MES Impact Drop Test

[Anonymous]

Hello All,

I am performing a "drop test" on a shipping container (approx 8ft x 8ft x 16ft long) using Algor 2012.  This involves an MES with non-linear material models simulation of the container dropping to an impact plane under the acceleration force of gravity. 

What I am ultimately concerned with is whether or not the container ruptures (if it breaks or splits open at any area; ie. if it was container which held water at the start of the analysis, would it leak at any time throughout the simulation).

Is there an easy way to determine this from the results options?  I am envisioning an option to check to see if nodes from adjacent elements that started the simulation with the same coordinates, ever resulted in different coordinates (indicating a "hole").

I am seeing very large stresses in the results which are beyond the "ultimate tensile strength" of the material - would this alone indicate that ultimate failure has occurred, even though no visible separation between nodes has occurred?

Thanks!

Joe Lowry, PE

Mechanical Engineer

[Anonymous]

That is how I would take it.  I had a similar MES with an assumed failure, a projectile hits a .1 inch thick  and dents it.  I wanted to see ripped steel.  And I did for a while, but I slowly found issues with my model. Once fixed the steel still did dent, but just under yield.

[Anonymous]

If the stresses are above ultimate, doesn't that mean cracking?  Are you wondering is if the cracks open?  Or is it your wondering if the stress wave causes enough deformation to temporarily allow liquid to escape?  Or both?

[Anonymous]

Overall, I am trying to determine whether or not the stress will cause the material to crack and deform enough for liquid to escape.  The structure is mainly welded steel panels, but there is one bolted panel where I am also interested in knowing whether or not the joint is deforming enough to to allow liquid to escape.

Thanks for response.

[bannecke]

Hello,

according to my understanding it is not possible to do fracture analysis with MES.

This analysis will take you a long way into the the depth of structural mechanics.

Afaik Simulation is not able do sepearate nodes. Connected Nodes will always stay connected, but eventually seriously distorted.

You can assume a fracture when the deflection is above the allowable reflection of the material. But whith the effect of weldings and internal stress of the material this is only a rough approach. 

When doing non linear analysis you should only get stress above the ultimate stress in the contact areas.

Otherwise you should double check your material parameters.

Regards

Boris

[zhuangs]

MES can help do the drop test.  After the simulation, you can check the maximum stress for comparison with the ultimate stress.  But I don’t think MES can do fracture analysis in the drop test process.

-Shoubing

[JohnTomasik6493]

I'm doing a very complex drop analysis (that's struggling in many ways we've been trying to resolve with Autodesk), and I'm seeing stress values above the stress limits of the plastic.  The problem is that I'm using an existing design (as a baseline) that's been proven in the field to not break under these conditions.  So, the analysis isn't accurate.  We've been trying to dial in the analysis to get the most accurate results, but I have at least one question and a comment:

Question-I'm using materials from the library that comes with the Autodesk Simulation product.  Do these materials have actual material curves built into them?  If not, how is it a "non-linear" analysis?

Comment:  My growing understanding of impact testing (still nowhere near an expert by any means) is that the strain rate has everything to do with a material's ability to absorb the impact energy.  So, the UTS of the material alone doesn't determine whether the analysis predicts failure.  Is that correct?

[AstroJohnPE]

Hi John,

I would describe the material properties included in simulation material library as a rough approximation of the behavior beyond yield. After all, the entire plastic region is based on two points on the stress-strain curve: the yield point and the ultimate point. This data may be good enough for a first approximation, but it would be better to enter real stress-strain data.

Secondly, the material properties from the library that are used in the analysis are determined by the material model that is chosen under the Element Definition. (The default is isotropic which assume the material is linear all the way to plus or minus infinity.) Once you have the material model chosen, edit the material properties for the part to see the properties used in the analysis.

Keep in mind that there are many types of nonlinear effects in an analysis, where nonlinear means that doubling the load does not result in double the displacement/stress/etc. Examples of nonlinear effects include:

1. material (which can be totally elastic but not linear, such as rubber)
2. geometry (some parts become stiffer or weaker as they deflect, such as a guitar string, or snap-through parts)
3. contact (the number of points in contact is not necessarily linear with the load)

I think you are correct about the strain rate and failure, but I cannot find my notes where I have something written down. If I can find those notes, I will follow-up with another post. But, I am not sure how you would include the strain rate in the analysis or interpretation of the results.

Specifically, what type of material are you using?

[JohnTomasik6493]

John, the housing is a Makrolon 2258 high impact, medical grade polycarbonate.  The housing is what contacts the impact plane.  Two other key materials are some type of polyurethane that act as hinges (I don't have a spec for that yet), and an RTP 399 X 136784 PTFE/Silicone impregrnated polycarbonate.  

If you find any information on the strain rate and how it could be used to interpret results, I'd like to know.