ANSYS Worbench - ARCLEN and pressure load

[xtremi]

In Ansys Workbench (19.0), in Mechanical, I have a Static Structural analysis. I've added a pressure load on a geometric surface, where I under Magnitude selected Constant and set a value. By doing this a frame in the UI shows "Tabular Data" with two points [tt]0,0[/tt] and [tt]1,<my value>[/tt]. This makes sense for a ramped load.

I added a ADPL Command on my analysis (Static Structural) with [tt]ARCLEN, on[/tt].

If I run my analysis, I get the error:
[tt] *** ERROR *** CP = 0.344 TIME= 10:15:11
There are tabular loads in the model. The Arc-length method does not
support tabular loads. Please replace tabular loads by other load
types and then run the analysis again.[/tt]

How can I use arc length control with a pressure load, if it forces me to have tabular data? What am I missing?

Thank you.

 

[Erik Panos Kostson]

It says that you can not use a tabular force (so the tabular load can not be used).

Typically when one runs arc-length say for buckling instability, one would apply a load higher than the instability-load all at once (not gradually), then the solver will do the required cut backs to solve. That is at least for the software which I use (Strand7).

 

[xtremi]

I understand, but Ansys Workbench doesn't allow me to create a non-tabular pressure load. Even if I select "Constant" it will set it to "ramped" and create a table with two values...

 

[xtremi]

Well if anyone else has this problem, I think I found the solution;
After setting the pressure load, instead of choosing [tt]Constant[/tt], choose [tt]Tabular[/tt], but under [tt]Tabular Data>Independent Variable[/tt] select [tt]Step[/tt].
Now the tabular data table will only contain one cell where you can add the "Constant" value for the step.

 

[Erik Panos Kostson]

That is funny. Not much on the internet about this, but some people report similar issue.

I would then try to use a command snippet to apply the pressure force as a constant value (use named selections of the face and apply pressure). I am not sure if it will like this since it will not see any loads in the GUI.

Otherwise if you do not expect any snap-through kind of problem, and just "collapse" followed by stiffening/softening, then neglecting arc-lengths should be OK (just use the normal solver with large deflection on and sub-steps on).