TITLE:

METHOD AND SYSTEM FOR CONTROLLING AN ELECTRIC AC MOTOR

AREA OF THE INVENTION

The present invention relates to a method and a system for controlling an electric AC motor having three stator windings, for example for an electric car drive system.

BACKGROUND OF THE INVENTION Electric car drive systems are previously known, see for example US Patent No.

5 294 876, which discloses a control system for an AC induction motor for driving the motor at different speeds and torques in all four quadrants.

Modern cars with combustion motors have achieved good performance thanks to more than 100 years of development. In order to compete with the traditional combustion engine, an electric drive system should give similar performance and at the same time be more energy efficient. The size, weight and cost of an electric drive system should be minimized.

Losses in the drive system should be minimized for two reasons. Losses in the drive system consume electrical energy and the lost energy, which is converted to heat, should be eliminated by a cooling system resulting in additional size, weight and cost. Drive systems for vehicles have a different drive cycle than most industrial drive systems. A vehicle needs high power during acceleration and relatively low power during "cruising". An industrial drive is often operating with full power most of the time. This difference may lead to different strategies for optimizing the drive system for an electric car.

SUMMARY OF THE INVENTION

Accordingly, an object of the present invention is to mitigate, alleviate or eliminate one or more of the above-identified deficiencies and disadvantages singly or in any combination.

According to an aspect of the invention, there is provided a method for controlling an electric AC motor having three stator windings, for example for an electric car drive system, comprising: connecting the stator windings in a star configuration during operation in a low speed region with high torque requirement, connecting the stator windings in a delta configuration during operation in a high speed region with low torque requirement, and operating a mechanical, multispeed gearbox to change a gear ratio to further increase a range of speeds.

In an embodiment, the method may comprise switching the stator windings to a delta configuration when available drive voltage is insufficient; while the stator windings normally

are connected in a star configuration. The motor may be operated with a current, which is higher than the nominal current, for at least shorter time periods. The motor may in addition be operated with field weakening.

In another embodiment, the mechanical gearbox may have a first and a second gear position, such as with a ratio of 1 :3, wherein the method comprises: a) operating the stator windings in a star connection and with the gear in the first position for low speeds, b) operating the stator windings in a delta connection and with the gear in the first position for medium-low speeds, c) operating the stator windings in a star connection and with the gear in the second position for medium-high speeds, and d) operating the stator windings in a delta connection and with the gear in the second position for high speeds. The mechanical gearbox may further comprise a gear reduction unit for adapting the electric motor speed to a wheel rotational speed.

In another aspect, there is provided a system for controlling an electric AC motor having three stator windings, for example for an electric car drive system, wherein the stator windings are connected in a star configuration during operation in a low speed region with high torque requirement, and the stator windings are connected in a delta configuration during operation in a high speed region with low torque requirement, and further comprising a mechanical multispeed gear box.

In a further embodiment, the stator windings may normally be connected in a star configuration, wherein the stator windings may be switched to a delta configuration when available drive voltage is insufficient. The mechanical multispeed gearbox may have an automatic switching capability. The mechanical gearbox may have a first and a second gear position, such as with a ratio of 1 :3, wherein the system comprises that: a) the stator windings are configured in a star connection and with the gear in the first position for low speeds, b) the stator windings are configured in a delta connection and with the gear in the first position for medium-low speeds, c) the stator windings are configured in a star connection and with the gear in the second position for medium-high speeds, and d) the stator windings are configured in a delta connection and with the gear in the second position for high speeds.

BRIEF DESCRIPTION OF THE DRAWINGS

Further objects, features and advantages will appear from the description below of several embodiments of the invention with reference to the drawings, in which:

Fig. 1 is a schematic diagram of an equivalent model of a PM motor.

Fig. 2 is a diagram over the motor torque versus motor speed. Fig. 3 is a diagram over the motor torque versus motor speed during field weakening.

Fig. 4 is a schematic diagram of a star configuration.

Fig. 5 is a schematic diagram of a delta configuration.

Fig. 6 is a diagram over the torque versus speed for a first embodiment of the invention.

Fig. 7 is a schematic diagram of a switch-over between delta and star configuration.

Fig. 8 is a schematic diagram of an embodiment in which and AC motor is combined with a mechanical gearbox.

Fig. 9 is a diagram over torque versus speed for the embodiment according to Fig. 8.

Fig. 10 is a diagram similar to Fig. 8 of another embodiment of the invention.

Fig. 11 is a diagram similar to Fig. 8 of a system including field weakening.

Fig. 12 is a schematic diagram of a delta configuration. Fig. 13 is a diagram over torque versus speed for a still further embodiment of the invention having extended operation range.

Fig. 14 is a diagram over torque versus speed for a yet further embodiment of the invention having extended operation range.

Fig. 15 is a diagram over torque versus speed for a yet still further embodiment of the invention having extended operation range.

Fig. 16 is a schematic diagram of an alternative design of a relay used in Fig. 7.

DETAILED DESCRIPTION OF EMBODIMENTS

The embodiments of the invention are intended for use with AC motors and there are several types of suitable motors.

The AC induction motor is the work-horse in industry and it is well suited for use also in vehicles.

Modern permanent magnet AC motors have become common in vehicle drive systems. There are two major variants of PM AC motors, namely the brushless DC motor and the synchronous PM motor. The brushless DC motor has a trapezoidal shape of the counter electro motive force, EMF, while the synchronous motor has a sine-shaped EMF. A modern PM motor has permanent magnets in the rotor and electrical windings in the stator. The permanent magnets may be surface mounted or "buried" in the rotor.

The present embodiments are intended for use with all types of AC motors, as described above. It may be used also with other motor types, for example the switched reluctance motor. The following description is made for a PM motor but it would have been essentially the same for any other motor type.

There are several types of motor control systems such as scalar control and vector control. Most systems use pulse width modulation (PWM) in the power circuits. Other modulation systems are possible. The modulation waveform may for example be pure sine waves, modified sine waves (space vectors) or 6-step modulation (block modulation). Any

type of modulation can be used with the present embodiments. The motor control system may be described as a "frequency inverter", which is well-known.

The embodiments are based on the unique characteristics of three-phase electrical systems for AC motor control. Other numbers of phases may be used. The minimum number of phases is two, because two phases are always required for describing a voltage or current vector. Higher phase numbers are possible. However, there are practical advantages with three phases.

Combustion motors operate efficiently only in a restricted speed range. In order to extend the speed range for the vehicle, there is always a multi-speed gearbox between the motor and the wheels, having for example 4 to 6 different gearings.

An electric motor has a wider speed range than a combustion motor but it would also benefit from a variable mechanical gearbox in order to make the most efficient use of the motor. However, there is an alternative to this costly solution, namely "field weakening".

Field weakening is a standard method to increase the speed range of an electric motor. The motor operates with full magnetic field up to the "base speed". The base speed is the maximum speed that can be obtained with the maximum control voltage. In order to run the motor with higher speed, the magnetic field is weakened. Field weakening is well-known and will not be described in detail here.

A field weakening range of 1 :5 is typical in a modern drive system. This can be compared with a multi-speed mechanical gearbox with the same ratio between the lowest and the highest gearing. The field weakening creates a continuously variable speed system that is equivalent with a mechanical gearbox with an infinite number of speed ratios or continuous speed ratio.

The field weakening range is normally described as the "constant power" region. The motor torque decreases when the magnetic flux goes down. The motor speed increases with the same factor as the torque goes down, and consequently the product of torque and speed remains constant, i.e. constant power. This is mathematically exact in theory, but there may be some extra losses in the field-weakening region. Constant power is a favorable mode of operation, especially in an electric vehicle, where the available power is limited. The power may come from a battery, a generator or a fuel cell.

The AC induction motor is well suited for field weakening because the magnetic field is created by the motor control system itself. There are no permanent magnets in such motors.

Field weakening is more difficult in PM motors, because the permanent magnets cannot be

"demagnetized". There are two different kinds of PM motors, those with surface mounted magnets and those with "buried" magnets. PM motors with buried magnets offer certain advantages, such as being suitable for high-speed operation. The magnets will be kept in place, also at high

speed, because they are buried inside the rotor. Surface mounted magnets must be kept in place by special means.

In a surface-mounted rotor, the flux cannot be moved tangentially within the magnet to provide a rotor-to-stator phase advance, while in a buried magnet rotor the flux can be moved tangentially above the magnets in the rotor so that quite a significant rotor-to-stator phase advance can result. Such a phase advance may be used for field weakening operation. However, all forms of field weakening, regardless of the PM motor type, represent a loss in motor efficiency.

Field weakening may be possible also for surface mounted magnets but in a different way and in a limited range, see for example "Control of Electrical Drives", 3rd edition by Werner Leonhard, chapter 14.1 (Springer- Verlag, Berlin, Heidelberg, New York).

An object of the present embodiments is to eliminate the need for field weakening or to minimize the range of field weakening.

It is assumed, although not proven here, that a PM motor with surface mounted permanent magnets makes the best possible use of the permanent magnets. This is one important reason to create a system that does not need field weakening. It is also assumed that the motor has its best efficiency up to base speed, i.e. without field weakening. In an optimal system the motor should be allowed to operate at or below base speed also when the vehicle is running at cruising speed and top speed. Field weakening creates some special problems. The rotor operates at increased speed during field weakening. This creates high mechanical stress in the rotor. The electric frequency becomes high and this causes extra losses. No detailed analysis of the problems is given here. It is, however, assumed that a system without field weakening has improved efficiency. A unique but well-known problem with field- weakening of PM motors is a safety problem. Suppose that the motor operates with high speed in the field weakening region. The electronic control unit creates the field weakening. If the control unit should fail for some reason, the field weakening will disappear. Consequently the magnetic field will revert to its full strength and create an induced EMF that may be much higher than the normal operating voltage. The high voltage may destroy the electronic control unit. There is no simple solution to this problem. However, a solution would be to avoid the use of field weakening.

The trend today is to build hybrid cars with a combination of combustion motor and electric motor instead of a purely electric drive system. The electric motor is used only part of the time in a hybrid system. Consequently, the performance of the electric motor is not as important as in an "all-electric" drive system. An object of the embodiments is to make possible a car with an all-electric drive system. Today, the power source is likely to be a

battery. The system should make the best possible use of the battery energy. So called "all- electric" cars will become dominating in the future when fuel-cells replace the batteries.

The efficiency of the drive system is important not only during driving. The system is supposed to regenerate energy during braking. Then the battery energy passes through the drive system twice. A low loss will help to recover more energy.

A one-quadrant drive system is described and shown in the figures. The complete system will be a four-quadrant system, i.e. with driving and braking in both directions. The features of the embodiments will automatically work properly also in four quadrants.

To summarize, there are four good reasons to avoid the use of field weakening: 1. To make it possible to use PM motors with surface mounted magnets.

2. To operate the PM motor with the best possible efficiency in the base speed region.

3. To avoid the safety problem when field weakening disappears during high speed.

4. To avoid the mechanical stress during high motor speed with field weakening. Below, several embodiments of the invention will be described with references to the drawings. These embodiments are described in illustrating purpose in order to enable a skilled person to carry out the invention and to disclose the best mode. However, such embodiments do not limit the invention. Moreover, other combinations of the different features are possible within the scope of the invention.

Figure 1 shows the equivalent model of a PM motor. The rotor, symbolized by a circle, creates a counter EMF, E, which is proportional to the rotor speed. The magnetic flux is assumed to be constant. There is a relatively small stator resistance R and a relatively small inductance L in the stator circuit. The current I creates a small voltage drop in R and L. The motor impedance Z depends on both R, L and E and is a complex electro-mechanical function. Figure 2 shows the first quadrant of the motor drive system, with rotor speed on the horizontal axis and motor torque on the vertical axis. Maximum torque is limited by the maximum current that is supplied from the frequency inverter to the motor. The maximum current may be defined by the frequency inverter but it may also be limited by the heating of the motor or by magnetic saturation in the motor or by mechanical limitations in the motor. It may be possible to use a higher torque for short periods if the inverter can supply enough current and if the motor can deliver the extra torque.

Maximum speed is defined by the available voltage from the frequency inverter. The highest speed is obtained during no-load operation. The speed decreases when the torque increases because of the voltage drop in the stator circuit. There is also a voltage drop in the inverter and in the power source itself. The sloped dotted line describes this. In reality the change in speed in a well-designed motor is only a few percent. For simplicity the vertical line will symbolize the maximum speed in the following figures.

The motor is a complex load that can be described as an electro-mechanical impedance. Consequently, the operating point may be located anywhere inside the rectangular area. This is true for figure 2 and all the following figures. The rectangles describe the maximum limits for torque and speed, but not necessarily the actual operating points. Figure 3 shows a similar diagram as in figure 2, however including the field- weakening region. The torque is 100 % up to the "base speed", here defined as 100 %. The motor is supposed to operate with constant rotor current. The torque decreases when the field strength is reduced, but then the speed can increase. The result is a curved line representing constant power. It is the purpose of the present invention to create a similar operating range, however without field weakening.

The three-phase system gives a valuable flexibility in the system. There are three separate windings in the motor with six end points. The windings can be connected either in a star or a delta configuration. The two possible winding connections result in two different motors from an electrical perspective. The impedance of the star connected motor is 3 times higher than in the delta-connected motor.

Figure 4 shows a star connected motor with three phases R, S and T. There are three identical phase windings with impedance Z connected in a star. The motor is supplied from a three-phase system with the phase-to-phase voltage U. This is normally referred to as the "main" voltage. A simple trigonometric relation tells that the voltage across one single-phase winding is U/ V3 . The total phase current I flows through the phase winding.

Figure 5 shows the same motor, now in delta connection. The "main" voltage and the phase current are assumed to be the same as in figure 4. The individual phase windings are the same as before, although connected in a new configuration. Seen from the outside this is electrically a different motor. The motor impedance has become 3 times lower. The voltage across a single phase winding is V3 times higher and the current through the same winding is V3 times lower than in the star connected motor. Consequently, the product of voltage and current is unchanged and the power is the same as for the star connected motor.

Figure 5 shows a delta-connected motor where the voltage across the impedance Z is V3 times higher than the voltage across the impedance Z in figure 4. Normally the current through the impedance Z in figure 5 would have been V3 times higher than in figure 4. Instead it is V3 times lower. This may be confusing. The explanation is that the impedance Z doesn't have a fixed value. It is a complex "electro-mechanical impedance" that changes in dependence of the operating conditions, especially the electrical frequency and the mechanical speed of the rotor. Compare with figure 1. A frequency inverter can create the correct operating conditions, both for figure 4 and figure 5.

However, the copper resistance R of the phase winding is the same in figure 4 and figure 5. The value of the copper resistance is of special interest because the "copper losses" are created in this resistance.

Historically, three phase AC induction motors have been operated from the three-phase grid. Then, the motor could be used with two different grid voltages, thanks to the possibility to change between star and delta connection. The grid voltage (in Europe) would typically be 230 V or 400 V (400 = 230-73 ). This was a fixed installation of the motor depending on the actual grid voltage. The motors were normally started by direct connection to the grid, so called line-starting. The starting current (the inrush current) was several times higher than the rated current. Especially when large motors with heavy loads were started, it was necessary to protect the grid from a too large inrush current. One common solution was to use a star-delta reconnection of the stator windings. The motor was started with star configuration and then reconnected to delta configuration. Starting currents were reduced by a factor 1/ v3 .

The present embodiments are based on the same star-delta transformation, however for a different purpose. The classical method was intended to reduce the surge or inrush current for an AC induction motor at the fixed line frequency. The present embodiments are intended to increase the frequency range of the motor and are useful for all kinds of AC motors.

The use of two different phase configurations is not intended to change between two different supply voltages. Instead, it is intended to operate with the same supply voltage but to change the operating range in terms of speed and torque.

Figure 6 explains the idea. It is the same type of diagram as in figure 2. It shows two different operating ranges under the assumption that the motor is supplied from the same frequency inverter with a fixed maximum supply voltage and a fixed maximum current. The star connection results in a smaller voltage across the individual phase windings but in exchange each phase winding gets the full inverter current. This results in a motor that can operate with maximum torque but at a reduced speed. A change to delta configuration results in a motor that can operate at maximum speed but with reduced torque.

This simple change of phase configuration has similar effect as a mechanical gear shift with the factor V3 would have had on a motor with constant winding configuration. Consequently, the embodiment has been named "electrical gearbox".

Figure 7 shows as an example how the winding configuration can be changed between star and delta by help of a three-phase relay. There exist standard industrial methods for performing this star-delta transformation.

The mechanical three-phase relay may be replaced by electrical switches, such as thyristors or transistors, see further below in connection with Fig. 16.

A valuable feature of the "electrical gearbox" is that the motor rotational speed remains constant during the gearshift process. An embodiment of a gearshift procedure would

be to decrease the motor current towards zero, make the relay switchover and then increase the current again until the required torque is obtained. It would be a "soft" no-load gearshift. If electric switches are used, such an electrical "gearshift" may take place within fractions of a second, almost without being noticeable by the user. Modern cars with combustion motors are equipped with a mechanical gearbox that can change between different gear ratios. 4 to 6 different ratios are common. This is in order to extend the speed range of the car. The same principle may be used for an electric drive system and for the same purpose. However, already 2 different mechanical gear ratios can increase the performance of an electric drive system substantially. Moreover, it is noted that the electric motor normally operates at a rotational speed much higher than the rotational wheel speed of a conventional car. A typical maximum rotational speed of an AC motor is 10000 rpm, while the wheels rotate at a maximum of about 1000 rpm. Thus, a speed reduction gearbox is required between the electric motor and the wheels. Such a reduction gearbox can with very few mechanical additional components be converted to a two-gear gearbox, see further below.

Figure 8 shows a combination of the "electric gearbox" with a classical mechanical gearbox. The left part shows a conventional frequency inverter together with the switch device for star-delta transformation. The inverter can supply a maximum voltage U and a maximum current I. The right part shows a mechanical system with a two-speed gearbox, which can have a simple mechanical design and consequently would be easy to incorporate in an electric drive system.

Figure 9 exemplifies a system according to figure 8. The two-speed gearbox may have a base ratio of 1 :3 and for higher output speeds the ratio is 1 : 1. As mentioned above, an electric motor has a much higher speed (rpm) than the wheels of a car. This is not considered in this example.

The motor starts operation with star connection from zero speed up to the base speed 100 %, because a high starting torque is desired. The gearbox ratio is 1 :3. The operation area is inside the rectangle with the upper right corner A. The motor can deliver 100 % torque to the output. For speeds between 100 % and 173 % the motor is switched to delta connection. The available speed range increases by a factor y/3 to 173 % speed. The gearbox ratio is still 1:3. The maximum torque is reduced to 58 %. (1/ v3 = 0.577). The operation area is inside the rectangle with the upper right corner B.

In order to increase the speed further, the mechanical gearbox switches to the new ratio 1:1 and the output speed may increase 3 times. However, in order to obtain maximum torque at the output, the motor is switched back to star-connection. The new operation area is inside

the rectangle with the upper right corner C. The speed goes up to 300 % and the torque is 33 %. (1/ 3 = 0.33).

Finally, for still higher speeds the motor is switched back to delta connection and the maximum speed goes up to 520 % (3 S = 5.20). Now the torque is 19 % (1/(3 V3 )= 0.19). The operation area is inside the rectangle with the upper right corner D.

The ratio 1:3 in the mechanical gearbox was chosen because it matches the ratio in the electrical gearbox. Every new speed range in figure 8 is increased by a factor V3 . It would be possible to use another mechanical gear ratio. It would also be possible to use a multi-speed mechanical gearbox, having more than two ratios. It was noted that the electric motor did not have to change speed during the switching at point A and point C. However, the switching at point B must be coordinated with a speed alteration of the motor.

The combination of a star-delta transformation and a simple mechanical gearbox has created an output speed range that is 5.2 times the basic speed range. This is considered to be a suitable speed range for the drive system in an electric car. It would be possible to extend the speed range further by help of a multi-stage gearbox with more than two gear ratios.

.It would also be possible to use a different ratio than 1 :3 in the mechanical gearbox, if a different speed range ratio than 5.2 times is required.

The combination of an electric switch and a mechanical gearbox has created a drive system with a large speed range, but without the extra losses that may exist in alternative solutions. The motor is allowed to operate within its base speed range all the time. The reliability of the system will be high because of its simplicity.

A PM motor that operates within a wide speed range including field-weakening often uses a speed sensor on the motor shaft for control purposes. If the field weakening can be eliminated it will be much easier to design a sensor-less system. This will give a cost reduction and an increased reliability.

The diagram in figure 9 is based on strict limits for the maximum current and under the assumption that the motor is fully magnetized. It will be possible to use higher currents if the hardware tolerates this, for example during short time accelerations. The real limits may depend on thermal properties in the frequency inverter and the motor.

It may also be possible to use a certain degree of field weakening in order to smooth the transitions between the different operating areas. Figure 9 shows a curved line that represents constant power output. The real output from the system is shown as a staircase. The available output power is always lower than or equal to the theoretical constant output power. It can be understood from figure 9 that it would be possible to operate with different combinations of electrical and mechanical gearbox ratios. There should be an automatic control of the drive system that chooses the best combination with regard to the highest

efficiency or some other criterion. The first rule would be to change from star connection to delta connection when the available voltage is used up to the limit. Another major rule would be to always operate the electric motor with the highest possible speed. This is because an electric motor has its best efficiency at maximum speed and full magnetic field strength. Figure 10 shows an example of a complete drive system according to the invention.

The motor has a maximum speed of 6000 rpm. The maximum wheel speed is assumed to be 1000 rpm. Consequently the gearbox should have a gear ratio of 1 :6 for high vehicle speeds. In the lower speed range the gearbox should have a gear ratio of 1:18.

Figure 11 shows a similar example with field weakening. The base speed of the motor is assumed to be 2000 rpm and the maximum speed with field weakening is assumed to be 5 times higher, i.e. 10 000 rpm. A gear box ratio of 1 : 10 gives the maximum wheel speed 1000 rpm, the same as in the earlier example.

The motor as well as the gearbox in figure 11 has to operate with a much higher speed than the system in figure 10. The higher speed may result in mechanical design problems and higher losses than in the system without field weakening.

Figure 4 shows the motor in star-connection. This may be considered to be the "base configuration". It is the best configuration for starting the vehicle, at least if maximum acceleration is desired. Study one single phase winding Z. The total current I from the frequency inverter flows through the phase winding. Figure 5 with the delta-connection shows that the current through the phase winding now is v3 times smaller at the same time as the voltage across the winding is V3 times larger. This has been described as constant power operation. Both figure 4 and 5 have been drawn under the assumption that the maximum available current from the frequency inverter is I. The motor torque is -v/3 times smaller in delta-connection and this is shown in figure 6. The copper losses in the phase winding are 3 times smaller thanks to the reduced current. This is not the optimum utilization of the motor. The motor windings have been designed for the full current I regardless if the motor is star-connected or delta-connected. Consequently, it would be possible to get 100 % torque from the motor also in delta-connection, if the frequency inverter had the capacity to deliver more current. The best strategy when designing a drive system for an electric car should be to use the motor up to maximum performance. Consequently, the frequency inverter should be designed for this purpose.

Figure 12 shows the same motor configuration as figure 5, however with a current from the inverter that is λ/3 times larger. It is possible to control the current amplitude by help of the frequency inverter. Figure 13 is the same as figure 6, now with an additional area that is defined as "Plus- power-delta". The motor can operate in delta-connection with 100 % torque up to maximum

speed, if the inverter can supply enough current. The same additional areas (not shown) can be added to the diagram in figure 9, one for each of the mechanical gear steps.

Often the motor is the weak link in the total system and consequently the best strategy may be to use the motor to its full capacity. This can be done if the inverter is designed to deliver full current and full voltage according to figure 12 and figure 13. The electric power source, for example a battery or a fuel cell, should be able to deliver the extra power that is indicated as Plus-power-delta in figure 13.

Figure 14 shows another operation area if the inverter is able to deliver 173 % of the rated motor current. It would be possible to get 173 % torque from a motor in star connection. This additional operation area has been indicated as "Plus-power-star". However, this is a more difficult case than the additional torque in figure 13. It requires that the motor can be overloaded and deliver more than the rated torque. The overload capability may be limited by magnetic saturation in the motor or by overheating of the motor windings. This will reduce the size of the Plus-power-star rectangle. Figure 15 shows an example where both Plus-power areas from figure 13 and figure 14 are used. It would be possible to introduce a new limit inside these areas, namely a constant power limit, as indicated in the figure. This may be a favorable mode of operation if the available power from the battery or fuel cell has a maximum limit.

Figures 12 to 15 have described the electrical "gear change" between star and delta connection of the motor windings. There is also a different type of gear change in the system, namely the mechanical gear change. A gear change factor 1 :3 has been described as a suitable solution. It is not possible to compensate fully for this change by help of electrical measures, in order to create a constant power system. It would require that the motor torque can change 3 times when the mechanical gearing is changed. However, some levels of compensation may be used.

The methods that are described in figures 12 to 15 are applicable both for the lower and for the higher mechanical gearing.

Figure 7 shows a circuit for switching between star and delta configuration. A three- phase relay performs the switching. Figure 16 shows a similar circuit, however with thyristors instead of the electro-mechanical relay. There are six pairs of thyristors. Each thyristor pair is used as a bi-directional switch.

The motor operates in star configuration when the thyristor pairs A ^R , A ^s and A ^r are conducting (turned ON), Then the other thyristor pairs must be non-conducting (turned OFF).

The motor operates in delta configuration when the thyristor pairs ^R , ^s and ^τ are conducting (turned ON). Then the other thyristor pairs must be non-conducting (turned OFF).

The drive circuits for the thyristor are of conventional design and are not shown in Figure 16.

Embodiments of the system can be designed as a compromise between all possibilities and limitations, especially thermal limitations that may allow maximum power during limited times only. The maximum power limit may be defined by the power source, the inverter or the motor itself. Thermal limitations are most likely to limit the performance. Thermal sensors on the individual components in the drive system may be used to monitor and control the drive system.

In the claims, the term "comprises/comprising" does not exclude the presence of other elements or steps. Furthermore, although individually listed, a plurality of means, elements or method steps may be implemented by e.g. a single unit or processor. Additionally, although individual features may be included in different claims or in different embodiments, such features may be combined in other combinations, and the inclusion in different claims or embodiments does not imply that another combination of features is not implied, feasible and/or advantageous.

Singular references do not exclude a plurality. The terms "a", "an", "first", "second" etc do not preclude a plurality. Reference signs in the claims are provided merely as a clarifying example and shall not be construed as limiting the scope of the claims.

Although the present invention has been described above with reference to specific embodiment, it is not intended to be limited to the specific form set forth herein. Rather, the invention is limited only by the accompanying claims and, other embodiments than the specific above are equally possible within the scope of these appended claims.