ML20046D437
| ML20046D437 | |
| Person / Time | |
|---|---|
| Issue date: | 06/29/1993 |
| From: | Office of Nuclear Reactor Regulation |
| To: | |
| Shared Package | |
| ML20046D435 | List: |
| References | |
| NUDOCS 9308200027 | |
| Download: ML20046D437 (3) | |
Text
. - _ - - - - - _ _ - _ _ _ - - _
O r
s UNITED STATES
\\
[(4 i
NUCLEAR REGULATORY COMMISSION 3
WASHINGTON. D. C. 20555 f
8 oy'..
f SAFETY EVALUATION BY THE OFFICE OF NRR RELATING TO TOPICAL REPORT WCAP-10484. ADDENDUM 1. " SPACER GRID HEAT TRANSFER EFFECTS DURING REFLOOD"
1.0 INTRODUCTION
By letter dated December 23, 1992, Westinghouse Energy Systems submitted an addendum to the subject report.
The primary purpose of the addendum was to correct a programming logic error discovered in the spacer grid heat transfer model subroutine contained in the BART code. This error affected all versions of the Westinghouse 1981 ECCS Evaluation Model which incorporate BART or BASH methodology.
Specifically, the logic error caused the grid heat transfer l
calculation to be performed twice instead of once each fuel time step iteration. Because the grid temperatures are reset and other parameters are changed after the first loop, the resulting grid temperature array deviates from the intended values.
In addition to correcting the code logic, Westinghouse has also made the following minor modifications to the spacer grid heat transfer model:
- 1) To obtain a better representation of the axial temperature profile, the number of grid nodes was increased by decreasing the axial mesh size from 0.0036ft to 0.0020ft.
- 2) To insure convergence of grid temperatures when steep axial temperature gradients exist, the sub-timestep for the grid model calculation has been limited to the thermal time constant corresponding to a lumped parameter transient conduction solution for a single node.
- 3) A refinement has been made to the calculation of the liquid film evaporation term in the grid mass balance.
930B200027 930629 PDR TOPRP EMVWEST PDR
I.
O o
i
! 1 These modifications have been made to improve convergence and consistency in.
{
grid model behavior under conditions where flow oscillations and surges can The overall performance result in repeated wetting and dryout of the grids.
of the model and the predicted results are not, however, significantly affected by these revisions.
It should be noted that no changes have been made to basic model or governing equations presented in WCAP-10484-P-A.
2.0 EVALUATION For purposes of validating the spacer grid heat transfer model, WCAP-10484-P-A presented comparisons of LOCBART code calculations for representative 3-loop.
and 4-loop plants incorporating the grid model, to data generated in rod bundle reflood heat transfer experiments performed by Westinghouse at their The experiments were selected so as to bound the FLECHT and G2 facilities.
range of conditions anticipated during a reflood transient.
l The logic error described above was present in the code versions used in these l
To re-validate the corrected grid model, including the validation studies.
minor modifications discussed above, code calculations were re-performed for As before, two i
the experiments selected in the original validation studies.
I calculations were performed for each of the experiments; one with the grid model included, and one without the grid model, in order to assess grid performance.
a The reported results indicate that the effect of correcting the programming e
logic error in the grid heat transfer model generally improved the ability of the grids to become wetted during reflood.
This, in turn, contributed to enhanced cooling of fuel rods and reductions in calculated peak clad r
In cases where the erroneous code logic previously temperatures (PCT).
predicted poor grid wetting, the corrected model produced substantial PCT On the other hand, in those cases where the-reductions on the order of 100F.
original calculations predic.ted extensive grid wetting, the corrected model
I l
h i producea only a small effect on the results.
Comparisons of predicted PCTs obtained using the corrected grid model with the experimental data employed in the original validation studies indicate that model predictions, overall, remain conservative.
For production type mixing For simple egg crate vane grids, the model consistently overpredicts PCTs.
(non-mixing) grid designs, the model predicts the PCTs at around the mean of the data with a slightly conservative capture fraction.
3.0 CONCLUSION
We have reviewed Addendum 1 to the subject topical report (WCAP-10484-P-A) and find that Westinghouse has adequately re-validated the spacer grid heat transfer model which had been corrected to eliminate a programming logic error The licensee has demonstrated that the corrected model discovered earlier.
Addendum 1 may therefore continue: to produce generally conservative results.
be referenced in future licensing analyses.