© GLOOR ENGINEERING, CH-7434 SUFERS, 31. JULI 1996
A drive concept is being developed for a passenger car starting from the currently available battery technology and electric motors. This concept will be the basis of an Integral Drive Electronics, compactly integrating all electronic components into one unit. The objective of the work is to reduce the cost and to increase the reliability to the level of a fuel driven car.
Motorised private vehicle traffic is one of the most important factor of air pollution. A solution to this problem must above all be looked for in an adapted urbanisation policy and in the development of public transport. Low emission cars and electric vehicles can contribute, provided their energy consumption is reduced. Today's battery technology allows the usage of electric vehicles for short distances, but they are often too expensive and provide an inadequate reliability. The present design shall contribute to the promotion of the electrical vehicle. During the first phase of this project different drive systems have been evaluated and the concept of the integral drive electronic has been defined.
The target vehicle can transport two persons and meets the traffic safety requirements. Its total weight is max. 1000 kg.
The drive system can be divided into the following 4 groups:
The integral vehicle electronics is a unit which consists of all electronic components from the mains socket, over the battery up to the motor of an electric vehicle.
Various solutions will be evaluated according to the following criteria:
Robustness: The robustness of the drive system must be equal to that of a conventional car in respect to its reliability, safety and servicing.
Inexpensive: The purchase price of the drive system (including battery, electronics, motor and gear) should not exceed $4500. The maintenance cost for energy storage (service, life, and disposal) should be as low as possible.
Performance: Apart from a maximum range per battery charge the vehicle must keep up with city traffic. It must be able to climb, although at a reduced speed, under unfavourable conditions (e.g. almost exhausted battery or winter temperatures).
Driving comfort: The advantages of the electric drive system such as a low noise level and continuous acceleration from standstill should be fully utilised.
The battery is known as the weakest part of the drive system. Having no affordable alternatives to lead acid batteries, this is the starting point for our design concept. Lead acid batteries provide a low power and energy density and they have to be protected from low discharging or overcharging. Neglecting this factors results in damaged battery cells and high service expenses. The reduction of power output at low battery temperatures is also a critical attribute, which makes a battery heating system necessary.
Because of the low power density at least one third of the total car mass is used up for batteries. An approximately .400 kg battery is able to deliver about 10 kW nominal power for 30 minutes and about 7.5 kWh electrical energy even under unfavourable conditions up to the end of its life span. Regarding the battery a minimal voltage would be favourable as this reduces the number of cells connected in series and allows the usage of more robust high capacity cells. This contradicts, however, the cost of the power electronics and cabling.
The integral battery management has to monitor battery conditions during both charging and discharging periods by evaluating all information available in the system:
The charging process has to be controlled according to the battery condition, and discharging and recuperation have to be limited if the battery condition is critical.
The voltage of each battery unit will be measured to detect possibly weak cells. Additionally a spider charger recharges some of the total energy back to these cells and relief's them during the ride and ensures that their missing energy will be stored back during the charging phase. This allows not only early detection of battery problems, it also offers the possibility to fix them and keep all battery cells balanced.
Currently available charging systems suffer from two major draw backs: High harmonic contents on the line current and a charging process which is controlled by the battery voltage instead of the actual remaining energy. The available power from a 230V/10A outlet is sufficient for a complete recharge during 10 hours.
The main aim of a drive system is to move a load (vehicle and passengers). The characteristic of the motor is its torque and of the battery it is the power. Four different forces act on the drive system of a car, 3 of them depend on the car mass:
|rolling resistance||FR||=||cR g m||[N]|
|air resistance||FW||=||cW A rL/2 (v + w)²||[N]|
|gradient resistance||FS||=||s g m||[N]|
|acceleration resistance||FA||=||a (m+mZ)||[N]|
|mZ||additional masses (rotational)||100||[kg]|
|cR||rolling resistance coefficient||0.015||[-]|
|A||frontal surface area||2||[m²]|
|rL||specific weight of air||1.25||[kg/m³]|
Another important requirement for a drive system is the ability to start from a ramp (peak torque) and to climb a Long slope (continuous torque). For a gradient of 20% for example a force of 2.2 kN is necessary, whereas for driving on the flat at 80 km/h 0.4 kN is sufficient.
Vehicle load characteristic
For the acceleration the inert mass of the motor is negligible (less than 1% of the vehicle mass). Since acceleration is of short duration, the drive system can be overloaded. The overload factor can reach up to 2.5 depending on the condition of the battery, the converter and the motor. For instance, a vehicle with a weight of 1000 kg accelerates from 20 to 80 km/h in 20 seconds.
The energy difference is 260 kJ and needs a medium acceleration power of 13 kW. However it must be possible to stop the vehicle in much less time. This increased brake power must be absorbed by the mechanical brake system.
Depending on the road profile, vehicle and driving style the driving diagram varies widely. A simulation with selected realistic driving diagrams is useful for the following reasons:
The requirements of the drive system are defined as follows:
The power overload capacity of the motor and inverter corresponds to the overload capacity of the torque up to half the maximum speed (for induction motors).
Single Motor Drive: A single motor drive has the following characteristics:
Multiple Motor Drive: A multiple motor drive has the following characteristics:
The size of the motor depends on its nominal torque. This torque decreases with increasing nominal speed because of the iron losses. The required torque can be matched to the motor by a gearbox. For a given motor (fix armature stamping and power losses but variable winding) a maximum nominal power with peak efficiency can be obtained at a given nominal speed.
Induction motor with different nominal values due to different windings and cooling systems.
An increase of the efficiency from 90 to 91% corresponds to a weight reduction from 1000 to 990 kg to provide the same performance. In other words no more than 10 kg of motor mass may be spent to increase its efficiency by 1%.
A motor with a high nominal speed and a wide constant power range requires a much higher maximum speed. This implies a high mechanical performance for the transmissions. Due to the large diameter of the conductor (10 to 30 mm²) only a coarse selection of the nominal speed is possible. The limited operating time (20 to 40 minutes full load per battery charge) and the limited life span (ca. 4000 hours) allow to increase the nominal torque of the motor by a factor 1.2 to 1.6, compared with the torque necessary for continuous operation for 100'000 hours.
The nominal torque depends also on the cooling capacity. Cooling with the natural airflow requires a special positioning of the motor and the efficacy varies according to the circumstances. An in-creased nominal torque is possible with an additional temperature regulated forced ventilation (20 to 50 W power). The best torque density [Nm/kg] is achieved with water cooling (motor and converter). It has to be kept in mind, that all external methods to increase the nominal torque are at the expense of efficiency since the increased current density also increases the copper losses. The constant power range is reduced as well.
The cost of the converter depends on the number and the nominal current of the power semiconductors (switches). A step-up converter to increase the DC bus voltage means additional cost and loss in performance and is therefore omitted. The cost for the control electronics is almost independent of the motor selection.
To supply the motor with the converter output power over a wide speed range, its impedance has to be adjusted. This can be achieved as follows:
For the dimensioning of the converter (UZ = 96 VDC) at a given nominal motor power (8.5 kW), the number of phases, the efficiency and the reactive power need are important. The following estimates are given for the different drive machines:
In traction applications (railway, fork lift truck etc.) the DC machine has proved successful. The basic equations for the DC machine are:
|armature voltage||U||=||Omega Psi + R I||[V]|
The excitation with permanent magnets up to 10 kW nominal power is used in servo drives built as disk, bell or cylindric motor. Keywords for DC machines are:
|excellent and simple control||brushes for commutating (contamination and wear)|
|parallel operation of several drives is possible||heavy and expensive (commutator and magnetic material)|
|low circuit complexity (four switches)||low efficiency (below 20 kW less than 90%)|
|low maximum speed (at 10 kW up to 6000 rpm)|
|Even if the converter is switched off, blocking is possible in case of a short circuit|
Due to the availability of modern power electronics, the permanent excited schronous machine has become widely used in three-phase servo drives. The optimum torque for an applied current corresponds to the induced voltage. For construction reasons most permanent excited synchronous machines use multiple poles. The simplified vector equations for the synchronous machine are:
|torque||M||=||3 I Psi||[Nm]|
|stator voltage||U||=||Omega Psi + (R + jwL) I||[V]|
At nominal speed the induced voltage (Omega Psi) is equal to the nominal voltage. The torque decreases rapidly to zero with a further increase of the frequency. A wide range of constant power can be achieved (at the cost of additional copper and iron losses) by increasing the stator impedance with a series inductivity. The additional voltage drop at nominal current must increase equal to the induced voltage. Keywords for synchronous machines are:
|light (especially for small sizes and with the use of rare earth magnets)||available power classes are mostly below 6 kW|
|high efficiency, small reactive current||maximum speed below 10'000 rpm|
|low circuit complexity (four switches)||expensive (magnetic material cost)|
|needs sensor if commutating angle is not acquired with the current measurement|
|Even if the converter is switched off, blocking is possible in case of a short circuit|
The induction machine is the most widely used industrial motor. It is robust, cheap and standardised. The simplified basic equations are:
|torque||M||=||3 LS IW IM||[Nm]|
|pull out torque||M||=||3 (1 - h) / (2 h LS) U² / w²||[Nm]|
|stator voltage||U||=||Omega LS IM||[V]|
|IW||active current (real part)||[A]|
|IM||magnetisation current (imaginary part)||[A]|
The maximum possible torque, the pull out torque, is a key indicator for the induction motor. It depends on the magnetic leakage (stray field) at a given voltage and frequency. The pull out torque decreases with the square of the nominal speed. The ratio pull out torque / nominal torque corresponds to the constant power range. Besides a decreased efficiency a thermally intensely loaded machine has also a narrower constant power range. Keywords for the induction machine are:
|machines with low leakage and a wide field weakening range (factor 4 to 8) are available||reactive power is necessary, cos(phi) with 10 kW machines is about 0.85, in the weakening field range up to 0.95|
|no sensor needed (no vector control needed)||low efficiency for small and slow motors|
|very high speeds are possible (over 20'000 rpm)|
|increased efficiency and cos(phi) over nominal speed|
Efficiency of an induction machine.
It is mainly the open loop step motor up to 200 W which is known. For the converter driven switched reluctance motor the following basic equations are applicable:
|torque||M||=||½ dL/dPhi i²||[Nm]|
|voltage||U||=||L di/dt + dL/dPhi i Omega + R i||[V]|
|Phi||rotor twisting angle||[rad]|
The inductivity depends on the rotor twist angle and on the current (saturation). Up to saturation the torque is proportional to the square of the current. For a switched reluctance machine the current is controlled according to the required torque during the respective phase on time. When operating above the nominal speed the trigger angle has to be advanced in order to build up a magnetic field against the induced voltage. Keywords for the switched reluctance machine are:
|robust, high power / weight ratio, characteristics and efficiency are similar or better than those of the induction machine||noise and torque pulsation require a sophisticated control|
|control with less than 6 switches is possible||a position sensor is necessary|
|no rotor copper losses||little dissemination and experience in the industry, no standards.|
|good internal cooling due to salient poles|
|cheap in large scale production|
Different industry motors have been compared in order to gain an overview of the characteristics of electrical motors.
An objective theory based comparison of the different drive Systems (working principles, material characteristics etc.) is very difficult and does not take into consideration the potential of the construction with the experience of the engineer and the motor manufacturer. It is therefore recommended to compare the data of existing drive systems. The appropriate Systems can be modified according to the specific need.
|legend||DC||Direct Current motor (permanent magnets)|
|SM||Synchronous Motor (brushless DC motor)|
|IM||Induction Motor (squirrel cage)|
|SR||Switched Reluctance motor|
The induction machine and the switched reluctance machine show the best properties for electric vehicles. Drive systems with switched reluctance motors are only available as complete systems for special applications. Based on our experience with industrial induction machines we give preference to the following drive System:
single induction motor with low magnetic leakage, size 100, 4 poles, copper cage rotor, 170 mm stator diameter, 140 mm iron length, aluminium housing, weight 35 kg, forced temperature controlled ventilation, sensor for winding temperature supeedsion, isolation class F
30 Nm nominal torque up to 2700 rpm, 80 Nm peak torque up to 3400 rpm, 8.5 kW nominal power up to 10'000 rpm, 20 kW peak power from 2400 to 5000 rpm, max. speed 15'000 rpm
reduction gear transmission ratio 1:13, wheel diameter 550 mm, max. gradient 30%, top speed 80 km/h, run away speed 120 km/h
nominal motor voltage 60 VAC, nominal motor current 100 AAC, star connected, inverter max. current 320 A for 1 minute
motor efficiency 91% and cos(phi) 0.85 at nominal operation, inverter losses 600 W at nominal output, total efficiency from the battery to the wheels ca. 80%.
All drive subsystems are connected through the central power electronic unit. Its main purpose is to adapt and control the power from the source to the sink. For this, the integral drive electronic unit combines the inverter, the battery charger, the on board power supply system, the spider charger and a central controller, which includes the battery management System. The unit is equipped with all necessary interfaces to commonly used control sensors and may be connected over a serial link with either a digital driver interface console or a service station, e.g. a PC.
Block diagram of the integral drive electronics.
By the integration of all these electronic functions into one single unit some important advantages for both designer and user may be achieved:
The on board power supply supplies 12 VDC to all standard vehicle equipment (e.g. head lights, windscreen wiper, radio etc.). It is combined with the so called spider charger, which stores up to 6% of the inverter output power back to the currently weakest battery block during charging and discharging periods. This combination is a cost optimal solution to enhance battery life.
The battery charger provides a sinusoidal line current and may be connected to any single-phase household outlet of 230V/10A. The battery charging characteristic is both controlled by the battery voltage and by the total energy consumption of the passed rides. For later optimisation and for adaptation to different lead acid batteries, the charging characteristic is programmable.
A major benefit of the integral drive electronics is the access to all system information needed for the battery management. This improves the estimation of the available energy and the prediction of the remaining operating range. Additionally it allows to monitor and to control the charging and discharging processes and to protect the batteries. Recorded battery data inform service personnel.
The inverter is implemented as a three-phase bridge with a nominal power of 14 kW. A 200% overload is accepted for 1 minute to ensure a reciprocally tuning of all drive components:
Reciprocally tuning of the drive components
The control part is built around an 80C166 CPU. whose control software is borrowed from industrial inverter designs currently developed in our labs. An universal parameter interface helps in adapting the unit to different requirements and particular operating conditions.
An optimal drive system for an electric vehicle was evaluated, which interlaces to a 96 V lead add battery, made of 8 individually controlled battery blocks. An induction motor with a wide constant power range up to 10'000 rpm was selected. This induction motor is very robust, cheap and meets the performance needs. The integration of all electronic components into a single unit saves cost and allows a comprehensive battery management thus ensuring a long battery life. With appropriate adaptation the integral drive electronic unit can also be used for fork lift trucks and other vehicles.
Schmidhauser AG, Electronic Drive Systems
© GLOOR ENGINEERING, CH-7434 SUFERS, 31. JULI 1996, update 31.12.2002