CN116569472A

CN116569472A - Driving circuit for brushless motor

Info

Publication number: CN116569472A
Application number: CN202180081570.6A
Authority: CN
Inventors: S·格里瑟姆; M·霍瓦特
Original assignee: Dyson Technology Ltd
Current assignee: Dyson Technology Ltd
Priority date: 2020-12-03
Filing date: 2021-11-15
Publication date: 2023-08-08
Also published as: WO2022117986A1; GB202019095D0; GB2601535A; GB2601535B

Abstract

A drive circuit for a brushless motor is described. The drive circuit includes a converter and a controller for connection to the phase windings of the motor. The converter includes a plurality of legs, each leg including a high side switch and a low side switch. Each switch includes four states: (i) on, wherein the switch is conductive in both the first direction and the second direction, (ii) D1, wherein the switch is conductive only in the first direction, (iii) D2, wherein the switch is conductive only in the second direction, and (iv) off, wherein the switch is non-conductive in both the first direction and the second direction. The controller controls the state of the switch to configure the converter in one of a plurality of configurations.

Description

Driving circuit for brushless motor

Technical Field

The present invention relates to a drive circuit for a brushless motor.

Background

The drive circuit for the brushless motor is responsible for controlling the excitation of the phase windings of the motor. When powered by an AC voltage, the drive circuit typically includes a rectifier, an active Power Factor Correction (PFC) stage, and a link capacitor. The rectifier, active PFC stage, and link capacitor together output a relatively stable DC voltage for energizing the phase windings.

Disclosure of Invention

The present invention provides a driving circuit for a brushless motor, the driving circuit comprising: a converter for connection to a phase winding of an electric machine, wherein the converter comprises a plurality of legs, each leg comprising a high side switch and a low side switch, each switch comprising four states corresponding to: (i) ON (ON), wherein the switch is conductive in both the first direction and the second direction; (ii) D1, wherein the switch is conductive in a first direction and non-conductive in a second direction; (iii) D2, wherein the switch is non-conductive in a first direction and conductive in a second direction; (iv) OFF (OFF), wherein the switch is non-conductive in both the first direction and the second direction; and a controller for controlling the state of the switch to configure the converter into one of a plurality of configurations, comprising: a second configuration in which the states of the high-side switch of the first branch and the low-side switch of the second branch are one of D1 and D2, and the states of the low-side switch of the first branch and the high-side switch of the second branch are the other of D1 and D2.

By providing a switch that can be controlled in both directions (i.e. can be conductive in both directions and can be non-conductive in both directions), the drive circuit is able to apply a voltage of either polarity to the phase windings, irrespective of the polarity of the voltage across the converter. That is, regardless of the polarity of the supply voltage, the drive circuit can apply a positive voltage to the phase windings to energize the phase windings in a first direction (e.g., left to right) and can apply a negative voltage to the phase windings to energize the phase windings in a second opposite direction (e.g., right to left). Thus, the drive circuit may be used with an AC power source without the need for a rectifier, PFC stage, or DC link capacitor.

However, having a switch that turns on in both directions does present challenges, particularly when the polarity of the supply voltage changes. Thus, the controller configures the converter in a manner that alleviates some of the problems.

The controller configures the converter to a second configuration in which the state of the high side switch of the first leg and the low side switch of the second leg is one of D1 and D2 and the state of the low side switch of the first leg and the high side switch of the second leg is the other of D1 and D2. Thus, the first pair of switches is D1 and the other pair of switches is D2.

The second configuration may be used to ensure that when the polarity of the supply voltage changes, i.e. when the supply voltage transitions through zero, the phase windings continue to be energized in the same direction. For example, when the polarity of the supply voltage is positive, the first pair of switches (i.e., the switch at D1) is forward biased and the second pair of switches (i.e., the switch at D2) is reverse biased. Conversely, when the polarity of the supply voltage is negative, the first pair of switches is now reverse biased and the second pair of switches is forward biased. Thus, the phase windings continue to be energized in the same direction despite the polarity of the supply voltage being changed.

The second configuration may also be used to ensure that when the polarity of the supply voltage changes, the phase current continues to be drawn from the phase windings. For example, the converter may have a first configuration when the polarity of the supply voltage is positive. In this first configuration, the first pair of switches is D1 and the second pair of switches is open. The first pair of switches is then forward biased by the self-induced voltage across the phase-crossing windings such that the inductive energy stored in the motor is transferred to the power supply, i.e. the self-induced voltage is opposite to the supply voltage. If the polarity of the supply voltage changes, the first pair of switches will be forward biased by the supply voltage. Therefore, when the polarity of the power supply voltage is negative, the converter may have a third configuration. In this third configuration, the first pair of switches is open and the second pair of switches is D2. The second pair of switches is then forward biased by the self-induced voltage across the phase-crossing windings so that the inductive energy stored in the motor continues to be transferred to the power supply. The second configuration may then be used to safely transition from the first configuration to the third configuration. In particular, when both pairs of switches are in diode mode, a path is continuously provided for the current in the phase windings.

While the converter may comprise additional configurations, such as the first and/or third configuration, such additional configurations should not be considered necessary merely because reference is made to a converter having the second configuration.

The controller may include an input for receiving a signal indicative of the polarity of the voltage, and the controller may configure the converter into the second configuration in response to a change in the polarity of the voltage. Thus, the converter is configured into the second configuration in response to a change in voltage polarity. By configuring the converter to the second configuration, some problems that may occur when the polarity of the supply voltage is changed may be alleviated.

The plurality of configurations may include a first configuration in which states of the high side switch of the first leg and the low side switch of the second leg are one of D1 and D2 and states of the low side switch of the first leg and the high side switch of the second leg are open, and the controller may configure the converter from the first configuration to the second configuration. As described above, the first configuration may be used to excite the phase windings in a particular direction. In response to a change in the polarity of the supply voltage, the controller may configure the converter into a second configuration such that the phase windings continue to be energized in the same direction. Alternatively, the first configuration may be used to transfer energy from the motor to a power source. For example, a switch in a diode state (i.e., D1 or D2) may be forward biased by a self-induced voltage across the phase windings, which is opposite to the supply voltage. In response to a change in the polarity of the supply voltage, the controller may configure the converter into the second configuration in preparation for safely converting to the third configuration.

The controller may include an input for receiving a signal indicative of the polarity of the voltage, and the controller may configure the converter in the first configuration such that the state of the high side switch of the first leg and the low side switch of the second leg is D1 when the polarity is positive and D2 when the polarity is negative. The choice of diode state (i.e. D1 or D2) is therefore dependent on the polarity of the voltage. Either diode state may be selected such that energy is transferred from the power source to the motor (i.e., the phase windings are energized) or from the motor to the power source.

The plurality of configurations may include a third configuration in which states of the high-side switch of the first leg and the low-side switch of the second leg are open and states of the low-side switch of the first leg and the high-side switch of the second leg are the other of D1 and D2, and the controller may configure the converter from the second configuration to the third configuration. When in the third configuration, the phase windings may be energized by a supply voltage. By transitioning from the second configuration to the third configuration, for example, if the back EMF induced in the phase windings exceeds the power supply, energy may be removed along a path that the motor is transferring to the power supply. Alternatively, the third configuration may be used to transfer energy from the motor to a power source. By switching from the second configuration to the third configuration, energy may be removed along the path from the power source to the motor, i.e. the excitation of the phase windings by the power source is prevented.

The plurality of configurations may include a fourth configuration in which states of the high-side switch and the low-side switch of the first and second branches are open, and the controller may configure the converter from the third configuration to the fourth configuration. As described above, the third configuration may be used to transfer energy from the motor to the power source. As the phase current decreases in the third configuration, the converter may safely switch to the fourth configuration when the phase current reaches zero (e.g., after a period of time or by monitoring the phase current). In case all switches are open in the fourth configuration, the driving circuit can safely switch to another state.

The controller may include an input for receiving a signal indicative of the current in the converter, and when the magnitude of the current is zero, the controller may configure the converter from the third configuration to the fourth configuration. This then ensures a secure transition from the third configuration to the fourth configuration.

Drawings

Embodiments will now be described, by way of example, with reference to the accompanying drawings, in which:

FIG. 1 is a block diagram of an electric motor system;

FIG. 2 is a schematic diagram of an electric motor system;

FIG. 3 shows in detail the different states of the motor system switch;

FIG. 4 is a first example of a configuration sequence of a converter of a motor system;

FIG. 5 is a second example of a converter configuration sequence;

FIG. 6 is a third example of a converter configuration sequence;

FIG. 7 is a fourth example of a converter configuration sequence;

FIG. 8 is a fifth example of a converter configuration sequence;

FIG. 9 shows example waveforms of supply voltage, back EMF, and phase current when the configuration sequence of FIG. 8 is employed;

FIG. 10 illustrates another example waveform of supply voltage, back EMF, and phase current when the configuration sequence of FIG. 8 is employed;

FIG. 11 is a sixth example of a converter configuration sequence;

FIG. 12 shows example waveforms of supply voltage, back EMF, and phase current when the configuration sequence of FIG. 8 or FIG. 11 is employed;

FIG. 13 illustrates example waveforms of supply voltage, back EMF, and phase current resulting from incorrect timing of phase excitation of a motor system;

fig. 14 shows in detail four arrangements of a specific configuration of the converter, which arrangement depends on the polarity of the supply voltage and the phase current; and

fig. 15 is a seventh example of a converter configuration sequence.

Detailed Description

The motor system 10 of fig. 1 and 2 includes a brushless motor 20 and a drive circuit 30. The motor system 10 is powered by an AC power source 40, such as a household power source.

Brushless motor 20 is a permanent magnet motor including, among other things, phase windings 21 and position sensor 22. The position sensor 22 senses the angular position of the rotor of the motor 20 and outputs a signal POS. Suitable examples of the position sensor 22 include a hall effect sensor or an optical encoder.

The drive circuit 30 includes a pair of power lines 31, an input filter 32, a converter 33, a voltage polarity detector 34, a current sensor 35, a current polarity detector 36, a gate driver 37, and a controller 38.

The power line 31 is intended to be connected to the live and neutral terminals of an AC power source 40. Thus, the power line 31 carries an alternating voltage.

The input filter 32 includes a capacitor C1 and an inductor L1. The capacitor C1 is used to smooth the relatively high switching effect of the converter. In addition, capacitor C1 provides storage for any energy generated by motor 2; this will be discussed in further detail below. The capacitor C1 is not required to smooth the AC voltage at the fundamental frequency. Thus, a capacitor having a relatively low capacitance can be used. The inductor L1 is used to smooth out any residual current ripple. Inductor L1 is intended to reduce ripple at the motor frequency, so a relatively low inductance inductor may be used, particularly when motor 20 is operating at a relatively high speed or has a relatively high number of poles.

The converter 33 is a single-phase full-bridge converter, sometimes referred to as an H-bridge converter. The converter 33 is connected to the phase winding 21 of the motor 20 and comprises two branches connected in parallel across the power line 31. Each leg includes a high side switch SW1, SW3 and a low side switch SW2, SW4.

Each of the switches SW1-SW4 is bi-directional and includes four states: ON (ON), D1, D2, OFF (OFF). When the state of the switch is on, the switch is conductive in both the first direction and the second direction. When the state of the switch is D1, the switch is conductive in the first direction and non-conductive in the second direction. Conversely, when the state of the switch is D2, the switch is non-conductive in the first direction and conductive in the second direction. Thus, D1 and D2 can be considered as diode states. For the particular arrangement of switches shown in fig. 2, the first direction may be considered downward (i.e., d1=downward conducting) and the second direction may be considered upward (i.e., d2=upward conducting). Finally, when the state of the switch is off, the switch is not conductive in both the first direction and the second direction.

Fig. 3 shows the different states of each switch and the equivalent circuit.

The bi-directional switch has two additional switch states compared to a MOFSET with a body diode or an IGBT with an equivalent anti-parallel diode. For example, when the MOSFET is on, the switch is turned on bi-directionally. When the MOSFET is turned off, the switch continues to conduct unidirectionally due to the inherent body diode. In contrast to the bi-directional switches described above, a MOSFET does not have an open state where the switch is non-conductive in both directions. Furthermore, while the MOSFET is capable of conducting in a first direction only when turned off (i.e., through the body diode), the switch cannot be conducting in a second opposite direction only.

As described in more detail below, an advantage of providing the converter 33 with a bi-directional switch is that the switch can be controlled regardless of the polarity of the voltage on the power line so that a voltage of either polarity can be applied to the phase windings. Thus, the drive circuit can be used with an AC power source without the need for a rectifier. However, the absence of anti-parallel diodes presents challenges when managing the inductive energy stored in the motor as well as any energy that the motor may produce.

Each switch may comprise a gallium nitride switch having a relatively high breakdown voltage and thus being well suited for operation at a supply voltage. However, other types of bi-directional switches that can be controlled in both directions may be used.

The voltage polarity detector 34 detects the polarity of the power supply voltage and outputs a signal v_pol. V_pol may be a digital signal that is logically high when the polarity of the supply voltage is, for example, positive, and logically low when the polarity is negative. Suitable examples of voltage polarity detector 34 include a ground-referenced comparator or a commercially available integrated package, possibly providing electrical isolation.

The current sensor 35 includes a pair of sense resistors R1, R2, each resistor being located on a branch of the converter. The voltage across the SENSE resistors R1, R2 is output as current SENSE signals i_sense1 and i_sense 2. These signals provide a measure of the current in the converter and phase windings. While the current sensor includes a pair of sense resistors, it should be appreciated that other means may be used to sense the current in the converter and phase windings, such as a current transducer or current transformer.

The current polarity detector 36 detects the polarity of the current in the converter 33 and the phase winding 21 and outputs a signal i_pol. I_pol may be a digital signal that is logically high when the polarity of the current is positive, and logically low when the polarity is negative, for example. In the schematic diagram of fig. 2, the polarity of the current may be said to be positive when the current flows through the phase winding 21 in a left-to-right direction, and negative when the current flows through the phase winding 21 in a right-to-left direction. In one example, the current polarity detector 36 may include a comparator having I_SENSE1 and I_SENSE2 as inputs.

The gate driver 37 is responsible for driving the switches SW1-SW4 of the converter 33. The gate driver 37 of the illustrated embodiment includes a pair of half-bridge drivers 37a, 37b, each half-bridge driver 37a, 37b being responsible for driving the switches of a respective leg of the converter 33. However, it is envisioned that door drive 37 may comprise a single full-bridge drive. Each of the half-bridge drivers 37a, 37b includes a plurality of inputs for receiving input signals and a plurality of outputs for outputting gate signals. In response to the input signals, the half-bridge drivers 37a, 37b generate gate signals for driving the switches of the branches of the converter. Each switch includes two gates, so the half-bridge drivers 37a, 37b generate a pair of gate signals for each switch.

The controller 38 is responsible for controlling the operation of the motor system 10. The controller 38 includes a plurality of inputs for receiving input signals and a plurality of outputs for outputting control signals. The input signals received by the controller 38 include a position signal POS, a voltage polarity signal v_pol, a current polarity signal i_pol, and current SENSE signals i_sense1 and i_sense2. In response to the input signals, the controller 38 generates and outputs four control signals S1-S4. Each control signal is used to control the state of a respective switch SW1-SW4 of the converter 33. Each switch has four states, namely: ON (ON), D1, D2, and OFF (OFF). Thus, each control signal has four levels. The control signal is output to the gate driver 37, and the gate driver 37 drives the switches SW1 to SW4 in response. As described below, the controller 38 outputs control signals to configure the converter 33 in one of a plurality of different configurations.

The operation of the motor system 1 will now be described.

To energize the phase windings 21, the controller 38 configures the converter 33 in a configuration in which the high side switch of the first leg and the low side switch of the second leg of the converter are on, and the low side switch of the first leg and the high side switch of the second leg are off. The specific choice of switch-on depends on the desired direction of excitation and the polarity of the supply voltage, determined by the signal v_pol. For example, to energize the phase winding 21 from left to right, the switches SW1 and SW4 are turned on when the polarity of the power supply voltage is positive, and the switches SW2 and SW3 are turned on when the polarity of the power supply voltage is negative. In contrast, in order to energize the phase winding 21 from right to left, the switches SW2 and SW3 are turned on when the polarity of the power supply voltage is positive, and the switches SW1 and SW4 are turned on when the polarity of the power supply voltage is negative.

Thus, the controller 38 is able to configure the converter 33 such that the phase winding 21 can be energized in either direction, i.e. a voltage of either polarity can be applied to the phase winding 21, regardless of the polarity of the supply voltage. Thus, the drive circuit 30 is able to energize the phase windings 21 with AC supply voltage without the need for a rectifier or PFC stage.

To freewheel (freewire) the phase winding 21, the controller 38 configures the converter 33 into an additional configuration in which (i) the high side switches of the first and second legs are on and the low side switches of the first and second legs are off, or (ii) the low side switches of the first and second legs are on and the high side switches of the first and second legs are off. In both cases, a freewheeling or cyclical path is provided for the current in the phase winding 21 around the converter 33. In the first scenario, where the high-side switch is on, current freewheels around the high-side loop of converter 33. In the second scenario, where the low-side switch is on, current freewheels around the low-side loop of converter 33. In the schematic diagram shown in fig. 2, the sense resistors R1, R2 of the current sensor are located in the lower part of the branch of the converter 33. Thus, by freewheeling around the low-side loop of converter 33, controller 38 is able to sense the current during freewheeling as well as excitation. However, freewheeling around the high-side loop of converter 33 is fully possible, particularly if no current sensing is required during freewheeling or if current is otherwise sensed.

During normal operation, the controller 38 may control the converter 33 to repeatedly change between excitation and freewheel. For example, the controller 38 may configure the converter 33 to energize the phase winding 21 for a set period of time, or until the current in the phase winding 21 exceeds an upper threshold, at which point the controller 38 may configure the converter 33 to freewheel the phase winding 21. The freewheel may then continue for a set period of time, or until the current in the phase winding 21 drops below a lower threshold, at which point the controller 38 may configure the converter 33 to energize the phase winding 21. The process of sequentially exciting the phase winding 21 and freewheeling the phase winding 21 may then be repeated.

The switches SW1-SW4 of the converter 33 do not have anti-parallel diodes, which presents challenges when changing between excitation and freewheeling. For example, consider the case where the switches SW1 and SW4 are on. For freewheeling around the low-side loop of the converter 33, the switch SW1 will typically be turned off and then the switch SW2 will be turned on. However, when SW1 is turned off, no path is provided for inductive energy stored in motor 20. This problem does not occur for switches comprising body diodes or anti-parallel diodes, as the diodes continue to provide paths. However, the lack of diodes presents challenges not only to the inductive energy stored in the motor 20, but also to any energy that may be generated by the motor 20. Thus, the controller 38 configures the converter 33 in a series of configurations when moving between different operating states.

Excitation to freewheel

Let us first consider the configuration sequence when moving from excitation to freewheel.

The sequence starts with the converter 33 in a first configuration in which the high side switch of the first branch and the low side switch of the second branch of the converter are on and the low side switch of the first branch and the high side switch of the second branch are off. This first configuration causes the phase windings 21 to be energized by the supply voltage. As mentioned above, the direction of actuation (i.e. the polarity of the applied voltage) will depend on which switch is turned on and the polarity of the supply voltage.

Fig. 4 (a) shows a specific example of the converter in the first configuration. In this particular example, switches SW1 and SW4 are turned on, and switches SW2 and SW3 are turned off. The polarity of the supply voltage is positive and thus the phase windings are energized from left to right.

The controller 38 then configures the converter 33 to a second configuration in which the high side switch of the first leg and the low side switch of the second leg are on, one of the low side switch of the first leg and the high side switch of the second leg is off, and the other of the low side switch of the first leg and the high side switch of the second leg is either D1 or D2. The only change that occurs in the first configuration is that one of the previously opened switches is now D1 or D2. The specific choice of switch will depend on whether freewheeling occurs around the high-side or low-side loop of the converter. The particular choice of diode state (i.e., D1 or D2) depends on the polarity of the supply voltage and is chosen such that the switch is reverse biased by the supply voltage. Thus, if the polarity of the supply voltage is positive, the switch transitions from off to D2 (i.e., turns on upwards). Conversely, if the polarity of the supply voltage is negative, the switch transitions from off to D1 (i.e., turns on downward). The two switches are still in the on state and thus the phase winding 21 continues to be energized in the second configuration.

Fig. 4 (b) shows a specific example of the converter in the second configuration. In this particular example, the polarity of the supply voltage is positive and freewheeling will occur around the low-side loop. Therefore, the low-side switch SW2 changes from off to D2.

The controller 38 then configures the converter 33 to a third configuration in which (i) the low side switches of the first leg and the second leg are off, the high side switch of the first leg is on, the high side switch of the second leg is D1 or D2, or (ii) the high side switches of the first leg and the second leg are off, the low side switch of the second leg is on, and the low side switch of the first leg is D1 or D2. The only change that occurs in the second configuration is that one of the switches that was previously turned on is now turned off. The particular choice of switch from on to off depends on whether freewheeling occurs around the high-side or low-side loop of converter 33. When the converter 33 is in the third configuration, the phase winding 21 is no longer energized by the supply voltage and the phase current freewheels or circulates around the high-side or low-side loop of the converter 33.

Fig. 4 (c) shows a specific example of the converter in the third configuration. Freewheeling occurs around the low-side loop, so the high-side switch SW1 changes from on to off.

When in the third configuration, current in the converter 33 flows through the first switch in the on state and the second switch in the diode state (i.e., D1 or D2). For example, in the example of fig. 4 (c), the current flows downward through the switch SW4 in the on state, and upward through the switch SW2 in the diode state. With the converter 33 in the third configuration, freewheeling may continue in this manner. However, conduction losses may be lower when the switch is in the on state than in the diode state. Thus, the controller 38 may configure the converter 33 in a fourth configuration in which (i) the high-side switches of the first and second legs are on and the low-side switches of the first and second legs are off, or (ii) the low-side switches of the first and second legs are on and the high-side switches of the first and second legs are off. The only change that occurs in the third configuration is that the switch that was previously in the diode state is now in the on state. While in the fourth configuration, current continues to freewheel around the high-side or low-side loop of the converter 33. However, current now flows through the switch in the on state, thereby reducing conduction losses.

Fig. 4 (d) shows a specific example of the converter in the fourth configuration. The current continues to freewheel around the low-side loop of the converter, so both low-side switches SW2, SW4 are on.

Freewheel to excitation

Next let us consider the configuration sequence when moving from freewheel to stimulus. Furthermore, let us consider the case where the phase windings are excited in a direction that keeps the phase currents in the same direction as during freewheeling.

The sequence starts with the converter 33 in a first configuration in which (i) the high side switches of the first and second legs are on and the low side switches of the first and second legs are off, or (ii) the low side switches of the first and second legs are on and the high side switches of the first and second legs are off. The phase current then freewheels around the high-side or low-side loop of the converter 33.

Fig. 5 (a) shows a specific example of the converter in the first configuration. In this particular example, the high side switches SW1, SW3 are off and the low side switches SW2, SW4 are on. The current then freewheels around the low-side loop of the converter.

The controller 38 then configures the converter 33 to a second configuration in which (i) one of the high side switches of the first and second legs is on and the other high side switch is D1 or D2 and the low side switches of the first and second legs are off, or (ii) one of the low side switches of the first and second legs is on and the other low side switch is D1 or D2 and the high side switches of the first and second legs are off. The only change that occurs in the first configuration is that one of the previously turned on switches is now either D1 or D2. The specific choice of switch depends on the intended direction of excitation and thus on the polarity of the supply voltage and the polarity of the phase current. As described below, this particular switch is not used (i.e., is not conductive) during actuation and is eventually turned off. The diode state of the switch (i.e., D1 or D2) is selected so that the phase current continues to freewheel around either the high-side loop or the low-side loop of the converter. The choice of D1 or D2 is therefore dependent on the branch of the converter to which the switch belongs and the polarity of the phase current. When the converter is in the second configuration, the phase current continues to freewheel around the high-side or low-side loop of the converter 33. However, current now flows through the first switch in the on state and the second switch in the diode state (i.e., D1 or D2).

Fig. 5 (b) shows a specific example of the converter in the second configuration. In this particular example, switches SW1 and SW3 are off, switch SW4 is on, and switch SW2 is D2. Thus, current continues to freewheel around the low-side loop of the converter through switches SW2 and SW 4.

The controller 38 configures the converter 33 to a third configuration in which the high side switch of the first leg and the low side switch of the second leg are on, one of the low side switch of the first leg and the high side switch of the second leg is D1 or D2, and the other of the low side switch of the first leg and the high side switch of the second leg is off. The only change that occurs in the second configuration is that one switch that was previously turned off is now turned on. In addition, the switch that is turned on is on the same leg as the switch that is in the diode state (either D1 or D2). The phase winding 21 is now excited by the supply voltage by means of the two switches in the on state. The switch in the diode state is now reverse biased by the supply voltage.

Fig. 5 (c) shows a specific example of the converter in the third configuration. In this particular example, switches SW1 and SW4 are on, switch SW2 is D2, and switch SW3 is off.

It is conceivable that the excitation may continue in the third configuration without any further change. However, if the polarity of the supply voltage changes, a pass-through will occur along the branch of the converter 33 with the switch in diode state. Thus, the controller 38 configures the converter 33 to a fourth configuration in which the high side switch of the first leg and the low side switch of the second leg are on and the low side switch of the first leg and the high side switch of the second leg are off. The only change that occurs in the third configuration is that the switch that was previously in the diode state (i.e., D1 or D2) is now open.

Fig. 5 (d) shows a specific example of the converter in the fourth configuration, in which the switches SW1, SW4 are on and the switches SW2, SW3 are off.

The particular configuration sequence described herein when moving from freewheel to freewheel is opposite to the configuration sequence described above when moving from freewheel to freewheel. For example, it can be seen that the configuration sequences shown in fig. 4 and 5 are reverse complements. Thus, it can be said that the controller 38 configures the converter 33 in a first sequence to excite the phase winding 21 and configures the converter 33 in a second reverse sequence to freewheel the phase winding 21.

Freewheel to reverse excitation (reversing)

Let us next consider the configuration sequence from freewheel to reverse excitation. In this case, the phase windings are energized in a direction opposite to the phase currents, resulting in the currents in the phase windings being commutated.

Also, the sequence starts with the converter 33 in the first configuration, wherein (i) the high side switches of the first and second legs are on and the low side switches of the first and second legs are off, or (ii) the low side switches of the first and second legs are on and the high side switches of the first and second legs are off. Thus, the phase current freewheels around the high-side or low-side loop of the converter 33.

Fig. 6 (a) shows a specific example of the converter in the first configuration, in which the switches SW1, SW3 are off and the switches SW2, SW4 are on. In addition, the current freewheels in a clockwise direction around the low-side loop of the converter.

The controller 38 then configures the converter 33 to a second configuration in which (i) the high-side switches of the first and second legs are on, one of the low-side switches of the first and second legs is D1 or D2, and the other low-side switch is off, or (ii) the low-side switches of the first and second legs are on, one of the high-side switches of the first and second legs is D1 or D2, and the other high-side switch is off. The only change that occurs compared to the first configuration is that one of the previously opened switches is now D1 or D2. The specific choice of switch depends on the intended direction of excitation and thus on the polarity of the supply voltage. As described below, this particular switch is used (i.e., turned on) during actuation and is ultimately turned on. The particular choice of diode state (i.e., D1 or D2) is also dependent on the polarity of the supply voltage and is chosen such that the switch is reverse biased by the supply voltage. Thus, if the polarity of the supply voltage is positive, the switch transitions from off to D2 (i.e., turns on upwards). Conversely, if the polarity of the supply voltage is negative, the switch transitions from off to D1 (i.e., turns on downward).

Fig. 6 (b) shows a specific example of the converter in a second configuration, in which the switches SW2, SW4 are on (and thus the current continues to freewheel around the low-side loop of the converter), switch SW1 is off and switch SW3 is D2.

The controller 38 configures the converter 33 to a third configuration in which the high side switch of the first leg and the low side switch of the second leg are off, one of the low side switch of the first leg and the high side switch of the second leg is on, and the other of the low side switch of the first leg and the high side switch of the second leg is D1 or D2. The only change that occurs in the second configuration is that one of the switches that was previously turned on is now turned off. Furthermore, the switch that is open is in the same branch as the switch that is in the diode state (either D1 or D2). When the converter 33 is in the third configuration, inductive energy stored in the motor 20 is transferred to the capacitor C1.

Fig. 6 (c) shows a specific example of the converter in the third configuration, in which the switches SW1, SW4 are off, the switch SW2 is on, and the switch SW3 is D2.

Finally, the controller 38 configures the converter 33 to a fourth configuration in which the high side switch of the first leg and the low side switch of the second leg are off and the low side switch of the first leg and the high side switch of the second leg are on. The only change that occurs in the third configuration is that the switch that was previously in the diode state (i.e., D1 or D2) is now on. When the converter 33 is in the fourth configuration, the phase winding 21 is energized by the supply voltage in the opposite direction as before. Any remaining inductive energy stored in the motor 20 is transferred to the capacitor C1. When all the inductive energy is transferred to the capacitor, the current in the phase windings commutates.

Fig. 6 (d) shows a specific example of the converter in the fourth configuration, in which the switches SW1 and SW4 are off and the switches SW2 and SW3 are on. Thus, the phase windings are now energized from right to left.

Excitation to reverse excitation (reversing)

There may be situations where it is desirable or necessary to reverse the direction of excitation (i.e., commutating the phase windings) during excitation, and the controller may employ the following sequence to configure the converter.

The sequence starts with the converter 33 in a first configuration in which the high side switch of the first branch and the low side switch of the second branch of the converter are on and the low side switch of the first branch and the high side switch of the second branch are off. Thus, the phase windings are energized in a direction that depends on the particular choice of switch on and the polarity of the supply voltage.

Fig. 7 (a) shows a specific example of the converter in the first configuration. In this particular example, switches SW1, SW4 are on, and switches SW2, SW3 are off. The polarity of the supply voltage is positive and thus the phase windings are energized from left to right.

The controller 38 then configures the converter 33 to a second configuration in which the high side switch of the first leg and the low side switch of the second leg are on and the low side switch of the first leg and the high side switch of the second leg are D1 or D2. The only change that occurs in the first configuration is that the two switches that were previously opened are now D1 or D2. The particular choice of diode state (i.e., D1 or D2) depends on the polarity of the supply voltage and is chosen such that the switch is reverse biased by the supply voltage. Thus, if the polarity of the supply voltage is positive, the switch transitions from off to D2 (i.e., turns on upwards). Conversely, if the polarity of the supply voltage is negative, the switch transitions from off to D1 (i.e., turns on downward). When the converter 33 is in the second configuration, the phase windings 21 continue to be energized in the same direction as the first configuration.

Fig. 7 (b) shows a specific example of the converter in the second configuration. In this particular example, the polarity of the supply voltage is positive. Thus, switches SW1, SW4 are on and switches SW2, SW3 are D2 (i.e., turned on upward). Thus, the phase windings continue to be energized from left to right.

The controller 38 configures the converter 33 to a third configuration in which the high side switch of the first leg and the low side switch of the second leg are open and the low side switch of the first leg and the high side switch of the second leg are either D1 or D2. The only change that occurs in the second configuration is that the two switches that were previously turned on are now turned off. When the converter is in the third configuration, inductive energy stored in the motor 20 is transferred to the capacitor C1 via the diode-state switch.

Fig. 7 (c) shows a specific example of a converter in a third configuration, in which the switches SW1, SW4 are off and the switches SW2, SW3 are D2 (i.e. turned on upwards). Inductive energy stored in the motor is then transferred to a capacitor (not shown) via switches SW2 and SW 3.

Finally, the controller 38 configures the converter 33 to a fourth configuration in which the high side switch of the first leg and the low side switch of the second leg are off and the low side switch of the first leg and the high side switch of the second leg are on. The only change that occurs in the fourth configuration is that the switch that was previously in the diode state (i.e., D1 or D2) is now on. When the converter 33 is in the fourth configuration, the phase winding 21 is energized by the supply voltage in the opposite direction as before. Any remaining inductive energy stored in the motor 20 is transferred to the capacitor C1. When all the inductive energy is transferred to the capacitor, the current in the phase windings commutates.

Fig. 7 (d) shows a specific example of the converter in the fourth configuration, in which the switches SW1, SW4 are off and the switches SW2, SW3 are on. Thus, the phase windings are now energized from right to left.

Reactive current

As the rotor of the motor 20 rotates, a back EMF is induced in the phase windings 21. Near the mains voltage zero crossing, the magnitude of the back EMF may exceed the mains voltage. Thus, the amplitude and polarity of the phase currents may become uncontrolled. In addition, the phase currents at this time are mainly reactive, since in the case of a relatively small or no supply voltage, the active power is relatively small or no active power. This reactive current may be large and may affect the efficiency of the motor system. The motor 20 may be designed such that machine parameters of the motor, such as peak back EMF and phase inductance, help mitigate reactive current. However, this inevitably compromises the performance of the motor. As will now be explained, the converter may be configured to maintain control of the phase current in the vicinity of the zero crossing of the supply voltage. In this way, the design of the motor can be decoupled from reactive current considerations.

In one example, controller 38 may configure converter 33 such that all switches SW1-SW4 are open at or near the zero crossing of the supply voltage. Therefore, no phase current flows. Thus, there is no reactive power, but likewise no active power. However, loss of active power is unlikely to be a problem. The active power at or near the zero crossing of the supply voltage is low or zero even if there is a phase current, because the amplitude of the supply voltage is low or zero. Thus, the reduction in total active power due to opening all switches may be small and less likely to adversely affect the overall performance of the motor.

In motor systems with switches comprising body diodes or anti-parallel diodes, the phase currents cannot be controlled in this way. In particular, the diode continues to provide a path for reactive current even when the switch is open. Thus, the motor system described herein is able to control phase currents in ways that are not possible with other motor systems at all.

By opening all switches of the converter 33, there is now a period of time when no current is drawn from the supply voltage. As a result, the harmonic content of the current drawn from the supply voltage may increase. Thus, the controller 38 may configure the converter in a different manner so that the phase current continues to be controlled while better shaping the current drawn from the supply voltage. Two configuration sequences will now be described.

Reactive current-first configuration sequence

The first sequence starts with the converter 33 in a first configuration, in which the high side switch of the first branch and the low side switch of the second branch of the converter are on and the low side switch of the first branch and the high side switch of the second branch are off. Thus, the phase winding 21 is energized in a direction dependent on the particular choice of switch being turned on and the polarity of the supply voltage.

Fig. 8 (a) shows a specific example of the converter in the first configuration. In this particular example, switches SW1, SW4 are on, and switches SW2, SW3 are off. The polarity of the supply voltage is positive and thus the phase windings are energized from left to right.

The controller 38 monitors the magnitude of the current in the phase windings via the current SENSE signals i_sense1 and i_sense 2. In the event that the rate of change of the amplitude of the phase current is less than the threshold, the controller 38 configures the converter 33 to a second configuration in which the high side switch of the first leg and the low side switch of the second leg of the converter are D1 or D2 and the low side switch of the first leg and the high side switch of the second leg are open. The only change that occurs in the first configuration is that the two switches that were previously turned on are now either D1 or D2. The particular choice of diode state (i.e., D1 or D2) depends on the polarity of the supply voltage and is chosen such that the switch is forward biased by the supply voltage. Thus, if the polarity of the supply voltage is positive, the switch transitions from on to D1 (i.e., turns on downwards). Conversely, if the polarity of the supply voltage is negative, the switch transitions from on to D2 (i.e., turns on upwards).

Fig. 8 (b) shows a specific example of the converter in the second configuration. Since the polarity of the power supply voltage is positive, the switches SW1 and SW4 are D1, and the switches SW2 and SW3 are turned off.

The controller 38 continues to monitor the magnitude of the current in the phase winding 21. If the amplitude of the phase current reaches zero, the controller 38 configures the converter 33 into a third configuration in which all switches of the converter 33 are open. The controller maintains the converter 33 in the third configuration until the phase winding 21 is energized again.

An implementation of this particular configuration sequence will now be described with reference to fig. 9, fig. 9 showing the supply voltage, the back EMF induced in the phase windings, and the phase current.

At T0, the converter is configured in a first configuration and the phase windings are energized (e.g., left to right). In this particular example, the phase windings are energized when the back EMF is opposite in polarity to the supply voltage. Thus, the supply voltage is boosted by the back EMF. The end result is a rise in current in the phase windings. In this particular example, the polarity of the phase current is positive. If the phase windings are energized in opposite directions (e.g., right to left), the polarity of the phase currents will be negative. However, the behavior of the amplitude (i.e., absolute value) of the phase current is the same. Between T0 and T1, the magnitude of the back EMF decreases, transitions from zero, and then increases. The polarity of the back EMF has now changed and is thus opposite to the supply voltage. Thus, the rate of change of the amplitude of the phase current decreases during T0 to T1. At T1, the magnitude of the back EMF is the same as the magnitude of the supply voltage. Further, as previously described, the back EMF is opposite the supply voltage. Therefore, the rate of change of the phase current is zero at T1. Between T1 and T2, the magnitude of the back EMF is greater than the supply voltage, so the magnitude of the phase current decreases, i.e., the rate of change of the phase current magnitude is now negative. In the event that the rate of change of the amplitude of the phase current is less than a threshold value, the controller configures the converter into a second configuration. In this particular example, the threshold is zero. Therefore, when the rate of change of the amplitude of the phase current becomes negative, the converter is configured in the second configuration. In the second configuration, the previously turned on switch is now in a diode state. The switch is forward biased by a combination of supply voltage and self-induced voltage across the phase windings. Thus, the current continues to flow in the same direction as before (e.g., left to right). When the amplitude of the phase current reaches zero, the switch is now reverse biased by the back EMF. Thus, the phase current is clamped (clamped) to zero. Finally, between T2 and T3, the controller configures the converter into a third configuration. When all switches are now open, the phase current remains zero.

With this particular configuration sequence, current continues to be drawn from power supply 40, thereby improving the shape of the current waveform. However, the phase current is always controlled. In particular, the switch is configured to a diode state that clamps the current at zero and prevents the back EMF from reversing the polarity of the phase current.

Fig. 10 shows another example of implementation of this particular configuration sequence.

Again, at T0, the converter is configured in a first configuration, the phase windings are energized, and the phase currents rise. Between T0 and T1, the magnitude of the back EMF decreases, transitions from zero, and then increases. Thus, the rate of change of the amplitude of the phase current decreases during T0 to T1. At T1, the magnitude of the back EMF is the same as the magnitude of the supply voltage, so the rate of change of the phase current is zero, and the magnitude of the back EMF is now opposite to the supply voltage. Between T1 and T2, the magnitude of the back EMF is greater than the supply voltage, and thus the magnitude of the phase current drops. Again, in this particular example, the controller configures the converter to the second configuration when the rate of change of the amplitude of the phase current is less than zero. Therefore, when the rate of change of the amplitude of the phase current becomes negative, the converter is configured in the second configuration. The switch is forward biased by a combination of the supply voltage across the phase windings and the self-induced voltage, so current continues to flow. Between T1 and T2, the magnitude of the back EMF rises, peaks, and begins to fall. At T2, the amplitude of the back EMF is again the same as the supply voltage, so the rate of change of phase current is zero. From T2 to T3, the amplitude of the supply voltage is greater than the back EMF, so the amplitude of the phase current rises again. Thus, contrary to the example of fig. 9, the phase current never reaches zero. The controller then maintains the converter in the second configuration until the time of freewheeling or commutating the phase winding comes.

In the example of fig. 10, switches SW1 and SW4 may remain on all the time. In fact, it may be beneficial to keep the switch on, because conduction losses may be higher when the switch is in the diode state. Thus, the controller may configure the converter to the second configuration when (i) the rate of change of the amplitude of the phase current is less than a threshold value, and (ii) the amplitude of the phase current is less than another threshold value. In this way, the efficiency of the motor system can be improved by switching to the second configuration only in case the phase current may drop to zero. In the example shown in fig. 10, the other threshold may be set low enough so that the switches SW1 and SW4 remain on despite the instantaneous decrease in the amplitude of the phase current.

Reactive current-second configuration sequence

The second sequence starts again with the converter 33 in the first configuration, wherein the high side switch of the first branch and the low side switch of the second branch of the converter are on and the low side switch of the first branch and the high side switch of the second branch are off. Thus, the phase winding 21 is energized in a direction dependent on the particular choice of switch on and the polarity of the supply voltage.

Fig. 11 (a) shows a specific example of the converter in the first configuration. In this particular example, switches SW1, SW4 are on, and switches SW2, SW3 are off. The polarity of the supply voltage is positive and thus the phase windings are energized from left to right.

The controller 38 monitors the amplitude of the phase currents in the converter 33. In the event that the rate of change of the amplitude of the phase current is less than the threshold, the controller 38 configures the converter 33 to a second configuration in which the high side switch of the first leg and the low side switch of the second leg of the converter are on and the low side switch of the first leg and the high side switch of the second leg are D1 or D2. The only change that occurs in the first configuration is that the two switches that were previously opened are now D1 or D2. The particular choice of D1 or D2 depends on the polarity of the supply voltage and is chosen such that the switch is reverse biased by the supply voltage. Thus, if the polarity of the supply voltage is positive, the switch transitions from off to D2 (i.e., turns on upwards). Conversely, if the polarity of the supply voltage is negative, the switch transitions from off to D1 (i.e., turns on downward).

Fig. 11 (b) shows a specific example of the converter in the second configuration. Since the polarity of the power supply voltage is positive, the switches SW1 and SW4 are turned on, and the switches SW2 and SW3 are D2.

The controller 38 then configures the converter 33 to a third configuration in which the high side switch of the first leg and the low side switch of the second leg of the converter are open and the low side switch of the first leg and the high side switch of the second leg are either D1 or D2. The only change that occurs in the second configuration is that the two switches that were previously turned on are now turned off. The switch in the diode state (i.e., D1 or D2) is now forward biased by the self-induced voltage across the phase windings. The self-induced voltage is opposite to the supply voltage and the back EMF. Thus, the amplitude of the phase current decreases.

Fig. 11 (c) shows a specific example of the converter in the third configuration, in which the switches SW1, SW4 are now open and the switches SW2, SW3 continue to be D2.

Controller 38 continues to monitor the phase current in the third configuration. When the amplitude of the phase current is zero, the controller 38 configures the converter 33 into a fourth configuration in which all switches of the converter 33 are open.

It is apparent that the two configuration sequences described in this section are similar. The common concept of both sequences is that the controller 38 initially configures the converter 33 into a configuration in which a pair of switches is on (and a pair of switches is off) to energize the phase windings and drive current through the phase windings 21 in a particular direction. The controller 38 monitors the phase current and, in the event that the rate of change of the amplitude of the phase current is less than a threshold, the controller 38 configures the converter 33 into another configuration in which a pair of switches are in a diode state (and a pair of switches are open) such that the phase current continues to flow in the same direction. Although the choice of switch (i.e., SW1/SW4 or SW2/SW 3) and the choice of diode state (i.e., D1 or D2) are different in the two sequences, it is common to both that when the phase current is zero, the switch is reverse biased by the back EMF, thereby preventing reversal of the phase current polarity. Although the second sequence includes one additional configuration (i.e., the second configuration), only that additional configuration is required to safely transition from the first configuration to the third configuration.

Although the two configuration sequences are similar, there is still a difference. When the switch of the first sequence is in the diode state, the conduction direction of the switch is the same as the conduction direction of the power supply voltage (see, for example, fig. 8 (b)). Thus, if the supply voltage subsequently exceeds the back EMF, the switch is forward biased and the phase current rises as shown in the example of fig. 10. In contrast, when the switch of the second sequence is in the diode state, the conduction direction of the switch is opposite to the conduction direction of the power supply voltage (see, for example, fig. 11 (b)). Thus, if the supply voltage subsequently exceeds the back EMF, the phase current continues to clamp at zero. Another difference between the two sequences is that the self-induced voltage across the phase winding is only opposite to the back EMF when the switch of the first sequence is in the diode state. Conversely, when the switch of the second sequence is in the diode state, the self-induced voltage across the phase winding is opposite to the back EMF and the supply voltage. Thus, the phase current drops at a faster rate and the controller has a longer time to open all switches before the next event.

The two sequences are not mutually exclusive and the controller may employ one or both sequences when controlling the energization of the phase windings.

Polarity change of supply voltage

Two configuration sequences have been described that better control the phase current near zero crossings in the supply voltage. In particular, both of these sequences prevent phase current polarity reversal that may occur when the back EMF exceeds the supply voltage. As will now be explained, both configuration sequences also prevent inversion of the phase currents in case the supply voltage changes polarity during excitation.

Let us consider the first of the two configuration sequences described above, namely the configuration sequence shown in fig. 8. Fig. 12 shows an example of a scenario in which the polarity of the power supply voltage changes during excitation.

At T0, the controller configures the converter to a first configuration (e.g., SW1/SW4 on, SW2/SW3 off). Thus, the phase windings are energized and the phase current rises. At T1, the amplitude of the back EMF is the same as the amplitude of the supply voltage, so the rate of change of the phase current is zero. Between T1 and T2, the magnitude of the back EMF is greater than the supply voltage, and thus the magnitude of the phase current drops. In this particular example, the controller configures the converter to the second configuration when the rate of change of the amplitude of the phase current is less than zero. Thus, the controller configures the converter to the second configuration shortly after T1 (e.g., SW/SW4 is D1, SW2/SW3 is off). At T2, the polarity of the supply voltage is changed. Thus, the supply voltage, together with the back EMF, is opposite to the self-induced voltage across the phase windings. At T3, the phase current has fallen to zero. The switch in the diode state is now reverse biased by the back EMF and the supply voltage. Thus, the phase current is clamped to zero regardless of the polarity of the supply voltage. Then, at some time after T3, the controller configures the converter to a third configuration (i.e., SW1-SW4 are turned off).

Let us now consider a second configuration sequence, namely the configuration sequence shown in fig. 11. As will become apparent, this situation is hardly different from the situation described above for the first configuration sequence. Reference is again made to fig. 12.

At T0, the controller configures the converter to a first configuration (e.g., SW1/SW4 on, SW2/SW3 off). Thus, the phase windings are energized and the phase current rises. At T1, the amplitude of the back EMF is the same as the amplitude of the supply voltage, so the rate of change of the phase current is zero. Between T1 and T2, the magnitude of the back EMF is greater than the supply voltage, and thus the magnitude of the phase current drops. The controller configures the converter to a third configuration when the rate of change of the amplitude of the phase current is less than zero. Thus, the controller configures the converter to a second configuration (e.g., SW1/SW4 on, SW2/SW 3D 2), and to a third configuration (e.g., SW1/SW4 off, SW2/SW 3D 2) shortly after T1. The supply voltage is now, together with the back EMF, opposite the self-induced voltage across the phase winding. However, at T2, the polarity of the power supply voltage changes. The power supply voltage now acts in the same direction as the self-induced voltage. However, the magnitude of the back EMF is significantly greater than the supply voltage, so the phase current continues to drop. At T3, the phase current reaches zero. The switch in the diode state is now reverse biased by the back EMF. Thus, the phase current is clamped to zero. Then, at some time after T3, the controller configures the converter in another configuration (i.e., SW1-SW4 are turned off).

It is conceivable that a situation may occur where the power supply voltage with polarity change may exceed the back EMF while the switch is still in the diode state. While this is unlikely to occur in the case shown in fig. 12, it is envisioned that this may occur at lower speeds where the back EMF amplitude is small. In this case the switch will be forward biased by the supply voltage and the phase current will rise again. However, the polarity of the phase current is unchanged. This situation is somewhat similar to fig. 10.

Incorrect timing

The phase windings 21 may be unintentionally energized at the wrong time. For example, noise in the position signal POS may cause the controller 38 to energize the phase windings 21 at the wrong time. Incorrect timing of the actuation may cause the phase currents to flow in a direction opposite to the desired direction.

Fig. 13 shows an example in which the excitation time is incorrect. At T0, the phase windings are energized by turning on switches SW1 and SW 4. T0 should occur before the back EMF crosses zero. However, for whatever reason, T0 happens to occur just after the back EMF zero crossing. Thus, when the back EMF is opposite the supply voltage, the phase windings are energized. Further, at T0, the magnitude of the back EMF is greater than the magnitude of the supply voltage. Thus, when the phase windings are energized (i.e., when the switches SW1/SW4 are turned on), the phase currents are driven in a direction opposite to the desired direction of the back EMF. Thus, in the example of fig. 13, the phase current is negative, rather than positive, as desired.

Once the polarity of the sensed phase current is opposite to that expected, controller 38 may configure converter 33 such that the switch that was turned on is now placed in a diode state. The diode state (i.e., D1 or D2) is selected such that a path for the phase current to flow from the phase winding 21 to the capacitor C1 continues. Thus, in the example of fig. 13, the controller may configure the switches SW1 and SW4 to D1 (i.e., conduct downwardly) shortly after T0.

Between T0 and T1, the magnitude of the back EMF increases, peaks, and then decreases. However, the back EMF is greater than the supply voltage throughout T0 to T1. Thus, during the period T0 to T1, the amplitude of the phase current increases, although having a negative polarity. At T1, the amplitude of the back EMF is the same as the supply voltage, so the rate of change of phase current is zero. Between T1 and T2, the magnitude of the supply voltage is greater than the back EMF. In addition, the back EMF becomes zero and the supply voltage is raised. Thus, the amplitude of the phase current decreases during the period T1 to T2. At T2, the phase current is zero and then clamped by the switch. The controller 38 then configures the converter 33 shortly after T2 so that all switches are open.

While the controller 38 may employ the above-described configuration sequence to manage phase currents having the wrong polarity, this situation may be avoided altogether. In each of the above examples in which the phase windings are energized, the sequence begins with the configuration in which the high side switch of the first leg and the low side switch of the second leg are on, and the low side switch of the first leg and the high side switch of the second leg are off. However, the sequence may alternatively start with a configuration in which the high-side switch of the first branch and the low-side switch of the second branch are D1 or D2, and the low-side switch of the first branch and the high-side switch of the second branch are open. D1 or D2 is selected such that the switch is forward biased by the supply voltage. Since the switch is activated in the diode state instead of the on state, the phase current is prevented from flowing in the wrong direction. The controller may then monitor the phase current, i.e. via a current sense signal. In the event of a rise in phase current, the controller may revert to one of the configuration sequences described above. In particular, the controller may configure the converter in a first configuration of the above sequence, wherein the high side switch of the first leg and the low side switch of the second leg are on, and the low side switch of the first leg and the high side switch of the second leg are off. Alternatively, if the phase current does not rise, the controller may configure the converter such that all switches are open in preparation for the next event.

In this particular configuration sequence, the switch first enters a diode state before turning on. Thus, this particular sequence cannot be used when the phase current has flowed through the converter in the opposite direction (i.e. opposite to the excitation direction). As described above, the phase current may be clamped to zero after each excitation period at or near the zero crossing of the supply voltage. Thus, the configuration sequence may be used in combination with either of the two sequences described above in connection with reactive current.

Closing

It may be necessary to shut down the motor system at any time. The shutdown may occur as part of normal operation or as a response to a fault condition. If the switches of the converter comprise body diodes or anti-parallel diodes, the switch-off may only require that all switches be turned off. The diode will then provide a path for transferring the inductive energy stored in the motor to the capacitor of the input filter. For the present motor system 10, the switches SW1-SW4 do not have such diodes, so the converter 33 must be configured in such a way as to handle the inductive energy of the motor 20 before the switches SW1-SW4 can be opened.

In response to closing, the controller 38 configures the converter 33 to a first configuration in which the high side switch of the first leg and the low side switch of the second leg are D1 or D2 and the low side switch of the first leg and the high side switch of the second leg are open. The choice of placing the switch (i.e., SW1/SW4 or SW2/SW 3) in the diode state depends on the polarity of the supply voltage and the polarity of the phase current and is selected such that inductive energy stored in the motor 20 is transferred to the capacitor C1. The choice of diode state (i.e. D1 or D2) depends on the polarity of the supply voltage and is chosen such that the switch is reverse biased by the supply voltage. Thus, if the polarity of the supply voltage is positive, the switch transitions to D2 (i.e., turns on upwards). Conversely, if the polarity of the supply voltage is negative, the switch transitions to D1 (i.e., turns on downward). Thus, the first configuration has four possible arrangements, as shown in detail in fig. 14.

Depending on the configuration of the converter 33 immediately prior to shutdown, the controller 38 may configure the converter 33 in one or more pre-configurations prior to the first configuration such that the converter 33 safely converts to the first configuration. For example, if the switches SW1, SW4 are on at the off point, the controller may configure the converter in a preconfiguration with the switches SW1, SW4 on and the switches SW2, SW3 in a diode state. The controller 38 then configures the converter 33 to a first configuration in which in this case the switches SW2, SW3 are in a diode state and the switches SW1, SW4 are in an off state.

When in the first configuration, inductive energy stored in the motor 20 is transferred to the capacitor C1, and the amplitude of the phase current decreases. When the current drops to zero, the current is clamped by the switches and the controller 38 configures the converter 33 to a configuration in which all of the switches are open.

As described above, the specific choice of switch and diode states in the first configuration depends on the polarity of the supply voltage. In particular, the switch and diode states are selected such that the supply voltage is opposite to the self-induced voltage across the phase windings. However, the polarity of the supply voltage may be changed when the converter 33 is in the first configuration. In this case, the controller 38 reconfigures the converter 33 so that the power supply voltage continues to act in opposition to the self-induced voltage.

In reconfiguring the converter, the controller 38 configures the converter 33 to a second configuration in which the high side switch of the first leg and the low side switch of the second leg are one of D1 and D2 and the low side switch of the first leg and the high side switch of the second leg are the other of D1 and D2. The only change that occurs in the first configuration is that the two switches that were previously open are now in the diode state. Furthermore, the two switches have opposite diode states as the other two switches. Thus, for example, if switches SW1 and SW4 are D1, then switches SW2 and SW3 are D2 and vice versa. Thus, the converter 33 has a configuration in which all of the switches SW1 to SW4 are in diode states.

The controller 38 then configures the converter 33 to a third configuration in which the high side switch of the first leg and the low side switch of the second leg are open and the low side switch of the first leg and the high side switch of the second leg are either D1 or D2. The only change that occurs in the second configuration is that the two switches that were previously in the diode state in the first configuration are now open. Inductive energy stored in the motor 20 continues to be transferred to the capacitor C1, and thus the amplitude of the phase current decreases. Upon decreasing to zero, the current is clamped by the switches and the controller 38 configures the converter 33 to a final configuration in which all of the switches are open.

Fig. 15 shows an example sequence during shutdown. Fig. 15 (a) shows the converter in the first configuration. In this particular example, the polarity of the phase current is positive (i.e., current flows from left to right) and the polarity of the supply voltage is positive. Thus, the switches SW2, SW3 are in the diode state D2, and the switches SW1, SW4 are off. The switches SW2, SW3 are forward biased by the self-induced voltage across the phase windings, so the phase current continues to flow in a left-to-right direction, and the inductive energy stored in the motor is transferred to the capacitor. Fig. 15 (b) shows the converter in the second configuration. The previously opened switches SW1 and SW4 are now in diode state D1. As described above, the converter is configured in the second configuration in response to a change in the polarity of the supply voltage. Thus, in this example, the polarity of the supply voltage is now negative. Thus, switches SW2 and SW3 are now forward biased by the supply voltage, and switches SW1 and SW4 are reverse biased. Fig. 15 (c) shows the converter in a third configuration, in which the switches SW2, SW3 are open. Then, the switches SW1 and SW4 are forward biased by the self-induced voltage across the phase windings. Thus, the phase current continues to flow from left to right, and the inductive energy stored in the motor is transferred to the capacitor. Finally, although not shown, when the phase current drops to zero, the converter is configured in a final configuration in which all switches SW1-SW4 are open.

Door driver

As described above, in the event of a fault, it may be necessary to disconnect the motor system 10. While it is generally desirable for the controller 38 to control the shutdown sequence, a fault may exist within the controller 38. Accordingly, the door drive 37 is also configured to shut down the motor system 10 in the event of a failure. The fault may be caused by a lack or collision of control signals from the controller or by an overcurrent on one or more of the switches.

The gate driver 37 receives signals v_pol and i_pol, which provide an indication of the polarity of the supply voltage and phase current, respectively. In the event of a fault, the gate driver uses these signals to shut down the motor system 10 using the same sequence as described above, i.e. the gate driver 37 generates a gate signal sequence that depends on the voltage polarity and the current polarity.

As shown in fig. 14, the first configuration of the sequence has four possible arrangements. Thus, in response to a fault, the gate driver generates a gate signal for: (i) When the polarity of the voltage is positive and the polarity of the current is positive, driving the first pair of switches to D1 and the second pair of switches to off; (ii) When the polarity of the voltage is negative and the polarity of the current is positive, driving the first pair of switches to D2 and the second pair of switches to off; (iii) When the polarity of the voltage is positive and the polarity of the current is negative, driving the first pair of switches to off and driving the second pair of switches to D1; and (iv) when the polarity of the voltage is negative and the polarity of the current is negative, driving the first pair of switches to off and driving the second pair of switches to D2.

Although not shown in fig. 2, gate driver 37 also receives an input signal that provides a measurement of the phase current amplitude. The input signal may be analog or digital. For example, the gate driver may receive current SENSE signals i_sense1 and i_sense2. Alternatively, the input signal may be a digital signal that is logically high when the phase current is non-zero and logically low when the phase current is zero. When the amplitude of the phase current is zero, the gate driver drives all switches open using the input signal.

If the polarity of the supply voltage changes in the first configuration, the gate driver 37 configures the converter 33 into the second configuration and then into the third configuration. In the second configuration, the gate driver 37 generates gate signals for driving the first pair of switches to D1 and the second pair of switches to D2. More specifically, the gate driver 37 generates a gate signal for: (i) When the polarity of the current is positive, driving the first pair of switches to D1, driving the second pair of switches to D2, and (ii) when the polarity of the current is negative, driving the first pair of switches to D2, driving the second pair of switches to D1.

When in the third configuration, the gate driver 37 adopts the same logic as in the first configuration. The polarity of the supply voltage has changed so that the different switch pairs are in different diode states, but the logic that determines which switch and which diode state (i.e., D1 or D2) has not changed. Thus, the gate driver generates a gate signal for: (i) When the polarity of the voltage is positive and the polarity of the current is positive, driving the first pair of switches to D1 and the second pair of switches to off; (ii) When the polarity of the voltage is negative and the polarity of the current is positive, driving the first pair of switches to D2 and the second pair of switches to off; (iii) When the polarity of the voltage is positive and the polarity of the current is negative, driving the first pair of switches to off and driving the second pair of switches to D1; and (iv) when the polarity of the voltage is negative and the polarity of the current is negative, driving the first pair of switches to off and driving the second pair of switches to D2.

Although it is not unknown for the gate driver to include fault protection logic, it is neither known nor unusual for the gate driver to monitor the polarity of the voltage and/or the polarity of the current and then responsively generate the gate signal. Although the gate driver described herein generates the gate signal using the polarity of the power supply voltage and the polarity of the phase current, there are cases where the gate driver may use only the polarity of the power supply voltage or the polarity of the phase current.

With the drive circuit described herein, the AC supply voltage may be used to drive the motor without the need for a rectifier or PFC stage. This is made possible by providing the converter with a bi-directional switch. In particular, the switches may be controlled such that a voltage of either polarity may be applied to the phase windings, regardless of the polarity of the supply voltage. However, providing a bi-directional switch is not without difficulty. In particular, the absence of anti-parallel diodes presents challenges when managing the inductive energy stored in the motor as well as any energy that the motor may produce. Thus, the controller configures the converter in a different configuration sequence to ensure that the motor system safely transitions from one operating state to the next. Although various configuration sequences have been described, the controller need not employ each configuration sequence.

Claims

1. A drive circuit for a brushless motor, the drive circuit comprising:

a converter for connection to a phase winding of the electric machine, wherein the converter comprises a plurality of legs, each leg comprising a high side switch and a low side switch, and each switch comprising four states corresponding to: (i) Turning on, wherein the switch is conductive in both the first direction and the second direction; (ii) D1, wherein the switch is conductive in a first direction and non-conductive in a second direction; (iii) D2, wherein the switch is non-conductive in a first direction and conductive in a second direction; (iv) Open, wherein the switch is non-conductive in both the first direction and the second direction; and

a controller for controlling the state of the switch to configure the converter into one of a plurality of configurations, the plurality of configurations comprising:

a second configuration in which the states of the high side switch of the first leg and the low side switch of the second leg are one of D1 and D2, and the states of the low side switch of the first leg and the high side switch of the second leg are the other of D1 and D2.

2. The drive circuit of claim 1, wherein the controller includes an input for receiving a signal indicative of voltage polarity, and the controller configures the converter into the second configuration in response to a change in voltage polarity.

3. The drive circuit according to claim 1 or 2, wherein: the plurality of configurations includes a first configuration in which states of the high-side switch of the first leg and the low-side switch of the second leg are one of D1 and D2, and states of the low-side switch of the first leg and the high-side switch of the second leg are off; and the controller configures the converter from the first configuration to the second configuration.

4. A drive circuit according to claim 3, wherein the controller includes an input for receiving a signal indicative of voltage polarity, and the controller configures the converter from the first configuration to the second configuration in response to a change in voltage polarity.

5. The drive circuit of claim 3 or 4, wherein the controller includes an input for receiving a signal indicative of voltage polarity, and the controller configures the converter to the first configuration such that when the polarity is positive, the states of the high side switch of the first leg and the low side switch of the second leg are D1, and when the polarity is negative, the states of the high side switch of the first leg and the low side switch of the second leg are D2.

6. The drive circuit according to any of the preceding claims, wherein: the plurality of configurations includes a third configuration in which states of the high-side switch of the first leg and the low-side switch of the second leg are open, and states of the low-side switch of the first leg and the high-side switch of the second leg are the other of D1 and D2; and the controller configures the converter from the second configuration to the third configuration.

7. The drive circuit of claim 6, wherein the plurality of configurations includes a fourth configuration in which states of the high side switch and the low side switch of the first and second legs are open; and the controller configures the converter from the third configuration to the fourth configuration.

8. The drive circuit of claim 7, wherein the controller includes an input for receiving a signal indicative of current in the converter, and the controller configures the converter from a third configuration to a fourth configuration when the magnitude of the current is zero.