DE102020202318A1 - Method for operating an electrical machine - Google Patents

Abstract

The invention relates to a method for operating an electrical machine by means of a converter with one power half-bridge per phase of the electrical machine for generating phase voltages, the power half-bridges being controlled in block commutation by triggering signals (Su, Sv, Sw) for the power half-bridges in the converter are generated, wherein the converter receives a clock signal (ST) that contains switchover information (lu) about switching times of the control signals (Su, Sv, Sw), and wherein in the converter based on the clock signal (ST) and the switch information contained therein (lu ) Levels of the control signals (Su, Sv, Sw) can be switched, as well as such a converter.

Description

The present invention relates to a method for operating an electrical machine by means of a converter, a converter, a system with converter and arithmetic unit, and a computer program for its implementation.

State of the art

In the field of inexpensive motorcycles or motorcycles with typically small internal combustion engines, permanent magnet synchronous generators can be used, which are rigidly coupled to the crankshaft of the internal combustion engine. These generators are used to supply the on-board network (with battery, control unit (s), ignition, injection, etc.) with electrical energy. The generator can be designed in such a way that the maximum energy requirement of the on-board network can be made available at any time. This means that in many operating points more electrical energy is provided by the generator than is required.

If there is an oversupply of electrical energy at an operating point, overcharging of the battery must be prevented. For this purpose, the generator can be decoupled from the other components of the on-board network and put into electrical no-load or short-circuit, since no change in the excitation (as in the case of separately excited electrical machines) or the speed is possible. In the event of electrical idling, the generator rotates synchronously with the internal combustion engine without placing any significant load on it. In the event of a short circuit, the resulting electrical power is converted into heat and thus loads the internal combustion engine parasitically. Since very high voltages can arise in electrical idling (proportional to the speed), a short-circuit solution can be preferred and the resulting power loss can be accepted.

One-phase and three-phase generators are typically used in this area, but other numbers of phases are also possible. Usual power ratings range between 150 W and 500 W (with an output voltage that corresponds to the battery voltage), depending on the design and energy requirements of the on-board electrical system.

In the case of small internal combustion engines, e.g. for motorcycles, the generator can combine three functions. In addition to electrical energy for the on-board network, the massive rotor can provide the flywheel to stabilize the running behavior of the internal combustion engine. In addition, the mostly metallic markings ("teeth") can be found on the rotor, which represent a sender wheel and, together with a speed sensor, can generate a speed and crankshaft position signal for an engine control unit. This is necessary for the operation of a modern internal combustion engine by means of an electronic engine control unit. This encoder wheel can be used, for example, to record the speed of rotation using an inductive sensor or Hall sensor. This sensor then registers metallic markings ("teeth") passing by and determines the speed of rotation based on the time difference between two marks.

To carry out functions that are synchronous to the crank angle, the absolute position of the crankshaft can also be recorded, for example to correctly determine the times for injection and ignition (in the case of a gasoline engine). This can be achieved via a tooth gap on the encoder wheel, which can be recognized by the speed sensor and thus assigned to a defined crank angle.

Disclosure of the invention

According to the invention, a method for operating an electrical machine, a converter, a system with converter and arithmetic unit, and a computer program for its implementation with the features of the independent claims are proposed. Advantageous refinements are the subject matter of the subclaims and the description below.

The invention is concerned with the operation of an electrical machine, in particular a permanent magnet, by means of a converter with one power half-bridge per phase of the electrical machine for generating phase voltages. A power half-bridge can contain two semiconductor switches or power switches such as MOSFETs or IGBTs, one of which is usually referred to as a high-side switch and the other as a low-side switch and which is connected to the positive or negative connection of an on-board network connect are.

In the case of a three-phase electrical machine, three power half-bridges must be provided. However, the invention is also suitable for electrical machines with a different number of phases, even if a three-phase electrical machine is used for explanation in the context of the present invention.

By using such power converters, in particular also referred to as inverters or inverters, instead of rectifier voltage regulators, an electrical machine can be operated not only as a generator but also as a motor, which increases the efficiency of electrical energy generation and the drive train is at least slightly hybridized can.

In order to control the power converter or inverter to match the respective torque requirement and to operate the electrical machine either as a motor or generator, at least one digital control signal can be output per power half-bridge via a computing unit or a control unit. Depending on the logic level of the control signal, the upper (high level) or lower (low level) power switch of a power half-bridge can be switched on by a suitable circuit (e.g. gate driver) and the corresponding other power switch can be blocked. Accordingly, a digital control signal is required for each power half-bridge. In the case of a three-phase electrical machine, three control lines are therefore required.

In order to be able to implement the respective torque request, the converter must output leading or lagging voltage signals for the stator relative to the current position of the rotor. Correspondingly, the current rotor position and thus the crankshaft position in the case of a rotor that is rigidly mounted on the crankshaft is required to generate the control signals.

Since the crankshaft position for operating the internal combustion engine is already available as a measured variable in the engine control unit, it is advisable to use this control unit to control the converter and thus achieve hybridization without a second control unit or rotor position sensors; it is only the converter for the electrical machine necessary.

However, as long as a suitable circuit for the converter is not integrated into the motor control unit (or cannot or should not be integrated), three control lines (or generally one per phase) must be provided between the converter and the control unit.

Against this background, it is now proposed within the scope of the invention that the power half-bridges are controlled in block commutation by generating corresponding control signals for the power half-bridges in the converter. For this purpose, the converter receives a clock signal which contains switchover information about switchover times of the control signals and which is generated and transmitted, for example, by a computing unit such as the motor control unit that is separate from the converter. In the converter, based on the clock signal and the switching information contained therein, levels of the control signals are switched or the control signals are generated in such a way that they accordingly contain a change in level.

While high-performance electrical machines often use fast pulse-controlled inverter processes to generate the phase voltages of the stator, for electrical machines such as those used specifically in the field of motorcycles, comparatively low powers (e.g. 150 W to 500 W or possibly 1 kW), a pure block commutation can also be used as the control method. The resulting curves of the phase voltages are similar in shape to the voltage curve using a conventional diode rectifier. However, the power output or power consumption of the electrical machine can be adjusted as desired by the phase position of the voltages relative to the rotor position.

With block commutation, the control signals for the power half-bridges are only phase-shifted with respect to one another, but otherwise have the same shape. This now makes it possible to store this form of control signals in the converter, for example in a suitable circuit or hardware for all power half bridges or phases, and to apply only the mentioned clock signal as an input or to receive it in the converter that indicates the switchover point or switchover time from a Block in the next describes, that is, those times at which the level of a control signal should or must be changed, for example from high to low.

A block describes the interval in which there is no change in the control signals. Correspondingly, each edge of a control signal, i.e. a level change, describes a transition from one in the next block. With three phases, for example, there are six different switching states and blocks before the first block pattern is repeated one period (or one control period) later, that is, this control period completes a full cycle.

The number of periods per rotor revolution results from the number of magnets or pole pairs used in the rotor. There is one electrical period per pair of poles, so with six pairs of poles there are, for example, six periods or control periods per revolution. Since the sequence of the blocks is fixed by the geometry of the electrical machine, the sequence of the blocks does not change for different torque requirements. Only the phase position of the blocks relative to the rotor position is shifted.

The clock signal preferably also contains synchronization information that preferably specifies each full cycle of a control period of all power half-bridges, the control signals being brought into a desired state in the converter based on the clock signal and the synchronization information contained therein, preferably synchronized with a rotor position of the electrical machine. A corresponding synchronization of the control is thus possible regularly and / or when required.

It is particularly preferred if the switchover information in the clock signal for a specific switchover time of a specific drive signal corresponds to the synchronization information within a full cycle of a drive period of all power half-bridges and is embodied as pulses with a first length, the switchover information for the other switchover times of the drive signals within this The full cycle of the control period of all power half bridges are each formed as pulses with a second length different from the first. The pulse for the synchronization information is expediently longer (the first length) than the remaining pulses for the switchover information (the second length).

In the exemplary case of a three-phase electrical machine, there are a total of six successive switching times, one of which is used for synchronization and has a different pulse length than the other five. The clock signal therefore contains a longer pulse after every five short pulses. A defined block transition is assigned to this longer pulse. This means that synchronization to the rotor and crankshaft position is possible and the circuit can be brought into a known state at any time. The clock signal can be designed with short voltage drops and a high quiescent level. Depending on the implementation, a clock signal with a low quiescent level and short rising voltage peaks or pulses can also be used.

Alternatively, it is preferred if the switchover information in the clock signal for the switchover times of all control signals of a full cycle of a control period of all power half-bridges are each embodied as pulses with mutually different lengths. In particular, one or more pulses can then be used as the synchronization information within this run of the control period of all power half bridges.

With this variant, not only can a specific block change be triggered by a longer pulse, but each block change is assigned a defined pulse length. This means that you can switch to any block at any time without having to output a sequence of several pulses. Depending on the reaction speed of the converter circuit, a rapid sequence of block changes can trigger brief oscillations in the phase voltages, etc., which is avoided with this alternative, even if this requires a somewhat more complex coding of the clock signal.

Regardless of the number of different pulse lengths, an implementation is preferred in which the first, leading edge of the pulse triggers a block change to the next switching state, that is to say is used as the switchover information. The length of time up to the second, final edge of the pulse (i.e. the length of the pulse itself) is expediently used to check or evaluate whether the currently set block corresponds to the desired block identified by the pulse length, i.e. the level of the control signal it's correct. If not, a new switching process into the expected block is preferably triggered. This ensures the earliest possible switchover to the next block and at the same time enables resynchronization in the event of an error as soon as the synchronization information with the end of the pulse is available (in the case of second pulse lengths after six pulses at the latest, in the second variant after evaluation of each pulse length).

Over the entire operating range of the electrical machine, there may be various requirements for the circuit or the converter and the clock signal to be generated or received.

During a starting process, the crankshaft position of the internal combustion engine (and thus the position of the rotor of the electrical machine) are not immediately known. Correspondingly, no specific control of the converter can take place and no plausible clock signal can be provided. In this respect, it is preferred if the control signals for the power half-bridges are only generated in the converter in order to control the power half-bridges in block commutation if a clock signal is received (at all).

It is also useful that when the converter does not receive a clock signal or the clock signal does not contain any switching information, the converter is operated as a diode rectifier or all high-side or all low-side switches of the power half-bridges are switched on.

An “Enable” signal can be generated from the clock signal so that the converter is in a defined state at such an operating point during the start-up process. If there is no clock signal (or if the switchover information is not included), the "Enable" signal can correspond to a "Disable" level (e.g. low). If the clock signal is present, an "Enable" level (then e.g. High) for the "Enable" signal is derived from this and the circuit in the converter is activated. If necessary, the "Enable" signal can only take on the "Enable" level when a synchronization pulse (e.g. a longer pulse, as already mentioned) is specified for the first time.

If the level is “Disable”, the converter can, as already mentioned, be operated as a diode rectifier. For this purpose, all power switches can be blocked and the body diodes typically present (parasitic diodes, e.g. of a MOSFET) can be used as diode rectifiers. Alternatively, the upper circuit breakers can all be blocked and the lower ones can all be switched on (or vice versa) and the electrical machine can thus be operated in a short circuit, decoupled from the DC voltage side (battery or vehicle electrical system side).

The activation can preferably take place by the (longer) synchronization pulse or the synchronization information in the clock signal. So that this can be output as soon as possible after the crankshaft position is available, the necessary number of short pulses (with the switchover information) can be output in very quick succession directly after the synchronization pulse in order to set the desired block for the current crankshaft position. When this block is reached, the clock signal is output synchronously with the rotational frequency of the rotor, with a longer pulse being output at the corresponding points as described to ensure that the phase position of the phase voltages is set as desired and not, for example, due to a brief disturbance on the Clock line is permanently shifted (faulty) operation.

As already mentioned, the electrical machine is preferably coupled to an internal combustion engine in a rotationally fixed manner, e.g. by fastening the rotor directly on or on the crankshaft of the internal combustion engine.

It is also useful if the clock signal is generated by a computing unit other than the converter - e.g. a motor control unit of the internal combustion engine - and transmitted to the converter, preferably via a single control line. To change a mechanical or electrical power output by the electrical machine, the switchover time information in the clock signal is then changed in such a way that the switchover times are shifted in time.

In order to change the torque of the electrical machine, the phase position of the phase voltages and thus of the clock signal must be changed. Based on a typical generator operation to supply the on-board network, the clock signal must be delayed for a more powerful generator operation and an increased braking torque (i.e. an increase in the electrical power output). For a weaker generator operation or motor operation and accelerating torque (i.e. an increase in the mechanical power output), the clock signal must be output earlier.

For a stronger generator operation, the switchover takes place accordingly by a corresponding delay of the next clock pulse. However, if the new resulting block were to correspond to a block other than the one currently present, this would have to be delayed for a correspondingly long time. Alternatively, for an immediate change to the desired block, a synchronization pulse or synchronization information followed by the necessary number of short pulses or switchover information.

For a change to less powerful generator operation or motor operation, the clock pulses must be moved forward in time. If a pulse were to "slip" into the past, a synchronization pulse or synchronization information followed by the necessary number of short pulses or switchover information must also occur here in order to achieve a quick and targeted transition to the desired block.

As an alternative to a synchronization pulse and the necessary number of further pulses to reach the desired new block, only the necessary number of short pulses can also be output. Since every first edge of the pulse triggers switching to the next block, it is desirable for a quick change to the new block to place all the necessary pulses in the shortest possible time, i.e. to forego the long synchronization pulse. This alternative enables faster switching to the new block, but with the risk that, in the event of an interference pulse, the electrical machine may be in an incorrect operating state until a regular synchronization pulse is output.

The electrical machine is expediently integrated into an on-board network with a battery, in which case it is preferred that the synchronization information and / or the switching information cause the converter to next generate a desired level of a desired control signal when an undesired change in a battery or signal occurs Vehicle electrical system voltage is detected.

A brief disturbance of the clock signal, for example a single disturbance pulse, leads to a change to an unwanted block or to a change in the level of a control signal. Accordingly, an undesired torque of the electrical machine occurs. Correspondingly, if there is a jump to a more powerful generator mode, more electrical energy is released into the vehicle electrical system and the battery voltage rises. In the case of a jump to a weaker generator or even motor operation, on the other hand, the amount of energy decreases or becomes negative. The battery voltage drops accordingly.

Upon detection of an undesired, particularly significant change in the battery voltage or on-board network voltage (or if a respective threshold value is exceeded or fallen below), a possible disruption of the signal line or of the clock signal transmitted via it can be concluded. In response, a synchronization pulse and the necessary number of short pulses (or synchronization information and switchover information) can be output in order to get to the desired block.

If the desired battery voltage is not established afterwards either, a change in load in the vehicle electrical system can be concluded and the setpoint value of a corresponding voltage regulator can be adapted to the new load conditions.

Furthermore, monitoring of the clock signal can be provided in the hardware of the converter. For example, the frequency of the clock signal can be checked for plausibility and the circuit deactivated ("Disable" level, etc.) if a high-frequency and persistent disturbance is present. Alternatively, a new clock pulse can only be allowed after a minimum time, so that the area in which disturbances have an effect is reduced. A maximum time between clock pulses can also be monitored. As the control by means of the control unit only takes place when the crankshaft is rotating and a minimum speed (e.g. 50-100 rpm or rpm), it can be prevented that faults on the clock line cause the converter to malfunction when the crankshaft is at a standstill or at low speeds.

The invention also relates to a converter for operating a permanently excited electrical machine, each with a power half-bridge per phase of the electrical machine for generating phase voltages, which is set up to carry out all method steps of a method according to the invention, insofar as these are carried out by the converter.

Accordingly, the subject matter of the invention is also a system with a converter for operating a permanently excited electrical machine, with one power half-bridge for each phase of the electrical machine for generating phase voltages, and with a computing unit that is preferably connected to the converter by means of a control line The system is set up to carry out all method steps of a method according to the invention if these are carried out, for example, by the converter and the computing unit (which can be a motor control device).

With regard to possible configurations of the converter or its circuit, reference is made to the figures and the associated description.

The implementation of a method according to the invention in the form of a computer program or computer program product with program code for performing all method steps is advantageous, since this causes particularly low costs, in particular if an executing control device is also used for other tasks and is therefore available anyway. Suitable data carriers for providing the computer program are, in particular, magnetic, optical and electrical memories, such as hard drives, flash memories, EEPROMs, DVDs, etc. A program can also be downloaded via computer networks (Internet, intranet, etc.).

Further advantages and embodiments of the invention emerge from the description and the accompanying drawing.

The invention is shown schematically in the drawing using an exemplary embodiment and is described below with reference to the drawing.

Figure list

• 1 shows schematically a system with internal combustion engine, electrical machine, converter and motor control device, with which a method according to the invention can be carried out.
• 2 shows schematically a clock signal and control signals when carrying out a method according to the invention in a preferred embodiment.
• 3 shows schematically a converter unit as part of a converter according to the invention in a preferred embodiment.
• 4th shows schematically with the converter unit according to FIG 3 generated signals.
• 5 shows schematically a circuit for achieving the functionality shown in FIG 3 is shown.

Embodiment (s) of the invention

In 1 is schematically a system 100 with an internal combustion engine 120 , an electrical machine 110 , a power converter 130 and a computing unit designed as an engine control unit 160 shown, with which a method according to the invention can be carried out.

The electric machine 110 has a rotor 112 with permanent magnets (not shown) and a stator 111 with three phases U, V, W, for example, and thus forms a permanently excited electrical machine (for example a so-called external rotor). The rotor 112 is rigid and non-rotatable with a crankshaft 121 the internal combustion engine 120 tied together.

By means of a speed sensor 122 For example, with a suitably designed crankshaft (with, for example, an encoder wheel as mentioned above), a speed and an absolute position can be recorded. These values can then also be sent to the engine control unit 160 are fed.

The converter 130 (also referred to as an inverter) has - corresponding to the three phases U, V, W - three power half bridges 140 each with a high-side power switch (e.g. a MOSFET) 141 and a low-side power switch (e.g. also a MOSFET) 142 which are (electrically) interconnected in a manner known per se and connected to the phases U, V, W. The power half bridges are also 140 in a manner known per se with an on-board network 170 or a battery 171 (electrically) connected.

The converter 130 furthermore has a converter unit 150 via which the converter can receive a clock signal via a (single) control line 131 , which - as will be explained in more detail later - is then evaluated and passed on to a control unit (or a gate control circuit) 151 via which the converter controls control signals for the power half-bridges (in the example shown, a signal for each power switch, but only conceivable a signal for each power half-bridge). Corresponding control lines to the individual circuit breakers are shown for this purpose.

Furthermore, the converter 130 for example a monitoring unit 152 on, by means of which, for example, a voltage of the vehicle electrical system 170 or the battery 171 can be monitored. Likewise with it eg an error reaction can be initiated. As soon as an inadmissibly high battery voltage is determined, the high-side circuit breaker can, for example, be switched to completely blocking and the low-side circuit breaker to be completely conductive (or vice versa), thus operating the electrical machine in short-circuit mode and protecting the battery from overcharging.

Should, for example, during the operation of the converter 130 the full, permissible range of the battery voltage can be used, the described monitoring can only be enabled, for example, if there is no clock signal.

In order to prevent the output voltage of the converter from rising when the electrical machine is operating in generator mode, if the battery connection is lost during operation, a second voltage monitoring above the permissible battery voltage range can be provided permanently. This intervenes in the event of a fault, for example, and also switches the electrical machine to short-circuit operation as described.

In 2 are schematically a clock signal S _T Control signals Su, Sv and Sw for the three phases U, V and W as well as associated phase voltages (plotted as voltage U over time t) are shown as they can occur in a preferred embodiment when carrying out a method according to the invention.

The clock signal S _T has a high level as quiescent level and contains switchover information lu and synchronization information ls. In the example shown, the switching information lu is designed as pulses with a short length (here short voltage dips), the synchronization information Is as pulses with a slightly longer length (here longer voltage dips), which also serve as switching information.

From the clock signal S _T are in the converter, in particular by means of the converter unit 150 as well as the control unit 151 as in 1 shown, the control signals Su, Sv and Sw for the three phases U, V and W obtained. In this regard, reference is also made to the remarks below.

It can be seen that with each pulse of the clock signal S _T or the level of one of the drive signals is changed for each switchover information lu and each synchronization information item Is. The form of the control signals corresponds to the usual form when using block commutation and is known per se. In this respect it is possible to use only the pulses of the clock signal S _T to generate these control signals.

The control of the power half-bridges or the power switches then takes place with these control signals in a manner known per se, ie the phase voltages obtained therefrom are, as in FIG 2 at the very bottom, the phase voltages customary for block commutation can be seen.

In 3 is schematically a converter unit - see also 1 - Shown as part of a converter according to the invention in a preferred embodiment. With the present invention, as already mentioned, three control lines (or one per phase) are no longer necessary, rather one control line is sufficient for the clock signal. A possible and exemplary form of implementation of the necessary additional circuit - the converter unit mentioned - is shown in FIG 3 shown.

The basic functionality can be achieved through a shift register 300 can be achieved with serial data input and parallel, clocked data output. This can be implemented as an integrated component or a circuit with appropriate functionality.

The shift register 300 When a first corresponding signal edge occurs at the clock input CLK, it transfers the logic level that is present at the data input DATA at this point in time to the first data output OUT 1. The corresponding signal edge can be a falling or rising signal edge, depending on the shift register selected. The corresponding other signal edge does not trigger any action in the shift register. With every further corresponding signal edge of the clock signal S _T passes the shift register 300 the logic level of the data output N-1 to the next data output N (OUT 1 to OUT 2, OUT 2 to OUT 3) and takes over the new current level of the data input for the first output OUT 1.

In order to obtain a high or low level alternately for three clocks, the last data output (OUT 3) is inverted and fed back to the data input (DATA). In this context, inverted means that if there is a high level at the output, a low level is specified at the input and vice versa.

This results in the following sequence of switching states: Tact OUT 1 OUT 2 OUT 3 DATA Measure 1 0 0 0 1 Measure 2 1 0 0 1 Measure 3 1 1 0 1 Measure 4 1 1 1 0 Measure 5 0 1 1 0 Measure 6 0 0 1 0 Measure 1 ' 0 0 0 1

Cycle 1 corresponds to the state that is also set initially and after a reset. After the sixth bar (bar 6th ) the pattern from cycle 1 (cycle 1 ') is set again both by a normal clock pulse and by an extended clock pulse (reset pulse). This means that each cycle 1 can be triggered with an extended reset pulse without affecting the signal sequence, so that constant resynchronization takes place and possible interference pulses can only briefly cause a disruption of the output pattern.

From the clock signal specified by the engine control unit S _T all required internal signals are generated. If necessary, in a first step the form and level of the clock signal are adapted to the additional circuit in the converter. To this end, level converters, inverting circuits or amplifier circuits, as well as protective circuits such as overvoltage limiter or current limiter, can be used if required.

From the processed clock signal (see step or block 310 ) the reset signal (see step or block 330 ) and, if required, an enable signal (see step or block 320 ) generated. The reset signal is derived from the longer pulse in the clock signal (the synchronization information mentioned) and clears the register content of the shift register 300 so that it is reset to its initial state.

This initial state defines the block in which the converter circuit starts initially and on which the position of the longer pulse in the clock signal must be triggered. The initial state typically corresponds to a low level at all data outputs. Accordingly, after a reset has been triggered, there is a high level at the data input for the first following clock signal.

As in 2 shown, only two of the drive signals have the same level, while the third drive signal has the corresponding other level. After a reset, however, all three data outputs of the shift register have the same level. Therefore, one of the three output signals (OUT 2) is inverted (OUT 2 ') and transferred to the following circuit parts. This ensures that the correct signal shape of all three control signals is present at all times and that after a reset (triggered by a longer pulse of the clock signal) the system always starts in the same state, so that the functionality described above is fulfilled.

In order to additionally ensure that the control signals are only passed on, i.e. are only available at the output of the shift register, when a clock signal is present, an “enable” signal can be provided. Many shift register circuits offer the option of activating the outputs only if there is a corresponding “Enable” level at the “Enable” input. Depending on the gate drivers used, the same signal can also be used as an “enable” signal for these driver circuits.

Reset and enable functions are often linked to falling signal edges or low levels. Therefore, when the circuit is not activated, the enable signal assumes a high level as the quiescent level and goes to a low level to activate the outputs. When the circuit is activated, the reset signal has a high level as the quiescent level and triggers a reset of the shift register with low pulses.

This results in the in 4th represented signals. The signals CLK, RESET and EN are shown above. As described, the clock signal has S _T or CLK high level as long as there is no external signal. By specifying a low level as the quiescent level, it is signaled that a clock signal can be expected from now on. Accordingly, the reset signal RESET is set to high level. By specifying a longer pulse at the desired position (depending on the desired torque on the electrical Machine) a corresponding low pulse is generated in the reset signal, which sets the shift register to the desired state. The enable signal is set to low level with a delay. This means that the shift register can be switched to the desired state using the required number of clock pulses before the outputs are activated and the inverter begins to work.

The signal outputs OUT 1, OUT 2 and OUT 3 are also shown. These are only activated when the enable signal is set to low level and they are all deactivated accordingly when the enable signal is set back to high level. During the activated time, the outputs switch to the next state or block with each clock pulse according to the table above.

Furthermore, the two signals OUT 2 'and DATA are shown, which are derived from the signals OUT 2 or OUT 3 can be generated and behave accordingly level inverted.

In addition, the three output signals OUT 1, OUT 2 'and OUT 3 are in the direction of the other hardware of the converter (i.e. to the control unit 151 according to 1 ) shown. The signal curves correspond to those in 2 Required control signals shown for operating the converter in block commutation.

After the fifth reset pulse is in the clock signal S _T or CLK to see an irregular pulse pattern. A single interference pulse is shown between the desired clock pulses. This triggers an early change to the next block. The resulting shifted output signal pattern lasts until the next reset pulse. At this point the circuit is resynchronized with the desired clock signal and the corresponding output pattern is output again.

The first extended clock pulse (ie the synchronization information) and the associated reset pulse can optionally only be output when the rotor of the electrical machine is in the desired position (depending on the desired operating point of the electrical machine), or even already earlier, if the converter is to start working as early as possible.

In this case, after the reset pulse, the required additional clock pulses must be output in quick succession in order to bring the shift register into the desired state before the outputs are activated (enable = low level). In the example shown here, a further pulse is added directly after the first reset pulse in the clock signal before the further clock pulses for switching over the respective blocks are specified in a rotor-synchronous manner.

In 5 is a possible implementation or circuit with additional function blocks in addition to the shift register according to FIG 3 shown. The circuit shown is based on an input signal S _T generated by an output stage with open collector wiring or a push-pull stage with suitable output levels in the engine control unit. A corresponding input circuit (see reference numerals 310 in 3 ) which adapt the signal of the control unit to the working range of the converter circuit and at the same time suppress interference.

If no input signal is given by the engine control unit, the clock signal S _T through the pull-up resistor R1 pulled to the supply voltage Vcc. However, if the engine control unit generates an input signal, this is high-active. This means that the quiescent level is low and the pulses are transmitted as short high levels.

This input signal is (in the case of an open collector generation) through the pull-up resistor R1 pulled to the required voltage level Vcc of the circuit. Through the condenser C1 slight fluctuations or short voltage peaks are filtered out.

The components R2 , R6 , C2 , M1 and D1 generate the required reset signal from the input signal. Due to the low-pass behavior of the resistor R6 and the capacitor C2 rising edges in the input signal are only slowly sent to the gate of the MOSFET M1 forwarded. Only with a sufficiently long pulse length (extended synchronization pulse of the input or clock signal) is the threshold voltage of the MOSFET exceeded and this briefly conductive. Most of the time, the reset signal is passed through the resistor R6 pulled to high level. Only if the input signal has a sufficiently long pulse (defined by the time constant of the RC element R2 , C1 ), the reset signal is pulled to low level for this period.

The diode D1 ensures that with falling edges of the input signal the gate of the MOSFET M1 is discharged quickly (without time delay of the RC element). This means that short pulses can be processed again just a short time after a synchronization pulse without a falsified reset signal being generated.

The components R3 , R7 , R8 , C3 , M2 and D2 generate the enable signal. The clock signal is idle while there is no input signal S _T to high level. Inverting with M1 becomes the gate of the MOSFET M2 not controlled. In this case, the enable signal is pulled to high level by R3. As soon as the input signal is present (mostly as a low level), the gate of M2 is slowly charged via R8 and C3 (time defined by the time constant of the RC element). As soon as the threshold voltage of the gate is reached, M2 becomes conductive and pulls the enable signal to low level. Since typical enable functionalities are activated by a low level, the state of the enable signal corresponds to the activation of the subsequent shift register outputs. As soon as there is no more input signal, an enable signal is no longer required. If this is to be brought back to high level as quickly as possible, the gate of M2 can be connected to an optional additional circuit consisting of the diode D2 and a resistor R7 be discharged to limit the current. Alternatively, these two components can be dispensed with and the gate of M2 can be slowly discharged via R8. In both cases, the enable signal returns to the high level (deactivated outputs and circuits) as soon as there is no longer an input signal.

The components R4 , R9 and M3 serve to invert the output signal OUT 3. This inverted signal DATA is fed back to the data input of the shift register.

The components R5 , R10 and M4 serve to invert the output signal OUT 2. This inverted signal OUT 2 'is sent as a control signal to the circuit for controlling the power switches (the control unit 151 according to 1 ) forwarded.

As an alternative to implementation using discrete components, the use of a digital controller on the converter can also be provided. Unlike in other converters, this does not take over the detection of the rotor position and a corresponding control of the output stage according to an external torque request. Instead, this control can be supplied with the same pulse signal as described above from an external control unit and the individual control signals for the output stages can be derived from this, as well as the plausibility check of the input signal and the generation of an enable signal.

Accordingly, the requirements placed on this digital control are significantly lower than with other power converters, and a cheaper control unit with less software effort can therefore be used. If necessary, an integrated unit can also be used, which combines a digital control unit with the necessary additional circuitry (e.g. power supply) and the circuit parts for controlling all circuit breakers in one component. As a result, the space requirement on the converter board can be greatly reduced.

Claims (15)

Method for operating an electrical machine (110) by means of a converter (130) each with one power half-bridge (140) per phase (U, V, W) of the electrical machine (110) for generating phase voltages, the power half-bridges (140) in block commutation are controlled by generating control signals (Su, Sv, Sw) for the power half-bridges (140) in the converter (130), the converter (130 _{) receiving a clock signal (S T} ) containing switchover information (lu) about switchover times of the control signals (Su, Sv, Sw), and the level of the control signals (Su, Sv, Sw) being switched in the converter (130) based on the clock signal (S _T ) and the switching information (lu) contained therein. Procedure according to Claim 1 , wherein the clock signal (S _T ) further contains synchronization information (Is) which preferably indicates each full cycle of a control period of all power half bridges, and wherein in the converter (130) based on the clock signal (S _T ) and the synchronization information (Is ) the control signals (Su, Sv, Sw) are brought into a desired state and are preferably synchronized with a rotor position of the electrical machine (110). Procedure according to Claim 2 , the control signals (Su, Sv, Sw) for the power half bridges (140) being generated in the converter (130) in order to control the power half bridges (140) in block commutation when the clock signal (S _T ) is received. Procedure according to Claim 2 or 3 , the switching information (lu) in the clock signal (S _T ) for a switching point in time of a control signal (Su, Sv, Sw) within a full cycle of a control period of all power half-bridges corresponds to the synchronization information (Is) and is in the form of pulses with a first length, and wherein the switching information (lu) for the remaining switching times of the control signals (Su, Sv, Sw) within this full run of the control period of all power half bridges are each formed as pulses with a second length different from the first. Method according to one of the Claims 1 until 3 , the switching information (lu) in the clock signal (S _T ) for the switching times of all control signals (Su, Sv, Sw) of a full run of a control period of all power half-bridges are each formed as pulses of different lengths from one another, and preferably one or more pulses can be used as the synchronization information (Is) within this run of the control period of all power half bridges. Procedure according to Claim 4 or 5 , wherein a leading edge of a pulse is used for the switchover time, and / or wherein a duration of a pulse is used for a check, in particular, of the currently used level of the control signal. Method according to one of the preceding claims, wherein if the converter (130) does not receive a clock signal or the clock signal (S _T ) contains no switching information, the converter (130) is operated as a diode rectifier or all high-side (141) or all low-side switches (142) of the power half bridges (140) are switched on. Method according to one of the preceding claims, wherein the electrical machine (110) is coupled to an internal combustion engine (120) in a rotationally fixed manner. Method according to one of the preceding claims, wherein the clock signal (S _T ) is generated by a computing unit (160) different from the converter (130) and transmitted to the converter (130), preferably via a single control line (131). Procedure according to Claim 9 wherein, in order to change a mechanical or electrical power output by the electrical machine (110), the switching time information (lu) in the clock signal (S _T ) is changed in such a way that the switching times are shifted in time. Procedure according to Claim 9 or 10 , wherein the electrical machine (110) is integrated into an on-board network (170) with a battery (171), and if an undesired change in a battery or on-board network voltage is detected, the clock signal (S _T ) is changed in such a way that the Synchronization information (l _s ) and / or the switching information (l _U ) cause the converter (130) to next generate a desired level of a desired control signal (Su, Sv, Sw). Converter (130) for operating a permanently excited electrical machine (110), each with a power half-bridge (140) per phase (U, V, W) of the electrical machine for generating phase voltages, which is set up to carry out all method steps of a method according to one of the Claims 1 until 7th perform. System with a converter (130) for operating a permanently excited electrical machine (110), with one power half bridge (140) per phase (U, V, W) of the electrical machine (110) for generating phase voltages, and with a computing unit (160 ), which is preferably connected to the converter (130) by means of a control line (131), the system being set up to carry out all method steps of a method according to one of the Claims 8 until 11 perform. Computer program that causes a converter (130) or a system to carry out all method steps of a method according to one of the Claims 1 until 11 carried out when it is executed on the converter (130) or the system. Machine-readable storage medium with a computer program stored thereon Claim 14 .

Images