Selection of constitutive model

**Nithyanand Kota** · 2011-10-31, 16:50

Hello,
Given the choice of many models for predicting elastomeric behavior, how would one go about selecting the right model. Is there a rule of thumb, or is it like curve fitting i.e as long as the model fits all the available data sufficiently well then its okay.

Specifically why should one choose Ogden over BB or vice versa?
(ogden constants are prony series based, i.e ogden is modified to include rate dependence)

Any leads or referance is helpful
Thanks

**Jorgen** · 2011-10-31, 19:55

There are indeed a number of different models that one can try to use for predicting elastomeric behavior. The Bergstrom-Boyce (BB) model will always be more accurate than the pure hyperelastic Ogden model. The only reason why you would not chose the BB model is if you only have very limited experimental data and cannot run a small number of additional experimental tests.

-Jorgen

**Nithyanand Kota** · 2011-10-31, 21:01

Thanks for the reply Jorgen, appreciate the quick response

If I have tensile test data at different rates, (not relaxation or creep), which one do you suggest?.

Also is there any fundamental reason why abaqus accepts only creep or relaxation data for calibrating the viscous response, Is there some non uniqueness that results if one uses tensile test data at various rates for characterizing the viscous response.

Thanks

**Jorgen** · 2011-11-01, 19:02

I am a big fan of the Bergstrom-Boyce (BB) model [no surprise since I developed it

]
The BB model actually works quite well for most elastomers.

No there is no fundamental reason by Abaqus only accepts creep or stress relaxation data. The MCalibration software allows you to calibrate any material model in Abaqus using tensile test data at different rates.

-Jorgen