Under Relaxation Factors
 

Hi,

I have a fluent model with predominantly conduction and radiation only. there are only solid zones currently in my model.

I am trying to compare my results with Patran since there is no flow in my model at least for now.

Any comments/suggestions on the following questions is greatly appreciated.

1. Does the under relaxation factor play a role in arriving at the final temperatures during a steady state simulation?

2. From my understanding, the under relaxation factor plays a role only in transient simulations!. Am i correct?

3. Do you think the under relaxation factors for energy could be the reason for different results between Patran and Fluent for a transient simulation?

Thank you

Venigall

Re: Under Relaxation Factors
 

Hi,

I will try to answer your question:

1. No, if the simulation arrives to convergence they should not change the final results.

2. They play a role during the convergence of the calculation towards the solution. If transient is a physical phenomenon which start from time (physical time) zero and brings to a steady state situation underelaxation factors should not play a role in the results obtainable for the different transient instants (time steps). They act during the iterative procedure which brings to the converged solution for each time step.

3. The problem can rise from an erroneous choice of the underelaxation factor. For energy, as far as I know from my experience, a small underrelaxation factor can freeze the convergence and, as a consequence, influence the final resul. This sentence is not saying the contrary with respect to points 1. and 2.! The important condition is that the underrelaxation factors choice is a subtle topic and requires some experience on the specific physical case. Practically this means try different cases and look to the effects.

I hope this helps.

Cheers,

MCB