Experimental Bench Testing of an Active-Twist Rotor

Abstract

Active Twist Rotor Blades have been developed for the use in secondary control such as higher harmonic control (HHC) and individual blade control (IBC). The basic principle of such blades is the implementation of piezoelectric actuators into the blades, using different types of coupling, causing the blades to twist. At the DLR five model scale rotor blades have been manufactured to demonstrate the feasibility of such systems in a complete four bladed rotor, leaving one spare blade. The rotor is meant to be tested in a wind tunnel experiment within the STAR consortium. The reference for this blade design is the BO105 model rotor with a scaling factor of approximately 1:2.5. The C-spar and the glas skin are also used in the active blades, as well as the rectangular planform. The blades are equipped with 24 piezoceramic actuators in the blade skin each, 12 on the upper and 12 on the lower side respectively. The blades are equipped with different amounts of strain gauges and pressure sensors, which cause small differences between the blades. This paper presents results of the experimental testing of those DLR active twist blades. A comparison to simulations is also given. The experimental testing includes cantilevered bench testing of blade dynamics and actuation performance, the experimental analysis of its elastic axis, stiffnesses as well as geometric and mass properties for each individual blade. The methods to derive this data is presented as well. The data that is shown in this paper represents all needed data for the use as an input for rotor simulation with comprehensive codes. Deviations are discussed and compared with simulations. Especially the actuation of the blades is discussed and different measurements of the actuation are presented, always comparing all five different blades and discussion of their differences. A special focus is put on the nonlinearities of the twist actuation, showing experimental results of just one blade for tip twist with different DC voltage offsets as well as different amplitudes. Especially a dependency of the tip twist in accordance with the actuation frequency is demonstrated for very low frequencies (<1 Hz), where the dynamics of the blade should not have any influence on the blade behaviour. The results indicate phenomena, which are caused by the piezoceramic actuators themselves. These dependencies will make a difference for any quasi steady excitation – as it may be needed for active tracking for example. Finally the electric power used in one of the blades is analysed in more detail, showing dependencies of the frequencies and applied voltage amplitudes. Piezoceramic actuators are usually modelled in FE and analytical models in a linear simplification using d33 as one constant value regardless of the frequency, amplitude and DC offset of the signal. This paper will help to show the limits of such simplifications for active twist blades.