i have maybe a possible bug in the load

[Jee-Bee]

I created a plate created of shells with in the center an (point)load
after creating the inp i checked the inp file and there was no load at all
see attachments

[bernd]

your analysis type is frequency. For frequency analysis type neither force nor pressure nor selfweight constraint are written to input file. FEM and CalcluiX experts around. Should load constraints be written on analysis type frequency?

frequency:
https://github.com/berndhahnebach/FreeC ... cx.py#L129

[Jee-Bee]

beamf2 one of the standard calculix examples have a force applied.

[fandaL]

Seems that CalculiX enables frequency analysis of loaded structure in different way: first compute step with loading and then compute frequencies in the next step where parameter PERTURBATION means taking loaded state from the previous step.
Details from the manual but rather look at example code
http://feacluster.com/CalculiX/ccx_2.11 ... de245.html

... If the PERTURBATION parameter is used in the *STEP card, the load active in the last *STATIC step, if any, will be taken as preload. Otherwise, no preload will be active.

and http://feacluster.com/CalculiX/ccx_2.11 ... de147.html

If the perturbation parameter is not activated on the *STEP card, the frequency analysis is performed on the unloaded structure, constrained by the homogeneous SPC's and MPC's. Any steps preceding the frequency step do not have any influence on the results.
If the perturbation parameter is activated, the stiffness matrix is augmented by contributions resulting from the displacements and stresses at the end of the last non-perturbative static step, if any, and the material parameters are based on the temperature at the end of that step. Thus, the effect of the centrifugal force on the frequencies in a turbine blade can be analyzed by first performing a static calculation with these loads, and selecting the perturbation parameter on the *STEP card in the subsequent frequency step. The loading at the end of a perturbation step is reset to zero.

In the beamf2.inp there are 2 steps

Code: Select all

...
*STEP
*STATIC
*CLOAD
LAST,3,-48.155
*END STEP
*STEP,PERTURBATION
*FREQUENCY
10
*END STEP

btw a formal detail
Is it needed to print into .inp file line *STATIC before *FREQUENCY, as FreeCAD currentli does? I thought these are 2 different analysis types with own keyword *STATIC or *FREQUENCY

[Jee-Bee]

btw a formal detail
Is it needed to print into .inp file line *STATIC before *FREQUENCY, as FreeCAD currentli does? I thought these are 2 different analysis types with own keyword *STATIC or *FREQUENCY

Is not needed. in the frequency examples they use just *FREQUENCY. I think only on case like described above it is needed.
however i have seen some example dat files with just the eigen frequencies. so it could be that static is needed just for displacement and stresses etc

[bernd]

for reference:

http://feacluster.com/CalculiX/ccx_2.11 ... de289.html

http://feacluster.com/CalculiX/ccx_2.11 ... de245.html
http://feacluster.com/CalculiX/ccx_2.11 ... de287.html
http://feacluster.com/CalculiX/ccx_2.11 ... de215.html

It will take some changes in the input file writer to change this. But we will need them anyway if we would like to support multiple steps at one point in the future. I will come up with some changes ...

[bernd]

I reimplemented the analysis type handling. It is all in one def now. This makes it easier once we would like support multiple steps. Any testing is highly appreciated.

branch:
https://github.com/berndhahnebach/FreeC ... femccxstep

commit:
https://github.com/berndhahnebach/FreeC ... dfc9b681c6

cheers bernd

[Jee-Bee]

Hi was by accident looking true CCX writer.
I saw by accident that (still?) no constraints are applied. see https://github.com/FreeCAD/FreeCAD/blob ... er.py#L137
I would asume this is (still) a bug that never made it to master...

[thschrader]

Should load constraints be written on analysis type frequency?

In my opinion yes. For example, the first eigenfrequency of a column with
external pressure loading is a little bit lower than without loading.

[bernd]

Thomas, have you tried on latest FreeCAD. My post is more than a year old!