Hybrid model material properties

[overaker]

Jorgen, what standard material tests are required to determine the Hybrid model parameters for a material? Uniaxial tension and compression (at different rates?), cyclic tension-compression, and what else? Thanks for preliminary info.

[Jorgen]

Good question. I typically use one of the following:
:arrow: Uniaxial tension (both loading and unloading, i.e. cyclic loading) at different strain rates.
:arrow: Uniaxial compression (both loading and unloading, i.e. cyclic loading) at different strain rates.

If you need predictions at different temperatures, you will also need to repeat the experiments for a few temperatures.

It is also possible to use a number of other tests instead of uniaxial loading, for example, simple shear, biaxial loading, etc.

- Jorgen

[overaker]

Ok, so I guess the testing strain rates should be within a range representative of the environment in which the material will operate?

And have you found from your work with UHMWPE for orthopaedics applications that the testing environment is important? Does the material need to be tested in buffered saline at 37C for accurate determination of material properties?

I'm surprised that you need only test in one direction. Aren't there differences between tensile and compressive behavior that require testing in both directions? Will the Hybrid model capture the tension/compression differences?

Thanks again for your advice!

[Jorgen]

As always, for best accuracy and most reliable results, the testing should be performed in the same environment and performed using similar strain rates and temperatures as for the real application.

Sometimes it is difficult to reproduce the environment accurately. For example, it takes quite specialized equipment to perform that tests at body temperature in a buffred saline environment. Under these circumstances it is often possible to simply reproduce the dominat aspect of the environment (e.g. the temperature) and ignore some other variables (e.g. the buffered saline). In my experience, for material parameter extraction of UHMWPE this can be sufficient. The thing to remember is that you can never get a 100% accurate material model, so having a few % error in the experimental data might be OK, as long as the data does not have systematic errors. The key to all of this is to perform a significant validation program, after the material parameters have been determined, in order to validate the accuracy of the material model predictions.

Your final point is important - I did not express myself quite right. Yes, you should perform both tension and compression experiments for the calibration. The Hybrid model was specifically designed to capture eventual differences between tension and compression. For calibration, you need to perform both tension and compression, or as an alternative, you can perform either tension or compression in combination with another loading mode such as shear.

Jorgen