A Test Rig for the Experimental Investigation on the Nonlinear Dynamics in the Presence of Large Contact Interfaces and Numerical Models Validation

Abstract

The simulation of the coupling between components modeled by finite elements (FEs) plays an important role for the prediction of the forced response of the assembly in terms of resonant frequencies, vibration amplitudes, and damping. This is particularly critical when the time-varying stress distribution must be limited for vibrating components with thin thickness coupled with large contacts. Typical examples can be found in aeronautical structures (plates, panels, and bladed disk components) assembled with bolted flanges, riveted lap joints, or joints without hole discontinuities like rail-hook joints, lace wire sealings, and strip dampers. In this paper, a new test rig is introduced for the experimental validation of a reduced-order model (ROM) based on the Gram–Schmidt Interface (GSI) modes applied to a friction contact whose dimensions are not negligible with respect to the size of the substructures. In this case, classical approaches like Craig–Bampton technique might be not effective in reducing the size of the problem when many contact nodes subjected to nonlinear contact loads cannot be omitted. The technique is implemented in a solution scheme in the frequency domain using penalty contact elements and the harmonic balance method. The preload on the joint is produced by permanent magnets to enhance the friction contact without introducing uncertainties due to bolting. Measurements are compared with the ROM simulations and with standard time-domain integration of the full FE model. The advantage of using the GSI technique is shown in terms of time computation and accuracy of the simulation.