Define a Polymer (PPHC) on Mechanical behavior section

[a21887]

Hi, I am new on abaqus and i'm trying to make an impact teste on a Polymer, "high crystallinty polyolefins" HCPO on sabic database, but i'm having problems to define de material property in the mechanical behavior section.
What's the best model?! Drucker-Prager??? there are a subroutine specific for that kind of material.

I have the Tensil teste results, and i think i have to introduce that on abaqus...


I need help....
I have already read all de abaqus documentation and i don't know what to do....


THANKS in advance!!! [/b]

[sq]

This is an interesting question, the way you put it.

I'm going to guess that you want to use Drucker-Prager as a failure model because of regions of high crystallinity, in a sense acting like a composite with local slipping and shearing? Otherwise, for an unreinforced polymer, where friction is not going to be an issue, I probably would have started with Johnson-Cook plasticity.

In any case, I believe very strongly that you will have to use the strain-rate dependence suboptions (available in ABAQUS). Along these lines, if you're going to model impact failure, you need to have some higher rate stress-strain (and, better still, failure) data.

[a21887]

Thanks for the answer sq, so on your opinion i must use Johnson-Cook plasticity, but that model isn't for metal, i think, and with that kind of model i don't put de stress-strain curves at diferent strain-rates that i have on hand...
I sorry but i'm new on this. Can you tel me the steps to define a polimer...
There are at least 3 part on a tensil test, isn't it,
viscoelastic - softning - hardening

how do i define this on abaqus, and do i introduce the curves on it???

I'm sorry for my bad english. I appreciate any help.

Thanks


Jorge From Portugal

[Jorgen]

Here are a few more comments/questions:

:arrow: What temperature are you interested in? and how brittle/ductile is the material at that temperature?

:arrow: How brittle/ductile is the material at the strain rate that you are interested in?

:arrow: What are you ultimately trying to do? predicting failure, or predicting deformation and failure?

:arrow: Do you have experimental stress-strain curves for the material?

:arrow: Do you have failure data for the material?

:arrow: How accurately and how much time do you have to model the polymer component?

These questions will help you decide the most appropriate material model and the most appropriate failure model. Note that the two are often decomposed.

I agree with SQ that the Johnson-Cook plasticity model can be a reasonable starting point. If accuracy is more important than time and
cost then there are other more specialized FE models that can be used.

Best of luck,
Jorgen

[a21887]

First of all i want to show you the experimental data that i have on hands:

http://polymerfem.com/polymer_files/external/tensilecurves.JPG
http://polymerfem.com/polymer_files/external/Moduluscurve.JPG

the tensile curves are at diferent strain rate...

The impact occurs at a velocity of 4m/s, at 23ºC and the impacter is a rigid body with a mass of 24kg. The deformable body has the shape of a " U " and can't move on the impact axis, blocked with a rigid wall.

The material has a density of 905kg and is considered as a ductil material...

What i'm trying to predict is deformation behavior not failure. the elastic deformation, i supose viscoelastic or hiperelastic. which is the best to predict elastic??? and plastic deformation, that i don't know which is the best...

the results that i want from that test is the absortion energy and the reaction force of the rigid wall.

i need help, from you experts, because i must have results at the end of february...

I'm finishing the last year of Mechanical engineering and this is my final project.

Thank you very much.

[Jorgen]

Very nice experimental data!

Since you need results quite fast, and I assume that you have a limited budget, I recommend that you use a simple J2-plasticity model. That is, use: *Elastic and *Plastic. These commands are very simple to use and calibrate. I would use your experimental data at room temperature to calibrate the model and then simply perform the FE simulation. That should enable you to get a rough estimate of the absorption energy and the reaction force.

Now, if you want more accurate results then I would recommend the Hybrid Model (http://www.polymerfem.com/modules.php?name=Downloads&d_op=getit&lid=12). This model is likely to be very accurate for your material and loading conditions.

Best of luck,
Jorgen

[sq]

I won't repeat what Dr. Bergstrom has already (correctly) said, but I would add that your initial FE solution will also give you some idea of the strains and strain rates likely encountered in this material.

This will be important information, if for no other reason than to include at the end of your report, in the section called "future work", as you will in the future want to investigate the material response (not only modulus, but loss as well) at higher strain rates.