ELECTRICAL MACHINES FOR AIRCRAFT POWER AND PROPULSION SYSTEMS

Abstract

An electrical machine for an aircraft electrical power system includes a stator having current-carrying coils and flux guiding stator iron defining one or more stator slots that house the current-carrying coils. The electrical machine further includes a rotor having a plurality of permanent magnets configured to interact with the stator to produce a torque. The electrical machine has an active parts mass, m.sub.act, which is a cumulated mass of components of the electrical machine that contribute to producing the torque. The electrical machine is configured to produce a peak rated torque, ?.sub.peak. The electrical machine has a slot current density, J.sub.slot,peak, when producing the peak rated torque, ?.sub.peak. A value of a machine parameter, ?, defined as: ?=?.sub.peak/(m.sub.act?J.sub.slot,peak) is greater than or equal to 5 ?Nm.sup.3 kg.sup.?1A.sup.?1.

Claims

1. An electrical machine for an aircraft electrical power system, the electrical machine comprising: a stator comprising current-carrying coils and flux guiding stator iron defining one or more stator slots that house the current-carrying coils; and a rotor having a plurality of permanent magnets configured to interact with the stator to produce a torque, wherein the electrical machine has an active parts mass, m.sub.act, the active parts mass being a cumulated mass of components of the electrical machine that contribute to producing the torque, wherein the electrical machine is configured to produce a peak rated torque, ?.sub.peak, wherein the electrical machine has a slot current density, J.sub.slot,peak, when producing the peak rated torque, ?.sub.peak, and wherein a value of a machine parameter, ?, defined as: $?=\frac{{?}_{peak}}{m_{act}?J_{\mathrm{slot},peak}}$ is greater than or equal to 5 ?Nm.sup.3 kg.sup.?1A.sup.?1.

2. The electrical machine of claim 1, wherein ? is less than or equal to 35 ?Nm.sup.3 kg.sup.?1A.sup.?1.

3. The electrical machine of claim 1, wherein ? is in a range of 6 to 25 ?Nm.sup.3 kg.sup.?1A.sup.?1.

4. The electrical machine of claim 1, wherein ? is in a range of 7 to 20 ?Nm.sup.3 kg.sup.?1A.sup.?1.

5. The electrical machine of claim 1, wherein the slot current density, J.sub.slot,peak, when producing the peak rated torque, ?.sub.peak, is in a range of 4 to 11 Amm.sup.?2.

6. The electrical machine of claim 1, wherein the rotor is an ironless permanent magnet dual rotor.

7. The electrical machine of claim 6, wherein the ironless permanent magnet dual rotor comprises a first ironless rotor portion having a first set of permanent magnets and a second ironless rotor portion having a second set of permanent magnets, wherein the second ironless rotor portion is spaced apart from the first ironless rotor portion, and wherein the stator is located between the first ironless rotor portion and the second ironless rotor portion.

8. The electrical machine of claim 7, wherein the second ironless rotor portion is axially spaced apart from the first ironless rotor portion, and wherein the stator is located axially between the first ironless rotor portion and the second ironless rotor portion.

9. The electrical machine of claim 1, further comprising: a cooling system configured to remove heat from the electrical machine, wherein the cooling system comprises a cooling system mass, m.sub.cool, and wherein a value of a machine parameter, ?*, defined as: ${?}^{*}=\frac{{?}_{peak}}{\left(m_{act}+m_{cool}\right)?J_{slot,peak}}$ is greater than or equal to 4 ?Nm.sup.3 kg.sup.?1A.sup.?1.

10. The electrical machine of claim 9, wherein ?* is less than or equal to 25 ?Nm.sup.3 kg.sup.?1A.sup.?1.

11. The electrical machine of claim 1, further comprising: an air-cooling system configured to supply a flow of air to the current-carrying coils of the stator to remove heat from the current-carrying coils, and wherein the current-carrying coils of the stator are directly exposed to the flow of air whereby the heat is transferred directly from the current-carrying coils to the air.

12. The electrical machine of claim 11, wherein the air-cooling system has a cooling system mass, m.sub.cool, less than or equal to 10 kg.

13. The electrical machine of claim 1, wherein the electrical machine is a transverse flux electrical machine.

14. The electrical machine of claim 13, wherein each respective stator slot of the one or more stator slots is a circumferentially extending and continuous stator slot, and wherein current is configured to flow through the current-carrying coils of the stator in a circumferential direction.

15. The electrical machine of claim 14, wherein the transverse flux electrical machine is a multi-lane transverse flux electrical machine comprising a first sub-machine and a second sub-machine, wherein the first sub-machine comprises a first stator having coils for carrying current and a first rotor configured to interact with the first stator to produce a torque for driving rotation of a propeller or a fan, wherein the second sub-machine comprises a second stator having coils for carrying current and a second rotor configured to interact with the second stator to produce a torque for driving rotation of the propeller or the fan, and wherein the first stator and the first rotor are coaxial with and axially spaced apart from the second stator and the second rotor.

16. The electrical machine of claim 1, wherein the electrical machine is configured to produce a maximum continuous rated torque, ?.sub.max,cont, with a rotor speed in a range of 75 to 150 rads.sup.?1.

17. The electrical machine of claim 1, wherein the peak rated torque, ?.sub.peak, is in a range of 800 to 2000 Nm, and wherein the active parts mass, m.sub.act, is in a range of 9 to 30 kg.

18. An electrical propulsion unit (EPU) for an aircraft, the EPU comprising: a propeller or a fan; and an electrical machine having: a stator comprising current-carrying coils and flux guiding stator iron defining one or more stator slots that house the current-carrying coils; and a rotor having a plurality of permanent magnets configured to interact with the stator to produce a torque for driving rotation of the propeller or the fan, wherein the electrical machine has an active parts mass, m.sub.act, the active parts mass being a cumulated mass of components of the electrical machine that contribute to producing the torque, wherein the electrical machine is configured to produce a peak rated torque, ?.sub.peak, wherein the electrical machine has a slot current density, J.sub.slot,peak, when producing the peak rated torque, ?.sub.peak, and wherein a value of a machine parameter, ?, defined as $?=\frac{{?}_{peak}}{m_{act}?J_{\mathrm{slot},peak}}$ is greater than or equal to 5 ?Nm.sup.3 kg.sup.?1A.sup.?1.

19. The EPU of claim 18, wherein the rotor of the electrical machine is configured to directly drive the propeller or the fan of the EPU, whereby an angular frequency of rotation of the rotor of the electrical machine is equal to an angular frequency of rotation of the propeller or the fan.

20. A vertical take-off and landing (VTOL) aircraft comprising: an electrical propulsion unit (EPU) comprising: a propeller or a fan; and an electrical machine having: a stator comprising current-carrying coils and flux guiding stator iron defining one or more stator slots that house the current-carrying coils; and a rotor having a plurality of permanent magnets configured to interact with the stator to produce a torque for driving rotation of the propeller or the fan, wherein the electrical machine has an active parts mass, m.sub.act, the active parts mass being a cumulated mass of components of the electrical machine that contribute to producing the torque, wherein the electrical machine is configured to produce a peak rated torque, ?.sub.peak, wherein the electrical machine having a slot current density J.sub.slot,peak when producing the peak rated torque, ?.sub.peak, and wherein a value of a machine parameter, ?, defined as $?=\frac{{?}_{peak}}{m_{act}?J_{\mathrm{slot},peak}}$ is greater than or equal to 5 ?Nm.sup.3 kg.sup.?1A.sup.?1.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0179] Embodiments will now be described by way of example only with reference to the accompanying drawings, which are purely schematic and not to scale, and in which:

[0180] FIG. 1 is a perspective view of an example of a vertical take-off and landing (VTOL) aircraft with six electrical propulsion units (EPUs);

[0181] FIG. 2 is a schematic illustration of an example of a propulsion system that may be used in the VTOL aircraft of FIG. 1;

[0182] FIG. 3 is schematic illustration of an example of a portion of the propulsion system of FIG. 2, showing the use of multiple power lanes per EPU;

[0183] FIG. 4 is a schematic illustration an example of an EPU, further showing the use of multiple power lanes;

[0184] FIG. 5A is a schematic cross-section of an example of a radial flux electrical machine;

[0185] FIG. 5B shows flux paths of an example of magnetic circuits formed in the radial flux electrical machine of FIG. 5A;

[0186] FIG. 6 is a plot illustrating an example of the optimization of the design of a radial flux electrical machine;

[0187] FIG. 7A is a schematic cross-section of an example of a transverse flux electrical machine;

[0188] FIG. 7B is a circumferential cross-section through the transverse flux electrical machine of FIG. 7A;

[0189] FIG. 8 illustrates the flux paths of the magnetic circuits formed in the transverse flux electrical machine of FIGS. 7A-B;

[0190] FIG. 9A is a schematic illustration of an example of the active parts of a stator of a transverse flux electrical machine that has three phases and two coils per phase;

[0191] FIG. 9B is a more detailed view of an example of a single coil of the stator of FIG. 9A, further showing a portion of the stator support structure;

[0192] FIG. 10A is an example of a circuit diagram showing one way of connecting the stator coils of the electrical machine of FIG. 9A;

[0193] FIG. 10B is an example of a hybrid circuit illustration combining FIG. 9A and FIG. 10A, showing the connection of the stator coils and the path of current through the stator coils;

[0194] FIG. 11 is an example of a cross-section of an electrical machine having the stator of FIGS. 9 and 10, further showing the rotor and additional support structures;

[0195] FIG. 12A is a schematic illustration of an example of a transverse flux electrical machine with radial air gaps;

[0196] FIG. 12B is a perspective view of an example of one stator coil of the transverse flux electrical machine of FIG. 12A;

[0197] FIG. 13 is a schematic illustration of an example of a transverse flux electrical machine in which a circumference of the stator is divided into sectors to implement first and second power lanes;

[0198] FIG. 14 is a cutaway view of an example of a transverse flux electrical machine in which first and second power lanes are implemented using axial stacking of active parts;

[0199] FIG. 15 is an example of a plot illustrating how the tangential force developed in a rotary electrical machine varies with air gap size and pole pitch;

[0200] FIG. 16A is a perspective view of an example of a transverse flux electrical machine;

[0201] FIG. 16B is another perspective view of an example of a transverse flux electrical machine, further showing components of a clutch mechanism of the electrical machine;

[0202] FIG. 17 illustrates an example of a cross-section of the transverse flux electrical machine of FIGS. 16A-B, showing additional components and a space in which the active parts of the machine are located;

[0203] FIG. 18A is perspective view of an example of a portion of the stator of transverse flux electrical machine of FIG. 17;

[0204] FIG. 18B is a perspective view of an example of the portion of the stator of FIG. 18A, further showing the end windings;

[0205] FIG. 19A is a further illustration of an example of a portion of the stator of the electrical machine of FIG. 17, showing how a stator slot may be divided into discrete regions to aid cooling;

[0206] FIG. 19B is a schematic illustration of an example of how portions of the stator coil of FIGS. 18A-18B and 19A may be connected;

[0207] FIG. 20A is a top-down view of an example of one stator coil of the electrical machine of FIGS. 16-19;

[0208] FIG. 20B is a side-on view of the stator coil of FIG. 20A;

[0209] FIG. 21 is a schematic cross-section of an example of an air-cooled multi-lane transverse flux electrical machine;

[0210] FIG. 22 is a perspective cutaway of an example of the transverse flux electrical machine of FIGS. 16A-B and 17, further showing the active parts;

[0211] FIG. 23A illustrates an example of how the conductor cross-section and slot packing of a transverse flux electrical machine may be varied to optimize slot current density, torque generation, and cooling; and

[0212] FIG. 23B illustrates an additional example of how the conductor cross-section and slot packing of a transverse flux electrical machine may be varied to optimize slot current density, torque generation, and cooling.

DETAILED DESCRIPTION

[0213] FIG. 1 illustrates a vertical take-off and landing (VTOL) aircraft 1 that may be used for Urban Air Mobility (UAM) applications. The VTOL aircraft 1 includes a fuselage 20 that incorporates a cabin for occupants, wings 30, a rear flight surface 40, and a distributed propulsion system 10. The distributed propulsion system 10 includes six electrical propulsion units (EPUs), four of which are front EPUs 100f and two of which are rear EPUs 100r. Also visible in FIG. 1 is a retractable undercarriage 50 in which a landing platform or gear, in this case having wheels, may be stowed during flight.

[0214] The size of the fuselage 20 and the cabin depends on the application requirements. In this example, the cabin is sized for five occupants including a pilot. Some UAM platforms, however, will not require a pilot and will instead be flown under the control of an autopilot system or may be controlled remotely.

[0215] Each EPU 100f, 100r has a propeller 110 driven to rotate by an electric motor. The four front EPUs 100f are attached to the wings 30 of the aircraft 1, and the two rear EPUs 100r are attached to the separate flight control surface 40 located towards the rear of the aircraft 1. The wings 30 and the rear control surface 40 are tiltable between a VTOL configuration (shown in FIG. 1) in which the axes of the propellers of the EPUs point upward to provide vertical lift for vertical take-off and landing and a horizontal flight configuration in which the axes of the rotors point forward. The horizontal flight configuration, while principally used for horizontal flight, may also be used for taxiing and possibly short take-off and landing (STOL) operation if supported. In other examples, the wings 30 and/or the rear control surface 40 may be fixed in a horizontal position, and the EPUs attached thereto may be tiltable in order to selectively switch between a horizontal flight mode and a vertical flight mode.

[0216] The electrical systems, including the electric motors that drive the EPUs 110f, 110r of the aircraft 1, receive electrical power from one or more battery packs and/or fuel cell packs located within the aircraft 1. The battery packs and fuel cells packs may be located within any suitable part or parts of the aircraft 1, including the EPUs 100f, 100r, the fuselage 20, and the wings 30.

[0217] While the illustrated aircraft 1 is a VTOL aircraft, UAM platforms may also be of the STOL or conventional take-off and landing (CTOL) type. Further, while an electric VTOL (eVTOL) aircraft is shown, the propulsion system may be a hybrid-electric propulsion system that includes both engines (e.g., one or more gas turbine engines) and batteries and/or fuel cells. Hybrid-electric platforms may utilize similar distributed propulsion system configurations, but the underlying power system may be a series-hybrid, parallel-hybrid, turboelectric, or other type of hybrid power system.

[0218] The configuration of the illustrated VTOL aircraft 1 is merely one example configuration, and other VTOL aircraft configurations are known and will occur to those skilled in the art. For example, a VTOL aircraft may have a different number of EPUs (e.g., eight EPUs, with four front EPUs 100f and four rear EPUs 100r). Alternatively, the VTOL aircraft may have a multi-copter (e.g., quadcopter) configuration in which the propellers or fans of the EPUs may not be tiltable and may be ducted. Other VTOL aircraft may have features of more than one type (e.g., a mix of open and ducted propulsors and/or a mix of tiltable and fixed propulsors). The present disclosure is not limited to any particular type of VTOL aircraft.

[0219] As noted above, each EPU of the six EPUs 100f, 100r includes a propeller or fan 110 driven to rotate by an electric motor that receives electrical power from an onboard power source. In some examples, each motor may receive the electrical power via its own dedicated power channel, possibly from its own dedicated power source (e.g., a battery module). In other examples, some of the EPUs may share a power channel. This is shown in FIG. 2, which is a simplified illustration of an electrical power and propulsion system 10 that may be used in the VTOL aircraft of FIG. 1.

[0220] The power and propulsion system 10 includes three power and propulsion sub-systems 10a, 10b, 10c. Each sub-system of the sub-systems 10a-c includes two of the six EPUs, and the two EPUs of each sub-system are electrically connected to a shared power channel. The first sub-system 10a includes a first of the front EPUs 100f-a and a first of the rear EPUs 100r-a connected to first power channel 140a. The second sub-system 10b includes a second of the front EPUs 100f-b and a second of the rear EPUs 100r-b connected to a second power channel 140b. The third sub-system 10c includes the remaining two front EPUs 100f-c, 100f-d, (e.g., one front-left EPU and one front-right EPU), connected to a third power channel 140c. Each power channel 140c receives DC electrical power from a battery module 150a-c. The battery modules 150a-c may be physically separate from one another or may be part of a common battery pack that outputs three separate power channels 140a-c.

[0221] In the present example, the power channels 140a-c receive the DC electrical power directly from the battery modules 150a-c. In other examples, the battery modules 150a-c may interface with the power channels 140a-c via DC:DC power electronics converters. This may allow the DC voltage level of the power channels to be kept substantially constant as the state of charge of the battery modules 150a-c decreases. Also, while in the present example the sole power source is in the form of battery energy storage 150a-c, alternative examples may include only fuel cells, or a mix of fuel cells and battery energy storage. In a hybrid system, the power source may include one or more engine-driven electric generators interfacing with the DC power channels via DC:AC power electronics converters (e.g., rectifiers).

[0222] Each EPU 100f-a, 100f-b, 100f-c, 100f-d, 100r-a, 100r-b includes a propeller or fan 110 with rotation that is driven by an electric motor 120. The motor 120 receives AC electrical power from a DC:AC power electronics converter 130 (e.g., an inverter 130). The inverter 130 receives DC electrical power from its power channel 140a-c and converts the DC electrical power to AC electrical power for supply to its motor 120.

[0223] Aircraft and power and propulsion systems of the aircraft are subject to strict safety and certification requirements. An aircraft and its safety-critical systems may be tolerant to faults; the aircraft and its safety-critical systems may be capable of continued, safe operation after the failure of a component. To this end, FIG. 3 illustrates the principle of a laned architecture that may be adopted for the power and propulsion system 10 of FIGS. 1 and 2 to improve fault tolerance.

[0224] FIG. 3 shows a power and propulsion sub-system 10n that has a laned architecture. The sub-system 10n may be any of the sub-systems 10a-c of FIG. 2. As in FIG. 2, the sub-system 10n of FIG. 3 includes two EPUs 100, each of which includes a propeller or fan 110 driven to rotate by an electric motor 120. However, unlike the motors of FIG. 2, the motors 120 of FIG. 3 are multi-lane electric motors. In particular, the motors 120 of FIG. 3 are dual-lane motors with two electrically independent AC inputs (e.g., two independent three-phase inputs).

[0225] The term multi-lane electric motor or multi-lane electrical machine, as used herein, refers to an electric motor that has at least two electrically independent sets of stator coils that may be separately excited and may separately interact with a rotor to produce torque. In this way, in the event of a fault in one lane of the motor (e.g., a stator terminal short circuit), the remaining lane(s) may remain functional, and the multi-lane motor 120 may thus continue to apply torque to rotate the propeller or fan 110. Depending on the number of lanes and the extent to which the motor 120 is overrated, the motor 120 may be able to supply all (i.e., 100%) or a large proportion (e.g., 70%, 80%, or 90%) of the torque that would be supplied to rotate the propeller or fan 110 during normal, fault-free operation. In this example, the motors 120 have two lanes and may be referred to as dual-lane motors, but a higher number of lanes (e.g., three or four lanes) may be used. The combination of an independent set of stator coils and an associated rotor with which the stator coils interact may be referred to as one lane of the multi-lane electric motor, or as a sub-machine of the multi-lane electrical machine.

[0226] A multi-lane motor may take one of a number of different forms. In one example, the dual-lane motor of FIG. 3 takes the form of two axially stacked motors having rotors that are mechanically coupled to the same output shaft. In this case, the motor has two completely separate sets of active parts (e.g., two axially spaced stators and two axially spaced rotors) but has some shared structures and features, (e.g., shared support structures, casing, and cooling features). In another example, each dual-lane motor of FIG. 3 has a single stator structure having a circumference that houses two electrically independent sets of coils, one set belonging to the first sub-machine and the second set belonging to the second sub-machine. The first set of coils and the second set of coils may interact with a common rotor. The above described approaches may be extended to a higher number of lanes (e.g., three axially stacked sub-machines) or may be combined to give a higher number of lanes (e.g., four lanes formed from two axially stacked arrangements, each having two sub-machines). Other examples will occur to those skilled in the art.

[0227] FIG. 3 also shows that each lane of the dual-lane electric motor 120 receives its multi-phase AC input from a separate, independent DC:AC inverter 130i, 130ii. Likewise, each inverter 130i, 130ii receives its DC input from a separate, independent DC power channel 140i, 140ii that itself receives power from its own battery module 150i, 150ii. Thus, the propulsion sub-system 10n is not only tolerant to faults in the multi-lane electric motor 120, but is also tolerant to faults in other sub-system components. For example, a failure in a converter circuit 130i of the first lane will not prevent the supply of electrical power to the second lane of the associated motor 120. Likewise, the loss of one battery module 150i or a fault in a power channel 140i will not prevent the operation of the inverters 130ii of the second lane or the second lanes of the motors 120. Thus, the use of a laned architecture, including multi-lane motors 120, increases the redundancy in and fault tolerance of the power and propulsion system 10.

[0228] Other measures, not shown in the simplified systems 10, 10n of FIGS. 2-3, are also possible. For example, a propulsion system 10 may provide for a degree of reconfigurability to improve the fault tolerance and power availability within the system. For example, in the event of a failure of one battery module (e.g., battery module 150i), the sub-system 10n may be reconfigured (e.g., by the selective opening and closing of switches such as contactors or solid-state power controllers (SSPCs)) to allow electrical power from the second battery module 150ii to be supplied to the first power channel 140i. As another example, following a fault in the first lane of one or both of motors 120, the sub-system 10n may be reconfigured to allow electrical power from the first battery module 150ii to be supplied to the second power channel 140ii.

[0229] FIG. 4 is a schematic illustration of a multi-lane EPU 100 such as may be used in the VTOL aircraft and propulsion systems of FIGS. 1-3. The EPU 100 includes a propeller or fan 110 mechanically coupled to, and thereby drivable by, a rotor of a multi-lane electric motor 120. In this example, the multi-lane motor 120 is a dual-lane motor in which each of the two sub-machines has three phases. Thus, the motor 120 is shown as having two three-phase inputs that receive power from the outputs of two DC:AC converters 130i, 130ii. The DC:AC converters 130i, 130ii interface with respective DC power channels 140i, 140ii, the positive and negative rails of which are indicated.

[0230] The inverters 130i, 130ii are controlled by controllers 135i, 135ii. The controllers 135i, 135ii may, for example, control the switching frequencies, switching timings, and duty cycles of MOSFETs of the inverters 130i, 130ii to adjust the magnitudes and frequency of the output AC voltage and current waveforms of the inverters 130i, 130ii. In this way, the controllers 135i, 135ii may control the lanes of the motor 120 to produce the required torque, for example. In this example, each lane has its own controller; again, this prevents a fault in one controller (e.g., controller 135i) from impacting the entire EPU 100.

[0231] FIG. 4 also shows an EPU cooling system 125. The EPU cooling system 125 is responsible for managing the temperature of the motor 120 and the inverters 130i, 130ii during use. For example, the cooling system 125 provides that the temperature of the insulation of the stator coils does not exceed its rated temperature. The cooling system 125 may take any suitable form, including a liquid cooling system that utilizes a pumped liquid coolant (e.g., an oil) or an air cooling system that directs a flow of air (e.g., ambient air) to cool the components. In some examples, the cooling system 125 includes substantially separate cooling systems for cooling the electrical machine 120 and the inverters 130i, 130ii. In other examples, the cooling system 125 is shared by the electrical machine 120 and the inverters 130i, 130ii. Further, in some examples, the cooling system 125 may include separate parts for each lane of the EPU 100 to prevent a cooling system fault from impacting the entire EPU 100.

[0232] The EPU 110 may include a gearbox 105. The optional gearbox 105 may be required to step down the speed of the rotor(s) of the motor 120 to a lower speed of rotation for the propeller or fan 110. VTOL aircraft are expected to have relatively large propellers or fans 110 (e.g., diameters of 2-4 meters) in order to limit disk loading while increasing propulsive efficiency during VTOL and hover. At the same time, there is a desire to keep aerodynamic noise low, wherein the aerodynamic noise is strongly dependent on the propeller tip speed. The combination of a large propeller diameter and a low propeller tip speed necessitates a relatively low propeller rotational speed. Unless the electrical machine 120 is capable of providing the required torque at a low rotational speed, which is relatively high given the large propeller, a gearbox 105 is to be provided.

[0233] Also shown in FIG. 4 are a propeller bearing unit 111 and an EPU lubrication system 115. The presence of and designs of the propeller bearing unit 111 and the EPU lubrication system 115 will depend on the design of the EPU 110 and are beyond the scope of the present disclosure and will not be described any further.

[0234] The VTOL aircraft 1, the propulsion system 10, and the EPU 100 described with reference to FIGS. 1-4 are only intended to be examples, and many other arrangements are possible and within the scope of the present disclosure. As noted previously, the aircraft 1 may additionally or alternatively include fuel cells or engines. The aircraft 1 may also be of a different VTOL design (e.g., multi-copter or different number and arrangement of EPUs) or utilize a different electrical power system layout. However, the above description explains certain principles of a VTOL aircraft.

[0235] The general design of VTOL aircraft, such as the one described above, results in a number of design challenges. Some of these are discussed below.

[0236] In one example, a need for redundancy and propulsive efficiency calls for a distributed propulsion system with a relatively large number of EPUs. In the above example, there are six EPUs, and most proposed VTOL aircraft designs include at least four EPUs. This increases the mass and reduces the power density of the VTOL aircraft because each EPU includes not only lift- and thrust-producing parts but support and attachment structures, cabling, etc.

[0237] In a second example, high torque and low speed requirements of the propeller or fan of the EPU, discussed above, may call for a gearbox to step down the output rotor speed of the electrical machine. A gearbox is a relatively heavy component and also introduces additional complexity as well as lubrication and maintenance requirements. Further, each EPU would require its own gearbox, multiplying the gearbox mass by, for example, six times. A direct drive arrangement would eliminate this mass and complexity and be of great advantage. However, designing a high torque, low speed electrical machine that is lightweight and efficient, yet does not have onerous cooling requirements, is a challenge.

[0238] Table 1 provides exemplary values of a hover parameter, ?, that may be achieved by motors described herein. The hover parameter ? is defined (see Equation (26)) as the continuous torque produced by the motor while the VTOL aircraft is hovering (?.sub.hover) divided by the continuous angular speed of rotation of the rotor of the motor while the VTOL aircraft is hovering (?.sub.hover), measured in radians per second. By hovering, it is provided that the EPUs of the VTOL aircraft are producing sufficient thrust to lift the aircraft and maintain a constant aircraft altitude, without requiring the assistance of airframe lift (e.g., wing-borne lift).

TABLE-US-00001 TABLE 1 ?(Nmsrad.sup.?1) Example 1 Example 2 Example 3 7.2 13.1 16.3

[0239] Motors described herein may have values of ? in the range 5 to 20 Nmsrad.sup.?1, which may allow for omission of the gearbox 105, resulting in reduced EPU and aircraft mass.

[0240] In another example, strict safety and certification requirements of aircraft call for a fault-tolerant electrical power system (e.g., the laned architecture described above with reference to FIGS. 3 and 4). This results in further multiplication (e.g., duplication) of components in an EPU: electrical machines with multiple sets of active parts; multiple power channels; multiple inverters; and multiple cooling systems. This results in a further increase in EPU mass. It would be desirable to reduce the mass of the active parts and cooling system associated with the electrical machine, to the extent that this is possible, while meeting the platform design requirements.

[0241] Table 2 provides examples of values of a take-off parameter, ?, that may be achieved by a VTOL aircraft with one or more EPUs incorporating an electric motor described herein. The take-off parameter, ?, is defined (see Equation (24)) as the maximum tip speed (v.sub.tip), measured in ms.sup.?1, of the propeller or fan of the EPU to occur during a vertical take-off operation of the VTOL aircraft divided by the active parts torque density (?.sub.actsee Equation (1)).

TABLE-US-00002 TABLE 2 ? (sm.sup.?1) Example 1 Example 2 Example 3 1.2 2.1 4.7

[0242] Motors described herein may have take-off parameters in the range of 0.5 to 7.5 sm.sup.?1, which may be associated with reduced system mass and noise.

[0243] In another example, multiplication of electrical components such as inverters results in a stacking of losses in the system. For example, a propulsion system with six EPUs and a dual lane architecture would include at least twelve inverters, each having its own losses. One mitigation is to design inverters with high efficiencies, for example, by using state-of-the-art semiconductor materials. Even then, however, it would be desirable to be able to operate the inverters in an operating regime close to their peak efficiency, which occurs when the electrical frequency of the inverter output waveform is relatively high. This is a design challenge, especially for direct drive, because it requires the use of a high inverter output frequency with a low rotor speed of rotation.

[0244] Electrical machine designs that are optimized for aircraft propulsion systems, particularly for VTOL aircraft, and may address one or more of the above problems and/or other problems are now described with reference to FIGS. 5A-23.

[0245] FIG. 5A illustrates a radial flux permanent magnet motor 200 (e.g., a motor) that may be used for EPUs of electric aircraft, including VTOL aircraft. As noted previously, permanent magnet motors may be provided for VTOL applications because their power density is relatively high compared with most other designs.

[0246] For clarity, FIG. 5A only shows the active parts of the motor 200. Active parts refers to the components of the motor 200 that contribute to the production of torque. Thus, active parts include magnetic flux generating components such as coils and magnets, and magnetic flux guiding components such iron, but active parts do not include support structures, cooling system features, etc., which do not contribute to the production of torque. The active parts of the motor 200 have a cumulated mass of m.sub.act, referred to herein as the active parts mass. For the avoidance of doubt, the active parts mass includes the mass of the end windings and coil insulation because these are integral features of the coils without which a motor cannot produce any torque.

[0247] The radial flux motor 200 includes a stator 210 and a rotor 220 arranged to rotate about an axis of rotation 230.

[0248] The stator 210 includes an annular stator back iron 211, which may also be referred to as a yoke, and a plurality of circumferentially arranged stator teeth 212 (e.g., stator teeth 212) that project radially inwardly from the back iron 211. The stator teeth 212 define stator slots 213, which may also be referred to as stator winding space, between the stator teeth 212. In the present example, there are twenty-four stator teeth 212 defining twenty-four stator slots 213 therebetween. The stator teeth 212 and/or the stator iron 211 may, for example, be formed of laminations of a ferromagnetic material to improve their flux guiding performance while reducing the induction of eddy currents in the stator teeth 212 and/or the stator iron 211. In another example, the flux guiding material includes a soft magnetic composite (SMC) such as, for example, a non-iron material with embedded iron particles. The active parts mass, m.sub.act, of the motor 200 includes any carrier material of the flux guiding iron (e.g., non-iron material included in laminations) or the non-iron material in an SMC.

[0249] The stator 210 further includes a plurality of electrically conductive stator coils 214 (e.g., stator coils) wound around the stator teeth 212. The stator coils 214 may be formed of any suitable material, such as copper or aluminum. Strands of the conductor that form the stator coils 214 have (e.g., are coated in) an electrically insulating material to prevent short circuits. In the present example, there are twelve stator coils 214, and each coil occupies two of the slots 213.

[0250] In this example, the motor 200 is a three-phase motor, and thus there are 12/3=4 stator coils 214 per phase. The three phases are designated U, V, W in FIG. 5A, and the stator coils 214 are labelled 1-4. Stator coils 214 of the same phase (e.g., U1, U2, U3, U4) are evenly distributed about the circumference of the stator 210, while circumferentially adjacent stator coils 214 (e.g., U1 and V1, or V1 and W1) belong to different phases. Many different stator winding arrangements are known, but, in this example, a distributed winding arrangement in which each coil is wound around two teeth that are located 2?/8=?/4 radians (45 degrees) apart is used. Stator coils 214 of the same phase are electrically connected (e.g., in series or in parallel). The terminals of each set of phase coils may be connected in, for example, a star or delta configuration, and the input terminals may be connected to an inverter (e.g., a two-level, three phase bridge circuit) from which the stator coils 214 receive current. In an alternative example, for increased fault tolerance, the stator coils 214 of each phase may be connected to its own inverter circuit (e.g., an H-bridge circuit).

[0251] The rotor 220 includes an annular rotor back iron 221 and a plurality of permanent magnets 222 (e.g., permanent magnets) arranged around a circumference of the rotor 220 forming a plurality of permanent magnet rotor poles (e.g., permanent magnet poles). Circumferentially adjacent permanent magnet poles are of opposite polarity. The permanent magnet poles are distributed evenly about the rotor circumference and define a pole pitch, P.sub.?, equal to 2? divided by the number of permanent magnet poles (N.sub.P):

[00027]$\begin{array}{cc}{P}_{?}=\frac{2?}{N_P}& \left(27\right)\end{array}$

[0252] In this example, there are eight permanent magnet poles, so P.sub.? is equal to 2?/8=?/4 radians (45 degrees).

[0253] In addition to the pole pitch, P.sub.?, a pole arc length, P.sub.L, of the motor is also defined. Herein, the pole arc length is equal to pole pitch, P.sub.?, measured in radians, multiplied by the active parts radius, the active parts radius being half the active parts diameter, D.sub.Act, of the motor:

[00028]$\begin{array}{cc}{P}_L=\frac{P_{?}?D_{Act}}{2}& \left(28\right)\end{array}$

[0254] The active parts diameter, D.sub.Act, is a diameter corresponding to a radially outermost active part of the motor 200. In this example, in which the rotor 220 is radially inward of the stator 210, a radially outer circumference of the stator iron 211 defines the active parts diameter. When the rotor 220 is instead radially outward of the stator 210, a radially outermost active part of the rotor defines the active parts diameter. In the present example, if the motor 200 is sized for an EPU of a VTOL aircraft, the active parts diameter may be about 0.45 meters, giving a pole arc length, P.sub.L, of about 17.7 cm.

[0255] FIG. 5A also labels an air gap 215 that separates the rotor 220 from the stator 210. The air gap 215 has a width, measured in the radial direction for the radial flux motor 200, equal to G.sub.Air.

[0256] In use, the stator coils 214 of the stator 210 are excited with AC power to generate a rotating magnetic field that interacts with the magnetic field of the permanent magnets 222 to produce torque. The torque causes the rotor 220 to rotate relative to the stator 210 about the axis of rotation 230. The motor 200 is configured to produce a maximum continuous rated torque of ?.sub.max,cont and a peak rated torque of ?.sub.peak. As used herein, ?.sub.max,cont is the highest torque the motor can produce and sustain for an extended period (e.g., at least three minutes) at ISA sea level conditions. ?.sub.max,cont depends to some extent on the capabilities of the cooling system of the motor, which is configured to remove heat at a rate sufficient to maintain the temperature of the stator coil insulation below its rated temperature while operating at ?.sub.max,cont. As used herein, ?.sub.peak is a highest torque the motor can produce for a short period (e.g., a three second transient period) at ISA sea level conditions without damaging the motor due to, for example, breakdown of the coil insulation due to excessive voltage or excessive heat generation.

[0257] FIG. 5B illustrates main magnetic circuits produced by the radial flux motor 200 during use. The current flowing through each stator coil 214 produces magnetic flux that flows in a radial direction. For a given stator slot 213, the flux produced by the stator coil 214 flows radially outward along one tooth 212 and radially inward along another tooth 212 on a circumferentially opposite side of the slot 213. At the radially outer side of the stator 210, the magnetic flux flows circumferentially along the stator back iron 211 between stator teeth 212. At the radially inner side of the stator 210, the magnetic flux crosses the air gap 215 to flow to/from a stator tooth 212 from/to a permanent magnet 222 of the rotor 220. In the rotor 210, flux flows from a permanent magnet 222 to the rotor iron 221 and then flows circumferentially along the rotor iron 221 to another permanent magnet 222.

[0258] FIG. 6 is a plot 1000 illustrating how the design of a radial flux motor 200 may be optimized for use in a EPU of VTOL aircraft. A three-dimensional surface 1000 is shown, with shading of the surface 1000 indicating the temperature of the insulation of the stator coils 214.

[0259] The vertical axis represents the active parts torque density, ?.sub.act, of the motor, defined in Equation (1) as the peak rated torque divided by the cumulated mass of the active parts (i.e., the components which contribute to the production of torque) measured in Nmkg.sup.?1. In the case of VTOL aircraft, it desirable for the active parts torque density, ?.sub.act, to be high because this implies the platform's torque production requirements are met at a low motor mass, which is a significant benefit in VTOL aircraft due in part to the multiplication of components (e.g., multiple EPUs). The remaining two axes show the slot current density, J.sub.slot,peak, (e.g., the slot current density when producing the peak rated torque), measured in A(mm).sup.?2, and the linear RMS current loading, A.sub.rms, measured in kA/m. In certain examples, the higher the current loading and the slot current density, the higher the torque production. However, if the current density is high, the stator coil temperature will be higher for a given nominal cooling rate because resistive losses (i.e., I.sup.2R losses) will also be higher.

[0260] On the surface 1000, an isotherm 1001 (i.e., a line of constant temperature) is shown. The isotherm 1001 represents operation at the rated temperature of the insulation, assuming operation of a liquid cooling system that cools the stator coils at a nominal rate. In other words, the isotherm 1001 divides the surface 1000 into two design space regions: a lower region 1002 below the isotherm 1001 in which operation is sustainable at the nominal cooling rate; and an upper region 1003 above the isotherm 1001 in which operation is not sustainable at the nominal cooling rate. Thus, the isotherm 1001 may be regarded as the optimal design. FIG. 6 further shows three possible stator tooth and slot designs, labelled (i), (ii) and (iii), and operating points thereof that lie on the isotherm 1001.

[0261] First referring to design (i), this shows a stator tooth 212 that is relatively long in the radial direction and includes a large volume of conductor in the slots 213 defined circumferentially adjacent to the tooth 212. The large volume of active parts (e.g., the iron stator teeth and the conductor) results in high torque for a given slot current density. However, the large volume of active parts also results in a high active parts mass, which limits the active parts torque density ?.sub.act. Further, the use of radially long stator teeth 212 provides that, for a motor of a given diameter (noting the diameter will be constrained by the EPU integration requirement), there is relatively little space to flow coolant around the active parts. Thus, while the slot current density is low for a given current value, the extent to which the slot current density may be increased without departing from the isotherm 1001 into the region 1003 is limited.

[0262] Design (ii) is a more optimized design for VTOL aircraft in that design (ii) better balances torque production, slot current density, and active parts mass. As shown, compared with design (i), design (ii) has radially shorter teeth 212 with a smaller volume of conductor in the slots. While this reduces torque production at a given value of the slot current density, radially shorter teeth 212 with a smaller volume of conductor in the slots reduces the active parts mass. This also provides additional room for coolant, which improves cooling and therefore allows for an increase in the slot current density without departing from isotherm 1001 into the upper region 1003. Further, the radially shorter teeth have a lower aspect ratio, which may improve flux guiding, and allow for the use of a larger radius rotor. The use of a larger radius rotor may produce a higher torque. Overall, as shown, tooth design (ii) has the peak value of ?.sub.act on the isotherm 1001.

[0263] Design (iii) illustrates the impact of further reducing the radial length of the stator tooth 212 and decreasing the volume of conductor in the slot. As before, the reduction in active parts volume results in lower torque production at a given slot current density but also a reduction in active parts mass. The additional free volume for coolant allows the slot current density to be increased, thus increasing active parts torque density ?.sub.act, without departing from the isotherm 1001. However, resistive losses increase with the square of the current density whereas the torque increases in a linear fashion. There is therefore a point on the isotherm 1001 after which the increase in torque that results from the increase in slot current density, and which is made possible by the increase in cooling volume, does not compensate the reduction in torque that results from the reduced volume of conductor. Therefore, design (iii) is associated with a lower value of ?.sub.act than design (ii).

[0264] While optimized radial flux designs may be used in the EPUs of VTOL aircraft, further performance improvements and optimizations may be provided. To this end, FIG. 7A, illustrates a transverse flux electric motor 300 that may be particularly suitable for use in VTOL aircraft. FIG. 7B is a circumferential cross-section of the motor 300 of FIG. 7A. As before, for clarity and ease of explanation, only the active parts of the motor 300 are shown. The illustrated motor 300 has only a single phase. This is for clarity of explanation; in practice, a motor may have a higher number of phases, and such a motor will be described below.

[0265] The transverse flux motor 300 has a stator 310 and a rotor 320a, 320b arranged to rotate about an axis of rotation 330.

[0266] The stator 310 includes flux guiding stator iron 311 that defines a circumferentially extending stator slot 313 (e.g., generally annular stator slot; annular winding space). In this example, the stator iron 311 includes a plurality of circumferentially arranged flux guiding stator elements 312 (e.g., stator elements 312) that together define and surround the annular winding space 313. In the present example, there are eight stator elements 312, alternately facing axially up and axially down, defining a single stator slot 313. The stator elements 312 may be formed of laminations of a ferromagnetic material or an SMC to improve their flux guiding performance while reducing the induction of eddy currents. In other examples, the winding space 313 may be defined by a continuous (e.g., a single piece) stator iron structure as is the case in the radial flux machine of FIG. 5A.

[0267] The stator slot 313 houses a circumferentially extending stator coil 314. As in the radial flux motor 200, the stator coil 314 may be formed of any suitable material such as copper or aluminum. The conductor that forms the stator coil 314 has (e.g., is coated in) an electrically insulating material to prevent short circuits. In FIGS. 7A and 7B, the stator coil 314 is a solid piece of conductor (e.g., the coil has a single turn). In practice, the stator coil 314 may have a plurality of turns; this is discussed in more detail below.

[0268] FIG. 7B is a circumferential cross-section through the active parts of the transverse flux motor 300, and more clearly shows how the flux guiding stator elements 312 may define the open slot 313. In FIG. 7B, two circumferentially adjacent stator elements 312i, 312ii defining a single stator pole are shown. Stator element 312i is in the foreground, and stator element 312ii is in the background. Each of the stator elements 312i, 312ii has a generally claw-shaped form factor and includes a body portion 3121 and two projections 3122, 3123 that project from the body portion 3121. In this example, the projections 3122, 3123 extend axially away from the body portion 3121, and the projections 3122, 3123 of circumferentially adjacent stator elements 312i, 312ii face axially opposite directions. In FIG. 7B, the projections 3122, 3123 of the first stator element 312i project axially downwards, whereas the projections 3122, 3123 of the second stator element 312ii project axially upwards. In this way, the two circumferentially adjacent stator elements 312i, 312ii cooperate to define the cross-section of a winding space (e.g., a slot 313) therebetween, which in this case has a hexagonal shape, though other shapes may be formed by modifying the shape and curvature of the stator elements 312i, 312ii. Collectively, the eight stator elements 312 of the stator 310 (see FIG. 7A) define an annular winding space, and the stator coil 314 is housed in the annular winding space. Other stator element form factors are possible and in accordance with the present disclosure.

[0269] The rotor 320a, 320b, is a dual rotor and has two rotor portions: an inner rotor portion 320a that is radially inside the stator 310; and an outer rotor portion 320b that is radially outside the stator 310. In this example, the inner rotor portion 320a and the outer rotor portion 320b are mechanically connected so that the inner rotor portion 320 and the outer rotor portion 320b rotate together. Each of the inner rotor portion 320a and the outer rotor portion 320b includes a plurality of circumferentially arranged permanent magnets 322a, 322b defining evenly spaced permanent magnet poles (e.g., poles). Circumferentially adjacent poles are of opposite polarity. In this example, the inner rotor portion 320a includes eight permanent magnet poles, and the outer rotor portion 320b includes eight permanent magnet poles. Thus, in the present example, the pole pitch, P.sub.?, of the motor 300 is 2?/8=?/4 radians (45 degrees).

[0270] The permanent magnets 322a of the inner rotor portion 320a are affixed to an outer surface of an inner rotor structure 321a. Similarly, the permanent magnets 322b of the inner rotor portion 320b are affixed to an inner surface of an outer rotor structure 321b. The inner rotor structure 321a and the outer rotor structure 321b may include flux guiding stator iron (e.g., laminations of a ferromagnetic material). However, the use of a dual rotor design, with permanent magnets 322a, 322b on both radial sides of the stator 310, may allow for the omission of iron material from the rotor 320a, 320b because the permanent magnets 322a, 322b may define closed magnetic circuits. This is described in more detail below. Other transverse flux motors 300 in accordance with the present disclosure may not feature a dual rotor, and, in this case, rotor iron or additional stator iron may be provided to define closed magnetic circuits.

[0271] As stated above, the pole arc length, P.sub.L, of the motor is defined as the pole pitch, P.sub.?, measured in radians, multiplied by half the active parts diameter, D.sub.Act, of the motor. In this example, which features an ironless dual rotor 320, the active parts diameter is defined by the outer diameter of the permanent magnets 322b of the outer rotor portion. Assuming the motor 300 is sized for an EPU of a VTOL aircraft, the active parts diameter may be about 0.45 meters, giving a pole arc length, P.sub.L, of about 17.7 cm.

[0272] FIGS. 7A-7B also label air gaps 315a, 315b that separate the inner rotor portion 320a and the outer rotor portion 320b from the stator 310. The inner air gap 315a has a width, for example, measured in the radial direction, equal to G.sub.Air,1. The outer air gap 315b has a width, for example, measured in the radial direction, equal to G.sub.Air,2. The air gap widths G.sub.Air,1, G.sub.Air,2 may be the same to equalize the motor loading.

[0273] The transverse flux motor 300 of the present example has radial air gaps 315a, 315b. In other words, the two rotor portions 320a, 320b are on radially opposite sides of the stator 310. In other examples, a transverse flux motor has axial air gaps. In other words, the two rotor portions would be on axially opposite sides of the stator. Such an example will be described with reference to FIGS. 9-11.

[0274] The transverse flux motor 300 of the present example has only a small number of pole pairs and relatively large values for the pole pitch, P.sub.?, and pole arc length, P.sub.L This is for ease of explanation. As will be described in detail below, the present disclosure provides the selection of a larger number of pole pairs to improve the characteristics of the motor.

[0275] FIG. 8 shows the magnetic flux paths of the main magnetic circuits formed in the transverse flux motor 300 of FIGS. 7A-B. The left-hand diagram of FIG. 8 is the axial end view of FIG. 7A, with the main magnetic circuits overlaid and labelled by the magnetic flux density vector {right arrow over (B)}. The dotted lines indicate where the flux lines are below the plane of the page. The right-hand diagram, taken through plane A-A, is a zoomed-in perspective view of two circumferentially adjacent stator elements 312, forming a single stator pole, and illustrates the three-dimensional shape of the flux path {right arrow over (B)}. The current and force vectors {right arrow over (I)} and {right arrow over (F)} are also shown.

[0276] The current flows through the stator coils 314 in the circumferential direction. This is illustrated in FIG. 8 by the current vector {right arrow over (I)}. At each circumferential position, the current flow produces a loop of magnetic flux in a plane perpendicular to the direction of current flow. Considering a circumferential position corresponding to a first stator element 312i, the loop of flux is guided along the radially extending body portion 3121i of the stator element 312i before entering an axial projection of the stator element 312i. From there, the flux crosses the inner air gap 315a and enters a permanent magnet 322a of the inner rotor portion 320a. The flux then flows circumferentially through the inner rotor magnets 322a to reach a circumferential position corresponding to the adjacent second stator element 312ii. From there, the flux again crosses the inner air gap 315a, this time into the second stator element 312ii. The flux is then guided along the radially extending body portion 3121ii of the second stator element 312ii before entering an axial projection of the second stator element 312ii. From there, the flux crosses the outer air gap 315b and enters a permanent magnet 322b of the outer rotor portion 320b. The flux then flows circumferentially through the outer rotor magnets 322b to reach another circumferential position (e.g., in this case, back to the circumferential position corresponding to the first stator element 312i). From here, the flux again crosses the outer air gap 315b to the first stator element 312i, thus completing the magnetic circuit. The left-hand drawing of FIG. 8 shows a similar magnetic circuit for each adjacent pair of stator elements 312.

[0277] Thus, FIG. 8 shows that the magnetic circuits of the transverse flux motor 300 are three-dimensional (e.g., the magnetic circuits have components in the radial, axial, and circumferential directions) and spiral around the annular stator winding region 313 that houses the stator coil 314.

[0278] As mentioned previously, a practical motor will include more phases than the single phase shown in FIG. 7A. As also mentioned previously, a transverse flux electrical machine may alternatively utilize axial air gaps instead of radial air gaps. To this end, a three-phase axial air gap transverse flux motor 60 will be described with reference to FIGS. 9-11.

[0279] FIG. 9A in axial end view of the active parts of a three-phase stator 61 of a transverse flux motor 60 that has axial air gaps.

[0280] In this example, the three-phase stator 61 includes six circumferentially arranged phase modules 610-1 to 610-6, distributed evenly about the stator circumference. Radially opposite phase modules (e.g., phase modules 610-1 and 610-4) are associated with the same phase (e.g., phase U) of the motor 60 to provide mechanical balance. Each phase module 610-1 to 610-6 includes flux guiding stator iron 611 defining a circumferentially extending and open winding space 613 (e.g., a slot), and a coil 614 (e.g., a stator coil) housed within the slot 613.

[0281] FIG. 9B shows one of the phase modules 610 fixed to a support structure 640. In this example, the slot 613 and the coil 614 housed within the slot 613 include first and second radially spaced portions. Specifically, the coil 614 includes a first, radially inner and circumferentially extending, coil portion 614a housed within a first, radially inner and circumferentially extending, slot portion 613a. The coil 614 further includes a second, radially outer and circumferentially extending, coil portion 614b housed within a second, radially outer and circumferentially extending, slot portion 613b. The first coil portion 614a and the second coil portion 614b are connected at respective circumferential ends by end windings 617. In this way, the current flowing through a coil 614 changes direction in the end windings 617, and the current flows through the first coil portion 614a in a circumferential direction opposite to (e.g., generally antiparallel to) the current that flows through the second coil portion 614b.

[0282] The flux guiding stator iron 611 includes two sets of flux guiding stator elements: a radially inner first set of flux guiding stator elements 612a and a radially outer second set of flux guiding stator elements 612b. The radially inner first set of stator elements 612a define the radially outer first slot portion 613a that houses the first coil portion 614a. The radially outer second set of stator elements 612b defines the radially outer second slot portion 613b that houses the second coil portion 614b. Each of the two sets of stator elements 612a, 612b includes a plurality of circumferentially arranged, evenly distributed stator elements 612. In the present example of FIG. 9B, the first, radially inner set of stator elements 612a has twenty-five stator elements 612, whereas the second, radially outer set of stator elements 612b has twenty-seven stator elements 612. Other numbers of stator elements are possible.

[0283] Each stator element 612 is substantially as described above with reference to FIGS. 7-8 (e.g., each element 612 includes a body portion 3121 and two projections 3122, 3123). However, in the example of FIGS. 9-11, the stator elements 612 are oriented so that the body portions 3121 extend axially and the projections 3122, 3123 extend radially. The projections 3122, 3123 of circumferentially adjacent stator elements 612i, 612ii of each set 612a, 612b alternately project radially inwardly and radially outwardly so as to define an open slot cross-section, similar to that shown in FIG. 7B but with the radial direction and the axial direction swapped. Collectively, all of the stator elements 612 within a set (e.g., set 612a) define a slot portion (e.g., radially inner slot portion 613a).

[0284] FIG. 10A is a circuit diagram of the stator 61 illustrating how the coils 614 of the six phase modules 610-1 to 610-6 may be connected together.

[0285] The stator 61 has three phase terminals U, V, W by which the stator 61 receives AC electrical power from an inverter arrangement. For example, each phase terminal may be connected to a two-level, one-phase H-bridge inverter circuit, or each phase terminal may be connected to one of the phase legs of a two-level, three-phase DC:AC inverter circuit.

[0286] The coil 614 of each phase module 610-1 to 610-6 has two terminals. Respective first terminals of the coils 614 of the radially opposite first phase module 610-1 and fourth phase module 610-4 are connected in parallel to the first phase terminal U. Respective first terminals of the coils 614 of the radially opposite second phase module 610-2 and fifth phase module 610-5 are connected in parallel to the second phase terminal V. Respective first terminals of the coils 614 of the radially opposite third phase module 610-3 and sixth phase module 610-6 are connected in parallel to the third phase terminal W. Respective second terminals of the first phase module 610-1, second phase module 610-2, and third phase module 610-3 are connected at a first star point 616-1. Respective second terminals of the fourth phase module 610-4, fifth phase module 610-5, and sixth phase module 610-6 are connected at a second star point 616-2. Thus, in this example, the phases are connected in a star configuration. In other motors in accordance with the present disclosure, the phases may be connected in a delta configuration.

[0287] FIG. 10A also shows the measurement of the voltage between the two star points 616-1, 616-2. A difference in the voltage between the star points may be used to diagnose a fault in the stator coils 614.

[0288] FIG. 10B is a hybrid diagram combining the axial end view of the stator 61 (FIG. 9A) and the circuit diagram of the stator 61 (FIG. 10A). As well as showing the connection of the coils 614 together, and to the phase connections U, V, W and the star points 616-1, 616-2, FIG. 10B shows the coils 614 in schematic form. FIG. 10B shows that the coils 614 have end windings 617 at circumferential ends, respectively, which allows the current to reverse direction between the inner coil portion 614a and the outer coil portion 614b. FIG. 10B also shows that each coil 614 has multiple winding turns 6140. Although only two turns are illustrated, in practice, each coil may have more than two turns.

[0289] FIG. 11 shows the motor 60 in cross-section. While the previous figures have only illustrated certain active parts of their respective motors, FIG. 11 also shows various other features, including features of an EPU in which the motor may be integrated.

[0290] The motor 60 includes a main motor housing 601 that includes, amongst other things, the stator 61. Various components of the stator 61 are visible and labelled in FIG. 11. This includes, for the two of the six phase modules 610-1 to 610-6 that are visible in the cross-section of FIG. 11, the first, radially inner set of flux guiding stator elements 612a, the second, radially outer set of flux guiding stator elements 612b, the first, radially inner slot portion 613a defined by the first set of stator elements 612a, and the second, radially outer slot portion 613b defined by the second set of stator elements 612b. The coil portions 614a, 614b are omitted from FIG. 11 to more clearly show the slots 613 (e.g., the winding space).

[0291] The rotor 62 includes a rotor housing 625 mechanically coupled to the EPU drive shaft 630 via a coupling structure 626 that, for example, may be disk-shaped. Thus, in this example, the rotor housing 625 rotates with the drive shaft 630 about an axis of rotation 63. The active parts of the rotor (e.g., the permanent magnets that interact with the active parts of the stator 61) are located within the rotor housing 625 and also rotate together with the housing 625.

[0292] The permanent magnets include four groups of permanent magnets 622a-1, 622a-2, 622b-1, 622b-2, each of which are circumferentially distributed around the rotor 62. In one example, each group of permanent magnets 622a-1, 622a-2, 622b-1, 622b-2 is arranged as a Halbach array. The first group of permanent magnets 622a-1 and the second group of permanent magnets 622a-2 form a first set of magnets 622a that interact with the magnetic field associated with the first, radially inner coil portion 614a and the first set of flux guiding stator elements 612a. The third group of permanent magnets 622b-1 and the fourth group of permanent magnets 622b-2 form a second set of magnets 622b that interact with the magnetic field associated with the second, radially outer coil portion 614b and the second set of flux guiding stator elements 612b.

[0293] The first group of magnets 622a-1 is located axially adjacent to (e.g., axially above) and facing a first axial end of the radially inner portions 612a, 613a, 614a of the active parts of the stator 61. The first group of magnets 622a-1 is separated from the first axial end of the active parts of the stator 61 by a first axial air gap, schematically indicated in FIG. 11 but too small to see, having a width G.sub.Air,1 in the axial direction. The second group of magnets 622a-2 is located axially adjacent to (e.g., axially below) and facing a second axial end of the radially inner portions 612a, 613a, 614a of the active part of the stator 61. The second group of magnets 622a-2 is separated from the second axial end of the active parts of the stator 61 by a second axial air gap, schematically indicated in FIG. 11 but too small to see, having a width G.sub.Air,2 in the axial direction. The values of G.sub.Air,1 and G.sub.Air,2 may be the same to balance loading.

[0294] The third group of magnets 622b-1 is located axially adjacent to (e.g., axially above) and facing a first axial side of the radially outer portions 612b, 613b, 614b of the active parts of the stator 61. The third group of magnets 622b-1 is separated from the first axial end of the active parts of the stator 61 by a third axial air gap, schematically indicated in FIG. 11 but too small to see, having a width G.sub.Air,3 in the axial direction. The fourth group of magnets 622b-2 is located axially adjacent to (e.g., axially below) and facing a second axial end of the radially outer portions 612b, 613b, 614b of the active parts of the stator 61. The fourth group of magnets 622b-2 is separated from the second axial end of the active parts of the stator 61 by a fourth axial air gap, schematically indicated in FIG. 11 but too small to see, having a width G.sub.Air,4 in the axial direction. The values of G.sub.Air,3 and G.sub.Air,4 may be the same to balance loading, and may be the same as G.sub.Air,1 and G.sub.Air,2.

[0295] In use, the stator coils 614 of the phase modules 610-1 to 610-6 are excited with current from inverter circuits. The current flows in a circumferential direction through the inner coil portion 614a and the outer coil portion 614b of the phase modules 610-1 to 610-6, changing direction in the end windings 617. The magnetic flux generated by the current is guided in magnetic circuits axially through the body portions of the stator elements 612, radially through the projections of the stator elements 612, axially across the axial air gaps, and circumferentially between rotor magnets 622. The magnetic field produced by the rotor magnets 622 interacts with the stator field to produce torque, which drives rotation of the rotor 62 and, via the coupling structure 626, rotation of EPU drive shaft 630. Rotation of the drive shaft 630 drives rotation of a propeller or fan, and a propeller interface 65 is shown in FIG. 11.

[0296] Thus, a three-phase transverse flux motor 60 with axial air gaps and two coils per phase has been described. For completeness, FIGS. 12A and 12B illustrate an equivalent transverse flux motor 70 with radial air gaps. It will be appreciated that a radial air gap transverse flux motor 300 was described with reference to FIGS. 7-8, but for a single phase having a single coil.

[0297] FIG. 12A is an axial end view of a transverse flux motor 70 having a stator 71 and a rotor 72. The rotor 72 rotates about an axis of rotation 73. Once again, the stator 71 includes six circumferentially arranged phase modules 710-1 to 710-6. Radially opposite phase modules (e.g., phase modules 710-1 and 710-4) are associated with the same phase (e.g., phase U) and are connected, for example, as shown in FIG. 10A. The rotor includes a radially inner rotor portion 72a and a radially outer rotor portion 72b, with the stator 71 positioned radially therebetween. The stator coils 74 are omitted from FIG. 12A for clarity, but a single stator coil 74 of one phase module 710 is illustrated in and will be described with reference to FIG. 12B.

[0298] As in the motor 60 of FIGS. 9A-9B, 10A-10B, and 11, each phase module 710-1 to 710-6 of the motor 70 of FIGS. 12A-12B includes two sets of stator elements 712a, 712b defining two circumferentially extending slot portions 713a, 713b housing two circumferentially extending coil portions 714a, 714b of the coil 714. However, while the two sets of stator elements 612a, 612b of the previously described motor 60 are radially spaced, the two sets of stator elements 712a, 712b of the motor 70 of the present example are axially spaced. This can be most easily appreciated from FIG. 12B, which shows one phase module 710. FIG. 12B shows the two sets of axially spaced stator elements 712a, 712b and the coil portions 714a, 714b, which are connected at circumferential ends of the phase module 710 by end windings 717. The direction of current flow is indicated by the circumferential arrows labelled {right arrow over (I)}. FIG. 12B shows that in this example, each coil 714 is formed of a plurality of winding turns.

[0299] In the present example, each set of stator elements 712a, 712b includes a plurality of circumferentially arranged, evenly distributed flux guiding stator elements 712 (e.g., stator elements 712). Each of the stator element 712 is substantially as described above with reference to FIGS. 7-8 (e.g., each stator element 712 includes a body portion 7121 and two projections 7122, 7123). The stator elements 712 are oriented so that the body portions 7121 extend radially and the projections 7122, 7123 extend axially from the body portion 7121. The projections 7122, 7123 of circumferentially adjacent stator elements 712i, 712ii of each set 712a, 712b alternately project axially inwardly and axially outwardly so as to define an open slot cross-section, similar to that shown in FIG. 7B. Collectively, all of the stator elements 712 within a set (e.g., set 712a) define a slot portion (e.g., first axial slot portion 713a).

[0300] Comparing FIG. 12A and FIG. 12B, the stator phase modules 710-1 to 710-6 extend axially into the plane of the page such that only one axial portion (e.g., portion 712a) of each phase module 710 is visible. Similarly, the two rotor portions 72a, 72b extend axially into the plane of the page. Each rotor portion 72a, 72b includes first and second axially spaced groups of permanent magnets such that, in total, the rotor 72 has four groups of magnets: a radially inner and axially inner first group 722a-a, a radially inner and axially outer second group 722a-b, a radially outer and axially inner third group 722b-a, and a radially outer and axially outer fourth group 722b-b. The dashed lines used for labels 722a-b and 722b-b indicate that the permanent magnets of the second group 722a-b and the fourth group 722b-b are axially behind those of the first group 722a-a and the third group 722b-a. Each group of permanent magnets 722a-a, 722a-b, 722b-a, 722b-b may be arranged as a Halbach array.

[0301] The magnets of the first group 722a-a face a radially inner side of the stator 71 at a first axial height and are separated from the radially inner side by a first radial air gap of width G.sub.Air,1. The magnets of the second group 722a-b face the radially inner side of the stator 71 at a second axial height and are separated from the radially inner side by a second radial air gap of width G.sub.Air,2. The magnets of the third group 722b-a face a radially outer side of the stator 71 at the first axial height and are separated from the radially outer side by a third radial air gap of width G.sub.Air,3. The magnets of the fourth group 722b-b face the radially outer side of the stator 71 at the second axial height and are separated from the radially outer side by a second radial air gap of width G.sub.Air,4. The first radial air gap width G.sub.Air,1 and the second radial air gap width G.sub.Air,2 may be the same to balance loading. Likewise, the third radial air gap width G.sub.Air,3 and the fourth radial air gap width G.sub.Air,4 may be the same to balance loading. In some examples, all four radial air gaps widths are the same. The radial air gaps are only schematically indicated in FIGS. 12A-12B because the air gaps are too small to resolve.

[0302] In use, the stator coils 714 of the phase modules 710-1 to 710-6 are excited with current from inverter circuits. The current flows in a circumferential direction through the axially inner portion 714a and the axially outer coil portion 714b of the phase modules 710-1 to 710-6, changing direction in the end windings 717. The magnetic flux generated by the current is guided in magnetic circuits radially through the body portions of the stator elements 712, axially through the projections of the stator elements 712, radially across the radial air gaps, and circumferentially between rotor magnets. The magnetic field produced by the rotor magnets interacts with the stator field to produce torque that drives rotation of the rotor 72a, 72b.

[0303] Thus, a three-phase transverse flux motor 70 with radial air gaps and two coils per phase has been described.

[0304] As described above with reference to FIGS. 3 and 4, for VTOL applications, it may be desirable to utilize multi-lane (e.g., dual-lane) electric motors. FIG. 13 and FIG. 14 illustrate how a multi-lane architecture may be implemented in a transverse flux electric motor.

[0305] FIG. 13 illustrates a transverse flux motor 70 (e.g., a motor) with radial air gaps and two independent power lanes (e.g., two sub-machines). The motor 70, which is similar in its construction to the motor 70 of FIGS. 12A-12B, includes six phase modules 710-1 to 710-6. While the motor 70 of FIGS. 12A-12B has two coils per phase with radially opposite coils connected and belonging to the same phase, the motor 70 has two three-phase sub-machines 70a, 70b with one coil per phase, and radially opposite coils correspond to different sub-machines 70a, 70b. Each sub-machine 70a, 70b receives its power from a different inverter so that if an inverter fails, one sub-machine 70a, 70b is not affected by the failure.

[0306] In more detail, the circumference of a stator of the motor 70 is circumferentially divided into two sectors each spanning ? radians (180 degrees): a first sector 70a and a second sector 70b. The first sector 70a corresponds to a first three-phase sub-machine 70a and has three phase modules 710-1 to 710-3, each corresponding to one phase of the first sub-machine 70a. The second sector 70b corresponds to a second three-phase sub-machine 70b and has three phase modules 710-4 to 710-6, each corresponding to one phase of the second sub-machine 70b. The stators of the sub-machines 70a, 70b share and interact with a common rotor, which is configured in a same or similar way to the dual rotor of the motor 70 of FIG. 12A.

[0307] By increasing the number of sectors into which the circumference is divided, the number of sub-machines may be increased. For a number of sub-machines equal to N.sub.L, there may be N.sub.L sectors each spanning 2?/N.sub.L radians (360/N.sub.L degrees). The number of coils per phase may be increased by increasing the number of phase modules per sector.

[0308] FIG. 14 illustrates a transverse flux motor 60 with axial air gaps and two independent power lanes (e.g., two sub-machines). The two sub-machines 60a and 60b are implemented by axially stacking two sets of active parts. Specifically, the motor 60 has two axially stacked sub-machines 60a, 60b, each of which is of similar construction to and operates in much the same way as the axial air gap motor 60 of FIGS. 9A-9B, 10A-10B and 11. Rotors of the two sub-machines 60a, 60b, each of which is of dual-rotor construction, are mechanically coupled so that the rotors rotate together, though it will be appreciated the two rotors may instead be independent and be separately connected to an output drive shaft.

[0309] An advantage of the axial stacking approach is that, as well as implementing multiple sub-machines for increased fault tolerance, the torque developed by the motor 60 is increased without requiring an increase in the motor diameter or the slot current density. Although the active parts mass does increase, the use of some common features (e.g., non-active features such as cooling and support structures) limits the overall increase in the mass of the motor 60. The number of power lanes may be increased beyond two, if this is desired, by axially stacking more than two sub-machines and/or dividing the circumference of each stator into multiple sub-machines as shown in FIG. 13.

[0310] Motors in accordance with the present disclosure may be configured to have particularly high active part torque densities, defined in Equation (1). For example, motors may have a value of ?.sub.act of at least 50 Nmkg.sup.?1. Table 3 illustrates the calculation of ?.sub.act for three motors in accordance with the present disclosure, each of which is a transverse flux motor.

TABLE-US-00003 TABLE 3 m.sub.act ?.sub.peak ?.sub.act (kg) (Nm) (Nmkg.sup.?1) 10.2 870 85.2 13.4 1300 97.0 17.8 1450 81.5

[0311] As shown, each of the example transverse flux motors has a particularly high value of ?.sub.act, in excess of 80 Nmkg.sup.?1. Noting that a VTOL aircraft may include at least four EPUs, such a high value of ?.sub.act results in a significant mass saving when compared even to VTOL aircraft utilizing optimized radial flux motors.

[0312] The increased active parts torque density may be understood by comparing the two-dimensional magnetic circuits of the radial flux motor (FIG. 5B) and the three-dimensional magnetic circuits of the transverse flux motor (FIG. 8). Referring first to FIG. 5B, the magnetic circuits are substantially two-dimensional (e.g., the magnetic circuits lie in planes perpendicular to the axis of rotation 230) and pass through the annular region of the stator 210 in which the coils 214 are housed. There is, therefore, competition for space in this annular region between the conductor, which carries the current that generates the stator field, and the stator teeth that guide the flux in the radial direction. Thus, for a motor of a given active parts diameter, any gain in performance that may be realized by increasing the conductor volume may be offset by the effects of a corresponding reduction in a stator iron volume, and vice versa. For example, a higher conductor volume may allow more torque to be produced, or the same torque to be produced at lower current density, the latter reducing the cooling burden. However, this would require more slender stator teeth, which are less efficient flux guides, and/or fewer stator teeth, which may result in higher torque ripple (e.g., especially at the low rotor speeds for VTOL aircraft). Thus, an improvement in one performance metric (e.g., peak torque) will likely require either a reduction in another performance metric (e.g., efficiency and torque ripple) and/or an increase in the mass of the active parts of the motor. Referring now to FIG. 8, in contrast with FIG. 5B, the magnetic circuits are three-dimensional and spiral around the annular winding space 313 that houses the conductor. There is therefore no competition or much more limited competition for space in the stator between the conductor and the flux guiding stator iron. Consequently, both the stator pole design and number may be optimized (e.g., without reducing the conductor volume), resulting in a higher active parts torque density.

[0313] In designing for a high value of ?.sub.act, it is useful to introduce a dimensionless machine parameter ?, defined in Equation (5) as the cumulated volume of the conductor (e.g., the stator coils), V.sub.conductor, included in the motor divided by cumulated volume of the flux guiding iron material, V.sub.iron, included in the motor. In accordance with the present disclosure, a notably high value of ?, greater than or equal to 0.25, may be selected to promote the production of high torque with a low active parts mass. Table 4 shows values of ? for three motors in accordance with the present disclosure and sized for an EPU of a VTOL aircraft:

TABLE-US-00004 TABLE 4 V.sub.conductor (cm.sup.3) V.sub.iron (cm.sup.3) ? Example 1 39.2 62.2 0.63 Example 2 34.1 92.2 0.37 Example 3 52.2 49.7 1.02

[0314] The iron material may be present in both the stator and the rotor. However, in the transverse flux motors of the examples described herein, only the stator includes iron material. This reduces the iron volume and promotes a higher value of ?.

[0315] Another characteristic parameter for the purposes of an EPU of a VTOL aircraft is ?, defined in Equation (3) as the ratio of the active parts torque density and the slot current density at the peak rated current. ? may be a useful parameter for optimizing a motor for a VTOL EPU because the parameter rewards torque production but penalizes the addition of active parts mass, which increases the EPU weight, and at the same time penalizes the use of a high slot current density, which creates onerous cooling requirements and increases the likelihood of failures. In accordance with the present disclosure, a particularly high value of ? (e.g., greater than or equal to 5 ?N.sup.3mkg.sup.?1A.sup.?1) may be selected. Table 5 shows values of ? for three transverse flux motors in accordance with the present disclosure:

TABLE-US-00005 TABLE 5 J.sub.slot,peak ?.sub.act (Nmkg.sup.?1) (Amm.sup.?2) ? (?N.sup.3mkg.sup.?1A.sup.?1) 84 6 14 108 7.5 12.5 76 11 6.9

[0316] As noted above, the radial flux motor 200 has magnetic circuits that are two-dimensional and pass radially through the annular region of the stator 210 in which the slots 213 are defined. This creates competition for space in the annular region of stator 210 between the flux guiding material (e.g., the stator teeth 212) and the slot 213 that houses the conductor (e.g., the coils 214). This provides that increasing the number of stator pole pairs requires a decrease in the volume of conductor. Equivalently, increasing the volume of conductor requires a decrease in the number of stator pole pairs. In contrast, in a transverse flux motor, there is no, or much more limited, competition for space in the annular stator region. Thus, the number of stator pole pairs, formed by the stator elements 312 in this example, may be increased with no impact on the volume of conductor. The impact of this may be appreciated from FIG. 15.

[0317] FIG. 15 is a plot 1100 illustrating how, for a motor of a given diameter, the tangential force (y-axis) developed by the motor varies with the pole pitch, P.sub.? (x-axis), and the air gap width, G.sub.Air. Two plots 1101, 1102 corresponding to two air gap widths are shown: a 3 mm air gap (plot 1101) and a smaller 1.4 mm air gap (plot 1102). These values are shown purely for the purpose of explanation.

[0318] As can be seen from both plots 1101, 1102, at small values of the pole pitch, P.sub.?, the tangential force increases as the pole pitch increases. However, the tangential force eventually reaches a maximum at a particular value of the pole pitch P.sub.?,max. Increasing the pole pitch beyond P.sub.?,max decreases the tangential force and thus reduces the torque developed by the motor. By comparing the two plots 1101, 1102, it is also shown that: (i) the tangential force increases as the air gap decreases; and (ii) the value of P.sub.?,max decreases as the air gap decreases.

[0319] From FIG. 15, it may be appreciated that to increase the torque density of a motor of a given diameter, it is desirable to decrease the air gap width. However, for a given air gap width, torque production may only be maximized if the pole pitch may be decreased to P.sub.?,max. In a radial flux motor, the extent to which P.sub.? may be decreased (e.g., by increasing the number of pole pairs) is limited by competition for space in the stator. In a transverse flux motor, however, the extent to which P.sub.? may be decreased (e.g., by increasing the number of pole pairs) may instead only be limited by manufacturing constraints and the flux guiding efficiency of the stator iron. Thus, a transverse flux motor may access an upper-left region of the plot 1100 corresponding to high torque density, whereas a radial flux motor may only access a relatively lower-right region of the plot.

[0320] In accordance with the present disclosure, to further optimize a motor for use in an EPU of a VTOL aircraft, the value of a motor parameter ?, defined in Equation (9) as the product of the pole pitch, P.sub.?, and the air gap width, G.sub.Air, may be selected to be in the range 5 to 100 micro radian-metres. Table 6 shows examples of values of ? for three example motors in accordance with the present disclosure. Values are provided in micro radian-meters.

TABLE-US-00006 TABLE 6 (10.sup.?6 radian-meters) Example 1 Example 2 Example 3 11.0 26.7 68.1

[0321] For motors sized for VTOL aircraft, the selection of a value of ? in this range may optimize the torque-producing tangential force and thus increase the active parts torque density, pact. Small values of P.sub.? (e.g., less than or equal to 10 degrees, or less than 5 degrees) and high values of the pole pair number (e.g., at least 15, or greater than or equal to 50) may be provided, along with small values of the air gap width (e.g., less than or equal to 1.5 mm). Motors in accordance with the present disclosure may have more than one air gap because of the use a dual rotor design and/or the use of axial stacking of active parts to implement multiple lanes. All air gaps of a given motor may be approximately the same size, such that the value of ? will be approximately the same for all air gaps of a motor. Where different air gaps are used, however, the largest air gap may be used to calculate ? as the largest air gap may limit the torque density.

[0322] Motor-inverter combinations in accordance with the present disclosure may also have optimized values of a parameter ?, defined in Equation (11). ? is the ratio of the pole arc length, P.sub.L (see Equation (28)) and the maximum value of the electrical frequency, f.sub.max, of the current output by the inverter and received by the stator coils of the motor during use. In accordance with the present disclosure, the value of ? may be between 1 and 30 ?ms, which is unusually low. Table 7 shows examples of values in accordance with the present disclosure.

TABLE-US-00007 TABLE 7 P.sub.L (mm) f.sub.max (kHz) ? (?ms) Example 1 7.0 1.5 4.7 Example 2 4.2 0.6 7.0 Example 3 17.5 1.2 14.6

[0323] As described above with reference to FIG. 4, particularly in point d), the propulsion system of a VTOL aircraft has a large number of inverters as a result of its distributed propulsion system and fault tolerant electrical architecture. This results in the stacking of inverter losses and provides that inverter efficiency may have a significant impact on performance (e.g., aircraft mission range). Selecting an unusually low value for ? (e.g., in a range of 1 to 30 ?ms or in a range of 3 to 15 ?ms) has been found to reduce inverter losses when operating at a relatively low rotor speed. The use of a low value of ? may therefore not only reduce inverter losses, but also allow for the omission of a speed-reducing gearbox in the EPU without sacrificing low aerodynamic noise or motor efficiency.

[0324] An important consideration in the context of aerospace electrical machines is fault tolerance. In accordance with the present disclosure in which the electrical machines may be the permanent magnet type, the tolerance to a stator terminal short circuit fault may be particularly important.

[0325] In the event of a stator terminal short circuit fault (e.g., a short circuit fault condition in the electrical network connected to the stator terminals), the rotation of the rotor will drive a fault current into the network for as long as the rotor excites the stator windings. In motor designs that feature rotor windings, it is possible to stop excitation of the rotor windings to prevent the excitation of a voltage in stator windings and thus stop the fault current. However, in a permanent magnet motor, the rotor is permanently excited and will, unless the permanent magnets are demagnetized or the rotor is moved away from the stator, continue to excite a voltage in the stator windings that will drive the fault current. With zero or little impedance in the short-circuited electrical network, this fault current may be very large. The heat dissipated by the stator windings, which causes heating of the coil insulation, increases with the square of the current (I.sup.2R losses).

[0326] One potential mitigation to this problem is for the EPU to include a mechanism or device to physically disconnect the permanent magnet rotor from the propeller fan so that the inertia of the propeller does not continue to force rotation of the rotor. For example, a freewheel transmission may be included in the EPU. However, this solution may add mass, complexity, and maintenance requirements to the EPU. Another potential mitigation would be to provide additional overrating to the cooling system of the motor, so that the cooling system may maintain the temperature of the insulation at or below its rated temperature even in the presence of a terminal short circuit fault. However, this also adds mass to the EPU and may make air cooling (described in more detail below) unfeasible, adding even more mass to the EPU due to the requirement to adopt liquid cooling.

[0327] In accordance with the present disclosure, an electrical machine may have a short-circuit insulation temperature parameter, ?, defined in Equation (17) that satisfies the inequality:

[00029]$?=\frac{{?}_{\mathrm{ins},\mathrm{cont}}\left(I_{\mathrm{SC}}\right)}{{?}_{\mathrm{ins},\mathrm{cont}}\left(I_{cont}\right)}?1.1.$

[0328] In the above equation, I.sub.SC is the steady-state short circuit current, and I.sub.cont is the continuous rated current (e.g., the highest current the stator coils are rated to carry for a sustained period; this is associated with production of the maximum continuation rated torque, ?.sub.max,cont). ?.sub.ins,cont(I.sub.SC) is the temperature of the insulation when carrying the steady-state short circuit current, and ?.sub.ins, cont(I.sub.cont) is the temperature of the insulation when carrying the continuous rated current. Designing a motor to have a short-circuit insulation temperature parameter, ?, less than or equal to 1.1 may allow the stator coils and their insulation to be sufficiently cooled following a terminal short circuit fault without additional overrating the cooling system. A value of ? in the range of 0.7 to 1.0 may be provided and may, for example, allow for the use of air cooling in a transverse flux motor without additional overrating of the cooling system or a reduction in the performance of the motor during normal operation.

[0329] Additionally or alternatively, a short circuit current ratio, ?, defined in Equation (16), may satisfy the inequality:

[00030]$0.5??=\frac{I_{SC}}{I_{peak}}?1.2$

[0330] In the above equation, I.sub.peak is the peak rated current (e.g., the current associated with production of the peak rated torque, ?.sub.peak). A value of this ratio in a range of 0.6 to 0.9 (e.g., in a transverse flux motor) may strike a good balance between fault tolerance and good electrical and mechanical performance.

[0331] A further motor design optimization in accordance with the present disclosure is to select a design with a value of a characteristic motor parameter ?, defined in Equation (6), greater than or equal to 65 Nmkg.sup.?1. Table 8 shows the calculation of ? for three exemplary transverse flux motors, sized for use in the EPU of a VTOL aircraft.

TABLE-US-00008 TABLE 8 ?.sub.act (Nmkg.sup.?1) cos(?) ? (Nmkg.sup.?1) Example 1 85 0.65 131 Example 2 95 0.75 127 Example 3 74 0.85 87

[0332] The selection of a value of greater than or equal to 65 Nmkg.sup.?1, particularly a value in a range of 80 to 190 Nmkg.sup.?1, may provide a surprising combination of low EPU mass and fault tolerance. For example, such a selection may correspond to a sweet spot in the combined mass of an EPU's motor, inverter, and cooling system while offering good tolerance against stator terminal short circuit faults. This may be understood in terms of the effect of the power factor and its relationship with the torque density of the motor. A motor with a low power factor may require oversized power electronics but will also have a lower steady state terminal short circuit current. Thus, the selection of the power factor affects the inverter mass and also the required cooling system mass, as the cooling system may be sized to cool the motor under short circuit conditions. At the same time, the value of the power factor is mediated by the inductance of the motor, which depends on the quantity and distribution of active parts. This affects the active parts mass and the peak rated torque. A value of ? in a range of 80 to 190 Nmkg.sup.?1 may strike an effective balance between these competing requirements.

[0333] It is also useful to introduce a motor parameter Z, defined in Equation (14) as the product of the power factor of the motor and active parts mass divided by the efficiency, ?, of the motor. The efficiency is defined as the efficiency when the motor is producing the maximum continuous rated torque, ?.sub.max,cont, at ISA sea level conditions. Table 9 illustrates values of Z in accordance with the present disclosure. The values of Z are notably low and may be associated with a strong balance between efficiency and fault tolerance in a motor sized for VTOL aircraft.

TABLE-US-00009 TABLE 9 Z (kg) Example 1 Example 2 Example 3 11 7.2 14.5

[0334] In accordance with the present disclosure, the value of Z may be less than or equal to 30 kg or in a range of 5 to 15 kg. This may be achieved most effectively in a transverse flux motor, where the inductance may be tuned to achieve a desirable power factor, (e.g., in a range of 0.6 to 0.9), without a significant negative impact on the efficiency of the motor. In a radial flux motor, the length of the magnetic circuits (illustrated in FIG. 5B) may be shorter than those in a similarly-sized transverse flux motor (illustrated in FIG. 8). The tuning of the inductance of the radial flux motor may therefore require the addition of significant active parts mass, or the selection of a sub-optimal design (e.g., the selection of long and narrow stator teeth; see FIG. 6, design (i)), which may increase the inductance but at the same time reduce the efficiency of the motor.

[0335] In a similar manner, a value of a motor parameter ?, defined in Equation (20) as the product of the efficiency and the inductance of the machine divided by its active parts mass, may be tuned to improve balance between efficiency and fault tolerance. Table 10 illustrates values of ? in accordance with the present disclosure. The values of ? are notably high.

TABLE-US-00010 TABLE 10 ? (?Hkg.sup.?1) Example 1 Example 2 Example 3 6.8 1.8 3.0

[0336] The value of ? may be selected to be greater than or equal to 1.4 ?Hkg.sup.?1, while values in the range of 2.1 to 5.5 ?Hkg.sup.?1 may provide a particularly good balance between the competing constraints. The machine inductance itself, L.sub.machine, may be relatively high, especially relative to the mass of the active parts (i.e., L.sub.machine divided by m.sub.act may be particularly high).

[0337] Another important consideration in the design of a motor for an EPU of an aircraft is the capability of the cooling system of the motor. The cooling system may be capable, at all relevant operating conditions, of removing heat from the motor at a rate sufficient to keep the motor below a rated temperature. Herein, the rated temperature may be a maximum rated temperature of the coil insulation, ?.sub.ins,max. The cooling system may significantly add to the mass of the EPU, and this additional cooling system mass (m.sub.cool) is multiplied by the number of EPUs on the aircraft. Thus, rather than designing an EPU with a high torque production capability and an aggressive cooling system, which may have a high mass, it may be desirable to consider a parameter ?, defined in Equation (12):

[00031]$\begin{array}{cc}?=\frac{{?}_{\max ,\mathrm{cont}}}{m_{act}?C_{\max ,\mathrm{cont}}}& \left(12\right)\end{array}$

[0338] In this equation, ?.sub.max,cont is the maximum continuous rated torque, and C.sub.max,cont is the heat capacity cooling rate required to maintain the coil insulation at or below its rated temperature, ?.sub.ins,cont, assuming operation at ISA sea level conditions. C.sub.max,cont may be defined as the product of the specific heat capacity of a coolant of the cooling system (at ISA sea level conditions) multiplied by the mass flow rate of the coolant required to maintain the coil insulation at or below ?.sub.ins,max. In accordance with the present disclosure, a value of ? may be selected so that the combined mass of the active parts and the cooling system may be optimized relative to the torque producing capability of the motor. Table 11 illustrates values of ? for a motor sized for an EPU of an aircraft in accordance with the present disclosure. The values of ?, which are notably high, are quoted in units of Kskg.sup.?1 (Kelvin-seconds-per-kg):

TABLE-US-00011 TABLE 11 ? (Kskg.sup.?1) Example 1 Example 2 Example 3 0.19 0.27 0.55

[0339] The value of ? may be selected to be greater than or equal to 0.1 Kskg.sup.?1, greater than or equal to 0.18 Kskg.sup.?1, or greater than or equal to 0.21 Kskg.sup.?1.

[0340] It is also useful to introduce a dimensionless figure of merit, F, for a motor of a VTOL aircraft. F is defined in Equation (22):

[00032]$\begin{array}{cc}F=\frac{{?}_{\max ,\mathrm{cont}}}{m_{act}}\frac{p_{\mathrm{air},0}}{C_p{\stackrel{.}{m}}_{\max ,\mathrm{cont}}\left({?}_{\mathrm{ins},\max }-{?}_{air,0}\right)}\frac{2??D_{ref}}{{?}_{mech,cont}}{\left(\frac{D_{ref}}{D_{act}}\right)}^2& \left(22\right)\end{array}$

[0341] In this equation, ?.sub.max,cont is the maximum continuous rated torque, m.sub.act is the active parts mass, C.sub.p is the specific heat capacity of the coolant at ISA sea level conditions, ?.sub.ins,max is the maximum rated temperature of the insulation for operation at the maximum continuous rated torque, {dot over (m)}.sub.max,cont is the mass flow rate of the coolant required to maintain the insulation at or below ?.sub.ins,max during ISA sea level operation at ?.sub.max,cont, ?.sub.mech,cont is the angular speed of rotation (in radians per second) of the rotor of the motor while producing the maximum continuous rated torque, and D.sub.act is the active parts diameter. The remaining values are fixed, nominal operational, values: p.sub.air,0 is a nominal ambient air pressure equal to 100 kPa, ?.sub.air,0 is a nominal ambient air temperature of 318 Kelvin, and D.sub.ref is a nominal motor diameter set equal to 0.5 meters.

[0342] For the purposes of comparing two motors, any value may be selected for D.sub.ref as long as the same value of D.sub.ref is used for both calculations. The value of F is decreased by using an active parts diameter, D.sub.act greater than D.sub.ref but increased by using an active parts diameter, D.sub.act, less than D.sub.ref. In other words, the equation for F penalizes the use of an arbitrarily large active parts diameter to meet the torque and speed requirements of the motor, as the use of an arbitrarily large diameter would create installation and aerodynamic drag issues. The selection of 0.5 meters for D.sub.ref reflects that 0.5 meters is a reasonable value for certain EPU designs. If calculating and comparing values of F for a smaller platform (e.g., an unmanned aerial vehicle (UAV) or drone), a smaller value of D.sub.ref may be selected (e.g., 0.1 meters). If calculating and comparing values of F for a larger platform (e.g., a larger aircraft), a higher value of D.sub.ref may be selected (e.g., 1.0 meters). Accordingly, in Equation (22), it is the value of D.sub.act, and not D.sub.ref, that characterizes the motor.

[0343] In order to provide a particularly good balance between the competing requirements of physical size (e.g., active parts diameter), mass, torque production, and cooling, electrical machines in accordance with the present disclosure may have a particularly high value of F. The first two rows of Table 12 illustrate values of F for motors in accordance with the present disclosure. For comparison, the third line of Table 12 illustrates the value of F for an exemplary radial flux motor designed for use in a CTOL aircraft having a more conventional value of F.

TABLE-US-00012 TABLE 12 F Air-cooled transverse flux motor (VTOL) 5.2 Liquid-cooled radial flux motor (VTOL) 2.3 Liquid-cooled radial flux motor (CTOL) 0.3

[0344] According to the present disclosure, the value of F may be greater than or equal to 1.9. In some examples of electric motors for EPUs of VTOL aircraft, particularly those utilizing a transverse flux arrangement, the value of may be greater than or equal to 2.5.

[0345] A motor and EPU for these applications may use an air cooling system. This is partly due to a reduction in the complexity and maintenance requirements associated with a liquid cooling system. However, a potentially more significant benefit is the reduction in the cumulated mass of the components of the cooling system, which may otherwise make a substantial contribution to the EPU mass and platform mass. For example, a liquid cooling system will include not only the mass of the liquid coolant, but may also include: the mass of the coolant tank; conduits (e.g., piping) through which the coolant flows; the mass of pumps, valves, and other fluid flow modulating components; the mass of filters; and the mass of heat exchangers. In one example, the mass of a liquid cooling system sized for a motor of a VTOL aircraft EPU is about 14 kg, representing approximately 20-25% of the overall mass of the motor. If each one of the six EPUs of the exemplary VTOL aircraft 1 of FIG. 1 had such a cooling system, the total motor cooling system mass for the entire aircraft would be about 84 kg, which is on the order of the mass of a passenger. In contrast, an example of an air-cooling system having a mass generally limited to the mass of filter components and flow directing components may only be about 3-5 kg per EPU.

[0346] While the advantages of selecting an air-cooling system may be clear, implementing an air-cooling system in a motor for an EPU of a VTOL aircraft requires more consideration. Air has a relatively low specific heat capacity compared with certain liquid coolants (e.g., 1006 Jkg.sup.?1K.sup.?1 for air, compared with 1745 Jkg.sup.?1K.sup.?1 for one oil-based coolant), and the available mass flow rate may be limited in VTOL applications due to both the low density of air compared to liquid and the relatively slow movement of the aircraft at some operating points. This may limit the rate at which heat may be removed from the motor. If the slot current density, J.sub.slot, is high, there will be high resistive losses (I.sup.2R losses) and/or a lack of free space in the slot to effectively cool the coils, which may make air-cooling impractical. If the slot current density, J.sub.slot, is too low, the motor may not be able to meet its torque production requirements.

[0347] In accordance with the present disclosure, the selection of a transverse flux motor with one or more of the optimizations described above (e.g., an optimized value of ? to access the peak of the torque curve illustrated in FIG. 15) may allow for use of air cooling in an EPU of a VTOL aircraft, particularly where the flow of cooling air is supplied to directly contact the coils. FIGS. 16 to 22 illustrate transverse flux electrical machines with air cooling systems and, more specifically, transverse flux motors with directly cooled conductors.

[0348] FIG. 16A and FIG. 16B show a motor 60 in perspective view. The motor 60 includes a rotor 62 that is coupled to a drive shaft 630 via a coupling structure 626 and rotates about an axis of rotation 63. The motor 60 further includes a bearing unit 64. Reinforcement ribs 670 are arranged circumferentially around the bearing unit 64 and the axis of rotation 63 and are fixed to the bearing unit 64 and to a base plate 672 of the bearing unit 64. The coupling structure 626 is visible in the perspective bottom view of FIG. 16B. At a radially outer region 674 of the coupling structure 626, the coupling structure 626 is connected to the rotor 62. At a radially inner region 676 of the coupling structure 626, the coupling structure 626 is connected to the EPU drive shaft 630.

[0349] FIG. 17 shows the motor 60 of FIGS. 16A and 16B in cross-section. FIG. 17 shows the same motor 60 as FIG. 11, but further shows cooling channels 602 in the stator 61 and omits the active parts for clarity. An empty volume 662 that would accommodate the active parts is labelled, as are the locations of axial air gaps 615 formed between the active parts of the stator 61 and the rotor 62. In operation, an external flow of ambient air that impinges on the EPU due to, for example, movement of the aircraft and/or wind enters the motor housing and is guided by the cooling channels 602 in a radially outward direction into the volume 662 where the active parts are located. Thus, the ambient air directly contacts and cools the active parts, including the stator coils.

[0350] For effective direct cooling, the volume 662 may not be completely filled and leaves space through which the cooling air may pass. For example, the stator coils 614 may define an effective cooling surface area that is directly exposed to air. FIGS. 18-20 illustrate an example of a stator phase module structure 610 by which the stator coils define an effective cooling surface area for direct cooling.

[0351] FIG. 18A and FIG. 18B show a stator phase module 610 mounted to its support structure 640. The phase module 610 is comparable to the one shown in FIG. 9B, which is referred to for a detailed explanation. FIG. 18A and FIG. 18B show the same embodiment; however, FIG. 18A is cut in a radial plane, enabling a view of the cross section of the first coil portion 614a and the second coil portion 614b of the coil 614, and the first slot portion 613a and the second slot portion 613b of the slot 613. FIG. 18B shows the end windings 617.

[0352] The stator according to FIG. 18A includes an assembly 680 that may be a modular (e.g., prefabricated) component. The assembly 680 extends in the radial direction (r) and in the circumferential direction (?), and includes two axially spaced, non-magnetic and non-magnetizable support structures 640i, 640ii. These have radially inner fastening areas 690, 692, at which the support structures 640i, 640ii may be connected to a plurality of ribs of the stator. This is done, for example, via retaining projections of the ribs, as will be described with reference to FIG. 21.

[0353] Flux guiding stator elements 612 extend between the support structures 640i, 640ii, and collectively provide the flux guiding stator iron 611 of the stator. The stator elements 612 define the slots 613 (i.e., the winding space) extending in the circumferential direction, in which the coil 614 extending in the circumferential direction is arranged.

[0354] According to the present example, a flow of air (e.g., the flow of external ambient air that enters the EPU and is directed by the stator cooling channels 602 of FIG. 17) flows radially through the assemblies 680 in the region between the two support structures 640i, 640ii and flows across the stator elements 612 and the coil 614. This flow of air cools the stator elements 612 and the coil 614. The stator elements 612 are each aligned radially. The stator elements 612 each have two radially aligned side surfaces 694, 696 spaced apart in the circumferential direction, both of which are cooled by a cooling air flow.

[0355] The coil 614 includes multiple individual winding turns 6140 (see FIG. 18A) that are formed from a continuous winding wire. Further, each coil portion 614a, 614b of the coil 614 includes two axially spaced winding packages: the first coil 614a portion has a first axially spaced winding package 614a-i and a second axially spaced winding package 614a-ii in the first slot portion 613a, and the second coil portion 614b has a third axially spaced winding package 614b-i and a fourth axially spaced winding package 614b-ii in the second slot portion 613b. The winding packages 614a-i, 614a-ii, 614b-i, 614b-ii each have sections extending longitudinally in the circumferential direction of the coil 614. As shown in FIG. 18B, through the deflected coil section that forms the end winding 617, the winding packages 614a-i, 614a-ii, 614b-i, 614b-ii form a coil 614. A corresponding end winding 617 is found at the other end of the coil 614.

[0356] The winding packages of each pair of winding packages 614a-i, 614a-ii, and 614b-i, 614b-ii are spaced apart in the axial direction from one another and from the support structures 640i, 640ii. In this way, cooling air may flow around the winding packages on their upper side and on their lower side. This is illustrated in FIG. 19A. The assembly 680 defines three radially extending and axially spaced cooling air flow passages A1, A2, A3 for cooling the winding packages 614a-i, 614a-ii, 614b-i, 614b-ii. A first cooling air flow passage A1 runs adjacent to the upper support structure 640i, a second cooling air flow passage A2 runs in an area between the winding packages 614a-i, 614b-i and the winding packages 614a-ii, 614b-ii, and a third cooling air flow passage A3 runs adjacent to the lower support structure 640ii. The division of the coil 614 into axially spaced winding packages 614a-i, 614a-ii, 614b-i, 614b-ii increases the surface area of the winding that is available for direct cooling. While two axially spaced windings packets per winding portion are shown in this example, more than two axially spaced winding packets may also be provided. Alternatively, it is also possible that only one winding package is arranged in each slot portion 613a, 613b.

[0357] According to FIG. 18A and FIG. 19A, two winding packets 614a-i, 614a-ii and 614b-i, 614b-ii, respectively, (or, e.g., one coil portion 614a, 614b) may each be fixed by a fixing material 6130 in the respective slot portion 613a, 613b. In the present example, the fixing material 6130 only slightly extends in the circumferential direction (e.g., being in the shape of a disk or plate) so as not to impair cooling by obstructing the cooling air flow. The fixing material may be arranged radially in front of or behind a stator element 612, so as to limit a reduction in the cross-section facing and exposed to a radial air flow.

[0358] To avoid physical contact of the coil 614 with the stator elements 612, a mechanical protective layer may also be applied to the stator elements 612 on the side facing the slot 613 (e.g., the slot portions 613a, 613b). For example, an aramid paper may be used, analogous to the use of slot papers in the slots of radial flux machines.

[0359] Referring again to FIG. 18A, the arrangement of the stator elements 612 for defining the slot portions 613a, 613b is shown. The stator elements 612 are arranged in four circumferential rows: two rows 612a-i, 612a-ii being of the first set 612a of stator elements 612 and defining the first slot portion 613a, and two rows 612b-i, 612b-ii, being of the second set 6122b of stator elements 612 and defining the second slot portion 613b.

[0360] The stator elements 612 are, like those shown in previous examples, curved and/or bent. For example, the stator elements 612 may be claw-shaped and/or curved in a C-shape. The stator elements 612 of the respective radially inner rows 612a-i, 612b-ii are concave, viewed from the radially outer side, and the stator elements of the respective radially outer rows 612a-ii, 612b-ii are convex, viewed from the radially outer side, so that their mutually facing sections together define the slot portions 613a, 613b. The stator elements 612 of each of the two rows delimit the slot portions 613a, 613b transversely to the circumferential direction. For this purpose, each stator element 612 of a given row (e.g., row 612a-i) forms a pair of stator elements with a circumferentially adjacent stator element belonging to a radially adjacent row (e.g. 612a-ii) of the same set of stator elements (e.g., 612a), and stator elements of a pair are oriented such that the stator elements of the pair oppose each other.

[0361] End portions (e.g., projections) of the stator elements 612 form pole heads (e.g., upper pole heads and lower pole heads; see, e.g., the pole heads 3122, 3123 in FIG. 7B). The end portions are positioned adjacent the permanent magnets of the rotor 62 and are separated from the permanent magnets of the rotor 62 only by an air gap (e.g., corresponding to the air gap 615 of FIG. 17). For this purpose, it is provided that the end portions or pole heads are each arranged in one of the support structures 640i, 640ii and terminate flush with their outer sides 641i, 641ii. Accordingly, the upper end portions of the stator elements 612 lie in the outer plane of the upper outer side 641i of the upper support structure 640i, as shown in FIG. 18A.

[0362] A motor includes a plurality of the assemblies 680, adjoining one another in the circumferential direction. For example, six assemblies 680 may be provided for the motor described with reference to FIGS. 9-11, with two per phase.

[0363] FIG. 19B is a schematic sectional view of an embodiment of a coil 614. The coil 614 has a continuous winding wire that is wound in a number of winding turns 6140, with each winding turn 6140 extending over an angle of 360?. A total of fourteen winding turns 6140 are provided in the exemplary embodiment considered. The coil 614 is configured such that the total of fourteen winding turns 6140 are arranged in four levels or coil layers L1, L2, L3, L4, with three winding turns 6140-1 to 6140-3 arranged in the first coil layer L1, four winding turns 6140-4 to 6140-7 arranged in the second coil layer L2, four winding turns 6140-8 to 6140-11 arranged in the third coil layer L3, and three coil turns 6140-12 to 6140-14 arranged in the fourth coil layer L4. In the embodiment shown, each coil layer L1, L2, L3, L4 is arranged in an axial plane, parallel to each other.

[0364] The winding order is indicated by the arrows 6142. From the winding sequence, it follows that in the case of the winding turns 6140 of the first coil layer L1, a turn diameter D.sub.Turn of the winding turns 6140 decreases as the number of winding turns increases. In other words, the continuous winding wire or conductor is moving inwards with every winding turn 6140 in the first coil layer L1. Thus, winding turn 6140-1 has a larger turn diameter than winding turn 6140-2, which has a larger turn diameter than winding turn 6140-3. Turn diameter, in this context, refers to the average diameter of a 360? loop of one winding turn 6140 around winding turn axis W, and not to the diameter of the wire or conductor. An example of the turn diameter D.sub.Turn is shown in FIG. 19B for the twelfth winding turn 6140-12. A winding turn with a smaller turn diameter lies radially (e.g., with respect to the winding turn axis) within an adjacent winding turn with a larger diameter. For example, winding turn 6140-2 lies within winding turn 6140-1.

[0365] In contrast, in the coil layer L2, the turn diameter of the winding turns 6140 increases with an increasing number of winding turns. For example, the winding turn 6140-5 has a larger turn diameter than the winding turn 6140-4. In the third coil level L3, the turn diameter of winding turns 6140 decreases again as the number of windings increases, and in the fourth coil level L4 the turn diameter increases again.

[0366] The described coil 614 forms winding packages 614a-i, 614a-ii, 614b-i, 614b-ii corresponding to the winding packages 614a-i, 614a-ii, 614b-i, 614b-ii of FIGS. 18A and 19A.

[0367] FIG. 20A shows an embodiment of a coil 614 with a structure according to FIGS. 19A-B in a view from above. FIG. 20A shows that the coil 614 may have a curved shape similar to that of a banana. Accordingly, the coil 614 includes longitudinally extending sections 614a, 614b that may be concavely bent. The longitudinally extending sections 614a, 614b are bent over and form deflected sections at the end windings 617. The top view of FIG. 20A shows the coil turns 6140-1, 6140-2, and 6140-3 of the first coil layer L1 of FIG. 19.

[0368] FIG. 20B schematically shows a coil 614 that corresponds to FIGS. 19A and 20A in terms of structure. The side view of FIG. 20B shows the individual coil layers L1, L2, L3, and L4 in which a plurality of winding turns 6140 are formed. Further, the coil 614 of FIG. 20B is additionally shown with fixing material 6130 in the form of fixing disks 6132 that correspond to the fixing material 6130 of FIGS. 19A and 20A, and serve to arrange and position the coil 614 in the slot 613.

[0369] The shape of the coil 614 in FIGS. 19A-19B and 20A-20B is a non-limiting example. In principle, the coil 614 (and/or the winding packages 614a-i, 614a-ii, 614b-i, 614b-ii) in the winding diagram and structure shown in FIG. 19B may have other shapes, (e.g., circular, elliptical, or with a plurality of concave and convex areas). The use of fixing disks 6132 according to FIGS. 20A, 20B is also optional.

[0370] As explained above with reference to FIG. 17, ambient air may enter a motor and be directed radially outwardly by channels 602 towards the active parts of the motor. As explained with reference to FIGS. 18A-B, 19A-B, and 20A-B, the stator phase modules may be arranged so that there are gaps (e.g., radial gaps between stator elements 612 and passages A1, A2, A3) for air flow for effective direct air cooling. FIGS. 21 and 22 illustrate this in more detail for a dual-lane transverse flux motor 80.

[0371] FIG. 21 shows an electric drive unit with the motor 80. In FIGS. 11, 16A-B, and 21, like reference numerals label like parts. The motor 80 includes a rotor 81, a stator 82, an axis of rotation 83, and a bearing unit 84. The bearing unit 84 includes an axially arranged, rotatable EPU drive shaft 830 and a static bearing part 822 that supports the EPU drive shaft 830. The coupling structure described with reference to FIGS. 16A and 16B is not shown in FIG. 21, but may be included in a corresponding manner. FIG. 21 shows a stator 82 as a ring structure with a large number of ribs 820 that adjoin one another in the circumferential direction and each form a cooling air passage 821 between the ribs 820. The active components of the stator 82 are held and positioned by the ribs 820. For this purpose, the ribs 820 have retaining projections 823.

[0372] A difference from FIGS. 11, 16A and 16B results from the fact that the motor unit of FIG. 21 includes two rotor-stator assemblies 8110, 8120, each forming a sub-machine of the dual-lane motor, which are axially stacked (e.g., arranged one behind the other in the axial direction) and are fixed to one another. Accordingly, the rotor 81 includes three axially spaced outer walls 811, 812, 814, each of which has or integrates permanent magnets 85, and two radially outer end walls 813, 815. The outer walls 811, 812, 814 and the end walls 813, 815 form two axially spaced volumes 882 of the two rotor-stator assemblies 8110, 8120, each containing the active components of the stator 82 of the respective assembly.

[0373] The permanent magnets 85 of the rotor are only shown on the right-hand side of FIG. 21 for the sake of clarity. The permanent magnets 85 are arranged on the insides of the outer walls 811, 812, 814. The air gap 615 shown in FIG. 17A runs between the permanent magnets 85 and the associated stator poles of the assembly.

[0374] FIG. 21 also shows how a cooling air flow may be provided through the cooling air passages 821 and the active components of the stator 82 arranged in the volume 882. The transverse flux machine may have a first end 8010 facing a mechanical load to be driven (e.g., a propeller) and a second end 8020 facing away from the load to be driven. In FIG. 21, the transverse flux machine forms openings 801 at its first end 8010, which enable an air flow 860 to enter the motor unit in an initially primarily axial orientation. This may be supported by a fan 891, which is, however, optional. For example, the airflow may come from a propeller driven by the EPU drive shaft 830.

[0375] The second end 8020 facing away from the load to be driven is hermetically sealed to prevent inflowing air from leaving the motor unit again in the axial direction. For this purpose, a cover plate 802 is provided, which is shown schematically. The cover plate 802 is connected to the stator 82 in FIG. 21, but may alternatively be connected to the rotor 81, or be formed by a coupling structure similar to the coupling structure 626 of FIGS. 11, 16A or the like, depending on the design.

[0376] By the cover plate 802, the inflowing air flow 860 flows radially outwards as an air flow 861 through the cooling air passages 821 and the active components of the stator arranged in the volume 882. The radial air flow 861 may also be optionally supported by fans 892.

[0377] The end walls 813, 815 of the rotor 81 are provided with radial openings 816 that enable the cooling air flow 861 to be directed into the environment. Alternatively, openings may be formed in the motor unit at the second end 8020 facing away from the load to be driven, while the first end 8010 facing the load to be driven may be sealed airtight in this case. A further alternative is that a cooling air flow is directed radially inwardly through the stator 82. For this purpose, an air flow located at the outer circumference of the rotor, which may originate from a propeller, for example, may be deflected by baffles or other deflecting mechanisms or devices, and guided through openings 816 in the walls 813, 815 of the rotor 81 into the stator 82. In this way, the radial direction of the air flow may be reversed, with the air flow going from radially outside to radially inside through the active components of the stator (e.g., that are arranged in the volume 882) and the cooling air passages 821.

[0378] FIG. 22 shows a detail cutaway and perspective view of the example motor 80 shown in FIG. 21. The first rotor-stator assembly 8110 and the second rotor-stator assembly 8120 (e.g., sub-machines) are visible. The connection between the radially inner fastening areas 690, 692 of stator assemblies 680 and retaining projections 823 of the stator 82 (e.g., integrally connected to the ribs 820, respectively) is also visible. Further, the axial air gaps 815, which are arranged between the permanent magnets 85 of the rotor 82 and each support structure 640i, 640ii, are indicated.

[0379] As noted above, for the purposes of aircraft (e.g., VTOL aircraft), an air cooling system may be used because of the associated reduction in EPU mass, complexity, and maintenance requirements. However, meeting the platform torque production requirements (e.g., with a high active parts torque density) while using air cooling may necessitate the use of direct air cooling (e.g., a cooling system in which heat is transferred directly from the coils into the cooling air, rather than via a heat exchanger). FIGS. 16 to 22 show motors 60, 80, stator modules 680, and coils 614 that incorporate spaces through which air may flow to directly cool the active parts. For example, the illustrated coils have multiple (e.g., two) winding packages to define an intermediate air passage A2 for direct cooling of axial inner regions of the coil, with additional spaces left in the axial outer regions to define axial outer air channels A1, A3.

[0380] Increasing the amount of free space in the active parts region (e.g., the empty volume 662 shown in FIG. 17) may improve the direct air cooling of the active parts, since the effective cooling surface of the coils (e.g., the percentage of the coil surface area directly exposed to the air) may be increased. However, this is balanced against the reduction in the slot current density that results from reducing the amount of conductor that is packed into the slot. Reducing the slot fill factor (e.g., packing factor) too far may necessitate an increase in the current, which may result in additional heat production (I.sup.2R) that is greater than the additional heat removal capability of the cooling system gained from the reduced slot packing.

[0381] In accordance with the present disclosure, the coil design may be selected to tune the effective cooling surface area to optimize the balance between direct air cooling efficiency and slot current density. For example, the number of winding packages, the number of turns per winding package, the arrangement of turns within a winding package, the turn cross-section, the turn radius, or any combination thereof may be adjusted to optimize the effective cooling surface area. The effective cooling surface area percentage, expressed as a percentage, is defined in Equation (29):

[00033]$\begin{array}{cc}\mathrm{Effective}\mathrm{cooling}\mathrm{surface}\mathrm{area}\%=\frac{\mathrm{Exposed}\mathrm{coil}\mathrm{surface}\mathrm{area}}{\mathrm{Total}\mathrm{coil}\mathrm{surface}\mathrm{area}}?100\%& \left(29\right)\end{array}$

[0382] In this equation, the total coil surface area is the sum of the surface areas of each winding turn of the coil. The exposed coil surface area is the sum of the areas that are exposed to the cooling flow of air for direct cooling.

[0383] FIGS. 23A and 23B illustrate how, for a given number of winding packages and turns (e.g., in this case, two winding packages each having seven turns arranged in a trapezium shape), the effective cooling surface area may be tuned by adjusting the conductor cross-section and radius.

[0384] In the first example, depicted in FIG. 23A, each turn 6140 of the winding package 614a-i has a circular cross-section. In the second example, depicted in FIG. 23B, each turn 6140 of the winding package 614a-i may have a rectangular cross-section with rounded corners. In both cases, the turns 6140 are arranged in two rows: a first row of three turns and a second row of four turns, wherein the turns are packed as closely as possible. The effective cooling surface area of the winding package 614a-i is the outer surface area of the winding package 614a-l, which is exposed to the flow of air. In both cases, the curvature of the turn cross-section results in empty central regions 6145 that are not occupied by conductor but are also inaccessible to cooling air, and thus do not form part of the effective cooling surface area.

[0385] Comparing the first and second examples, the empty central regions 6145 are larger in the example in FIG. 23A than in the example in FIG. 23B due to the greater curvature of the turns in the example in FIG. 23A. This reduces the slot current density, which reduces torque production, without offering any improvement in cooling. However, portions of the effective cooling surface areas adjacent regions A1, A2 (e.g., corresponding to the air flow passages A1, A2 in FIG. 19A) are larger in the example in FIG. 23A than in the example in FIG. 23B. Specifically, the rectangular shape of the turns in the example in FIG. 23B results in linear, planar effective cooling surfaces adjacent regions A1, A2. The result of this is that the surfaces adjacent to the air flow passages A1, A2 only expose one side of the rectangle (e.g., slightly over 25%) to the cooling air. In contrast, the curvature of the circular turns 6140 in the example in FIG. 23A results in cooling surface areas adjacent regions A1, A2 that are non-linear. This exposes more of the turn surface area (e.g., slightly under 50%) to the cooling air. Overall, the example in FIG. 23B has a slightly higher packing factor and slot current density, resulting in slightly greater torque production. However, the example in FIG. 23A has superior cooling due to the greater cooling surface area adjacent the air passages A1, A2.

[0386] In accordance with the present disclosure, the effective cooling surface area may be at least 25% of the overall surface area of the coil. Values of between 35% and 70% may strike a particularly good balance between cooling and torque production in a transverse flux motor.

[0387] For completeness, Table 13 summarizes the configuration and properties of a transverse flux electrical machine that is optimized for the use in the EPU of a VTOL aircraft. This is merely one example and does not limit the present disclosure to an electrical machine of this configuration.

TABLE-US-00013 TABLE 13 Air-cooled 150 KW Dual-Lane Transverse Flux Motor Air gap configuration Double rotor, axial air gap Lane Configuration Two lanes, axially stacked active parts Cooling system type Direct air cooling Continuous rated power (P.sub.cont) 150 KW Peak rated power (P.sub.peak) 175 KW Maximum continuous rated torque 1300 Nm Peak rated torque 1440 Nm Hover torque 975 Nm Rotor speed at maximum continuation 120 rads.sup.?1 rated torque (?.sub.max,cont) Rotor speed at hover (?.sub.hover) 96 rads.sup.?1 EPU tip speed (v.sub.tip) 171 ms.sup.?1 Efficiency (?) 0.94 (94%) Maximum continuous rated current 200 A (RMS) Peak rated current 230 A (RMS) Steady-state terminal short circuit 174 A (RMS) current Slot current density (peak) 8.1 A/(mm).sup.2 Continuous rated voltage 900 V Active parts diameter 0.46 meters Air gap 0.7 mm Active parts mass 15.2 kg Cooling system mass 4.2 kg Total motor mass 56 kg Conductor volume 38.8 cm.sup.3 Iron volume 61.1 cm.sup.3 [00034]$\mathrm{Pole}\mathrm{pair}\mathrm{number}\left(\frac{N_P}{2}\right)$ 80 [00035]$\mathrm{Pole}\mathrm{pitch}\left(P_{?}=\frac{2?}{N_P}\right)$ 0.039 radians (2 degrees) [00036]$\mathrm{Pole}\mathrm{arc}\mathrm{length}\left(P_L=\frac{P_{?}?D_{Act}}{2}\right)$ 8.97 mm Machine inductance 43 ?H Power Factor (cos (?)) 0.72 Max. electrical frequency 1.4 kHz (f.sub.max) Insulation rated temperature 475 K (?.sub.ins, max) Coolant specific heat capacity 1006 Jkg.sup.?1K.sup.?1 (C.sub.p) Required mass flow rate at 0.27 kg ?.sub.max,cont ({dot over (m)}.sub.max,cont) C.sub.max,cont = C.sub.p ? {dot over (m)}.sub.max,cont 272 JK.sup.?1 [00037]${?}_{act}=\frac{{?}_{peak}}{m_{act}}$ 94.7 Nmkg.sup.?1 [00038]${?}_{act+\mathrm{cool}}=\frac{{?}_{peak}}{m_{act}+m_{cool}}$ 74.2 Nmkg.sup.?1 [00039]$?=\frac{{?}_{peak}}{m_{act}?J_{\mathrm{slot},peak}}$ 11.7 ?Nm3kg.sup.?1A.sup.?1 [00040]${?}^{*}=\frac{{?}_{peak}}{\left(m_{act}+m_{cool}\right)?J_{slot,peak}}$ 9.3 ?Nm3kg.sup.?1A.sup.?1 [00041]$?=\frac{V_{condu\mathrm{ctor}}}{V_{\mathrm{iron}}}$ 0.64 [00042]$?=\frac{{?}_{act}}{\cos \left(?\right)}$ 132 Nmkg.sup.?1 [00043]${?}^{*}=\frac{{?}_{peak}}{\left(m_{act}+m_{\mathrm{cool}}\right)?\cos \left(?\right)}$ 103 Nmkg.sup.?1 ? = P.sub.? ? G.sub.Air 27.3 micro radian-meters ?* = P.sub.L ? G.sub.Air 6.3 ?m.sup.2 [00044]$?=\frac{P_L}{f_{\max }}$ 6.4 ?ms [00045]$?=\frac{T_{\max ,\mathrm{cont}}}{m_{act}?C_{\max ,\mathrm{cont}}}$ 0.31 Kskg.sup.?1 [00046]${?}^{*}=\frac{{?}_{\max ,\mathrm{cont}}}{\left(m_{act}+m_{cool}\right)?C_{\max ,\mathrm{cont}}}$ 0.24 Kskg.sup.?1 [00047]$Z=\frac{\cos \left(?\right)?m_{act}}{?}$ 11.6 kg [00048]$Z^{*}=\frac{\cos \left(?\right)?\left(m_{act}+m_{cool}\right)}{?}$ 14.9 kg [00049]$?=\frac{I_{\mathrm{SC}}}{I_{peak}}$ 0.76 [00050]$?=\frac{{?}_{\mathrm{ins},\mathrm{cont}}\left(I_{\mathrm{SC}}\right)}{{?}_{ins,cont}\left(I_{cont}\right)}$ 0.94 ? = L.sub.machine ? ?.sub.act 4.1 mHNkg.sup.?1 [00051]$?=\frac{??L_{machine}}{m_{act}}$ 2.7 ?Hkg.sup.?1 [00052]${?}^{*}=\frac{??L_{machine}}{\left(m_{act}+m_{cool}\right)}$ 2.1 ?Hkg.sup.?1 F (see Equation (22)) 6.2 F* (see Equation (23)) 4.9 [00053]$?=\frac{v_{tip}?m_{act}}{{?}_{peak}}$ 1.8 sm.sup.?1 [00054]${?}^{*}=\frac{v_{tip}?\left(m_{act}+m_{cool}\right)}{{?}_{peak}}$ 2.3 sm.sup.?1 [00055]$?=\frac{{?}_{h\mathrm{over}}}{{?}_{h\mathrm{over}}}$ 10.2 Nmsrad.sup.?1 Effective cooling surface area 42% Slot fill factor 28%

[0388] Various examples have been described, each of which features various combinations of features. It will be appreciated by those skilled in the art that, except where clearly mutually exclusive, any of the features may be employed separately or in combination with any other features, and the disclosure extends to and includes all combinations and sub-combinations of one or more features described herein.

[0389] It is to be understood that the elements and features recited in the appended claims may be combined in different ways to produce new claims that likewise fall within the scope of the present disclosure. Thus, whereas the dependent claims appended below depend on only a single independent or dependent claim, it is to be understood that these dependent claims may, alternatively, be made to depend in the alternative from any preceding or following claim, whether independent or dependent, and that such new combinations are to be understood as forming a part of the present specification.