ASME TCVS Seminar on Dynamics of Bolted Interfaces, May 3, 2024
One of the greatest challenges in structural dynamics is to predict the stiffness, damping and nonlinearity in a structure caused by the interfaces between its parts. Over the past few decades, finite element modeling of structures has improved dramatically, to the point that we can create a predictive model of a structure, but this is only true if the structure is constructed from one solid piece of a well characterized material. Realistic structures are assembled from many parts and uncertainties in the damping and stiffness of the joints are the major source of modeling errors. Furthermore, joints are the most common source of nonlinearity in civil, automotive and aerospace structures. Current methods are not adequate to predict joint behavior, so industry relies on testing to determine the damping in the built up system, validate models for the stiffnesses of the joints and to detect any nonlinearities. This can only be done once the first prototype has been built, at which point design changes may be very expensive.
This talk first highlights methods that can be used to test these types of structures to understand how the joints influence the effective natural frequencies and damping ratios of the structure, and to identify cases when more highly nonlinear models are needed. We then review a new method which uses quasi-static analysis to efficiently predict the dynamic behavior when a structure vibrates in one of its modes. This method has enabled some of the first predictive simulations of the damping and nonlinearity of simple structures with bolted joints. Predictions from a detailed finite element model, with Coulomb friction describing the contact at the interface, are compared to experimental measurements to evaluate the performance of the method and highlight areas where further research is needed.
by Matt Allen, Professor, BYU. Given to the ASME TCVS on May 3, 2024