A Comparison of Two Microslip Contact Models for Studying the Mechanics of Underplatform Dampers

Chao Xu,Dongwu Li,Muzio M. Gola,Chiara Gastaldi

DOI: https://doi.org/10.1115/gt2018-76007

2018-01-01

Abstract:In turbine blade systems, under-platform dampers are widely used to attenuate excessive resonant vibrations. Subjected to vibration excitation, the components with frictionally constrained interfaces can involve very complex contact kinematics induced by tangential and normal relative motions. To effectively calculate the dynamics of a blade-damper system, contact models which can accurately reproduce the interface normal and tangential motions are required. The large majority of works have been developed using macroslip friction models to model the friction damping at the contact interface. However, for those cases with small tangential displacement where high normal loads are applied, macroslip models are not enough to give accurate results. In this paper two recently published microslip models are compared, between them and against the simple macroslip spring-slider model. The aim is to find to which extent these models can accurately predict damper mechanics. One model is the so called GG array, where an array of macroslip elements is used. Each macroslip element of the GG array is assigned its own contact parameters and for each of them four parameters are needed: normal stiffness, tangential stiffness, normal gap and friction coefficient. The other one is a novel continuous microslip friction model. The model is based on a modification of the original classic IWAN model to couple normal and tangential contact loads. Like the GG array the model needs normal and tangential stiffness, and friction coefficient. Unlike the GG array the model is continuous and, instead of the normal gap required by the GG array, the Modified IWAN model needs a preload value. The two models are here applied to the study of the mechanics of a laboratory under-platform damper test rig. The results from the two models are compared and allow their difference, both for damper mechanics and for the complex-spring coefficients, to be assessed.