Learning Forum

[Dharani]

Hi sir,

Wheel is made up of three materials . Core is composite and the core is covered with metal and the outermost part is Hyperelastic material (Elastomer). While solving, I'm facing convergence problems. Please suggest some solution.

I'm attaching my image for your perusal.

 

Thanks

[peteroznewman]

There are many, many posts on convergence failures on this site. Use the Search field  on this page to find some of them. Use words from the specific Error such as distorted.

Use Google with a search term like this: site:studentcommunity.ansys.com highly distorted

Also search for hyperelastic as there are several methods of helping convergence of hyperelastic materials.

[Dharani]

I go through all the convergence issues and hyperelastic materials. I went upto 0.79 sec out of 1 sec. Still I'm unable to solve this convergence issues. I tried APDL command whatever given in this site. For this, I'm getting error that "For element type 33 (Solid 187), keyopt 2 =1 is invalid" and "Solid 187 not associated with fully incompressible hyperelastic materials, no suggestion is available and no resulting is fixed" warning messages are coming. And I'm havingone more doubt that why this many Element types are coming? 

[peteroznewman]

Don't worry about how many element types the solver creates.

Overcoming convergence problems requires many iterations, so you want a model that runs as fast as possible. The image in your first post has a large volume of material to support the wheel. It would waste a lot of time solving elements in that volume. If 99% of the deformation is in the elastomer, you could discard all parts except the wheel, tire and block. You could also make the wheel and block rigid parts so the only flexible part is the elastomer. Is that bonded to the wheel?  You could even cut the tire down to a +/- 30 degree slice on each side of the contact point.  Finally, if the block and wheel is symmetric across the width of the wheel, you could cut the model in half again.

Can you make your model to run faster by making it much smaller, before you spend more time getting the last part converged?

[Dharani]

Hi sir,

Thank u so much for your suggestions. I make it as symmetric, applied reduced integration and yeoh type. For that it got converged and solved.

Now I have to do another model also. For that, same convergence problem is happening. I make it as symmetry, still not converging. any other suggestions?

 

Thanks

Dharani

[peteroznewman]

Glad to hear you have had some success.

The error here is Highly Distorted Element.

Here is a Google Search that will provide links to many discussions that you may find helpful.

Change the SOLID186 element formulation using keyopt(2) and keyopt(6) to make the elements work better with incompressible materials is one suggestion from one of those links.  I see from one of your images above that the solver is automatically setting some of these for you.