These slides were presented at the NAFEMS World Congress 2025, held in Salzburg, Austria from May 19–22, 2025.

Abstract

An interference fit is a common joining technology used to connect a shaft and a hub. In the case of dynamic loads, characteristic variables such as contact pressure or safety against slip are load and state dependent. Such effects could either not be investigated using common simulation methods or just in a highly simplified manner. The reason for this is the non-linear contact situation between shaft and hub, which made a finite element simulation with fine meshing and the consideration of all dynamic effects impossible. In this work so-called "contact modes" are applied to interference fits. Contact modes are trial vectors for describing the deformations within a contact area, leading to a dramatic reduction of the number of degrees of freedom, and thus the computing time, without any significant loss of accuracy. This method closes the former mentioned blank spot on the simulation map because they allow non-linear, accurate and fast numerical time integration of fine meshed finite element models of interference fits without simplifications regarding the dynamics. In the latest versions of ABAQUS (FEM) and SIMPACK (MBS), the automated computation of contact modes, their import into the multi-body simulation and the definition of the contact force elements have been implemented in a user-friendly way. The accuracy and efficiency of this new method are demonstrated by static and dynamic load cases. For the static load cases, the resulting stresses inside an interference fit, computed with the multi-body simulation software SIMPACK, are compared to those, computed with the nonlinear Finite Element code ABAQUS. With fine meshes and dynamic loads, a direct comparison of the MBS result with an FEM computation is no longer possible. For this reason, convergence analyses are presented for dynamic loading. The stresses within the contact area converge with an increasing number of contact modes. The CPU times are very short for all computations.