Experimental noise characterisation of different pitch propellers in forward flight Nur Syafiqah Jamaluddin a,1 , Alper Celik a,b,2 , Kabilan Baskaran a,c,3 , Djamel Rezgui a,4 , Mahdi Azarpeyvand a,5

a University of Bristol, Department of Aerospace Engineering, Bristol, UK, BS8 1TR b Swansea University, Swansea, UK, SA1 8EN c Indian Institute of Technology, Department of Mechanical Engineering, Dhanbad 826004, India ABSTRACT The present study investigates the influence of blade pitch on propellers' aerodynamic load and noise characteristics during forward flight operations. A series of experiments have been carried out on an isolated propeller configuration in the aeroacoustics facility at the University of Bristol. Thrust, torque and far-field noise are measured at varying inflow speeds. Acoustic spectral and directivity analyses are performed to understand the distribution of acoustic radiations with frequency and emission angles, respectively. The results are presented in propeller loading variation, power spec- tral density of acoustic pressures, and overall sound pressure level directivity. The results reveal that under matching operating conditions, the acoustic spectral and directivity of lower-pitched propel- lers are strongly characterised by the high-frequency broadband noise contributions. In contrast, higher-pitched propellers are significantly influenced by the low-frequency tonal noise contributions for both spectrum and directivity features. 1. INTRODUCTION

The use of propellers in the new generation of urban air vehicles offers many potential benefits. However, noise emission is one of the significant challenges for a propeller-powered system in gain- ing public and regulatory acceptance. Propellers or rotors are one of the vital components used to power Urban Air Mobility (UAM) technologies [1]. The tonal and broadband noise components, ubiquitously generated by rotating blade operations, can potentially cause noise pollution and sub- stantially impact human health. Therefore, understanding the aeroacoustics of the propeller, particu- larly in terms of noise radiation, is imperative for developing the most pragmatic designs for this urban aerial vehicle. Knowledge of the correlations between propeller's aerodynamics and aeroacous- tics is also essential to help build efficient UAM vehicles [2]. A propeller is a device that generates propulsion by using the lift and drag force generated by the propeller blades [3]. The pressure distri- bution around the propeller blades determines the lift and drag and creates acoustic pressure variation, which produces sound as it radiates into the far-field. Propeller noise is classified into two types: harmonic tonal noise and broadband noise [4] [5]. The fluid volume displacement and steady loading distribution on the rotating propeller contribute to the generation of tonal noise components.

1 syafiqah.jamaluddin@bristol.ac.uk 2 alper.celik@swansea.ac.uk 3 kabilan@iitism.ac.in 4 djamel.rezgui@bristol.ac.uk 5 m.azarpeyvand@bristol.ac.uk

On the other hand, circumstantially non-uniform loading on the propeller, and non-linear turbu- lence interactions contribute to the generation of broadband sound components that also radiated to the far-field [6]. Previous studies in the literature show that propeller sizing has a pronounced effect on propeller efficiency and performance [7]. Moreover, the geometrical properties such as chord twist and blade taper ratio significantly affect the aerodynamic and aeroacoustics characteristics [8]. This paper presents the experimental results on the noise and loading characteristics of different pitched propellers operating in forward flight configuration. A similar experimental setup to this study has partly been published in Jamaluddin et al. (2021) [9]. This paper is organised as follows: Section 2 explains the aeroacoustics facility, experimental rig and the techniques used; Section 3 provides the results and discussions from the experiments conducted; and finally, Section 4 reports the conclusion for this manuscript.

2. EXPERIMENTAL METHODOLOGY

2.1. Experimental Facility

Experiments were conducted at the University of Bristol's closed-circuit open-jet anechoic wind tunnel facility. The wind tunnel has a less than 0.2% turbulence intensity and a flow uniformity of more than 90% at the exit of the nozzle. The anechoic chamber is 7.9 m long, 5.0 m wide, and 4.6 m high, with acoustically treated interiors that achieve approximately 160 Hz of the cut-off frequency. The contraction nozzle exit is 500 mm in width and 775 mm in height, allowing a steady operation from 5 to 40 m/s. For a more detailed description of the facility, the reader is referred to Mayer et al. (2018) [10]. Far-field noise measurements were taken from 23-units of ¼ inch-diameter GRAS Sound and Vibration - model 40PL microphones, which have an upper limit of 142 dB and cover a frequency range between 10 to 20,000 Hz. These microphones are installed in a far-field array arch, allowing measurement of polar angles between 40° and 150° at a radial distance of 1.75 m from the centre of the propeller system.

2.2. Test Rig and Setup

The isolated propeller rig used in the present study was positioned in the middle of the open jet nozzle, centred approximately 0.6 m away from the nozzle exit. The APC 10"x10" and APC 10"X5" constant-pitch type two-bladed propellers were tested. The propellers measure 10-inches in diameter and have geometrical pitches of 10 and 5 inches, corresponding to pitch-to-diameter ratios of 1 and 0.5, respectively. A pitch ratio is a dimensionless number determined by the propeller pitch (P) to diameter (D) ratio. Figure 1 illustrates the experimental setup used in the present work.

A T-Motor Antigravity MN4006 brushless motor with a diameter of 44.35 mm and a maximum output of 420 W was used to operate each propeller. The motor speed was controlled by Robotbird 100A pro electronic speed controller. The setup was powered by a DC bench power supply, regulated up to a maximum of 25 V. Electrical current was measured at the power supply as the throttle setting of ESC was varied to change the speed of the motor. The rotation speed of the propeller was obtained by detecting the electrical pulse signal from one of the three wires of the brushless DC motor, taking into account the motor's 24 poles. The propeller speed was also measured using a digital optical laser tachometer DT-2234C+.

Figure 1: Schematic view of the experimental setup with the details of the test rig configuration, the

wind tunnel nozzle and far-field microphone array.

2.3. Data Acquisition and Processing A National Instruments PXIe-1082 data acquisition system was used to collect the far-field noise data. The interface between the data acquisition system and the microphones was programmed using Matlab R2016a. Far-field noise measurements were collected for 16 seconds at a sampling rate of 2 16 Hz. An ATI Mini40E 6-axis load cell was used to acquire aerodynamic loading data (i.e. thrust and torque). The measurements lasted 16 seconds and were taken at a sample rate of 2 14 Hz. Both noise and loading measurements were performed at various propeller rotation speeds ( n ) and freestream inflow velocities ( U ∞ ), corresponding to advance ratios between 0.47 to 0.66. The advance ratio is calculated as 𝐽= U ∞ 𝑛 𝐷 Τ , where D is the propeller diameter and n is the rotational speed in revolutions per second.

3. RESULT AND DISCUSSION

The aerodynamic performance of isolated propeller with different blade pitch ratio are presented in this section. The acoustic spectral and directivity characteristics of the propellers are also investi- gated and discussed. For brevity, the APC 10 " x10 " propeller (P/D=1) will be referred to as propeller A and the APC 10 " x5 " (P/D=0.5) as propeller B.

3.1. Aerodynamic Performance

The thrust (T) and torque (Q) forces generated at a fixed rotation speed are measured to ensure a constant tip Mach number in the investigation. Results are presented in non-dimensional terms of coefficient of thrust, power and efficiency, which are calculated using Equations 1, 2 and 3:

𝐶 𝑇 = 𝑇 𝜌𝑛 2 𝐷 4 (1)

𝐶 𝑃 = 2𝜋nQ

𝜌𝑛 3 𝐷 5 (2)

𝜂= 𝐽 𝐶 𝑇

, (3)

𝐶 𝑃

where 𝑛 is the propeller’s rotational speed in rev/s, and 𝜌 is air density.

Figure 2 presents the aerodynamic coefficients of the tested blade, which are compared against published data from Brandt et al. (2011). The experimental results are shown in error bars, consider- ing the measurement uncertainty of a 95% confidence level based on the calibrated load cell used. It is worth mentioning that the obtained findings sufficiently corroborate with the published reference values [11], as illustrated by the solid lines in Figure 2. Propeller A, which has a larger blade pitch ratio, clearly demonstrates greater thrust, power, and efficiency coefficients at the same operational setting as propeller B. The thrust and torque forces are the derivatives of the blade’s lift and drag, while torque is directly proportional to power. Therefore, an increase in blade pitch results in in- creased aerodynamic forces. In terms of performance efficiency, propeller A has a broader range of efficient operating zone, extending from J=0.4 to 0.8, when tested at a fixed rotational setting. On the other hand, propeller B's efficiency peaks at around J=0.5, that is, η=0.61, beyond which any further increase in inflow speed results in blade stall.

Figure 2: Comparisons of the aerodynamics performance data from the present study (solid circles

with error bars) with the published reference source (solid lines). Figure 3 shows the radial distribution of the estimated angle of incidence (α), from the blade root (r/R=0) to the blade tip (r/R=1), at varying advance ratio operations. The angle of incidence was calculated ignoring the contribution of the induced velocity, which is relatively lower than the incom- ing flow speed. The following equation was used to calculate α:

𝛼= 𝑡𝑎𝑛 −1 𝑃

2𝜋𝑅 − 𝑡𝑎𝑛 −1 𝑈 ∞

2𝜋𝑛𝑅 . (4)

Both propellers show the highest incident angle at around 0.2 R, near the blade root. It is important to note that propeller B, with a smaller pitch ratio, has a significantly lower incident angle distribution across the blade locations than propeller A. Furthermore, propeller B depicts negative ranges of α for advance ratio operations higher than 0.47. It is important to note that the blade is still able to produce positive thrust in the operation setting of interest because it has cambered aerofoil.

Figure 3: Geometric distribution of incidence angle (α) at varying advance ratio operations

for propellers A and B

3.2. Aeroacoustics Performance

This section presents the far-field noise characteristics of propellers A and B, which are investi- gated using power spectral and directivity analyses. The former analysis provides information on the frequency-dependent energy content of the far-field noise spectrum. Meanwhile, the latter provides insights into the directivity trends of the radiated acoustic energy in the far-field. This work demon- strates the noise directivity characteristics at several polar observation points above the propeller. The magnitude of the radiated noise is expressed in terms of overall sound pressure level (OASPL), which is derived by integrating the energy spectrum with respect to frequency using Equation 4, that is,

𝑂𝐴𝑆𝑃𝐿= 10 𝑙𝑜𝑔 10 ൤ ׬ 𝑃𝑆𝐷 ሺ 𝑓 ሻ 𝑑𝑓

𝑝 2 𝑟𝑒𝑓 ൨ , (5)

where 𝑓 is the resolved range of frequency, 𝑝 𝑟𝑒𝑓 is the conventional reference pressure of 20 μPa, and PSD is the power spectral density derived using the Welch method based on the unsteady pressure [12]. Figure 4 compares the OASPL directivity features from 100 to 25,000 Hz at advance ratio op- erations of 0.47, 0.57 and 0.66 for propellers A and B. It should be noted that this range of advance ratio corresponds to a favourable performance zone for propeller A, as can be seen by the efficiency plot in Figure 2.

Meanwhile, propeller B operates at a reasonably reduced performance at advance ratio higher than 0.57. The results show a concentric ring pattern for propeller A that is highly directional near the plane of rotation (θ=90°). Meanwhile, propeller B exhibits a dipolar radiation pattern, with minimum radiation observed just upstream of the plane of rotation (θ=80°). Another interesting observation is at J=0.47 and 0.57, where propeller B was operating at its peak and critical efficiency, unlike propeller A, which had yet to reach its peak performance. Regardless, noise radiation was comparable at the

upstream and downstream in both cases. This trend may be attributed to the significant flow separa- tion on the blade surface near the tip of propeller B, which will be further discussed in the following discussions.

Figure 4: Directivity of overall sound pressure level for advance ratio operation of a) 0.47, b) 0.57,

and c) 0.66. Figure 5 presents the acoustic spectra of propellers A and B obtained at a polar observation loca- tion of θ = 90° for 5000 rpm and J=0.47 operating conditions. The results characterise the frequency- energy content of the propellers' acoustic spectra for several frequency domains, including the total frequency, broadband frequency, and blade pass frequency (BPF). The acoustic energy content in the spectra are presented as PSD, which are calculated using Equation 5, that is,

𝑃𝑆𝐷= 10 𝑙𝑜𝑔 10 Φ 𝑝𝑝 𝑝 2

𝑟𝑒𝑓 ൗ , (6)

where Φ 𝑝𝑝 is the power spectrum of acoustic pressure calculated using Welch method.

Figure 5: The propeller’s acoustic spectral content resolved at different frequency domains: a) total

noise, b) broadband noise, and c) BPF noise for operation settings of n=5000 rpm and J=0.47. The tonal noise spectra show that propeller A produces significantly higher PSD levels in the overall frequency domain, especially at BPF (f=167Hz) and its harmonics, compared to propeller B. The magnified BPF spectra show an apparent decrease in tonal peak amplitude of propeller B at f/BPF=1

relative to propeller A by approximately 6 dB/Hz. The broadband noise spectra in Figure 5 show the extracted mid-and high-frequency broadband signal without the sharp tonal peaks, which were re- moved using the median filtering technique. A significant broadband hump is observed for propeller B relative to A, which could be due to the smaller blade's incidence angle that reduces the blade loading. This geometric limitation of propeller B may lead to laminar separation bubbles on the trail- ing edge and consequently contributes to the associated high-frequency vortex shedding noise [13].

The comparisons of noise directivities for different frequency range is presented in Figure 6 to emphasise the effect of blade pitch ratio on the characteristics of noise radiation patterns. The total noise OASPLs for propeller operating at 5000 rpm and J=0.47 are calculated with a resolved fre- quency ranging from 100 to 25,000 Hz. Meanwhile, the frequency range for the narrowband BPF and high-frequency broadband OASPLs are resolved from 150 to 180 Hz, and from 7,000 to 25,000 Hz, respectively. Propeller A has significantly higher BPF noise levels, while propeller B exhibits higher broadband noise levels across the polar noise emission angle. The total noise radiation pattern of propeller A resembles the concentric ring directivity pattern of its BPF noise, which is highly direc- tional in the plane of rotation. In contrast, the total noise directivity of propeller B is comparable to its broadband noise radiation pattern, which is dipolar and has its axis aligned just upstream from the plane of rotation. This trend suggests that at J=0.47, the high-frequency broadband noise generated by propeller B significantly influenced the overall noise directivity when radiated into the far-field, as opposed to propeller A.

Figure 6: OASPL directivity comparison with polar emission angles presented as a) total noise, b)

broadband noise, and c) BPF noise for propeller operation settings of n=5000 rpm and J=0.47.

4. CONCLUSIONS

The effects of blade pitch ratio on the noise characteristics of propellers in forward flight operation were examined. Two propellers with identical diameters but different blade pitches were tested in an isolated configuration. The experiments are performed under matching operating conditions. The higher-pitched propeller (propeller A) operates at a more favourable efficiency zone than the lower- pitched propeller (propeller B) due to the geometric limitation in the blade angle of incidence. The tonal blade passing frequency harmonic noise contents of propeller A are significantly higher than those generated by propeller B under matching operating conditions. In contrast, the far-field acoustic spectra of propeller B contain significant levels of high-frequency broadband noise compared to pro- peller A. The high-frequency broadband noise generated by propeller B radiates to the far-field in a dipolar pattern, strongly influencing its overall noise radiation trend. On the other hand, the overall noise directivity pattern of propeller A is mainly influenced by the BPF tonal noise radiation. The

content of high-frequency broadband in the acoustic spectra of the lower-pitched propeller, under matching operating conditions as the higher-pitched propeller, suggests an apparent influence of blade pitch ratio on both noise level and radiation characteristics. This geometrical-dependent influ- ence is expected to cause boundary layer separation on the blade trailing edge for lower-pitched pro- pellers when operated under matching settings as higher-pitched propellers. 5. ACKNOWLEDGEMENTS

The first author would like to acknowledge the financial support of Majlis Amanah Rakyat Malaysia. The second author would like to acknowledge the EPSRC (Engineering and Physical Sciences Re- search Council) for post-doctoral sponsorship at the University of Bristol from June 2020 to Decem- ber 2021 (Grant No. EP/S013024/1). The first and third authors would like to acknowledge Horizon 2020 research and innovation programme under grant agreement number 882842 (SilentProp pro- ject). 6. REFERENCES

1. Filippone, A. & Barakos G. N. Rotorcraft systems for urban air mobility: A reality check. Aero-

nautical Journal, 125 , 3 – 21 (2021). 2. Rizzi, S. A., Huff, D. L., Boyd, D. D., Bent, P., Henderson, B. S., Pascioni, K. A., Sargent, D. C.,

Josephson, D. L., Marsan, M., He, H. B. & Snider, R. Urban air mobility noise: Current practice, gaps, and recommendations (No. NASA/TP-20205007433) (2020). 3. Hubbard, H. H. Aeroacoustics of flight vehicles: theory and practice. Volume 1. Noise sources.

National Aeronautics and Space Admin Langley Research Center Hampton VA (1991). 4. Wright, S. The acoustic spectrum of axial flow machines, Journal of Sound and Vibration , 45(2) ,

165 – 223 (1976). 5. Glegg, S. & Devenport, W. Aeroacoustics of low Mach number flows: fundamentals, analysis,

and measurement, Academic Press (2017). 6. Kurtz, D., & Marte, J. A review of aerodynamic noise from propellers, rotors, and lift fans, JPL

Technical Report , 32-1462 (1970). 7. Bass, R. M. Small Scale Wind Tunnel Testing of Model Propellers, AIAA Paper , 86 – 392, January

1986. 8. Ol, M., Zeune, C., & Logan, M. Analytical/experimental comparison for small electric unmanned

air vehicle propellers, AIAA Paper 2008, 7345, August 2008. 9. Jamaluddin, N.S., Celik, A., Baskaran, K., Rezgui, D. and Azarpeyvand, M., Aeroacoustic per-

formance of propellers in turbulent flow, In AIAA AVIATION 2021 FORUM, p. 2188, (2021). 10. Mayer, Y., Jawahar, H. K., Szoke, M. & Azarpeyvand, M. Design of an aeroacoustic wind tunnel

facility at the University of Bristol, in Proceedings of the 24th AIAA/CEAS Aeroacoustics Con- ference (2018). 11. Brandt, J., & Selig, M. Propeller performance data at low reynolds numbers, in 49th AIAA Aero-

space Sciences Meeting including the New Horizons Forum and Aerospace Exposition, 1255, January 2011. 12. Liu, W., Zagzebski, J. A., & Liu, W. Trade-offs in data acquisition and processing parameters for

backscatter and scatterer size estimations, IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control , 57 , 340 – 352 (2010). 13. Grande, E., Romani, G., Ragni, D., Avallone, F. & Casalino, D. Aeroacoustic investigation of a

propeller operating at low reynolds numbers, AIAA Journal , 1 – 12 (2021).