10:30
Session 9: Experimental Aeroacoustic Measurements – Laboratory 1
Chair: John Kennedy
10:30
20 mins
|
Understanding and Controlling Laminar Separation Bubbles for Quieter Drone Propellers
SIMON WATKINS
Abstract: Small multi rotor drones and emerging Advanced Air Mobility vehicles rely on propellers that operate at low Reynolds numbers, where laminar boundary layer separation and the formation of laminar separation bubbles are common. These flow phenomena reduce aerodynamic efficiency and contribute significantly to increased broadband and tonal noise, limiting endurance, payload capability, and community acceptance in noise sensitive environments. Despite the importance of this regime, experimental data on low Reynolds number propellers and effective flow control strategies remains scarce.
This work presents an experimental investigation into the aerodynamic and acoustic performance of a propeller operating at low Reynolds numbers, with a focus on boundary layer augmentation through surface tripping. The primary research questions address how trip type and placement influence blade surface flow behaviour, propeller efficiency, and noise generation, and whether spanwise specific tripping can offer further performance benefits. A 9×6 propeller operating at a tip Reynolds number range of 4.3×10⁴ to 4.38×10⁴ was tested in a wind tunnel over wind tunnel speeds from 0 to 40 km/h. Aerodynamic forces and moments were measured using a calibrated force and moment balance, while acoustic measurements were obtained using a Brüel and Kjær sound level meter. Surface oil flow visualisation was employed to directly observe boundary layer behaviour, separation, reattachment, and vortex shedding associated with laminar separation bubbles. Tests were conducted in both static conditions and in a controlled inflow corresponding to a max of 40 km/h.
The results show that the chordwise location of boundary layer tripping plays a critical role in performance enhancement. When trips are positioned upstream of the natural laminar separation location, propulsive efficiency improves due to increased thrust generation with nominally similar torque levels. Oil flow visualisations confirm that appropriate tripping suppresses or weakens laminar separation bubbles and associated unsteady vortex shedding. Spanwise variations in Reynolds number were found to produce distinct flow responses along the blade, motivating the investigation of spanwise specific tripping strategies. This approach demonstrates potential efficiency gains beyond those achieved using uniform tripping methods (trips all the way to the tip of the propeller).
Overall, this study provides new experimental insight into low Reynolds number propeller aerodynamics and acoustics, introduces the concept of spanwise specific boundary layer tripping, and contributes data that can inform quieter and more efficient propeller designs for drones and future AAM applications.
|
10:50
20 mins
|
Effect of Large scale Turbulence on Propeller Aerodynamic and Aeroacoustic performance
Gokhul Venkataraman Swaminathan, Lourenco Tercio Lima Pereira, Yannick G.A. Chance
Abstract: The noise generated by UAM in urban environments is a key factor in their certification. Urban flight is expected to subject aerial vehicles to a broad spectrum of large-scale turbulent structures. To investigate this problem, an experimental work is conducted in Delft University’s anechoic tunnel, where large-scale turbulence in the order of the rotor diameter is generated using bluff-body shedding, and its influence on propeller loading and noise emission is analyzed. It is observed that the highly turbulent inflow has a severe impact on the noise emissions, with an observed increase of up to 10-20 dB in the broadband content and retention of discrete tones emitted at BPF harmonics across the higher advance ratios. Furthermore, with increased rotor-scale turbulence in the inflow, haystacking is observed in the noise emissions, which dominates the high-frequency noise levels, yielding predominantly broadband content above the 2nd BPF.
|
11:10
20 mins
|
Rotor Speed Estimation from Onboard Audio Recordings of Multirotor Drones
Dmitrii Mukhutdinov, Lin Wang
Abstract: The acoustic signature of a multirotor unmanned aerial vehicle (UAV) is dominated by rotor harmonics whose fundamentals are set by the instantaneous rotational speeds (RPS) of individual rotors. Recovering per-rotor RPS directly from audio captured by onboard microphones would enable passive acoustic monitoring of the drone state without requiring access to onboard telemetry, with potential applications in flight-mode and manoeuvre inference, payload and fault diagnostics, model-based source separation, and conditioning-signal generation for downstream speech or event-detection systems. We conduct a pilot study of audio-only RPS estimation in the practically relevant regime where the rotor signal is mixed with non-rotor sound. We propose a lightweight convolutional regressor that takes the log-magnitude short-time Fourier transform (STFT) of the mixture as input and outputs the four per-rotor speeds at every STFT frame. The network is trained by frame-wise regression against telemetry-aligned ground truth. Experiments on synthetic mixtures and free-flight recordings demonstrate that the proposed model can reliably recover per-rotor RPS across both low- and high-SNR conditions.
|
|