Hi everybody,
I'm trying to model heat transfer by convection using the conjugate heat transfer node. For doing this I need to model a control volume to represent the domain of the surrounding air. As it is explained in one of the COMSOL tutorials (Natural Convection Cooling of a Vacuum Flask), the best way to do it is to start with a small control volume, and expand it until the results do not change (the dimensions of the control volume influences the model). I'm using the parametric weep option to solve the model for different dimensions of the control volume (I'm making a parametric sweep of the dimensions of the control volume). I already know that at some point, the solver will crash for some control volume, do to not convergence or high computational cost. The question is: it is possible with COMSOL to make a parametric seep and to keep the solver running, even if for some of the parametric values the solver crashes? Or something like this?
Thank you!
David
I'm trying to model heat transfer by convection using the conjugate heat transfer node. For doing this I need to model a control volume to represent the domain of the surrounding air. As it is explained in one of the COMSOL tutorials (Natural Convection Cooling of a Vacuum Flask), the best way to do it is to start with a small control volume, and expand it until the results do not change (the dimensions of the control volume influences the model). I'm using the parametric weep option to solve the model for different dimensions of the control volume (I'm making a parametric sweep of the dimensions of the control volume). I already know that at some point, the solver will crash for some control volume, do to not convergence or high computational cost. The question is: it is possible with COMSOL to make a parametric seep and to keep the solver running, even if for some of the parametric values the solver crashes? Or something like this?
Thank you!
David