Strand 7: Error Solution has diverged. Rotation at node 7372 (1.9370E+08 deg) is too large. 2

[Totobolz]

Good Day.

To All Strand 7 FEA experts,

I am using Strand 7 to perform Nonlinear Analysis in a Plate Model where I encountered this error:

ERROR[ 13]:Solution has diverged. Rotation at node 7372 (1.9370E+08 deg) is too large.

and this warning.
*WARNING[ 73]:Internal angle is out of range in plate 15137.

*WARNING[ 73]:Internal angle is out of range in plate 14130.

Plate properties are properly defined.
The thing is the DOF/node restraint applied.
I used the fixed restraint with no error but when applying ZX symmetry as restraint. Error occurs.

I also attached the model as reference. Feel free to check.

Thanks in advance.

 

[FEA way]

Could you attach a picture showing your model with boundary conditions ? I think that it’s underconstrained and thus rigid body motions occur.

 

[ThomasH]

Based on what FEA_way mentioned,

run an analysis for Eigenmodes. If there are any with 0.0 Hz frequency you have rigid body modes.

Another thought, does nonlinear analysis i Strand automatically include large deformations?

Thomas

 

[Denial]

Totobolz.[ ] Do you get a similar sort of problem if you perform a simple linear analysis on your model?

ThomasH.[ ] I haven't used Strand7 for quite a while, but when I last used it the inclusion / non-inclusion of large deformations was optional.[ ] I think the default was non-inclusion.

 

[Totobolz]

@Denial, linear analysis worked fine.
Okay, I'll check the settings for the large deformations.

 

[IDS]

Only a few people here are Strand7 users, so it's always a good idea to post some images so others can comment.

You have your nodes restrained in the Y direction only, but gravity is in the Z direction.

I'm surprised the linear analysis completed, but if you look at the log file you will see the deflections are a little on the large side (about 1E10).

If you want symmetry about the XZ plane then the nodes on the plane of symmetry need to be fixed in the Y direction, but you also need to restrain global movement in the X and Z directions.


Doug Jenkins
Interactive Design Services

http://newtonexcelbach.wordpress.com/

 

[Totobolz]

@IDS, thanks for passing by. I read many of your replies.
I tried the "automatic sub-incrementation" of Strand 7 particularly the Displacement scaling and Displacement Control.
The solver proceeded but the iterations were painful. more than 50(about 15mins before I stopped it).
I agree with what you said about the ZX symmetry restrain.

 

[IDS]

Just a suggestion if a non-linear analysis is giving problems: set the first iteration load factor to a very low value (say 0.01). If that fails to converge there must be a problem with the model, other than non-linear behaviour effects. In this case the analysis failed at the first iteration, suggesting there was a problem with the restraints.

Doug Jenkins
Interactive Design Services

http://newtonexcelbach.wordpress.com/

 

[jhardy1]

Your model is not adequately restrained. I ran a Linear Static Solver (ALWAYS a good place to start!), and the solver log file gives the following displacements.

Results for 1 Load Case...

MAXIMUM DISPLACEMENT MAGNITUDES
Case DX DY DZ RX RY RZ Name
1 9.45004E+10 1.23440E-01 1.27770E+11 2.46870E-03 1.98140E+07 5.19257E-04 "Load Case 1"

You have not applied any X-displacement or Z-displacement restraint in your model. Fix this, then run a Linear Static solver (to validate overall model behaviour etc), and then and only then, start looking at non-linear static, buckling modes, natural frequencies, etc.

http://julianh72.blogspot.com