High resolution absolute encoder

Abstract

A high resolution encoder device using a number of static sensors distributed on a circumference, and a rotating disc, having several sections of two different properties on an annular track according to a predefined pattern, placed so that the sensors can sense the properties of the sections of track in proximity. An auxiliary unit is also provided and by itself or in combination with the sensor signal values, provides a first low resolution position value. In a first processing step, the signals of each sensor are compared to a threshold, and bit values zero or one for each sensor are set according to the comparison result. All bits are then set in a digital word, in order to create a code number, which in combination with the output of the auxiliary unit, is characteristic of a first low resolution position value. For each low resolution position value, a mathematical combination of signals values is defined. The values of the result of said mathematical combination of the signals value is then used as an entry variable to pre-recorded tables to output high resolution position value.

Claims

1. A high resolution encoder device to measure the angular position of a rotating element, comprising: a) first means to provide a first angular position value; b) a rotating disc fixed to the rotating element, said rotating disc including a circular track having sections associated with a first and second property according to a given pattern, c) a number of fixed sensors positioned in proximity of said circular track, each sensor outputting an electrical signal having a first range of analog values when in proximity to said circular track section associated with said first property, and a second range of analog values when in proximity to said circular track section associated with said second property, each sensor receiving continuously changing intermediate values when said rotating disc rotates from a position at which said sensor is in proximity to said circular track section associated with said first or second property to a position at which said sensor is in proximity to said circular track section associated with said second or first property respectively, d) processing means to process the analog values of said electrical signals; and e) a memory to store pre-recorded tables of signal selection, signal combinations and table position values, wherein for every said first angular position value there is selected: i) a number of said electrical signals, ii) at least one mathematical combination of analog values of said selected electrical signals and iii) at least one of said pre-recorded table position values, wherein said selected electrical signal analog values are combined using said selected mathematical combination to provide a combined value, and wherein at least one high resolution angular position value of said rotating disc is retrieved from at least one of said selected table position values using said combined analog value as an entry variable.

2. The encoder device of claim 1 wherein said first means to provide a first position value comprises an auxiliary unit providing a low resolution position value and wherein said electrical signal analog values are digitized by comparison to threshold values to provide a digital word code, and wherein said code and low resolution position value taken together are characteristic of said first position value.

3. The encoder device of claim 1 wherein said first means to provide a first position value includes processing steps of: a) associating a Boolean value to each electrical signal by comparing it to a first threshold to compose a digital word code; b) selecting for said digital word code at least one electric signal, at least one second threshold value and at least one table of said first position values; c) obtaining at least one Boolean value by comparing said selected electrical signal analog value to one of said selected second threshold values; and d) retrieving said first position value from said selected position value table using said Boolean values as entry variable to said selected table.

4. The encoder device of claim 1 wherein at least one section associated with said first or second properties includes permanent magnets generating a magnetic field, and wherein said sensors output said electrical signal analog values related to said magnetic field.

5. The encoder of claim 2 wherein said auxiliary unit includes two magnetic sensors distanced angularly by approximately one quarter of a turn and at least one magnet fixed to said rotating disc, said magnet inducing variable electrical signals in said magnetic sensors during disc rotation.

6. A method for measuring the angular position of a rotating element, using a high resolution encoder device comprising: providing a high resolution encoder device comprising: a) first means to provide a first angular position value; b) a rotating disc fixed to the rotating element, said rotating disc including a circular track having sections associated with a first and second property according to a given pattern, c) a number of fixed sensors positioned in proximity of said circular track, each sensor outputting an electrical signal having a first range of analog values when in proximity to said circular track section associated with said first property, and a second range of analog values when in proximity to said circular track section associated with said second property, each sensor receiving continuously changing intermediate values when said rotating disc rotates from a position at which said sensor is in proximity to said circular track section associated with said first or second property to a position at which said sensor is in proximity to said circular track section associated with said second or first property respectively, d) processing means to process the analog values of said electrical signals; and e) a memory to store pre-recorded tables of signal selection, signal combinations and table position values, and selecting, for every said first angular position value: i) a number of electrical signals; ii) at least one mathematical combination of analog values of said selected electrical signals and iii) at least one of said pre-recorded table position values, wherein said selected electrical signal analog values are combined using said selected mathematical combination to provide a combined value, and wherein at least one high resolution angular position value of said rotating disc is retrieved from at least one of said selected table position values using said combined value as an entry variable.

7. The method of claim 6 wherein said first means to provide a first position value comprises an auxiliary unit providing a low resolution position value and wherein said electrical signal analog values are digitized by comparison to threshold values to provide a digital word code, and wherein said code and low resolution position value taken together are characteristic of said first position value.

8. The method of claim 6 wherein said first means to provide a first position value performs the processing steps of: a) associating a Boolean value to each electrical signal by comparing it to a first threshold to compose a digital word code; b) selecting for said digital word code at least one electric signal, at least one second threshold value and at least one table of said first position values; c) obtaining at least one Boolean value by comparing said selected electrical signal analog value to one of said selected second threshold values; and d) retrieving said first position value from said selected position value table using said Boolean values as entry variable to said selected table.

9. The method of claim 6 wherein at least one section associated with said first or second properties includes permanent magnets generating a magnetic field, and wherein said sensors output said electrical signal analog values related to said magnetic field.

10. The method of claim 7 wherein said auxiliary unit includes two magnetic sensors distanced angularly by approximately one quarter of a turn and at least one magnet fixed to said rotating disc, said magnet inducing variable electrical signals in said magnetic sensors during disc rotation.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIG. 1 shows an absolute encoder with sensors and auxiliary unit according to the present invention;

(2) FIG. 2 shows an implementation of the encoder with a pattern of 5 magnets, 5 sensors and two digital Hall sensors;

(3) FIG. 3 shows the variations of the sensors signals for the implementation shown in FIG. 2;

(4) FIG. 4 shows an example of code values obtained with a pattern and 5 sensors where same code value 25, shown as an example, is obtained at two different sectors;

(5) FIG. 5 shows a combined signal at a given angular position;

(6) FIG. 6 shows a general block diagram for the processing of the sensor signals to output a high resolution position value;

(7) FIG. 7 shows sensor signal variations in the range of two sectors for which the same code is obtained; and

(8) FIG. 8 is a block diagram describing the processing steps for obtaining a low resolution position value.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

(9) In FIG. 1 there is shown an encoder arrangement, according to the present invention. An encoder disc 101, fixed to a rotating shaft 103, includes at its outer circumference, a track with a number of sectors like 102a and 102b of different properties. Sectors like 102a filled in black color represent sectors of a first property, and sectors like 102b filled in white color represent sectors of a second property. Sensors 105a, 105b, 105c, 105d and 105e are static and placed in proximity of the track. The sensors provide output analog signals respectively A0, A1, A2, A3, A4 and A5. The output of the sensor varies from a first range of values when the sensor is in proximity to a sector of a first property like 102a, and a second range of values when the sensor is in proximity to a sector of a second property like 102b. The rotating encoder disc also includes a second track 110.

(10) A static auxiliary unit 109, placed in proximity to the second track 110, is connected to a microcontroller (CPU) 107, and provides a low resolution position value available to the CPU 107. The second track 110 and the auxiliary unit 109 may be designed according to any prior art method. Optimally, the auxiliary unit 109 and the second track 110 will be designed to provide very low resolution for low cost considerations. The analog signals A0-A5 are input to an analog to digital converter 106 and their digital values are then input to the CPU 107. The CPU 107 also receives low resolution position value from the auxiliary unit. The CPU 107 then processes the signals values A0-A5 and the low resolution position value according the method of this invention described further herein, and outputs a high resolution position value.

(11) FIG. 2 shows an example of pattern of 50 sectors that can be used with 5 sensors. The encoder disc includes an external cylindrical wall 200. Internally to this wall, 5 magnets having an arc shape 201a to 201e are affixed to form the pattern. The magnets are all magnetized radially with the same direction, for example inward. A pattern is obtained wherein sections of the internal circumference of the cylindrical wall 200 are covered with magnets (first property) or not covered with magnets (second property). According to the specific pattern shown in FIG. 2, the magnets 201a-201e cover respectively and approximately 6, 4, 4, 3 and 8 sectors, and the uncovered sections of the wall 200 circumference adjacent to magnets 201a-201e in the counter clock wise direction extend respectively and approximately over 3, 4, 4, 6 and 8 sectors. Five static Hall sensors 202a-202e are uniformly positioned on a circular line, internal and concentric with the external wall 200 and these sensors output a signal in relation with the magnetic field. Whenever there is a rotation of the disc, external wall 200 and magnets 201a-201e rotate around the sensors.

(12) During rotation, magnet-covered sectors and magnet-uncovered sectors alternate and are positioned in proximity to the sensors, generating a variable sensor signal, which is a function of the disc rotation angle. Also shown in FIG. 2 is a circle magnet 204 fixed to the rotating disc and magnetized so that a first half circle is North pole, and the other half a South pole. Two static digital Hall sensors 203a and 203b are placed close to the magnet, are connected to the CPU and can sense which magnet pole is in proximity. These two digital Hall sensors are placed at a 90 degree angular position distance. In this embodiment, these digital Hall sensors implement the above mentioned auxiliary unit and provide position information with a .sup.th of a turn resolution to the CPU according to their two possible respective states i.e. in proximity to a North or South pole. The Boolean state of these digital Hall sensors will be further referred as H1 and H2.

(13) FIG. 3 shows an example of the signals variations as a function of the angular rotational position of the encoder disc for the pattern of FIG. 2. The five signals S1-S4 have basically the same shape with an angular shift according to the respective sensor angular position. Apart from angular shift, signals S1-S4 may have small shape differences due to tolerances in the geometrical arrangement or in the Hall sensor characteristics. We refer here to a zero cross of signal as the position where the value of this signal crosses a threshold T0, typically close to the average value. The pattern has been designed so that zero crossings are approximately equidistant. At each position, the Boolean value 1 is associated with a signal if its value is greater than the threshold T0, and zero if smaller. Thus, five associated Boolean values are set in a five bit word, thus creating a code for this position. The number of sectors is equal to the number of zero crossings, and for the specific pattern of FIG. 2 there are 50 sectors. Since the number of codes is limited by the number of sensors (5) to 2{circumflex over ()}5=32, then several sectors may have the same code.

(14) FIG. 4 is a graph showing the code value for each sector. In FIG. 4 there is also shown 4 ranges of positions for which the digital Hall sensors have a given state. These ranges are shown by double arrows labeled with the state of the digital Hall sensors H1 and H2. The combination of the code of a sector, and the state of the digital Hall sensors provides a unique corresponding identifier of the sector with the resolution of a sector size. As an example, in FIG. 4, it can be seen that the same code 25 is obtained for sectors A and B which are sectors 9.sup.th and 33.sup.rd when counting sectors from angle 0 of the graph. At sector A, digital Hall sensors have states H1=0 and H2=0, while at sector B digital hall sensors have different states H1=1 and H2=1. The code of a sector combined with the digital Hall sensor's state thus gives a unique corresponding identifier of the sector, and thus gives low resolution position value of the disc rotation angle.

(15) It is of particular advantage that this low resolution position value is obtained with a resolution greater than 2{circumflex over ()}n, (32 in this particular case). This increased resolution is obtained requiring neither an increased number of sensors nor miniaturization of the pattern sections.

(16) This is in contrast with U.S. Pat. No. 9,007,057 by Villaret; in this prior art patent, in order to obtain a high resolution, the number of sensors must be increased. This also requires the use of small pattern sections, precisely positioned. According to this prior art patent, the pattern section should have position and length precision better than one half a sector size in order to produce the Gray code. In total, the number of sensors is increased and precision requirements of the pattern are increased, resulting in higher complexity and cost. In particular, if magnets are used for pattern, increasing resolution of the digital code becomes impractical.

(17) In a pre-processing step, for each sector, a least one particular combination of a number of Hall sensor signals is selected and stored in the form of tables in the CPU 107 memory. These tables are referred to as combination tables. A combination of Hall sensor signals may include any mathematical operation. The purpose of the combination is to obtain a combined signal which is monotonous with the angular position in the range of the sector. A further purpose of this combined signal is to create a signal which is less sensitive to mechanical and electronic tolerance of the particular implementation. Examples of signals combinations are:

(18) A weighted sum of all the signals that are monotonous within the particular sector. For example, weight may be 1 if the signal is growing, or 1 if the signal is decreasing

(19) The arc-tangent of the ratio of two signals that are zero crossing at the two sector ends

(20) The arc-tangent of the ratio of two combined signals each combined signal being the weighted sum of monotonous signals in the code range.

(21) It must be understood that many other combinations can be used.

(22) Using combination tables, for each sector, identified as explained above, a particular combination of signals is defined. The CPU 107 acquires the low resolution position value and uses it as an entry variable to the combination tables, and calculates the correspondent combined signals.

(23) FIG. 5 shows the variation of the signals S1-S5 (also of FIG. 3) within a short angular position range. At the particular position P, the CPU 107 acquires the low resolution position value, and selects the correspondent combination of signals to be used. In the particular example shown in FIG. 5, the combination is defined by C=S1S3. The variation of the combined signal C is shown.

(24) During the pre-processing step, for each low resolution position value and in the respective low resolution position range, combined signals values and angular position are stored in position tables, as a function of the combined signal values. Thus for each low resolution position, at least one table is defined. Optionally, a mathematical model that approximates the variations of the combined signal values with position may be found. In this case, these tables will contain the parameters of the mathematical model.

(25) Referring again to the particular case shown in FIG. 5, at position P, the particular position table is defined. This table defines a function of the position P=f(C), where C represents the value of the combined signal, and P the position value. The sector with code 25 is defined between the two zero crossing of signals S3 and S1, occurring respectively at positions P1 and P2. As can be seen, the combined signal C is monotonous over the whole range of the sector. The function can thus be inversed. This inversed function P=f(C) has been recorded in the pre-processing step in a specific position table of high resolution. Alternatively, the signals S3 and S1 could also have been used, since in this sector they are monotonous. At position P there is shown in FIG. 5 the slope of the signals C, S3 and S1 with respective double arrows labeled DC, D3 and D1.

(26) As can be seen, the slope of signals S3 and S1 (D3 and D1) are much smaller than the slope of the combined signal. This means that a small error in reading of signal S3 or S1 would result in a relatively large error of the position reading P. In contrast, the slope of the combined signal is always higher, and thus same error in reading results in a smaller position reading error. For clarity purpose, the combined signal has been shown using two signals. In a practical embodiment, a greater number of signals can be used with various mathematical combinations to reduce the sensitivity to errors in the analog signal value readings.

(27) FIG. 6 is a simplified block diagram of the encoder processing method.

(28) In first block 601, a position value of low resolution is acquired from the auxiliary unit. As an example, in the above-described embodiment, this is achieved by reading the state of the two digital Hall sensors (203a and 203b of FIG. 3),

(29) In block 602, the output of the sensors is read providing analog signals.

(30) In block 603, the said analog signals are compared to a threshold to provide digital values and combined to make a digital word code.

(31) In block 604 the code value combined with the position value of low resolution of the auxiliary unit provides a unique corresponding identifier of a position value of intermediate resolution. The obtained code value may be ambiguous due to the fact that it is characteristic of a number of positions, however the low resolution position value acquired from the auxiliary unit resolves this ambiguity.

(32) Other means of reading the position value of intermediate resolution may be used.

(33) In block 605, the position value of intermediate resolution is used as an entry variable to the tables:

(34) A first table defines the type of mathematical combination to use with the signals.

(35) A second table defines which position table to use.

(36) The defined combination and position tables are then selected.

(37) In block 606 a combined signal is calculated.

(38) In block 607, the position value of high resolution is read from the selected position table, using the combined signal value as an entry variable to the selected position table.

(39) The output of block 607 is thus a high resolution position value output by the encoder.

(40) This process is executed in a continuous cycle 608.

(41) In the above described embodiment, an auxiliary unit is required in order to resolve code ambiguities. The auxiliary unit is preferably of very low resolution in order to reduce its cost.

(42) In order to avoid additional cost of this auxiliary unit, a preferred embodiment will be shown which is able to provide the position information of intermediate resolution by means of further analog signal processing.

(43) In FIG. 7 there are shown the variations of the signals S1-S5 of the embodiment shown in FIG. 2. The signals are the same signal as shown in FIG. 3, and are shown in two graphs around the positions of sectors A and B of FIG. 4. At these two positions, the same digital code 25 is obtained. In the left graph (FIG. 7) around A, it can be seen that signal S4 always has a value greater than a threshold value T1, while in the right graph around B, the signal S4 always has value smaller than threshold value T1. The two sectors can thus be distinguished by comparing the value of the signal S4 to the threshold value T1. In the pre-processing step, the value of the threshold T1, and the number of the signal to be compared to this threshold are recorded and stored in the CPU 107. In case there are several sectors with the same code, then several thresholds and signals like T1 and S4 can be used. For example, for the particular case shown in FIG. 7, two thresholds T1 and T2 can be used respectively with signals S4 and S2.

(44) Finally, it is thus possible to obtain the position value of intermediate resolution by further processing of the analog signals.

(45) FIG. 8 shows a block diagram for the processing of the analog signals in order to obtain a position value of intermediate resolution.

(46) In block 801, all analog signals values are acquired.

(47) In block 802, all signal are compared to a threshold T0 to provide a Boolean value and calculate a digital code.

(48) In block 803 the digital code is used as an entry variable to the tables to select:

(49) 1A number of threshold values T1,T2.

(50) 2Signals to be compared to these thresholds (n1, n2 . . . )

(51) 3Table of positions of intermediate resolution.

(52) In block 804 binary values are associated to selected signals (n1, n2 . . . ) by comparing each to the correspondent threshold. These binary values are set in additional bits of the code, thus creating a secondary code.

(53) In block 805, secondary code is used as entry to the selected table of low resolution position, and low resolution position value is obtained.

(54) Many variations can be conceived within the scope of this invention, wherein the use of several thresholds provides a position value of intermediate resolution.

(55) Having described the invention with regard to certain specific embodiments, it is to be understood that the description is not meant as a limitation, since further modifications will now become apparent to those skilled in the art, and it is intended to cover such modifications as fall within the scope of the appended claims.