Ansys Workbench Contact-problem

[JoelHenrik]

I have a problem that requires me to do a simple simulation of a sheet metal stamping. I have access to Ansys Workbench v12 Academic version, and have familiarized myself with the program and tried to read up on its features and limitations.

The simulation is selected as a static structural.

The boundary conditions are set as:
* The lower die-tool underside is set with a Fixed Support
* The upper die-tool vertical sides are set with Frictionless support, to allow a movement in the vertical direction.
* The sheet metal plate´s long sides are also set with Frictionless support, to prevent a possible Rigid-body motion, but still allow a vertical press movement.
* The upper tool is applied with a displacement towards the second tool.
* There is a initial contact between the platesurfaces and the tool's pressure surfaces.
* There a two sets of contacts. First, the upper tool´s whole stamping-area and the surface of the plate that it intersects with. Second, the lower tool´s surface and the bottom surface of the plate.

My problem occurs when I am applying the contact conditions for the contact area. Ansys choice is initially a bonded contact, and simulations can be done. However, this does not simulate the reality as we do not allow room to stretch and slide the plate. I would therefore choose a Frictional contact and puts friction to about 0.3, just to test out. Then the simulation would not converge, and the upper tool glides right through the plate without being aware of when a contact occurs.
Could it be that Workbench is not allocated for this type of simulation, or do I choose the wrong settings?
Would any kind person be able to explain if this is not possible, and if so, what restrictions of the program that causes this?

Best regards, J-H

 

[Drej]

You'll need to initialise the contact in the model. You can try either telling ANSYS that you have initial contact between the parts and solve (check the gap between the parts and set the initial contact tolerance based on this value), or you can create a multi-step analysis. The first step in the multi-step case will be a small displacement applied to the part to establish initial contact, and then a second step to apply the full load on the model.


------------
See faq569-1083 for details on how to make best use of Eng-Tips.com

 

[JohnyGluebag]

Ansys is capable of this type of problem. However there are a few issues that the user has to verify/check to get convergence.
- contact stiffness
- mesh size
- time step/load increments.

Ask yourself, is this problem non-linear? Will the material encounter plastic deformation? If so you will have to input a material curve and use a non-linear material model.

If you are not using the correct physics for the problem your results will be meaningless.

Can you take advantage of symmetry? At first glance it looks like you could have two planes of symmetry and thus reduce the size of your problem.

What are trying to get out of this analysis?

____________
JohnyGluebag

 

[seymours2571]

You can also try manually setting the pinball radius in the contact settings within Ansys WB. I have had instances in the past where the "Automatic" setting did not work and required me to manually enter the pinball radius.

Good luck,

Steve


Stephen Seymour, PE
Seymour Engineering & Consulting Group

http://www.seymourecg.com