ML20205E525

Topical Rept Evaluation of WCAP-10107(P), Condor, Thermal-Hydraulic Code for Bwrs. Rept Acceptable for Referencing in Licensing Actions

ML20205E525
Person / Time
Issue date:	10/11/1985
From:	Office of Nuclear Reactor Regulation
To:
Shared Package
ML20205E478	List: ML20205E474 ML20205E525
References
TAC-48567, NUDOCS 8510170320
Download: ML20205E525 (8)

Text

.

- ENCLOSURE SAFETY EVALUATION REPORT'ON THE WESTINGHOUSE CONDOR CODE FOR THERMAL-HYCRAULIC ANALYSES OF BOILING WATER REACTORS (TACS 48567)

Report Number: WCAP-10107(Proprietary)

Report

Title:

CONDOR-A Thermal-Hydraulic Code for Boiling Water Reactors Report Date: June 1982

~

Originating Organization: Westinghouse Reviewed By: Core Performance Branch / Pacific Northwest Laboratory

1.0 INTRODUCTION

In WCAP-10107 (Refs. I and 2), Westinghouse presented the CONDOR computer code which will be used to calculate the axial distribution of coolant flow, enthalpy, pressure, void fraction and other associated parameters in a BWR core under steady-state conditions.

The CONDOR code solves the steady-state conservation equations of mass, momentum, and energy in one dimensional channels. Since it only' solves the steady-state equations, it can not calculate the thermal-hydraulic conditions during transients. The code takes the total core flow, reactor power, inlet enthalpy, system pressure, axial and radial power distributions, core geometries and component pressure loss coefficients as inputs for the boundary conditions to the numerical solution, and then iterates on the flow distribution among the channels until a converged solution is obtained. The constitutive relations for the void fractions, friction losses, two-phase pressure drop and bypass flow are also discussed in the report (WCAP-10107).

$k c

k l

I

l l

.s 2.0 STAFF REVIEW AND EVALUATION .

The objective of this review is to evaluate the validity and applicability of the CONDOR code for use in thermal-hydraulic calculations for licensing applications. The review includes the evaluation of the hydraulic models, numerical schemes, code verification and assessment of the limitations arising from modeling assumptions.

_. The results of our review are discussed in the following sections.

2.1 HYDRAULIC MODELS 2.1.1 Conservation Equations The mass continuity used in CONDOR is satisfied by having the sum of the flows from all core channels equal to the total flow to the core. No cross flow between heated channels in the core is assumed. The energy balance is achieved by requiring that the energy released to a given channel axial node divided by the flow rate is equal to the enthalpy rise for that channel node.

The heat transfer across the fuel channel wall from the active flow to the bypass flow is accounted for in the energy balance. We find that the approaches used for mass and energy conservation in the code'are conventional for modeling a BWR core and are acceptable. However, during the course of the review we requested that a sensitivity study be performed to establish the accuracy in the calculation of the total pressure drop and the enthalpy rise with respect to the axial node size.

In response, Westinghouse submitted the results of a sensitivity study (Ref. 8) indicating that the changes in the calculated total pressure and minimum critical power ratio are small (less than 0.26% and 0.9%, respectively) when the number of axial nodes is reduced from 24 nodes to 12 nodes. Westinghouse indicates that since the CONDOR code used 24 nodes for code verification, this same nodal scheme will be used for licensing calculations. Westinghouse also agrees that 2

==

if a different number of nodes is used, additional calculations will be performed to identify the uncertainties associated with the change of the number of nodes. We find that the Westinghouse responses adequately address the concern related to the sensitivity of number of nodes on the CONDOR calculations and are acceptable.

The CONDOR axial momentum equation (Equation 3.1) expresses the total axial pressure drop in terms of the pressure drops in elevation, acceleration,

. friction, and local form loss. We find that axial area variations are not considered in the acceleration term. However, this will not result in any deviation in channel pressure drcp and flow calculation because of the absence of flow exchange between channels in a BWR core.

In summary, the steady-state conservation equations of mass, momentum, and energy given in the report (WCAP-10107) are fonnulated correctly for the intended use of the code and are acceptable.

2.1.2 Bypass Flow Models The bypass flow is not solved through the rigorous differential momentum equations. Instead, an empirical correlation (Equation 3.8) establishing the bypass flow in terms of the pressure differences across the leakage paths is used as follows:

C W= C1+ cap2 1/2 + C3a P 4, This approach is acceptable for the steady-state calculations of CONDOR when the dominating driving force for the bypass flow is the pressure differences across the leakage paths. However, the accuracy of using this approach depends on the selection of the constants C , C , C , and C . These constants 1 2 3 4 have to be determined through comparison with experimental data for the prototypical assemblies with bypass geometries. Therefore, we require that the determination of these constants be made through comparison with appropriate 3

m _ _ _ _

test data on a plant-specific basis. Factors affecting these values, such as the crud buildup, geometry changes due to irradiation, and mixed core due to fuel designs provided by different vendors, have to be considered.

With the coefficients Cy to C carefully selected and justified through 4

comparison with appropriate test data, we conclude that the bypass flow model used by the COND0R code is acceptable.

2.1.3 Yoid Models The void model in both subcooled and bulk boiling regions is based on a modifi-cation of Zuber's drift flux model. The formulation is the Zuber model (Ref. 3) with coefficients adjusted to give agreement with test data. In the subcooled

~

region, the voids are predicted using the Zuber model in conjunction with the effective flow quality for the subcooled region obtained from the Levy correlation (Ref. 4). These models give acceptable results for void fractions for both sub-cooled and bulk boiling. This was verified by the comparisons of the void models with five FRIGG data sets discussed in the topical report.

We conclude that the void models are acceptable.

2.1.4 Pressure Drop Correlations The single ~ phase friction factor, f, is based on the Blasius equation. The multiplier for the two-phase pressure drop is based on the modified Baroczy (Ref. 5) and Chisholm (Ref. 6) correlations. Data comparison by Westinghouse with FRIGG experiments showed that these correlations are satisfactory.

The single and two-phase form losses are based on a multiplier, k, to the velocity head and a homogeneous two-phase multiplier to k. These models are adequate and acceptable. Westinghouse indicates that in the actual coding of CONDOR, another factor is multiplied to k to account for the effects of crud buildup at the grid spacers on pressure drop. As a result of our review we 4

. l l

require that the selection of the form loss coefficients with the effects of the crud buildup, geometry change due to irradiation, different fuel designs by different vendors, be considered on a plant-specific basis.

In sumary, the pressure drop correlations in the CONDOR code are acceptable provided that the multiplier accounting for the effects of crud buildup and geometry change on the pressure drop are considered on a plant-specific basis.

2.2 NUMERICAL SOLUTION TECHNIQUES In the CONDOR solution, the core and bypass regions are divided into a given number of coolant channels with given inlet conditions. The power to each channel is given. Each channel is axially subdivided into a number of nodes.

The code then iterates on the flow distributions among the channels until a converged solution is obtained.

We conclude that the solution scheme is a common scheme for BWR core flow calculations and is, therefore, acceptable.

2.3 CODE VERIFICATION CONDOR predictions were compared with two analytical solutions. One was for the case of homogeneous equilibrium two-phase flow in a vertical heated tube with uniform heat flux. The other is for the case with a cosine-shaped heat flux. To do this, the void fraction expression and two-phase friction multiplier expression used in the analytical model were inserted into the CONDOR code.

There is a small difference between the analytical solution and the COND0R evaluations. This difference results from the different calculational methods:

CONDOR evaluates the local steam / water properties using the local coolant pressure of the calculational nodes whereas the analytic solutions are based on evaluation of these properties at a given constant system pressure. The effect of this difference in calculation is insignificant 1y small at typical BWR steady-state operational conditions, because the pressure variations along a channel are usually very small when compared to a system pressure of 1000 psia.

5

The good comparison between the analytical solutions and the CONDOR predictions shows that the numerical solution in the code is correct in predicting the total and component pressure drop for a given inlet flow rate.

In addition to the comparison with analytical solutions, CONDOR predictions were compared with the P1 process computer output. The results showed that the CONDOR predictions matched the P1 computer output closely. However, the P1 results were not measured data; instead, they were based on a numerical cal-

._ culation with given boundary conditions such as core power, total flow, system pressure, and inlet enthalpy similar to CONDOR. Therefore, they are not a truly independent source for comparison purposes. However, the benchmark of CONDOR with P1 results may be combined with the independent check of the void model, the pressure drop calculations, and the analytical solution, as discussed in the CONDOR topical report, to verify that the code is able to calculate the flow and enthalpy distributions satisfactorily.

In summary, we conclude that the verification calculations showed that different models in CONDOR performed correctly, the void and pressure drop correlations compared favorably with experimental data and comparisons of CONDOR predictions with analytical solutions verified the numerical and programming algorithms.

3.0 CONCLUSION

S Based on the review which is described above we conclude that topical report WCAP-10107 is acceptable for referencing in licensing actions by Westinghouse with respect to the steady-state thermal-hydraulic performance of a BWR. The following restrictions apply to the use of the CONDOR code:

(1) The CONDOR code is claimed, in the topical report, to be able to perform BWR loop calculations. However, no description of loop modeling is given.

Therefore, the current version of CONDOR is restricted to the calculation of core flow and enthalpy distribution.

l 6 1

(2) CONDOR does not have a verified CHF correlation in the code at this time. Any correlation to be incorporated in the code for MCPR licensing analysis has to be reviewed and approved by the NRC separately.

(3) Since the core bypass flow calculation is based on a simplified correlation with AP as the independent parameter, the correlation coefficients should be determined by comparing with the test data on a plant-specific basis. Factors affecting the coefficients, such as mixed core with fuels from different vendors, the crud buildup, and

-irradiation effects have to be considered.

(4) The CONDOR code uses 24 nodes to represent a flow channel for code verification, this number of nodes should be used for licensing cal-culations. If any reduced number of nodes is used, additional calculations should be performed to identify uncertainties associated with the reduced number of nodes.

(5) Selection of the loss coefficients with the effects of the crud build-up, geometry change due to irradation, different fuel designs by ,

different vendors should be considered on a plant-specific basis.

(6) During the course of our review, we raised questions regarding the proposed models and data. Westinghouserespondedtotbesequestions in Reference 7. These questions and answers should be included (Ref. 2) in the final topical report on CONDOR submitted by Westinghouse.

i 7

4.0 REFERENCES

1. C. A. Olson, " CONDOR-A Thermal-Hydraulic Performance Code for Boiling Water Reactors", WCAP-10107, June 1982.

2. C. A. Olson, "COND0R-A Thermal-Hydraulic Performance code for Boiling ,

Water Reactors", WCAP-10107 (Rev.1), December 1983.

3. N. Zuber, et al., " Steady-State and Transient Void Fraction in Two-Phase

_ Flow Systems", Final Report, Vol. 1, GEAP-5417, January 1967.

4. S. Levy, " Forced Convection Subcooled Boiling Prediction of Vapor Volumetric Fraction," GEAP-5157, April 1966.

5. C. Baroczy, "A Systematic Correlation for Two-Phase Pressure Drop,"

Heat Transfer Conference (Los Angeles), Chemical Engineering Program Symp. Series No. 64, Vol. 62, 1966.

6. D. Chisholm, " Pressure Gradients Due to Friction During the Flow of Evaporating Two-Phase Mixtures in Smooth Tubes and Channels",

Intl. J. Heat & Mass Transfer, Vol.16, pp. 347-358,1973.

7. LetterwithAttachmentfromE. Rahe (Westinghouse) toc. Thomas (NRC),

dated December 1983.

8. Letter from E. Rahe (Westinghouse) to C. Thomas (NRC), dated July 23, 1984.

8

_