Reaction forces in axisymmetric model

General Questions
1

Good morning,
I am having some troubles obtaining correct reaction forces from axisymmetric model.
After some “strange” results from more complex models, I resolved myself to create a very simple model to compare results with analytical calculations.

image

I applied an imposed displacement (v = 0.01mm) on the cylinder top face, in the vertical direction. Behaviour is elastic and non linear geometries aren’t considered. The bottom of the cylinder is constrained in vertical direction only.
From my understanding, reaction forces determined using 3D model (FEA 3D, in the attached image) through History Output are correct (the small difference analytical/numerical doesn’t bother me at this stage). It is evident that in the axisymmetric case, RF2 forces (total) from History Output are quite different from other cases. Am I missing something?

Thank you for your help!

2

CalculiX expands axisymmetric elements to one layer of 3D elements in a 2° sector. So you have to multiply the reaction force by 180.

4 Likes
3

Thank you so much! this is very important for me, since comparing force obtained from 3D model and 2D axisymmetric I obtained a ratio of almost 180… I suspected something similar but I couldn’t explain by myself.

Thanks again!

4

I think this peculiar feature shoul appear somewhere in the PPM interface, so that the user won’t have doubts.

One possibility would be to write

F.. for a 2 deg angle [N]

in the Edit window of the History data.

image
image286×593 8.45 KB

Best regards

5

Actually I guess the best “solution”, as all the other FEM software do (as far as i know) would be to multiply the force for the 2 degree angle by 180 and give the user directly the 360 degree force in the results.

6

The problem is that it would be inconsistent with the convention CalculiX uses and thus confusing for someone who checks the ccx manual. So PrePoMax should at least highlight this difference in the output window.

1 Like