Problem of convergence

[moliPHD]

Hello friends

I try to do a nonlinear analysis "using ansys APDL" of a 3-point bending of a beam made in composite materials (face + heart + face).
When:
The faces and heart are made in isotropic material, I easily obtained the solution without errors.
But if:
The faces are made in isotropic material, and the core is made in orthotropic material, I have the problem of convergence, even if I increase the number of substeps.
I didn't understand why, what's the solution in this case.

best regards

 

[rickfischer51]

Why is the model not converging? What errors, warnings or other messages are printed in the output screen or the .err file?

Rick Fischer
Principal Engineer
Argonne National Laboratory

 

[moliPHD]

Ok Mr.Rick Fischer,

When I apply a force that has a value equal to 100N, I choose the following values ​​in the "Solution controls Dialogue Box":
Analysis options: Large Displacement Static
Time at end of loadstep: 1
Automatic time stepping: On
Number of substeps: 10
No. Max. substeps of: 1000
Min no. substeps of: 1

The error mentioned by Ansys is as follows:
  *** ERROR *** CP = 645953 TIME = 4:26:32
  Not converged solution at time 0.435375 (load step 1 substep 6).
   Run terminated.

And 2 warnings:
*** WARNING *** CP = 10812 TIME = 4:12:37
  The / command is only valid PMACRO inside a macro or an input file.
   Command ignored.
*** WARNING *** CP = 646156 TIME = 4:26:41
  The unconverged solution (identified as time 1 substep 999999) is
  output analysis for debug purposes. Results shoulds not be used for
  Any other purpose.

best regards

 

[rickfischer51]

Are your materials non-linear? Do you have contact in your model?

What is not converged? Force, moments, deflections?

You have a converged solution at loadstep 1, substep 5. Load that and look at it (set,1,5). Look for distorted elements, large deflection, high stress, etc. Then load the last unconverged loadstep and look at that. Compare the two and look for obvious differences in stress deflection, etc.


Rick Fischer
Principal Engineer
Argonne National Laboratory

 

[moliPHD]

I use SOLID45 as element type, is linear.
I think that the deflections are those who don't converge.

Please Mr. Rick Fischer I need more details, it will be very clear if you show me what can I do step by step, because I'm a new user of Ansys APDL.
I don't understand how can I load loadstep 1 and the last unconverged loadstep, can you give me any example explain this clearly?

thank you so much for your time.
best regards.