ML20135D838
| ML20135D838 | |
| Person / Time | |
|---|---|
| Site: | 05200003 |
| Issue date: | 12/03/1996 |
| From: | Mcintyre B WESTINGHOUSE ELECTRIC COMPANY, DIV OF CBS CORP. |
| To: | Quay T NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| NUDOCS 9612100187 | |
| Download: ML20135D838 (3) | |
Text
r j
. \\
g zy Q
Westinghouse Energy Systems Ba 355 Electric Corporation Pinstugh Pennsylvania 15230-0355 December 3,19%
NSD-NRC-96-4894 DCP/NRC0667 Docket No.:STN-52-003 Document Control Desk U.S. Nuclear Regulatory Commission Washington, D.C. 20555 ATTENTION:
T.R. QUAY
SUBJECT:
REQUEST FOR INFORMATION ON NRC MODEL FOR THERMOPHORETIC REMOVAL OF CONTAINMENT AEROSOLS IN THE AP600
Dear Mr. Quay:
In an October 9,1996 letter, in response to a Westinghouse request, NRC provided information on the thermophoresis equations used in a Sandia National Laboratories (SNL) report, " Monte Carlo Uncertainty Analysis of Aerosol Behavior in the AP600 Reactor Containment." After reviewing this material, we find that additional information is required in order to understand why there is such a large discrepancy between the SNL calculations of thermophoretic deposition in the AP600 containment and those performed by Polestar Applied Technology for Westinghouse. As noted below, we would like to obtain from you a description of the modeling used by SNL to calculate the wall temperature gradient (i.e., the temperature gradient at the interface between the gas boundary layer and the liquid film on the AP600 containment shell) that is used to calculate the thermophoretic removal of aerosols and the results of this wall temperature gradient calculation for the AP600.
We have reviewed the models used in the SNL report for aerosol sedimentation, turbulent deposition, diffusiophoresis, and thermophoresis and we are in general agreement with these models. However, there is a significant difference in the values for thermophoretic deposition velocity. The difference between the SNL and Polestar results is clearly shown by referring to Figure 8 of Part 1 of the SNL report where the thermophoretic deposition velocity is reported to be approximately three orders of magnitude lower than the diffusiophoretic deposition velocity. (Figure 8 is described in the SNL report as a " single, typical caelulation from the Monte Carlo uncertainty analysis" for AP600.) The Polestar calculation, on the other hand, indicates that the deposition velocities for thermophoresis and diffusiophoresis are of the same relative magnitude. Also, the Polestar calculation shows that the KO deposition velocities (for both thermophoresis and diffusiophoresis) are roughly equivalent to the values for the diffusiophoretic deposition velocity reported on Figure 8 of the SNL report. This difference in the calculation of thermophoretic removal may explain, at least in part, the differences between the SNL and Polestar reported results for total aerosol removal coefficient.
9612100187 961203 PDR ADOCK 05200003 A
PDR 080152
1 Page 2 -
December 3,1996 He deposition velocities for thermophoresis and diffusiophoresis are each proportional to their respective driving forces. Dese are sensible heat transfer rate for thermophoresis and condensation heat transfer rate for diffusiophoresis. For the AF600 design, these two proportionality constants t
(determined by the properties of the gases and particles) are found to be of the same order.
Consequently, the Polestar determination that the two deposition velocities are approximately the same is consistent with the fact that the condensation hear transfer rate and the sensible heat transfer rate are roughly the same.
)
It appears that the factor most likely to be causing the large variation between the Polestar and the SNL values for thermophoretic deposition velocities is the wall temperature gradient. In calculating i
the wall temperature gradient, Polestar used a relatively straightforward approach which was discussed in our August 5,1996 letter report to the NRC. This approach is consistent with the j
following relationship from the first equation in the October 9,1996 NRC letter:
l l
g AH(f,)m(H O) k VT 2
=
A A
8 Polestar used this equation to calculate the wall temperature gradient (i.e., VT). The values for Q*
(the total heat transfer rate), m(H O) (the condensation rate), and AH(f,) (the latent heat of water) 2 were based on the same calculation which generated the containment temperature and pressure inputs for the analysis. This choice was made out of the decision to maintain consistency among the thermal i
hydraulic parameters used in the analysis.
In the Polestar calculation, if the condensation and sensible heat transfer rates are taken to be approximately the same (as predicted in the AP600 containment calculations), the wall temperature gradient can be estimated as:
VT=
2A k, 2
With k, = 0.03 w/mK, Q* = 2x10' watts, and A = 7000 m, the wall temperature gradient is approximately 5 x 10' K/m.
The first equation in the October 9,1996 NRC letter also includes the following relationship:
h *A T= k,VT e
4 Page 3 December 3,1996 1
However, use of this equation would necessitate recalculating the heat transfer coefficient which is a complex calculation and, if this approach is taken, it would be advisable to check the results for consistency against the AP600 decay heat rate since this is the major driving force for heat transfer (and is the only driving force after the transient energy effects from blowdown and zirconium - water I
reaction during core degradation have completed).
l It would be useful if the NRC would provide the specifics on how the wall temperature gradient was calculated in the SNL report and the values for the wall temperature gradient calculation. If a method different from the identified Polestar approach has been used, the results should be checked for consistency with the AP600 design heat transfer rates.
During the AP600 scheduling meeting on November 20-21,1996, it was agreed that Westinghouse and the NRC would meet prior to January 15,1997 to reach closure on the Lambda values, it is therefore essential that the NRC respond to this request for information prior to December 31,1996.
Please contact John C. Butler on (412) 374-5268 if you have any questions concerning this j
transmittal.
.n. f $
B, A. McIntyre, anager Advanced Plant Safety and Licensing
/jwh cc:
T. Kenyon, NRC i
R. Emch, NRC D. Powers, SNL N. J. Liparulo, Westinghouse
,