WO2009093214A2

WO2009093214A2 - Electric drive and method for controlling it

Info

Publication number: WO2009093214A2
Application number: PCT/IB2009/050280
Authority: WO
Inventors: Pietro De Filippis
Original assignee: SPAL Automotive SRL
Current assignee: SPAL Automotive SRL
Priority date: 2008-01-25
Filing date: 2009-01-23
Publication date: 2009-07-30
Anticipated expiration: 2010-07-25
Also published as: WO2009093214A3; ITBO20080047A1

Abstract

An electric drive (1) comprises: a permanent magnet brushless motor (2), a motor (2) power supply bridge, a circuit for driving the power supply bridge according to the position of the rotor and of the phase currents (is); the drive (1) comprises a circuit (40) for detecting the zero crossings of a phase current (Is) in order to drive the electric motor (2).

Description

Electric drive and method for controlling it

Technical Field

This invention relates to an electric drive for a brushless motor with permanent magnets and to a method for controlling the drive.

By way of non-limiting example, this specification describes a drive comprising a three-phase brushless motor with permanent magnets that generates a sine-wave counter electromotive force (c.e.m.f.) for driving axial, radial and other types of fans used in electric ventilators.

Background Art Considering that the field of application of these electric fans is that of climate control and cooling systems for installation in motor vehicles, it should be observed that the main aims of developing electric fans for this purpose are: low acoustic noise, limited energy consumption and reduced costs.

These requirements have led to the adoption of sine-wave c.e.m.f. brushless motors (AC brushless motors) driven by inverter capable of generating sine- wave currents and making obsolete the use of PWM six-step driven trapezoidal c.e.m.f motors (more commonly known as DC brushless motors) .

The sine waveform of the c.e.m.f. and of the related phase current minimizes active torque ripple (virtually zero), thus reducing mechanical vibrations and acoustic noise.

It is also known that it is possible to minimize current draw to generate a certain drive torque, thereby maximizing electromechanical conversion efficiency through optimum drive of AC brushless motors which are normally driven by current-controlled, impressed voltage inverters. To obtain this type of drive, the static switches must change state in such a way that the polar axis of the rotor magnetic field remains at 90 electrical degrees to the polar axis of the magnetic field generated by the current circulating in the stator windings, whatever the torque supplied and the rotation speed. To obtain information about the angular position of the rotor, relatively expensive devices are normally used, including absolute encoders or Hall effect sensors, integral with the stator and suitably positioned angularly, to detect the sine waveform of the magnetic energizing field along the periphery of the rotor. The output signals generated by the sensors are then suitably decoded to drive the static switches in such a way as to keep the angular shift of 90 electrical degrees between the rotor and stator magnetic fields.

This type of drive requires the use of the above mentioned position sensors, whose cost is relatively high. In an attempt to reduce the cost of drives, driving strategies that do not use sensors of this type have been developed.

These driving strategies are based on the consideration that if drive is optimum, the c.e.m.f. and the phase current are in phase and vice versa at each point in the operating field (torque, rotation speed, DC supply voltage). Consequently, these driving strategies and drives, which have come to be known as "sensorless", are based on the reading of electrical quantities (e.g. voltage at motor terminals or current circulating in motor windings) to detect the points where the c.e.m.f. and the current cross zero (zero crossings), calculate the relative phase between c.e.m.f. and current and implement appropriate methods of driving the inverter static switches which tend to keep the two quantities in phase. One disadvantage of these methods lies in the fact that to detect the zero crossing of the c.e.m.f., that is to say, to read the sign of the c.e.m.f, the current flowing through the windings must remain zero long enough to enable the reading to be taken, which contrasts with the desired sinusoidal waveform of the current. For the deviation from the ideal to have negligible effects, the length of the time interval during which the current remains zero must be reduced to the minimum and, to eliminate the distortion induced by the controlled phase current interruption, however brief, and the risks of not reading the desired signal, sophisticated algorithms are introduced to calculate the angular position of the rotor in real time: in practice, these algorithms are an integral part of field-oriented controls (FOC in the jargon of the trade) and require the use of sophisticated and expensive controllers with high processing capacity (known as DSP controllers in the jargon of the trade). Disclosure of the Invention

In this context, the main purpose of the present invention is to propose an electric drive which is free of the above mentioned disadvantages.

One aim of this invention is to provide a low-noise and low energy consumption drive.

Another aim of the invention is to provide an electric drive based on a simple and inexpensive control architecture.

The stated technical purpose and specified aims are substantially achieved by an electric drive with the characteristics described in claim 1 and in one or more of the dependent claims. The invention also relates to a method of controlling the drive.

Brief Description of the Drawings

Further characteristics and advantages of the invention are more apparent in the description below, with reference to a preferred, non-limiting embodiment of an electric drive for permanent magnet brushless motors, as illustrated in the accompanying drawings, in which:

- Figure 1 shows a principle diagram of the electric brushless motor drive according to the invention;

- Figure 2 illustrates an equivalent circuit of a phase of an AC brushless motor;

- Figure 3 illustrates a vector diagram of the circuit of Figure 2;

- Figure 4 illustrates a vector diagram representing optimum operation of the circuit of Figure 2;

- Figure 5 is a diagram showing an example of a portion of the drive according to the invention;

- Figure 6 shows a circuit diagram of a first detail of the drive of Figure 1;

- Figure 7 shows a circuit diagram of a second detail of the drive of Figure

1;

- Figure 8 illustrates a procedure for controlling the drive of Figure 1 until optimum operating conditions are reached;

- Figure 9 is a diagram showing the voltages applied to the brushless motor;

- Figure 10 shows the diagram of Figure 9 in a particular operating condition; - Figure 11 shows a circuit diagram of a second embodiment of a detail of the drive according to the invention;

- Figure 12 shows a diagram of a preferred form of a part of the second embodiment of the drive according to the invention, with some parts removed for greater clarity; - Figure 13 illustrates the result obtained by simulating the circuit of Figure

12.

Detailed Description of the Preferred Embodiments of the Invention

With reference to the accompanying drawings and in particular with reference to Figure 1, the numeral 1 denotes an electric drive according to this invention.

This invention is based on the principle of obtaining information continuous in time from which to derive the supply voltage values for optimizing control of the motor powered by the electric drive.

The drive 1 comprises an electric motor 2, for example for driving a fan not illustrated.

As becomes clearer as this description continues, the purpose of the drive 1 is to obtain information relating to the position of the motor 2 rotor by detecting the zero crossing of the counter electromotive force (also referred to in the abbreviated form c.e.m.f.) in a simple and economical manner. By way of example, without limiting the scope of the invention, this specification refers to a permanent magnet brushless motor 2 with an isotropic, two- pole rotor and a three-phase stator winding.

The stator winding comprises three windings with identical shape and number of turns, spatially phase-shifted by 120° and connected by a wye connection whose centre is not accessible.

Figure 2 illustrates a circuit model of the motor 2.

Each of the three windings is characterized by a phase resistance Rs and a synchronous inductance Ls-

A voltage Vs is applied to the motor 2, while a vector Es represents the induced c.e.m.f. in each of the three stator windings.

The c.e.m.f. has a sine waveform and is due to the rotation of the permanent magnet rotor. Is is the phase current, also sinusoidal, which flows through each of the three windings.

Figure 3 shows the vector diagram of the electrical quantities Vs, Is, Es just mentioned. The direct axis d is oriented in the rotor flow direction Φ_r and the quadrature axis q makes an angle of 90° with the direct axis d.

According to the law of induction (e = dΦ/dt) the induced c.e.m.f. Es in the stator winding is always directed along the quadrature axis q, i.e. it is phase- shifted by 90° with respect to the rotor flow Φ_r.

The voltage Vs applied by the drive 1 to the stator windings for a predetermined operating condition is, as mentioned, represented by the vector Vs-

The stator current vector Is makes an angle ψ with the vector difference Vs - Es. The angle ψ depends on the characteristic parameters of the motor and on the supply frequency according to the relation: ψ = arctan(ωLs / Rs).

The electromagnetic power yield of the motor is given by 3EgIscos(γ) where γ is the angle made by Es and Is.

The power absorbed by the motor 2 is essentially the sum of the electromagnetic power yield and of the power losses due to the Joule effect in the three phase resistors.

Hence, given a certain electromagnetic yield, the absorbed power is lowest when the angle γ is zero, that is, when the c.e.m.f. E_s and the current I_s are in phase, as illustrated in Figure 4. As illustrated in Figure 1, the drive 1 comprises a three-phase bridge 3 or inverter for powering the motor 2.

Preferably, the drive 1 comprises a low-inductance shunt 3 a connected as shown in Figure 1 to the three legs of the inverter 3 and crossed by the currents circulating in the inverter, as described in more detail below. The drive 1 also comprises a direct current stage 4 for powering the bridge 3 and in turn comprising a levelling capacitor 5 (Cbus) and a filtering inductor 5a (Lbus).

By way of example, the three-phase bridge 3 generates, through sine-delta PWM modulation of substantially known type, three voltages phase-shifted by 120° from each other at variable frequency "freq".

Advantageously, the amplitude of the fundamental supply voltages can be programmed both as a linear function of the frequency f and independently of the latter.

It should be noticed that, as is well known, the permanent magnet brushless motor 2 develops torque only at its synchronous speed and therefore it will rotate exactly at a speed directly proportional to the frequency f of the applied voltages according to the known relation RPM = 120 x freq/p, where p is the number of poles of the permanent magnet rotor.

The drive 1 comprises a circuit 6 for detecting the counter electromotive force and, more specifically, the passage through zero of the c.e.m.f. Es, hereinafter also referred to as c.e.m.f. zero crossing detection circuit 6.

The detection circuit 6 comprises a first stage, illustrated in Figure 5, and a second stage, illustrated in Figure 6. The second stage processes the output signal from the first stage. In the preferred embodiment illustrated, as will become clearer as this description continues, the information on the position of the rotor is obtained by detecting the zero crossings of the c.e.m.f. generated by only one of the three phases of the motor.

In alternative embodiments, for example in more sophisticated applications that require higher transient response speeds, a rotor position signal is detected for all the phases by replicating the circuit described above for each phase.

Considering, for convenience of description, the phase quantities of a wye connected motor 2 (it is also known that a delta connected motor is functionally indistinguishable from its wye connected equivalent), the value of the c.e.m.f. is given by the relation:

To find the value of e_s it is therefore necessary to know both the value Vs of the voltage Vs applied to the phase of the motor 2 and the resistive-inductive drop due to the flow of current in the windings of the motor 2. As described below, the drive 1 according to the invention has for an aim to find the resistive - inductive drop and the voltage applied to the phase of the motor 2.

For finding the resistive - inductive drop, the drive 1 comprises, as illustrated in Figure 5, an inductive - resistive element 9 connected in series with one of the three phases of the motor 2, as described below.

The element 9 comprises a first inductor L₁₁ and a second inductor Lj₂ with a mutual magnetic coupling coefficient very close to 1.

The two inductors La and L₁₂ are connected to form an autotransformer 9a and, preferably, are coiled around a magnetic core illustrated schematically and labelled 60 in Figure 5.

By way of example, the magnetic core 60 is in the shape of a double E and is made either of high-frequency ferrite or of plain steel for magnetic plates.

The first inductor L]₁ is connected in series with one of the phase windings of the motor 2 and constitutes the primary of the autotransformer 9a.

Preferably, the inductor L₁] has a low number Nl of large diameter turns to minimize the power loss due to the Joule effect.

R₁₁ represents the resistance of the winding of the first inductor L₁₁.

In practice, the inductor Lj₁ and resistor R₁] series form an inductive - resistive component 100 constituting the primary of the inductive - resistive element 9.

The second inductor Lj₂, which constitutes the secondary of the autotransformer, has a number N2 of turns much higher than the number Nl of turns of the inductor Lj₁ and is not crossed by the current is, and therefore provides a voltage Va that depends on the derivative of the current is flowing on the primary.

With reference to Figure 5, if:

V_t = voltage at the terminal 10, 11 of Figure 5, that is, at the terminals of the inductive - resistive element 9; V_R = voltage drop on the resistor R₁₁

V_tl = voltage drop on the first inductor L₁₁;

V_t2 — voltage drop on the second inductor Lj₂;

M - mutual inductance between L_U and L;₂; then:

V₁ = V_{x +} V_n + V₁₂

It is important to note that the expression for V_t is formally identical to that for the resistive inductive drop in the windings of the motor 2 due to the flow of current. The following equation can therefore be written:

D where a = — — is an attenuation coefficient.

Since the synchronous inductance Ls of the motor 2, the self-inductance of the primary Lu and the related number of turns Nl are known, the number of turns N2 of the secondary L^ can be found by solving the following equation:

Thus, by making a resistive - inductive element 9 with the parameters specified above, the attenuated value of the resistive - inductive drop on the phase of the motor 2 can be obtained using, in practice, a measuring circuit corresponding to the equivalent circuit of the phase of the motor 2.

It should be noticed that the attenuation coefficient α indicates the impact of the c.e.m.f. detection circuit 6 on the total loss of the drive 1 : the lower the coefficient, the lower the loss. The information on the voltage Vs applied to the motor 2 is obtained using a circuit 12 for measuring the applied voltage.

The measuring circuit 12 comprises three wye connected resistors 13, 14, 15 illustrated in particular in Figure 5.

By attenuating by the coefficient α both the contribution of the applied equivalent voltage v$ measurable, as described in more detail below, by the set of wye connected resistors 13, 14, 15, and the contribution of the resistive — inductive drop supplied by the mutually coupled inductors Lj₁ and Lj₂, the circuit of Figure 5 supplies an attenuated c.e.m.f. signal whose amplitude is given by:

a-e_s = a-v_s —a -R_s -ι_s —a -L_s — ~ dt Based on the attenuation coefficient α defined above, the wye connected resistors 13, 14, 15 used for measuring the supply voltage Vs is suitably unbalanced.

Figure 9 shows the real first harmonic voltages Vl, V2, V3 generated by the inverter 3 and applied to the motor 2, each schematically represented with a respective ideal voltage generator.

As illustrated, the voltage αVl must appear at the terminals of the resistor 13, also labelled Ra.

Preferably, the values of the resistors 13, 14, 15 can be calculated, at a given point in time, with reference to the symmetrical three-phase triad that powers the motor 2.

For example, at the point in time where Vl reaches its maximum value Vm, the values of V2 and V3 are -Vm/2.

The circuit to be analysed is therefore the one shown in Figure 10. Applying the overlapping effects principle to calculate the voltage drop on

Ra in the circuit of Figure 10 gives:

from which the relation between Ra and R can be derived for the previously fixed value of α:

Ka ^{~ K}

3 - 2a where R is a generic resistance value for the resistors 14 and 15.

Applying the signal αes to the signal conditioning circuit shown in Figure 6 gives a signal for the zero crossing of the counter electromotive force that can be processed by a microcontroller 26. It should be noticed that the behaviour obtained is substantially the same as that obtained with a digital output Hall sensor.

As illustrated in particular in Figure 6, this circuit comprises two comparator stages connected in cascade: the first stage 7, with a respective comparator 16, does not exhibit any hysteresis and spurious switching can be detected at its output.

At low rotation speeds, the circuit of Figure 5 for detecting the zero crossing of the c.e.m.f. gives voltage values of a few hundred millivolts and, owing to the low attenuation factor α, the signal-noise ratio is low and causes the above mentioned spurious switching at the output of the first stage 7 of the circuit of Figure 6.

The second stage 8 comprises a second, hysteresis comparator 17 having an RC input filter 18 to limit the signal oscillations that might erroneously trigger the comparator 17.

The RC filter 18 comprises a capacitor 19 and a resistor 20 and is of substantially known type.

The second stage 8 also comprises a resistive network to fix the switching threshold and the related hysteresis.

In the embodiment illustrated, the resistive network comprises four resistors 21, 22, 23, 24 suitably connected to each other.

Thus, there is no spurious switching at the output of the hysteresis comparator stage, with obvious advantages in terms of the processing efficiency of the microcontroller 26.

To maximize the efficiency of the motor 2 the current flow in the stator windings must be in phase with the related c.e.m.f. .

With reference to the vector diagram of Figure 4, it is possible to obtain an approximated expression of the optimum advance angle δ_opt for the applied voltage Vs with respect to the c.e.m.f. Es.

If the resistive drop in the phase is negligible (the higher the efficiency of the motor 2, the more negligible the drop), assuming K_E as the c.e.m.f. constant measured in V/rpm andp as the number of poles, then: ω_el -Ls Is π -p π - Ls - p tgV_nnt = = ®_PI - LS IS = — ^■ I_s o^pt Es ^el 60 - K_E - ω_el 60 - K_E ^s Moreover, if the optimum value of the angle δ_opt is less than 20 electrical degrees, the tangent of the angle can be approximated with the angle itself and hence:

where the synchronous inductance L₈ is preferably expressed in Henrys. In other terms, if the resistive drop RsIs is negligible with respect to Es and the tangent of the advance angle δ_opt can be approximated with the angle itself, then, in practice, the advance angle δ_opt depends linearly only on the phase current

Is

Since the drive 1 according to the invention comprises an extremely simple and economical microcontroller 26, for example an 8-bit microcontroller, the above mentioned linear relation between δ_opt and the current draw Is; can be stored in it, for example in table format, making available to the microcontroller 26 a signal proportional to the current Is and the microcontroller 26 will be able to control the bridge 3 according to the corresponding δ_opt.

It is for this purpose that the drive 1 comprises a circuit 25 for indirectly detecting phase current amplitudes. More specifically, the circuit 25 comprises an enveloping detector or detector stage 27 which processes the voltage signal present at the terminals of the shunt 3 a, directly proportional to the current flow through the shunt 3 a itself.

As is known from the literature, the maximum value of the voltage peaks on the shunt 3 a is proportional to the phase current peak of the motor 2. Since the phase current is sinusoidal, the reading of the enveloping detector

27 is equal to the effective value Is of the phase current multiplied by ^ ^

The enveloping detector 27 keeps track of this information and the microcontroller 26 samples it at a much lower frequency than that of the carrier

PWM: the validity of the information is guaranteed by the fact that the changing speed of the enveloping detector 27 output, which is directly linked to the changing speed of the mechanical load, is very low.

It should be noticed that the discharge constant of the detector 27 is suitably dimensioned to correctly follow the envelope of the current peaks on the shunt 3 a.

Basically, the microcontroller 26, by sampling the output signal of the enveloping detector stage 27 through its analog-to-digital converter, indirectly measures the current value of the phase current and, as a result, determines the corresponding optimal angle to be applied to keep the current in phase with the c.e.m.f. .

Figure 7 shows a diagram of an embodiment of the shunt current enveloping detector 27.

The detector 27 comprises an RC filter 28 for filtering the shunt current envelope.

The detector 27 also comprises a circuit 29 for charging a capacitor in such a way that when the non-inverting signal is lower than the inverting signal, the capacitor can be discharged through the resistors 30 and 31.

The resistors 30 and 31 are suitably connected to enable the device to follow the shunt current peaks. hi an alternative embodiment, illustrated in Figures 11 and 12, the drive 1 comprises a circuit 40 for detecting the zero crossings of the phase current, otherwise referred to as circuit for detecting the zero crossings of the current. The circuit 40 is provided as an alternative to the above mentioned circuit 25 for indirectly detecting the phase current to supply the microcontroller 26 with a signal significant of the phase current.

The circuit 40 comprises a suitably rated resistive-capacitive network or resistive-capacitive element 41.

The resistive-capacitive element 41 is connected in parallel with the terminals of an inductive-resistive component 100, comprising a resistor and an inductor, crossed by the power supply current of one of the phases of the motor 2.

Preferably, the component 100 comprises the series defined by the inductor L₁₁ and resistor R₁₁ constituting the primary of the inductive-resistive element 9 described above, to which express reference is made but without limiting the scope of the invention.

The sign of the voltage drop at the terminals of the resistor R₁] of the primary of the inductive-resistive element 9 coincides with the sign of the phase current.

Retrieving this information, as described below, provides the microcontroller 26 which controls the drive 1 with the information on the positions where the current crosses zero.

In the example illustrated in Figure 11, the element 41 comprises a capacitor 42 in series with a compensating resistor 43.

If C* is the generic value of the capacitor 42 and R* is the generic value of the resistor 43, where the resistive-capacitive element 41 absorbs a negligible current compared to that flowing through the component 100, the voltage Vc at the terminals of the capacitor 42 is:

1

C^* dt

Vc = V

^;1 dt ^l]

R + dⁱs _c* dt

^{Rή +} dt ^Al

= h - —77 Λ dι, dt

In the light of the above, by setting the condition:

the voltage Vc obtained on the capacitor 42 is proportional to the current flowing through the inductor L₁₁. hi the preferred embodiment illustrated in Figure 12, the compensating resistor 43 comprises a first and a second resistor 43 a and 43b since the reading on the capacitor 42 is preferably performed with a differential configuration.

The resistors 43 a and 43 b have the same rating which, with reference to the above for example, may be R*/2.

The reference characters 44a, 44b, 44c denote the three phases of the motor 2, each schematically represented by a respective resistor 45a, 45b and 45c, a respective inductor 46a, 46b and 46c and a respective voltage generator 47a, 47b and 47c. Also, as may be observed from the diagram of Figure 12, showing the circuit 40 for detecting the zero crossings of the current, a first comparator stage 48a is connected in parallel to the capacitor 42.

The stage 48 a has a pair of resistors 49, 50 and a capacitor 51 mounted at the input of a comparator 52 provided with a power supply V. A network of resistors 53, 54, 55 and 56, connected in parallel with the input of the comparator 52, suitably referenced to the voltage V, prevents the signal level on the inputs of the comparator 52 from exceeding the latter's supply voltage or dropping below the GROUND reference G.

In practice, the circuit comprising the resistors 49, 50, the capacitor 51 and the network of resistors 53, 54, 55 and 56 is used to keep the voltage at the inputs of the comparator 52 within finite values permitted by the comparator 52 itself.

At the output of the comparator 52 there is a resistor 57 connected to a voltage Vl.

Downstream of the first comparator stage 48a just described, the circuit 40 that detects the zero-crossings of the current comprises a second hysteresis stage 48b for "cleaning" the signal before sending it to the controller 26.

The second stage 48b, substantially the same as the second comparator stage 8 of the c.e.m.f. zero crossing detection circuit 6, comprises an RC filter 58 at the input of a comparator 59 and a resistive network that sets the switching threshold and the related hysteresis.

In the embodiment illustrated, the resistive network comprises four resistors 60, 61, 62, 63 connected to each other as shown in Figure 12.

In particular, the resistors 60 and 63 are connected to the voltage Vl .

The diagram of Figure 13 shows a point-to-point adaptation of the suitably amplified, reconstructed current in the current zero-crossing detection circuit 40, represented by the curve IR₅ to the real power supply current, represented by the curve IS.

It is important to observe that the voltage measured at the output of the comparator 51 , represented by the curve VC, is also in phase with the real power supply current and thus significant of the zero crossings of the power supply current. The microcontroller 26, on receiving the input signal relating to the zero crossings of the current and the input signal relating to the zero crossings of the counter electromotive force, detects the temporal distance between two homologous signal edges.

Depending on this temporal distance, the microcontroller 26 applies to the bridge 3 a correction of the phase angle between Es and Vs in such a way as to bring the temporal distance to zero.

In practice, as a result of these continuous corrections, the temporal distance between two homologous signal edges oscillates around zero, thus optimizing the operation of the motor 2. Figure 8 illustrates the procedure for controlling the brushless motor 2, comprising the following steps:

A) step of parking or alignment;

B) step of accelerating according to a predetermined ramp V/f;

C) step of "engaging" the c.e.m.f. zero crossing signal, where "engaging" means reaching an optimum operating condition; D) step of optimized drive.

In steps A and B the inverter 3 powers the motor 2 entirely in "open loop" mode, that is to say, without using either of the two available feedback signals, namely, c.e.m.f. zero crossing and shunt current envelope or phase current zero crossing.

In step C only the c.e.m.f. zero crossing signal is used.

Lastly, in step D both the c.e.m.f. zero crossing signal and the shunt current envelope or phase current zero crossing signal are used and the inverter 3 drives the motor 2 under optimum operating conditions, that is to say, with the c.e.m.f. and the phase current in phase with each other.

During step A three constant voltages are applied to the motor, suitably defined to enable the current to flow in such a way as to make the rotor turn until it is at a known position where the stator field and the rotor field are aligned.

This step ensures that the maximum drive torque possible under "open loop" control conditions can be generated in the next step B.

In step B the motor 2 is powered by three sinusoidal voltages phase shifted by 120° from each other so as to create a rotary stator field at increasing frequency.

During this step, the amplitude of the mean voltage applied to the motor 2 varies linearly with its frequency "freq", as illustrated in Figure 8.

The frequency "freq" starts at zero and is increased until it reaches the "fset" value shown Figure 8, set in the software of the microcontroller 26.

The "fset" value is greater than the minimum electrical frequency at which the c.e.m.f. zero crossing signal can be surely detected, so that the phase relation between the counter electromotive force Es and the applied voltage Vs can be measured in the next steps C and D.

This relation is the angle δ made by the quantities Es and Vs shown in Figures 3 and 4.

The slope of this acceleration ramp is a parameter of the drive and must be modified according to the inertia of the motor 2 + load system.

The brushless motor 2 subjected to the rotary stator field generated by the drive accelerates until it exactly reaches the synchronous speed related to "fset".

In step B the brushless motor 2 is controlled in exactly the same way as an asynchronous motor but, unlike the latter, it reaches the end of ramp speed since, during the ramp itself, the angle between rotor field and stator field never exceeds 90 degrees representing the condition necessary and sufficient to generate the drive torque for the permanent magnet brushless motor which, as is known, is a "synchronous" motor.

The slope value of the ramp V/f is chosen in such a way as to guarantee that the motor receives sufficient current, and hence torque, to accelerate it in the required time to the speed corresponding to the frequency "fset", for example as a function of environmental parameters such as inverter 3 supply voltage and ambient temperature.

Reaching the frequency "fset" triggers step C, during which the frequency remains constant at the value "fset" and the applied voltage Vs is decreased at a predetermined rate.

As mentioned, in step C the c.e.m.f. zero crossing signal is available and the phase between Vs and Es is therefore measured through the microcontroller 26.

The gradual decrease of Vs reduces the current draw Is until it reaches the minimum value required to keep the motor turning: when this condition is reached, Is and Es are substantially in phase, the microcontroller 26 detects the in phase condition between Is and Es and considers step C ended.

Step C is followed by step D.

In step D, only the amplitude Vs is set and not the frequency "freq". The microcontroller 26 continuously detects the electrical frequency, acquiring the time interval between two consecutive signal edges, whether homologous or non-homologous, of the output signal from the c.e.m.f. zero crossing detection circuit 6 to which the frequency of the output voltage fundamental of the inverter 3 corresponds. In order to obtain optimum operation of the motor 2, an iterative procedure comprising the steps described below is also implemented.

The peak value of the phase current is measured by the microcontroller 26 through the shunt current enveloping detector 27.

The microcontroller 26 detects the c.e.m.f. zero crossing through the respective detection circuit 6.

The microcontroller 26 then applies the advance δ_opt between Vs and Es, since the software installed in the microcontroller 26 incorporates the relation between the advance angle made by Vs and Es and the peak value of the phase current corresponding to optimum operation. At this point, the procedure restarts from the measurement of the peak value of the phase current.

If the drive 1 comprises the circuit 40 for detecting the zero crossings of the power supply current, the motor is driven as described below.

Steps A, B and C are performed in substantially the same way. Similarly, in step D, only the amplitude Vs is set and not the frequency

"freq".

The microcontroller 26 continuously detects the electrical frequency, acquiring the time interval between two consecutive signal edges, whether homologous or non-homologous, of the output signal from the c.e.m.f. zero crossing detection circuit 6 to which the frequency of the output voltage fundamental of the inverter 3 corresponds.

In order to obtain optimum operation of the motor 2, an iterative procedure similar to that described above and comprising the steps described below is implemented. The zero crossings of the power supply current are detected by the detection circuit 40.

The microcontroller 26 then applies the advance δ between Vs and Es to cancel out the temporal distance between two homologous edges of the c.e.m.f. zero crossing signal and of the current zero crossing signal thereby obtaining the optimum operation of the motor 2.

Both the above mentioned optimizing procedures cause the brushless motor 2 to operate with the c.e.m.f. in phase with the respective phase current. In this situation, as mentioned, absorbed power is minimized; by suitably setting the time interval of the optimizing procedure, it is possible to make the system reasonably reactive even to sudden load variations due, for example, to temporary choking of the delivery and/or suction ducts of the air-hydraulic circuit and the subsequent removal of the choking itself. The control methods described also allow maximum efficiency of motor drive by causing the current in each stator winding to be in phase with the respective c.e.m.f. .

Thus, the motor generates the maximum torque possible. Expressed in other terms, the brushless motor is driven efficiently because the stator current has no components in the axis d but only in the axis q. The drive operates in such a way that, once the starting transient is over, motor power consumption is minimized under all load conditions and at all speeds of rotation: in terms of the vector diagram, the phase current is in phase with the respective c.e.m.f.

Claims

1. An electric drive comprising: a brushless motor (2) comprising a stator and a rotor; a motor (2) power supply bridge (3); means for detecting the angular position of the rotor; a circuit for controlling the power supply bridge (3) according to rotor position; the drive being characterized in that it comprises a circuit (40) for detecting the zero crossings of a phase current, said control circuit driving the bridge (3) according to the zero crossings of the phase current

2. The drive according to claim 1, characterized in that the circuit (40) for detecting the zero crossings of the phase current comprises an inductive-resistive component (100) connected in series with one of the phases (44a) of the motor (2) to determine the resistive - inductive drop due to the current flowing in the winding of said phase of the motor (2).

3. The drive according to claim 2, characterized in that the circuit (40) for detecting the zero crossings of the phase current comprises a resistive-capacitive element (41) connected in parallel with the inductive-resistive component (100) to obtain a voltage proportional to the phase current flowing through the inductive- resistive component (100).

4. The drive according to claim 3, characterized in that the ratio between the inductance (Lj₁) of the inductive-resistive component (100) and the resistance of the inductive-resistive component (100) is such as to obtain, on a capacitive part (42) of the resistive-capacitive element (41), the voltage that is proportional to the current flowing through the inductive-resistive component (100).

5. The drive according to claim 4, characterized in that the ratio between the inductance (Li₁) of the inductive-resistive component (100) and the resistance of the inductive resistive component (100) is equal to the product of the capacitance of the resistive-capacitive element (41) and the resistance of the resistive- capacitive element (41), that is to say:

^ = C -R^* Ra where: L₁₁= inductance of the inductive-resistive component (100); resistance of the inductive-resistive component (100) ; C*= capacitance of the resistive-capacitive element (41); R*= resistance of the resistive-capacitive element (41),

6. The drive according to any of the claims from 3 to 5, characterized in that the resistive-capacitive element (41) comprises a compensating resistor (43) in series with a capacitor (42) respectively constituting the resistive part and the capacitive part of the resistive-capacitive element (41).

7. The drive according to claim 6, characterized in that the compensating resistor (43) comprises a first and a second resistor (43a, 43b) in series with, and mounted on opposite sides of, the capacitor (42) in order to measure the voltage proportional to the current flowing through the inductive-resistive component (100) with a differential configuration.

8. The drive according to claim 7, characterized in that the first and second resistors (43a, 43b) have the same resistance value.

9. The drive according to any of the claims from 3 to 8, characterized in that the circuit (40) for detecting the zero crossings of a phase current comprises a first comparator stage (48a) for stabilizing the voltage proportional to the phase current flowing in the inductive-resistive component (100) and obtained through the resistive-capacitive element (41).

10. The drive according to claims 6 and 9, characterized in that the first comparator stage (48a) is connected in parallel to the capacitor (42).

11. The drive according to claim 9 or 10, characterized in that the first comparator stage (48a) comprises a comparator (52) and a network of resistors (53, 54, 55, 56) connected in parallel to the input of the comparator (52) to stabilize the input signal to the comparator (52) itself.

12. The drive according to any of the foregoing claims from 9 to 11, characterized in that it comprises a second hysteresis comparator stage (48b) downstream of the first comparator stage (48a).

13. The drive according to any of the foregoing claims, characterized in that the means for detecting the angular position of the rotor comprise a circuit (6) for detecting the zero crossings of the counter electromotive force (Es) induced in the stator windings by the rotation of the rotor, said control circuit driving the motor (2) according to the zero crossings of the counter electromotive force and the zero crossings of the phase current.

14. The drive according to claims 2 and 13, characterized in that the circuit (6) for detecting the zero crossings of the counter electromotive force (Es) comprises an inductive - resistive element (9) connected in series with each phase of the motor (2) to determine the resistive — inductive drop due to the current flow in each winding of the motor (2), said inductive - resistive element (9) comprising the inductive resistive component (100).

15. The drive according to claim 14, characterized in that the inductive - resistive element (9) comprises a first and a second inductor (Li₁ , Lj₂), mutually coupled, the first inductor (Lj₁) defining the inductance of the inductive - resistive component (100) and being in particular crossed by the phase current (Is).

16. The drive according to claim 15, characterized in that the first and second inductors (Lu, L₁₂) are connected to form an autotransformer (9a), said first and second inductors (L₁₁ , L₁₂) being in particular coiled around a magnetic core (60).

17. The drive according to claim 15 or 16, characterized in that the second inductor (Li₂) has a number (N2) of turns much higher than the number (Nl) of turns of the first inductor (L₁₁).

18. The drive according to any of the claims from 15 to 17, characterized in that the number (N2) of turns of the second inductor (Lj₂) is the product of the number (Nl) of turns of the first inductor (Li₁) and the difference between the ratio of the synchronous inductance (Ls) of the motor (2) multiplied by an attenuation coefficient (α) and the inductance value (Lj₁) of the first inductor (L₁₁) and one, that is to say:

19. The drive according to claim 18, characterized in that the attenuation coefficient (α) is calculated as the ratio of the resistance (Rj₁) of the first inductor (L₁₁) to the phase resistance (Rs) of the motor (2), that is to say:

a =—^ύ- .

20. The drive according to any of the foregoing claims from 13 to 19, characterized in that the control circuit comprises a controller (26) in communication with the circuit (6) that detects the zero crossings of the counter electromotive force and with the circuit (40) that detects the zero crossings of the phase current, the controller (26) controlling the power supply bridge (3) according to the zero crossings of the electromotive force and according to the zero crossings of the phase current.

21. The drive according to claim 20, characterized in that the controller (26) measures the temporal distance between two signal edges, that is, between the edge of a signal generated by the circuit (6) that detects the zero crossings of the counter electromotive force, being significant of the zero crossings of the counter electromotive force, and the edge of a signal generated by the circuit (40) that detects the zero crossings of the current, being significant of the zero crossings of the phase current, and, according to said temporal distance, applies a correction to the supply voltage of the motor in such a way as to cause the temporal distance to oscillate around zero.

22. The drive according to claim 21, characterized in that the microcontroller (26) applies an advance (δ) between the supply voltage (V_s) and the counter electromotive force (E_s) to cancel out the temporal distance between two homologous edges of the c.e.m.f. zero-crossing signal and of the current zero- crossing signal.

23. A method for controlling an electric drive that comprises a brushless motor (2) comprising a stator and a rotor; a motor (2) power supply bridge (3); a circuit for controlling the power supply bridge (3) according to signals significant of the operation of the motor; the method being characterized in that it comprises the steps of: defining a first signal significant of the zero crossings of the counter electromotive force induced in the stator; defining a second signal significant of the zero crossings of the phase current; and driving the power supply bridge according to said first and second signals.

24. The control method according to claim 23, characterized in that the driving step comprises the step of estimating the temporal difference between an edge of the first signal and an edge of the second signal and driving the power supply bridge (3) in such a way as to cancel said temporal difference.

25. The control method according to claim 24, characterized in that the edge of the first signal and the edge of the second signal are homologous to each other.

26. The control according to any of the claims from 23 to 25, characterised in that it comprises the steps of;

- starting the motor (2) by driving the power supply bridge (3) in open loop mode, that is to say, without using the first and second signals;

- accelerating the motor (2) according to a predetermined ratio between supply voltage and supply frequency; - decreasing the supply voltage until it reaches the minimum phase current draw value required to keep the motor (2) turning:

- optimizing the drive by setting the amplitude of the supply voltage according to the first and second signals.

27. The method according to claim 26, characterized in that during the starting step, three constant voltages are applied to the motor (2), the motor (2) being in particular a three-phase motor, so as to make the rotor turn until it is at a known position.

28. The method according to claim 27, characterized in that at the known position the rotor field and the stator field are aligned.

29. The method according to claim 27 or 28, characterized in that it comprises the step of powering the motor (2) with three sinusoidal voltages phase shifted by 120 electrical degrees from each other so as to create a rotary stator field at increasing frequency.

30. The method according to any of the foregoing claims from 26 to 29, characterized in that during the step of accelerating the motor (2) according to a predetermined ratio between supply voltage (Vs) and supply frequency, the amplitude of the average voltage applied to the motor varies linearly with the supply frequency (freq).

31. The method according to claim 30, characterized in that the supply frequency (freq) starts at zero and is increased until it reaches a predetermined frequency (fset), which is, in particular, a value set in the controller (26).

32. The method according to claim 31, characterized in that the predetermined frequency (fset) is set in the controller (26).

33. The method according to claim 31 or 32, characterized in that the predetermined frequency (fset) is greater than the τniτriτmiτn electrical frequency at which the first signal is detected.

34. The electric drive according to any of the foregoing claims from 1 to 22, the control method according to any of the foregoing claims from 23 to 33 and according to what described and shown in the enclosed drawings and for the above mentioned aims.