Uncertainty in identifying contact stiffness in a dovetail attachment for turbine blades

Designing bladed disks in turbo engines requires accurate dynamic models to correctly estimate resonance frequencies and related stresses. The contact parameters – stiffness and damping in the blade attachments – are currently among the most significant uncertainties of such models. Dry friction and alternating relative motions between the contact interfaces determine the transition between stick and slip that causes the nonlinear behavior of attachments. Commercial and in-house finite element software make use of specific contact elements to simulate dry friction and the resulting nonlinear behavior. These elements require the friction coefficient as input while the normal and tangential contact stiffness can be directly evaluated by the software or set up by the user. The main objective of the present research is to discuss the uncertainty associated with identifying with experimental data the normal and tangential stiffness in a dovetail coupling. In addition, the reliability of available theoretical contact models will be addressed. The response of the blade/attachment system was measured as a function of the axial load (simulating the centrifugal force) and of the blade vibration amplitude. An identification procedure using a finite element model was set up to identify the normal and tangential contact stiffness of the attachment. These stiffnesses were compared with the values predicted by a theoretical model. The comparison highlights a significant difference between measured and predicted stiffness. The final discussion focuses on the significance of theoretical contact stiffness and its use in finite element models.