Overview
Automotive manufacturers are under growing pressure from regulations across the globe which are demanding substantial emissions reductions. Therefore, the number of passenger cars and commercial vehicles with electrified drivetrains need to be increased remarkably.
Understanding vehicle dynamics and customer needs
The process of optimizing a drivetrain according to individual client expectations needs a continuous dialogue among the engineering teams to precisely specify the OEM’s requirements and to agree on all details. Developers should consider the entire vehicle system and its components to understand their impact on the drivetrain. Moreover, they must have the experience and capabilities in designing electrification solutions that conform to the Functional Safety standard ISO26262. OEMs can only comply with this standard at the overall vehicle level if the drivetrain sub-system (known as an ‘item’ in ISO26262) meets all of the specific requirements.
In order to define required wheel torque as well as the highest rotational speed of the wheels and to relate them to the electric drivetrain, vehicle dynamics such as top speed, desired acceleration and maximum gradient need to be understood. They are key factors in combination with vehicle mass and wheel diameter.
This also applies to the vehicle architecture: Decoupling the electric drivetrain from the combustion engine is not possible with P0 and P1 systems. P2, P3 and P4 configurations enable pure electric driving.
Gear ratios and transmission requirements
(Figure 1). Solutions for all architectures
The design of an electric drivetrain is substantially impacted by gear ratios. Different solutions are feasible but cost, weight, package size and the required performance must be considered in advance. For example, direct drive is possible but using a single motor would be too large to package in most vehicles, while a system for providing torque to several wheels results in a more compact package that is complex and expensive. Lowest cost and highest efficiency can be achieved with a single speed transmission, but it requires an electric motor sized for maximum speed and torque.
Two or more gear ratios would reduce the motor size because less torque and speed are needed. Even so, this solution will not lower the required peak power. A multi-speed transmission offers benefits but can result in a bigger package that costs more and might be less efficient due to the additional components.
Electric motor types
Figure 2.
The most commonly used options for an automotive traction application can be seen in Figure 2: a synchronous electric motor with Interior Permanent Magnets (IPM) and an asynchronous Induction Machine (IM).
Both have advantages and disadvantages which need to be considered to define the most suitable motor type for each individual powertrain architecture: The IPM provides higher torque density for the motor which is beneficial at low speed and an efficiency advantage for city driving cycles.
With the IM configuration, the motor performs better at higher speeds and low torque which is typical for highway cruising.
An IPM needs no current in the stator windings to produce rotor magnetic flux because it is provided by its high energy NdFeB magnets. At higher speeds, however, the voltage induced in the windings has to be controlled by a flux opposing that of the magnets. Additional current in the stator windings is required that can affect overall efficiency.
An IM needs current in the stator windings so that magnetic flux in the rotor is generated. The conduction losses in stator windings and rotor bars can negatively impact the efficiency of an IM at low motor speed. However, as the speed increases, the IM stator voltage can be held at the required level by reducing the magnetizing current, which is a performance advantage compared to the IPM application.
The next thing to decide on is the type of wire for the windings: Rectangular wire is the best choice for a high-powered electric motor. It offers optimum utilization of the available slot area and shows improved thermal performance. It has been used in BorgWarner High Voltage Hairpin (HVH) motors for over 15 years, along with the patented and award-winning ‘S-wind’ technology. In comparison, round wire is an inexpensive option, but has poor thermal properties and does not make optimal use of the available stator slot area.
Defining power electronics design and specification
The electric motor characteristics, for example the number of magnetic poles, have various effects on the specifications for the power electronics. More poles usually mean higher torque density, but it may negatively impact the inverter and system efficiency. Inverters use the Pulse Width Modulation (PWM) technique to produce sinusoidal alternating motor current from a direct current battery. A high-fidelity current waveform requires the PWM frequency to be substantially higher than the desired current’s fundamental frequency. For example, with a PWM switching frequency of 10kHz, the fundamental frequency of the motor current would maximally be around 1kHz. The rotational speed of a motor with 12 magnetic poles would be 10,000rpm at 1kHz and 15,000rpm for an 8-pole motor. Increasing the top speed of the motor to 20,000rpm would reduce size and mass, but require an increase of the inverter PWM frequency to 20kHz for a 12-pole motor design. Consequences for the power electronics would be less efficiency and increased temperature, ultimately reducing their operating life, due to doubled switching losses in the power switches. To reduce these, different modulation schemes like six-step can be applied, which also increase high speed motor torque. But this solution not only requires a significantly larger DC link capacitor, it may also result in worse NVH characteristics.
The decision between an IPM or an IM solution impacts overall system behavior and has an effect on Functional Safety. In the event of an inverter error while it is connected to an energized and spinning IM all the inverter power stage switches can be switched off, bringing the magnetic flux in the rotor down immediately. The voltage induced in the stator windings is low and no torque is produced. In an IPM configuration, however, with the magnets continuously producing flux, turning off the inverter’s power stage switches at high speed can cause the electric motor to operate as an uncontrolled generator because voltage induced in the stator may be greater than the battery voltage. A substantial braking torque can act on the vehicle wheels and create critical driving conditions.
Opening a switch between battery and inverter to prevent current flow can help by removing the motor torque but the inverter’s DC bus voltage can be increased considerably by the voltage that is generated in the stator windings of the electric motor. Components of the inverter or of subsystems like DCDC converter or battery charger could be damaged.
Active Short Circuit (ASC) technology offers a possible solution by short circuiting the motor terminals using the inverter power stage. This prevents regeneration into the battery and reduces voltage-related risks.
The efficiency of inverters using Silicon IGBT and Silicon Carbide MOSFETs.
The linear on state characteristic and fast switching of the Silicon Carbide MOSFETs gives improved efficiency at part load which is beneficial for WLTP drive cycles and ultimately enables more mileage on each battery kWh, balancing the higher inverter cost. Positive effects of this technology also include a smaller inverter package, smaller vehicle cooling system size and improved aerodynamics due to less air flow over the radiator.
Finding the optimum solution
Striving for the best possible solution often involves objectives that are mutually exclusive. Reduced weight, high mileage, peak performance, compact size and low cost cannot be achieved at the same time. For optimal results, the evaluation of a potential design must be based on several weighted criteria and restrictions placed on parameters such as package space or mass. It is impossible to improve a system on the basis of only one parameter as it will most likely result in poor performance regarding other metrics.
The optimization algorithm, which enables a choice of design options that are modular and scalable to meet different customer requirements. Vehicle specifications as well as client feedback and design parameters along with further constraints for the various sub-systems have been taken into consideration before creating potential design solutions. These are further refined with the help of Genetic Algorithms and multi-physics optimization to make sure that they ultimately meet the overall goals, both technically and financially.
Conclusion
Designing an electric drivetrain requires not only a deep understanding of the entire vehicle system and its components, engineers also have to meet their clients’ wishes and needs without losing sight of technical and financial constraints.
Contributors
- Peter Barrass, Engineering Director, BorgWarner Gateshead (UK)
- Steve Stover, Senior Manager Product Strategy, BorgWarner Noblesville (USA)
- Dave Fulton, Director of Electric Innovation, BorgWarner Noblesville (USA)
