving force for the fuel cladding swelling and rupture.

21 How is it calculated?

22 A.

The perfect gas law is used to calculate the fuel rod 23 internal pressure.

This law states that the fuel rod internal 24 pressure (PI) is linearly proportional to the number of l

25

c. oles of gas (N) and to the temperature of the gas (T) and 26 inversely proportional to the gas volume (V).

i 27 NT PI E V-28 l

1 -

2 This equation is written for both the plenum gas and the 3

fuel gas.

4 The initial fuel rod internal pressure (PI) of the i

5 unirradiated fuel is 3 at=ospheres at room te=perature.

As 6

the fuel is irradiated, the available gas volumes change 7

with te=perature and irradiation effects while the a=ount of l

8 gas to fill the available volume increases due primarily to 9

the release of fission gas.

The nu=ber of moles of gas 10 present prior to the LocA is calculated by making bounding 11 assu=ptions on the operating history prior to the LOCA.

12 I==ediately following the accident an additional amount of 13 fission gas is assumed to be released from the fuel.

14 The initial plenum gas volume is calculated from the 15 cold plenu= volume and the fuel expansion.

During the 16 accident, the increase in plenu= volu=e is calculated con-17 sidering only diametral growth and conservatively ignoring 18 the lengthening of the plenu= due to the increasing te=perature.

19 The fuel gas volu=e at the start of the accident is 20 assumed to remain constant throughout the accident.

The 21 additional volume due to increasing gap size during the 22 accident is ignored yielding a conservative value for fuel 23 rod internal pressure.

The hottest cladding temperature is 24 used as the fuel gas temperature during the accident.

This 25 will always over-estimate the fuel gas temperature and thus 26 yield a conservative internal pressure.

27 The plenum gas temperature during the accident is 28 co=puted by modeling the heat transfer exchange between both

1 2 the plenum and the top of the fuel and between the plenum 3

and the surrounding coolant.

4 This information allows the fuel rod internal pressure 5

to be calculated during the LOCA.

6 Q.

What type of testing has been done to quantify the 7

amount of swelling and incidence of rupture in a BWR?

8 A.

There have been many test programs to investigate the 9

important parameters of fuel swelling and rupture. General 10 Electric has carried out full scale bundle tests in which 11 prototypical BWR fuel bundles were tested under simulated 12 LOCA conditions.

These tests revealed localized fuel cladding 13 swelling and rupture.

Fuel c1 adding perforation failure 14 propagation was not observed.

The cross-sectional flow area 15 of the bundle was found to be re/.uced by up to 40% as a 16 result of the fuel cladding s elling and rupture, but this 17 extent of flow blockage did not reduce the coolability of 18 the bundle.

The results of these tests confirmed the 19 adequacy of the GE swelling and rupture models presented in 20 NEDO 20566.

21 In addition to GE's full bundle tests, there have been 22 numerous single rod and small bundle tests carried out by 23 other vendors and various research laboratories.

The 24 findings of these tests substantiate GE's test results.

1 l

25 0

Are fuel perforations expected for a LOCA at the Allens 26 Creek plant?

27 A.

A conservative assessment of the consequences of a LOCA 28 for Allens Creek has been done using the models given in

1 2 NEDO 20566 and Section 6.3 of the Allens Creek PSAR.

The 3

resultant hoop stress due to internal pressure for all fuel 4

rods is lbelow the ultimate strength of the clad, and results I

5 in no fuel damage in the form of perforations.

6 Q.

What are your conclusions?

i I

7 A.

Extensive testing has been done on fuel rod swelling i

8 and incidence of rupture.

General Electric has used con-9 servative interpretations of the available data to formulate i

10 its computer modeling of fuel swelling and perforation.

11 Q.

Mr! Holtzclaw, what is the purpose of your testimony 12 with respect to Mr. Doherty's contention 20 (a) ?

13 A.

The purpose of this testimony is to address Mr. Doherty's 14 Contention 20(a) which alleges that:

15 fission gas release due to fuel rupture of fuel rods with burn-up of greater than 20,000 megawatt-oays 16 per ton of uranium will be greater than applicant's estimate during a LOCA.

Applicant's underestimation 17 means fission gas release will be greater than pre-dicted, resulting in lower pellet-cladding gap con-18 ductance which results in higher initin1 stored energy and consequently higher peak cladding temper-19 ature for the ECCS to control during a LOCA.

This higher peak cladding temperature vill increase 20 the calculated peak cladding temperature to one in excess of 2,200*F.

The underprediction is caused by 21 the fact that in the Dutt and Baker co2;rection factor, the only independent variable considered was fuel burn-22 up.

Fuel operating temperature is an independent variable also.

Further, much of the data in support of 23 the correction factor was taken from fuel rods fabricated many years before and tested in 1973.

These older rods 24 differ from those to be used in ACNGS in several ways, some of which may have increased fission gas release, 25 while others decreased fission gas release.

There is no certainty the differences cancel out one another, so 26 that the data are applicable to a calculation to the ACNGS.

Intervenor contends Applicant should not be 27 permitted to use fuel rods once the threshold for significant fission gas release occurs.

This would be 35 at 24,000 MWD / metric ton for a BWR according to an article in Nuclear Safety, 20 (4), p. 418, 1979.

2

. 1 2

0 What is fission gas and fission gas release?

3 A..

A result of the fissioning of U235 is the r oduction of 4

fission gases.

The dominant gases produced are isotopes of 5

krypton, xenon, and iodine with a variety of half-lifes.

6 The total yield for these three gases is approximately 30 7

atoms per 100 fissions.

Most of these gaseous fission g

products are trapped within the fuel pellets.

However, a g

small fraction is released to the gap between the fuel 10 pellets and the cladding.

This is what is referred to as 11 fission gas release.

The amount of fission gas released 12 from the fuel is a strong function of the fuel temperature 13 and a lesser function of the fuel burn-up.

The fission gas 14 release model used by General Electric explicitly accounts 15 for the temperature dependence of fission gas release.

16 Q.

What is gap conductance and what is the effect of the 17 release of fission gas?

18 A.

Gap conductance is a measure of the rate of energy 19 transferred across the pellet-clad gap and determines the 20 temperature difference between the outer surface of the fuel 21 pellet and the inside surface temperature of the cladding.

22 Gap conductance is affected by the gap size, which is a 23 function of thermal expansion of fuel and cladding, fuel 24 irradiation swelling, fuel relocation and fuel densificatien 25 and by the thermal conductivity of the gas mixture.

Fission 26 gas release does not directly affect gap size but does 27 directly affect gas thermal conductivity.

For Allens Creek, 28 the gas mixture between the pellet-clad consists of helium, s

1 the gas trapped in the fuel pellet during manufacturing 2

3 (volatile gas), and fission gas.

Helium has a thermal conductivity that is more than an order of magnitude larger 4

than the conductivity of the dominant fission gases.

Thus, 5

release of fission gas during irradiation dilutes the helium 6

7 fill gas and lowers the thermal conductivity of the gas g

mixture.

Overpredicting the fission gas release would lead to g.

higher predicted fuel temperatures and peak cladding tempera-10 tures and, on this basis, would be conservTtive for LOCA 11 12 calculations.

An underestimation of fission gas release would erroneously lower the predicted fuel temperatures and 13 14 the peak clad temperature.

15 O.

What method is employed by General Electric to predict 16 fission gas release?

17 A.

General Electric's gap conductance model, which is 18 referenced in the Allens Creek PSAR, is celled GEGAP III.

GEGAP III is used to initialize the stored energy in the 19 fuel and the fuel rod fission gas inventory at the onset of 20 21 the postulated LOCA.

In order to do this the model must j

22 address gap size which is affected by thermal expansion of 23 fuel and clad, fuel irradiation swelling, fuel relocation, 24 and fuel densification and the gas mixture thermal con-25 ductivity which is affected by the quantities of initial 26 fill gas, volatile gas release, and fission gau release.

27 The fission gas release model which is part of the 28 GEGAP III gap conductance model, is an empirical model which

1 1 2 calculates the amount of fission gas released during a fuel 3

cycle by dividing the fuel into specified annular regions 4

based on temperature and then swnming the fission gas release 5

from each region.

Given a temperature profile in the fuel, 6

the model predicts the amount of fission gas release.

There 7

is no burn-up dependence in the model.

g The model was verified using gas release data described 9

in NEDO-10506.

This data comes from UO fuel irradiated in 2

10 the vallecitos Boiling Water Reactor (VBWR) and the General 11 Electric Test Reactor (GETR).

The fuels had an average 12 burn-up ranging from 300 to 73,000 MWD /Mtu.

Since the model 13 depends only on temperature and not on fuel rod design, the 14 verification applies to ACNGS fuel as well as to the fuel 15 used in VBWR and GETR.

16 Q.

What is the Dutt-Baker correction factor referred to by 17 Intervenor?

18 A.

The Dutt-Baker correction is an enhancement factor used 19 to predict fission gas release for fuel with burn-up greater 20 than 20,000 MWD /Mtu.

The factor modifies the fission gas 21 release results which the temperature dependent GEGAP III 22 model yields for conditions of fuel burn-up in excess of 23 20,000 MWD /Mtu and thus makes the results dependent not only 24 on temperature but also on burn-up.

25 The Dutt-Baker correction factor for BWR's was obtained 26 from the Dutt and Baker Liquid Metal Fast Breeder Reactor 27 (LMFBR) corralation in "Siex:

A Correlated Code for the 28 Prediction of LMFBR Fuel Thermal Performance," Westinghouse,

1,

2 Hanford Report, HEDL-TME 74-55, June, 1975.

3 Q.

Has General Electric done a safety analysis comparing 4

peak clad temperature using GEGAP III with and without the 5-Dutt-Baker correction factor.

6 A.

Yes.

General Electric has done a generic analysis for 7

the LOCA.

Based on the analytical results, the Dutt-Baker 8

correction factor makes no changes in the calculation of 9

peak cladding temperature up to approximately midlife but 10' does predict up to an 85'F increase in peak cladding tempera-11 ture at end-of-life.

However, at end-of-life, the exposure 12 of the high burn-up bundles will result in lower power 13 generation levels in these bundles.

At these reduced power 14 levels, the calculated peak clad temperature is significantly 15 below the 2200*F limit.

Therefore, the 85'F increase in 16 peak clad temperature late in life will not result in ex-17 ceeding the 2200*F peak clad temperature limit so that 18 actual plant operation would not be affected.

19 Q.

Has Gene.ral Electric submitted a new fuel performance 20 model to the NRC Staff for review?

21 A.

General Flectric has submitted to the NRC for review a 22 new fuel performance model, GESTR, which has an explicit 23 exposure enhancement factor in addition to an altered tempera-24 ture dependence.

Once the NRC has approved GESTR, the 25 Applicant will use GESTR in its Final Safecy Analysis Report.

26 Q.

What are your conclusions?

27 A.

The GEGAP III model used in the Allens Creek safety The inclusion 28 analysis is strictly temperature dependent.

w-

19 -

1 2

of fuel burn-up dependence in the GEGAP III model via the 3

Dutt-Baker correction factor changes the calculated peak 4

clad temperature for a LOCA but this small increase in 5

temperature is still well within the maximum allowed clad 6

temperature of 2200*F.

7 8

9 10-11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 w

e v--

1 Exhibit KWH-1 2

EDUCATION AND PROFESSIONAL QUALIFICATIONS 3

Mr. Kevin W.

Holtzclaw 4

Mr. Holtzclaw is a Senior Licensing Engineer in the 5

Safety and Licensing Operation working for the General Electric 6

Company, Nuclear Power Systems Division, in San Jose, California T

U.S.A.

His employment with General Electric began in 1969 as 8

an engineer in the Fuel Development Section, where he worked on 9'

a fuel capsule irradiation program and performed failure 10 analyses for operating reactor fuel.

From 1971 to 1980 he 11 worked in the Fuel Rod Thermal and Mechanical Design Unit 12 performing thermal and thermal-hydraulic design analyses for 13 initial core and reload fuel.

In May 1975, he was made 14 technical leader of a group of engineers performing design 15 and licensing calculations for initial core fuel projects.

16 This work scope included defining acceptance criteria and fuel 17 thermal-mechanical properties during steady-state, transient 18 and accident conditions.

In January, 1977 Mr. Holtzclaw also 19 acted as Program Manager of the Fuel Rod Prepressurization 20 Program having responsibility for the planning, scheduling, 21 budgeting and implementation of the GE BWR prepressurized fuel 22 design.

In February, 1980 Mr. Holtzclaw transferred to the 23 GE Safety and Licensing Operation.

As a Senior Licensing 24 Engineer, he is responsible for defining and planning programs 25 relating to degraded core rulemaking.

26 Prior to working at General Electric, Mr. Holtzclaw 27 worked for one year as an engineer at the San Francisco Bay 28 Naval Shipyard Nuclear Power Department.

In this capacity

I he worked in areas of vendor qualification and piping system 2

stress analysis.

3 Mr. Holtzclaw is a past member of the National Society 4

of Professional Engineers and the California Society of 5

Professional Engineers.

In 1976 he was named as the Young 6

Engineer of the Year by the Santa Clara Valley Chapter of 7

the California Society of Professional Engineers.

8 Mr. Holtzclaw is a 1968 graduate of San Jose State 9

College with a B.S. Degree in Mechanical Engineering (Nuclear 10 Power option).

In 1973, he received a M.S. Degree in Mechanical 11 Engineering from the University of California, Berkeley.

12 13 14 15 16 17 18 19 20 21 22 23 l

24 l

25 l

26 l

27 l

28

1 Exhibit RJW-1

~

2 EDUCATION AND PROFESSIONAL QUALIFICATIONS 3

Dr. Richard J. Williams 4

Dr. Williams is an Engineer working in the Fuel Rod 5

Thermal and Mechanical Analysis Unit at General Electric's 6

Nuclear Ene,rgy Business Group in San Jose, California.

His 7

employment with General Electric began in January 1979.

Dr.

8 Williams is responsible for nuclear fuel rod integrity under 9-normal, off-normal, transient and accident conditions.

In 10 this capacity he has performed analyses of fuel integrity under 11 the loss of coolant accident (LOCA), the reactivity initiated 12 accident (RIA) and the flow blockage event.

He has also 13 participated in the Three Mile Island Utility Support Program 14 providing analyses of the reactor fuel condition following the 15 incident.

Dr. Williams is a major General Electric engineering 16 interface with government and regulatory agencies on the Fuel l

17 Rod Research Programs being carried out in the United States, 18 Europe and Japan.

He is currently involved with the OPTRAN 19 test series being carried out in Idaho, the NRU LOCA Program 20 being carried out in Canada and the Super-Ramp Project 21 Committee, which governs the fuei rod testing being carried 22 out-in the Studsvic reactor in Sweden.

28 Prior to working at General Electric, Dr. Williams held 24 a National Research Council, (National Academy of Sciences) 25 resident research fellowship at NASA Ames Research Center, 26 Moffett Field, California where he directed cryogenic diode 27 heat pipe research.

In this capacity he was responsible for 28 the thermal diodes to be used on the Long Duration Exposure

1 Facility to be launched from the space shuttle.

~

2 Dr. Williams has published many papers in International 3

Technical Journals and holds both a BSc (1973) and Phd (1976) 4 in Mechanical Engineering from Swansea University (UK).

5 6

7 8

9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 l

- - -,, - -. -,