Prediction of Structural Rotor Loads: When to Consider Drivetrain Dynamics?

The constant, undisturbed rotor hub rotational speed is a commonly applied boundary condition and simplification in computational analyses of helicopter rotors. Revoking this simplification and considering rotor-drivetrain interactions in the hub's rotational degree of freedom can - but doesn't necessarily - improve the predictions of structural blade loads, especially in the lead-lag direction. To estimate the drivetrain's potential to influence the lead-lag loads, this paper proposes the systematic evaluation of the modified collective lead-lag modes. These eigenmodes, as well as the resulting modification of lead-lag loads in the aeromechanic simulation, are presented and compared for the rotordrivetrain configurations of the Eurocopter Bo105 and the Sikorsky UH-60A. The study focuses on understanding the drivetrain's influence rather than on making high fidelity predictions. In the Bo105 case, the drivetrain impact on the lead-lag moments is significantly more pronounced than for the UH-60A. The discussion of this difference includes the assessment of the blade passage frequency magnitudes (4/rev for a number of nb = 4 rotor blades) and their sensitivity to the modified collective second lead-lag eigenfrequency, which in turn is changed by a varying drivetrain stiffness. While a very stiff drivetrain causes an eigenfrequency of 4/rev with high-magnitude resonance for the Bo105, the UH-60A configuration maintains an eigenfrequency above 4/rev for the whole range of applied stiffness values. For rotor-drivetrain systems, especially near resonance, the nb/rev magnitudes of the lead-lag loads are very sensitive to changes in the drivetrain properties as well as changes in the nb/rev excitations. The accurate modeling of these nb/rev excitations, e. g. by airloads, is essential to capture changes in the nb/rev dynamic response that are caused by the drivetrain. Therefore, including the drivetrain in the structural model may only be useful if the fidelity of other models is increased simultaneously.