Synchronous Reluctance Next Generation Efficient Motors for Electric Vehicles

Final Report Summary - SYRNEMO (Synchronous Reluctance Next Generation Efficient Motors for Electric Vehicles)

Executive Summary:
Road transportation is the only major sector in the EU where greenhouse gas emissions are still increasing, contributing to as much as one-fifth of the EU’s total carbon dioxide emissions. Electric vehicles are at the heart of EU actions to reduce transport emissions, and the European Commission has set the objective of halving the use of conventional fuel vehicles in urban transport by 2030, and phasing them out entirely in cities by 2050. However, before becoming the most common vehicles on our roads, electric vehicles need to deliver better range performance in a lightweight design and at lower cost, together with the development of an adequate charging infrastructure. The objective of this project is to deliver novel drivetrain architecture for electric vehicles, based on Permanent Magnet assisted Synchronous Reluctance (PMaSYRM) machine technology. Such architecture aims at delivering better power and torque performance while reducing weight and volume, thus increasing power density and reducing weight. Additionally it is designed to be: (1) as little dependent as possible on scarce and hence expensive materials, featuring an innovative stator winding and rare-earths free magnets, (2) easy to manufacture, dismantle, and recycle in large volume, (3) feature a smart packaging of the power electronics and an integrated air-based thermal management system, to increasing efficiency and (4) increase energy efficiency over a wide range of operating conditions, including real-world operational constraints. The project consortium has designed, prototyped and tested a ferrite permanent magnet assisted synchronous reluctance machine, suited for A/B-segment M-category light duty electric vehicles at TRL 5-to-6. The machine was conceived ab initio to avoid the use of rare-earth magnets, implementing a novel hairpin winding for the stator, an innovative lightweight modular design for the rotor, and an air-cooled case. Results show that the final motor design provides a maximum torque performance of 133 Nm at 3,700 rpm and a maximum power of 56.7 kW at 4,900 rpm, with peak efficiency above 96% around 4,000±500 rpm and 50±20 Nm, decreasing to 93-94% on including the inverter efficiency, according to the simulation results of the final design. The prototype machines were tested in a Hardware-in-the-Loop configuration according to a comprehensive Validation Test Plan accounting for 12 different tests. Final test results showed that the drive system efficiency is 92% on the NEDC and 89% on the WLTC, with motor peak efficiency is measured at 96% with system peak efficiency beyond 92%. The maximum delivered power is measured at 46 kW (compared to the 56 kW of the simulation), while the maximum torque performance is measured at 125 Nm (compared to the 133 Nm of the simulation). Such deviations can be ascribed to the operation of the control algorithm, which, on purpose, did not stress current and voltage levels to ensure safe operation of the drive systems under all the conditions tested. Additionally, the electric and mechanical safety margin of the test environment kept the machine operation below the peak performance, in order not to stress the components and not damage the hardware. This resulted in an underrated measured performance of the machine, therefore it can be assumed that the simulated values are achievable with the present design. The consortium also delivered a final set of recommendation to improve the current design including the: (1) adoption of a liquid cooling system with the implementation of a spiral cavity aluminium housing, (2) increase the phase current capability of the power inverter, (3) installation of a higher grade of ferrite permanent magnets in the rotor stack, (4) adoption of a single typology of non-oriented silicon steel, in order to limit cost and make the machine more capable for series-production, (5) harmonisation of the permanent magnet height to reduce cost of the ferrites and (6) implementation of a shortened pitch winding to decrease torque ripple and be able to use a smooth rotor surface. The final design has been also subjected to a cost assessment. The NPV of the prototype is estimated at 9,145 €, decreasing to 3,170 € assuming a production line of 10,000 units with 432 work-hours for the production of the machine. The system is finally priced at 2,455 €, accounting for 20% market surcharge and 6% profit. It is important to underline that this price has been calculated based on a very low-volume of 10,000 units, and that, if scaled up to automotive numbers, i.e. typically between 100,000 and 1,000,000 units/year can significantly decrease. However, this implies a scale up also of the TRL of the drive, that can be made at the level of follow-up projects. However, it is important to highlight that the technical achievements of the proposed motor design were evaluated based on two key-performance parameters, i.e. the machine constant of mechanical power and the torque density. These two parameters have been calculated for the SyrNemo machine design, either in air-cooled and liquid-cooled configuration. The liquid cooled design results to be the best-in-class design among the 2016 state-of-the-art rare-earths free PMaSYRM. This result proves the quality of the presented design as well as the potential for further improvements, thus enabling the presented machine to constitute a key enabling technology for next generation electrified vehicles.

Project Context and Objectives:
Introducing electric vehicles in the mass consumers’ market is a vital part for future mobility. In order to facilitate mass adoption of electric vehicles by consumers, one of the most important issues is bringing their cost in line with that of the internal combustion engine. Together with this, the “Next generation Electric Motor” is expected to:

- deliver higher and tailored output while reducing weight and volume, thus increasing power density and reducing weight;
- be as little dependent as possible on scarce and hence expensive materials (e.g. as rare earths);
- be easy to manufacture, dismantle, and recycle in large volume;
- have smart packaging of power electronics and integrated thermal management, thus increasing efficiency;
- increase energy efficiency over a wide range of operating conditions, accounting for diverse real-world operative constraints;

In order to address these issues, the EU-FP7 SyrNemo project had the objective of:

- designing an electric machine suited for the relevant drive cycle and real world operational constraints. To achieve this, the maximum efficiency region has to be located where the highest probability of operation and the higher amount of energy exchange in driving situations is expected, as per NEDC and WLTC duty cycle combined with the A-segment vehicle requirements. Additionally, an efficient control ensures that the drive is always controlled so that the energy efficiency of the system is as high as possible.
- delivering an electric machine whose usage pattern ideally suits automotive applications. That means that the lifetime of the machine should be chosen according to the life expectancy of the vehicle, thus allowing maximum performance, or, minimum used space within the car.
- including, within the design phase, of a number of criteria such as robustness, easiness of the manufacturability and high power density of the machine for vehicular applications and the achievement of no rare-earths content by using ferrite magnets.

Based on these objectives, the SyrNemo Project aimed at delivering the design and the prototype of an innovative Permanent Magnet assisted SYnchronous Reluctance Machine (later referred as PMaSYRM) with higher power density and higher driving cycle efficiency at lower cost than the current state-of-the-art Permanent Magnet (PM) synchronous machines. The delivery of this technology implies a number of significant technological advancements, such as:

- the increase of the mass and volume specific power densities by approximately 5%. This must achieved through an innovative magnetic reluctance rotor design with optional ferrites. Bar windings are used to reduce the required winding space. A natively integrated and efficient air cooling concept is used to cool both the power electronics and the motor to further reduce the system’s energy consumption and total mass and volume.
- to avoid the use of rare earths by adopting either non-permanent magnets or ferrites. The proposed rotor design also allows for using future magnet materials with high energy density once they will be available on the market.
- the easiness of manufacturing of the proposed SYRM technology, including the dismantling and recycling phases. This way manufacturing cost can be reduced by 20% compared to PM Synchronous Machines (PMSMs).
- the native inclusion of the eco-design and Life-Cycle Assessment (LCA) to ensure minimum environmental impact, and avoid hidden social costs.
- the rotor design robustness.
- the proper design of the insulation system for a total lifetime of 10 years and 10,000 operating hours.
- the high efficiency of the delivered technology for a wide range of driving speed and torque. Therefore, the overall driving cycle efficiency of SYRM is improved by 5–15% compared to state-of-the-art PMSMs.
- the implementation of the control of the drive, to achieve the maximum efficiency in each operating point, accounting for the stator winding temperature.
- the possibility to industrialize the machine.

The project included eight partners coordinated by the Austrian Institute of Technology Ges.m.b.H. It ran for 36 months. The project has delivered a fully electric powertrain technology at TRL 5-to-6, explicitly designed to be integrated in large-scale production of future electric vehicles, enhancing driving efficiency and environmental compatibility of the next generation of electric motors, in line with the EU policies in the field of low-carbon transport and GHG reduction for road transport. This objective has been achieved with significant advancements and progresses beyond the state-of-the-art in several technological aspects. The results achieved are:

- (WP1): the PMaSYRM economic, environmental, and social impacts is calculated in the full life cycle. The NPV of the prototype is estimated at 9,145 €, decreasing to 3,170 € assuming a production line of 10,000 units with 432 work-hours for the production of the machine. The system is finally priced at 2,455 €, accounting for 20% market surcharge and 6% profit. The environmental assessment, carried out according to the principles from the ISO 14040 standard, included ten different impact categories: climate change, fossil depletion, human toxicity, ionizing radiation, metal depletion, ozone depletion, particulate matter formation, photochemical oxidant formation, freshwater and terrestrial eco-toxicity. The results show that for the majority of these impact categories the use stage of the motor has the largest impact, because of the impact of the generation of the electricity in the operational life. The highest impact categories are climate change, fossil depletion, and ionizing radiation, where the impact of the use stage compared to the total life cycle impact ranges between 72% and 93%. The social assessment revealed a benefit of the machine in EU (Austria, machine production), Southern African region (materials mining and early processing), as well as Eastern Europe and Central Asia (extraction and distribution of fossil fuels for electricity generation).

- (WP2): The PMaSYRM design constraints are identified. The machine is designed for native integration in a real-axle driven hatchback fully electric vehicle. The performance requirements correspond to the one imposed by the current and the future reference type approval duty cycles for light duty vehicles: the NEDC and the Worldwide harmonized Light-duty Test Cycle (WLTC). The torque and power specifications of the SYRM are derived from the lumped model predictions of a hatchback vehicle with a curb weight of 1,030 kg (A-segment passenger car) driving the NEDC and WLTC driving cycles. Additionally, a vehicle maximum speed of 130 km/h and the capability of climbing a road with 20% slope are also set as constraints to determining the minimum design parameter set for the motor. Larger vehicles, i.e. a B-segment passenger car with curb weight of 1,760 kg and a light commercial vehicle with a curb weight of 3,500 kg, are also simulated to determine the motor requirements for a scaled-up version of the PMaSYRM design.

- (WP3): The PMaSYRM is designed, by undertaking the following steps: (i) design of the motor stator by adopting an innovative hairpin winding that is composed of solid conductors with rectangular cross sections. This solution, due to the higher cross section area and to the higher slot fill factor (i.e. approximately 65 %) compared to typical round wire, allows also reducing the DC resistance. Stator insulation is also designed and guaranteed for 10,000 hours of operation with a target reliability level of 99.5%; (ii) design of an innovative rotor concept made of 5 skewed rotor stacks, assisted by optional ferrite magnets. The rotor is designed for maximising torque and power performance, including a smooth torque waveform, sufficient mechanical strength, short axial machine length and improved demagnetization behaviour in case of three-phase short-circuits during load operation; (iii) the design of a new power electronics, with custom inverter and controller; (iv) the design of an innovative air-cooling concept via Computational Fluid Dynamics for tuning a motor mono-dimensional lumped thermal model, resulting in finned housing suited for passive cooling, eventually enhanced with active cooling via fans for suppling additional cooling flow. The simulated performance of the machine consists in a maximum torque performance of 133 Nm at 3,700 rpm and a maximum power of 56.7 kW at 4,900 rpm, with peak efficiency above 96% around 4,000±500 rpm and 50±20 Nm, decreasing to 93-94% by including the inverter efficiency.

- (WP4): Two full PMaSYRMs have been prototyped, consisting of the motor (i.e. stator and rotor), the inverter and the housing, as well as the needed test bench equipment, tooling, dummy parts and mock-up components to manufacture the machines. One machine, i.e. later referred as machine #1, contains a rotor with a smooth surface, while the other, i.e. later referred as machine #2, contains the rotor with a structured surface. Furthermore, two additional rotors and three additional inverters have been manufactured. The prototypes have been delivered after passing an End-of-Line (EoL) in order to verify the conformity of the prototyped parts to the design specification as well as the possibility to safely operate the machine in controlled environment. Additionally, the Over-Speed (OS) test was also performed, as complement to the machine prototype delivery procedure. This consists in driving the rotor at a rotational speed 10% higher than the maximum speed of the machine, i.e. 13,200 rpm versus a maximum machine speed of 12,000 rpm, to verify the mechanical integrity of the rotor under extreme operational conditions. The test was passed successfully.

- (WP5): The two full PMaSYRMs are tested and validated on a HIL test bench, according to the Validation Test Plan made of 12 tests. The smooth rotor machine (i.e. Machine #1) has been tested following the full validation test plan, whereas the structured rotor machine (i.e. Machine #2) has been tested only under the specific conditions to highlight the differences between the two machines. The overall performance of the system is evaluated according to the relevant duty cycles, featuring appropriate cooling environment to replicate the motor’s real world operating conditions in different environments. Data are analysed to derive the driving efficiency for different load conditions, evaluating the specific high current low speed points to assess the limits of operation. The overall efficiency of the system results to be 92% on the NEDC and 89% on the WLTC, with motor peak efficiency at 96% and system peak efficiency beyond 92%. The maximum delivered power is measured at 46 kW (compared to the 56 kW of the simulation), while the maximum torque performance is measured at 125 Nm (compared to the 133 Nm of the simulation). Such deviations can be ascribed to the operation of the control algorithm, which, on purpose, did not stress current and voltage levels to ensure safe operation of the drive systems under all the conditions tested. Additionally, the electric and mechanical safety margin of the test environment kept the machine operation below the peak performance, in order not to stress the components and not damage the hardware. This resulted in an underrated measured performance of the machine, therefore it can be assumed that the simulated values are achievable with the present design. Additionally, based on the test results, a set of recommendation to improve the final design has been delivered, including the: (1) adoption of a liquid cooling system with the implementation of a spiral cavity aluminium housing, (2) increase the phase current capability of the power inverter, (3) installation of a higher grade of ferrite permanent magnets in the rotor stack, (4) adoption of a single typology of non-oriented silicon steel, in order to limit cost and make the machine more capable for series-production, (5) harmonisation of the permanent magnet height to reduce cost of the ferrites and (6) implementation of a shortened pitch winding to decrease torque ripple and be able to use a smooth rotor surface.

- (WP6): a number of dissemination and exploitation activities have been undertaken during the project execution. In summary these activities have led to (i) a constantly updated website, (ii) the organisation of an electro-mobility workshop, (iii) the presentation of the project at the TRA 2014 and the presentation of the project results at the TRA 2016 with an exhibition stand, (iv) the invitation for the web article “New Generation of Electric Cars for a Cleaner Europe” and (v) four oral presentations of the project and of the project’s findings within scientific events, seven articles in conference proceedings, one poster presentation to a conference and five peer reviewed journal articles. The project finally delivered a final video material, to be used for web and social media dissemination.

The final assessment of the proposed motor design shows how it can evaluated based on two key-performance parameters, i.e. the machine constant of mechanical power and the torque density. These two parameters have been calculated for the SyrNemo machine design, either in air-cooled and liquid-cooled configuration. The liquid cooled design results to be the best-in-class design among the 2016 state-of-the-art rare-earths free PMaSYRM. This result proves the quality of the presented design as well as the potential for further improvements, thus enabling the presented machine to constitute a key enabling technology for next generation electrified vehicles.

Project Results:
The scientific and technological results achieved during the execution of the SyrNemo project are presented broken-down as per project work-package structure. Only RTD WPs are here reported, i.e. from WP1 to WP5.

WP1: Eco-Design

The eco-scan of the SyrNemo machine has been carried out on the preliminary design, for the identification of Key Environmental Performance Indicators (KEPIs). The environmental impact of each of the stages in the life cycle of the electric motor has been evaluated by including several material alternatives, establishing which material option has the potential to decrease the overall environmental impact of the motor. The Eco-Scan also covers general Eco-Design recommendations for selection of “eco-materials”, management of the supply chain, and manufacturing and assembly to enhance recycling. This also included an overview of environmental European legislation concerning electric motors. The results show that during the usage stage of the motor produces the highest environmental impact. This impact is many times higher than comparing to any other life stage phase, for all the 18 impact categories considered in the study, except for metal depletion, which is highest during materials production. The amount of energy consumed during the usage of the motor is key to decrease the environmental impact. A high efficient motor has a big potential to produce much less environmental, comparing to a motor of lower efficiency, especially if it uses electricity producing from a renewable source. A reduction to minimum levels of the overall amount of materials used, while adequately fulfilling other functional requirements, it is also important to decrease the impact of the e-motor, especially for housing, magnetic core and winding components. A reduction of copper used is especially important to improve the toxicity produced to humans and ecosystems. The same case occurs for aluminium and steel components, which are very energy intensive, where a reduction of their use can improve depletion of metal resources and total GHG emitted.
After the eco-scan and during the execution of the design and prototyping phases of the SyrNemo drive, the complete environmental life cycle assessment has been carried out. This task aims at identifying the critical stages, materials, and processes, affecting the environmental performance of SyrNemo integrated drive, as well as the validation of these results by means of benchmarking, quantification of uncertainty, and sensitivity analyses on critical parameters. Several strategies are proposed to further reduce the environmental impact of the electric drive when this one is produced beyond prototype stages. In addition, the LCA also includes analysis on the manufacturability, dismantling and recycling of the machine. To achieve the previous, a series of actions were completed:

- Definition of the goal and scope of the study.
- Establishment of the study boundaries.
- Definition of necessary inventory data.
- Collection of primary data and secondary data related to the SyrNemo design, as well as three other benchmark motors.
- Creation of environmental models for representing the life cycle stages of the drive.
- Selection of an adequate LCA impact category, according to the established goal.
- Interpretation and validation of the results in terms of Key Environmental Performance Indicators.
- Proposal of impact reduction strategies for decision support.

An in-depth analysis of the recyclability of the SyrNemo machine was performed, identifying critical metals, minerals and energetic resources related to large depletion impact, according to the different impact methods available. A gap is identified between the critical materials and the real rates at which these materials can be recovered after effective dismantling and recycling. A study on the environmental criticality of rare earth metal used in permanent magnet was performed, highlighting the advantages of replacing this one by ferrite materials. The main outcomes can be summarized as:

- Electricity produced to power the electric drive is the main contributor to impact in the life cycle of the machine. Efficiency of the machine is key for the reduction of this impact.
- Components of larger mass and precious metals produce the most significant share of impact, mainly due to emission related to extraction and early processing of the raw materials.
- The manufacturability of the machine is enhanced by a series of decisions taken during design, as highlights the advanced integration of motor and inverter, and the simplicity of winding arrangement.
- Dismantling and recycling of the machine can be done without difficulties, with the use of specific machinery, obtaining a significant avoidance of emissions and depletion impact.
Together with the LCA, the micro and macro-economic impacts in the life cycle of the machine are estimated. The objective is to identify economic costs and benefits by accounting for the different monetary flows, including capital expenditures (materials, production costs, R&D), operating costs (fuel, damages), and disposal costs (residual and disposal). While a macroeconomic assessment shows how these activities have an impact in the each country along its supply chain. These estimations are carried out through the following actions:
- Definition of every stream of punctual and periodic costs for the manufacturing, use and disposal of the electric drive.
- Collection of all necessary costs and data.
- Selection of an appropriate discount rate, and calculation of the net present value of one-time and recurring costs.
- Comparison between life cycle stages and identification of potential hidden costs, such as the costs associated to the repair of health damages due to electricity consumption.

All cost flows are identified, aggregated and compared to show the main driver for economic impact when the machine is produced at prototype level of with 10,000 production units. This analysis shows that the manufacturing of the SyrNemo machine is the main factor in the generation of life cycle economic impact, as well as in the generation of direct and indirect economic activities. The main costs are the acquisition of materials and the manufacturing of rotor and stator stacks. However, when the machine is produced at larger scale, the costs of manufacturing decrease, and a comparison made shows that the costs of electricity for the powering of the machine during its use can become the main source of economic impact. Several scenarios are analyzed to observe the effect of assumptions taken for the study, including prices of electricity, R&D costs and external costs. A description of costs associated to the dismantling of the machine is done, including initial investment necessary, later operational costs, and the potential revenue expected from selling of scrap metals. A macroeconomic input-output assessment was performed, in a scenario where the machine is manufactured, use and disposed mainly in Austria. The study shows that, the main economic contributions take place in this same country and its European neighbors. Countries with large economies (i.e the United States, China and Russia) also have an indirect economic stimulus, while smaller economies in the rest of the world contribution mainly in the sectors of mining and production of basic metals.

The WP1 is finally concluded by performing the Social Life Cycle Assessment (S-LCA), which has the objective to identify positive and negative social impacts in the supply chain of the materials required in the life cycle of the SyrNemo electric drive. The identified social impacts are translated into recommendations to consider once the electric drive is manufactured in large scale. Given the complexity of global supply chains, the locations for the extracted materials and resources and processing used in the design, are estimated by means of global input-output tables. By combining these probable locations with internationally recognized reports on social issues, the potential social hotspots in the supply chain are calculated. In addition, a life cycle sustainability assessment is performed, gathering the results from each pillar of sustainability (environment, economy and society). A selected number of indicators are aggregated representing each pillar. The results are compared to the defined state of the art technology, and recommendations for optimization of the sustainability conditions are proposed. The results of the social assessment show potential opportunities for improvements in the social footprint, especially in the raw materials extraction from South African region countries, and energetic resources extraction and transit from Central Asia regions. Comparing to the state of art technology, the sustainability impact of the SyrNemo machine is relatively lower as a consequence of the efficient material use of the electric motor design, the higher efficiency, smaller size resulting in slightly lower costs, and the better social conditions of employment in European countries such as Austria.

WP2: Specification and OEM constraints and needs

The preliminary feasibility study is performed to verify the scalability of the SyrNemo electric motor concerning different vehicle types. Vehicle performance targets and specifications are defined, in order to define the baseline electric motor that will be manufactured and applicable for different vehicle classes. After simulations a baseline electric motor specification is achieved, and based on this baseline electric motor, different configurations are defined, able to handle with all the different application requirements. The results show that and after minor modifications in the vehicle preliminary specifications, a baseline electric motor has been defined for the heavy quadricycle and the A segment BEV vehicle, able to cope with the requirements of all the vehicle types applying the scalability feature required within SyrNemo project. For this reason, in the specification activity and in the design phase of the SyrNemo project the baseline motor is used as reference e-machine, and the performance figures identified are set as the performance targets for the e-machine design. The maximum acceleration target has been the most restrictive of all the requirements, in terms of torque vs. motor speed curve definition as predicted. Nevertheless, the use of the very demanding WLTP duty cycle has showed that even if the optimised motor size for each vehicle type is able to fulfil the vehicle dynamic requirements, in some cases like in the LCV or the Bus with full load, the motor cooling at high torque demands can be critical with the selected electric machine size. The NEDC-based optimization of power train components is compared against approaches using real world cycles and attempts considering component ageing, deriving efficiency maps and real-world requirements. The results show that the NEDC approach gives satisfactory results for the case because (i) the densely populated areas on the torque-speed plane are similar in both cases (i.e. real world vs. NEDC) and (ii) the NEDC is quite demanding in terms of component ageing. The results of this activity allowed making correct assumption on the main figures for the specification of the e-machine, particularly the required mechanical performance. The specification has been organised in order to recall the typical specifications for traction e-machines in automotive applications, in order to have also a focus on the problems that should be faced far beyond the prototyping phase. Environmental aspects have been briefly analysed, determining the environmental limits for the e-machine, and defining a possible set of environmental / endurance tests that a mass production shall withstand in case of Design Verification or Product Validation. Materials requirements have been developed according to the outputs of WP1, related to the Eco-Design, in order to give advices about how to sustainably design and manufacture the target e-machine. From mechanical point of view, some hypotheses have been set for the dimensions available for the e-machine, considering both production intent and testing requirements. On this basis, the electrical interface has been defined for power and signal wirings, considering that for the preliminary samples the wiring shall take into account also the need for a telemetry system in order to monitor all critical aspects during testing activities. From power side, the e-machine preliminary design deals with HV, and voltage levels defined in agreement with the inverter limitations, and battery system constraints. The manufacturing requirements have been set trying to remind constantly the aim of the project that is the realisation of a high efficiency electric machine; for this reason, practices aiming to maximise the energetic performance have been strongly recommended. The main goal of the preliminary design is to deliver the concept of an e-machine capable of providing the desired performance without using permanent magnets, but in case a minor contribution is required, the specification is already predisposed to this circumstance, provided that rare earth materials are not allowed in any case. The mechanical performance target of the e-machine has been based on the outputs from other tasks within WP 2, leading to a preliminary sizing in torque and power. These figures have been evaluated as feasible according to the design dimensions provided in the document; some efficiency target have been set, based on similar e-machine arrangements and technologies. In the end, a lifetime target has been set for the production version of the device, considering all the aspects that can affect this parameter from electrical and mechanical point of view, guaranteeing an acceptable reliability to customers. Finally, the test bench requirements and list of procedures have been agreed in detail to carry out different types of tests. Main scope is to collect testing and validation requirements for the e-motor and e-drive, from which to derive all the detailed procedures to be used for the final prototype on-bench evaluation and the following assessment. All these procedures are recommended and valid for generic electric machine testing. In the SyrNemo project an additional activity, that involved all the partners, is focused on the definition of a shorter list of recommended tests specifically defined for the electric machines manufactured at prototype level.

WP3: Next Generation Motor Design

According to the guidelines defined by the consortium, the preliminary motor and the final motor should not differ significantly, since different mechanical designs would impose additional manufacturing cost, which would exceed the planned budget. The partners agreed on an intermediate design freeze, (at least for the stator active parts), at the stage of the preliminary SYRM design. At the beginning of WP3 Next Generation Motor Design, it soon became apparent, that the design freeze limits the optimization possibilities considerably. In order to develop the SYRM control already at an early stage of the SyrNemo project, CRF thankfully provided the HySYS machine for testing purpose. The proposal to use the HySYS machine as Preliminary SyrNemo Motor was accepted by the project officer and it will serve as test case for the used design tools within the consortium. The preliminary design also accounts for the calculations followed by detailed FEA studies to assess individual loss components and the balance between best performances at specific operational points, inverter constraints, and manufacturability. Integrated inverter design and inverter-motor thermal cooling integrated systems are included in the preliminary design also together with boundaries for detailed mechanical functional design. The task finally provides the 3D digital Mock-up of the drive and the winding scheme.
The HySYS machine from CRF is used as preliminary motor prototype as well as a suitable inverter for this motor from TEC facilities. The HySYS machine (rare earth permanent assisted synchronous reluctance topology) important data and reports are shared among the partners and are adopted to validate the SyrNemo design methodology. The preliminary control is also adopted and implemented to carry out the preliminary testing at the test bench, by using standard power inverted capable of operating the HySYS motor. Safety goals are accounted at the prototype level according to the standard ISO 26262, i.e. functional safety requirements of SYRM eMotor. This approach ensures a power train concept, which will be inherently safe from a functional safety perspective, to be adopted for different road vehicles, and duty cycles, such as heavy quadricycle, segment A-B vehicles, FC vehicles, LDVs or buses. The results have led to the Safety Goal definition, thus identifying the Safe State for the system being analysed. Finally, a list of high level Functional Safety Requirement has been defined, focusing of course on those requirements that could be in charge of the MCU. Preliminary tests are carried out on the HySYS eMotor’s controller, following the testing requirements identified above. Based on the performance, robustness and reliability of the designed control algorithm all over the HySYS operating range results have been evaluated, together with efficiency maps and driving cycles of the system. The current controller has been tested in different scenarios (normal operating areas and in the boundary of the curve torque vs speed) and stability test including NEDC and WLTP driving cycles as well as efficiency maps (reaching an efficiency of 93%) have been carried out in order to verify control’s response. In all the previous cases mentioned, the system remains under control. After detailed assessment and some experimental evidences, the system performance is considered acceptable.
Based on these results the SYRM design has been carried out. As far as the stator winding concerns, three different wire solutions have been analysed: (1) the “solid” wire with a rectangular cross section, (2) the typical “round” wire and (3) “thin wire bundle” (or litz wire). To find the most suitable wire type for the SYRM, these three candidate solutions have been evaluated versus a variety of criteria that are directly related to the efficiency and the manufacturing of the machine. The solid wire technology with two parallel sub-conductors presents the best trade-off for the intended use as a stator winding for the SYRM drive. Based on this result, the 3-phases winding with the hairpin technology manufacturing process has been designed for the proposed drive by using two solid wires in parallel. As far as the rotor design concerns, it has been designed to maximize the power density and the efficiency of the motor together with minimising torque ripples throughout the whole speed range by (1) applying a short pitch winding, (2) skewing of the rotor and (3) adding dips across the rotor surface. The lowest torque ripple is reached by skewing the rotor and applying slots along the perimeter. After an intensive simulation campaign, the rotor segmentation has been designed with five different stacks displaced by a relative rotation angle. The segmentation of the rotor stack gives a significant improvement in terms of voltage exploitation at the very high speed, reaching a peak increase equal to +35% of torque in the top speed working point. However, a higher current request for the torque production is the downside of skewing. This was partly compensated by slotting the rotor perimeter. The optimization measures resulted in (1) 5% reduction of the rotor stack weight, keeping the same manufacturing process, (2) 4% reduction of the torque ripples, (3) better exploitation of the flux weakening especially at top speed. Additionally a worst case scenario of an all-phase electrical short circuit in the stator windings was investigated to find the optimal permanent magnets position to avoid the demagnetization due to the magnetic field caused in such a situation. As far as the mechanical design is concerned, the final electromagnetic design of the rotor has been submitted to a mechanical analysis, in order to find a suitable compromise between electromagnetic and mechanical properties. It is important to highlight that high torque output and mechanical robustness are two contradicting goals, resulting in the fact that the better the flux guiding to achieve high reluctance properties, the lower the mechanical robustness. On the other hand, if the mechanical robustness is high, usually the magnetic flux passes through ribs that are meant to assist in the mechanical structure of the rotor rather than guiding the flux. This usually amounts in a less channeled flux and therefore the output torque is slightly smaller. The mechanical analysis has been carried out by considering the worst mechanical stress operating conditions and calculating that the rotor total deformation would remain below 0.1 mm. The electrical steel lamination chosen for the rotor is very robust, i.e. TKES-500Y40-35HS with elastic limit Rp0,2 of 500 N/mm².
The motor final design has been coupled with a fully integrated inverter and power electronics, which include a novel control system. On one hand the inverter design has to meet cooling requirements given the available installation space in the motor/inverter housing, while, on the other hand, the inverter has to sustain the currents that are necessary to reach the operation points. In drive operation, the semiconductor losses cause heat dissipation that can damage the Insulated Gate Bipolar Transistors (IGBTs) and diodes of the power module if insufficient cooling is provided. Therefore, to guarantee the functionality of the inverter, the electro-thermal conditions in the IGBTs and diodes have been analyzed by means of an electro-thermal-mechanical model tailored for the PMaSYRM drive system. On one hand, the drive model describes the electro-mechanical interaction of the machine connected to a load profile with the 3-phase power-module and the battery providing the DC-link voltage. On the other, relations for the electro-thermal interaction within the module are implemented. Consequently, it is possible to calculate the IGBTs and diodes temperatures at any given machine operation point with regard to an assumed temperature of the housing that will be connected to the module in the final design. In the electric model, relations for switching and conduction losses, including temperature dependence, have been implemented. Using the controller block the speed in the machine model (only copper losses considered) can be regulated. The thermal model of the 3-phase power-module consists of an RC-network representing the thermal impedances of the module. Preliminarily a combined liquid cooling system for the inverter and the machine was planned. However, the loss calculations for the inverter as well as the machine showed that an air cooling system is sufficient for the drive. Different working points of the machine have been simulated to calculate peak temperatures in the IGBTs and diodes with respect to a fixed heat sink temperature. For the inverter it was defined that below ambient air temperature of 50 °C, normal operation can be carried out. At ambient temperatures above 50 °C the power of the drive needs to be limited, in order to preserve drive integrity. The most crucial working points are at low machine speeds and high phase currents because the thermal impedances in the 3-phase power-module do not sufficiently limit the temperature rise in the semiconductors. For a reliable and safe operation of the PMaSYRM the junction temperatures should not significantly exceed 125 °C. Operation at higher temperatures is possible up to 150 °C, but this condition reduces the lifetime of the power module. Table 3 reports the simulation results of maximum semiconductor temperatures at different operation points as per simulation results.
As far as the thermal design of the motor concerns, the high specific power densities of the air-cooled motor and power electronics require an accurate design of the cooling fins and system layout. The complex velocity field that arises from the integration of the e-drive with the car bodywork makes the use of simplified approaches to estimate the heat exchange (e.g. empirical correlation) unsuitable. Three-dimensional CFD (Computational Fluid Dynamics) could be instead an effective method to predict the complex thermal-velocity field, enabling a more realistic estimation of the drive thermal performance. Therefore, the cooling concept has been designed and optimized with the aid of CFD. The dynamic thermal behaviour of the particular PMaSYRM design however, has been evaluated with a customized mono-dimensional thermal model, where the Heat Transfer Coefficient (HTC) has been computed from CFD results. The cooling concept design has been carried out in two main steps. At first, an optimal drive layout (motor-inverter location) and external shell shape (fins distribution and sizing) has been found in order to maximize the heat exchange. Then further improvement of the inner design of the electrical machine has been investigated to enhance the heat flow from the machine to the external shell. The different layouts and geometries derived from the optimization steps have been numerically modelled and tested with three-dimensional CFD simulations, carried out in ANSYS Fluent. The final design of the PMaSYRM case implements asymmetric fins, designed to have higher area of heat exchange in the back of the motor where low flow velocities are observed. The power electronics are mounted so that the copper base plate (where the heat flow is exchanged) is averted from the motor. Its shell (heat sink) has a quite large number of fins and it is designed to mechanically and thermally connect the power electronics to the motor. With this solution practically, an increased thermal inertia of the inverter heat sink is achieved. Consequently, the temperature peaks under severe unsteady loads are reduced and semiconductor aging is mitigated. Moreover, the fins on the inverter-motor connection allow increasing the area of heat exchange by keeping a high compactness of the system. The layout design has been developed also to include cooling fans for boosting and suppling additional cooling flow. This could be necessary when the car suddenly stops or when severe unsteady loads appear at driving speeds with insufficient convection. The inner part of the motor has also been analysed by using a three-dimensional numerical model developed to evaluate the influence of different technical solutions to the heat exchange between solid (rotor, magnet, etc.) and inner air. The higher HTC between solid and inner air was found by introducing bladed parts connected to the rotor. During operation, the rotating blades increase the turbulence inside the motor allowing a higher HTC between the solids and the internal air. The dynamic thermal analysis has been carried out by using a customized mono-dimensional lumped thermal model based on thermal resistances, thermal capacitances, and power losses, tuned with the three-dimensional CFD numerical results. Due to the shell geometry and to the different conditions of the air flow around the shell, this element is characterized by using 23 thermal control points (i.e. locations where the thermal exchange coefficient is set, later on referred as nodes), while the stator, the rotor, the shaft, the fan, the bearings and the interior air are characterized by using 32 additional nodes, summing up to a total 55 nodes. The time dependent temperature distribution on the motor has been evaluated for three different test cases: cruise speed test, gradient test (car climbing a slope) and the WLTC driving cycle. The vehicles selected for the simulation were: an A-segment passenger car, B-segment passenger car and a light duty commercial vehicle. The results of the simulations shown that the cooling concept integrated in the PMaSYRM allows keeping the temperature under control in all the dynamic operational condition mentioned above.
Due to the performance and robustness of the designed Sliding Mode Control (SMC) algorithm all over the HySYS operating range, the same control strategy has been selected to validate the SyrNemo control algorithm. Based on the HySYS controller, the new control design has been re-tuned in order to be adjusted to the new machine electrical parameters. A Model in the Loop (MIL) based platform has been developed in Matlab/Simulink in order to validate the SyrNemo control approach. An in deep simulation analysis of the control algorithms has been carried out. The present document shows the most significant results obtained for the SyrNemo machine control. Concerning the control algorithm validation: (1) the control is stable and precise in the whole operation range, including four-quadrant operation and (2) the current control algorithm regulates the d-q axis currents according to their set points. Concerning the sensorless strategy, a hybrid algorithm has been developed including Extended Kalman Filter (EKF) at medium/high speed regions and High frequency Injection (HFI) at low speeds and standstill. The results are: (1) the EKF technique cannot be used at low speeds due to its lack of back-Electromotive force (EMF) at these operation points and (2) the HFI method achieves to correctly estimate the rotor position at low speed regions and at standstill. High Frequency Injection provokes a perturbation addition that can be translated into torque ripple and power losses, thus the use of such technique should be minimized.

WP4: Smart integration and realization of the drive

The prototyped parts for two complete SyrNemo drives are realised. These are the motor plus inverter, the housing components plus the needed test bench equipment, tooling and dummy parts as well as mock-up components. Regarding to the object structure plan two motors (object ID 144191 and 145603, i.e. prototype #1 and #2 respectively) with different rotors have been built, that is one motor containing a rotor with a smooth surface and the other one with a structured surface to see the effect of the structure on the torque ripple. To have some backup in case of demagnetization, which might occur during some uncontrolled inverter shutdowns or similar critical situations and for additional tests as the overspeed test, two additional rotors have been built separately. The two rotors use the same shaft as the original ones and this can be used to implement temperature sensors for telemetry measurements. The backup rotor with the smooth surface (object ID 145606) is used for the overspeed test and it has been equipped with strain gauges at critical positions. The one with the structured surface (object ID 145605), which might technically be the better solution, is only used for backup purposes and therefore no critical tests are planned to perform with this one. To perform the overspeed test a separate test equipment has been designed (object ID 145573). First, it was planned to perform the overspeed test in idle operation of the complete motor, but for reasons of risk minimization an external motor drives the rotor to the maximum speed. It is placed in an empty housing, which only carries the bearings and prevents parts from flying away in case of self-destructing with overspeed. The strain gauges are placed on the axial surface of the rotor. Also for the regular testing of the motor an adapter flange has been designed (object ID 143516), which can be mounted on the test benches of THIEN and Tecnalia for a variety of tests. Last part of the test equipment is used for mounting the parts of the telemetry system and to keep them fixed during operation and the measurements. The energy-providing module is screwed on a base plate, which can be fixed on the test bench. The telemetry-system consists in a small printed circuit board, plugged into the telemetry-adapter. Both parts rotate in the energy-supply-coil fixed in the stand. The first stator mock-up was built as a dummy to test and adjust all the tools for winding production and to have a lightweight part to handle. To build the stator and rotor stacks, separate tools for piling and stacking in the wished tolerances and for baking the rotor stacks with self-bonding resign were also built.
Together with the machine, the prototypes of the power electronics are realised. The power electronics motherboard has been prototyped in 5 pieces, and designed to mount on the stator housing. A variety of temperature sensors is placed on the inverter, on the housing and on top and inside of the winding to develop control algorithms and to verify the temperature model of the drive. The schematic diagram and layout of the inverter was constructed with a graphic Layout editor. The production of the PCBs and the main assembly of the components were done by subcontractors, while testing the final assembly was done at AIT. Control algorithm was developed by AIT based on the technical specification of the inverter.
The drive components have been then integrated, i.e. the joining of the inverter parts into the housing, the assembly of the system, the integration of the telemetry measurement equipment as well as the end of line tests and the synchronization of the resolver. Regarding to the object structure plan both motors (object ID 144191 and 145603) have an identical geometry from outside, that means, both use the same shaft, motor housing and inverter housing. For driving the motor both drives use the same inverter hardware also. The first steps of the integration of the drive are the mounting of the inverter into the inverter housing and the mounting of the complete motor. Both parts are connected via three motor phases and some signal lines of temperature sensors and the resolver. The inverter is fed by two cables for power supply and a CAN signal line. After assembling of the motor parts, some safety-relevant tests are performed with the motor, that is the high voltage test, an insulation test, the measurement of the ohmic resistance, the test of the temperature sensors and with the connected inverter the determination of the direction of rotation. Geometrically the position of shaft and centering ring are checked. Motor and inverter then are placed on the test bench and the back EMF is measured. Also the resolver signals are measured and the resolver is adjusted electronically. With the given voltage, supply the motor can run up to its full speed with low torque, which is the End-of-Line test. Additionally, the Over-Speed (OS) test is also performed, as complement to the machine prototype delivery procedure. This consists in driving the rotor at a rotational speed 10% higher than the maximum speed of the machine, i.e. 13,200 rpm versus a maximum machine speed of 12,000 rpm, to verify the mechanical integrity of the rotor under extreme operational conditions. The OS test has been performed with one of the additional rotors manufactured, appropriately equipped with strain gauges to measure the mechanical stress in four control points. In order to minimise the risk of carrying out such test, the rotor is driven by an external motor and placed into an empty housing, which only carries the bearings and prevents parts from flying away in case of physical damage of the rotor. The measured stress is then compared with numerical prediction, thus ensuring a level within the maximum allowed limits as per rotor design specifications. Results show how the measured stress deviation from calculations is rather limited and within the safety margin designed for the rotor.

WP5: Testing and Validation

The optimal control of the SyrNemo drive has been implemented, enabling the final implementation issues of the drive and the start-up of the testing phase. In the frame of this task the synchronous reluctance machine optimal control developed in WP3, is integrated into the test bench together with the inverter and the SyrNemo motor’s prototypes. The activities performed in this task consisted:
- test bench platform setup and adaptation (including the manufacturing of a plate flange, a coupling affix and two bearing to ensure mechanical connection of the motor to the HiL test-bench);
- integration and verification procedure of the drive within the experimental test-bench following three verification stages: (2.1) optimal software control verification, (2.2) inverter verification (i.e. coupling of the power module with the control module), and (2.3) complete drive verification.
The activities carried out under this task enabled to perform the validation test plan, whose details are presented in Task 5.2. The SyrNemo powertrain is tested following the test procedures defined by the Validation Test Plan (VTP) agreed by all partners. This covers different test configurations and test cases in order to ensure that it complies with the specifications and objectives. The tests are: (1) no load test; (2) resolver offset approximation test; (3) short circuit test; (4) over speed test; (5) current control verification, torque control verification and torque ripple assessment; (6) continuous and overload rating test; (7) efficiency requirements test; (8) driving cycles test (NEDC and WLTP); (9) electromagnetic characterization test; (10) thermal test; (11) efficiency maps derivation test; (12) Human in the Loop test.
The smooth rotor machine (i.e. Machine #1) has been tested following the full validation test plan, whereas the structured rotor machine (i.e. Machine #2) has been tested only under the specific conditions to highlight the differences between the two machines. The results have been processed after the test for first level verification, and then presented in D5.23.
The overall performance of the system is evaluated according to the relevant duty cycles, featuring appropriate cooling environment to replicate the motor’s real world operating conditions in different environments. Data are analysed to derive the driving efficiency for different load conditions, evaluating the specific high current low speed points to assess the limits of operation. The overall efficiency of the system results to be 92% on the NEDC and 89% on the WLTC, with motor peak efficiency at 96% and system peak efficiency beyond 92%. The maximum delivered power is measured at 46 kW (compared to the 56 kW of the simulation), while the maximum torque performance is measured at 125 Nm (compared to the 133 Nm of the simulation). Such deviations can be ascribed to the operation of the control algorithm, which, on purpose, did not stress current and voltage levels to ensure safe operation of the drive systems under all the conditions tested. Additionally, the electric and mechanical safety margin of the test environment kept the machine operation below the peak performance, in order not to stress the components and not damage the hardware. This resulted in an underrated measured performance of the machine, therefore it can be assumed that the simulated values are achievable with the present design. Additionally, based on the test results, a set of recommendation to improve the final design has been delivered (see task below).
The design review process of the SyrNemo includes a number of different aspects, which can be further developed to improve the energy efficiency of the system. Hereunder the list of the main aspects, which have been identified:

- Adopt a liquid cooling system with the implementation of a spiral cavity aluminium housing;
- Increase the phase current capability of the power inverter (bigger power module size);
- Install higher grade of ferrite permanent magnets in the rotor stack;
- Use a single typology of non-oriented silicon steel, in order to limit cost and make the machine more capable for series-production;
- Harmonize the permanent magnet height to reduce cost of the ferrites (so far every layer demands a different magnet size), with the possibility to have a modular PMs size;
- Implement a shortened pitch winding to decrease torque ripple and be able to use a smooth rotor surface;

Potential Impact:
Dissemination Results

The SyrNemo project has been characterised by a large share of dissemination and exploitation activities. These activities include: (1) the project website design and on-line release of up-to-date information on the project achievements and advancements; (2) participation to a number of relevant conferences and workshops, representing the consortium and providing the scientific community with the technical and scientific advancements resulting from the project execution. In summary, the dissemination results reached a worldwide audience through:

- a constantly updated website;
- the organisation of an electro-mobility workshop (initial), presentation of the final SyrNemo demonstrator and production of public dissemination material (leaflet & video);
- the presentation of the project at the TRA 2014 and the presentation of the project results at the TRA 2016 with an exhibition stand;
- the invitation for the web article “New Generation of Electric Cars for a Cleaner Europe”;
- four oral presentations of the project and of the project’s findings within scientific events;
- seven articles in conference proceedings, one poster presentation to a conference and five peer reviewed journal articles;

Exploitation results (1): PMaSYRM with ferrite PMs and SyrNemo e-drive as an innovative solution for a rear-axle e-vehicles and parallel hybrid applications (CRF)

The SyrNemo e-drive was designed either thinking to an A-segment fully electric vehicle, or to a parallel hybrid installation. Beyond this primary application, the e-drive had the target to be installed as a P4 unit. In this type of architecture, the e-machine must respect a series of dimensional constraints and significant limitations for the installation of some auxiliary components of the e-drive, such as:
- Three-phase power connector;
- Signal connector;
- Cooling ports;
- Power/signal cable routing;
- Cooling pipes routing.

This is basically due to the shape and overall dimensions of the rear cradle and the rear suspension system. Other consistent constraints in terms of dimensions and available space are fixed by the mandatory clearance between the P4 unit and:

- The ground: 210 mm;
- The rear floor: 20 mm;
- The rear cradle: 20 mm;
- The exhaust system: 30 mm;
- The fuel tank: 30 mm.

Every carmaker defined different targets for the clearance and the distance from the rear crash area, so at the end the available room is very restrictive for the P4 unit; this means that the e-drive must be as small as possible and very compact in all its auxiliary components. A valid solution to reduce the auxiliary components complexity and dimensions is to integrate as much as possible the power electronics, the electric machine and the reduction gearbox, by this way is possible to reduce the wirings and the cooling system complexity. The SyrNemo e-drive concept was devised, developed and manufactured to enhance the benefits of a fully integrated P4 unit. The goal was to reduce the overall dimension and the complexity of cabling and piping, in order to satisfy the space claim in the rear axle of a B segment parallel hybrid vehicle. The simplification of the system has a positive impact at multiple levels:

- Simplification of the layout and packaging study for the cable/pipes routing;
- The reduced number of the electrical and cooling connections reduces the complexity of the mounting process in case of production intent vehicle;
- The reduction of cable and pipe’s length, the reduction of power connector, cooling ports and cooling connections reduces both the costs and weights, and this could be a huge benefit for the business case in a production intent initiative.

As already mentioned, the main advantage of a fully integrated P4 e-drive is the reduction in size, weight and complexity. CRF, as OEM, has already studied and designed in detail this type of hybrid architecture and is able to provide target dimension and weight for a feasible installation of the e-drive in the rear axle of a B-segment vehicle. The other key parameter for a P4 unit is the weight, because the additional weight on the rear axle degrades the handling and the stability of the vehicle especially during the regenerative braking and e-AWD in low traction condition (ice, snow, sand, loose gravel). The weight for a P4 unit installed on a B segment vehicle is typically targeted at about 80 Kg, without considering power inverter, lubricant in the gearbox, power cables, cooling system pipes, and coolant liquid.

The SyrNemo e-drive presents an overall weight of 75 kg, with the DC power cables already considered, but without the reduction gearbox. The initial e-drive design was studied to be compliant with a Carraro gearbox. Considering the complete e-drive, the overall weight is 91.5 Kg, considering that this value including also the power cables and the power inverter the value is still on target for a B segment parallel hybrid vehicle. Moreover, the aluminium housing is not refined for on-vehicle installation, but with a specific study is possible to optimize the aluminium housing, with a consequent reduction of weight and dimensions. The complexity of a liquid cooling system could be an issue for the packing aspects of a hybrid vehicle, this because the cooling system for an e-drive forces to provide package protection for the coolant ports and to install a series of additional and dedicated component in the vehicle:

- Dedicated pump;
- Dedicated expansion tank;
- Dedicated piping;
- Dedicated radiator;
- Additional thermostat;
- Additional hydraulic valves.

For the project the partners decided to investigate the pros and cons of an innovative forced air cooling system applied to a P4 e-drive unit. Specific tasks and studies were dedicated to the implementation of innovative aluminium housing with customized fins profile. A lot of resources and time were invested in this task, and the synergy between the partners was excellent, since the beginning, the first results of the FEM and CFD simulations were very promising and after a series of recursive optimizations, the finned housing profile was frozen. The forced air cooling system allows reducing the complexity of the vehicle and the packaging of the P4 e-drive; moreover, the cost and weight of the additional components are neglected.
The only cons of this type of cooling system is the thermal stability of the active components, this means that the peak and continuous performance must be reduce respect to potential ones with more performing cooling systems. In particular, the peak torque working condition has to be reduced in accordance with possible over temperature in the inverter power modules and e-machine stator winding. This because in peak torque working condition the speed of the vehicle is typically low with consequently reduced cooling air flow on the finned body, moreover the joule losses are very significant because in that working condition the phase current is the highest possible (power inverter max phase current fixed at 175 Arms).

Normally, with a liquid cooling system, the peak performance is the 155÷160% of the continuous working point, whereas air cooling system allows just 125÷130%. In the case of the SyrNemo e-drive, if we consider the rated speed, it can provide a peak torque of 125 Nm and a continuous torque of 95 Nm, therefore the peak performance is 131% of continuous one.

The rare earth magnetic material, Neodymium Iron Boron, forms the basis for the traction motors used in many of today’s leading Battery and Hybrid Electric Vehicles. These magnets enable the design of motors, which offer extremely high torque densities, making them compact and lightweight, whilst also offering high efficiencies. However, there are a number of arguments that this technology may not offer the best long-term solution for use in this application. In particular, rare earth magnets are expensive, doubling or more the raw material cost of the electric motor, whilst perhaps not being particularly sustainable in terms of their mining and refinement. The main motivations for the interest in SynRM for the HEV/P-HEV and BEV applications are:

- Improved saliency ratio makes the SynRM competitive with an induction machine, particularly in terms of power factor and inverter kVA requirement;
- Small to medium size high performance drives may have simpler control using the SynRM as compared to the field oriented controlled induction machine;
- It can be operated stably down to zero speed at full load unlike an induction motor which may suffer overheating problems. In addition, SynRM appears to be more efficient at low speed than an induction machine;
- By adding appropriate amount of magnet into the rotor core, efficiency improves without having significant back-EMF and without necessary change in the stator design. Because of the existence of flux barriers, demagnetization is hard to occur if strong magnets are used. Demagnetization due to the machine overloading and high ambient temperature is a significant problem in IPMs.

The ferrite magnet and switched reluctance motors may offer the lowest cost in volume manufacture, though care must be taken not to increase system costs (power electronic converter and battery) and neither technology is yet fully proven in this application. The final choice is the ferrite permanent magnet used since the 1950s. This class of hard magnetic materials is manufactured from Iron Oxide combined with the metals Strontium, Barium or Cobalt. In order to increase their remnant flux density, companies such as TDK have had success with introducing quantities of Lanthanum; this is also a rare earth metal, however Lanthanum based ferrite costs remain much lower than for NdFeB. Ferrite PMs impact on the e-machine design could be summarized as follows:

- The major challenges of a ferrite based motor design concern the low remnant flux density and low coercivity of these magnets;
- Having only one third of the remnant flux density of NdFeB magnets, to obtain a competitive torque density requires substantial flux focusing (concentrating the flux from multiple magnets), coupled with a design which also offers reluctance torque;
- Equally, the coercivity of ferrite magnets is about one fifth to one third of the NdFeB magnets; as a result very careful design of the motor is necessary to withstand demagnetizing fields during field weakening and under short circuit conditions.

One advantage, however, is that the ferrite permanent magnets have a characteristic that their coercivity increases with temperature, being the opposite behaviours observed in materials such as NdFeB, making ferrites less sensitive to sudden demagnetisation in demanding applications. In rare earth based electric machine, the high operating temperature forces the designer to choose an NdFeB PM with a very high grade of Dysprosium and consequently with an higher price. For example, if we consider that for an electric traction machine application the PM’s working temperature could be up to 180°C, the Ferrite PMs present a significant advantage in terms of relative cost respect to the NdFeB PMs. In rare earth PMs based electric machine, the high operating temperature forces the designer to choose an NdFeB PM with a very high grade of Dysprosium. In fact, the mainstream approach to increasing the coercivity of NdFeB magnets is by adding heavy rare earth such as Dysprosium to increase its intrinsic coercivity. Grain Boundary Dysprosium Diffusion (GBD), has been developed toward enhancing the coercivity of NdFeB magnets while minimizing the use of heavy rare earths, approximately 4%. It also offers demagnetization protection around the magnet corners where they are typically most vulnerable since the finished magnet will now have the highest Dy concentration in those areas. While this approach is demonstrated to work and widely accepted in the industry with mature processes in place, it also brings along a significant price tag associated with the cost and amount of required heavy rare earth material, approximately 8-12% in automotive applications. In addition, the forecast for the 2020 PMs Market indicates a significant increase of NdFeB PMs cost, so the choice of the Ferrite PMs for the electric traction machines seems to have a solid strategic advantage.

Several scientific papers and studies highlighted and consolidated this perspective about the future of electric/hybrid mobility. The SyrNemo e-drive, in line with this analysis, provides a clear and tangible example of what are the potentialities of this material for sustainable electric/hybrid mobility. In fact, it has also been widely reported that the extraction and refinement of rare earth oxides is a potentially environmentally damaging process. This aligns with research by the automotive industry suggests that NdFeB magnets may be, per unit mass, more damaging than other materials commonly used in electrical machines. If the breakdown of materials is estimated for a representative rare earth permanent magnet electric traction motor, it can be calculated that the NdFeB magnets may be responsible for perhaps 25% of the material related greenhouse gas emissions, despite being less than 5% of the motor by mass. The environmental impact of the rare earth mining process is an already investigated and in-depth topic, seems that it has an appalling environmental impact that raises serious questions over the credibility of so-called green technology.

In last decade, China dominates the rare earth magnet materials production, but the impact on the environment in sites where the materials are extracted and post processed is dramatic. For example, the mining of the Neodymium it is contributing to a vast synthetic lake of poison in northern China. In Baotou, Inner Mongolia, the world’s largest rare earth mineral refinery pumps toxic and radioactive tailings into an adjacent artificial lake. Jamie Choi, an expert on toxics for Greenpeace China, says villagers living near the lake face horrendous health risks from the carcinogenic and radioactive waste: “There’s not one step of the rare earth mining process that is not disastrous for the environment. Ores are being extracted by pumping acid into the ground, and then they are processed using more acid and chemicals”. These aspects should guide the automotive OEM to find other viable solution for the future of the electric/hybrid mobility. The SyrNemo e-drive has been conceived and designed to investigate a valid alternative to the massive use of the rare earth PMs in electric traction machines, making it a successful trade-off between cost saving, sustainability and performance. Some carmakers started to invest effort in this solution and already produced the first hybrid vehicles with PMaSYRM electric machines. For example in the Chevrolet Volt, first-generation, the Voltec system, both motors used NdFeB magnets. In the second generation, to optimize the EV range of the system, motor A was designed with a Ferrite magnet rotor while motor B was designed with an NdFeB magnet rotor. Since motor A rotates during most operating conditions but is mostly at zero torque, it was designed with weaker magnet flux to minimize speed related losses, while motor B was designed as a stronger magnet flux machine to minimize losses at its operating points which are typically under load. GM engineers developed other ways to increase the torque capability of the machine using a topology referred to as Permanent Magnet Assisted Synchronous Reluctance Machine (PMaSynRM).

The effects of inserted magnets on the d-q inductances and their difference and ratio can be seen by comparing the inductances of the SynRM without PMs in rotor core. As it can be inferred by inserting magnets in the rotor, the inductance ratio and their difference have increased which implies increase of power factor and torque density of the motor. These improvements are caused by saturation of the tangential and radial ribs inside the rotor core due to the permanent magnets fluxes. As already explained the PMaSynRM performances improvements are caused by saturation of the tangential and radial ribs due to the permanent magnets fluxes. In the SyrNemo context, another additional solution was developed to maximize the saliency: special ThyssenKrupp non-oriented silicon steel with very high mechanical robustness (yield strength = 500 MPa). This allows reducing extremely the tangential and radial ribs thickness (down to 0.35 mm) and therefore maximising the saliency and the reluctance torque.
At the same time, the new lamination presents a higher magnetic reluctance this produces a more marked local saturation on the tangential and radial ribs. The con is that the iron losses coefficient in this type of lamination is higher but the impact on the overall efficiency is very marginal (max rotor iron losses =150 W).

Currently the parallel hybrid architecture takes its place in the hybrid vehicle scenario, and all the carmakers are focused on this solution. The most of the European carmakers already designed and commercialized a significant number of parallel hybrid vehicles with electrified rear/front axle. The most of the carmakers install in their vehicle IPM machine with NdFeB permanent magnets, this choice forces the installation of a reduction gearbox with a disconnect clutch. In fact over a predefined vehicle speed (normally 110÷130 km/h) the electric machine is disengaged, this because in terms of efficiency and energy consumption is more advantageous to stop the P4 e-drive and run the vehicle using the only internal combustion engine. Many times the gear ratio of the reduction gearbox forces the disengagement of the electric machine, in fact for this type of application the gear ratio normally is about 10÷12, therefore when the vehicle speed is about 110÷130 km/h the e-machine is already rotating at 12000÷15000 rpm. For an electric machine is possible to run at higher speed but with a considerable increasing mechanical and thermal stress on rotating parts, therefore the cost of the components could increase significantly.

During the normal operation a sudden faults could occurs at power inverter level and the e-machine is forced to run in an uncontrolled mode, this situation could be potentially destructive for a rare earth PMs based e-machine. This because the e-drive could work as uncontrolled generator, and could occurs the so called “uncontrolled generator operation” or UCG. This event occurs when the inverter stops controlling the rotating electric motor, typically because of a fault that forced the inverter to shut down. Even though the shutdown mode is by itself non-destructive, the high amplitude of the machine’s back-EMF causes current to flow back through the inverter’s freewheeling diodes to the DC link, charging the DC link capacitor ideally up to the rectified back-EMF voltage; if in this process the voltage exceeds the rated value of the capacitor or power module then eventually these components will get irreversibly damaged, making the electric drive unusable.
Electric motors with strong permanent magnets contribution have strong back EMF, and at high speeds the aforementioned event is highly likely to occur. Possible countermeasures can be taken by design, trying to achieve higher saliency ratios and reducing the magnet contribution; from this point of view, PMaSynRM with ferrite conjugates both goals, resulting in the safer design in terms of UGC mitigation. In P4 architecture the e-machine is permanently engaged to the rear axle, so the uncontrolled generator operation and all the safety aspects correlated is a tangible risk, especially if the P4 e-drive is a rare earth based e-machine (e.g. IPM machine). Moreover, this type of e-machine, due to the high remanence of the rare earth PMs, is permanently magnetized also in absence of torque request. This means that in case of permanently engaged P4, the Power Inverter is forced to implement a flux weakening strategy in case of non-operating condition of the e-machine when it is running over the base speed. This implies a waste of energy also in case of non-operational condition of the P4 e-drive. In order to prevent these issues and reduce cost and complexity of the rear e-drive module, the possible solution is to use rare earth free e-machines, and as we already demonstrated, the best trade-off between performance and costs is represented by the PMaSynRM e-machine.

In case of three-phase short-circuit, a dangerous amount of current flow through the phases of the electric machine and generating a huge electromagnetic flux that can easily demagnetize the ferrite permanent magnets. A verification of this working condition is very important because in the actual electric machine sometimes could happen that a fault of the Power Inverter or a loss of insulation can generate a dangerous short circuit. Specific FEM simulations were performed in order to quantify the three-phase short circuit current and the electromagnetic reaction in the rotor stack.
In order to reduce the impact of the short circuit on the ferrite permanent magnet an additional re-design of the rotor lamination where designed. The spin stress analysis developed on the rotor stack moreover highlighted the possibility to remove the central ribs on the second and third layers of the rotor lamination. These two feedbacks lead us to the final version of the SyrNemo electric machine rotor.

Exploitation results (2): SyrNemo for multi-purpose Hybrid Electric Vehicles (AVL)

The SyrNemo machine meets the requirements of an A-segment passenger car, targeting urban mobility as a primary objective for the electrification of road vehicles. This is achieved by combining desirable features, such as close integration of motor and power electronics to reduce architectural complexity and associated losses, and a rare-earth magnet free rotor. Beyond this application, the motor has a scalable layout, delivering higher power and torque performance to cover upper vehicle segments, up to a curb weight of 3,500 kg. The integrated thermal management for both motor an inverter together with the possibility of pure air-cooling can be a significant cost benefit for electric vehicles where water-cooling is not mandatory, especially for small power rates . The lightweight design and the integrated e-drive unit, comprising both e-motor and power electronics, increase the effectiveness of production and maintenance of electric vehicles. Experience gained by the authors in a previous automotive project for a German OEM (details cannot provided due to confidentiality commitments) shows that the PMaSYRM would be a good candidate as a traction machine as well as for a range extender unit. The concept presented in this paper certainly has the advantage of integration of motor and power electronics. In both applications, i.e. traction motor and range extender generator, the three-phase AC harness can be eliminated, thus saving weight, losses, and cost. Furthermore, the architectural simplicity greatly facilitates dismantling at the end of the product’s lifespan. Recycling of the components is easy because the bulk of the e-drive is solely made of steel, copper, ferrites and aluminium, none of which is environmentally hazardous. Moreover, the processes of dismantling and recycling of electric motors are rather mature. Apart from its use in pure electric vehicles, the presented motor couples well with internal combustion engines for hybrid vehicle applications. The only part that would need a minor adaptation is the motor and inverter housing. Hybrid drive trains aim at improving fuel efficiency by recuperation of kinetic energy when braking the vehicle, by operating of the ICE in a more efficient manner, e.g. by load point moving, and at enhancing performance and driveability of the vehicle. Typically, hybrid drive trains employ at least two different energy converters, i.e. a combustion engine together with one electric motor or more. In such a configuration, long-range energy storage (i.e. the fuel tank) and short-range energy storage (i.e. the battery) together with the respective control units complete the system. In the case of hybrid vehicles, the e-motor can deliver better driving performance in the low speed region of the ICE because of its high torque. The specific design of the PMaSYRM for minimum weight consequently increases the power density of the whole power train and reduces the specific weight (per Nm or per kW).
Besides the most common serial and parallel hybrid configurations, power-split hybrids are receiving considerable interest because of their inherent advantages. They allow nearly continuous speed ratio and several operation modes such as engine start, electric drive, hybrid drive and pure internal combustion engine drive. Dual or multi-mode hybrid transmissions employ two electric machines coupled by two planetary gear sets, while, for rear-wheel-drive vehicles, the whole hybrid power train resides in the transmission bell housing where both e-machines, the clutches and gear stages are accommodated. For this reason, only machines with very high power density, capable of running at elevated speeds, are valuable design choices. In a previous project for a Chinese OEM (details cannot be provided for confidentiality reasons), such a configuration was examined in detail. Unfortunately, the SyrNemo machine in its air-cooled version did not meet the power requirements of this specific request; nevertheless, it has been considered as secondary option because of its dimensions and speed range. In most cases, direct integration of the machine into the gearbox housing is realized with the inverter disconnected from the gearbox due to expected vibration levels at the transmission.
Since this machine can employ liquid cooling as well - for very high requirements additionally combined with an oil-spray cooling of the winding heads - it is an ideal candidate for electric rear-axle drives of HEVs, for instance SUVs with the option of electric 4WD. In a project for another Chinese OEM, such a scheme was investigated using the presented PMaSYRM, although it was not selected as final choice because of the strict “off-the-shelf” policy of the OEM itself. Nevertheless, its choice would have been justified by the fact that the machine has the proper speed and power range of a rear axis drive of such a vehicle. The efficiency map of the motor shows high efficiency in a region that naturally matches the stringent requirements for cycle efficiency over a wide torque-speed ratio, thus reducing CO2 emissions. That will become increasingly important when emission regulations will move towards Real-Driving Emissions (RDE) and emission targets within the next few years. Additionally, the large reluctance torque helps with keeping armature currents low since much of the torque comes from the magnetic anisotropy of the rotor. Furthermore, the diameter-to-length ratio of the machine is well suited for packaging the motor into the confined space near the rear axle. The close integration of e-machine and power electronics would furthermore reduce the complexity of the HV wiring harness. However, a suitable ingress protection (IP) class of the e-drive would have to be developed and validated. Liquid cooling would help to reach high performance, albeit at the expense of a long piping system. The favourable speed range of this PMaSYRM could make a gear stage at the rear axle obsolete, which would be an important benefit in terms of cost and complexity.

Exploitation results (3): beyond automotive, a SME perspective on the applications of technologies developed in SyrNemo (Thien E-Drives GmbH)

The development of the SyrNemo machine has posed a number of technical challenges which go beyond the automotive field, and that can be applied in diverse fields. More specifically, the presented motor features an in-built telemetry system for measuring temperature and mechanical stress on the rotor. Such a solution is particularly interesting for monitoring the performance of any kind of permanent-magnet motor, which strongly depends on the thermal condition of the rotor. Additionally, the telemetry system also allows efficient OS testing at the design level, improving and speeding up the delivery of final components. Another important innovation consists in the hairpin winding. Beyond the specific application presented, this technical solution provides a significant improvement in power density and efficiency of the machine, with an expertise that can be transferred to other products and to electric motors whose application field goes beyond automotive.

List of Websites:
Project Coordination and Management: Dr. Michele De Gennaro (michele.degennaro@ait.ac.at);
Full list of the partners and updated news are available online at: www.syrnemo.eu;

Final Report Summary - SYRNEMO (Synchronous Reluctance Next Generation Efficient Motors for Electric Vehicles)

Condividi questa pagina

Scarica