Learning Forum

[ansysuser]

Hello,

I am trying to use the dampening parameter beta in a simulation I have built.  The simulation runs just fine when there is no damping parameter beta in the material properties, but as soon as I add beta to get rid of some oscillations I am seeing, the force stops converging. My material is hyperelastic and nearly incompressible.  I put this command under Solid_Body in the Geometry:

ET, matid, 226
keyopt, matid, 11, 1

Because the documentation says, "The current-technology elements with mixed u-P formulation and SOLID285 are available for nearly incompressible hyperelastic materials."  I find 226 is a current technology element and without the keyopt it will not converge any better than the solid186 element I tried before. So I wanted to try the mixed u_P formulation as the documentation says.  However, I get this error:

 *** ERROR ***                           CP =       2.625   TIME= 16:56:53
 For element type = 1 (SOLID226), KEYOPT(11) = 1 is invalid.

 

Why does the documentation say I can use keyopt 11 under solid226 and yet transient structural says it is invalid?  Is there anything else I can do to help "force convergence" when I add dampening to the material property?  I have tried a finer mesh, smaller time steps and so I am at a loss. I have tried dampening parameter values of 0.1, 0.01 and they both fail to converge.

 

Thanks

[ansysuser]

Hello,

I am still trying to solve this.  Any input is appreciated.  I am wanting to dampen the oscillations in my solution, but I have not been able to get convergence by any means.  The solution without damping has oscillations in the compression and release stages, as shown in the image of displacement below.  Without damping, I get a solution just fine.

So I added beta damping to the material properties, and the damping follows the same displacement curves but without the oscillations, so that is good. But the solution only goes to t=0.375 at best with damping. Usually the force does not converge. In the below image you can see that the solution gets partway there, and there are no oscillations.  But it never makes it further than this.

Things I have tried:

1.  Using solid226 with u-P formulation.  Here I get the "VOLUMETRIC COMPATIBILITY IS NOT SATISFIED " problem, even for tol set as high as 0.8 through the cnvtol command.

2.  Using solid186 with and without keyopts to get u-P.  This did not help.

3.  Increasing the number of equilibrium iterations to 100. No effect.

4.  Increasing the force convergence tolerance to 2% from the default 0.5%.  This resulted in "severely distorted elements" error.

5. Refining the mesh.  This didn't seem to have much impact either.

There are several things that don't make sense to me about this.  I looked in the documentation at this page:

https://ansyshelp.ansys.com/account/secured?returnurl=/Views/Secured/corp/v194/ans_elem/Hlp_E_SOLID226.html

and based on that page, I assigned keypts this way:

ET, 1, 226
keyopt, 1,1,1
keyopt, 1, 6, 1  
keyopt, 1, 11, 1

But in the solution output I see this:

 *** WARNING ***                         CP =       2.453   TIME= 18:45:08
 For element type = 1 ( SOLID226 ), KEYOPT 1 = 1 is undocumented.       

 *** WARNING ***                         CP =       2.453   TIME= 18:45:08
 For element type = 1 ( SOLID226 ), KEYOPT 1 = 1 is undocumented.       

 *** WARNING ***                         CP =       2.453   TIME= 18:45:08
 For element type = 1 (SOLID226), KEYOPT(6) = 1 is an undocumented      
 option.                                                                

 *** WARNING ***                         CP =       2.578   TIME= 18:45:08
 For element type = 1 ( SOLID226 ), KEYOPT 1 = 1 is undocumented.       

 *** WARNING ***                         CP =       2.578   TIME= 18:45:08
 For element type = 1 (SOLID226), KEYOPT(6) = 1 is an undocumented      
 option.

Q1:  Why do I see those keyopts are undocumented when they are right there in the documentation??  Those keyopts are definitely not undocumented, as you can see from the link! So why the warnings??

Q2:  Are there any suggestions as to how to get convergence with damping?  Any suggestion is helpful.

[ansysuser]

Hello,

I have linked to a file on Drive with the workbench project in it.  There are three transient structural blocks in there, labeled for ease of use.  Please look over my settings in the two with damping and let me know what I can do to get convergence.

https://drive.google.com/file/d/1LjsgLoaVKmtMZuiGcMGTdnMUF17u79cN/view?usp=sharing 

Thanks