External force versus Reaction Force Output

Feature Requests
1

Hi,

CCX only outputs external force, which is reaction force and applied force together.

I have always found that ccx output kind of weird. Taking into consideration that looking at the reactions is part of any study’s basic check , could you consider postprocessing that value internally subtracting the imposed forces to have real reactions in the solution window.?

Regards

1 Like
2

Then there could be even 3 force outputs like in Abaqus:

  • TF - total forces (reaction forces + point loads) - the current RF output
  • RF - reaction forces only
  • CF - point loads and concentrated moments

In Abaqus, this distinction is particularly helpful for pretension section nodes.

1 Like
3

Hm, this is a really important distinction. But for some nodal forces, it is easy to compute them (concentrated force, surface traction), but other loads like *Dload are computed as nodal forces in CalculiX. I will have to see if it is possible to obtain them in some other way. Any ideas?

4

In Abaqus, only *CLOAD is used to calculate CF and TF. *DSLOAD/*DLOAD is not taken into account for those outputs. CF is even automatically removed from output requests (with a warning) when no *CLOAD is present.

1 Like
5

I tried using Gravity load, and the result is that it creates nodal forces, which are taken into account when outputting the history result RF. So the history output contains nodal forces and not only reaction forces. To get the reaction forces I would need the entire force vcector to subtract it from the RF field.

6

The best method I see for the moment is to compute the *Dload nodal forces by myself. I think I can do that for constant loads, such as gravity, but for spatially varying loads (E.g., Centrifugal), it could become tricky and inaccurate. And this will only work for linear analysis. So the effort of doing it is questionable.

The other method is to save all the resulting displacements and use them as a load in a separate analysis. Since no nodal forces are applied, the correct reactions should be the result. This is computationally expensive, but maybe only the elements at the supports could be used. Could be automatic, but would require some effort to integrate into PrePoMax.

7

Thanks Matej,

I see. Maybe I could request in the github webpage to be done directly in ccx.?

Regards

8

That would be the only correct way of doing it.

1 Like