Authors: Eng. Potito Cordisco, Senior Project Manager, Vicoter
Eng. Mauro Terraneo, Chief Technical Officer, Vicoter
Drones are liable to high vibratory levels generated by the rotors and eventually amplified by the resonances of the structure. Slenderness of the arms supporting the motors, light weight of the structure, aerodynamic forces often generated by not perfectly balanced blades, can give birth to very complex and high-amplitude spectra. The situation is furthermore complicated by the payload that, with its weight comparable with the one of the drone, couples its dynamic behaviour with the hexacopter one.
ANT-X (www.antx.it) is an italian spin-off company of Politecnico di Milano providing drone solutions for education, research and industrial application. Its flagship is Aries, an innovative drone designed to accommodate different payloads according to the various tasks to perform. Peculiar characteristic of this drone is the non-conventional arrangement of the rotors with the axis not perpendicular to the central body but tilted with pre-defined angles that make Aries able exerting force component in the plane of the central body. This feature makes Aries unique compared to drones in conventional configuration, which are able to generate a force component only in the direction orthogonal to the body plane.
To investigate the dynamic behavior of its product and the difficulties concerning the coupling with different payloads, ANT-X calls Vicoter (www.vicoter.it), a society specialized in the sector of the vibrations from numerical and experimental point of view, to perform a series of tests.
The six double-bladed rotors rotate each at different speeds, producing a large broadband force spectrum potentially able to excite various resonances. To investigate each motor’s contribution to the whole vibratory level, a test campaign able to exploit all the various aspects of the research has been programmed and performed.
In particular, following tests have been carried out:
- Flight tests in controlled configuration to assess the magnitude of the accelerations during the various phases of the flight.
- Modal analysis without the payload to investigate the eventually excited resonances of the structure and furnish reliable data to correlate a future finite element model.
- Modal analysis with a payload to study the coupling of the structural modes with different equipment.
- Vibration measurement with only one motor on (six tests, one per motor), to measure relative contribution of each rotor.
- Measurement of the FRFs (Frequency Response Function) between each motor location and the Flight Control Unit (FCU) to assess the vibratory paths.
Sensor positions were selected to obtain information regarding the first, low frequencies, bending/torsional modes. Thirteen locations have been measured, ten on the aircraft and three on the payload. All the locations on the arms have been acquired along two mutual orthogonal directions. In total, twenty-seven channels were acquired simultaneously: twenty-six accelerometers during the flights, plus the hammer load cell during the modal tests. Special care was given to install accelerometers on the structure and on the FCU, to evaluate the transmissibility of the soft foam dampers used to reduce the vibrations read by the flight controller.
Speeds of the rotors has been derived by the log file of the flight control system using a trigger, due to the impossibility to duplicate such signal or to fit sufficiently small instruments.
Results of the experimental study clearly show that the main source of the vibration comes from the electrical motors and from the aerodynamic loads generated by the propellers. The balancing of the folding blades would reduce the 1st harmonics, strongly decreasing the whole vibratory level.
The main acceleration peaks during the flight are related to the orders of the loads and their frequencies are out of the band of the main modes during hovering.
In particular, the performed experimental tests allowed to understand the cause of a very annoying increment in the signal recorded by the gyro along the roll axis and not detected by the other sensors in the flight control. Such incremented response comes at a very specific frequency located in the band of the flight dynamics and so impossible to be filtered out without affecting the flight control performances. It has been demonstrated that this disturbance is produced by a coupling effect between the payload and the drone structure which generates a torsional mode with a translational node on the flight controller sensors location.
Investigations on the filtering capability of the flight controller dampers, located between the FCU and the drone frame, demonstrated that the cut-frequency is too high compared with the controller band, leaving the mechanical vibrations to be transmitted from the structure to the FCU. Methods to improve the transmissibility and its passive damping capability was suggested and will be implemented soon.
Vicoter warmly thanks Simone Panza and Mattia Giurato of ANT-X for their valuable support during the flight tests and the numerous hints concerning the flight control of a drone. A well done even at Lorenzo Ciuti, who discussed his master thesis in aerospace engineering on the correlation of a finite element model of the Aries drone based on the data acquired during this test campaign.