Learning Forum

[AliminShafie]

 Hi. Im using Ansys 18.1 and conducting gird independence test for my jet flow (20-40 m/s) simulation on a curved surface. However as I increase my element quantity, the residual plot becoming hard to converge (as shown). At first I thought the problem was my mesh, but nothing changed even after I improve it by doing some geometry simplification. Any advice?

[Keyur Kanade]

Please check orthogonal quality. The min should be above 0.1. 

From image it looks you have refined mesh well. 

There are no universal metrics for judging convergence. Residual definitions that are useful for one class of problem are sometimes misleading for other classes of problems. Therefore it is a good idea to judge convergence not only by examining residual levels, but also by monitoring relevant integrated quantities such as drag or heat transfer coefficient.

 

Regards,

Keyur

 

If this helps, please mark this post as 'Is Solution' to help others.

Guidelines on the Student Community

How to access ANSYS help links

 

[AliminShafie]

Thanks for your reply. My orthogonal quality was over 0.9 while the min element quality was 0.1 (due to inflation).

Im aware of the need to monitor the integrated quantities as well as the residual. Does that means that I can disregard the residual if when the quantity of interest beginning to remain unchanged (or al least for some cases)? Since in my case the residual cant even reach 10^-5.

Or perhaps i could solve this problem by changing the input setting? I used k-omega SST with curvature correction, SIMPLE scheme with 2nd order discretization and the fluid properties was constant instead ideal gas since the flow is incompressible and at sea level

[Rob]

Not exactly.  We use a combination of fluxes, monitors and residuals to judge convergence. Where one of these doesn't show convergence we then look at why before deciding if the data is good enough.