Controlling a brushless DC motor

[0002] This application is a continuation application of and claims priority to U.S. application Ser. No. 09/557,994, filed on Apr. 25, 2000, now U.S. Pat. No. 6,462,495, the entire disclosure of which is incorporated by reference.

[0003] This invention relates to systems and methods for controlling a brushless DC (direct-current) motor.

[0004] Brushless DC motors, also known as self-synchronous or electronically commutated motors, may be used in a variety of mechanical systems, including motor vehicles and aerospace systems. For example, brushless DC motors may be found in automotive engine cooling systems and in heating, ventilation and air-conditioning (HVAC) equipment. A brushless DC motor typically includes a plurality of windings (or coils) wound about a stator, and a plurality of permanent magnets mounted on a rotor. An electronic control system switches current in the stator windings in a process known as commutation. The control system senses the position of the rotor and applies a coordinated sequence of control signals to electronic switches (e.g., transistors) that control the flow of current through the stator windings. The sequential switching of current through the motor windings produces a magnetic flux that rotates the rotor. Drive currents may be applied to the motor windings continuously in a linear mode of operation, or discontinuously in a nonlinear mode of operation using, e.g., pulse width modulation (PWM) techniques.

[0005] In order to apply the proper drive currents to the stator windings, the position of the rotor with respect to the conducting (or active) stator windings must be known. Various techniques may be used to detect the position of the rotor. For example, some systems derive rotor position information from sensors (e.g., Hall-effect sensors and optical sensors) coupled to the motor shaft. Other systems derive rotor position information from the back electromotive force (EMF) voltages generated in the windings as the motor rotates.

[0006] The invention features systems and methods for controlling a polyphase brushless direct current (DC) motor having a plurality of phase windings energizable by timed application of drive voltages to a plurality of motor terminals resulting in back electromotive force (EMF) conditions at the motor terminals.

[0007] In one aspect, the invention features a system comprising a meter coupled to a motor terminal and adapted to produce a range of digital output values representative of a motor terminal signal from which a back EMF signal may be derived.

[0008] In another aspect, the invention features a system comprising a meter, and a controller coupled to the meter and operable to monitor the range of meter output values and to compute a motor commutation time based upon a monitored local minimum meter output value and a monitored local maximum meter output value.

[0009] The invention also features a system comprising a pulse width modulation (PWM) circuit, and a meter coupled to a motor terminal, synchronized with the PWM circuit and configured to produce an output representative of a motor terminal signal from which a back EMF signal may be derived.

[0010] Embodiments may include one or more of the following features.

[0011] The meter may include an analog-to-digital (A/D) converter. A buffer may be coupled between the motor terminal and the meter. The buffer preferably is configured to scale the terminal signal values to a selected range. The buffer may include a voltage divider.

[0012] The motor commutation time preferably is computed based upon an identified phase-to-phase zero crossing in a back EMF signal. The controller may be configured to store a local minimum meter output value and a local maximum meter output value every motor terminal driving cycle. The zero crossing preferably is identified when the meter output value is substantially equal to one-half of the difference between the local maximum meter output value and the local minimum meter output value.

[0013] When the meter is synchronized with a PWM circuit, the meter preferably is adapted to sample motor terminal voltages in a non-conducting motor phase at times when each conducting motor phase is being driven.

[0014] In another aspect, the invention features a method of controlling a polyphase brushless direct current (DC) motor. In accordance with this inventive method, a range of signal values representative of a motor terminal signal from which a back EMF signal may be derived is monitored. Over the range of monitored signal values, a local maximum signal value and a local minimum signal value are stored. A motor commutation time is computed based upon the stored local maximum signal value and the stored local minimum value.

[0015] Among the advantages of the invention are the following.

[0016] The invention determines when zero crossings in the back EMF signals occur based upon direct measurements of the motor terminal signals. In this way, the invention avoids the sensors and complex circuitry, such as an ASIC or a plurality of comparators, that typically are used to determine when zero crossings occur. Furthermore, because the local maximum back EMF signal values and the local minimum back EMF signal values may be measured periodically (e.g., once every driving cycle), the invention readily accommodates component drifts and temperature variations.

[0017] Other features and advantages of the invention will become apparent from the following description, including the drawings and the claims.

[0018] FIG. 1A is a block diagram of control system and a driver for a polyphase brushless DC motor.

[0019] FIG. 1B is a plot over time of a voltage signal at a terminal of a motor in a linear mode of operation.

[0020] FIG. 1C is a plot over time of a voltage signal at a terminal of a motor in a PWM mode of operation.

[0021] FIG. 2 is a block diagram of a control system, a driver and a buffer coupled between the control system and a polyphase brushless DC motor.

[0022] FIG. 3 is a schematic diagram of the buffer of FIG. 2.

[0023] FIG. 4A shows plots over time of signals applied to the transistors of a three-phase inverter driving circuit and the voltages at the terminals of a brushless DC motor in a linear mode of operation.

[0024] FIG. 4B is a plot over time of digital output values of an A/D converter in the control system of FIG. 2.

[0025] FIG. 5 is a flow diagram of a method of controlling a polyphase brushless DC motor.

[0026] FIG. 6A shows plots over time of output signals of two timers configured to implement the method of FIG. 5.

[0027] FIG. 6B is a flow diagram of a method of controlling a polyphase brushless DC motor implemented using two timers.

[0028] Referring to FIG. 1A, a system 10 for controlling a polyphase brushless DC motor 12 includes a meter 14, a control circuit 16, a memory 18, and a driver interface 20. Driver interface 20 supplies control signals 22 to a driver 24, which includes a plurality of switching transistors for driving the terminals A, B, C of motor 12. In operation, during each driving cycle, two of the terminals (e.g., terminals B and C) are energized (one terminal is driven high and one terminal is driven low) and the third terminal (e.g., terminal A) is floating. The terminal drive currents are commutated to operate the motor in a stable, substantially jitterless way. The motor coils are commutated based upon information derived from back EMF voltages (V_EMF) that are measured during a floating (non-conducting) driving phase of each terminal A-C. The back EMF voltages may be measured between terminals A, B, C and a reference voltage (e.g., the voltage at the center tap N of the motor windings).

[0029] As explained in detail below, control system 10 is configured to identify “zero crossings” in the back EMF voltages based upon direct measurement of the terminal voltages. In particular, zero crossings are identified when the floating terminal voltage is halfway between a local minimum back EMF voltage and a local maximum back EMF voltage. The time span between adjacent zero crossings corresponds to the driving voltage phase delay. The motor coils preferably are commutated at a time corresponding to one half of a phase delay after each identified zero crossing; although the motor coils may be commutated at other times.

[0030] Meter 14 is coupled to terminals A, B, C of motor 12 and is configured to produce a range of output values representative of terminal voltages from which the back EMF signals may be derived. In a PWM mode of operation, meter 14 is synchronized with a PWM driving circuit 26 and samples voltage values from the non-conducting phase terminal at times when the switching transistors driving the conducting phase terminals are active. Zero crossings in the back EMF signals are computed from direct measurements of the voltages at the motor terminals. Control circuit 16 preferably stores in memory 18 a local minimum meter output value and a local maximum meter output value, and computes motor commutation times based upon the stored meter output values.

[0031] Referring to FIGS. 1B and 1C, in a linear mode of operation (FIG. 1B), each cycle of the terminal voltage has a characteristic “trapezoidal” profile over time. The measured motor terminal voltages range between a local minimum value V_Land a local maximum value V_H. Zero crossings in the back EMF signals are identified at times when the terminal voltage reaches a value that is equal to one half of the difference between the local maximum value and the local minimum value. In a PWM mode of operation (FIG. 1C), the terminal voltage includes a plurality of voltage spikes that are the result of the PWM switching. A trapezoidal profile 30, however, may be recovered from the back EMF signal by sampling the minima of the spike voltages, which correspond to the times when the on-phase switching transistors of driver 24 are active (i.e., when the conducting phases are active).

[0032] Referring to FIGS. 2 and 3, in one embodiment, a motor control system 40 includes an analog-to-digital (A/D) converter 42 that samples the voltages at terminals A-C of motor 12 and produces a digital output 44 that is representative of the sampled voltage values. An input buffer 46 is coupled between terminals A-C of motor 12 and the input of A/D converter 42. Buffer 46 scales the motor terminal voltages to a range (e.g., 0 volts to 5 volts) that is suitable for the inputs of A/D converter 42. As shown in FIG. 3, in one embodiment, buffer 42 includes a voltage divider 48, 50, 52 for each motor terminal A, B, C, respectively. In other embodiments, buffer 46 may include an operational amplifier or other circuit components that are configured to scale the motor terminal voltages to appropriate voltage range. Driver 24 includes a three-phase inverter 54 with a plurality of transistors 56 configured to drive the terminals A-C of motor 12 high (i.e., transistors A+, B+, C+) and low (i.e., transistors A−, B−, C−).

[0033] Referring to FIG. 4A, in one motor driving sequence, control system 40 choreographs the application of drive currents to motor terminals A, B, C, as follows: from times t₁to t₃, transistor A+ drives terminal A high, and from times t₄to t₆, transistor A− drives terminal A low; from times t₃to t₅, transistor B+ drives terminal B high, and from times t₀to t₂, transistor B− drives terminal B low; and from times t₀to t₁and from times t₅to t₆, transistor C+ drives terminal C high, and from times t₂to t₄, transistor C− drives terminal C low. The resulting voltages at terminals A, B and C have the characteristic trapezoidal profiles V_AN, V_BNand V_CN, respectively. In FIG. 4A, the winding commutation times are labeled t₀through t₆; the zero crossings preferably occur midway between the commutation times.

[0034] As shown in FIG. 4B, A/D converter 42 produces digital outputs (N_ADC) that are representative of the relative voltages produced at terminals A-C while motor 12 is being driven. The digital A/D converter output tracks the characteristic trapezoidal profile of the terminal voltages (V_AN, V_AN, V_AN). As explained above, control system 10 may identify the zero crossings by detecting when the A/D converter output reaches a value that is equal to one-half of the difference between the local maximum (N_H) and the local minimum (N_L). From the zero crossing information, control system 10 may determine the proper times to commutate the motor winding.

[0035] Referring to FIG. 5, in one embodiment, control system 10 controls a polyphase brushless DC motor as follows. The back EMF voltage (V_EMF) produced at a floating motor terminal is measured (step 60). If the measured voltage corresponds to a local minimum (step 62), the local minimum (V_L) is stored in memory 18 (step 64). If the measured voltage (V_EMF) is equal to one-half of the difference between a previously measured local maximum and a previously measured local minimum (step 66), control system 10 determines that a zero crossing has occurred. Control system then sets T₁to the time (T₂) of the previous zero crossing, and sets T₂to the current time (T), which corresponds to the most recent zero crossing (step 68). If the current time (T) follows the time (T₂) of the most recent zero crossing by one half of the time span between the last two zero crossings (½(T₂−T₁)) (step 70), the motor is commutated (step 72). If the back EMF voltage (V_EMF) corresponds to a local maximum (step 74), the local maximum (V_H) is stored in memory 18 (step 76). The process (steps 60-76) is repeated for each motor terminal and each motor commutation driving cycle.

[0036] By updating the values of the local maximum (N_H) and the local minimum (N_L) every cycle, control system 10 readily accommodates component drifts that might be caused by aging or temperature variations.

[0037] Referring to FIGS. 6A and 6B, in one embodiment, the control method of FIG. 5 may be implemented using two timers (T1, T0), as follows. After an initial zero crossing is detected (step 78), the timers T1, T0 are reset and timer T1 is started (step 80). When the next zero crossing is identified (step 82), the value of timer T1 corresponds to the period (Δt) between two adjacent zero crossings. At this time, a commutation phase delay (½Δt) is computed (step 84), timer T1 is reset (step 86), and timer T0 is started (step 88). When the value of timer T0 is equal to the computed phase delay (step 90), the appropriate motor winding is commutated (step 92) and the timer T0 is stopped and reset (step 94). The process (steps 82-94) is repeated every motor commutation cycle.

[0038] Other embodiments are within the scope of the claims. For example, although the above embodiments have been described in connection with three-phase brushless DC motors, these embodiments easily may be extended to operate with any polyphase brushless DC motor.