Inconsistent Total Heat Transfer Rate

Ken Adams

Hello all,

I modeled a steady-state natural convection problem involving a heatsink and surrounding air using segregated solver. After residuals converged, I checked Flux Reports. Mass Flow Rate over the boundaries of the air domain was to the order of 10e-4, which makes sense. Total Heat Transfer Rate over all the boundaries of the heatsink is around 4W, which does not seem right, compared with the total input power of 23W. However, the total heat transfer rate calculated by Surface Integrals gives 0.2W. Could anyone explain what may cause the inconsistency? I read in Fluent 6.2 manual that Flux Reports gives more accurate results than Surface Integrals. So does that mean there is something wrong with the simulation? Some information of the model is listed below:

1. Segregated solver. Steady-state

2. Discretization scheme: Body Force Weighted for Pressure. Second Order Upwind for Density, Momentum, and Energy.

3. Ideal gas model for air and pressure inlet and pressure outlet on the boudaries of the air domain.

4. Simulation was iterated with Fluent 6.2 for 3,000 times. Residuals and total heat transfer rate over all surfaces were monitored. After all iterations, all residuals converged and total heat transfer rate leveled off.

Thanks in advance,

Ken