ML19320B490

Forwards Addl Info Re Fuel Design & Physics Calculations Per RA Clark 800620 Request.Response to Question 10 Withheld (Ref 10CFR2.790)

ML19320B490
Person / Time
Site:	Millstone
Issue date:	07/07/1980
From:	Counsil W NORTHEAST UTILITIES
To:	Clark R Office of Nuclear Reactor Regulation
References
B10028, TAC-11348, TAC-11561, TAC-12505, TAC-42846, NUDOCS 8007140344
Download: ML19320B490 (14)

Text

T NORTHEAST IFFII. FRIES 3

l:Mi"70T',$'"lli""27~

IOnfro*nE CONNECTICUT 06101

• :::::::ll":*"";'::'"~

. am) men k L J 200;;;:'=lll!l0l July 7, 1980 Docket No. 50-336 B10028 Director of Nuclear Reactor Regulation Attn: Mr. Robert A. Clark, Chief Operating Reactors Branch #3 U. S. Nuclear Regulatory Commission Washington, D.C. 20555

References:

(1) W. G. Counsil letter to R. Reid dated March 6, 1980 transmitting the Millstone Unit No. 2 Basic Safety Report (BSR).

(2) R. A. Clark letter to W. G. Counsil dated June 20, 1980.

Gentlemen:

Millstone Nuclear Power Station, Unit No. 2 Response to Questions on the Basic Safety Report for Cycle 4 In Reference (1), Northeast Nuclear Energy Company (NNECO) docketed the Basic Safety Report (BSR) in support of Cycle 4 operation of Millstone Unit No. 2. The BSR is intended to serve as a reference fuel assembly and safety analysis report for the use of Westinghouse fuel at Millstone Unit No. 2.

In Reference (2), the NRC Staff requested that NNECO provide additional information regarding fuel design and physics calculations to complete the review of the BSR.

Accordingly, NNECO hereby submits the attached information in response to the enclosure of Reference (2).

Please note that the response to Question 10 is not included in the attach-ment. The proprietary nature of the response to Question 10 requires that it be docketed in a separate submittal.

THIS DOCUMENT CONTAINS POOR QUAUTY PAGES 800ngo Jf! P

O- O We trust you find this information satisfactorily dispositions the Reference (2) concerns.

Very truly yours, NORTHEAST NUCLEAR ENERGY COMPANY I' j. . }. l h 4 'Il]t W' 'd. Coundil Senior Vice President At tachment

o .

DOCKET NO. 50-336 n

ATTACIDIENT MILLSTONE NUCLEAR POWER STATION, UNIT NO. 2 RESPONSE TO QUESTIONS ON Tile BASIC SAFETY REPORT, CYCI.E 4 JULY, 1980

F

.m., . . . ~. u m ..w..uu.n.u w.> u A*

Responses to NRC Questions on Westinghouse Basic' Safety' Recort for Millstone Unit 2

1. Please submit page 5-14 which is missing from the BSR. .

RESPONSE: Please find enclosed with this submittal copies of the missing page. .

2. Coolant pressum drop calculaticns for the Westinghouse fuel assembly design indicates a matching of the overall pressure drop with that for the original Combustion Engineering Millstone 2 fuel assembly design.

However, at each axial elevatien the pressure drops do not match up hetween the Westinghouse and/Cor.bustion Engineering fuel designs. The largest variatien in pressure drops for the two designs occurs at the lower nozzle where the Westinghouse design has a higher pressure-loss coefficient. This variation in pressure drop will result in an inlet flow maldistribution with less direct flow through the Westinghouse bottom nozzle. The BSR should provide justification as to why the msulting cross ficw dcwnstream of the botton nozzle will not produce an unacceptable degree of fretting wear at sites where spacer grid springs and dimples contact-fuel rods.

PESPOUSE: As stated in Section 3.3 and shown in Table 3.1 of the BSR, the Westinghouse fuel assembly design is hydraulically compatible with the Corbustion Engineering Millstone 2 fuel assembly design. The geometric similarity of all the assembly cor:'ponents and subsequent testing have assumd that the overall asserrbly pressure drops as well as the pressure drops,at each axial elevation are matched for the two designs.

The lower nozzle loss coefficient mismatch which was a concern at the Feb. 14,1977 meeting (Reference 1) was resolved by a re-design of the lower nozzle and subsequent testing to determine the effects of rods lifted and not lifted. This nozzle _ re-design and test results were presented at the January 26,1979 meeting (Reference 2) and showed that the W lower nozzle was hydraulically compatible with the CE lower nozzle.

Thus there will be no inlet flow maldistribution between the two designs.

r-WESTINGHOUSE PROPRIETARY CLASS 2 References -

(1) M. M. Mendonca, NRC, Memorandum to R. L. Baer, NRC,

Subject:

" Westinghouse Design and Testing Program for Millstone 2 Reload Number 1 Peeting - February 14,1977", April 19,1977.

(2) W. G. Counsil, Northeast Nuclear Energy Company, letter to R. "Reid, NRC, " Millstone Nuclear Power Station, Unit No. 2 Cycle 4 Reload Regarding January 26, 1979 Meeting", March 21,1979.

3. The Westinghouse fuel assembly design has 4 holddown springs, while tiie original Combustion Engineering design has 5 springs. Discuss the dif-ferences in the static r.nd dynamic response of each fuel assently design.

The raised pad on the center of the top nozzle orifice plate prevents the Westinghouse holddown springs from being compressed solid. Does this pad limit the axial distance that the Westinghouse fuel assemblies can grow relative to that of the original Combustion Engineering fuel assemblies?

Will the spacer grids of the two fuel assently designs always line up?

What is the safety significance if grid-to-grid alignment cannot be assured (i.e., will there be neutronic anomalies, will assembly peripheral fuel mds be punctured)?

RESPONSE: The W fuel assenbly holddown springs are designed to pmvent lift off of the fuel assembly from the' bottom core plate during normal operation.

This design requirement is consistent with the CE fuel assembly design of the holddown spring as discussed in section 3.3.1.4 of the Millstone Unit 2 FSAR. Since both the CE and W fuel assembly holddown loads are similar, the fuel assembly ' spring rates are judged to be similar, thus the static and dynamic response of each fuel assembly should not differ significantly.

The raised pad at the center of the W top nozzle orifice plate will not interfere with the free growth of the fuel assembly. It is located such that it contacts the holddown flower prior to the springs being compressed solid and prevents the nozzle extensions fmm topping out in the blind holes in - the ' upper core plate. This promotes uniform loading of the guide tubes during any hypothetical accident which could cause the fuel assembly to lift.

l n

WESTINGHOUSE PROPRIETARY CLASS 2 Since the raised pad does not prevent fuel assembly growth, there will always be grid to grid overlap between W and CE fuel asseslies even

. when considering irradiation growth.

4. The BSR states that cladding flattening is precluded during the projected exposum of the fuel. Provide the minimum tirre to collapse as calculated with the COLLAP code. What is the design maximum value of fuel assembly burnup?

RESPONSE: The clad flattening time is predicted to be 2 50000 EFPH for region 6 fuel using the current Westinghouse evaluation model.(I) Region 6 fuel, comprised of 6-1 and 6-2, has a projected residence time through 3 cycles of ~27000 EFPH.

5. What is the calculated minimum shoulder gap which allows for differential growth between fuel rods and the fuel asserbly? Provide the two Zircaloy growth correlations used in this calculation and describe or provide the data base from which these correlations were determined. How were the growth correlations combined with (a) fabrication tolerances, (b) dif-ferential thermal strains of the fuel assembly and reactor internals, and (c) elastic compression and creep of the guide thimble tubes? For steady-state operation, at what axially-averaged assembly burnup will

, interference result in rod bow?

RESPONSE : To insure that no axial interference between the fuel rod and the assembly nozzles can occur, the W Millstone Unit 2 fuel rod was de-signed so that the minimum roogemperature clearance value would be equal to or greater than 1.08 percent times the fuel rod length (1.583 a - --

inches, minimum) ,c. Extensive operating experience with other fuel assemblies using-the same rItaterials and having equivalent fuel duty has shown that this design value for clearance is conservative with mspect to temperature and irradiation induced length changes of the fuel rod and fuel assembly.

Therefore, under normal steady-state operatien, there will be no inter-forence of fuel rods since the growth allowance will preclude rod bow. ,

def.1 - WCAP-8381, July.1974, "Revisec Clad Flattening rodel"

WESTINGHOUSE PROPRIETARY class 2

6. The NRC staff has not commenced the review of the Westinghouse generic ,

topical report WCAP-8691, Revision 1, " Fuel Rod Bowing Evalsation",

which is mferenced in the BSR. Specifically, the BSR uses a formula

• from WCAP-8691 that projects anticipated rod bow magnitudes due solely to geometrical changes in the fuel rod thickness and diameter and spacer grid span length. This formula has been somewhat controversial an'd has not been accepted by the staff. Therefore, we will require that the degree of rod bowing in the Westinghouse reload fuel be calculated with the existing approved method, which is relatively more conservatt ve. In spite of this additional conservatism, however, we do not calculate a need for a DNBR penalty until an assembly burnup of 36,300 MWD /MTU is attained at which exposure the 50% gap closure value is reached. We mquire that Westinghouse confirm our calculations and verify that no other changes in fuel design variables (i.e., grid spring preload, degree of cladding cold work, etc.) are significant to the rod 'uowing extrapolation for the Millstone, Unit 2 reload fuel.

RESPONSE _: The degree of fuel rod bowing in the Westinghouse reload fuel has.been recalculated with the existing, more conservative approved nethod:

Lg (M.S.II)/Ig (M.S.II) '

= 0.59 l

15x15 /I 15x15 where L = span length I = cross-sectional morrent of inertia .

The average burnup at which a gap closure of 50% is attained is 32,000 MWD /MTU. By the time the fuel attains a burnup of 32,000 IGD /MTU, it is not capable of achieving limiting peaking factors due to the decrease i in fissionable isotopes and the buildup of fission product inventory.

This physical burndown effect is greater than the rod bowing effects which would be calculated at those burnups. Therefore the effect of md bow need not be considered in the analysis of the Millstone II core.

The fuel design variables were selected according to standard Westinghouse design practice. Therefere, the red ; tim  : tin metbd i t '.ali d.

L_ m m

WESTINGHOUSE PROPRIETARY class 2

7. The Combustion Engineering supplied fuel for Millstone, Unit 2 was ,

designed according to a specific set of Specified Acceptable Fuel Design Limits (SAFDLsl. Please list all of the Westinghouse SAFDLs for the Millstone reload fuel and pmvide the basis for omissions or additions to the original Combustion Engineering set of SAFDLs. .

RCSp0NSE : There are only two Specified Acceptable Fuel Design Limits (SAFDL's) covered by the Reactor Protection System (RPS), and these are discussed in Chapter 6 of the Basic Safety Report (BSR). These SAFDL's are:

1) The peak linear heat rate must be below that which would cause incipient fuel centerline melting. The melting point limit is con-servatively taken as 4700*F to bound the effects of fuel burnup and uncertainties in the melting point.

2) T.he DNB themal limits must not be exceeded.

The SAFDL's as defined by Westinghouse are equivalent, or mom conser-vative, when compared to the CE SAFDL's as given in Chapter 1 of CENPD-199:

a) The reactor fuel shall not experience centerline melt.

b) The departure from nucleate boiling ratio (W-3 DNBR) shall have a minimum allowable limit of 1.3.

Further discussion of fuel design criteria may be found in Chapter 2 of the BSR and Chapter 3 of the FSAR.

8. Some of the accident anlayses described in the BSR were performed with the computer codes FACTRAN (WCAP-7908, "FACTRAN, A Fortran IV Code for Themal Transients in a U0 Fuel 2

Rod") and LOFTRAN (WCAP-7907, "LOFTRAN Code Description"). Our review of these topicals has progressed to the point that there is reasonable assurance that the conclusions based on these analyses will not be appreciably altered by completion of the analytical myiew, and therefore that there will be no effect on the decision to issue a license amendment. If the final approval of these topical-reports indicates that any revisions to the analyses are required. Millstene Unit 2 will be required to implement the results of such changes.

=t-

WESTINGHOUSE PROPRIETARY CLASS 2 Resp 0tiSE : We do not expect that any revised analyses will be required as .

a result of your review of FACTRAN or LOFTRAN.

9. Please either reference or provide a thorough desc*iption of the Westing-house Corrputer Analyses Code (WECAN), which was used to perform the stress analyses of fuel assembly components.

RESP 0NSE: A description and benchmark problem solutions have been sub-mitted to the NRC via Reference 1 below.

Ref.1 - WCAp-8929, " Benchmark Problem Solutions Employed for Verification of the Wr.CAN Computer Program". April 1977.

11. The fuel rod support grid for Cycle 4 will be Inconel-718 rathar than Zircaloy-4 as used in Cycle 3. What are the effects of this material change on power distributions and other physics parameters?

RESP 0 :SE : The effects of the nuclear and thermal expansion properties of Inconel-718 grids in Westinghouse - supplied fuel assemblies and Zircaloy-4 grids in CE supplied fuel asserblies were considered in the evaluation of physics parameters (e.g., reactivity Coefficients) for the Millstone 2, Cycle 4 core.

Calculations of Fg(Z) include a multiplicative factor, applied to the axial ,

peaking factors, to ' account for axial inhomogeneities introduced by 4

csserbly grids. The inclusior. of the grid multiplicative factor bounds the inhomogeneities due to either Zircaloy or Inconel grids.

l

CEsTINGHOUsE PROPRIETARY CLASS 2

12. Power distributions calculated by TUR 1.E appear to undcrpredict thi power in the. peripheral asserrblies while overpredicting the power in the center asserblies. In view of the large errors in prediccing CEA worth near the core periphery due to these power distribution inaccuracies, justify the use of. TURTLE without some type of baffle correction scheme.

RESPONSE: As noted in the B5R, CEA-3 control bank worth on the periphery f of the core was underpredicted relative to measurement in Cycles 1, 2 and (

3. This results from a slight underprediction of the power in the peripheral 5 asser.blies.

I The BOL, HZP control worths wem calculated again with a baffle correction I of [0.9 x D)]a,c and an shown in Table 1 along with the original results.

As seen, the CEA-3 agreement is irc. proved by the baffle correction. The total control worth, 2-7, remains virtually unchanged as expected. The '

baffle corrt : tion will be used for Cycle 4 design.

4 l

l

. t 1

CESTINGHOUSE PROPRIETARY CLASS 2 i

Table 1 Effect of Baffle Correction on Cycle 3 CEA-3 Worth

% Difference

• CEA Worth, %An Measured No Corr. Baffle Corr.

7 0.64 0. 31 -1.56 6 0.25 -16.80 -12.00 5 0.17 -24.70 -13.53 4 0.88 - 8.64 -10.80 3 0.67 7.61 3.23 2 1.15 0.70 -0.35 Center (7-1) 0.03 -27.90 -21.67 Sequential Worth, 2-7 3.76 - 3.21 -3.72

• (Measured-predicted)/ measured

WESTINGHOUSE PROPRIETARY CL ass 2

13. The CEA drop analysis was performed without automatic rod control (CEA motion inhibit) or turbine load reduction. Is this the operational plan for Cycle 4 -

RESPONSE: CEA motion inhibit and turbine load reduction tend to mitigate the consequences of a CEA drop. Therefore, no credit was taken fo'r these functions in the CEA drop analysis. Turbine load reduction will not be operational; but the CEA motion inhibits will be available during Cycle 4.

14. The parameters used in the analysis of the CEA ejection accident appear to be representative of Westinghouse coms and differ from the previous Millstone, Unit 2 fuel supplier in several areas such as ejected rod worths, ejection time, delayed neutron fraction, feedback reactivity weighting, and power peaking (Fq ). Please provide a comparison between the Cycle 3 and Cycle 4 values of these rod ejection initial assumptions and discuss the reasons for and effects of any diffennces.

RESPONSE: Comparison of CEA ejection accident parameters are given in Table 1 for Cycles 3 and 4. '

In Cycle 3, the ejected rod worth is larger than the Cycle 4 value at HFP; at HZp the values are the same. The difference at HFP is probably due to the assumption used for control bank insertion prior to ejection.

As explained in Ref.1, the method used by Westinghouse in Cycle 4 is to assume that the rod is ejected from control bank CEA-7 [at the insertion limi t]. (a,c)

The delayed neutron fraction used in Cycles 3 and 4 are the same.

The Fg after ejection at HFP is slightly higher in Cycle 3 than in Cycle 4.

Differences in radial and axial power distribution due to the burnup characteristics and location of the fuel in Cycle 4 account for the di fference.

1. WCAP-9272, March 1978, " Westinghouse Reload Safety Evaluation flethodology"

. . . . . . . .. .~....~.~ ....... _ . . .

D

WESTINGHOUSE PROPRIETARY CLASS 2 At HZP, the Cycle 4 gF after ejection is larger than the Cycle 3 value. '

Again, some of the diffemnce can be accounted for by the change in

, fuel burnup and location between Cycles 3 and 4. Another important contribution is the axial shape assumed as a pre-condition. The Westing-house methodolog is to [ assume the HFP xenon distribution exists .at HZP. This produces a very skewed flux distribution towards the top of the core]a.c which accounts for the large value of Fgat HZP.

The CEA ejection time was assumed to be 0.1 seconds in Cycle 4. This value has no impact on the results campared to the ejection time of 0.05 seconds used in Cycle 3.

The feedback mactivity weighting used in Cycle 4 was applied only to the Doppler feedback, and conservatively accounts for the increased feedback due to the highly peaked power distribution following the CEA ejection. In addition, this weighting factor was applied to a conser-vative pmdiction of the zero to full power normal operation Doppler power defect of only 0.84% Ak. This methodology is described more fully in WCAP-7588 Rev.1-A and in reference safety analysis reports. Although the Cycle 3 analysis employed a spatial Doppler feedback weighting factor, the value used was not reported for that cycle. However, values of 1.24 to 1.34 wem reported fc= the full power cases and 1.94 to 2.52 for the zero power cases analyzed for Cycle 1 and reported in the Millstone 2

~ FSAR. These values are very close to the values used in the Cycle 4 analysis.

l WESTINGHOUSE PROPRIETARY CLASS 2 i

9

, Table 1 Comparison of CEA Ejection Accident Parameters HFP HZP Parameter Cycle 3 II) W Cycle 4 5) Cycle 3(I) W Cycle 4(3)

Ejected rod worth, Lk 0.29 0.17 0.65 0.65 Delayed neutron fraction 0.47 0.47 0.47 0.47 F after ejection q 5.83 5.70 14.5 18.8 Ejection time .05 .1 .05 .1 Feedback reactivity (4) 1.30 (4) 2.50 weighting

1. Letter, Counsil to Reid, Millstone Unit No. 2 Power Uprating, Feb. 12, 1979

2. WCAP-9660, BSR, February,1980

3. RSE, Millstone Unit 2, Cycle 4, May,1980

4. Value not reported i

I

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

, .-, . - , , .- -