[kandelabr]

Hello.

Does one need a y+<1 mesh for accurate calculation of wallHeatFlux or are alphat wall functions taken into calculation?

I've been trying to decode the wallHeatFlux::calcHeatFlux function but have no idea where alphaBf and heBf come from. I suspect snGrad() is just linear gradient from wall to the end of the first cell. In that case, if y+ >> 1 this can be much different than the actual gradient at the wall (that is in fact calculated by wall functions).

https://develop.openfoam.com/Develop...wallHeatFlux.C
wallHeatFluxBf[patchi] = alphaBf[patchi]*heBf[patchi].snGrad();

[Bloerb]

he is h or e the solution variable depending on your solver settings. (internal energy or enthalpy) and alpha is your effective conductivity field. This is thermal conductivity divided by the heat capacity (cp or cv depending on the solution variable h or e)

And to the best of my knowledge this should give you a heatflux that incorporates the effects of the wall function. Simply because the wall function modifies the values and hence you do have a different gradient.

For accurate heat transfer simulations (most importantly in CHT) you should however always aim for y+<1.

[kandelabr]

Thank you for your reply. Yes: I figured out that mesh requirement myself the hard way over the last three months .

By the way, what boundary condition should I use in case of y+<1 mesh for alphat?

Thanks again.