No abstract

Axial and torsional fatigue tests at different stress ratios were performed on a structural adhesive designed for wind turbine rotor blades. By employing previously optimized specimens, fatigue properties were recorded without influences of manufacturing-induced defects such as pores. The Stüssi S-N model was an excellent fit to the data and was combined with a Haibach extension line to account for uncertainties in the gigacycle fatigue regime. A comparison of the results with hand-mixed specimens revealed significant and load leveldependent differences, indicating that manufacturing safety factors should be applied to the slope of the S-N curve. The experiments were accompanied by stiffness degradation measurements, which enabled an analysis of Young's and shear modulus degradation interactions. The degradation was modeled using power law fits, which incorporated load level-dependent fitting parameters to allow for a full description of the stiffness reduction and a prediction of the residual fatigue life of run-out specimens.

No abstract

No abstract

Biaxial tension/compression–torsion fatigue tests with varying levels of non‐proportionality were performed employing a structural adhesive designed for wind turbine rotor blades. The cycles to failure were found to be independent of the level of non‐proportionality. It is demonstrated that numerical fatigue life predictions via rainflow‐counted equivalent stress histories are not able to replicate these experimental observations and overestimate the fatigue life up to a hundredfold. The tension–compression asymmetry of the adhesive resulted in significant damage prediction differences depending on the stress space representation of the Haigh diagram. If not properly taken care of, the asymmetry will also lead to non‐conservative results. While demonstrated with a short fiber‐reinforced adhesive, the results can be transferred to other materials.