Hi

I have never made the kind of simulation you are asking about. For what it is worh, what you are writing should be correct, but only when your system is lossless, i.e zero damping and zero friction. Also, numerical damping needs to be zero for things truly to mimic theory.

I guess some pitfalls can be - are you tracking the velocity both of plate and ball when comparing with KE? If Abaqus takes into account also local flexibility, I would expect there to be also some KE in the part of the plate that is flexing.

Hope this helps more than confuse the issue

Claes

you only get zero kinetic energy in this situation if the system is fully ideal. since abaqus uses so many predefined assumptions then never you can get the ideal situation so it is normal not to have zero kinetic energy.

Hi,

I assume you are using the explicit solver. If so, you may want to go through the odf document at the following url:

http://imechanica.org/files/0-overview%20Explicit.pdf

In particular, have a look at the section "energy balance"; as you'll see, there are several criteria to be fulfilled in order to have consistent results, and these involve other forms of energy other than kinetic and strain.

I hope this helps