# Expansion within a defined area

• 31 Views
• Last Post 28 February 2020
SamHitchin96 posted this 25 February 2020

Below is an image of my geometry, I am trying to make the geometry in the middle thermally expand within the circular geometry, which produces a compression through those arms. However when i run the simulation the arms go through the circular geometry as shown in the second image. I have tried playing around with contacts etc and had no luck. I want that circular geometry to act as full solid so its impenetrable and the arms wont go through but instead get squashed. Anyone have any suggestions.

Aniket posted this 26 February 2020

have you defined contact between the parts that are not in touch initially and will be in touch in the future due to expansion? Also, use smaller timesteps.

-Aniket

Guidelines on the Student Community

How to access ANSYS help links

SamHitchin96 posted this 26 February 2020

What would be the best way to define those contacts?

Sam

Aniket posted this 26 February 2020

You can define them as frictional if you know the coefficient of the friction between them.

-Aniket

Guidelines on the Student Community

How to access ANSYS help links

SamHitchin96 posted this 26 February 2020

I don't have a friction coefficient as i am just using that ring to demonstrate the boundary i required.Is it worth making up a friction coefficient?

peteroznewman posted this 27 February 2020

What version of ANSYS are you using?

Edit the mesh element size to get at least two elements through the thickness of the arms.

I assume you picked all the faces on the arm to be in the contact definition.

Click on the Contact and show the Details window. There are some settings you can change.

Set Trim Contact to Off.

Change the Pinball Region to Radius and change the Pinball Radius to a number that is as large as any gap you expect to close during the simulation.

Some of the other changes may not be required, but the may help such as the Detection method and Update Stiffness.

SamHitchin96 posted this 28 February 2020

That seems to be working a lot better

SamHitchin96 posted this 28 February 2020

I am having an issue with it coming up with excessive memory ram when trying to run the simulation at a mesh of 0.1mm. The geometry itself is only 12.5mm in diameter so i was not expecting this issue. I am finding when i run the simulation at 0.5mm it shows the results fine with a warning of running over the cpu ram, however when I run it at 0.4 the centre piece seems to just disappear and when you press play on the simulation it just fires off. Any idea what the cause of this issue could be?

peteroznewman posted this 28 February 2020

When convergence fails, the solver will leave an unconverged deformation plot for diagnostic purposes only. Is that what you see?  If you put some screen snapshots up, it will be easier to say what happened.