Document Type



Available under a Creative Commons Attribution Non-Commercial Share Alike 4.0 International Licence


Materials engineering

Publication Details

Presented at ECCMR 2013, San Sebastian, on the 26th of June 2013 .


Strain amplitude control is most often employed when carrying out fatigue testing of rubber components. Often the design engineer requires fatigue test data that is based on load amplitude control. This is analogous to engineering or nominal stress amplitude control. This usually makes it easy to maintain the load within the specified test limits during extended testing. Values of true stress and strain can be obtained from this approach, but the magnitudes of the maximum true stresses in these tests increase with accumulated cycles, whereas the equivalent maximum engineering stress values remain constant. This is easily demonstrated in the uniaxial case; however the nature of the applied load in equi-biaxial testing adds complexity. The effect in the equi-biaxial case is amplified as the principal stretch ratios have a more pronounced influence on both true stress and strain than in the uniaxial case. Engineering stress-strain data is particularly useful when representing the fatigue behaviour of elastic materials; however it does not represent reality in the case of rubber samples subjected to repeated high strains. True stress control testing can provide data to verify viscoelastic models used to describe elastomeric behaviour under bi-axial fatigue conditions. A test programme has been devised to examine the effect on fatigue life and stress strain relationships of a material tested under conditions of maximum biaxial true stress. This paper compares the difference in fatigue test predictions when using engineering stress and true stress amplitudes as the control parameters. The Experiments are in the form of equi-biaxial bubble inflation fatigue tests employing the DYNAMET system as described in previous CER publications.