Introduction to Adiabatic Heating
If you deform a tension specimen it will exhibit viscoplastic energy dissipation, and that energy will increase the specimen temperature. Most people ignore the increase in specimen temperature during the deformation, but what if you want to be more accurate? What if you want to include the specimen heat up due to the dissipated energy during your material model calibration. How can you do that? One way to do this is to use an infrared thermometer to directly measure the specimen surface temperature during the test. Another way is to use MCalibration and include adiabatic heating as part of the material model calibration. In this article I will show you how to do this!
The default setting in MCalibration is to not convert the calculated dissipated energy to a temperature rise of the test specimen. The following image shows the experimental stress-strain data for a a polyethylene material, and the predictions from a calibrated Three-Network (TN) model. The figure shows that the model accurately predicts the stress-strain behavior. The figure to the right shows the specimen temperature, as calculated by MCalibration. The energy dissipation is not converted to heat!
Adiabatic heating can be activated by opening the material model dialog box and switching to the “Material Info and Properties” tab. To activate adiabatic heating you simply need to specify the specific heat for the material. Optionally, you can also specify how the adiabatic heat conversion depends on the strain magnitude and strain rate. If you are using the unit system [mm, N, s, K], then the specific heat should be given in μJ/(kg K). For polymers, Cp is typically between 1000 and 2000 J/(kg K).
If the heat conversion beta-factor is 0, then the dissipated energy is not converted to heat, and if the beta factor is 1 then all of the dissipated energy is converted to heat.
If I select Cp=1500 J/(kg K) then I get the specimen temperature history that is shown in the figure below. In this case the specimen temperature increases by almost 20K during the test. If I had used a temperature-dependent material model then that temperature increase would have influenced the predicted stress-strain response. Pretty cool!