WO2004102124A1

WO2004102124A1 - Sensor for measuring a length or an angle

Info

Publication number: WO2004102124A1
Application number: PCT/EP2004/004409
Authority: WO
Inventors: Daniel Brunnschweiler; Christian Staudenmann
Original assignee: Thyssenkrupp Automotive Ag
Priority date: 2003-05-17
Filing date: 2004-04-27
Publication date: 2004-11-25
Also published as: DE10327680A1; EP1627206A1

Abstract

The invention relates to a device for recording a rotational angle or a length, comprising a sensor assembly (26) for recording the position of the displacement. Said sensor assembly contains a support plate (8) that responds synkinetically in a relative or corresponding manner to the displacement and comprises at least one code track (4) and at least one fixed sensor (1) that acts on the code track (4). The latter (4) is located on the support plate (8) with each point running parallel to the displacement direction and has markings (41) that generate a successively changing bit pattern. A segment of the displacement in the displacement direction, for example the 360° of a circumference of a circle, divided by the number of markings (41) on said segment defines a rate (ß) and the bit pattern is determined by a phase-shifted arrangement of markings (41) in relation to the fixed separation of the rate (ß). To enable the rapid determination of the absolute angular position or the absolute length, selected markings (41) at several pre-defined points are offset (δ) on the support plate (8), causing the change of the bit pattern to be delayed at these points in relation to the rate (ß).

Description

Sensor for measuring a length or an angle

The invention relates to a device for detecting an angle of rotation or a length with a sensor arrangement according to the preamble of claim 1 and to a method according to claims 12 and 13.

The subject of the invention is concerned with the sensory detection of the angle or a length, as is required as the basis for the optimal control of systems, for example in motor vehicles, preferably for the electronic control of electric motors for various actuating devices.

Hall sensors are often used in the prior art for angle measurement. Magnetic elements are attached to the rotatable part, the polarity of which changes in a fixed time pattern. A Hall probe accordingly records the Hall voltage caused by the change in the magnetic field and outputs counting pulses. The difference angle can be determined by counting the counting pulses. The absolute angle of rotation can only be determined by marking a starting point, for example marked by several pulses with the same polarity or by an additional sensor arrangement. In alternative embodiments, as presented for example in DE 101 04 855 AI, the sequence of the counting pulses is used as additional information. However, three sensors are required here.

The counting of counting pulses offers the advantage over the use of analog signals that the signal-to-noise Distance is significantly increased, which means that interference influences the quality of the signal evaluation less.

However, the known systems have several significant disadvantages. In this way, absolute amounts of the angles can only be recorded after passing through a relatively large angular range, for example an entire revolution. There is also the risk of missing a marked starting point or a counting pulse, so that the angle measurement is incorrect. A starting point must also be specified. This either leads to a reduction in the resolution in the vicinity of the starting point by using a longer pulse duration or to the need for a separate starting point detection sensor. In addition, most systems require multiple sensors, sometimes even more than 2 sensors to record the angular position.

A device according to the preamble of claim 1 is disclosed in US Pat. No. 4,628,298. In this document, an embodiment of a generic device is described in which the measurement of the angle of rotation is possible with the help of only one code track (cf. column 9, line 66 to column 10, line 23 and FIG. 6). In this embodiment, the position data and the timing data are embedded in one and the same code track. This is realized by a code track that has markings made of highly reflective material, the extension of which has a fixed grid of 1/3 to 2/3. The code track is formed here by a series of code cycles of the same length. If the measured portion of the signal of the highly reflective material of the markings is 66 percent within a code cycle, this signal is rated as logic 1. is however, this proportion is only 33 percent, so this is rated as a logical 0.

Disadvantages of this known device are that markings of different sizes have to be used and the shorter length of the marking determines the minimum resolvable angle and thus the resolution of the measuring system. As a result, the achievable resolution of the measuring system is relatively low, so that the measuring system cannot meet very high demands on the measuring accuracy. In addition, the choice of the type of marking is restricted because, for example, due to the required 1/3 to 2/3 division no simple magnetic markings with equally long north / south poles can be used. Furthermore, in the known device the cycle length can be minimally only three times as long as the shorter of the applied markings because of the 1/3 to 2/3 division. A shorter cycle length cannot be realized.

The invention is based on the object of specifying a device according to the preamble of patent claim 1, in which a faster and / or more precise determination of the absolute angular position is possible compared to the aforementioned prior art. It is also an object of the invention to provide a method which enables the absolute angular position to be determined more quickly and / or more precisely.

Furthermore, the invention is intended to increase the freedom in the choice of the type of markings, it also being possible to use markings whose 0 and 1 lengths are each of the same length, as is the case, for example, with the north and south poles of permanent magnets , In addition, the fineness of the resolution of the angular values compared to the state of the Technology increased and the cycle length should be reduced.

With regard to the device, this object is achieved by a device having the features of patent claim 1. Advantageous developments of the device according to the invention are specified in the subclaims. With regard to the method, the object is achieved by a method having the features of claims 12 and 13.

In the invention, the sensory detection of the angles between the rotating components takes place in a simple manner and with comparatively little sensory effort.

According to the invention, the device for detecting an angle of rotation or a length comprises a sensor arrangement (26) for detecting the position of the movement, said device comprising a carrier plate (8) which is moved relative to or corresponding to the movement and has at least one code track (4) and at least one fixed one contains sensor (1) acting on the code track (4), the code track (4) being arranged in each point parallel to the direction of movement on the carrier plate (8) and having markings (41) for generating a successively changing bit pattern and a Section of the movement in the direction of movement, for example the 360 ° of a circumference, divided by the number of markings (41) on the section defining a clock (β) and the bit pattern by a phase-shifted arrangement of markings (41) in relation to the fixed division of the clock (ß) is determined, with selected markings (41) au at some predefined locations with an offset (δ) f of the carrier plate (8) are applied, so that the bit pattern changes at these points with a delay compared to the clock (β). A device according to the invention can advantageously be used, for example, in electric power steering devices. However, it is also suitable for use in completely different areas of technology. In a preferred embodiment of the invention, the device contains a sensor arrangement with two parallel or concentrically arranged code tracks, as will be described in more detail below.

The measuring sensor comprises a sensor unit which is fixedly arranged with the non-rotating component, consisting of 2 sensors and two code tracks on a carrier plate which are assigned to the sensors and are connected to the rotatable component. On each of the two code tracks there is an alternating 'l' / 'o' pattern at equal angular intervals for the generation of digital signals. The '1' / 'O' patterns of the two code tracks are arranged exactly the same to each other. This means that if the pattern of one code track changes from '1' to '0' for a certain angular position, the pattern of the other code track changes in exactly the same direction. The pattern can preferably be formed by permanent magnets or with a magnetization, for example with a corresponding north-south polarity or through a shadow mask or by other comparable means, depending on the type of sensor used. Correspondingly, in the case of the use of permanent magnets, the sensors can be Hall sensors or, for shadow masks, simple optical transmitted light sensors consisting of light source and photo semiconductor. In the preferred use of permanent magnets to represent the code tracks, a wide code track, as a two-dimensional code track pattern, can also be simply applied, which are scanned at two distances from the fulcrum by two sensors. The use of 2 different sensors with 2 associated code tracks serves two important purposes. On the one hand, the one code track indicates a fixed cycle in relation to the angle of rotation due to the uniform arrangement of the markings. On the other hand, the two sensors are slightly offset from one another. As a result, the change in the signal, caused by the change from marking '0' to '1', is first detected in one of the two signals of the two code tracks. The direction of rotation of the carrier plate can then be determined from this.

According to the invention, the alternating '1' / 'O' pattern does not change at exactly predefined angular intervals on one of the two code tracks, but is somewhat delayed or offset. This means that this respective selected marking is applied to the carrier plate with an offset. With these changes, which are delayed in relation to the clock, a specific pattern can be formed. The use of this pattern then allows a quick determination of the absolute angular position. After a predetermined amount of the change in angle, the initial detection angle, which corresponds to the pattern length, is run through for the first time, the absolute value of the angle can then be determined within a predetermined coding length in relation to the zero position and multiples of the coding length. The sample length should always have the same angular lengths, ie the same number of cycles, but should consist of different numbers of offsets of markings. The resulting patterns and their assignment to fixed angle values are stored in a data memory in the form of a truth table or the patterns are designed so that they can be mathematically converted to the respective angle. For this purpose, the ones used today electronic memory and processors can be used advantageously. It is particularly preferred here if the bit pattern determined by the phase-shifted arrangement of markings on the code track changes during the rotation of the carrier plate at every cycle (β) and this is unique and unique in order to be able to identify the position immediately and unambiguously. These predefined, unique patterns or words are stored in the memory and can then be compared with the actual pattern of the coding, from which the position can be determined.

The advantage of the method according to the invention is that after the first passage through several changes of the marking until the passage through the defined change in angle - the initial detection angle - the exact angle value can be determined immediately with each further change of the marking.

In a further development of the invention, the clock signal of one of the two code tracks is dispensed with. This cycle is then determined after passing through several changes in the marking of the one remaining code track, for example by simply dividing the number of changes divided by the elapsed time. The offsets of the markings are then related to the clock calculated in this way and processed further in the same way. The direction of rotation is then determined either by means of a truth table which contains the direction of rotation or by an offset in the markings in which both the start of the '1' and the end of the '1' are delayed with respect to the clock. As a result, a pattern with positive and negative signal peaks can be generated, for example by simply forming the difference between the sensor signal and the sensor signal constructed from the clock become. The direction of rotation is determined directly from the simple test of whether the positive or negative signal peak occurs first.

In all embodiments of the method according to the invention, the real sensor signal form (s) can additionally be sampled. In this way, the signal change can be resolved further and more precisely. By sampling the signals, further edge changes can be generated, from whose payment intermediate values for the angle can be determined. This increases the measurement resolution even further.

The aim of all the measures shown in the invention is to determine the angle as quickly and accurately and safely as possible. On the one hand, the physical limits with regard to the geometrical sizes that can be produced by corresponding markings, the signal-to-noise ratio and others must be taken into account. The recording of the angles in digital computers requires word widths of different sizes depending on the resolution. The computing speed can be decisively influenced by the correct size of the word width, for example 4 bit or 8 bit. A distinction must be made between the word widths both for arithmetic operations and for storage operations. The optimal design depends on the architecture of the hardware and can be brought into line with the required measurement resolutions and speeds using the measurement method presented. The length of the truth table, the sample rate in relation to the speed, diameter and geometric dimensions of the markings and other sizes serve as parameters for this.

The same method can be used not only for the preferred angle measurement, but also for length measurements. net. Then the rotation must be replaced by a translation and the angle reference by a length reference. The rotatable component is then a longitudinally displaceable component and the angular units and angular offsets are length units and offsets accordingly. The magnetic means are particularly suitable and thus preferred for code track arrangements with the associated sensors. However, optical, capacitive or inductive means or combinations thereof can also be used.

A special application for the use of the angle measurement is the determination of the angle of the rotor in relation to the pole shoes of the stator of an electric motor, for use as an electrical assistant in a steering system (power steering). This is particularly important for the correct and precise control of electronically commutated motors. Above all, the correct direction of rotation must be ensured when starting the motor. The coding length can advantageously be set equal to the angle of one or more pole pieces.

The use of the bearing cover of the gear arrangement, as in particular when using a ball screw drive, is particularly advantageous as a carrier plate for the code tracks.

In the preferred example of using permanent magnetic markings on the code tracks and using Hall sensors, the method for signal processing and evaluation is shown using schematic figures.

Show it: Figure 1 shows the schematic structure of a steering device with power assistance.

FIG. 2 shows the arrangement of the sensor arrangement, its position with respect to the pole pieces of the electric motor and the schematic assignment to the signal code bits.

Figure 3, consisting of Figures 3a and 3b, shows the waveforms of the Hall signals and the reference to the markings on the code tracks.

FIG. 4, consisting of FIGS. 4a, 4b and 4c, shows the signal profiles of the Hall signals corrected for the angular offset of the two sensors and the combination signal

FIG. 5, consisting of FIGS. 5a, 5b, 5c and 5d, shows the signal profiles of the digitized Hall signals adjusted for the angular offset of the two sensors as well as the reference to the markings on the code tracks and the combination signal and the signal code bits.

For another example, FIG. 6 shows the signal profiles of the digitized Hall signals and the digitized combination signal adjusted for the angular offset of the two sensors.

FIG. 7 shows a truth table for determining the current angle.

Figure 8, consisting of Figures 8a, 8b, 8c and 8d, show other examples of encodings with the associated truth tables. The representations in FIGS. 2, 3, 4, 5, 7 and 8a all relate to one and the same example. FIG. 6 is only intended to show the possibility of signal evaluation of positive signal peaks 111 and negative signal peaks 112. In FIGS. 8b, 8c and 8d, examples of other variants of the coding are shown in order to give an indication of how, after specifying the coding length C and the initial detection angle Δφ, a scheme for suitable angular offsets δ on the two code tracks 4, 5 and thus a suitable one Fixing the markings 41, 51 can be found.

When representing the digitized signals, the signal edge is clearly shown in the respective figures. The increase is determined by the electronics used. However, the rise in the signal edge plays only a subordinate magnitude for the evaluation. It has therefore not been shown in the diagrams to display the flanks at an infinitely steep angle.

The schematic structure of a steering device (29) with auxiliary power support shown in FIG. 1 essentially corresponds to the prior art. It consists among other things of a steering wheel 20, a steering column 21, the steering gear 22 and the two tie rods 24. The tie rods 24 are driven by the rack 23. The drive unit formed from the components servo motor 25, the sensor arrangement 26 and the ball screw drive 27 is used for auxiliary power support. The invention relates to the sensor arrangement 26 and in the special development in the arrangement in a steering device for a motor vehicle. The driver's request is fed into a control unit 28 by the control wheel 20 via a sensor system (not shown) as a signal 281. The sensor arrangement is gear signal 282, the current angle of rotation and derived from it the steering angle fed into the control unit 28. The corresponding control voltage 283 for the electric motor or servo motor 25 is determined therefrom in the control device and output to the servo motor 25. Fast, precise detection of the current angle φ is required for sensitive and fast control. It is sufficient for the control of the servo motor 25 to know the current position of the stator (not shown here) in relation to the pole shoes 6 of the stator (not shown here). The entire angle φ over a complete or even several revolutions, from which the steering angle can be determined, does not have to be determined at high speed and can be determined by counting the passes through the angular range of the coding length C.

FIG. 2 shows, for example, the preferred arrangement of the sensor arrangement 26 consisting of the carrier plate 8 connected to the rotatable component with the two code tracks 4 and 5 with the markings 41 and 51 as well as the axis of rotation Z and the sensor unit 3 connected to the non-rotating component and the sensor unit 3 thereon arranged sensors 1 and 2. The two sensors are advantageously arranged offset by an angular offset 32. The offset sensor arrangement enables better detection of the running direction. The non-rotating components include, for example, the pole shoes 6 of the stator of the servo motor 25. The zero position of the angle of the system is marked with the zero position 31 on the sensor unit 3 and the zero position 81 of the carrier plate 8. In order to clarify the markings, for example the alternating north-south pole orientation of a permanent magnetic track, the designations '0' and '1' have been selected here or synonymous L-Qj and H or synonymous LOJ and | chosen in the representations.

The arrangement of the code tracks in the same polarity is shown here. It is also possible to reverse the code tracks. To clarify the opposite arrangement of the polarity of the two code tracks, only the '1' in one of the two tracks must be replaced by the '0' and the '0' by the '1' label.

The marks' O'and '1' each have the angle of ß / 2. A pole pair 'O' / 'l' thus has the angle = the clock β and corresponds to one bit in the digital further processing of the signals. The coding length C is the angular range for which the angle can be determined exactly after it has passed through the initial detection angle Δφ. In the example, the angle Δφ corresponds to exactly 4 bits or 4 cycles. The second code track 5 has the task of specifying the clock β. If the carrier plate 8 is rotated, the clock signal β is generated from the Hall signal 10 of the sensor 2. On the first code track 4, as already mentioned above, an offset δ is introduced in some markings 41, that is to say one of the polarities, here for example the '1' polarity, is extended by the angle δ.

In the example, the servo motor 25 has 8 pole pairs and a coding length of 90 °, which corresponds to a length of 16 bits, with an initial detection angle Δφ of 4 bits, which corresponds to an angle of 16.875 °. The signal code bits 7, which are used to determine the angle φ, are formed from the offsets δ. Larger word widths allow better resolution, but with more effort. FIG. 3a shows the Hall signal 9 with the signal size I of the sensor 1 in relation to the markings 41 of the first code track 4. In FIG. 3b the corresponding Hall signal 10 of the sensor 2 is shown in relation to the markings 51 of the second code track 5. The angle φ increases with rotation in the direction of rotation 12. The illustration comprises a coding length C, that is 16 angles of a '0' / '1' marking 51. An angular offset 32 of the two sensors 1 and 2 to one another is shown as an example.

By the angular offset 32 of the two sensors 1 and 2, the direction of rotation can be determined in a simple manner. If the change from '0' to '1' occurs in Hall signal 9 before Hall signal 10, carrier plate 8 rotates in direction of rotation 12. Accordingly, carrier plate rotates counter to direction of rotation 12 when the change from '0' to '1' first in Hall signal 10. For this purpose, the angular offset 32 must be smaller than β / 2.

The angular offset 32 is not required for further evaluation, so that it is corrected electronically or numerically. Therefore, in all the other figures, the Hall signals are corrected by the angular offset 32 of the two sensors 1 and 2, i.e. shifted to an angular offset of 0 °, shown.

FIGS. 4a and 4b show the Hall signals 9A and 10A of the sensors 1 and 2, corrected by the angular offset, plotted against the angle of rotation. Furthermore, the combination signal 11 with the signal peaks 111 is shown in FIG. 4c. Here, as in FIG. 3, the signals for the case where the polarities of the two code tracks are arranged in the same direction are shown. In the simplest case, the combination signal can be formed as a difference from the Hall signals 9A and 10A. In the next step, the Hall signals are digitized. The further FIGS. 5 and 6 show a simple digitization to 2 threshold values of the signal, for example I signal> 0 and I signal <0. To increase the resolution, however, sampling can also take place, for example to a plurality of threshold values of the Hall signals 9 and 10.

5a and 5b show the digitized Hall signal 9B or 10B of the sensor 1 or 2 plotted over the angle of rotation φ and shifted in phase by the angular offset 32 of the two angle sensors. The position of the markings 41, 51 of the code tracks 4 and 5 is shown for clarification. FIG. 5c shows the digitized combination signal 11 with the signal peaks 111, which result from the angular offsets δ of the markings 41 of the code track 4. FIG. 5d shows the signal code bits 7 generated from the signal peaks 111 with the values 0 or 1, which can be assigned to the respective clock β. The illustration shows how the sequence of signal code bits 7 “1 0

I 0 "was generated from the combination signal 11. In our example, this value corresponds to exactly 16,875 ° from the starting angle 81, 31 or the starting angle 81 plus the coding length C (in the example 90 °) measured.

The truth table shown in FIG. 7, which matches the representations in FIGS. 2, 3, 4 and 5, shows how, based on that from the signal peaks 111 of the combination signal

II and the associated signal code bits of the angle φ is determined.

Table 1 shows the course of the angle detection and the structure of the truth table using the information shown in the figures. shown example. If the carrier plate 8 begins to rotate in the direction of rotation 12 from the zero position 81, 31, the cycle β, ie a number of '0' / '1' changes, is run through. The cycle ß corresponds to the angle φ, which is shown here in the following 5.625 °, 11.25 °. The signal peaks 111 in the combination signal 11 and thus the signal code bits 7 are generated by the angular offset δ of the markings. This corresponds to the signal code bits 7 shown in FIG. 5d with the digital values 0 or 1. The memory, which in our example has a 4-bit word width, fills bit by bit, as shown in the lower part of Table 1. After passing through the angle of 16,875 °, the memory is filled with a complete 4-bit word from the signal code bits 7. From now on, the current angle value can be read directly from the truth table, as shown in FIG.

the lowest bit of the word in memory, shifting the bits already in memory.

Table 2

Table 2 shows the memory content during construction from the zero position 81, 31 with the word width of 4 bits for the respective clock β / angle φ. As can be seen from the structure of Table 1 and Table 2, the word width also corresponds to the number of cycles until the first angle can be clearly recognized and thus the initial recognition angle Δφ / ß. Figure 8 shows several patterns of truth tables as

Example of how a sequence of signal code bits 7 can be defined for different coding lengths C and initial detection angle Δφ with the associated word widths Δφ / ß, so that after the first run through the initial detection angle Δφ the absolute value of the angle of rotation can be determined with each further run of the angle ß.

Our example is shown again in FIG. 8a. FIGS. 8b, 8c and 8d show the coding lengths of 5, 8 and 10 and word widths of 3bit, 3bit and 4bit, respectively.

It can be seen from the examples with the tables that the successive bit patterns, offset by the angle β, are selected such that they are always unique and appear as unique patterns, i.e. occur singularly, which means that the angular position can be identified at any time by comparing the measured pattern with the saved patterns.

For another example, FIG. 6 shows the digitized Hall signal 9B of the sensor 1 and the digitized Hall signal 10B of the sensor 2 plotted against the angle of rotation and shifted in phase by the angle offset 32 of the two angle sensors. Here, however, the markings are offset by the angle δ such that a combination signal 11 with positive signal peaks 111 and negative signal peaks 112 is produced. The direction of rotation is determined from the chronological order of the occurrence of the positive signal peak 111 and the negative signal peak 112. If the positive signal peak 111 comes before the negative signal peak 1112, the carrier plate 8 rotates in the direction of rotation 12. If the negative signal peak comes 112 in front of the positive signal peak 111, so turns

Carrier plate 8 against the direction of rotation 12.

In the event that it cannot be decided which signal peak 111, 112 comes earlier, the signal edge of the signal 9B is also used. If the signal 9B changes from the low to the high value and the positive signal peak 111 is determined at the same time, the carrier plate rotates in the direction of rotation 12. However, if the negative signal peak 112 is determined at the same time, the carrier plate rotates counter to the direction of rotation 12.

In the particularly cost-effective further development of the system, sensor 2 is omitted and the resulting clock signal 10B is determined numerically from the elapsed time and the '0' / '1' changes of signal 9B as already described above. In this way, the clock signal 10B can be generated mathematically and all of the above-described methods for determining the angle can be used.

However, the use of this inexpensive further development causes a deterioration in the resolution, since at least three, rather more '0 V 1' changes are required to determine the digitized (clock) signal 10B.

Very good results can be achieved with an average reading radius of around 42mm and a magnetic pole distance of 1.5-3.5mm. For example, 64 pole changes of the magnetic field can be displayed. That means 64 complete periods of the Hall signal. With a sample rate of 1/32, this results in 4096 edge changes for each digitized signal 9B or 10B. Since two coding tracks are available, this results in a total of 8192 edge changes. This results in an angular resolution of the system of approximately 0.044 °.

To improve the resolution, a component rotating at a higher speed can be coupled to the rotatable component via a gear with a transmission, to which the carrier plate 8 is then connected. In this way, the angular resolution can be increased by the transmission factor of the transmission with the same number of permanent magnets and Hall sensors.

It is clear that all the designs described above can also be transferred to optical, electrical, inductive or capacitive sensors. Lengths can also be measured in the same way.

A special application is the use of the device described above for the controlled actuation of an electric motor or servo motor 25 for driving a steering system for motor vehicles with electrical assistants. The problem here is to ensure the commutation of the current flow depending on the angular position of the rotor in relation to the stator of the servo motor 25. The current flow must be switched so evenly that no uneven torque is emitted by the electric motor. For this purpose, the carrier plate 8 is coupled to the rotor and the sensor unit to the stator of the electric motor. The position of the rotor with respect to the pole shoes of the stator is then determined with the aid of the measurement results. The coding length C and the clock β is to be determined according to the angle between the pole pieces 6. List of names:

1 sensor 1

2 sensor 2

3 sensor unit

31 Zero position on sensor unit

32 Angular misalignment between sensor 1 and sensor 2

4 code track 1

41 Marking '0' or '1' on code track 1

5 code track 2

51 Marking '0' or '\' on code track 2

6 pole shoe of the electric motor

7 signal code bit

8 carrier plate

81 zero position on carrier plate

9 Hall signal of code track 1

9A Hall signal of code track 1, corrected for offset 32

9B digitized Hall signal of code track 1, adjusted for offset 32

10 Hall signal of code track 2

10A Hall signal of code track 2, corrected for offset 32

10B digitized Hall signal of code track 2, adjusted for offset 32

11 digital combination signal

111 positive signal peak

112 negative signal peak with opposite polarity of the code tracks

12 direction of rotation

20 steering wheel

21 steering column

22 steering gear

23 rack

24 tie rod 25 servo motor

26 Sensor arrangement

27 Ball screw drive

28 control unit

29 steering device

281 Driver request signal

282 sensor output signal

282 Electric motor control voltage

C coding length

I signal size

Z axis of rotation δ angular offset ß cycle = angle of a "0 '/' 1 'mark φ angle of rotation

Δφ initial detection angle

Claims

claims

1. Device for detecting an angle of rotation or a length with a sensor arrangement (26) for detecting the position of the movement, said movement being a support plate (8) with at least one code track (4) and at least one fixed, relative to or corresponding to the movement. contains sensor (1) acting on the code track (4), the code track (4) being arranged in each point parallel to the direction of movement on the carrier plate (8) and having markings (41) for generating a successively changing bit pattern and a section the movement in the direction of movement, for example the 360 ° of a circumference, divided by the number of markings (41) on the section defines a cycle (β), and wherein the bit pattern is defined by a phase-shifted arrangement of markings (41) in relation to the fixed one Classification of the clock (ß) is determined, characterized in that selected markings (41) at some predefined locations with an offset ( δ) are applied to the carrier plate (8), so that the bit pattern changes at these points with a delay compared to the clock (ß).

2. Device according to claim 1, characterized in that the markings (41) not arranged with an offset (δ) are arranged with a division of half a cycle (ß / 2).

3. Device according to claim 1 or 2, characterized in that parallel to the first code track (4), a second, clock-generating reference code track (5) and at least one fixed a standing sensor (2) acting on the code track (5) is provided, this code track (5) having the same number of markings (51) as it is formed on the first code track (4) and this over the range of movement, for example the circumference , are arranged equidistant and each mark determines the clock (ß).

4. The device according to claim 3, characterized in that the two sensors (1, 2) acting on the code tracks (4, 5) are offset by a fixed distance (32), which is preferably at most half of the clock (β) are.

5. Device according to one of claims 1 to 4, characterized in that the bit pattern determined by the phase-shifted arrangement of markings (41) on the code track (4) during the movement of the carrier plate (8) within a fixed predetermined movement section (C) , which comprises the total length or only a partial length of the movement, each time after passing through a further cycle (β).

6. The device according to claim 5, characterized in that the signal code bits (7) generated by the bit pattern with a fixed predetermined word length of, for example, 3, 4 or more bits result in a data word which is clearly a fixed number of clocks (ß) within the fixed range of movement (C) and thus a defined path (φ), ie can be assigned to a track length or an angle.

7. Device according to one of the preceding claims 1 to 6, characterized in that by the phasing- shifted arrangement of markings on the code track (5) certain bit patterns (7) after each clock position (ß) is changed and this is unique and unique, singular, whereby the position is clearly identifiable.

8. Device according to one of the preceding claims 1 to

7, characterized in that the code track (4, 5) and the sensor (1, 2) have magnetic, optical, capacitive or inductive means for determining and recognizing the markings (41, 51), the magnetic means being preferred.

9. Device according to one of the preceding claims 1 to

8, characterized in that the device has electronic means (28, 281, 282, 283) for controlling and evaluating the rotation angle sensor arrangement (26), such as preferably a microprocessor control.

10. The device according to claim 9, characterized in that the singular patterns of the signal code bits (7) are stored in an electronic memory as a reference with the assigned path, i.e. a track length or an angular position, the carrier plate (8).

11. The device according to claim 10, characterized in that electronic means for the angular position-dependent control and regulation of an electrically commutated motor (25) of an actuator system in the motor vehicle, for example for controlling a headlight height control, an adjusting shaft for a mechanical variable valve train or a power steering (29) , are provided.

12. A method for determining a length or an angle by means of a device according to one of claims 1 to 11, characterized in that the bit pattern, which is determined by a phase-shifted arrangement of markings (41) in relation to the fixed division of the clock (ß) a signal code bit (7) is determined by detecting the markings with a sensor (1) after moving the carrier plate by one or more clocks (β) in any direction by comparison with the fixed clock (β), the sequence of signal code bits (7) is stored in a first memory with a fixed, but freely selectable word length of two or more bit word lengths, the stored signal code bits (7) with a table stored in a second memory, in which an assignment between Positions, ie Path lengths or angles of rotation, and corresponding to the word length stored in the first memory of two or more bit words, are compared, and the position is determined therefrom.

13. A method for determining a length or an angle by means of a device according to one of claims 1 to 11, characterized in that the bit pattern, which is determined by a phase-shifted arrangement of markings (41) in relation to the fixed division of the clock (ß) , by reading the markings with a sensor (1), after moving the carrier plate (8) by one or more clocks (ß) in any direction by comparison with the fixed clock (ß) a signal code bit (7) for each clock is determined the sequence of signal code bits (7) is stored in a memory with a fixed, but freely selectable word length of two or more bit word lengths, the stored signal code bits (7) based on a fixed rule, with which an assignment between position and path length or Angle of rotation, is determined, is evaluated, and from this the position is determined.

14. The method according to claim 12 or 13, characterized in that the markings (41) are designed so that they are out of phase in such a way that both the beginning and the end of the shift forming the bit pattern are out of phase with the clock (β).