Abaqus Explicit initial rotation

[bozosprzoj]

Dear all,

I have a very simple but at the same time a very frostrating problelm

I want to freely rotate the cilinder around its axis with an initial rotational velocity, while the inner surface is constrained to the reference point in the middle, as shown on the figure below. The simulation has to be done using Abaqus/explicit

I tried to simply use the coupling constraint to connect the inner surface with the reference point in the middle and prescribing the inital rotational velocity to the whole cylinder, but I get very strange results.

Since there is no boundary condition on load action on the cilinder, it sould freely rotate in space with its initial rotational velocity, but insted the results show very unstable results with high levels of stress, which doesnt make sence.

Does anyone know how this constrains should be modeled to get reasonable results?

I am also attaching the .inp of the problem.

http://www.filedropper.com/tryex2

Many thanks,

Uroš

 

[akabarten]

Hi,

You have to define roatation initial velocity for all nodes not only for reference point.
You can do it with keyword *INITIAL CONDITIONS, TYPE=ROTATING VELOCITY.

Regards,
Bartosz


VIM filetype plugin for Abaqus

https://github.com/gradzikb/vim-abaqus

 

[bozosprzoj]

Hi Bartosz,

thanks for your reply. If you open the .inp file you can see that the initial velocity is defined to all the nodes not only the Reference point, so the problem shouldnt probably be there. Do you have any other idea maybe?

Regards,

Uros

 

[rstupplebeen]

How are you applying the initial rotational velocity to the other nodes since they don't have a rotational DOF? You could instead create an initial step that accelerates the rotational velocity of the reference point. I hope this helps.

Rob

Rob Stupplebeen
Rob's Engineering Blog
Rob's LinkedIn Profile

 

[bozosprzoj]

Thanks, Rob

This is a valuable point. Solid elements indeed do not have a rotational DOF. Your advice that proposes incorporation of an initial step could be a solution. I am though interested especially in the case where such step would not be needed,(since i will use this approach in a large assembly later, where such step would not be efficient) and the initial velocity vould be prescrived in the initial conditions.

Since it seemes you have some experiences, would you have any other sugestion?

Regards,

URoš

 

[rstupplebeen]

You may be able to create a cylindrical coordinate system and apply the initial conditions that way. It's been a while since I've worked on rotating equipment though.

I mainly work on medical devices such as implants and the body doesn't usually spin.

I hope this helps.
Rob

Rob Stupplebeen
Rob's Engineering Blog
Rob's LinkedIn Profile

 

[bozosprzoj]

Thanks again Rob,

if you look at picture at the first figure you can see that cylindrical coordinate system has already been made.

I know it is a very strait forward problem that should not take a lot of efford to come accros, but it has been bodering me for weeks now. Any more suggestions would realy help me out.

Regards,

Uroš