I am trying to modeling solidification using a single phase model in which the solid is simulated by using temperature specific source terms. As a result some locations are turbulent (liquid, about Tcrit) and other locations are laminar ('quasi-solid', below Tcrit). I have been using a k-epsilon turbulence model; however I am concerned that the heat transfer in the laminar regions is not accurate due to the k-epsilon model.

Does anyone have advice about how I can improve the thermal modeling in the laminar regions?

-Should implement a thermal phase change so that phase specific flow models can be used

-Or is it possible to turn off the k-epsilon model in another way.

Advice is greatly appreciated

Malcolm

Hi,

I don't know if it will help you, but take a look at: http://www-waterloo.ansys.com/cfxcom...minar_flow.htm

Rui,

That looks great!!

Thanks for the link.

Malcolm