Research engineer, Fraunhofer-Institute fuer Betriebsfestigkeit (LBF), Darmstadt, West Germany
Pages: 30 Published: Jan 1986
An improvement of the accuracy of fatigue life prediction methods used for multiple-row riveted or bolted joints can be expected only if the rivet load distribution is considered during design and fatigue analysis. To calculate the individual load transfer and the bypassing load the fastener flexibility must be taken into account. Semiempirical formulas for the calculation of fastener flexibility existing in the literature turn out to be not exact or at least not applicable for a wider range of joint geometries. This unsatisfactory situation was the reason for performing an extensive experimental investigation during which fastener flexibilities for a wide range of joints of practical interest were determined.
The effects of primary joint parameters, such as the plate material, clamping length, and diameter and material of the fastener, as well as the effects of secondary parameters, such as the clamping force, condition of the faying surfaces, and fit and type of head of the fastener, were investigated through specific variations. For this purpose load-deformation measurements under quasi-static and flight-by-flight loading conditions were performed using single- and double-shear specimens with known load transfer.
A formula for fastener flexibility, valid for riveted and bolted metallic and graphite/epoxy joints, was derived from the test results and proved to be significantly superior to those found in the literature. This formula for load transfer calculations of multiple-row fatigue loaded joints was used to predict accurately the measured values of load transfer. This improvement also leads to a better fatigue life prediction. The load transfer measurements showed a quasi-linear relationship between the applied load and the total load transfer which was little affected by fatigue loading, although the load transferred by bearing was not constant. By measuring bearing loads and local strains close to fastener holes in a multiple-row joint the author shows that with increasing friction the bearing stress and the local strains decreased. The redistribution of loads or the changes in the mechanisms of load transfer result in a relief of the fatigue critical location. Since these changes of the local stress situation cannot be predicted, the application of fatigue life prediction methods on the basis of local strains must lead to inaccurate results.
fastener flexibility, load transfer, riveted joint, bolted joint, singleshear joint, double-shear joint, fatigue life estimation, Falstaff sequence, graphite/epoxy, aluminum alloy, deformation measurements
Paper ID: STP29062S