Innovative and Highly Integrated Modular Electric Drivetrain

Article Innovative and Highly Integrated Modular Electric Drivetrain

Jonas Hemsen 1,*, Daniel Kieninger 1, Lutz Eckstein 1, Mathias R. Lidberg 2 , Henk Huisman 3, Juris Arrozy 3, Elena A. Lomonova 3, Daniel Oeschger 4, Charley Lanneluc 5, Olivier Tosoni 5 , Patrick Debal 6, Michael Ernstorfer 7 and Rémi Mongellaz 8

1 Institute for Automotive Engineering (ika), RWTH Aachen University, 52056 Aachen, Germany; [email protected] (D.K.); [email protected] (L.E.) 2 Department of Mechanics and Maritime Sciences, Chalmers University of Technology, 41296 Gothenburg, Sweden; [email protected] 3 Electromechanics and Power Electronics Group, Eindhoven University of Technology, 5612 AZ Eindhoven, The Netherlands; [email protected] (H.H.); [email protected] (J.A.); [email protected] (E.A.L.) 4 BRUSA Elektronik AG, 9466 Sennwald, Switzerland; [email protected] 5 Commissariat à l’énergie atomique et aux énergies alternatives–CEA, 38054 Grenoble Cedex, France; [email protected] (C.L.); [email protected] (O.T.) 6 Punch Powertrain, 3800 Sint-Truiden, Belgium; [email protected] 7 MAHLE ZG Transmissions GmbH, 85386 Eching-Dietersheim, Germany; [email protected] 8 Simulations and Test Solutions – Siemens Industry Software, S.A.S, 69006 Lyon, France; [email protected] * Correspondence: [email protected]; Tel.: +49-241-8025-690

Received: 11 October 2019; Accepted: 5 December 2019; Published: 11 December 2019

Abstract: A highly integrated electric drivetrain module with 157 kW peak power is presented, which incorporates novel technologies in the field of electric machines, power electronics and transmissions: 1. High-speed electric machine with six phases and injection mould polymer-bonded magnets; 2. High-ratio dual-speed transmission with double planetary gear set (Ravigneaux gear set); 3. Gallium nitride (GaN) power electronics with winding reconfiguration feature.The combination of these components in one single housing makes the drive module flexible to integrate and to combine with conventional or alternative propulsion technologies, thus allowing various hybrid and electric drivetrain topologies. All technologies are selected in accordance with mass production potential and can therefore have a high impact on the automotive market in the future. Currently, the drive module is under development; the first models will be assembled in winter 2019. The integration into a demonstrator vehicle in 2020 will prove the potential of many new technologies and the suitability for the automotive market.

Keywords: electric drive module; hybrid synchronous machine; gallium nitride; ravigneaux gear set; multi-phase electric machine; holistic drivetrain design; injection moulding

1. Introduction The electrification of passenger cars and light-duty vehicles—which are still responsible for around 20% of Europe’s greenhouse gas emissions [1]—will have an enormous effect on the reduction of greenhouse gas emissions. Additionally, pollutant-free air in dense urban areas is possible with electrified propulsion [2,3]. However, the maturity of electric drives needs a final push for better performance and efficiency so that battery capacity can be saved and driving range can be increased. Combined with a cost reduction, a massive adoption of such transport in Europe and worldwide can be generated [4]. This paper presents a new generation of a modular electric drive module, which can

World Electric Vehicle Journal 2019, 10, 89; doi:10.3390/wevj10040089 www.mdpi.com/journal/wevj World Electric Vehicle Journal 2019, 10, 89 2 of 16 be used for various configurations of Full and Hybrid Electric Vehicles. The focus is on cost reduction, low environmental impact, efficiency, and mass manufacturing readiness. The multiphase electric machine based on permanent magnets will be combined with the latest Gallium Nitride (GaN) power switches, advanced control with higher fault tolerance, and cooling features with reduced dimensions and a higher efficiency. It will be linked with a performant two-speed transmission, also adapting new regenerative braking strategies. The development takes industrial- and user requirements into account as well as environmental impacts through a life cycle analysis, to gear the activities towards a full vehicle design approach and the realization of each component and the complete powertrain. The project gathers nine cutting-edge partners from the fields of automotive, power electronics, transmissions and electric motors including three research centers, three Tier-1 suppliers and “Small and Medium Entrepreneurs”.

2. Results

2.1. Permanent Magnet Hybrid Synchronous Machine

2.1.1. Machine Design In order to achieve very high power densities even with low available current and voltage, sophisticated electromagnetic designs are required. Additionally, the automotive market has strong requirements, adding additional effort to the design of a motor which is safe and strong at the same time [5]. To meet these challenges, the Hybrid Synchronous Motor (HSM) was invented a few years ago, which is an evolution of the internal permanent magnet motor (IPM) [6]. It is called Hybrid, because it utilizes a combination of reluctance and electromagnetic torque; it is sometimes also referred to as a “permanent magnet assisted synchronous reluctance machine” (PMaSynRM). The torque production by these two effects is the reason why it is difficult to design this type of machine. Small changes in the geometry or materials influence both effects and therefore, can have a significant impact on performance, torque ripple, efficiency, power factor, controllability, demagnetisation and more. The main goal for the EM development was to save more than 50% magnet mass compared to existing permanent magnet synchronous machine (PMSM) with similar power, to reduce the active weight and to improve the efficiency. As a reference, the electric motor of the BMWi3 is considered, since it is a state of the art motor which has proven its matureness on the road. Its specifications together with the developed motors are shown in Table1.

Table 1. Motor speciﬁcations of developed motors and reference motor.

ModulED Motor with ModulED Motor with Reference Motor: Parameter Sintered Magnets Injected Magnets BMW i3 Peak phase current (A rms) 225 225 410 Winding type Formed Litz Formed Litz Classical Litz Number of phases 6 6 3 Nominal speed (rpm) 8600 8600 5000 Max speed (rpm) 22,500 22,500 11,400 Peak torque (Nm) 160 156 250 Peak power (kW) 157 158 131 Battery voltage (VDC) 320 320 360 Stator outer diameter (mm) 176 176 242 Stator inner diameter (mm) 118 118 180 Stack length (mm) 179 179 130 Magnet Weight (kg) 1.32 1.73 2.02 Peak power @360 VDC (kW) 177 178 131 Magnet weight per kW (g) 7.5 9.7 15.4 Magnet reduction compared to 52 40 0 reference motor (%) World Electric Vehicle Journal 2019, 10, 89 3 of 16

Even though ferrite magnets have the lowest environmental impact among the available magnets, they were not selected, mainly due to the following reasons:

- Since the current capability of the gallium nitride inverter is expected to be low, the missing torque through current must be compensated by highest power magnets. - Due to the high speeds of the rotor, the small rotor cross section does not provide sufficient space to design a mechanically stable rotor with a high mass of ferrite magnets (which would be needed in order to produce the same air gap flux as with NdFeB magnets). - Ferrite magnets of currently available grades are not sufficiently safe against demagnetization.

Many iterations were needed to satisfy all of the previously deﬁned drive requirements. The resulting electric machine is a six-pole, six-phase HSM for high speeds up to 22,500 rpm. Since power is proportional to both speed and torque, the torque requirement for the motor is relatively low. For decent accelerations from 0 to 50 km/h of a C-class vehicle with the given gear ratio of 21.65, a max. torque of only 150 Nm is required. With this gear ratio and motor torque, the vehicle can accelerate from 0 to 50 km/h in around 3 s. As the required motor torque is comparatively low, the required phase current is also low, in this case only 225 A (rms) per phase. A challenge for high-speed motors is the centrifugal forces at the rotor which require mechanical stabilization and special rotor geometries to hold the permanent magnets in place [7]. The equivalent von-Mises stress of the rotor was tested at speeds higher than rated speeds. At an over-speed of 27,000 rpm (20% higher than rated speed), the stress does not exceed the material limit Figure1b. Unfortunately, the injected magnet rotor cannot Worldbe shown Electric here,Vehicle since Journal a 2019 patent, 10, for the design is pending. 4 of 17

(a) (b)

FigureFigure 1. 1. EquivalentEquivalent (von-Mises) (von-Mises) stress stress at at (a (a) )rated rated speed speed of of 22,500 22,500 rpm rpm and and ( (bb)) over-speed over-speed of of 27,000 27,000 rpm rpm forfor the the sintered sintered magnet magnet rotor. rotor.

TheThe stator stator windings are mademade fromfrom aa specialspecial wire wire type type called called “Formed “Formed Litz Litz Wire” Wire” (FLW, (FLW, Figure Figure2b). 2b).This This wire wire type providestype provides low DC low resistance DC resistance (similar (similar to massive to massive copper bars)copper and bars) low ACand resistancelow AC resistance(similar to (similar conventional to conventional Litz wires) atLitz the wires) same timeat the [8 ].same FLW time technology [8]. FLW combined technology with combined the high-speed with themotor high-speed concept leadsmotor to concept a very low leads phase to resistancea very low and phase high resistance continuous and power, high because continuous a small power, motor becausealso has a short small wires. motor Additionally, also has short iron wires. losses Additionally, are reduced iron because losses of are the reduced low ﬂux because harmonic of the content, low fluxeven harmonic if the motor content, speed even is very if the high. motor speed is very high.

(a) (b) (c)

Figure 2. (a) Classical Litz wire; (b) Formed Litz Wire (FLW) with increased fill factor in stator slot; (c) microscopic view of FLW [8].

The project target torque of 150 Nm is surpassed from the simulations by 156 Nm at a current of 225 A (rms) and 320 VDC. Since the nominal speed will also be higher than expected (8650 rpm instead of 6600 rpm), the power at the new nominal speed rises to 141 kW. Above the nominal speed, the power does not drop, it rises again to more than 157 kW (213 hp) between 12,000 and 16,000 rpm. This is very impressive if we consider the low remanence flux density (Br) of only 0.75 T @ 25 °C, compared to sintered NdFeB magnets which are available with more than 1.4 T @ 25 °C. For this project, sintered magnets of the grade 45UH were chosen. These provide a good compromise between costs and energy density (BHmax of 374 kJ/m3) while at the same time being robust to demagnetization at high temperatures [9]. Due to the typically low temperature rating of Dy-free magnets, it is not possible to use them here with a sophisticated rotor design. To achieve high levels of power with only 320 VDC and 225 A (rms), perfect magnetisation, optimal motor control and also very good mechanical design such as low friction bearings, a stiff World Electric Vehicle Journal 2019, 10, 4 of 17

(a) (b) (c)

Figure 1. Equivalent (von-Mises) stress at (a) 22,500 rpm, (b) 27,000 rpm and (c) 29,250 rpm for the sintered magnet rotor.

The stator windings are made from a special wire type called “Formed Litz Wire” (FLW, Figure 2b). This wire type provides low DC resistance (similar to massive copper bars) and low AC resistance (similar to conventional Litz wires) at the same time [8]. FLW technology combined with the high-speed motor concept leads to a very low phase resistance and high continuous power, becauseWorld Electric a small Vehicle motor Journal 2019also, 10has, 89 short wires. Additionally, iron losses are reduced because of the4 low of 16 flux harmonic content, even if the motor speed is very high.

(a) (b) (c)

FigureFigure 2. ((aa)) Classical Classical Litz Litz wire; wire; ( (bb)) Formed Formed Litz Litz Wire Wire (FLW) (FLW) with with incr increasedeased fill ﬁll factor factor in in stator stator slot; slot; ((cc)) microscopic microscopic view of FLW [[8].8].

TheThe project project target target torque torque of of 150 150 Nm Nm is is surpassed surpassed from from the the simulations simulations by by 156 156 Nm Nm at a at current a current of 225of 225 A (rms) A (rms) and and 320 320 VDC. VDC. Since Since the the nominal nominal speed speed will will also also be be higher higher than than expected expected (8650 (8650 rpm insteadinstead of of 6600 6600 rpm), rpm), the the power power at at the the new new nomina nominall speed speed rises rises to to 141 141 kW. kW. Above Above the the nominal nominal speed, speed, thethe power power does does not not drop, drop, it it rises rises again again to to more more th thanan 157 157 kW kW (213 (213 hp) hp) between between 12,000 12,000 and and 16,000 16,000 rpm. rpm. This is very impressive if we consider the low remanence flux density (B ) of only 0.75 T @ 25 C, This is very impressive if we consider the low remanence flux density (Brr) of only 0.75 T @ 25 °C,◦ comparedcompared to to sintered sintered NdFeB NdFeB magnets magnets which which are are avai availablelable with with more more than than 1.4 1.4 T T @ 25 25 °C.◦C. For For this this project, sintered magnets of the grade 45UH 45UH were were chosen. chosen. These These provide provide a a good good compromise compromise between between 3 costs and energy density (BH of 374 kJ/m3 ) while at the same time being robust to demagnetization costs and energy density (BHmax of 374 kJ/m ) while at the same time being robust to demagnetization atat high temperatures [9]. [9]. Due Due to to the the typically typically low low temperature temperature rating rating of of Dy-free Dy-free magnets, magnets, it it is is not not possibleWorld Electric to use Vehicle them Journal here 2019 with , 10, a sophisticated rotor design. 5 of 17 To achieve high levels of power with only 320 VDC and 225 A (rms), perfect magnetisation, optimalTo motorachieve control high levels and also of power very good with mechanical only 320 VDC design and such 225 A as (rms), low friction perfect bearings, magnetisation, a stiff shaftoptimal and motor many control other intelligent and also featuresvery good are mechanical needed. Furthermore, design such not as only low afriction strong andbearings, light motora stiff isshaft strived and for;many it other is no lessintelligent important features to have are needed a good. eFurthermore,fficiency. Simulations not only a showedstrong and that light with motor this high-speedis strived for; drive, it is no a peak less important efficiency ofto morehave a than good 97% efficiency. can berealized Simulations (Figure showed3). However, that with the this key high- to savingspeed drive, energy a andpeak battery efficiency size of is more mainly than not 97% peak can effi beciency, realized but (Figure a high drive 3). However, cycle effi theciency. key Sinceto saving the majorityenergy and of drive battery cycles size demand is mainly only not low peak power, efficiency, the final but HSM a high design drive was cycle selected efficiency. with aSince specific the focusmajority on of low drive load cycles efficiency. demand This only helps low to power, improve the thefinal overall HSM design energy was consumption selected with and a tospecific save batteryfocus on capacity. low load efficiency. This helps to improve the overall energy consumption and to save

FigureFigure 3.3. Simulated eefficiencyﬃciency mapmap ofof aa HybridHybrid SynchronousSynchronous MotorMotor (HSM).(HSM).

2.1.2. Magnet Injection Moulding Injection moulding of polymer-bonded magnets in the rotor cavities instead of insertion of sintered magnets was identified as one of the possible ways to further reduce the content of critical raw materials in the motor. Unfortunately, bonded magnets have less remanence flux density (Br) than sintered magnets but they contain no heavy rare-earth (Dy and Tb which are more critical than Nd and whose price might raise more severely if a supply crisis were to occur) [10]. They also allow complete freedom in the shape of the magnets, yielding some magnetic optimization potential of the rotor. Furthermore, all 66 magnets of a rotor slice can be injected in one single step, which lowers the production effort. As the mechanical strength of bonded magnets is lower than of sintered magnets (flexural strength 52 MPa vs. 285 MPa [9,10]), centrifugal forces are a serious issue. Hence, a mechanically stable rotor geometry was developed for which the calculated power and efficiency of this injected magnet rotor are comparable to the sintered magnet one. Magnetostatic finite-elements calculations were performed in order to design a magnetizer tool to produce the proper magnetic field in the rotor cavities during the injection process. The principle of magnetization of the permanent magnet material is illustrated in Figure 4a. A particular effort had to be made in order to comply magnetic requirements with mechanical constraints in the design of the mold: the magnetic field needs to be produced as homogeneously as possible in all the cavities simultaneously because injection must occur in all the magnetic poles at once in order to preserve the rotor integrity. The rotor also needs to be tightly maintained from all sides during the injection, but has to be released for extraction. A complete design of the mould has been issued (Figure 4b) and is now being realized. The main dimensions of the tool are 650 × 500 × 354 mm for a rotor segment with only 117 mm diameter and 30 mm height. First injection experiments in rotors are planned for the beginning of 2020. World Electric Vehicle Journal 2019, 10, 89 5 of 16

2.1.2. Magnet Injection Moulding Injection moulding of polymer-bonded magnets in the rotor cavities instead of insertion of sintered magnets was identified as one of the possible ways to further reduce the content of critical raw materials in the motor. Unfortunately, bonded magnets have less remanence flux density (Br) than sintered magnets but they contain no heavy rare-earth (Dy and Tb which are more critical than Nd and whose price might raise more severely if a supply crisis were to occur) [10]. They also allow complete freedom in the shape of the magnets, yielding some magnetic optimization potential of the rotor. Furthermore, all 66 magnets of a rotor slice can be injected in one single step, which lowers the production effort. As the mechanical strength of bonded magnets is lower than of sintered magnets (flexural strength 52 MPa vs. 285 MPa [9,10]), centrifugal forces are a serious issue. Hence, a mechanically stable rotor geometry was developed for which the calculated power and efficiency of this injected magnet rotor are comparable to the sintered magnet one. Magnetostatic finite-elements calculations were performed in order to design a magnetizer tool to produce the proper magnetic field in the rotor cavities during the injection process. The principle of magnetization of the permanent magnet material is illustrated in Figure4a. A particular e ffort had to be made in order to comply magnetic requirements with mechanical constraints in the design of the mold: the magnetic field needs to be produced as homogeneously as possible in all the cavities simultaneously because injection must occur in all the magnetic poles at once in order to preserve the rotor integrity. The rotor also needs to be tightly maintained from all sides during the injection, but has to be released for extraction. A complete design of the mould has been issued (Figure4b) and is now being realized. The main dimensions of the tool are 650 500 354 × × mm for a rotor segment with only 117 mm diameter and 30 mm height. First injection experiments in rotorsWorld Electric are planned Vehicle Journal for the2019 beginning, 10, of 2020. 6 of 17

(a) (b)

Figure 4.4.( a(a)) Principle Principle of magnetizationof magnetization of permanent of permanent magnets; magnets; (b) Moulding (b) Moulding tool for tool one rotorfor one segment. rotor segment. With regards to materials, several grades of composites are available; the most suited one among aboutWith 10 available regards to grades materials, has been several identified grades and of composites injected to are parallelepipedic available; the samplesmost suited (Figure one5 among) with anabout existing 10 available mold. Criteriagrades has for been the choice identified of the and 5SP-12ME injected gradeto parallel wereepipedic the following samples threefold: (Figure 5) binder with hasan existing to be stable mold. at Criteria 150 ◦C (PPS for the better choice than of PA), the 5SP-12ME the remanence grade has were to be the high following (mixture threefold: of SmFeN binder and NdFeBhas to be better stable than at 150 pure °C NdFeB),(PPS better and than the coercivityPA), the remanence has to be suhasffi tocient be high (MF18P (mixture powder of SmFeN better than and MF15P)NdFeB better [10]. This, than combined pure NdFeB), with someand the rheological coercivity studies, has to gavebe sufficient some useful (MF18P hints powder regarding better injection than parametersMF15P) [10]. to This, achieve combined proper fillingwith some of the rheological complex rotor studies, cavities. gave The some dependence useful hints of the regarding magnet strengthinjection onparameters the field intensityto achieve during proper injection filling of was the investigated complex rotor experimentally cavities. The independence order to set of tothe a propermagnet field strength intensity on the target field in intensity the mold, during allowing injection to set the was dimensions investigated of the experimentally magnetizing coilsin order and into particular,set to a proper the cooling field intensity system. target in the mold, allowing to set the dimensions of the magnetizing coils and in particular, the cooling system.

Figure 5. Injected magnet samples.

2.2. GaN Inverter

2.2.1. Advantages of Gallium Nitride Devices The low-voltage Gallium Nitride (GaN) power device technology (650 V) provides a lot of advantages compared to its Silicon (Si) and Silicon Carbide (SiC) competitors [11,12]. In terms of performance, the Gallium Nitride devices have • an unrivalled switching performance with switching on/off in less than 10 ns, • a low package inductance, • a low on-resistance • and reduced switching losses: five times better than Si devices and two times better than SiC devices.

With these advantages, the switching frequency can be raised and hence, the size of passive components can be reduced: the inverter becomes smaller and more efficient at the same time. World Electric Vehicle Journal 2019, 10, 6 of 17

(a) (b)

Figure 4. (a) Principle of magnetization of permanent magnets; (b) Moulding tool for one rotor segment.

With regards to materials, several grades of composites are available; the most suited one among about 10 available grades has been identified and injected to parallelepipedic samples (Figure 5) with an existing mold. Criteria for the choice of the 5SP-12ME grade were the following threefold: binder has to be stable at 150 °C (PPS better than PA), the remanence has to be high (mixture of SmFeN and NdFeB better than pure NdFeB), and the coercivity has to be sufficient (MF18P powder better than MF15P) [10]. This, combined with some rheological studies, gave some useful hints regarding injection parameters to achieve proper filling of the complex rotor cavities. The dependence of the magnet strength on the field intensity during injection was investigated experimentally in order to World Electric Vehicle Journal 2019, 10, 89 6 of 16 set to a proper field intensity target in the mold, allowing to set the dimensions of the magnetizing coils and in particular, the cooling system.

Figure 5. Injected magnet samples.

2.2. GaN Inverter 2.2.1. Advantages of Gallium Nitride Devices 2.2.1. Advantages of Gallium Nitride Devices The low-voltage Gallium Nitride (GaN) power device technology (650 V) provides a lot of The low-voltage Gallium Nitride (GaN) power device technology (650 V) provides a lot of advantages compared to its Silicon (Si) and Silicon Carbide (SiC) competitors [11,12]. In terms of advantages compared to its Silicon (Si) and Silicon Carbide (SiC) competitors [11,12]. In terms of performance, the Gallium Nitride devices have performance, the Gallium Nitride devices have • an unrivalled switching performance with switching on/oﬀ in less than 10 ns, • an unrivalled switching performance with switching on/off in less than 10 ns, • aa low low package package inductance, inductance, • • aa low low on-resistance on-resistance • • andand reduced reduced switching switching losses: losses: five ﬁve times times better better than than Si devices Si devices and and two two times times better better than than SiC • devices.SiC devices.

WorldWith Electric these Vehicle advantages, Journal 2019, 10 the, switching frequency can be raised and hence, the size of passive7 of 17 components can be reduced: the inverter becomes smaller and more eefficientﬃcient atat thethe samesame time.time. TheThe GaNGaN technologytechnology isis youngyoung andand underunder ongoingongoing development.development. AsAs anan example,example, thethe currentcurrent capacitycapacity perper diedie isis stillstill low:low: 6060 AA inin 20172017 andand 120120 AA inin 2019.2019. In the framework of this research project,project, inin orderorder to fulﬁlto fulfil power power requirements, requirements, extensive extensive work was work done was in developing done in adeveloping reliable parallelization a reliable ofparallelization dies. of dies.

2.2.2.2.2.2. GaNGaN PowerPower BoardBoard ConceptionConception InIn orderorder toto obtainobtain thethe requiredrequired torquetorque ofof 150150 NmNm inin thethe six-phasesix-phase high-speedhigh-speed motor,motor, aa currentcurrent ofof 225225 AA (rms)(rms) perper phasephase isis requested.requested. EachEach phasephase isis independentlyindependently connectedconnected toto aa fullfull bridgebridge inin whichwhich eacheach ofof twotwo switchesswitches isis realizedrealized withwith fourfour parallelizedparallelized GaNGaN devices,devices, speciﬁedspecified inin [11[11].]. TheThe topologytopology ofof thethe fullfull bridgebridge isis depicted depicted in in Figure Figure6 a6a with with each each switch switch (K (K1,2,3,41,2,3,4),), realizedrealized byby fourfour GaNGaN devicesdevices inin parallel.parallel. AA ﬁrstfirst powerpower boardboard prototypeprototype ofof oneone legleg ofof thethe bridgebridge is is shown shown in in Figure Figure6 b.6b.

(a) (b)

FigureFigure 6.6. ((aa)) TopologyTopology ofof thethe powerpower boardboard forfor oneone of of six six phases. phases. ((bb)) First First prototype prototype of of one one leg. leg.

Hardware development started with the parallelization of two devices. The experimental waveforms in Figure 7 show the results for two parallelized GaN devices with V 300 V at a double pulse test with an inductive load ( L 330 uH, 𝑖 200 A ). As shown, the switching of the two devices is synchronous, which is essential to maintain an equal current in the two of them. Moreover, oscillations on the DC bus at a frequency of approximately 1.2 MHz can be observed, which were induced by an exchange of reactive energy between the required decoupling capacitors.

Figure 7. Measured waveforms for two parallelized GaN devices.

The oscillations can be explained by the characterization of the circuit using a Bode plot shown in Figure 8. A resonance can be observed at 1.2 MHz, matching exactly the frequency of the oscillations. These oscillations were subsequently reduced by a better dimensioning of the decoupling and DC bus capacitors. World Electric Vehicle Journal 2019, 10, 7 of 17

The GaN technology is young and under ongoing development. As an example, the current capacity per die is still low: 60 A in 2017 and 120 A in 2019. In the framework of this research project, in order to fulfil power requirements, extensive work was done in developing a reliable parallelization of dies.

2.2.2. GaN Power Board Conception In order to obtain the required torque of 150 Nm in the six-phase high-speed motor, a current of 225 A (rms) per phase is requested. Each phase is independently connected to a full bridge in which each of two switches is realized with four parallelized GaN devices, specified in [11]. The topology of the full bridge is depicted in Figure 6a with each switch (K1,2,3,4), realized by four GaN devices in parallel. A first power board prototype of one leg of the bridge is shown in Figure 6b.

World Electric Vehicle Journal 2019(a), 10, 89 (b) 7 of 16

Figure 6. (a) Topology of the power board for one of six phases. (b) First prototype of one leg. Hardware development started with the parallelization of two devices. The experimental waveformsHardware in Figure development7 show the started results forwith two the parallelized parallelization GaN devices of two with devices.VDC =The300 experimentalV at a double pulsewaveforms test with in Figure an inductive 7 show loadthe results (L = 330 foruH two, i satparallelized= 200 A). GaN As shown,devices the with switching V 300 of the V twoat a devicesdouble pulse is synchronous, test with an which inductive is essential load ( to Lmaintain 330 uH, an 𝑖 equal 200 current A ). As in shown, the two the of them.switching Moreover, of the oscillationstwo deviceson is thesynchronous, DC bus at which a frequency is essential of approximately to maintain 1.2an MHzequal can current be observed, in the two which of them. were inducedMoreover, by oscillations an exchange on of the reactive DC bus energy at a betweenfrequency the of required approximately decoupling 1.2 capacitors.MHz can be observed, which

Figure 7. Measured waveforms for two parallelized GaN devices. Figure 7. Measured waveforms for two parallelized GaN devices. The oscillations can be explained by the characterization of the circuit using a Bode plot shown in FigureThe8. Aoscillations resonance can can be be explained observed atby 1.2 the MHz, characte matchingrization exactly of the thecircuit frequency using a of Bode the oscillations.plot shown WorldThesein Figure Electric oscillations 8.Vehicle A resonanceJournal were 2019 subsequently, 10can, be observed reduced at by 1. a2 better MHz, dimensioningmatching exactly of the the decoupling frequency and of8 of DCthe 17 busoscillations.World capacitors. Electric Vehicle These Journal osci 2019llations, 10, were subsequently reduced by a better dimensioning of8 ofthe 17 decoupling and DC bus capacitors.

Figure 8. Bode diagram of the power circuit. Figure 8. Bode diagram of the power circuit. Figure 8. Bode diagram of the power circuit. WhileWhile the projectproject progressed,progressed, higher higher power power GaN GaN devices devices became became available available and and 650 650 V, 120 V, A120 GaN A GaNdevices devicesWhile could the could be project used be used forprogressed, the for secondthe second higher generation genera powertion power Ga powerN devices boards. boards. became The The power poweravailable board board and was was 650 designed designed V, 120 asA asshownGaN shown devices in Figurein Figure could9a,b. 9a,b.be The used The topology for topology the ofsecond the of boardthe genera board istion almost is poweralmost the same boards.the same as onThe as the power on prototype; the board prototype; onlywas designed anonly extra an extrapoweras shown power device in device Figure was added, was 9a,b. added, The which topology which acts as acts of series theas series switchboard sw is (functionalityitch almost (functionality the same explained as explained on inthe Section prototype; in Section 2.2.3 ). only2.2.3). an extra power device was added, which acts as series switch (functionality explained in Section 2.2.3).

(a) (b) (c) (a) (b) (c) FigureFigure 9. 9. ((aa)) Top Top side side of of power power PC PCBB with with GaN GaN device; device; ( (bb)) bottom bottom side; side; ( (cc)) electrical electrical circuit circuit of of one one leg. leg. Figure 9. (a) Top side of power PCB with GaN device; (b) bottom side; (c) electrical circuit of one leg. 2.2.3. Series-independent Winding Reconfiguration 2.2.3. Series-independent Winding Reconfiguration In this section, a six-winding EM is used to explain the winding reconfiguration principle. Its windingsIn this are section, independent a six-winding of each other,EM is meaningused to ex thatplain there the iswinding no star reconfig point, whichuration enables principle. a free Its choicewindings of the are machine independent winding of eachconfiguration. other, meaning Furthermore, that there a fault is no tolerant star point, operation which (with enables reduced a free power)choice ofis possiblethe machine [13]. winding Figure 10 configuration. shows the implem Furthermore,entation afor fault one tolerant of the three operation phases, (with i.e., reducedfor two windings.power) is possible [13]. Figure 10 shows the implementation for one of the three phases, i.e., for two windings.

Winding 1 A B Winding 2 C D Supply Winding 1 A Series switch 1 B Winding 2 C Series switch 2D Supply Series switch 1 Series switch 2

Figure 10. Switch matrix for one phase (winding pair). Figure 10. Switch matrix for one phase (winding pair). The two windings are connected to the supply by means of a switch matrix, comprising two full bridgesThe and two two windings series switches, are connected depicted to the as contactors,supply by meansand each of amachine switch matrix, winding comprising is represented two fullby abridges single inductor. and two seriesThe matrix switches, is able depicted to interconnect as contactors, the windings and each in machine series or winding to connect is represented each winding by independentlya single inductor. to the The supply. matrix Inis ablethe se tories interconnect configurations, the windings either series in series switch or to 1 connector 2 is closed. each winding Hence, theindependently current in the to theseries-connected supply. In the windingsseries configurations, can be controlled either seriesby means switch of 1two or 2 phase is closed. legs Hence, only, whichthe current is advantageous in the series-connected with respect towindings switching can losses be controlled in the phase by legs. means of two phase legs only, which is advantageous with respect to switching losses in the phase legs. World Electric Vehicle Journal 2019, 10, 8 of 17

Figure 8. Bode diagram of the power circuit.

While the project progressed, higher power GaN devices became available and 650 V, 120 A GaN devices could be used for the second generation power boards. The power board was designed as shown in Figure 9a,b. The topology of the board is almost the same as on the prototype; only an extra power device was added, which acts as series switch (functionality explained in Section 2.2.3).

World Electric Vehicle(a) Journal 2019, 10, 89 (b) (c) 8 of 16 Figure 9. (a) Top side of power PCB with GaN device; (b) bottom side; (c) electrical circuit of one leg. 2.2.3. Series-independent Winding Reconfiguration 2.2.3. Series-independent Winding Reconfiguration In this section, a six-winding EM is used to explain the winding reconfiguration principle. In this section, a six-winding EM is used to explain the winding reconfiguration principle. Its Its windings are independent of each other, meaning that there is no star point, which enables a free windings are independent of each other, meaning that there is no star point, which enables a free choice of the machine winding configuration. Furthermore, a fault tolerant operation (with reduced choice of the machine winding configuration. Furthermore, a fault tolerant operation (with reduced power) is possible [13]. Figure 10 shows the implementation for one of the three phases, i.e., for power) is possible [13]. Figure 10 shows the implementation for one of the three phases, i.e., for two two windings. windings.

Winding 1 A B Winding 2 C D Supply Series switch 1 Series switch 2

Figure 10. Switch matrix for one phase (winding pair). Figure 10. Switch matrix for one phase (winding pair). The two windings are connected to the supply by means of a switch matrix, comprising two full The two windings are connected to the supply by means of a switch matrix, comprising two full bridges and two series switches, depicted as contactors, and each machine winding is represented by a bridges and two series switches, depicted as contactors, and each machine winding is represented by single inductor. The matrix is able to interconnect the windings in series or to connect each winding a single inductor. The matrix is able to interconnect the windings in series or to connect each winding independently to the supply. In the series configurations, either series switch 1 or 2 is closed. Hence, independently to the supply. In the series configurations, either series switch 1 or 2 is closed. Hence, the current in the series-connected windings can be controlled by means of two phase legs only, which the current in the series-connected windings can be controlled by means of two phase legs only, is advantageous with respect to switching losses in the phase legs. which is advantageous with respect to switching losses in the phase legs. In the independent configurations, the winding currents are controlled individually. Nevertheless, due to the symmetric construction of the motor, both currents will, in principle, be controlled to the same value. In a first implementation of this concept [14], a single series switch was used, and the reconfiguration was applied depending on the actual speed of the motor. However, this way of operating has to include a certain margin for varying supply voltage, stator impedance, and magnetic field strength in the machine. Additionally, once the independent configuration is used, it applies during a complete cycle of the voltage wave of the machine phase, although in the sections close to a zero crossing, the series configuration could still be used. Finally, with only a single series switch available, the loading of the four phase legs is asymmetric; as always, the same two legs were disabled in the series operation. Thus, in order to spread the losses more evenly over the phase legs, a second series switch was introduced [15]. Therefore, in the present work, it was decided to extend the reconfiguration concept: when possible, a series configuration was used, and only where necessary (starting at the peak of the voltage wave applied to a machine phase), the windings were controlled independently. Moreover, in order to spread the losses more evenly over the phase legs, a second series switch was introduced, as shown in Figure 10. As an ultimate refinement, the PWM patterns in the independent configuration were shifted 180 degrees with respect to each other, thereby minimizing the net torque ripple in the machine. Due to the reconfiguration, the system order changes during operation: in the series mode, only one current per phase needs to be controlled, but in the independent mode, two currents have to be regulated. The resulting control method for one phase is depicted in Figure 11. As shown, the switching on/off of the Difference Current controller (by definition zero in the series configuration) is governed by a binary signal derived from the output of the Sum Current controller (the desired winding current): once the demanded voltage for the series connection becomes too large in view of the DC link voltage, the configuration is changed to independent, the Difference Current controller is enabled, and the controller gain halved in order to avoid a transient in its output due to the changed load configuration. World Electric Vehicle Journal 2019, 10, 9 of 17

In the independent configurations, the winding currents are controlled individually. Nevertheless, due to the symmetric construction of the motor, both currents will, in principle, be controlled to the same value. In a first implementation of this concept [14], a single series switch was used, and the reconfiguration was applied depending on the actual speed of the motor. However, this way of operating has to include a certain margin for varying supply voltage, stator impedance, and magnetic field strength in the machine. Additionally, once the independent configuration is used, it applies during a complete cycle of the voltage wave of the machine phase, although in the sections close to a zero crossing, the series configuration could still be used. Finally, with only a single series switch available, the loading of the four phase legs is asymmetric; as always, the same two legs were disabled in the series operation. Thus, in order to spread the losses more evenly over the phase legs, a second series switch was introduced [15]. Therefore, in the present work, it was decided to extend the reconfiguration concept: when possible, a series configuration was used, and only where necessary (starting at the peak of the voltage wave applied to a machine phase), the windings were controlled independently. Moreover, in order to spread the losses more evenly over the phase legs, a second series switch was introduced, as shown in Figure 10. As an ultimate refinement, the PWM patterns in the independent configuration were shifted 180 degrees with respect to each other, thereby minimizing the net torque ripple in the machine. Due to the reconfiguration, the system order changes during operation: in the series mode, only one current per phase needs to be controlled, but in the independent mode, two currents have to be regulated. The resulting control method for one phase is depicted in Figure 11. As shown, the switching on/off of the Difference Current controller (by definition zero in the series configuration) is governed by a binary signal derived from the output of the Sum Current controller (the desired winding current): once the demanded voltage for the series connection becomes too large in view of the DC link voltage, the configuration is changed to independent, the Difference Current controller Worldis enabled, Electric and Vehicle the Journal controller2019, 10 ,gain 89 halved in order to avoid a transient in its output due to the changed9 of 16 load configuration.

Figure 11. Control system for one phase (winding pair). Figure 11. Control system for one phase (winding pair). To control all three phases, the sum currents were transformed to the d-q-0 frame by means of the ParkTo transformation. control all three For phases, brevity, the this sum approach, currents which were is transformed quite standard to forthe thed-q-0 control frame of by three-phase means of themachines, Park transformation. is not detailed For further brevity, in this this paper. approach, which is quite standard for the control of three- phase machines, is not detailed further in this paper. 2.2.4. Simulation Results of Winding Reconﬁguration WorldWorld Electric Electric Vehicle Vehicle Journal Journal 2019 2019, ,10 10, , 1010 ofof 1717 2.2.4. Simulation Results of Winding Reconfiguration A ﬁrst result of the control concept discussed above is shown in Figure 12. A first result of the control concept discussed above is shown in Figure 12.

Figure 12. Simulation result for one phase when generating a sine-wave voltage. FigureFigure 12.12. SimulationSimulation resultresult forfor oneone phasephase whenwhen generatinggenerating aa sine-wavesine-wave voltage.voltage. Here, the circuit is generating a sine wave voltage across one winding pair. As shown, once the Here, the circuit is generating a sine wave voltage across one winding pair. As shown, once the modulationHere, the index circuit for theis generating series connection a sine wave reaches volt 1 orage 1,across the configuration one winding is pair. changed As shown, to independent once the modulation index for the series connection reaches− 1 or −1, the configuration is changed to andmodulation correspondingly, index for the the modulation series connection index used reaches to generate 1 or − the1, PWMthe configuration signals is halved. is changed Once theto independent and correspondingly, the modulation index used to generate the PWM signals is halved. resultingindependent index and drops correspondingly, below 0.5, the the series modulation configuration index is used restored. to generate In the implementation, the PWM signals a hysteresisis halved. Once the resulting index drops below 0.5, the series configuration is restored. In the implementation, betweenOnce the theresulting value ofindex the modulationdrops below index 0.5, the that series triggers configuration the series is and restored. independent In the implementation, operation can be a hysteresis between the value of the modulation index that triggers the series and independent applieda hysteresis to avoid between ringing the between value of series the modulation and independent index operationthat triggers command the series in the and control independent system. operation can be applied to avoid ringing between series and independent operation command in Theoperation shape can of the be individualapplied to currentsavoid ringing in a similar between situation series isand shown independent in Figure operation 13. It can command be seen that in the control system. The shape of the individual currents in a similar situation is shown in Figure 13. eachthe control current system. follows The the shape set-point of the and individual the ripple ofcurrents the sum in currenta similar remains situation low is in shown both the in Figure series and 13. It can be seen that each current follows the set-point and the ripple of the sum current remains low independentIt can be seen configurations. that each current follows the set-point and the ripple of the sum current remains low inin bothboth thethe seriesseries andand independentindependent configurations.configurations.

((aa))

((bb))

FigureFigure 13.13.13.Simulation SimulationSimulation result resultresult for for onefor one phaseone phasephase when when generatingwhen gegeneratingnerating a sine-wave aa sine-wavesine-wave voltage and voltagevoltage current. andand Individual current.current. currentsIndividualIndividual (a ) currentscurrents and set point ((aa)) andand and setset actual pointpoint sum andand actual currentactual sumsum (b) Horizontalcurrentcurrent ((bb)) scale:HorizontalHorizontal (s), vertical: scale:scale: (s),(s), (A). vertical:vertical: (A).(A).

2.2.5.2.2.5. ExperimentalExperimental ResultsResults ofof WindingWinding Reconfiguration.Reconfiguration. ToTo bebe ableable toto alreadyalready testtest thethe windingwinding reconfreconfigurationiguration conceptconcept andand thethe machinemachine controlcontrol inin thethe absenceabsence ofof thethe finalfinal machinemachine andand powerpower electronicselectronics hardware,hardware, anan experimentalexperimental setup,setup, shownshown inin FigureFigure 14,14, waswas configured.configured. InIn thethe setup,setup, existingexisting siliconsilicon IGBTIGBT powerpower modulesmodules werewere re-usedre-used toto supplysupply anan availableavailable six-phasesix-phase openopen windingwinding surfacesurface mountedmounted permanentpermanent magnetmagnet synchronoussynchronous machinemachine (PMSM)(PMSM) loadedloaded byby aa DCDC machine.machine. TheThe circuitcircuit configconfigurationuration isis similarsimilar toto thethe oneone inin FigureFigure 10,10, butbut extendedextended forfor aa fullfull doubledouble three-phasethree-phase setup.setup. However,However, duedue toto limitationslimitations ofof thethe availableavailable hardware,hardware, onlyonly oneone seriesseries switchswitch isis presentpresent onon eacheach phasephase insteadinstead ofof two.two. TheThe powerpower ratingrating ofof thethe availableavailable PMSMPMSM (4(4 kW),kW), asas inin [16],[16], isis substantiallysubstantially lowerlower thanthan thethe finalfinal target,target, butbut thisthis setupsetup stillstill allowsallows forfor realisticrealistic testingtesting ofof thethe devdevelopedeloped algorithms,algorithms, andand asas fafarr asas thethe electronicelectronic modulesmodules areare concerned,concerned, full-full- powerpower testingtesting willwill bebe possiblepossible onceonce thethe finalfinal SPMSMSPMSM becomesbecomes available.available. World Electric Vehicle Journal 2019, 10, 89 10 of 16

2.2.5. Experimental Results of Winding Reconfiguration. To be able to already test the winding reconfiguration concept and the machine control in the absence of the final machine and power electronics hardware, an experimental setup, shown in Figure 14, was configured. In the setup, existing silicon IGBT power modules were re-used to supply an available six-phase open winding surface mounted permanent magnet synchronous machine (PMSM) loaded by a DC machine. The circuit configuration is similar to the one in Figure 10, but extended for a full double three-phase setup. However, due to limitations of the available hardware, only one series switch is present on each phase instead of two. The power rating of the available PMSM (4 kW), as in [16], is substantially lower than the final target, but this setup still allows for realistic testing of the developed algorithms, and as far as the electronic modules are concerned, full-power testing will World Electric Vehicle Journal 2019, 10, 11 of 17 beWorld possible Electric Vehicle once theJournal final 2019 SPMSM, 10, becomes available. 11 of 17

GateGate DriverDriver PowerPower ModuleModule SPMSMSPMSM DC Machine TorqueTorque SensorSensor DC Machine

Figure 14. Prototype setup for the windings reconfiguration concept verification. Three power converter Figure 14. Prototype setup for the windings reconfiguration concept verification. Three power modules,Figure 14. each Prototype for one phase,setup andfor the mechanicalwindings reconfig setup. uration concept verification. Three power converter modules, each for one phase, and the mechanical setup. converter modules, each for one phase, and the mechanical setup. Figure 15a shows the SPMSM in the prototype setup being accelerated from 50 rpm to 90 rpm. Figure 15a shows the SPMSM in the prototype setup being accelerated from 50 rpm to 90 rpm. As a result,Figure the15a circuit shows that the previouslySPMSM in runsthe prototype in all-time setup series being operation accelerated enters from the series-independent 50 rpm to 90 rpm. As a result, the circuit that previously runs in all-time series operation enters the series-independent operationAs a result, (see the Figure circuit 15 thatb). Thepreviously modulation runs indexin all-time for each series phase operation in the series-independententers the series-independent operation operation (see Figure 15b). The modulation index for each phase in the series-independent operation isoperation further detailed (see Figure in Figure 15b). The 16. modulation The initial experimental index for each results phase verify in the the series-independent feasibility of the windingsoperation is further detailed in Figure 16. The initial experimental results verify the feasibility of the windings reconfigurationis further detailed concept in Figure and thus16. The open initial upthe experimental possibility results of the final verify setup the tofeasibility be implemented of the windings with it. reconfiguration concept and thus open up the possibility of the final setup to be implemented with However,reconfiguration since inconcept the planned and thus final open setup up two the series possibility switches of the will final be used, setup further to be implemented investigation with into it. However, since in the planned final setup two series switches will be used, further investigation theit. However, most optimal since way in the to utilize planned the final two seriessetup switchestwo series will switches be done will in thebe used, future. further investigation into the most optimal way to utilize the two series switches will be done in the future.

(a) (b) (a) (b) Figure 15. (a) Acceleration of the permanent magnet synchronous machine (PMSM) from 50 rpm to FigureFigure 15.15. ((aa)) AccelerationAcceleration of of the the permanent permanent magnet magnet synchronous synchronous machine machine (PMSM) (PMSM) from from 50 rpm50 rpm to 90to 90 rpm; (b) the resulting modulation index during acceleration. rpm;90 rpm; (b) ( theb) the resulting resulting modulation modulation index index during during acceleration. acceleration.

Figure 16. Detailed measured modulation index for each phase in series-independent operation. Figure 16. Detailed measured modulation index for each phase in series-independent operation.

World Electric Vehicle Journal 2019, 10, 11 of 17

Gate Driver Power Module SPMSM

Torque Sensor DC Machine

Figure 14. Prototype setup for the windings reconfiguration concept verification. Three power converter modules, each for one phase, and the mechanical setup.

Figure 15a shows the SPMSM in the prototype setup being accelerated from 50 rpm to 90 rpm. As a result, the circuit that previously runs in all-time series operation enters the series-independent operation (see Figure 15b). The modulation index for each phase in the series-independent operation is further detailed in Figure 16. The initial experimental results verify the feasibility of the windings reconfiguration concept and thus open up the possibility of the final setup to be implemented with it. However, since in the planned final setup two series switches will be used, further investigation into the most optimal way to utilize the two series switches will be done in the future.

(a) (b)

WorldFigure Electric Vehicle15. (a) JournalAcceleration2019, 10 ,of 89 the permanent magnet synchronous machine (PMSM) from 50 rpm to11 of 16 90 rpm; (b) the resulting modulation index during acceleration.

Figure 16. Detailed measured modulation index for each phase in series-independent operation. Figure 16. Detailed measured modulation index for each phase in series-independent operation. 2.2.6. Industrialization Potential of GaN Devices

GaN is a young and emerging technology, which has a lot of advantages and great potential for automotive applications [12]. With increased switching frequency and low losses, it allows to build very efficient and compact power converters and inverters. Moreover, as it is a non-mature technology, technical improvements are to be expected, whether for the device performance itself (e.g., current capability and reliability) or the manufacturing process. As of today, the reliability of these devices cannot completely be assured, but the projects focus lies in the first application of this technology. Suppliers of GaN components foresee no problem in terms of reliability once the production ramps up and gets more mature. The prediction of CEA is that SiC devices will stay at the foreground for high voltage (>1 kV) applications. However, GaN technology will, in a few years, dethrone low voltage (<1 kV) technologies which have been installed for quite some time in power electronics, such as IGBTs and Cool-MOS [12]. Interestingly, it was noticed during the power board development that auxiliary components, such as passive components drivers and measurement probes, are not always adequate for use with GaN devices and progress has to be made also in those fields, without which the use and development of GaN-based inverters remains difficult.

2.3. Transmission For the transmission, two options were considered in the project. The first option is a fixed-ratio single-speed transmission. The second option is a dual-ratio switchable transmission. Several aspects were considered to help make the decision: Overall powertrain efficiency during type approval cycles: no clear winner between the two options • A dual ratio transmission with adapted ratios provides higher launch torque in first gear and higher • top speed for the same motor speed in the second gear: this benefits the projects demonstrator that has limited motor torque due to the current limitations of the available GaN devices. Market demand: some customers prefer a dual ratio transmission for EV powertrains for particular • reasons (e.g., gradeability, large difference between empty vehicle mass and fully loaded vehicle mass). As for the projects targets advantages of dual ratio transmission surpass disadvantages, it was decided to develop a dual-ratio switchable transmission. The ratios of the transmission are 21.65 for the first gear and 12.18 for the second gear, which leads to a ratio step of 1.78. The development process is described as follows: With the help of the gear synthesis software PlanGear from ZG, different gear configurations could be elaborated on the level of stick diagrams. Out of 34 possible configurations which fulfil the drivetrain’s requirements, the four configurations with the highest efficiencies were depicted for further investigation. These configurations were then evaluated with respect to expected costs and overall size. Finally, the combination of a Ravigneaux planetary system and a pair of cylindrical gears was identified as the most promising solution. In a next step, gear and bearing dimensioning was elaborated with the use of the transmission calculation World Electric Vehicle Journal 2019, 10, 89 12 of 16 software MASTA by SMT. The shaft and bearing loads were based on highly demanding drive cycles. FigureWorld Electric 17 illustrates Vehicle Journal the 2019 design, 10, process of the gear and bearing configuration. 13 of 17

FigureFigure 17. 17.From From geargear synthesissynthesis toto simulationsimulation model. model. Several configurations were reviewed whilst taking the manufacturing methods for parts and Several configurations were reviewed whilst taking the manufacturing methods for parts and the way of assembly of the powertrain into account. Subsequently, detailed gear and bearing lifetime the way of assembly of the powertrain into account. Subsequently, detailed gear and bearing lifetime calculations based on highly demanding drive cycles were performed. Load-capacity, efficiency, weight calculations based on highly demanding drive cycles were performed. Load-capacity, efficiency, and costs of possible bearing arrangements were evaluated and compared. Where possible, ball weight and costs of possible bearing arrangements were evaluated and compared. Where possible, bearings were applied to reduce transmission losses. Once the transmission layout was known, the oil ball bearings were applied to reduce transmission losses. Once the transmission layout was known, routing was developed to ensure that all bearings and gears were lubricated sufficiently. the oil routing was developed to ensure that all bearings and gears were lubricated sufficiently. The specific challenge of finding a suitable high-speed motor bearing is eased by a smart motor The specific challenge of finding a suitable high-speed motor bearing is eased by a smart motor shaft connection, which decouples the axial and radial forces of the transmission from the motor shaft connection, which decouples the axial and radial forces of the transmission from the motor bearings. This makes it possible to choose very small, efficient and speed-resistant motor bearings. bearings. This makes it possible to choose very small, efficient and speed-resistant motor bearings. This can be seen in more detail in Figure 18: The motor shaft is connected with the sun gear of the This can be seen in more detail in Figure 18: The motor shaft is connected with the sun gear of the planetary gear set through a hollow transmission shaft. By using a flexible spline connection, only planetary gear set through a hollow transmission shaft. By using a flexible spline connection, only torque is transmitted between motor and gear set and no axial or radial forces. The sun gear bearing is torque is transmitted between motor and gear set and no axial or radial forces. The sun gear bearing mountedWorld Electric in Vehicle the planet Journal carrier2019, 10, and thus, their relative speed is lower than the motor speed. 14 of 17 is mounted in the planet carrier and thus, their relative speed is lower than the motor speed.

2.4. Module Assembly and Final Specifications One of the ambitions of this project was to obtain a highly integrated, compact electric powertrain. The integration of motor, inverter and transmission into a single unit simplifies the powertrain assembly in the vehicle. Furthermore, the integration reduces the cost of interfacing by eliminating connectors, cables and covers. At the same time, a modular approach should allow exchanging (a part of) the motor, inverter or transmission for repair. From different layouts, the consortium chose an arrangement in which the inverter boards are placed around the drive shaft, as can be seen in Figure 18.

FigureFigure 18.18. CrossCross sectionsection ofof chosenchosen layoutlayout andand arrangementarrangement ofof componentscomponents inin thethe housing.housing.

This choice provides a compact unit with the best balance of length (L), width (W) and height (H) (Figure 19).

Figure 19. CAD drawing and main dimensions of developed powertrain module.

An important feature of the integrated housing is the cooling of both motor and inverter. Prior experience has shown that a too high back pressure in the cooling system requires a fairly high coolant pump power for the required flow. For this reason, the cooling subsystem puts the cooling of the motor and inverter in parallel. The motor cooling itself uses several parallel channels.

Having only one cooling circuit and all components integrated in one housing leads to a very compact overall design. Table 2 shows the resulting dimensions as well as peak power and torque.

Table 2. Resulting module specifications.

Parameter Value Width, length & height (W × L × H as in Figure 19) 405 × 513 × 275 mm Peak power at wheels 157 kW Peak torque at wheels 3460 Nm Required battery DC voltage 320 V World Electric Vehicle Journal 2019, 10, 14 of 17

World Electric Vehicle Journal 2019, 10, 89 13 of 16

2.4. Module Assembly and Final Speciﬁcations One of the ambitions of this project was to obtain a highly integrated, compact electric powertrain. The integration of motor, inverter and transmission into a single unit simpliﬁes the powertrain assembly in the vehicle. Furthermore, the integration reduces the cost of interfacing by eliminating connectors, cables and covers. At the same time, a modular approach should allow exchanging (a part of) the motor, inverter or transmission for repair.

From diﬀerent layouts, the consortium chose an arrangement in which the inverter boards are placed aroundFigure the 18. drive Cross shaft, section as of can chosen be seen layout in Figureand arra 18ngement. of components in the housing. This choice provides a compact unit with the best balance of length (L), width (W) and height (H) (FigureThis 19). choice provides a compact unit with the best balance of length (L), width (W) and height (H) (Figure 19).

FigureFigure 19. 19.CAD CAD drawingdrawing andand mainmain dimensions dimensions of of developed developed powertrain powertrain module. module.

AnAn important important featurefeature of of the the integrated integrated housing housing isis the the cooling cooling of of both both motor motor and and inverter. inverter. Prior Prior experienceexperience has has shown shown that that a too a too high high back back pressure pressure in the in cooling the cooling system system requires requires a fairly higha fairly coolant high pumpcoolant power pump for power the required for the flow.required For thisflow. reason, For this the reason, cooling the subsystem cooling subsystem puts the cooling puts the of thecooling motor of andthe invertermotor and in parallel.inverter in The parallel. motor coolingThe motor itself cooling uses several itself uses parallel several channels. parallel channels. Having only one cooling circuit and all components integrated in one housing leads to a very compactHaving overall only design. one cooling Table2 showscircuit theand resulting all componen dimensionsts integrated as well in as one peak housing power leads and torque. to a very compact overall design. Table 2 shows the resulting dimensions as well as peak power and torque. Table 2. Resulting module specifications. Table 2. Resulting module specifications. Parameter Value Width, lengthParameter & height (W L H as in Figure 19) 405 513 275Value mm × × × × Width, length & heightPeak (W power× L × H at as wheels in Figure 19) 157405 kW × 513 × 275 mm Peak powerPeak torqueat wheels at wheels 3460 Nm157 kW PeakRequired torque at battery wheels DC voltage 320 V3460 Nm Required battery DC voltage 320 V 2.5. Holistic Design Approach The best electric machine, the best inverter and the best transmission do not necessarily combine to the best drive module. This is because properties of fully electric drive modules result from multi-physical interaction of the powertrain components (electrical, mechanical and thermal). A design of the components with regard to the performance on vehicle level is necessary in order to find a global optimum for a specified vehicle and fixed driving requirements [17,18]. Furthermore, the control system design for the drive module is important to consider because it can increase the interaction between different subsystems but also be a remedy for unwanted interaction. World Electric Vehicle Journal 2019, 10, 15 of 17

2.5. Holistic Design Approach The best electric machine, the best inverter and the best transmission do not necessarily combine to the best drive module. This is because properties of fully electric drive modules result from multi- physical interaction of the powertrain components (electrical, mechanical and thermal). A design of the components with regard to the performance on vehicle level is necessary in order to find a global optimumWorld Electric for Vehicle a specified Journal 2019 vehicle, 10, 89 and fixed driving requirements [17,18]. Furthermore, the control14 of 16 system design for the drive module is important to consider because it can increase the interaction between different subsystems but also be a remedy for unwanted interaction. In parallel to the engineering of the drive module, a design process is being developed which In parallel to the engineering of the drive module, a design process is being developed which aims to identify the best combination of components for the demonstrator vehicle [18]. “Best”, herein, aims to identify the best combination of components for the demonstrator vehicle [18]. “Best”, herein, means the ideal compromise between costs, eﬃciency, performance, weight, volume and complexity. means the ideal compromise between costs, efficiency, performance, weight, volume and complexity. The result is an iterative and computational intensive process, which involves detailed component The result is an iterative and computational intensive process, which involves detailed component modelling of transmission, electric machine and inverter, as well as vehicle longitudinal dynamics modelling of transmission, electric machine and inverter, as well as vehicle longitudinal dynamics simulation in order to evaluate the components on system level (Figure 20). simulation in order to evaluate the components on system level (Figure 20).

Figure 20. Holistic design process of electric drive modules. Figure 20. Holistic design process of electric drive modules. In the first step, the requirements on vehicle level (e.g., acceleration from 0 to 60 km/h in 3.9 s) are brokenIn the first down step, to componentthe requirements level (1 on and vehicle 2). This level results (e.g., in acceleration a set of component from 0 to requirements 60 km/h in 3.9 such s) areas torque,broken down power, to gear component ratio, etc. level (3). (1 and The 2). component This results level in a calculation set of component blocks requirements layout and design such ascomponents torque, power, to fulfil gear the ratio, requirements etc. (3). andThe returncomponent the results level (4calculation and 5). The blocks components layout areand virtually design componentsassembled and to fulfil the performance the requirements on system and levelreturn is the evaluated results through(4 and 5). numerical The components and analytical are virtually vehicle assembledsimulations. and By the weighting performance different on designsystem criteria, level is such evaluated as costs, through performance, numerical and weight,and analytical the best vehicleset of components simulations. for By the weighting specified different target vehicle design is criteria, found (6). such As as this costs, study performance, is ongoing andand targetedweight, theto bebest completed set of components in the beginning for the ofspecified 2020, no target resulting vehicle drive is found module (6). specification As this study can is beongoing shown and yet. targetedHowever, to the be outcomecompleted of thisin the design beginning study is of not 2020, only no a suitableresulting specification drive module for drivespecification modules can for thebe shownVolvo C30e,yet. However, but also to the give outcome insights of into this the design component study interactionis not only betweena suitable electric specification machine, for inverter drive modulesand transmission. for the Volvo The C30e, following but also findings to give have insigh beents gained into the to component date during interaction the development between and electric usage machine,of the presented inverter design and transmission. environment: The following findings have been gained to date during the development and usage of the presented design environment: Electric Machine: High-speed EM generates power over high speeds and therefore, does need •• Electricless winding Machine: current High-speed than low-speed EM generates EM. Ohmic power losses over in thehigh windings speeds and are reduced therefore, while does the need iron lesslosses winding slightly current rise. However, than low-speed the savings EM. in Ohmic winding losses losses in surpassthe windings the additional are reduced iron losseswhile andthe ironthe losses machine slightly is more rise. effi However,cient. the savings in winding losses surpass the additional iron losses and the machine is more efficient. Inverter: Classical Si-based inverters are still a good choice for slow to medium switching •• Inverter: Classical Si-based inverters are still a good choice for slow to medium switching frequencies. At high switching frequencies, GaN and SiC provide a suitable characteristic. frequencies. At high switching frequencies, GaN and SiC provide a suitable characteristic. The The latter option leads to very compact inverters, as the size of the DC-link capacitor can be reduced with higher switching frequencies. Transmission: With higher maximum EM speeds, higher reduction ratios are required and the • transmission gets bigger and less efficient. For high reduction ratios, more complex topologies show advantages compared to simple spur gear designs. Module Level: A module based on a high speed electric machine as presented above is more • efficient and compact compared to conventional modules, even though the transmission itself World Electric Vehicle Journal 2019, 10, 89 15 of 16

is not. The drawbacks on the transmission side are overcompensated by the advantages on the machine and inverter side. However, due to high switching frequencies, conventional IGBTs produce high switching losses and thus, wide bandgap materials (GaN and SiC) should be used. Based on the current cost structures, the use of these materials results in a higher price on module level.

3. Conclusions This project sets the stage for many innovations to prove its readiness for the automotive market. Many new technologies and the reason for their selection are presented here. Among these are Formed Litz wires, six-phase series-independent winding connection, injected magnets and a two-gear shiftable ravigneaux gear set. The justification for the specific combination of these technologies comes from the striven high-speed concept, which makes the motor compact and efficient but introduces challenges for the transmission and inverter development. Furthermore, cost-effectiveness and environmental impact play an important role in technology selection. The high rotational speed favors fast switching semiconductors, hence, a switch technology gallium nitride was chosen. It is specifically preferred over SiC due to its lower switching losses. In addition, high reduction ratios and two gears are required, and as the most compact and smart solution, a ravigneaux gear set was identified. The components were assembled in one housing with one cooling circuit and hence, a compact and flexible to use drive module was realized. Component tests are promising; the complete module assembly and the integration in a demonstrator vehicle will take place in the near future, in 2020. As the overall conclusion, it can be said that the outcome of this project is a highly performant, compact and (potentially) cost effective drive module. It has the potential to increase attractiveness of electric drives over conventional vehicle and hence, to improve the adoption of them together with a reduction of worldwide CO2 and pollutant emissions.

Author Contributions: Conceptualization, J.H.; resources, L.E. and E.A.L.; writing—original draft preparation, J.H., H.H., J.A., D.O., C.L., O.T. and P.D.; writing—review and editing, J.H., M.R.L. and R.M.; supervision, D.K., L.E., E.A.L. and M.E.; project administration, C.L. Funding: This research was funded by the European Comission under the Horizon 2020 programme, grant number 769953. Acknowledgments: The authors want to express their gratitude for the opportunity of this research to the ModulED project (http://moduled-project.eu/), co-funded by the European Union under H2020 programme. Conﬂicts of Interest: The authors declare no conﬂict of interest.

References

1. Todts, W. CO2 Emissions from Cars: The Facts; European Federation for Transport and Environment AISBL: Brussels, Belgium, 2018. 2. Auvinen, H.; Sampsa, R.; Juha, O.; Anu, T.; Toni, A. Process to Support Strategic Decision-Making: Transition to Electromobility. In Proceedings of the 2013 World Electric Vehicle Symposium and Exhibition, Barcelona, Spain, 17–20 November 2013; IEEE: Piscataway, NJ, USA, 2014. 3. Damousis, Y.; Amditis, A.; Naberezhnykh, D. Electromobility: A Market Readiness Study—Preliminary Findings. In Proceedings of the IEEE IEVC’14, Florence, Italy, 16–19 December 2014. 4. Lienkamp, M. Status Electromobility 2016: Or How Tesla Will not Win; Technische Universität München: Munich, Germany, 2016; p. 4. 5. Boglietti, A.; El-Refaie, A.M.; Drubel, O.; Omekanda, A.M.; Bianchi, N.; Agamloh, E.B.; Popescu, M.; Gerlando, A.D.; Bartolo, J.B. Electrical Machine Topologies—Hottest Topics in the Electrical Machine Research Community; IEEE: Piscataway, NJ, USA, 2014; Volume 18, p. 24. World Electric Vehicle Journal 2019, 10, 89 16 of 16

6. Vagati, A.; Boazzo, B.; Guglielmi, P.; Pellegrino, G. Ferrite Assisted Synchronous Reluctance Machines: A General Approach. In Proceedings of the 2012 International Conference on Electrical Machines, Marseille, France, 2–5 September 2012; IEEE: Piscataway, NJ, USA, 2012. 7. Tenconi, A.; Vaschetto, S.; Vigliani, A. Electrical Machines for High-Speed Applications: Design Considerations and Tradeoffs. IEEE Trans. Ind. Electron. 2014, 61, 3022–3029. [CrossRef] 8. Benedikt Vogel (on behalf of: Swiss Federal Office of Energy SFOE, Bern, Switzerland). Electric Power Package with High Efficiency. Available online: https://www.researchgate.net/publication/332112701_ Electric_Power_Package_with_high_Efficiency (accessed on 21 November 2019). 9. Arnold Magnetic Technologies Co. N45UH Sintered Neodymium-Iron-Boron Magnets. Available online: https://www.arnoldmagnetics.com/wp-content/uploads/2017/11/N45UH-151021.pdf (accessed on 21 November 2019). 10. Aichi Steel Corporation. MAGFINE® NdFeB Anisotropic Resin-Bonded Magnet—Technical Datasheet. Available online: https://www.aichi-steel.co.jp/ENGLISH/products/electromagnetic/bonded_magnet/index. html (accessed on 21 November 2019). 11. GaN Systems Inc. GS66516T Top-Side Cooled 650 V E-Mode GaN Transistor—Preliminary Datasheet. Available online: https://gansystems.com/gan-transistors/gs66516t/ (accessed on 13 March 2019). 12. Kaminski, N.; Hilt, O. SiC and GaN Devices—Competition or Coexistence? In Proceedings of the 2012 7th International Conference on Integrated Power Electronics Systems (CIPS), Nuremberg, Germany, 6–8 March 2012; IEEE: Piscataway, NJ, USA, 2012. 13. Gerrits, T.; Wijnands, C.G.E.; Paulides, J.J.H.; Duarte, J.L. Fault-Tolerant Operation of a Fully Electric Gearbox Equivalent. IEEE Trans. Ind. Appl. 2012, 48, 1855–1865. [CrossRef] 14. Gerrits, T.; Wijnands, C.G.E.; Paulides, J.J.H.; Duarte, J.L. Electrical Gearbox Equivalent by Means of Dynamic Machine Operation. In Proceedings of the 14th European Conference on Power Electronics and Applications, Birmingham, UK, 30 August–1 September 2011. 15. Gerrits, T.; Duarte, J.L.; Wijnands, C.G.E.; Lomonova, E.A.; Paulides, J.J.H.; Encica, L. Twelve-Phase Open-Winding SPMSM Development for Speed Dependent Reconfigurable Traction Drive. In Proceedings of the 2015 10th International Conference on Ecological Vehicles and Renewable Energies (EVER), Monte Carlo, Monaco, 31 March–2 April 2015; pp. 1–7. 16. Daniels, B.; Gurung, J.; Huisman, H.; Lomonova, E.A. Feasibility Study of Multi-Phase Machine Winding Reconfiguration for Fully Electric Vehicles. In Proceedings of the 2019 9th International Conference on Ecological Vehicles and Renewable Energies (EVER), Monte-Carlo, Monaco, 8–10 May 2019; pp. 1–6. 17. Albers, A.; Reichert, U.; Ott, S.; Radimersky,A. Software Supported Development Process of a Battery Electric Vehicle Powertrain Considering the Efficiency; VDI-Berichte: Düsseldorf, Germany, 2016; Volume 2276, pp. 321–337. 18. Kieninger, D.; Hemsen, J.; Köller, S.; Uerlich, R. Automated Design and Optimization of Transmissions for Electric Vehicles. MTZ Worldwide 2019, 11, 94–98. [CrossRef]

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).