Learning Forum

[Johannes]

Hi,

during the simulation (Fluent 19.5) I get the warning that temperature was limited to 1 K in a couple of cells. So I stopped the simulation and tried to find the area by using the Mark/Adapt Cells by Iso-value function (I switched to the 19.2 GUI for that function).

According to the dialog box, the temperature range is from 1 K to 700 K, but when I try to mark the cells from 0 K to 270 K (as the temperature should actually not be below 280 K) no cells are marked. The same with cells of temperature upwards from 300 K. So I wonder if the warning is wrong or how I can identify where those cells are.

Thank you in advance for your help

Johannes

[Karthik Remella]

Hello,

Could you please try to plot the following and let us know?

Cell Registers -> Field Variable

Change the Type to Cells in Range (please see the screenshot).

Let me know what this says in your case.

Thanks.

Karthik

[Johannes]

Dear Karthik,

Thank you for your answer! Here is what I got:

In the console, 3 cells are limited to 1 K, and when I use the compute button in the Field Variable Register, it also shows me the range between 1 and almost 700 K. However, trying to mark the cells between 0 and 10 K for example results in 0 cells marked (see screenshot).

Don't know if I need to mention that I use a polyhedral mesh, almost 10,000,000 cells in parallel mode, also using shell-conduction. My inlet temperature is 281 K (outdoor air flow around buildings) including radiation. The temperature should be in the range of 280 K to maybe 300 K. When I try to mark the cells between 300 and 700 K also 0 cells are marked. So in fact when I enter 280 K to 300 K, all cells are marked, but I don't really get why I get the error then. I assume there are some bad cells (although inverse orthogonal quality is below 0.88) that cause the problem, so I just wanted to see where in order to adjust the mesh accordingly.

Best regards,

Johannes

[Johannes]

Does anybody know what the reason could be?

[DrAmine]

Maybee the values reported corresponds to extrapolated face values or the virtual cells due to shell. Not sure though. Anyway having such a low value is nonphysical. Better to focus on having better mesh quality metrics. Increase mesh check verbosity to 2 and re-do a quality check. Very important are skenwss, orthogonal quality, aspect ration and cell volume change.

[Johannes]

Yes, I just wanted to see where the problem areas are so I can edit the mesh there. Skewness, orthogonal qualiy and aspect ratio is reported quite OK in my case that's why I wondered...

[DrAmine]

If they are as you say than check the boundary you are using.

[Johannes]

I just found out that the temperature warnings occur in (deleted) shadow zones of walls that were sewn together. That is probably the reason why the cells are not displayed.