Composites of polyamide matrix and short glass fibers are increasingly used in the automotive industry. They are usually subjected to fatigue loadings during their service life. Short glass fiber reinforced (SGFR) thermoplastics are used in structural parts such as intake manifolds and engine mounts wherein they are subjected to complex positive and negative loading conditions. As they are injection-molded, SGFR thermoplastics show a complex fiber orientation distribution which clearly has an effect on their mechanical properties. In addition, the polyamide matrix in SGFR makes it sensitive to humidity and temperature of the atmosphere. Therefore a fatigue criterion proposed for this material should take care to unify different load ratios and fiber orientation distribution and account for atmospheric conditions. In literature, a wide variety of energy-based fatigue criteria has been proposed; nevertheless, single unified fatigue criterion has space for improvement. This article proposes a fatigue criterion combining several indicators to predict the lifetime of the sample with precision close to one decade.

PA66-GF50 (Polyamide 66 reinforced with 50wt.% of glass fibers) composite dog-bone samples are used to generate a classical fatigue test database for various load ratios (-0.5, 0.2, 0.1, 0.3, 0.5, 0.7) and different orientations (0°, 45°, 90°) from the injection direction. The atmospheric conditions are maintained at humidity ratio 50% and temperature 80°C during the fatigue tests to simulate the conditions experienced by composite materials in engine mounts. The testing apparatus includes optical and infrared cameras to monitor strain fields using digital image correlation and self-heating using thermal fields, respectively. This through experimental campaign was achieved by the Solvay Lab. It is verified that, the number of cycles to failure is correctly predicted by the proposed criterion for various load radios and loading directions. A discussion will also be provided on the capability of several fatigue criteria to unify this wised database.