Geometric nonlinear static analysis using OptiStruct

Altair Forum User

Hello All, 

I am doing geometry nonlinear static analysis on a lattice structure using Optistruct and during model organization, I am facing lot of problems that I am not able to solve, like - 

1. I have a lattice CAD model in form of a 2D mesh ( 441964 nodes, 892864 elements), then I am doing initially this 2D mesh convert into the 3D tetrahedral mesh (608588 nodes, 2171299 elements) using the 3D panel in HM.
2. Before doing this conversion, I checked the quality of 2D elements and then convert into the 3D tetra mesh. But during the 2D elements quality check, there is some large number of elements are failed in skew, trias min and max angle  e.g. (Skew (60) 163 of 892864 (0%) failed. It is difficult to correct the failed elements because the number of the failed elements are very large, it will take a lot of time.  
3. Next, after conversion 2D trias to 3D tetra elements, I deleted 2D elements then provided input details like material, properties, load collector and load step.

 

My question is that

1. when I do the static linear analysis so analysis was completed, but when I do geometry nonlinear static analysis some error showing and analysis has not completed. 

2. I provide all input data that required for geometry nonlinear static analysis. I am thinking that due to a large number of 3D tetra elements fails, therefore the nonlinear analysis not complete. 

   Please guide me, how can I do fastly to correct those fails elements.  

 

I am trying but I am not able to do it fastly.  

 

Altair Forum User

Hi Prashant,

 

Can you let us know what are the error message you see. 

 

Please share more information about errors and warning or share the .out file. 

Altair Forum User

Hi Prakash,

 

Thanks for replying, the error message that I have seen is:

 A fatal error has occurred during computations:

 

*** ERROR # 4079 ***
  Non-convergent nonlinear iterations for subcase 1.

  ANALYSIS COMPLETED.

  *** SOME ERRORS WERE REPORTED DURING THE RUN,
  *** PLEASE VERIFY RESULTS AS THEY MAY BE INVALID.

 

Can you please elaborate the meaning of fatal error. 

I am performing geometric non-linear analysis on a lattice structure using Optistruct.

1. This is a new one problem, In which I have a lattice CAD model in form of a 2D mesh ( 107140 nodes, 223216 elements), then I am doing initially this 2D mesh convert into the 3D tetrahedral mesh (124130 nodes, 3481714 elements) through the 3D panel in HM.

2. Before conversion 2D to 3D elements, firstly, I checked the quality of 2D elements. During 2D elements check, seen a lot of elements have failed e.g. 

   Aspect > 5, (1920 elements failed)
   Skew> 60, (4092 elements failed)
   Length<0.1 (7991)
   Trias
   Min angle < 15 (2536)
   Max angle > 135 (1837),I have a CAD lattice model, which I modeled in Rhino software, and save it into stl. file, before saving into stl. file, firstly I have ensured that my 2D mesh of CAD lattice model is fully valid, and there are some details of the model are:

Mesh has 0 naked edges. 

Mesh is manifold. 
Mesh is solid. 

 

Next, I Imported that stl. into HyperMESH software, and convert it into 2D trias mesh into 3D tetra, and deleting that 2D elements and then we again checked 3D tetra elements quality. Similarly, seen a lot of element were failed during 3D elements quality checked, like

Aspect > 5, (1925 elements failed)
Skew> 60, (4242 elements failed)

Tet collapse < 0.3 (2481 elements failed)
Length<0.1 (31250)

Trias
Min angle < 15 (2281)
Max angle > 135 (1975)

 

Again my question is same,

 

1. when I do the static linear analysis so analysis was completed, but when I do geometry nonlinear static analysis the above error showing and analysis has not completed. 

2. I provide all input data that required for geometry nonlinear static analysis. I am thinking that due to a large number of 3D tetra elements fails, therefore the nonlinear analysis not complete. 

   Please guide me, how can I do fastly to correct those fails elements.  

I have shared all three .out files, please find in attachments.

 

Please help me out. 

 

 

 

Unable to find an attachment - read this blog

Altair Forum User

Hi @Pr@shant

 

I see there are no modelling issues. 

 

Is it possible to share the model file?

Please use the file transfer link in my signature to share the model file

Altair Forum User

Hello Prakash,

 

Thanks for quick reply,

 

I am sharing my model .fem file. So please having you look once. 

 

If you want .hm file, so please let me know. 

 

Thank you, It means a lot to me.

 

 

 

 

Altair Forum User

Thank you,

 

I will check and update you soon. 

Altair Forum User

Thanks, Prakash

 

I will be waiting for your reply. 

Altair Forum User

Hi Prakash,

If you have checked my model file, so please do update me. I am still waiting.  

Or 

Give me an alternative way, that I could perform the geometric nonlinear static analysis of the same lattice model structure.

Actually, We have to perform 15-20 number of nonlinear simulation, for those 15-20 simulations, we have changed lattice structure model variables like porosity, pore size etc. 

Ahead, I am waiting for your reply. 

Thanks

Altair Forum User

Hi Prashant,

 

During analysis I see few elements are penetrating and this causing divergence.

 

I would suggest to try below:

 

1) Use RADIOSS to model self contact and run the analysis

or

2) Use NLSTAT with Large displacement (LGDISP) to run the same model. In this case you need to convert the existing Johnson Cook's material model to MATS1 format