I have having some problems with a a model I have set up. I am runing v3.3 and I have set up a time dependent model outline below.

A heat source (flux varing in time)in a room. The room has adequate venting, both for venting air and make up air.

When I move the heat source around the room, the model crashhes. It works in some locations and not others. The models that do work converge satifacorly - for hte problem at hand.

I have kept a consitent approach to meshing and relaxation parameters.

Does anyone have any ideas.

Any help would be greatly appreciated.

Leon

There is very little to go on here, but if the density is a function of pressure and temperature through the ideal gas law, then a common cause of crashes is when the density goes negative. The density should be stored and linearly relaxed, and limited through VARMIN to prevent unrealistically low densities. Likewise, the energy equation should be adequately relaxed by using inertial relaxation such that the temperature rise at the heat source does not exceed say 10degC per sweep at a given time step. If density variations are large, as say in a fire representation, then set DENPCO=T so as to activate a form of the pressure-correction equation that accounts for strong spatial variations in density.