A. Korbel

An Experimental Investigation on the Fatigue Performance of Riveted Lap Joints

Results of an experimental research on the influence of several factors on the fatigue behaviour of simple riveted lap joint specimens under constant amplitude loading conditions are presented. The variables considered are the rivet type and material, sheet material and the squeeze force. Also, the effect of sheet thickness staggering in the overlap region on the joint fatigue life is studied in the context of secondary bending. The measurements of the driven head dimensions for a range of squeeze force levels and fatigue test results for rivets installed with various squeeze forces indicate that the squeeze stress rather than the rivet driven head dimensions is a safe standard for the quality of the rivet installation. The superior fatigue performance of a rivet with the compensator compared to the round head and universal rivet is noted. The underlying reason is the better hole filling achieved due to the presence of the compensator. Fatigue lives observed for the staggered thickness specimens are consistently longer than for the standard specimens. The analysis of these preliminary results suggests that a primary reason for the improved fatigue performance is the reduction of secondary bending.

An Experimental Investigation on Crack Initiation and Growth in Aircraft Fuselage Riveted Lap Joints

Effects of variables related to design and production of riveted lap joints representative of longitudinal sheet connections for a pressurized transport aircraft fuselage were experimentally investigated. The specimens from an aircraft Al alloy D16 Alclad sheets of three different thicknesses (1.9, 1.2 and 0.8 mm) were assembled under load control using round head rivets and rivets with the compensator from a P24 Al alloy. For the joints from 1.9 mm thick sheets fatigue tests indicated a dependency of the crack initiation site and crack path on the squeeze force level and on the rivet type. At the same time, increasing the squeeze force led to improved fatigue properties of the joints, specimens assembled using the rivets with the compensator showing fatigue lives consistently longer than joints with the round head rivets. All observed trends have been explained based on hole expansion and load transfer measurements. For thin sheets connected using the round head rivets, local deformations and indentations under the driven rivet head promoted crack initiation and failure in the adjacent sheet. Fatigue test results indicated that the detrimental effect of this type imperfections could outweigh the benefits associated with a decrease in secondary bending due to thinning the sheets. The rivets with the compensator were observed to cause significant local imperfections beneath the manufactured head, which adversely affected the joint fatigue performance.

Strip Yield Model Applicability to Crack Growth Predictions for Various Types of Fatigue Loading

The capability of the strip yield (SY) model to predict crack growth in structural steel is investigated. To this end the SY model implementation developed by the present authors is applied to simulate crack growth observed in S355J2 steel specimens under constant amplitude and simple variable amplitude loading. A particular attention is given to the calibration of the model using the constraint factors and examining whether tuning the model, based on constant amplitude loading, allows the adequate predictions for variable amplitude loading.

Applicability of Empirical Formulae for the Fatigue Notch Factor to Estimate Riveted Lap Joint Fatigue Strength

A semi-empirical fatigue life prediction model under development by the present authors for riveted lap joints used in aircraft structures is outlined. In contrast to existing models, it will account for the influence of the rivet squeeze force on the fatigue life of riveted joints. To determine the effect of rivet-hole interference on the fatigue behaviour of a riveted joint, a series of fatigue tests on filled hole coupons with different amounts of interference will be carried out under loading conditions representing the bypass load, transfer load and secondary bending. These experiments will allow evaluating of the dependency of the fatigue notch factors on rivet hole expansion. Preliminary results obtained so far are presented in this paper.