Modeling Impact of one object on another (COSMOSWORKS)

I'm trying to figure out how to model the plastic strain that results
from the impact of one object on another.

Basically, I'm trying to answer a question like, "How fast does a
baseball have to be traveling for it to damage (plastically) an
aluminum baseball bat?"

I have COSMOSWORKS both the main and the motion portion, but I cant
quite figure it out. I can't use the "drop height" simulator because
it only involves the impact of one object on a flat surface, not the
impact of two specifically modeled objects with material properties
and internal characteristics.

Is this even possible in COSMOS? Do I have to go and learn how to use
ANSYS? Ugh.

Any help appreciated.

 

Since you're not worried about what happens to the ball, don't model
it.

Work backwards, of sorts. See how much it takes to cause a local
failure in the bat. You'll have to figure out the contact area first,
though.

Pick a "fast" number and model a drop test and see if that gives you
a rough idea of contact area size. I'm not very familiar with the
drop test functionality. Otherwise set it up as a contact analysis
with the appropriate numbers.

Take advantage of the bat symmetry and refine your mesh only
in/around the contact area to keep your solution times down.

Once you know your failure criteria for the bat you can work
backwards to find out how fast the ball would've needed to go.

-Tony

 

Thanks for the feedback Tony. I really appreciate it.

However, your approach isn't really gonna work for me, as I miss-typed
the original question. It *should* have read: "How much plastic
deformation is done to a given surface (or subsurface) for a given
projectile velocity?"

I think you've generated a viable approach, namely:

1) Find out what sort of contact area we're looking at from the
baseball at X m/s contacting a perfectly hard surface or more
reasonably, contacting a surface with some given Young's modulus and
Poison's ratio (L-E material w/ no plastic deformation). I know this
is possible via the COSMOS drop test (well the scenario is... I'll
have to play with it to find the given contact area).

2) Define this area as the contact area for my baseball bat, taking
appropriate steps to modify this contact area based on the curvature
of the bat. Perform a static force analysis(?) given the force at
impact... The thing is, the ball has only momentum, there's no force
behind it, so I'm not exactly sure how to go about modeling the force
distribution over my contact area.

Thoughts on this last part? Does anyone else have a different
suggestion on how to go about modeling the damage done to the baseball
bat?

Thanks again, Tony.

-Mike

 

This is not a CosmosWorks problem. It is an LS Dyna or Abaqus Explicit
problem. It might also be tackled in Cosmos/M though not as
elegantly.

One of the problems you face is that the problem is highly non-linear.
It is also the type of problem that requires lots of very small time
steps and a very refined mesh where plasticity occurs. Being non-
linear (contact, friction, non-linear material behavior, geometric
stiffening) it is path dependent. The material properties for the ball
are likely to be a bit difficult to obtain. The information for the
aluminum under high strain rates might be easier to find. So even if
you could set it up it will take some time and trial and error to get
it to run and give meaningful answers.

TOP

 

Probably. This is a hard problem. You need to know the non-linear material
properties of the bat and the ball at very high strain rates.Then you have
to solve the problem with the very short time steps for a long time. One of
my colleagues worked on something roughly similar for his Master's degree.
He spent a lot of hours running analyses, running real tests on bats,
comparing the results, tweaking the analyses, running tests...

Jerry Steiger