Modeling Plastics and how to determine material failure 3

[ABQfm]

Hi
I have a general question on whether to use Compression or tension test data for plastics in FEA modeling? I have come across couple of PTFE based plastics where compression modulus is twice as much the tension modulus. For metals I observed it would be nearly the same so doesnt matter which to use. However, for PTFE plastics which one to use? In the past, I have looked at the part whether it is loaded in tension or compression and have used appropriately. But I still am not satisfied and am looking for some concrete procedure. I am more towards using tension data as it is more conservative and I can determine the failure by looking at my strain or stress levels and comparing to the ultimate limits. For compression, this seems to be limiting factor.

Please throw some suggestions.

Thanks
FM

 

[TERIO]

I have no background in polymers. That being said I don't think people typically model them using linear elasticity, they use hyperelastic elastic models. Some material model names to look for are Neo-Hookean and Mooney-Rivlin. Some information (pretty mathematical) can be found at:

http://solidmechanics.org/text/Chapter3_5/Chapter3_5.htm

 

[ABQfm]

Thanks for the reply.Unfortunately the material has significant plasticity involved and hence I cant model using hyperelastic laws.

 

[EdR]

Don't know what software you are using but many systems allow use of different moduli for tension and compression properties....

Ed.R.

 

[ABQfm]

I am using Abaqus (with CAE) and not sure if I can specify tension and compression modulus together.

 

[seymours2571]

PTFE (aka Teflon) is a pretty stiff material. I don't think it would be classified as a hyperelastic material such as rubber.

This may be crude, but if you software does not allow you to set differnt moduli for compression and tension could you run two seperate simulations using the compression and tension moduli that you have uniformly.

I read a procedure published by someone several years ago with respect to Ansys software where they would run the simulation using either the tension or compression modulus uniformly. Then using the built in programming language would change the material properties for the elements in compression to have a modulus that corresponds to the compression modulus. Then they would rerun the simulation. I never tried it, but it might be an option for you.

Steve

 

[kellnerp]

You are asking about failure and looking at stresses in a material that is both temperature dependent and exhibits creep. You haven't said anything about what this part does and what might be considered a failure of the part.

That material is probably going to exhibit a lot of elongation in tension. At what point will you consider it failed?

When you do analysis on plastics always consider:

1. Temperature history
2. Load/Time history
3. Environment (particularly humidity and solvents)
4. The actual behavior of the plastic you are going to use, not just the online datasheet numbers.


TOP
CSWP, BSSE

http://www.engtran.com

http://www.niswug.org

"Node news is good news."

 

[flash3780]

It's been a while since I've looked at analysis of plastics, but based on what I know it get's pretty hairy pretty fast. The properties of many polymers are often nonlinear, temperature and rate dependent. Some polymers exhibit hysteresis in the stress/strain curve (i.e. a different stress/strain curve when loading and unloading).

Example of ?-? curve showing temp dependence: ABS.

You'll need tabular stress-strain data, or material constants related to one of the stress-strain relations available (Neo-Hookean, Mooney-Rivlin, etc). ABAQUS seems to work reasonably well in the elastic range.

It may be difficult to find material data, let alone a suitable mathematical model for plastic behavior of plastics (ironic, eh?).

On top of the material property challenges, often small displacement/rotation assumptions go out the window.

ABAQUS touts their FeFp material model for analyses like you're describing. However, this paper thoroughly trashes it's accuracy. Mind you, they're also selling another solver.

All-in-all, I'd rank plastic analysis of polymers right up there with forced response vibration analysis on the voodoo scale. It's not that I think it's impossible, I just think the material data is tough to get a hold of and the plastic response of polymers under combined loading is not well understood. Of course, there's been a lot of work done in that area recently, so perhaps things have progressed. Good luck.