ATEFA – Project’s results on UAM air traffic noise and air-taxi certi- fication Michael Bauer 1 Munich Aeroacoustics 85551 Kirchheim b. Muenchen, Germany Daniel Redmann 2 Kopter Germany GmbH 85635 Höhenkirchen-Siegertsbrunn, Germany

ABSTRACT ATEFA, Germany’s first nationally funded research project on UAM community noise, aimed to pro- vide first answers, how novel air traffic noise from air-taxis can be assessed in a realistic scenario. Two eVTOL air-taxi systems – diverse in design and acoustics – and one regional STOL were acous- tically described and implemented in a traffic scenario which was covering the greater area around Munich in southern Germany, interlinking surrounding small towns, five larger cities of the region and the international airport. So far, in most studies generic noise emission data were used to discuss quite complex, but also generic air traffic scenarios. ATEFA did follow a different path, based on realistic technical data and passenger number predictions. Besides this, simulated certification pro- cedures were carried out for the two air-taxi systems. This activity was performed for an understand- ing of the applicability of already existing rules and regulations regarding noise certification. The project will close by end of 2022, but significant results for air-taxi flyover simulations, large area noise mapping, and certification aspects are already part of this paper.

1. INTRODUCTION

New air vehicles may enter service around the year 2025+. ATEFA – Air-taxis: First Operational Noise Assessment – as a first German approach, will include noise source modeling, air-taxi opera- tions and certification aspects, and will provide first insight to answer questions related to community noise coming from future air-taxi operations [1]. One main task of ATEFA was collecting sufficient technical data of two air-taxi systems and the reference aircraft, in order to describe their main acoustic properties which will be used for all fly- over simulations in the project [1]. These acoustic properties were predicted according to existing semi-empiric methods, especially for noise sources related to rotor and propeller noise, as, e.g., pro- vided in [2]. Main results from the ATEFA project, will be summarized here, together with simulation results from flyover situations. A full noise mapping, performed for the greater area of Munich City in southern

1 michael.bauer@muc-aero.com 2 daniel.redmann@koptergroup.de

worm 2022

Germany are completing the work shown here. The STOL regional aircraft was not included since the focus here is on the urban air-traffic with the possibility of operating in vertical take-off and landing mode. 2. AIR-TAXI CONCEPT SELECTION AND DESCRIPTION

2.1. Selection Criteria The criteria for the selection of the two UAM aircraft described in this paper have been defined in the project’s Task 1.1 already. One main idea – which is also part of ATEFA’s objectives – was not to design new air-taxi concepts but start from existing ones. However, the technical specifications should be close to real air-taxi systems. For this reason, it has been decided that each of the selected air-taxis will have a similar counterpart in “real life”, but with deviating technical details to avoid any assessment of an existing or currently developed product, and to remain with general statements. In some cases, this deviation was inevitable, anyway, due to the lack of publicly available data.

One main requirement has been that both aircraft and the reference aircraft should have a clearly different acoustic character. This requirement appears to be necessary for later conclusions on com- munity noise effects and certification aspects. But all three vehicles should have in common to be operated in VTOL conditions.

As reference aircraft, the small helicopter type “CABRI” [3] was selected because it is expected to cover the same range for maximum take-off weight (MTOW) like most of the promising air-taxi concepts. Another advantage is the availability of data from noise certification, which will later help comparing with air-taxi noise and assessing possible certification aspects .

To include acoustically highly diverse air-taxis, their propulsion concept including its installation situation should be represented by a different aircraft different design. In any case, it was never the aim of ATEFA to model and simulate any existing air-taxi concept from the various (potential) man- ufacturers. Therefore, the two finally selected air-taxis will be labelled and described according to a specific classification, which will be shortly described in the next section.

2.2. eVTOL Classification for ATEFA Since the currently discussed future air-taxi systems show a big variety of different technical solu- tions, a classification will be necessary for a better “clustering” of the different aircrafts. There are already several classifications in use. In the principle, two major properties appear to be useful for classifying a certain air-taxi: Its technical design or the range of its intended operations. By technology, the following five major types may be used [4]:

• Vectored Thrust • Lift + Cruise • Wingless (Multicopter) • Hover Bikes/Personal Flying Devices • Electric Rotorcraft By operational range, four classes seem to be useful [5]

• urban to urban (intra-urban) • urban to next urban (inter-urban) • urban to outland (regional) • cross border (long distance)

worm 2022

Both of ATEFA’s air-taxis are capable to operate intra-urban as well as inter-urban, so the description regarding the aircrafts’ basic technology will be used here.

2.3. Technical Data Only those data, preliminarily relevant for the acoustic description of the vehicle, were of interest. Table 1 provides some of the technical specifications developed in ATEFA.

Table 1: Basic Technical Specifications for Reference Aircraft and Air-taxi Systems.

Helicopter Multicopter Vectored Thrust ATEFA ID REF AT#1 AT#2 UAM Classification (helicopter) multicopter vectored thrust MTOW [kg] 700 820 920 wings w/o w/o 2 tiltable wings propulsion combustion electric electric propulsion type main/tail rotor rotors propeller number of rotors/propellers 1+1 15 6+6 propeller/rotor mounting above cabin above cabin puller config required power [kW] 108 340 375 number of blades 3+7 2 3 average blade chord [cm] 18+4.2 6.7 3.6 - 8.9 rotor/propeller diameter [m] 7.2+0.6 2.2 1.8 nominal rpm [1/min] 530+5.148 1207 2180 number of passengers 2 2 2 In general, propeller and rotor related properties are well understood and described by theory and/or semi-empiric methods and were used for the further work in ATEFA. The most critical assumption appeared to be the estimation of a realistic value for the required electrical power. Therefore, a simple relation power vs. mass has been established from the data already identified in the frame of selecting the air-taxi configurations [6]. This relation is shown in Figure 1. Some other specific aerodynamic details, e.g., such as blade lift coefficients, are not listed in Table 1 but were still required for the noise source modelling. They were in more detail defined for the pre- diction of the aircrafts’ individual noise emission.

worm 2022

Figure 1: Simple relation between MTOW and required electric power.

osname Me hl

3. ACOUSTIC CHARACTERISATION AND FLYOVER SIMULATIONS

3.1. Air-Taxis’ Noise Emission The noise emission from the main sources was computed basically by using semi-empiric methods like [2,7]. Figure 2 shows the sound power spectra for both air-taxis, in comparison to the reference aircraft. The tonal components of the spectra were computed up the 10 th BPF. All sources were re- garded as monopoles, their spatial noise radiation is characterised by 3D-directivities, which are pro- vided by Figure 3. Figure 2: 3 rd octave spectra of the noise emission for all three a/c, from 20 Hz to 4 kHz. In the lower frequency range, the rotor/propeller tones are predominant, in the upper range the broadband noise is mainly contributing. Figure 3: Directivities on half hemispheres for the undisturbed tonal noise sources of AT#1 and AT#2. But from the propeller and rotor installation it appears to evident, that there are additional effects to be included, such as rotor-rotor interaction [8] or shielding (e.g., by the cabin). The latter resulted in

worm 2022

AT#1 AT#2

Sound Power Level [dB] au wart mretafe ” 6 100 160 250 400 «20 31000 "4 Octave Center Frequency [H2] 2500 4000

somehow relatively small effects, which may result from the cabin’s size to be relatively small com- pared to the wave lengths of the tonal components. Regarding installation effects, the situation seems to be more significant. Here an overall increase of noise emission can be observed, due to the propel- lers, operating close to each other. But since the single propeller’s overall emission is already on quite low level, the effects from installation still keep both air-taxis’ noise radiation non-critical. Figure 4: SPL on 10m-half-hemispheres of AT#1 for three exemplary 3 rd octave frequencies 25 Hz, 50 Hz and 100 Hz (upper w/o, lower with shielding effects). The differences are visible. Figure 5: SPL on 10m-half-hemispheres of AT#2 for three exemplary 3 rd octave frequencies 50 Hz, 100 Hz and 160 Hz (upper w/o, lower with shielding/installation effects). The differences are visible, too.

worm 2022

3.2. Flyover results and noise certification 3.2.1 FLYO® Toolchain The single flyover events were performed according to requirements defined in ICAO Chapter 11 [9]. This certification procedure for light helicopters was selected because chosen reference aircraft in ATEFA is a helicopter certified according to this regulation. The operational parameters, which determine the aircraft’s power setting and enabling a constant, stabilized flyover in cruise condition are shown in Table 2.

Table 2: Main operational parameters for flyover simulations according to Chapter 11 [2,3].

REF AT#1 AT#2 Thrust setting [%] 85.0/10.0 1 87.5 60.0 rotational speed [rpm] 475/3113 1 777 1057 flyover speed [km/h] 166.5 108.0 162.0 flyover altitude [m] 150 150 150 shaft power [kW] n.a. 2 91 45 rotor/propeller AoA 3 [°] 0 88 0 Temperature [°C] 25 25 25 Humidity [%] 70 70 70

1 tail rotor 2 not required for noise computation 3 for air-taxi propeller this is the tilt angle The flyover situation consisted of simple flight path at 150 m altitude and one observer (or micro- phone position in terms of certification measurement) on the centre line below the aircraft’s path at 1.2 meter above the ground. The flight path’s length was chosen to be 40 km, thus (virtually) “infinite long” compared to the altitude, i.e., long enough to capture the maximum sound pressure level during flyover and its 10dB-down time for integration [10]. The simulations were carried out using own software FLYO®. Each flyover took about 1 second of computation time. The results for all three aircraft are summarised in Table 3.

Table 3: Main results for flyover simulations according to Chapter 11.

REF AT#1 AT#2 SEL [dB(A)] 76.4 63.1 67.2 SPL max [dB(A)] 76.6 54.1 64.5 t 1,10dB down * [s] 431.3 667.9 440.7 t 2,10dB down * [s] 434.5 686.7 445.4

* absolute time on flight path For the reference aircraft, the SEL can be compared directly to the certification value [3] of 75.7 dB(A), which is only 0.7 dB below the simulation result.

worm 2022

3.2.2 Kopter Toolchain Another way to determine certification values is by using a tool chain that is also the closest to the real certification process based on the flight tests for the light helicopters that is approved by author- ities. In this case, the software developed for the analysis of measured test data according to ICAO Chapter 8, 11 and 13 requirements [10] is used as the core program and the acoustic data obtained by simulation at virtual microphone positions is used as input. This approach allows for direct compari- son and validation of results in the future when real flight test data will be available.

Figure 6: Kopter tool chain for prediction of certification values by simulation for new A/C based on combination of semi empirical pre-diction (FLYO®), commercial acoustic propagation tool (COM- SOL) and Kopter Certification Software (KARIN). The method for prediction of the certification values as kind of reference condition for the comparison of different aircrafts/eVTOLs is summarized in Figure 6 and contains following steps:

worm 2022

 First, half hemispheres described in Section 3.1 are used as an input for commercial acoustic

propagation tool COMSOL regarding overflight simulation.  Second, a propagation down to the ground from the hemisphere is performed taking into ac-

count the atmospheric attenuation according to ANSI standard S1.26-2014 [13] by means of the ray acoustic method in the frequency domain by COMSOL (see Figure 7). The resolution for this propagation calculation is the third octave in the range from 50 Hz to 10 kHz, as required by certification regulations.

 In the third step, depending on the trajectory, the speed of the simulated aircraft and the posi-

tion of the virtual microphones, the relevant acoustic data for complete manoeuvre are calcu- lated and divided in the one-third octave spectrum for 0.5 seconds each and stored as an input for the certification software.

 Finally, the certification values are determined. The results for all three aircraft are summa-

rised in Table 4.

Figure 7: Exemplary representation of emitted overall sound pressure levels on the ground for refer- ence helicopter and two selected eVTOL AirTaxis at 150m altitude.

Table 4: Results for flyover maneuver according to ICAO Chapter 11 regulations by Kopter tool-

chain.

REF AT#1 AT#2 SEL [dB(A)] 74.1 63.9 66.0 SPL max [dB(A)] 66.9 57.5 59.3 v ref [kts] 90 58 87 t 1,10dB down [s] 18 25 16 t 2,10dB down [s] 28 35 26.5 t SPL,max [s] 23 30 21.5 For the reference helicopter, the certification value according to EASA database [14] is given by 75.7 SELdB(A), which is only 1.6 dB(A) above the simulation result. A relatively good agreement for the predicted results can be observed for the AT#1 and AT#2 con- cepts independent of the chosen method (FLYO® vs. semi-empirical acoustic source description  COMSOL  KARIN). The values are only 1.2 dB from each other in the maximum. 4. UAM NOISE SIMULATIONS FOR A LARGER UAM TRAFFIC SCENARIO

worm 2022

Since the single aircraft appears to be less noisy than a small helicopter which is already on a quite low-noise level, the number of air-taxi movements per day will be the most important factor for the perceived noise on the ground [11]. Many studies, which are trying to establish a first understanding of air-taxi operations with a higher number of flights per day are using generic quantities. ATEFA’s approach was to establish realistic numbers for air-taxi traffic in larger region on the South of Ger- many [1]. Those numbers have been generated by using traffic models, based on algorithms of project partner ILR [12]. This model helped to locate the vertiports and their interconnections within the area of interest around the city of Munich. The flightpaths between these locations were merged at some knods, to create corridors where the air-taxi traffic was collected before being re-distributed on indi- vidual paths towards their actual destination. The two air-taxi types AT#1 and AT#2 were operated on total 43 flight paths with lengths between 9 km and 119 km. Both air-taxi types were distributed over these paths according to their maximum range, related to their design. The numbers for air-taxi movements were then allocated to these aircraft, taking into account that about 80 % of the flights are taking place during the time between 07:00 a.m. and 07:00 p.m., approximately 17 % will occur

between 07:00 p.m. and 10:00 p.m. [12]. The remaining flights are being operated during the hours of night. This distinction is important because of different noise penalties for flights during evening and night hours when computing the DNL. Some important results are collected in Table 5. Figure 8 shows the noise map for the complete computation area of 13.400 km 2 and for a smaller area around the City of Munich, covering 1.285 km 2 .The full area is consisting of 334,998 grid points with a distance of 200 m between each other, the smaller part consists of 128,557 grid points with a higher grid resolution of 100 m. This was necessary to avoid large regions where no AT#1 or AT#2 activities occur, which would distort the results of Table 5. At a later stage of ATEFA, an STOL regional small aircraft will being operated in those regions, too. In total, 2,648 air-taxi operations were included in the simulations to occur within 24 hours.

worm 2022

Figure 8: Noise map for the total scenario, including both air-taxi concepts (topographic map [12]). The smaller area is indicating the city’s urban area computed with a higher resolution

Table 5: Population of the inner city area of Figure 8, related to a certain DNL.

DNL [dB(A) ]

Population

[%] ≤ 20 23.11 20 - 30 44.58 30 - 40 29.93 40 - 50 2.28 50 - 60 0.11 The value of 0.11 % may appear small but can represent several thousand inhabitants for a densely populated city. 5. SUMMARY AND CONCLUSIONS

For the better understanding of future air-taxi traffic and its related community and en-route noise under realistic conditions, two diverse air-taxi concepts were described acoustically. The use of two different air-taxi systems appeared to be necessary, because future UAM traffic will consist of fleets, composed by different air-taxy types. Thus, this study included already the minimal fleet, by assum- ing that the use of more different systems does not appear to be likely in rather limited area around a city of 1.5 million people. Nevertheless, with predictions for UAM demand in the region of Greater Munich, a scenario was simulated, and the aircraft borne noise on the ground was visualized by noise mapping. It could be shown that a broader application of this novel, individual air transport can lead to relatively high sound pressure levels, averaged over 24 hours, also related to some indications of potential annoyance resulting from flyover noise created by air-taxis. These results are representing the first quantified assessment for possible UAM related noise, at least in Germany. Another task in this project is the possible comparison of already existing and established means of air transport with vertical take-off characteristics - helicopters with the new AirTaxi concepts in terms of certification. For the new air vehicles - eVTOLs, there are no certification regulations as yet. For this reason, the certification regulation for light helicopters represented by ICAO Chapter 11 was selected here to serve a possible comparison of the VTOL members. As it was shown above, a kind of comparison could be made in this way. Due to distributed propulsion and the partial use of aero- dynamic lift devices, the new representatives seem to be less noisy as helicopters, but it should be noted that this comparison does not provide any information about the annoyance. Nonetheless, this study can be used in the future as a starting point for the development of possible certification requirements for eVTOLs, as well as to make initial statements regarding the develop- ment of some new metrics that take annoyance into account for single events. The project’s results show that a further increase of knowledge about potential impacts by UAM – including annoyance – will have to follow, requiring more detailed and specific research, e.g., such as annoyance in the presence of high numbers for air-taxi movements in general.

worm 2022

4. ACKNOWLEDGEMENTS

The provision of the geographic data (vertiport positions and connecting flight paths) and the numbers for air-taxi operations by ATEFA partner Institute of Aerospace Systems/RWTH Aachen University, is acknowledged here.

ATEFA is funded by the German Federal Ministry of Economic Affairs and Climate Action, in the frame of the national project ATEFA (funding ID 20V1903), which is part of the “Luftfahrtforschungsprogramm” LuFo VI-1. 5. REFERENCES

1. Bauer, M., Redmann, D., and Weilandt, L., “ATEFA – A first German approach on UAM com-

munity noise and air-taxi certification”, INTER.NOISE 2021, 1-5 August 2021, Washington, D.C., USA. 2. Prediction Procedure for Near-Field and Far-Field Propeller Noise, SAE Aerospace Report

AIR1407, Version 08/2012. 3. Cabri G2, Technical Specifications, https://www.cabri.co.nz/technical-specifications [cited on 09

March 2022]. 4. eVTOL Classifications, http://evtol.news/classifications [cited on 09 March 2022]. 5. Air-taxi classification by range, urban air taxi - urban air taxi (urban-air-taxi.com ) [cited on 09

worm 2022

March 2022]. 6. Technical Report/Deliverable, LuFo VI-1 Project “ATEFA”, Report 1/4: Concepts’ Technical

Description, 5 February 2021. 7. Pegg, Robert J., A Summary and Evaluation of Semi-Empirical Methods for the Prediction of

Helicopter Rotor Noise, NASA Technical Memorandum 80200, December 1979. 8. Bernardini, G., et al., Numerical Characterisation of the Aeroacoustic Signature of Propeller Ar-

rays for Distributed Electric Propulsion, Appl. Sci. 2020, 10, 2643, 11 April 2020. 9. ICAO, Environmental Technical Manual on the Use of Procedures in the Noise Certification of

Aircraft, 3rd Edition, 2004. 10. ICAO, International Standards and Practices, Annex 16 to the Convention of International Civil

Aviation: Environmental Protection, Volume I: Aircraft Noise, 5th Edition, July 2008. 11. Bauer, M., First Assessment of Community Noise for a Simulated Scenario of New Urban Air

Traffic, 26th International Congress on Sound and Vibration (ICSV), Montréal, 2019. 12. Source of topographic maps (for noise map underlay): Bayerische Vermessungsverwaltung –

www.geodaten.bayern.de “, Creative Commons Namensnennung 3.0 Deutschland Lizenz (CC BY 3.0 DE). 13. ANSI S1.26-2014 (supersedes ANSI S1.26-1995) “American National Standard method for cal-

culation of the absorption of sound by the atmosphere” (Acoustical Society of America, New York, 2014). 14. EASA Rotorcraft noise database, Issue 37 of 30 March 2022

https://www.easa.europa.eu/domains/environment/easa-certification-noise-levels