Initial velocity For shell elements

Analysis issues
1

I’m running a very simple case: a thin plate falling with an initial velocity.

I’ve been able to simulate this process correctly using solid elements.

However, when I tried using the shell element, it diverged after running for a few time steps:

Average force = 264599194316220896.000000
Time avg. forc = 3834770932119144.000000
Largest residual force = 98587194676042121216.000000 in node 239 and dof 1
Largest increment of disp = 6.918282e+02
Largest correction to disp = 2.305770e+02 in node 3416 and dof 3

*ERROR: solution seems to diverge; please try

automatic incrementation; program stops

Best solution and residuals are in the frd file

I also found some similar problems:*INITIALCONDITIONS, TYPE=VELOCITY for shell elements

But I still haven’t solved them.

My inp is as follows:

*MATERIAL, Name=EL
*ELASTIC
210000000000, 0.3
*DENSITY
7850
*SHELL SECTION, Elset=Eall, Material=EL, OFFSET=0.0
0.001

*STEP, NLGEOM, INC=1000000
*DYNAMIC, ALPHA=0.0, DIRECT
1E-5, 1.0

*BOUNDARY, AMPLITUDE=RAMP_Y
Nlrmoving, 2, 2, 1

*CLOAD
Nsurface, 1, 0.0
Nsurface, 2, 0.0
Nsurface, 3, 0.0

I hope you can all share your opinions. Thank you very much.

2

If it’s this issue: Initial velocity is not transferred to the expanded nodes in dynamic analyses · Issue #63 · Dhondtguido/CalculiX · GitHub then it has a workaround with defining a dummy surface. Otherwise, can you share the whole input deck (you could use some hosting website) ?

3

Thank you so much for your reply, it was very helpful. I’m currently trying out this dummy face, and I’ve also attached my complete files so you can help me even more.

4

This workaround can only ensure that the initial velocity is correctly applied to the expanded 2D element. But non-convergence is likely due to something else.

From my experience, the dynamic implicit step in CalculiX quite often fails to converge, especially when there’s no damping. However, firstly, I would switch from direct to automatic incrementation.

5

I’ve tried using damping and automatic incrementation, but it’s still diverging.

*STEP, NLGEOM, INC=1000000
*DYNAMIC, ALPHA=0.0, DIRECT
1E-5, 1.0, 1E-15, 1E-05

6

It’s worth noting that my structure was successfully assigned an initial velocity. However, it diverged after running for a while. This is different from some similar topics.

7

Sorry, I should reply to you. Can you please view my comment?