Calculating angular and linear acceleration from a hollow object on drop test

[jonny17]

Hi all,

I was wondering if anyone could suggest a method for calculating the angular and linear acceleration from a hollow part during a drop test. I have attempted creating a set of the whole object to no avail. Is there a way to history output directly from the centre of mass? I have tried creating a RF point at the coordinates of the centre of mass but as it is hollow it has no mass properties and hence will not work.

Any help is greatly appreciated
Thanks
Jon

 

[burningDiesel]

If you are using rigid bodies (which I doubt you are), you can easily use:

*RIGID BODY, REF NODE=<node>,
ELSET=<element set>, POSITION=CENTER OF MASS

To put a ref node at the center of mass. I encourage you to run your simulation using rigid bodies anyways (depending on what you are dropping), at least for a first preliminary result.

If you are using deformable bodies, you would indeed need to put a point at the centre, and link it (using mpc, or connectors or *equation) to your model. How you connect depends on what you are doing. If you e.g. do an experiment with an accelerometer in the centre of your hollow part, the accelerometer is attached to the part in a way also.

If abq complains that the ref node has no mass/inertia, just give it an extremely small one that doesn't influence the result.

In short: link ref node with model, give it small inertia.