eVTOLUTION - Experimental dataset - Configuration A1 - VKI

eVTOL hybrid desIgn mUlti-fideliTy for low Noise and high aerodynamic performance HE Research and Innovation Action Grant Agreement No 101138209 Deliverable Reference Number: D6.2 Deliverable Title: Technical file dataset - Configuration A1 - VKI Andrea Bresciani, Aren Vardanyan, Julien Christophe, Christophe Schram 15 December 2025 Type R Document, report X DEM Demonstrator, pilot, prototype, etc. DEC Websites, patent fillings, videos, etc. ORDP Open Research Data Pilot Dissemination Level PU Public X SEN Sensitive, only for members of the consortium (incl. Commission Services) 1 Introduction This document contains the description of the test campaign concerning configuration A1, i.e. single isolated propeller in hovering conditions. The objective is to create an experimental database to validate simulation data. 2 Facility and test rig All experiments are conducted in the ALCOVES anechoic chamber at VKI, which meets ISO:3745 standards, demonstrating free-field behaviour down to 150 Hz for both broadband and tonal noise [1,2]. The facility, shown in fig. 1a, consists of two rooms, a main test room and a discharge room. The main room hosts the test rig, and it is equipped with 3 linear microphone arrays with a total of 24 G.R.A.S. 40PL condenser microphones. The room is equipped with an external fan that was not operated during this test campaign. A schematic of the room and the microphone layout is shown in Fig. 1. The 24 G.R.A.S. type 40PL microphones were arranged in three linear horizontal arrays of 8 microphones (fig. 1b), mounted on a motorized vertical traversing system spanning from −0.4 m to 0.4 m with respect to the height of the propeller axis. The microphones are located between 4 and 5.5 radii from the hub (0.65 m and 0.89 m). A Br¨uel & Kjær Type 4231 pistonphone producing a 94 dB, 1 kHz tone was used to calibrate the microphones. Following individual calibration, the amplitude accuracy is within ±1 dB from 100 Hz to 5 kHz and ±2 dB from 10 Hz to 20 kHz. The nets used to reduce the flow recirculation in the room are shown in fig. 2. Three nets, each consisting of 10 layers of mesh with a size of 1.5 cm, were installed downstream of the propeller. The nets were positioned at approximately 0.5D, 2D, and 4.3D. The assembly of the test rig is shown in Fig. 3, and an exploded view is given in Fig. 4. The test rig consists of an AT4130 300 KV brushless DC motor, with a peak power of 3200 W. The motor has a KV of 300, which yields a maximum RPM of 13,320 RPM, at no-load conditions. To deliver this power to the motor, the Mean Well RSP-3000-48 power supply is used. This is a 3 kW power supply with a DC output of 48V and a rated current of 62.5 A. Alternatively, a 5 Ah 12S LiPo battery is also available for power delivery. The power is supplied to the Scorpion Tribunus II 12-80A ESC. Two encoders are included in the test rig. The first encoder is the AMT10 incremental encoder. It is used to measure the RPM of the motor by detecting changes in the encoder signal as the motor shaft rotates. The encoder can track rotational speeds up to 15,000 RPM, making it suitable for this motor. The second encoder is the AMT22 absolute encoder. This is a 14-bit encoder that can measure the absolute position or phase of the propeller. Two TCRT5000 phototransistors are mounted adjacent to the motor. These sensors generate a pulse with each full rotation, but only when facing a reflective surface. To use this functionality, reflective tape is applied to the motor. The generated pulse is used to trigger the PIV laser, allowing synchronization between the PIV system and the motor’s rotation. The ATI Mini45 load cell is used to measure the loads in the test setup. This is a multi-axis load cell capable of measuring forces and torques along all six axes. Its small diameter and high accuracy make this a perfect fit for the setup. The sensor has a measurement range of 580 N in the Fz direction, 290 N in Fx and Fy, and 10 Nm in Tx, Ty, and Tz. Lastly, the test rig is encapsulated in a 3D printed nacelle, which limits flow and acoustic interactions between the propeller and the test rig components. A schematic of the control system architecture of the test rig is shown in Fig. 5. The test rig is controlled using a combination of NI’s cDAQ-9174 chassis and an Arduino UNO microcontroller. The first module of the cDAQ is an analog input (AI) module with a ±10 V input range, with a maximum sampling rate of 250 kHz. Six of these channels are used to acquire data from the load cell, while two additional channels measure the power supply’s output voltage and current. Since the power supply exceeds the ±10 V input limit, a voltage divider circuit is implemented to scale down the voltage for accurate measurement. The second module is the NI 9211, which is used for thermocouple data acquisition, with a sample rate of 14 Hz. The third module is the NI 9401, which is a digital input/output module (DIO). It is used to send the PWM signal to the motor via digital output (DO) and read the phototransistor signal via digital input (DI). Lastly, the encoder signals are acquired on the microcontroller. The RPM and phase calculations are done in Arduino and are sent to LabVIEW. Within LabVIEW, all sensor data 2 / 7 (a) (b) Figure 1: (a) ALCOVES anechoic chamber, (b) microphones layout, and (c) installed propeller in ALCOVES. 3 / 7 (a) (b) (c) Figure 2: Nets used to reduce the flow recirculation in ALCOVES. (a) 1 layer, (b) 10 layers, and (c) nets installed in ALCOVES. 4 / 7 AP BO CN DM EL FK H I G J F K E L D M C N B O A P 54 5 4 63 6 3 72 7 2 81 8 1 A-A BLDC Motor AT4130 KV300 5-Bladed forward propeller Multi-axis loadcell ATI Mini 45 Phototransistor 3D printed support Nacelle Absolute encoder AMT 22 Incremental encoder AMT 10 Figure 3: Assembly view of the test rig. is collected, and a PID controller is implemented to regulate the motor’s speed. 3 Data processing The loads have been acquired by means of LabView software over a time acquisition of 60 s at 10 Hz and averaged. The temperature of the load cell was allowed to stabilize between 26 and 32◦C, depending on the RPM, before the acquisition process started. After each acquisition, the propeller was stopped, and the zeros were acquired for 1 s, compensating for the temperature drift. The acoustic data acquisition system consisted of a cDAQ-9178 chassis equipped with six NI-9234 modules, to allow the simultaneous acquisition of the 24 channels at a sampling frequency of 51.2 kHz for 30 seconds. The signals were then bandpass-filtered between 10 Hz and 20 kHz using a 5th-order Butterworth filter. The power spectral density was computed with Welch’s method, using a window size of 214 points, with 50% overlap, and a Hanning window. The acoustic data have been acquired at a fixed RPM of 8300 for the 5-bladed propeller and 8900 for the 4-bladed one. References [1] Erica Gallo, Julien De Decker, Andrea Bresciani, Pauline Haezebrouck, Emanuele Garone, and Christophe Schram. Development and commissioning of an aeroacoustic test bench for the investigation of single and coaxial propeller noise. Acta Acust., 9:16, 2025. [2] M Bilka, J Anthoine, and C Schram. Design and evaluation of an aeroacoustic wind tunnel for measurement of axial flow fans. The Journal of the Acoustical Society of America, 130(6):3788–3796, 2011. 5 / 7 Figure 4: Exploded view of the test rig. 6 / 7 Figure 5: Control system architecture. 7 / 7