[manuc]

Hello all


I have a bottom heated top cooled cavity with conducting packed bed of sphere in side. The Rayleigh number based on box height is 10^8. I am comparing my numerical results with the experimental data.
What do I compare

Nusselt_expriment is determined from the power input required to maintain constant bottom wall temperature

h=power/(Area *(Th-Tc))
Nu_exp=h*L/kf
where Th,Tc are hot and cold wall temperatures, Kf =fluid conductivity.

It is compared with Nusselt number obtained from simulations calculated as
Nu_sim= sngrad(T)*L/(Th-Tc)
The Nusselt number so calculated is equivalent to assuming the fluid with conductivity 1. In order to compare it with the experiments
Nu_f_sim =Nu_sim*1/kf .

When I compare Nu_exp and Nu_f_sim,the Nu_f_sim is almost off (high) from experiments by 50%.

Could some one suggest possible reasons for such over prediction.

(Note: No turbulence models are used.The flow is fully resolved. The flow at the chosen Ra is assumed to be laminar. Behaviour is same for even fine grids.).