Questions in the residual and SST model

 Register Blogs Members List Search Today's Posts Mark Forums Read

June 1, 2020, 08:24
Questions in the residual and SST model
#1
New Member

Sumin Park
Join Date: Jul 2019
Posts: 26
Rep Power: 3
Hello! I'm graduate student in Republic of Korea.

I'm analyzing the Dry Cask to cool down the spent fuel using CFX.

I set up the SST model and now the calculation is on going.

But in the residual monitor in CFX solver, there are some problems.

First, I don't know why the "wall boundary and scale" is empty.

Second, In the Turbulence residual, the value of RMS K-TurbKE was reach to 10^-100. I wonder whether there is no problem or not.

I attached two figures about those.

Please teach me the meaning of two situations.

In addition to those, I want to know the criterion of y plus in the SST model.

Thank you!!
Attached Images
 K-216.png (23.9 KB, 14 views) K-217.png (21.6 KB, 10 views)

 June 1, 2020, 20:03 #2 Super Moderator   Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 16,541 Rep Power: 128 Once the wall boundary scale converges it stops calculating that equation. So the display is blank for the iterations it is not calculated for. If a residual reaches 1E-100 it means that there is effectively zero happening in that equation. So I suspect it means the value of that equation is zero everywhere (or close to it), which means either the equation is not required (as it is doing nothing) or your setup is wrong (and the equation is not working properly). __________________ Note: I do not answer CFD questions by PM. CFD questions should be posted on the forum.

June 2, 2020, 04:04
#3
New Member

Sumin Park
Join Date: Jul 2019
Posts: 26
Rep Power: 3
Quote:
 Originally Posted by ghorrocks Once the wall boundary scale converges it stops calculating that equation. So the display is blank for the iterations it is not calculated for. If a residual reaches 1E-100 it means that there is effectively zero happening in that equation. So I suspect it means the value of that equation is zero everywhere (or close to it), which means either the equation is not required (as it is doing nothing) or your setup is wrong (and the equation is not working properly).

So, Can i understand that because the "wall boundary scale" is converged just first iteration, so it looks like empty?

 June 2, 2020, 08:25 #4 Super Moderator   Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 16,541 Rep Power: 128 That is possible. It could also have read in the wall boundary scale values from a previous run as an initial condition. __________________ Note: I do not answer CFD questions by PM. CFD questions should be posted on the forum.

 June 2, 2020, 08:45 #5 Senior Member   Join Date: Jun 2009 Posts: 1,301 Rep Power: 25 Your setup has multiple disconnected domains, and I assume it is somehow incorrect. Have you created the appropriate domain interfaces to allow the flow to pass through? Are you running with Beta Features on? Did you remove the "Constant Domain Physics" setting? Something is off __________________ Note: I do not answer CFD questions by PM. CFD questions should be posted on the forum.

June 3, 2020, 23:10
#6
New Member

Sumin Park
Join Date: Jul 2019
Posts: 26
Rep Power: 3
Quote:
 Originally Posted by Opaque Your setup has multiple disconnected domains, and I assume it is somehow incorrect. Have you created the appropriate domain interfaces to allow the flow to pass through? Are you running with Beta Features on? Did you remove the "Constant Domain Physics" setting? Something is off

Actually, I solved my simulation using laminar(turbulence-none) first. So, I think the connection between domains is correct.

I use the Beta Features to set up the different fluid in domains.

Finally, I don't check the "Constant Domain Physics" setting.

The solver is running. So, if the solver is finished, I will check that.

The problems that i faced with come from the "Beta physics"?

June 3, 2020, 23:11
#7
New Member

Sumin Park
Join Date: Jul 2019
Posts: 26
Rep Power: 3
Quote:
 Originally Posted by ghorrocks That is possible. It could also have read in the wall boundary scale values from a previous run as an initial condition.

 Tags cfx 14, turbulence analysis