ML18057A859

Benchmarking & Validation of In-House DOT Calculation Methodology.

ML18057A859
Person / Time
Site:	Palisades
Issue date:	03/31/1991
From:	CONSUMERS ENERGY CO. (FORMERLY CONSUMERS POWER CO.)
To:
Shared Package
ML18057A858	List: ML18057A857 ML18057A859
References
NUDOCS 9104240036
Download: ML18057A859 (22)

Text

BENCHMARKING AND VALIDATION OF IN-HOUSE DOT CALCULATION METHODOLOGY March 1991 Performed by the Reactor Engineering Department Palisades Nuclear Plant Consumers Power Company

_J

)

TABLE OF CONTENTS Section 1.0 . Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 2.0 Summary . . . . . . . . . *. . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 3.0 Benchmarking with Measured Data . . . . . . . . . . . . . . . . . . . 3 3.1 A-240 Surveillance Capsule Data . . . . . . . . . . . . . . . . . . . . 3 3.2 W-290 Surveillance Capsule Data . . . . . . . . . . . . . . . . . . . . 3 3.3 Ex-vessel Dosimetry Data (Preliminary) ................ 4 4.0 Validation .. * . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 5 .0 Further Benchmarking of In-House DOT Calculations . . . . . . . 11 6.0 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 7 .0 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 8.0 Appendices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 8.1 Final Report on Westinghouse Review of Consumers Power

.PTS Calculations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

LIST OF TABLES 3.1 Ex-Vessel Dosimetry Flux Measurements (Preliminary) . . . . . . 5

4. 1 Comparison of Cycles 1 Through 7 Flux . . . . . . . . . . . . . . . 9 4.2 Comparision of Cycle 8 Flux . . . . . . . . . . . . . . . . . . . . . . . . 10 ii

LIST OF FIGURES Figure 3.1 Location of the Palisades Surveillance Capsule Assemblies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 3.2 Ex-vessel Dosimeters Locations . . . . . . . . . . . . . . . . . . . . . 7 iii

1.0 Introduction Consumers Power Company developed in-house methodology utilizing the DOT 4.3 discrete ordinate transport code for the calculation of pressure vessel fluence. A report to the NRC was submitted describing fluence reduction measures for the Palisades Nuclear Plant reactor pressure vessel[ 11

• NRC communicated to Consumers Power Company that additional information regarding benchmarking, uncertainty analysis in the calculated vessel wall fluence and their effects on the plant end-of-life need to be addressed[ 21

• This report is being prepared to answer those questions. Benchmarking with the measured flux data from the Palisades surveillance capsules and preliminary ex-vessel dosimeter, is described. In addition, validation of the in-house method-ology is obtained by comparing the calculations with the independent calculations performed by Westinghouse for the Palis9des core geometry and Cycles 1 through 8 operations data.

1

2.0 Summarv In-house methodology calculates the spatial distribution of neutron flux in the reactor using the DOT 4.3 computer code (available from Radiation Shielding Information Center, Oak Ridge). The DOT Program solves the Boltzmann transport equations in two-dimensional geometry using the method of discrete ordinates. Third order scattering P3 and S8 angular quadratures were used.

Cycle-by-cycle neutron flux distributions were calculated using the cycle-dependent neutron sources and material compositions. This methodology is basically equivalent to that proposed in the draft Regulatory Guide131

• The results of this analysis predict that the calculated flux/fluence by in-house methodology are slightly higher than the measured data. Thus the calculated date 9/2001 111 to exceed the PTS screening criteria for the Palisades Nuclear Plant reactor pressure vessel is still valid. In the future, Palisades will continue to enhance it's flux calculation methodology by comparing with the measured data from either the surveillance capsules or supplemental dosimetry installed inside and outside the reactor vessel.

2

3.0 Benchmarking with Measured Data 3.1 A-240 Surveillance Capsule Data Surveillance capsule A-240 was attached just outside of the core support barrel (Figure 3.1). This surveillance capsule was removed at the end-of-Cycle 2 and was analyzed by the Battelle Columbus Laborato-ries[41. The neutron fluence of specimens within the capsule was deduced from the neutron induced activity of several iron wires from the capsule. The measured neutron fluence for neutron energies E > 1.0 MeV was determined to be 4.42 x 1019 n/cm 2

• On the top and bottom compartments of the capsule, the variation was less than 10%. In-house methodology predicts the calculated fluence at the center of capsul~ location as 4.53 x 10 19 n/cm 2

• This provides C/E ratio, calculated to measured fluence as 1 ;02. This shows that the calculated fluence by in-house methodology is slightly higher than the measured value for the A-240 capsule.

3.2 W-290 Surveillance Capsule Data W-290 capsule was located on the inner vessel wall (Figure 3.1 ). This capsule was removed at the end-of Cycle 5 and was analyzed by Westinghouse 151 . Measured flux was obtained from the radiometric monitors of iron, copper, nickel, titanium and uranium. Measured average flux at the W-290 capsule was 6. 73 x 10 10 n/cm 2 -sed 11

• A variation between the measured data among various dosimeters was 3

about 20.he calculated in-house flux at th9enter of W-290 capsule is 7 .02 x 1010 n/cm 2-sec. This provides a C/E calculated to measured flux ratio as 1.04. Similar to A-240 capsule data, W-290 wall capsule data shows that the calculated flux/fluence are higher then the measured values.

3.3 Ex-vessel Dosimetry Data (Preliminary)

At the end-of Cycle 7, ex-vessel dosimetry was installed outside the reactor vessel (Figure 3.2). This dosimetry has been removed at the end-of Cycle 8 and currently is being analyzed by Westinghouse.

Preliminary measured results are available 171 and summarized with the in-house calculations in Table 3.1. In general, calculated fluxes are conservative compare to measured fluxes. Seven more ex-vessel dosimeters are to be analyzed.

4

TABLE 3.1 Ex-Vessel Dosimetry Flux Measurements (Preliminary)

Flux (x 109 n/cm 2 -sec)

• Dosimeter 280° 290° 315° Locations (1 QO) (20°) (45°)

Measurement by Westinghouse (Reference 7) 1.426 1.094 0.714 Calculation by Westinghouse (Reference 7) 1.336 1.350 0.780 In-House Calculations 1.381 1.428 0.879 NOTE: Angles in parentheses correspond to 1 /8th core DOT model configuration.

• Seven more ex-vessel dosimeters are to be analyzed and fine tuning of these results may occur.

5

Accelerated Wall W-80 A-60 Reactor Vessel Wall W-100 Wall Core Sliroud W-110

~---Accelerated Wall W-290 A-240 Wall --------~~~~~'---~~~~.....~------Wall W-260 W-280 Plan View Core----- - - - - - - - - - - Accelerated Capsule Support Barrel Assembly

---

Thermal Capsule Assembly Reactor---~

Vessel Elevation View Wall C;ipsule Assembly FIGURE 3.1 LOCATION OF THE PALIS.ADES SURVEILLANCE CAPSULE ASSEMBLIES 6

tao*

lSO 0 270° 315°

• 0 Location of Gradient Chains D Location of Core Mid-9la.ne Dosimeters 0 Location of Core Mid-plane and Sot:tom of Core Dosimeters

• Preliminary Measured Data in Table 3.1 Fig~re 3.2 Ex-vessel Dosi~eters Locations 7

4.0 Validation In-house DOT calculation methodology has been validated by comparing with the Westinghouse methodology of flux/fluence calculations. Basically, the two methodologies are similar but there are some significant differences also. DOT geometry model, core size, reactor internal dimensions were developed independently from the plant drawings. Pre-processing codes for. the conversion of Cartesian (x,y) neutron source to (R,6) geometry are different.

Effect of plutonium burn up of the peripheral assemblies has been accounted for differently. Westinghouse has performed two sets of calculations for Cycles 1 through 7 combined case and Cycle 8 for the Palisades core geometry.

Comparison of vessel wall flux from in-house calculation methodology with Westinghouse has been presented in Tables 4. 1 and 4. 2, at the selected azimuthal angles. Tables 4. 1 and 4.2 show that there is a positive bias of approximately 12% in the flux calculations at the reactor vessel wall fluxes and it increases with the azimuthal angle. Prior to sending the fluence report to the NRC[ 11 , Westinghouse was contracted to perform an independent review of the in-house DOT calculation methodology. Comments of their review for Cycles 1 through 9 are presented in Appendix 8. 1.

8

TABLE 4.1 Comparison of Cycles 1 Through 7 Flux Flux (x 1010 n/cm2 -sec) oo 17° 30° 45° Axial Peak Axial Calculations Weld Flux Weld End Performed By: Location Location Location Ray In-house Methodology DOT Version 4.3 4.47 5.94 4.62 2.94 Westinghouse

. Methodology DOT Ill W (Reference 6) 4.23 5.62 4.17 2.64

• Difference +5.67% +5.69% + 10.8% + 11.4%

• ci> (In-House) - ci> (Westinghouse) x 100 ci> (Westinghouse) 9

TABLE 4.2 Comparison of Cycle 8 Flux Flux (x 1010 n/cm 2-sec) oo 17° 30° 45° Axial Peak Axial Calculations Weld Flux Weld End Performed By: Location Location Location Ray In-house Methodology DOT Version 4.3 2.08 4.87 2.31 1.81 Westinghouse

- Methodology DOT Ill W (Reference 6) 2.07 4.81 2.21 1.62

• Difference +0.48% + 1.25% +4.52% + 11.7%

• cl> (In-House) - cl> (Westinghouse) x 100 cl> (Westinghouse) 10

5.0 Further Benchmarking of In-House DOT Calculations For the present Cycle 9, Combustion Engineering has installed a dosimetry capsule at the W-290 capsule holder location vacated following Cycle 5.

Westinghouse has installed dosimetry outside the vessel. At the end of Cycle 9, both in-vessel and ex-vessel dosimeters are planned to be removed and analyzed. Combined in-vessel and ex-vessel dosimetry data will provide an excellent through wall correlation of the vessel wall flux. This approach will further enhance the in-house flux calculation methodology.

11

6.0 Summary Sections 3 and 4 clearly indicate that the flux/fluence calculated by in-house methodology are conservative. Calculated date of 9/2001 to exceed the PTS screening criteria at the axial weld locations of Palisades reactor vessel is valid.

12

7 .0 References

1. Letter and Attached Report (Reactor Vessel Fluence Analysis) from CPCo (R.W. Smedley) to NRC dated May 17, 1990.

2. Telephone conversation Between CPCo (R.W. Smedley, G.H. Goralski and 0.P. Jolly) and NRC (Brian Holian and Lambros Lois) on 10/25/90.

3. Status of a new Regulatory Guide on Methods and Assumptions for Determining Pressure Vessel Fluence. Paper presented by J.F. Carew, M. Todosow, (BNL), E:D. McGarry, J.A. Grundl (NIST); F.8.K. Kam, R.E.

Maerker and F.W. Stallman (ORNL) at the Seventh ASTM-EURATOM Symposium on Reactor Dosimetry, August 27-31, 1990, Strasbourg, France.

4. Battelle Report BCL-585-12, March 13, 1979 "Palisades Nuclear Plant Reactor Pressure Vessel Surveillance Program: Capsule A-240".

5. WCAP-10637, Analysis of Capsules T-330 and W-290 from the Consumers Power Company Palisades Reactor Vessel Radiation Surveillance Program, M.K. Kunka and C.A. Cheney, September 1984.

6. Letter from R.W. Smedley (CPCo) to NRC, "Docket 50-255 - License DPR Palisades Plant - Compliance with Pressurized Thermal Shock Rule 10CFR50.61 and Regulatory Guide 1.99 Revision 2 - Fluence Reduction Status (TAC No. 59970)," April 3, 1989.

7. Westinghouse Report (March, 1991 ), "Preliminary Results of Palisades Cavity Dosimetry".

13

8.0 Appendices 8.1 Final Report on Westinghouse Review of Consumers Power PTS Calculations 14

Westinghouse Energy Systems

• • CPAL-90-508 Box 355 Electric Corporation Plttsburgn Pennsylvania i 5230*Cl355 April 27, 1990 Mr. R. Klavon Ref.:~ G.O. DTE90841 Palisades Nuclear Plant CPAL P.O. CP11-7383Q Consumers Power Company 27780 Blue Star Memorial Highway Covert, Michigan 49043 Consumers Power Company Palisades Nuclear Plant Final Report in Support of the PTS Issue

Dear Mr. Klavon:

Attached is the Final Report to Consumers Power on the Review of your Neutron Transport Calculations in Support of the PTS Issue.

This completes the scope of the Westinghouse work on Consumers Power P.O.

CPll- 7383Q.

Please feel free to call me at (412) 374-3355, if you have any questions.

Very truly yours,

~-

J. C~ct

~~ Manager Customer Projects Department attachment ll, lA 538V:GLA/0427'90

Final Report on Westinghouse Review of Consumers Power* PTS Calculations E. P. Lippincott A review of calculations carried out by Consumers Power to determine the reactor vessel fluence rate for evaluation of pressurized thermal shock (PTS) concerns has been completed. The following Engineering Analyses

{EA) were incl~ded in the final review:

EA-P-PTS-89-015 DOT Calculation--Cycle 8 EA-P-PTS-90-001 Flux Calculations for Cycles 1-5 EA-P-PTS-90-005 Flux Calculations for Cycle 9 Additional calculations necessary for the preparation of input to the DOT calculations were previously reviewed and this review was documented in an interim report (Reference 1). In this report, the preliminary Cycle 8 DOT calculation was also reviewed and suggestions were made for improvement.

These suggestions have been incorporated in the updated calculation reviewed here. It is concluded that the present DOT calculations are based on a good model of the Palisades Reactor geometry and that the results are valid for determination of reactor vessel fluence and for comparison of fuel cycle design impact on vessel exposure.

The validity of the Consumers Power (CP) DOT calculations was checked in several ways. These are sununarized as follows:

I. Comparison with Westinghouse calculation.

A comparison with the Westinghouse (W} Cycle 8 calculation indicates that a bias exists that varies with azimuthal angle. As indtcated in the interim report, this bias may be mainly attributed to the more accurate model of the fuel element geometry which results in a slightly larger core. In other studies, an approximate value of 13i flux increase per cm increase in core radius was found. This is about equal to the average increase in flux for the CP model. In addition, the fact that the core size increase extends in both the x and y directions will result in a larger increase in the core radius at 45 degrees than at 0 degrees, which p~rtially accounts for the increased bias between the calculations at the larger angles. Additional variation in. the bias between the calculations appears to be due*to other differences in the models that relate to details of the core edge representation and in locating the fuel source.

As noted in the interim report, the CP modeling of the source using the PTHETA code can possibly lead to a bias because the source is placed entirely in the volume element containing the center of each fuel pin.

The modeling differences other than the core size difference all fall within the expected accuracy of this type of transport calculations. Thus it is concluded that the CP calculation is basically consistent with the Westinghouse Cycle 8 result.

0813I:EPL/js

. -~

J

2. . he Comparison w1t Measurement **

The Cycle 1~5 calculations were made to benchmark the calculations with the dosimetry measurements- made on capsule W-290 which integrated the fluence over these 5 cycles. The calculational result was 4% higher than the measured fluence value which is agreement to well within the calculational and measurement error. It is concluded that the calculations well represent this measurement and produce a slightly conservative value. It should be noted that Westinghouse typically corrects the measured dosimetry values to the center of the surveillance capsule and derives a fluence value at that point. The fluence value averaged over the capsule is almost exactly equal to the value at the center.

3. Comparison of Cycle Calculations.

A comparison was made between the Cycle 9 and Cycle I calculation to determine if the calculations are giving a qualitatively correct estimate of the change expected between a fresh core and a low leakage core. The fluence outside the core is mainly determined by the power in the peripheral assemblies. Taking an average of four peripheral assemblies (assemblies 3, 9, 12, and 15) the ratio of power for cycle 1 to cycle 9 is 2.3. The ratio of flux (E>l MeV} for points at the inside of the core barrel at 0, 16.5, 30, and 45 degrees averages to 2.2, which is very close to the power ratio and would be even closer if changes due to burnup were included in this simple analysis. Similarly at the vessel IR the average ratio for these points is 2.1. Thus it 1s concluded that the calculations are correctly reflecting the power differences in the peripheral

~ssemblies in the leakage flux calculations and predictions for the cycle 9 fluence are valid.

The general conclusion of this review is that Consumers Power has demonstrated through comparison with measured results and through consistency checks that calculations can be carr;ed out with a high degree of confidence in the results. A number of major and minor details have been included ;n the establishment of the calculational capability and in its verification. Further comparisons w;th measurements are reco11111ended as they become available to ensure continued validity as chang;ng core patterns are calculated, and to confirm the single W-290 capsule measurement point. It is concluded that at present the Consumers Power calculations represent the best (slightly conservative relative to the capsule measurement) value to use in assessing the reactor vessel neutron exposure.

08131:£PL/js

- ~ .

• ~

The c~lculations for Cycles 8 and 9 have indicated the flux reduction that tan be achieyed with two different fuel schemes. In cycle 8, a* stainless steel pin region was inserted in place of fuel in two assemblies in each octant. This resulted in a significant decrease in flux to the vessel welds. In cycle 9, a low leakage fuel pattern was introduced that resulted in about the same flux to the weld at O degrees but reduced flux at other locations. This pattern did not require any stainless steel rods, but used poison rods in the assemblies critical to weld exposure.

Other reactors have adopted this. type of low leakage pattern to reduce

• vessel exposure and in some cases to improve fuel efficiency. It is probable that Palisades can refine this type of fuel pattern to maintain the low leakage while reducing costs and extending the cycle length.

Further flux reductions through fuel management will not be easily obtained.

Reference 1: E. P. Lippincott, "Interim Report on Westinghouse Review of Consumers Power PTS Calculations", August 1989.

0813I:EPL/js