The overall performance of an electric car depends primarily on its battery system and motor drive system. The drive system of an electric car generally consists of three main parts: a controller, a power converter, and a motor. The drive system of an electric car not only requires a high torque-to-weight ratio, a wide speed range, and high reliability, but also the torque-speed characteristics of the motor are affected by the power supply, which necessitates a drive system with the widest possible high-efficiency range. The motors we use are generally DC motors, mainly permanent magnet DC motors, servo motors, and stepper motors. DC motor control is simple, performance is excellent, and DC power supplies are easy to implement. This article mainly introduces the drive and control of this type of DC motor. 1. H-bridge Drive Circuit The most widely used DC motor drive circuit is the H-bridge circuit. This drive circuit can easily achieve four-quadrant operation of the DC motor, corresponding to forward rotation, forward braking, reverse rotation, and reverse braking. Its basic schematic diagram is shown in Figure 1. In the full-bridge drive circuit, all four switching transistors operate in chopper mode. S1 and S2 form one group, and S3 and S4 form another. The states of the two groups are complementary; if one group is on, the other must be off. When S1 and S2 are on, S3 and S4 are off, and a positive voltage is applied across the motor, enabling forward or reverse braking. When S3 and S4 are on, S1 and S2 are off, and a reverse voltage is applied across the motor, enabling reverse or forward braking. During the trolley's movement, the motor needs to be continuously switched between the four quadrants, i.e., between forward and reverse rotation. This involves switching between the states where S1 and S2 are on and S3 and S4 are off, and then between the states where S1 and S2 are off and S3 and S4 are on. Theoretically, the two sets of control signals should be completely complementary in this situation. However, since actual switching devices have on and off times, absolutely complementary control logic will inevitably lead to a short circuit between the upper and lower bridge arms. For example, the lower bridge arm might be on while the upper bridge arm is off. This process can be illustrated by Figure 2. The overall performance of an electric car depends primarily on its battery system and motor drive system. The drive system of an electric car generally consists of three main parts: a controller, a power converter, and a motor. The drive system of an electric car not only requires a high torque-to-weight ratio, a wide speed range, and high reliability, but also the torque-speed characteristics of the motor are affected by the power supply, which requires the drive to have the widest possible high-efficiency range. The motors we use are generally DC motors, mainly permanent magnet DC motors, servo motors, and stepper motors. DC motor control is simple, performance is excellent, and DC power supplies are easy to implement. This article mainly introduces the drive and control of this type of DC motor. 1. H-bridge Drive Circuit The most widely used DC motor drive circuit is the H-bridge circuit. This drive circuit can easily realize the four-quadrant operation of the DC motor, corresponding to forward rotation, forward braking, reverse rotation, and reverse braking. Its basic schematic diagram is shown in Figure 1. In the full-bridge drive circuit, all four switching transistors operate in chopper mode. S1 and S2 form one group, and S3 and S4 form another. The states of the two groups are complementary; if one group is on, the other must be off. When S1 and S2 are on, S3 and S4 are off, and a positive voltage is applied across the motor, enabling forward or reverse braking. When S3 and S4 are on, S1 and S2 are off, and a reverse voltage is applied across the motor, enabling reverse or forward braking. During the trolley's movement, the motor needs to be continuously switched between the four quadrants, i.e., between forward and reverse rotation. This involves switching between the states where S1 and S2 are on and S3 and S4 are off, and then between the states where S1 and S2 are off and S3 and S4 are on. Theoretically, the two sets of control signals should be completely complementary in this situation. However, since actual switching devices have on and off times, absolutely complementary control logic will inevitably lead to a short circuit between the upper and lower bridge arms. For example, the lower bridge arm might be on while the upper bridge arm is off. This process can be illustrated by Figure 2. Therefore, in order to avoid short circuits and ensure the coordination and synchronization between the actions of each switch, the two sets of control signals are theoretically required to be inverted logically, but in practice they must be separated by a sufficient dead time. This correction process can be implemented in hardware, i.e., by adding a delay between the two sets of control signals of the upper and lower bridge arms, or it can be implemented in software (see the following text for specific methods). The drive current can flow not only through the main switch, but also through the freewheeling diode. When the motor is in braking state, the motor works in generating state, and the rotor current must flow through the freewheeling diode, otherwise the motor will heat up and burn out in severe cases. The selection of the switch has a great impact on the drive circuit. The selection of the switch should follow the following principles: (1) Since the drive circuit is a power output, the output power of the switch should be large; (2) The turn-on and turn-off time of the switch should be as small as possible; (3) The power supply voltage used by the car is not high, so the saturation voltage drop of the switch should be as low as possible. In actual production, we chose the high-power Darlington transistor TIP122 or the MOSFET IRF530, both of which performed well. To simplify the circuit, it is recommended to use a dedicated motor driver chip with integrated bridge circuitry, such as L298 or LMD18200, which offers stable and reliable performance. Since the motor significantly interferes with the power supply during normal operation, using only one power supply would affect the microcontroller's normal operation. Therefore, we opted for dual power supplies: one 5V supply for the microcontroller and control circuit, and another 9V supply for the motor. Optical couplers are used to separate the control section and the motor drive section to avoid affecting the power quality of the control section. A transistor is added to the base of the Darlington transistor to provide a sufficiently large base current. Figure 3 shows the motor drive circuit using TIP122. When IOB8 is "0" and a PWM wave is input to IOB9, the motor rotates forward. The motor speed can be adjusted by changing the PWM duty cycle. When IOB9 is "0" and a PWM wave is input to IOB8, the motor reverses direction. The motor speed is adjusted by changing the PWM duty cycle. Figure 4 shows a motor drive circuit using the L298 dedicated chip, which integrates two bridge circuits. The L298 is a dedicated chip for driving two-phase and four-phase stepper motors. We utilize its internal bridge circuit to drive DC motors, a method with several advantages. Each set of PWM waves controls the speed of one motor, while the other two I/O ports control the forward and reverse rotation of the motor. The control is relatively simple, and the circuit is also simple. One chip contains eight power transistors, simplifying the circuit complexity. As shown in the figure, IOB10 and IOB11 control the direction of the first motor, and the PWM input to IOB8 controls the speed of the first motor; IOB12 and IOB13 control the direction of the second motor, and the PWM input to IOB9 controls the speed of the second motor. The LMD18200 is an H-bridge component from National Semiconductor specifically designed for driving DC motors, integrating CMOS control circuitry and DMOS power devices on the same chip. This type of chip can instantaneously drive a current of up to 6A and normally operate at a current of up to 3A, exhibiting strong driving capability and no "shot-through" current. Furthermore, it incorporates an internal overcurrent protection measurement circuit. Overcurrent protection is achieved by simply measuring the voltage at pin 8 of the LMD18200 and comparing it to a given voltage. A motor drive circuit using the LMD18200 is shown in Figure 5. Pin 5 of the LMD18200 is the PWM input terminal. Changing the PWM duty cycle adjusts the motor speed, while changing the high/low level of pin 3 controls the motor's forward and reverse rotation. This circuit offers significant advantages over the aforementioned drive circuits: high driving power, good stability, ease of implementation, and high reliability. 2. PWM Control PWM (Pulse Width Modulation) control, typically used in conjunction with a bridge drive circuit, is very simple to implement for DC motor speed regulation and offers a wide speed range. Its principle is based on DC chopping. As shown in Figure 1, if S3 and S4 are off, and S1 and S2 are controlled by PWM, assuming high-level conduction and ignoring switching losses, the conduction time in one cycle is t, the period is T, and the waveform is shown in Figure 6. The average voltage across the motor is: U = Vcc t/T = αVcc, where α = t/T is called the duty cycle, and Vcc is the power supply voltage (power supply voltage minus the saturation voltage drop of the two switching transistors). The motor speed is proportional to the voltage across the motor, and the voltage across the motor is directly proportional to the duty cycle of the control waveform. Therefore, the motor speed is proportional to the duty cycle; the larger the duty cycle, the faster the motor rotates. When the duty cycle α = 1, the motor speed is at its maximum. The PWM control waveform can be implemented using analog or digital circuits, such as a trigger circuit built with a 555 timer. However, the duty cycle of such circuits cannot be automatically adjusted and cannot be used for automatic speed control of the vehicle. Most microcontrollers currently in use can directly output this PWM waveform or output it through timing simulation, making them most suitable for vehicle speed control. We use the SPCE061 microcontroller from Sunplus Technology, a 16-bit microcontroller with a maximum frequency of 49MHz. It provides two direct PWM outputs with adjustable frequency and 16 adjustable duty cycles, offering a wide speed control range for the motor and ease of use. The SPCE061 microcontroller has 32 I/O ports and two independent internal counters, capable of simulating PWM signal outputs of arbitrary frequency and duty cycle for motor speed control. In actual production, we believe the control signal frequency does not need to be too high, generally below 400Hz is preferable, and 16 adjustable duty cycles are sufficient for speed control. Furthermore, the duty cycle should not be too high during the vehicle's movement, and should be treated differently for straight-line movement and turning. If the speed is too high, the direction is difficult to control when turning; if the speed is too low, it wastes time. Figure 6 shows the adjustment process as needed. In the actual production of the "Simple Intelligent Electric Vehicle" in 2003, the duty cycle of our vehicle drive signal was generally below 8/16. 3. Avoiding Shot-Short Circuits Through Software As the previous analysis shows, bridge drive circuits suffer from shot-through short circuits between the upper and lower bridge arms due to the turn-on and turn-off times of the switching transistors. Shot-through short circuits easily cause the switching transistors to overheat, potentially burning them out, and also increase energy loss, wasting the vehicle's valuable energy. Since many integrated driver chips now have built-in dead-time protection (such as the LMD18200), this section mainly introduces methods to add dead time when using discrete components like switching transistors or integrated chips without dead-time protection to create drive circuits. The dead-time issue only exists during forward to reverse transitions; it doesn't exist during forward or reverse start-up, so no correction is needed. If the turn-on and turn-off times of the switching transistors are very short, or if a delay element is added to the hardware circuit, the switching transistor's loss and heat generation can be reduced. Of course, avoiding shot-through short circuits through software is the best method, as it is simple to operate and offers flexible control. Implementing dead time through software involves inserting a delay element during sudden reversals, allowing the switching transistor to turn off before turning it on again. Figure 7 shows a flowchart of software-based dead-time correction. When the switching transistor reverses direction, the direction is not immediately switched; instead, the transistor is turned off for a period of time until it is completely off before the other transistor is switched on. This turn-off time is implemented by a delay in the microcontroller software. 4. Summary The above mainly analyzed the full-bridge drive circuit for motors, which is the most commonly used speed control method for DC motors. Currently, there are many integrated circuits for motor drivers on the market, which are highly efficient, have simple circuits, and are widely used, but their driving methods are mostly the same as those of the full-bridge drive. PWM control combined with a bridge drive circuit is currently the most common method for DC motor speed control.