Learning Forum

[georg.gruber]

Dear Ansys Community,

I want to set up a viscoelastic material model based on creep curves (stress vs. time curves for various stress levels). The curves could look like this:

The material model should be based on prony series, what is already implemented in Ansys. The material model can easily be set up based on shear relaxation curves (which I don't have). Since creep curves - as shown above - are much more common (the experimental setup is much simpler) I asked myself, if there is also a method to get the prony series parameters based on creep curves.

I know creep material models (like mod. time hardening) can directly by setup based on creep curves (there is a curve fitting procedure accessible in Ansys Classic). In my case, I'm specifically interested in the viscoelastic approach.

Many Thanks in advance.
Greetings,
Georg

 

[John Doyle]

The Prony Series expression is intended to be superimposed on a rate independent base material model (either linear elastic or hyperelasitic).  The experimental data input for curve fitting must be shear modulus vs time (or frequency) and/or bulk modulus vs time (or frequency). Please refer to Table 7.19 of the MAPDL Material Reference Guide (Release R2025-R1). Since the Prony coefficients (alpha factors) are unitless ratios representing relaxation for a corresponding relaxation time, you can substitute stress components that are proportional to modulus values to get the same result.  Since strain equals stress/modulus, you might be able to convert the data you have to stress vs relaxation time, if you have the modulus values and/or if you know the load and cross sectional properties of the test coupon used to generate this test data.

[Dennis Chen]

 

Hello, georg.gruber.   linear viscoelasticity, which is implemeneted as a linear elastic material model with a set of prony series terms, is usually generated from a DMA master curve.  From there, you can compute the instantaneous and long term modulus. Using a maxwell model, one can establish the viscoelastic response using a general form such as one shown below

Below paper provides a good summary – https://www.redalyc.org/journal/404/40465092006/html/#B16

I post these because I don’t get how one can get this from looking at a creep curve.   If you have a good reference you are willing to share that explains the methods, I would really love to learn.

Thank you!