How should I constrain a moving cylinder in abaqus?

Question

How should I constrain a moving cylinder in abaqus?

SommerK (Student)

(OP)

24 Sep 23 09:31

Hello together,

I have a question about the interactions and constraints in Abaqus.

I am doing a static analysis and I am modeling a cylinder that is rotatable on two sides (see picture). The cylinder rod is movable in y-direction and pushes with a force on a plate via a lever. I am now wondering what constraints and interactions I need to use for the rotating support and the moving cylinder rod. Currently I have Contact defined for the cylinder rod and the hole of the cylinder. However, an eigenvalue problem arises during the simulation. Then I defined a reference point in the center of the cylinder rod and used Coupling to relate the area of the rod to this point. With the help of a displacement constraint I restricted the movement, but the result does not correspond to reality. I have the same problem with the rotatable bearing. I have defined a Contact and then used an RP and Coupling to constrain the movement. However, the cylinder does not rotate in the result although it should through the lever. Do any of you have ideas? In itself, I understand how the system works, but I do not know how to get it translated into Abaqus.

I would be very grateful for any tips and suggestions.

Many greetings!

FEA way (Mechanical) · Answer 1 · 2023-09-24 11:00:11

Mechanisms are not easy to properly constrain for static analyses where you have to avoid rigid body motions. You can run a modal analysis to see if boundary conditions and contacts (those will be frozen) are correctly defined. In the actual static analysis, it would be best to use prescribed displacement instead of force. Automatic/contact stabilization may also help. If it doesn't work, you may have to switch to a dynamic implicit step.

rb1957 (Aerospace) · Answer 2 · 2023-09-24 13:59:05

I guess you don't see the problem with doing a static analysis on a mechanism, that'll move.

From your sketch it is hard for me to see how the purple rod moves in the y-direction ... that rectangular block looks fixed in y- ?

As above some part of your model is fixed, and some part is moving ... use enforced displacement.

"Hoffen wir mal, dass alles gut geht !"
General Paulus, Nov 1942, outside Stalingrad after the launch of Operation Uranus.

SommerK (Student) · Answer 3 · 2023-09-24 16:51:23

Thank you for the feedback and hints.

I have also thought about the dynamic aspect, but in my analysis I want to look at the exact moment when the piston rod presses on the plate via the lever. In this case it should be static or not? In other words, the rod should not move first.

However, since the plate can deform, the cylinder rod would theoretically move minimally in the cylinder guide (the guide is a simple hole with the same diameter as the rod). It is also possible to twist the cylinder due to the upper rotatable cylinder holder (I fixed the holder to the left and right of the cylinder with a Tie-Constraint).

In my case I have specified a cylinder force and I want to see how much the plate moves in the y-direction.

rb1957 (Aerospace) · Answer 4 · 2023-09-25 13:52:45

sure you can do a quasi-static analysis ... of your mechanism in a moment in time. Each component is connected as it would be in the real world, transferring forces and moments (or not) as you (well, the structure) sees fit.

But I still don't understand your model ...
are the two grey lus fixed ? (but then the block isn't moving in y ... unless the block has a slot in it?)
the little grey Tee fitting ... this is a slider ('cause he can't apply load to the rod)

so you need to figure out how is load being applied to the model, and how is the model reacting this load (how is it attached to "the rest of the world" ?)

"Hoffen wir mal, dass alles gut geht !"
General Paulus, Nov 1942, outside Stalingrad after the launch of Operation Uranus.

Come Join Us!

Posting Guidelines

How should I constrain a moving cylinder in abaqus?