Learning Forum

[aquadragon19]

I am trying to look at reaction forces on my part that would be caused by a rotation of the part. I tried using a remote displacement to apply a 30 deg rotation, but I think it deforms the part in a way that would be unrealistic. I applied it both to the center I would want to rotate it about, as well as the faces that would be near it. Would it be better to apply the displacement to the entire body? 

(applied across top face vs applied at an edge through middle of top face, ideally what the part would be rotating around) 

[peteroznewman]

If you set the Remote Displacement to have Behavior = Rigid, that will help.

[aquadragon19]

Ah, I missed that setting, much easier thanks! Also, if I were to use coupled, that wouldn't be like adding in a moment from the fixed point would it?

[aquadragon19]

Also, how is setting it as rigid really any different than applying it just at the edge? Doesn't remote displacement work by applying the displacement at one point? Or conversely, if I make it rigid, it looks like the part doesn't really experience any stress, which wouldn't be true. 

My next through would to separate the part into two bodies, and make them connected via a revolute joint so the part rotates around the location that I want, but I feel like that is overdoing it.

[Wenlong]

Hi,

Yes, remote displacement works by applying displacement one remote point, but that remote point controls the behavior of the whole related geometry. By applying direct displacement to the edge, you are making every point on the edge move the same amount, but if you have a remote displacement with behavior as rigid, that edge won't deform and moves like a rigid body. Points on the edge can move for different amounts in this case. (Imagine you have a stick, and you are pulling one end laterally, the stick can rotate and move but won't deform, that's how remote displacement with rigid behavior works).

Regards,

Wenlong

 

[aquadragon19]

I see your point. I am trying to simulate moving the entire object above around a 'fixed location' (where I placed the edge). So I want everything rigid, but I when doing that in the simulation the top of the bar doesn't really undergo any stress. I tried applying the remote displacement to the whole top face, and the results are very different, so I guess what I am really struggling with is where the most realistic place to attach the displacement is. I attached a couple different examples below showing how where it is applied drastically changes the stress. 

[peteroznewman]

Another part touches the top and causes it to move.

If you model that other part and move it, the top section will make contact with that other part and the top face will deform in a more natural way.

[aquadragon19]

I have tried this and while it gives the 'realistic' deformation, I don't know if I can use it because I am trying to look at the forces and moments produced by the object above. If I used a force probe, wouldn't that include the effects of the other object? (Sorry I am relatively new to FEA)

[peteroznewman]

Do you have gravity or is the only force coming from deformation of that stalk?  

Without gravity and with no contact with the stalk, the forces and moments required to move the other part with a remote displacement are zero.

If there is no gravity but there is contact with the stalk, when you apply a Remote Displacement to the other part, you can probe the forces and moments used to move that remote point. It will only report the forces transferred from the stalk.

[aquadragon19]

I am using a shape memory alloy, so the only forces would be coming from the stalk. I am trying to simplify the model a bit, but I don't think I would get an accurate reading unless I had the other attachments on the part for stresses etc on the part itself? I think the other issue is coming from the fact that this is a shape memory alloy (Nitinol) and the engineering data given to me is not properly defined, from what I can tell.