Propeller-Driven Rotor Twirl Phenomenon: Hover Performance, Dynamics, and Controls

Abstract

Propeller-driven rotors use compact electric motors and propellers, mounted on the main rotor blade, to generate the mechanical torque needed to rotate the main rotor. The presented work details experiments and model developments in three major areas: hover performance, dynamics, and controls. The idea of a propeller-driven rotor has become attractive over the last several years due to the advances in electrification of aircraft, for several reasons. First, large mechanical transmissions and gears associated with single main rotor helicopter designs are replaced with efficient electric motors, simplifying the rotorcraft design, maintenance, and potentially reducing weight. Second, modulating the propeller speed can generate the control moments for the rotor, eliminating the need for a conventional swashplate. Third, an unmanned design can eliminate a conventional fixed frame fuselage and tail rotor -- significantly reducing vehicle complexity and weight. Fourth, given these first three advantages, the concept is essentially a “flying rotor design”. Propeller-driven rotor performance and feasibility was investigated in a series of experiments with increasing complexity. First, high quality commercial-off-the-shelf electric motors and propellers were selected, and were were characterized in axial flight in the Low Turbulence Wind Tunnel at the Naval Surface Warfare Center Carderock Division. Next, hover experiments using a 5.5 ft diameter 2-bladed composite rotor blade on a custom hover test stand identified performance trends and areas for improvement, such as reduction in power. During these experiments, several propellers suffered root failures. The next hover experiment focused on measuring propeller loads during propeller-driven rotor operation. The results indicated that a large, previously unmodeled, propeller pitching moment was present during propeller-driven rotor operation. Finally, propeller placement on the rotor blade was investigated. A 2.7 ft diameter test stand was built for this experiment, and showed that a propeller placed at the rotor blade mid-span location resulted in a 50% reduction in the electrical power coefficient compared to a tip mounted propeller. While attempting additional propeller-driven rotor hover experiments, several propeller blades suffered root bending failures. This halted the hover test campaign, and prompted an investigation into the dynamics of spinning a fast propeller on a rotating rotor blade. The resulting failure has since been termed as the “propeller-driven rotor twirl phenomenon.” A rotor-propeller Coriolis interaction results in a large propeller pitching moment that bends the propeller. Additionally, this propeller pitching moment results in a rotor blade torsional load. The coupling of the propeller pitching moment with the rotor blade torsion mode is the suspected cause of the propeller failures during propeller-driven rotor operation. The propeller-driven rotor dynamics were investigated in two vacuum chamber experiments, as vacuum chambers remove aerodynamic effects and allow for only rotor-propeller inertial interactions to be studied. The first vacuum chamber experiment measured propeller flapping motion on an articulated propeller. The results indicated that a large, one per propeller rev, propeller flapping motion was present. A simplified physics model was derived, and showed the flapping was a result of the rotor-propeller Coriolis interaction. The second vacuum chamber experiment investigated the rotor and propeller loads, and showed that the propeller pitching moment resulted in a rotor blade torsional moment. The findings of the various hover and vacuum chamber experiments highlighted the need to study the propeller-driven rotor concept using modern comprehensive rotorcraft analysis tools. A finite-element based propeller-driven rotor comprehensive model was created, and validated using the various datasets. Special attention was given to ensure proper modeling of the rotor-propeller Coriolis moment. The rotorcraft comprehensive analysis model showed good agreement with the test data. After validation, a systematic study of various propeller-driven rotor configurations was performed, such as: impact of propeller placement on performance, propeller loads during propeller-driven rotor operation, and propeller-driven rotor scaling. The final section of this thesis explored a swashplateless control concept for a propeller-driven rotor. The propeller was mounted at a slight upward angle with respect to the rotor blade chord, which resulted in a portion of the propeller thrust to be in the vertical direction. The vertical portion of propeller thrust was controlled by varying the propeller speed around the rotor azimuth, resulting in a main rotor roll and pitch moment. The control concept was fabricated, characterized, and compared to a conventional rotor swashplate. The results indicated that, for certain propeller configurations, a 2.3% variation in propeller motor throttle resulted in the same control authority as 1 deg of swashplate excitation. Additionally, the propeller-driven rotor control scheme had excellent on-axis response, and little off-axis response. This exploratory study observed several areas of future work. First, the suitable propeller design for propeller-driven rotor is still an open question. The propeller must have two properties: it has high propulsive efficiency, and it can withstand the significant inertial loads. Vacuum chamber and hover experiments are expected. The second area is forward flight investigations. Edgewise rotor flight brings about harmonic, or time varying, airflows on the propeller that are not seen in hover. Forward flight stability investigations are needed to identify new technical challenges related to propeller-driven rotors.