impact simulation

[gbon]

Hi Everybody,


I have to analyse a plastic part that fall on a concrete floor for determining if it breaks or not.
So I have to simulate a sort of crash, impact, shock.
I know that dynamic shock simulation is not for this.


So is there any documentation or tutorial that would help me to do this ?
Thanks for any advice,

 

[cwilson]

Hello,


I need to do something almost identical to this and was just curious if you made any progress that you could share. I haven't had the greatest of luck but if I fine something I'll be sure to pass it on.


Thanks, Christina

 

[nschlueter]

Tough subject...

I have not had much luck with this.

You would have to use Mechanism Dynamics to perform the
drop test and use the force resultants in Mechanica to
perform a static analysis.

Wildfire 5 is supposed to deliver 3D contact to Mechanism
and it may have something new to get this done...

 

[Dynamica]

take a look at this link:
http://www.nxrev.com/WEBCASTS/Q3_2008/5-16-08.wmv

about 3/4 in Dave briefly explains a technique for using a drop test of a Moen shower head.

Cheers

 

[cwilson]

That was a very helpful presentation thank you.

 

[mneal]

Darn,
Seems like the link is broken.
:-(

 

[Roger@Mallett]

gbon,


You'll have better luck trying to solve these types of problems with an Explicit code like LD-Dyna or ANSYS Explicit STR. They are designed for short-duration/impact/large deformation type problems and can solve these problems in hours whereas an Implicit code could take days.

 

[jimshaw]

gbon,


There is a big difference between determining IF it breaks versus HOW it breaks. Perhaps you could do some sort of dynamic/static combination like what's mentioned above to determine if any of the materials see stresses past their ultimate strength... But with the nonlinear material properties of plastics, andto really understand what happens in that situation you need 3D contact, re-mesh, element death capabilities with an explicit code. LS-Dyna (not LD-Dyna) will get you what you want.If you want the most accurate solution in the shortest amount of time, then I suggest you farm it out to a place that specializes in these types of analyses.


Good Luck!


Jim
http://www.linkedin.com/in/shawengineering

 

[athurart09]

take a look at this link:

http://www.nxrev.com/WEBCASTS/Q3_2008/5-16-08.wmv



about 3/4 in Dave briefly explains a technique for using a drop test of a Moen shower head.



Cheers

Such a very amazing link!
Thanks you for the post.

 

[gbon]

take a look at this link:
<a href="http://www.nxrev.com/WEBCASTS/Q3_2008/5-16-
<br / target="_blank">08.wmv">http://www.nxrev.com/WEBCASTS/Q3_2008/5-16-
08.wmv</a>
about 3/4 in Dave briefly explains a technique for using
a drop test of a Moen shower head.
Cheers

Such a very amazing link! Thanks you for
the post.

The link doesn't seems to work anymore.
Any other link ?

In an other way, I've seen that Solidworks have a built-
in drop test module.
GB

 

[SW]

There is a danger with built in modules like this. The physics of a
drop are complex, apart from the impact and contact mechanics,
there is the fit of parts in the assembly, and the tolerances or surface
imperfections meaning the contact area is not what it appears to be.

The functionallity can be useful for modifictions used in conjuction
with physical testing but I am extremely cautious about hanging my
confidence in a design off the results.

Sam

 

[Mr.helmet]

Hi Gbon,


This type of analysis is typically done with Explicit dynamics (time based). The code is for ideal for figuring out things like how your clip will react, with respect to time, given gravity, an initial velocity, material properties, etc...


If all you have is implicit, you can solve for deflection given a certain load, so you can determine how bent your plastic clip can get before it yields, but not how high you can drop it from... hope that helps.


word of caution, if you're modeling large strain plasticity (>5%) you'll need a hyper-elastic model with plasticity (uses True stress/Log strain rather than nominal stress and strain) Elastic plastic models using nom stress/strain data aregood forstiff thermoplastics, if you consider plastic yeild failure... my opinion


I use Abaqus explicitfor impact analysis, another common code is Ansys's LS-Dyna


feel free to shoot me an email ifyou have any questions.


[email protected]


I'd be happy to offer some pointers, and/or consult if that's of interest?