Problems when using different element types (no error given)

[bjorn_fallqvist]

Lately, I have a large model of solid elements that I also wish to use beam elements and shell elements in. However, the analysis (AMG) starts to iterate, but then the analysis quits, without any error messages at all. After the first iteration line, the log file ends.

If I suppress the shell and beam elements, the analysis can be run, but if not, it just crashes without any error message.

What's the problem here, do you think?

Thanks,

Björn

[S.LI]

May I know

1.) the simulation is LSS or MES?

3.) Do the element data for shells and beams set correctly, such as thickness?

2.) How about other equations solvers, such as BCS? (I know it's really painful to simulate a huge model with BCS solver.)

3.) Could you share more information here, such as log or summary files?

Thanks.

[bjorn_fallqvist]

Hi,

1) It's an MES

2) Yes, they are all defined correctly

3) I tried using the BCS solver also, but the same error occured

4) I will share the log and summary files the next time I attempt to run it. I might be able to share the model file as an archive file also, depends on how big it is.

Thanks,

I will get back to you

/Björn

[bjorn_fallqvist]

Here is the entire log file:

 
 Autodesk Algor Accupak/VE Mechanical Event Simulation 
 Version 2012.00.08.0004-W64/X64 24-Nov-2010
 Copyright (c) 2010, Autodesk, Inc. All rights reserved.
 
 ** Total available memory = 14065276 (KB).
 ** Options executed are:
    NOMIN   - NO optimization for matrix storage
    RUN     - Execute processor
    STRAIN  - Output strain data for Superview
    DUPLEX  - Duplex communication with tWAIT=5 seconds
    PRRO    - Output reaction force data, n=2


    NUMNP    DOF NEGL NEGNL MODEX      NSTE        DT    TSTART IPRI IPSF 
   143937 000000    0    64     1        11 0.100E+00 0.000E+00    1    3

 ** Perform analysis
 ** Open output files ...
 ** Open temporary files ...
 ** Reading input file ...

                   PROGRAM VERSION: 201200080004
           alg-win-x64.dll VERSION: 201200080004
     algconfig-win-x64.dll VERSION: 201200080004
      agsdb_ar-win-x64.dll VERSION: 201200080004
              amgsolve.exe VERSION:   0360000000
 
    Model=C:\Users\Public\Documents\Projekt\Teroc AB\1089\CAD\Square bolt holes, split concrete with plates, split surfaces, square top pipe\Tower_assy.ds_data\3\ds

 **** Model Unit System Settings:
      --------------------------------------------
       Unit System              : Custom
       Force                    : N
       Length                   : mm
       Time                     : s
       Temperature (Absolute)   : deg C (K)
       Thermal Energy           : J
       Voltage                  : V
       Current                  : A
       Electrical Resistance    : ohm
       Mass                     : N*s^2/mm
      --------------------------------------------

    No. of nodes             = 143937     No. of equations = 480016
    Maximum no. of iteration =     15     No. of steps     =     11
    Convergence tolerance    =  1.0000E-02
 ** Input elements...
 ** 3-D continuum element calculation, number of element =      1320
 ** 3-D continuum element calculation, number of element =      4339
 ** 3-D continuum element calculation, number of element =      7903
 ** 3-D continuum element calculation, number of element =      6095
 ** 3-D continuum element calculation, number of element =      2781
 ** 3-D continuum element calculation, number of element =      2912
 ** 3-D continuum element calculation, number of element =      2507
 ** 3-D continuum element calculation, number of element =     12245
 ** 3-D continuum element calculation, number of element =     11108
 ** 3-D continuum element calculation, number of element =     11436
 ** 3-D continuum element calculation, number of element =      3112
 ** 3-D continuum element calculation, number of element =      7974
 ** 3-D continuum element calculation, number of element =      4156
 ** 3-D continuum element calculation, number of element =     11255
 ** 3-D continuum element calculation, number of element =     11035
 ** 3-D continuum element calculation, number of element =     11154
 ** 3-D continuum element calculation, number of element =      3715
 ** 3-D continuum element calculation, number of element =      3514
 ** 3-D continuum element calculation, number of element =      2947
 ** 3-D continuum element calculation, number of element =      3143
 ** 3-D continuum element calculation, number of element =      6059
 ** 3-D continuum element calculation, number of element =      4793
 ** 3-D continuum element calculation, number of element =     14009
 ** 3-D continuum element calculation, number of element =     13208
 ** 3-D continuum element calculation, number of element =     13203
 ** 3-D continuum element calculation, number of element =      4056
 ** 3-D continuum element calculation, number of element =      4128
 ** 3-D continuum element calculation, number of element =      3890
 ** 3-D continuum element calculation, number of element =      3372
 ** 3-D continuum element calculation, number of element =      4607
 ** 3-D continuum element calculation, number of element =      4904
 ** 3-D continuum element calculation, number of element =     16114
 ** 3-D continuum element calculation, number of element =     16759
 ** 3-D continuum element calculation, number of element =     13905
 ** 3-D continuum element calculation, number of element =      3897
 ** 3-D continuum element calculation, number of element =      4006
 ** 3-D continuum element calculation, number of element =      3635
 ** 3-D continuum element calculation, number of element =      3032
 ** 3-D continuum element calculation, number of element =      3225
 ** 3-D continuum element calculation, number of element =      5636
 ** 3-D continuum element calculation, number of element =     14244
 ** 3-D continuum element calculation, number of element =     11158
 ** 3-D continuum element calculation, number of element =     14194
 ** 3-D continuum element calculation, number of element =      3691
 ** 3-D continuum element calculation, number of element =      3674
 ** 3-D continuum element calculation, number of element =      3591
 ** 3-D continuum element calculation, number of element =      3827
 ** 3-D continuum element calculation, number of element =      2728
 ** 3-D continuum element calculation, number of element =      4970
 ** 3-D continuum element calculation, number of element =     11028
 ** 3-D continuum element calculation, number of element =     14812
 ** 3-D continuum element calculation, number of element =     15087
 ** 3-D continuum element calculation, number of element =      3638
 ** 3-D continuum element calculation, number of element =      3855
 ** 3-D continuum element calculation, number of element =      3618
 ** 3-D beam element calculation, number of element =        35
 ** 3-D beam element calculation, number of element =        35
 ** 3-D beam element calculation, number of element =        55
 ** 3-D beam element calculation, number of element =        54
 ** 3-D beam element calculation, number of element =        60
 ** 3-D beam element calculation, number of element =         4
 ** 3-D beam element calculation, number of element =       261
 ** 3-D beam element calculation, number of element =        18
 ** 3-D shell element calculation, number of element =     22077
 ** Nonzero entries initially in upper triangle matrix = 13473282
 ** Parallel AMG Iterative Solver Solution

Time        Percent     DT       L Iter. Residual
   0.000000   0.00     0.100000  1   0   0.0000E+00

[S.LI]

Thanks.

Is "Version 2012.00.08.0004-W64/X64 24-Nov-2010" the beta release?

How about Algor 2011?

If possible, could you share your model?

Sorry for the inconvenience.

[bjorn_fallqvist]

Yes, I forgot to mention that. It is the Beta version.

I tried to make the model as small as possible (re-meshed the surface mesh so it's not solid meshed and made an archive of it), but it's still 95mb. Do you have an FTP where I can upload it somewhere?

I'll try with Algor 2011 instead tomorrow.

[S.LI]

Sorry for I don't have a ftp site.

Please contact your sale person or client support, and forward it to me shanhu dot li at autodesk dot com

Thanks.

[bjorn_fallqvist]

Argh, I got the dreaded Error reading displacements message in 2011. I wonder why this didn't show in 2012. I guess I'll have to try the BCS again, then.

[bjorn_fallqvist]

With BCS I got

**** Error detected in the BCSLIB-EXT solver. **** Code: -503 Exactly zero pivot encountered during numeric factorization. Possible reason: The model may not be tied down enough or you may have a change in stiffness somewhere in the model which is too abrupt. Suggestion: check model and boundary conditions.

[S.LI]

According to the error message, it seems the model is not fully constrained.

This is only happened when the analysis type is "static stress with nonlinear ...".

What is yours, "MES with ..." or "static stress with ..."?

Another thing you might need to check is the connections between solid parts and shell/beam elements.

For a regular connection, the rotational DOFs are not constrained on shell/beam sides. This will make the "non-constrained" trouble.

[bjorn_fallqvist]

Yeah, it works in LSS, but I'm running an MES.

There shouldn't be any problems with the rotational DoFs; I have made sure that each beam element is connected to several nodes on the solids, so rotation shouldn't be a problem. Even if it was, this is a MES, it should be able to handle it...

[S.LI]

Given that to suppress shell/beams make model work, could you try to suppress them part by part to see which part makes the trouble?

Thanks.

[bjorn_fallqvist]

I'll do that and get back to you.

[bjorn_fallqvist]

It must've been something wrong with the beam elements. Now I have re-done the entire model from scratch and it works. I have no idea what was wrong with the first model.

[S.LI]

Great. At least you have results now.

[bjorn_fallqvist]

Yeah. It's very strange, I have no idea what was wrong with the old model.

Anyway, thanks for your time.