Prediction of fatigue life of rubber mounts using stress-based damage indexes

Abstract

Prediction of fatigue lives of a rubber mount necessitate formulation of models for estimating fatigue life of the rubber materials used in the mount. Moreover, the prediction accuracy of the model is strongly dependent upon the choice of damage index that are based on different strain, energy or stress measures in the vicinity of critical locations of the rubber mount. In this study, relative performance of models employing different damage indices are evaluated for prediction of fatigue lives of rubber material and a drive-train rubber mount. A combined stress and an effective stress function are proposed as a damage index for predicting fatigue lives of rubber materials and the mounts. Different damage indices, identified from the finite element models of the rubber dumbbell cylindrical specimen are applied for formulations of fatigue life prediction models. The model parameters are identified from the measured data acquired for the rubber dumbbell cylindrical specimen under 31 different uniaxial displacement loads, using least squared error minimization technique. The identified models employing different damage indices are subsequently applied for predicting fatigue lives of rubber mounts under different magnitudes of loads applied along two different directions. The correlations of the predicted lives of the rubber mount from the models employing different damage indices with measured fatigue life data were subsequently investigated for the rubber mount subject to different load conditions. It is shown that the models identified for the rubber material could be effectively used for predicting fatigue lives of the mounts, which are made of same material. The fatigue lives predicted by the models considering either effective stress or combined stress as the damage index correlated with the measured data within a factor of two for the two, suggesting that stress-based damage indices could yield more accurate predictions of fatigue lives of typical mounts.

Fatigue of natural rubber under different temperatures

“…Cadwell et al (1940) initially used cylindrical samples, which were replaced by a Diabolo-like geometry in the work by Beatty (1964). This type of geometry was then used as the reference one (Svensson, 1981;Lu, 1991;Xie, 1992;Bathias et al, 1998;André et al, 1999;Robisson, 2000;Saintier, 2000;Kim et al, 2004;Ostoja-Kuczynski et al, 2003;Le Cam, 2005;Raoult, 2005;Bennani, 2006;Oshima et al, 2007;Le Cam et al, 2008Woo et al, 2009;Ayoub, 2010;Moon et al, 2011;Flamm et al, 2011;Poisson, 2012;Wang et al, 2014;Shangguan et al, 2017;Neuhaus et al, 2017). Such geometry is well adapted to fatigue since no buckling is induced under compression and fatigue damage generally occurs at the surface of the sample, at half of its height, due to stress and strain concentration.…”

“…It applies for initiation of external as well as internal cracks and is therefore easily applicable to fatigue tests. Authors linked the sample end-of-life to a loss of the chosen parameter: 15% of the maximal force at the 128 th cycle (Mars, 2001), 50% of the stabilized shearing effort (Xie, 1992), 20% of the first cycle effort (Woo et al, 2009;Moon et al, 2011;Shangguan et al, 2017), 20% of the maximal displacement at the 1000 th cycle for tests under prescribed force (Neuhaus et al, 2017); (iv) the brutal decrease of the maximal reaction force (Lu, 1991;Xie, 1992), it was later formalized by Ostoja-Kuczynski et al (2003) as the number of cycles at which the derivative of the stiffness is not constant anymore (Le Chenadec, 2008;Le Saux, 2010;Masquelier, 2014). In practice, this number of cycles is correlated with the occurence of a a few mm long crack at the sample surface (2 mm for Ostoja- Kuczynski (2005) and Le Cam (2005)).…”

“…Iso-lifetime curves obtained highlighted the influence of σ mean on the rubber fatigue life in a (σ mean ,σ amp ) plot. The Haigh diagram was then systematically used for characterizing the fatigue reinforcement of crystallizable rubbers (Saintier, 2000;Le Cam, 2005;Bennani, 2006;Oshima et al, 2007;Le Chenadec, 2008;Poisson, 2012;Shangguan et al, 2017). Note that the effect of σ mean can however be represented with Wöhler curves.…”

Strain energy-based rubber fatigue life prediction under the influence of temperature

“…Among them, the maximum error is 22.26%, which is within the acceptable range. For rubber material, the acceptable error range of the fatigue life prediction is within 100% [ 9 , 43 ]. Therefore, the fatigue life predicted by the prediction model is close to the experimental one.…”

Multiaxial fatigue prediction on crack initiation for rubber antivibration design – Location and orientation with stress ranges

“…the ratio between the predicted fatigue life in cycle over the measured fatigue life in cycle was two). 19,20 Other factors have also been investigated in relation to rubber fatigue, such as temperature, [21][22][23][24][25] straininduced crystallisation, 26 and the S-N (W€ ohler) curves for stress amplitude in multi-axial stress states. 27 In order to improve the accuracy of such fatigue predictions, the probabilistic methodology, 28 the artificial intelligence method, 29 and the finite element approach for interfacial fatigue damage in rubber composites have also been studied.…”