Learning Forum

[rdk9000]

Hello I would like to add a force to the bottom right corner of the model in the y direction. I would like the force to stop once the horizontal displacement of the top of the cylinder reaches a threshold in the -x direction.

[peteroznewman]

There may be other ways to achieve what you want. Here is how I would achieve your goal.

Analysis Settings

   Auto Time Stepping, On

      Initial Substeps 10

      Minimum Substeps 10

   Large Deflection, On

In the Solution Branch, insert a directional deformation plot using the horizontal coordinate direction and scope it to the top of the cylinder.

After you solve, you will have tabular data for Time, Applied Force and Horizontal Deformation.  Pick the value of force when the horizontal deformation reaches the threshold.

This assumes that the applied force is large enough.  If the horizontal deformation didn't reach the threshold, make it larger. If you want more resolution in the tabular output, use more substeps.

 

[rdk9000]

Great I will try this!

[rdk9000]

I gave it a shot but hitting the deflection at the correct threshold proves to be very time consuming. I have samples of various geometry I have to try this with. I would ideally like to design this in APDL entirely.  Is there a way to incorporate APDL commands to say when the x displacement reaches 50 microns deactivate force?

[peteroznewman]

In Workbench, you can add a Direct Optimization component to the Static Structural analysis and let the optimizer adjust the force until the displacement is exactly 50 microns.

I hope an expert in APDL can answer the question for how to use APDL code to do the same.

How linear is the force-deflection data?  If it is basically linear, then just apply the same force to all designs and use the deformation of the top as the response. Even better, divide the force by the deformation to put the response in terms of stiffness K in N/micron.

[rdk9000]

Hello Sir,

So I changed the model a bit to more accurately reflect my device. The moment on the right side will eventually cause the two pillars to touch. This is what I'm looking for. However I have not had much luck with enforcing contact (thus why I'm trying to use displacement). I have read a bit about parameterizing but cannot find a solid source. Ideally I would like to optimize the gap between the pillars for a given strain. But for now I would just like to increase the moment until contact is made (or the difference of displacement in the X is 0). I'm not exactly sure how to add a direct optimization component as you mentioned above. Thank you again for your time. 

 

[peteroznewman]

Please attach your Workbench Project Archive .wbpz file after you Post your reply and I will try adding the Direct Optimization.

[rdk9000]

Hello Sir,

After looking around I found a great tutorial about direct optimization (

). It worked like a charm. Thank you very much for your help!