WO2014048467A1

WO2014048467A1 - System for processing electric signals delivered by a sensor, sensor unit comprising such a system, instrumented bearing and method for processing electric signals

Info

Publication number: WO2014048467A1
Application number: PCT/EP2012/068992
Authority: WO
Inventors: Matthieu RIOTEAU
Original assignee: Aktiebolaget Skf
Priority date: 2012-09-26
Filing date: 2012-09-26
Publication date: 2014-04-03
Also published as: DE112012006942T5

Abstract

This system (S) for processing electric signals delivered by a sensor unit, said electric signals comprising a sine electrical voltage (SIN) and a cosine electrical voltage (COS), comprises at least four comparators (C1, C2, C3, C4) adapted to compare a first parameter to a second parameter, said parameters being a mean output electrical voltage (Um) of the sensor unit and the voltage of the sine electrical signal (SIN) for a first comparator (C1), said mean output voltage (Um) and the voltage of the cosine electrical signal (COS) for a second comparator (C2), the voltages of the sine signal (SIN) and the cosine signal (COS) for a third comparator (C3), said mean output voltage (Um) and the average of the voltages of the sine (SIN) and cosine (COS) signals for a fourth comparator (C4). This system also comprises a first XOR gate (XOR1), having in inputs the output signals (U1, U2) of the first and second comparators (C1, C2), and a second XOR gate (XOR2), having in inputs the outputs signals (U3, U4) of the third and fourth comparators (C3, C4). The output signals (Ur1, Ur2) of the system (S) are the output signals of the first and second XOR gates (XOR1, XOR2).

Description

SYSTEM FOR PROCESSING ELECTRIC SIGNALS DELIVERED BY A SENSOR, SENSOR UNIT COMPRISING SUCH A SYSTEM, INSTRUMENTED BEARING

AND METHOD FOR PROCESSING ELECTRIC SIGNALS TECHNICAL FIELD OF THE INVENTION

The invention concerns a system for processing electric signals delivered by a sensor and a sensor unit for sensing the angular position of a rotatable element with respect to a fixed element comprising such a processing system. The invention also concerns an instrumented bearing comprising such a sensor unit. The invention further concerns a method for processing electric signals delivered by a sensor.

BACKGROUND OF THE INVENTION

Some rotation sensing applications need square signals having a determined resolution. Square output signals, often obtained thanks to magnetic sensors, do not provide a high enough resolution for such applications, mainly because the magnetization of the encoder of such sensors is not efficient enough. It is therefore needed to interpolate sine and cosine primary output signals by a certain number, for example 2 or more, to increase the resolution.

It is known to interpolate output cosine and sine signals thanks to digital processings and resistor arrays. The issue of these techniques is that they have a determined resolution because of digital processing, or need a relatively large number of electronic components.

SUMMARY OF THE INVENTION

The aim of the invention is to provide a new system for processing electric signals delivered by a sensor, which permits to obtain an interpolated signal in a more efficient way than in the prior art.

To this end , the invention concerns a system for processing electric signals delivered by a sensor unit, said electric signals comprising a sine electrical voltage and a cosine electrical voltage. This system is characterized in that it comprises at least four com parators adapted to com pare a first parameter to a second parameter, said parameters being:

the mean output electrical voltage of the sensor unit and the voltage of the sine electrical sensor output for a first comparator;

- said mean output voltage and the voltage of the cosine electrical output for a second comparator; the sine voltage and the cosine voltage for a third comparator; said mean output voltage and the average of the voltages of the sine and cosine outputs for a fourth comparator,

in that the system further comprises a first XOR gate, having in inputs the output signals of the first and second comparators, and a second XOR gate having in inputs the outputs signals of the third and fourth comparators, and in that the output signals of the system are the output signals of the first and second XOR gates.

Thanks to the invention, the square signals obtained by interpolation of the sine and cosine signals delivered by a sensor are completely analogically processed and have virtually no resolution limits. Moreover, the number of components used to perform the interpolation is relatively small. This permits to reduce the cost and the space occupied by the system.

According to further aspects of the invention which are advantageous but not compulsory, such a system may incorporate one or several of the following features:

- The system comprises resistors connected between the sine and cosine inputs of the system and the positive inputs of the comparators, whereas it further comprises resistors connected between the positive inputs and the outputs of the comparators.

The system comprises a voltage generator adapted to deliver a voltage equal to the mean output voltage of the sensor unit, whereas said voltage generator is connected to the negative inputs of the first, second and fourth comparators.

The comparators are operational amplifiers.

The output value of each comparator is 1 when the value of the second parameter is superior or equal to the value of the first parameter.

The invention also concerns a sensor unit for sensing the angular position of a rotatable element with respect to a fixed element, comprising a processing system as mentioned here above.

The invention also concerns an instrumented bearing comprising a sensor unit as mentioned here-above.

The invention also concerns a method for processing electric signals delivered by a sensor unit, said electric signals comprising a sine electrical voltage and a cosine electrical voltage. This method is characterized in that it comprises the following steps:

- a) compare a mean output electrical voltage of the sensor unit to the voltage of the sine electrical signal;

- b) compare said mean output voltage to the voltage of the cosine electrical signal;

- c) compare the voltages of the sine and cosine signals to each other; - d) compare said mean output voltage to the average of the voltages of the sine and cosine signals,

- e) on the basis of the comparisons made at steps a) to d), generate signals having the values 1 or 0 depending on the result of the comparison,

- f) generate a square signal having the value 1 when the values of the signals generated at step e) on the basis of the comparisons made at steps a) and b) are different, and the value 0 in the other cases;

- g) generate a square signal having the value 1 when the values of the signals generated at step e) on the basis of the comparisons made at steps c) and d) are different, and the value 0 in the other cases.

According to further aspects of the invention which are advantageous but not compulsory, such a method may incorporate one or several of the following features:

Steps a) to g) are performed permanently.

The method comprises a further step h) consisting in adding a hysteresis effect in the comparisons made at steps a) to d).

BRIEF DESCRIPTION OF THE DRAWINGS

The invention will now be explained in correspondence with the following figures, as an illustrative example. In the annexed figures :

- figure 1 is an electrical scheme of a processing system according to a first embodiment of the invention,

- figure 2 is an electrical scheme of a processing system according to a second embodiment of the invention,

- figure 3 is a scheme of a trigonometric circle depicting the method according to the invention,

- figure 4 is a time versus voltage chart showing output sine and cosine signals and, with a different scale, interpolated square signals obtained with the processing system and the method according to the invention;

- figure 5 is a sectional view of an instrumented bearing including a sensor unit according to the invention.

DETAILED DESCRIPTION OF SOME EMBODIMENTS

The processing system S of the invention is adapted to generate square signals from a sine electrical voltage SI N and a cosine electrical voltage COS delivered by a sensor unit 2 for sensing the rotation speed of a rotatable element 6 with respect to a fixed element 8 around a rotation axis X-X'. Sensor unit 2 comprises an encoder 20 fast in rotation with rotatable element 6 and at least one sensing element 22 which detects the angular position of encoder 20.

Sensor u n it 2 is advantageously cou pled to a rol li ng bearing 4 to form an instrumented bearing represented on figure 5. Rolling bearing 4 comprises an inner ring 40, an outer ring 42 and rolling elements, such as balls 44, arranged between inner ring 40 and outer ring 42. Encoder 20 is fixed to outer rotating ring 42 while sensing element 22 is fixed to non-rotating inner ring 40. Outer ring 42 is fixed to rotatable element 6 while inner ring 40 is fixed to fixed element 8. The various assembly possibilities of encoder 20, sensing element 22, inner and outer rings 40 and 42 and rotatable and fixed elements 6 and are not limited to the embodiments represented on figure 5.

Processing system S is advantageously integrated in a printed circuit board, which also includes sensing element 22.

SIN and COS electrical voltages oscillate around a mean voltage value Um, which is the voltage value delivered by sensing element 22 in its idle state.

The aim of the invention is to generate, from SIN and COS electrical signals, two square electrical signals which are interpolated by two with respect to SI N and COS electric signals. Interpolation by two means that the frequency of output square signals Ur1 and Ur2 shown on figure 4 is the double of the frequency of SIN and COS signals.

To obtain square signals U r1 and U r2, processing system S comprises four comparators C1 , C2, C3 and C4. Comparators C1 to C4 are operational amplifiers having a positive input, a negative input, a positive power supply and a negative power supply connected to the ground. Comparators C1 to C4 deliver an output value of 1 when the positive input is superior or equal to the negative input, and 0 when the positive input is inferior to the negative input. Comparators C1 to C4 have respective output signals U1 , U2, U3 and U4.

Processing system S comprises a reference generator subsystem adapted to generate a voltage equal to mean value Um thanks to a voltage divider. This subsystem comprises a voltage source G1 . A resistor R1 is connected between voltage source G1 and a point P 1 . A resistor R2 is connected between point P 1 and the ground. The electrical potential on point P1 is equal to mean voltage Um. Depending of the type of sensor used, mean voltage Um can be different. The nominal voltage of voltage source G1 and/or the values of resistors R1 and R2 are therefore determined so as to obtain a voltage at point P1 approximately equal to Um.

The negative input of comparator C1 is connected to point P1 . The positive input of comparator C1 is fed with SIN signal. Comparator C1 therefore compares the value of the voltage of SIN signal to mean voltage Um. When SIN voltage is superior or equal to mean value Um, U1 equals 1 . This means that the electrical angle corresponding to SIN has a value comprised in the area of the trigonometric circle where the sinus is superior to 0, in the upper half of the trigonometric circle. This area is represented by angle A1 on figure 3.

The negative input of comparator C2 is connected to a point P2, which has the same potential as point P1 . The negative input of comparator C2 is therefore at mean voltage Um. The positive input of comparator C2 is fed with COS electrical signal. This means comparator C2 compares the value of the voltage of COS signal to mean voltage Um. U2 therefore equals 1 when COS is superior or equal to Um. This corresponds to an area of the trigonometric circle of figure 3 where COS is superior to 0. This area corresponds to the right portion of the trigonometric circle and is represented by angle A2.

The negative input of comparator C3 is fed with SI N electrical signal, while the positive input of comparator C3 is fed with COS electrical signal. This means comparator C3 compares the values of the voltages of SI N and COS signals one to the other. U3 equals 1 when COS is superior or equal to SIN. U3 therefore equals 1 when the electrical angle is comprised in an area of the trigonometric circle of figure 3, which corresponds to the area of the trigonometric circle located below a line L1 oriented with respect to an horizontal line by an angle β1 of 45°, this line L1 crossing the first quadrant of the trigonometric circle in its middle. This area is represented by angle A3 on figure 3.

The negative input of comparator C4 is connected to point P2 and is therefore at mean value Um. The positive input of comparator C4 is connected to a point P3. A resistor R3 is connected between point P3 and a wire W3 which is fed with COS signal. A resistor R4 is connected between point P3 and a wire W4 which is fed with SIN signal. R3 and R4 having the same value, point P3 is therefore at a potential equal to the average of the voltages of COS and SIN signals. Comparator C4 therefore compares the average of the voltages of COS and SIN to mean voltage Um. U4 is therefore equal to 1 when the average of COS and S I N is superior or equal to Um. This corresponds to an area represented by angle A4 on figure 3. This area is located above a line L2 passing on the center of the trigonometric circle, perpendicular to line L1 and oriented with respect to an horizontal line by an angle β 2 of 45°. L2 crosses the second quadrant of the trigonometric circle in its middle.

The values of U 1 , U2, U3 and U4, which can be either 1 or 0, are representative of the position of a specific reference point of encoder 20, which corresponds to the magnetic origin of encoder 20, with respect to the trigonometric circle. In order to obtain a square signal, one must generate a 1 value or a 0 value depending on the angular sector of the trigonometric circle in which the reference point of encoder 20 is located. To this end, processing system S comprises a first gate XOR1 , which is an "exclusive or gate". XOR1 has two inputs and one output. The inputs of XOR1 are signals U 1 and U2. The output of XOR1 gate is 0 when U 1 and U2 equal 1 , 0 when U1 equals 1 and U2 equals 0, 1 when U 1 equals 0 and U2 equals 1 , and 0 when U 1 and U2 equal 0. XOR1 therefore equals 1 when U 1 and U2 are different. This situation corresponds to the upper left and lower right quadrants of the trigonometric circle.

Processing system S also comprises a second "exclusive or gate" XOR2, whose inputs are U3 and U4, and which has an output value of 1 when U3 is different from U4 and 0 when these values are equal. This situation corresponds to the upper and lower quadrants of the trigonometric circle on figure 3.

To sum up, the value of XOR1 equals 0 when the reference point of encoder 20 is at an angle comprised between 0° and 90°, equals 1 when the angle is comprised between 90° and 180°, equals 0 when the angle is comprised between 180° and 270° and equals 1 , when the angle is comprised between 270° and 0°. This permits to generate a square signal Ur1 , which reaches its high value twice and its low value twice in one full rotation of encoder 20, while COS and SIN only have one positive period and negative period. The frequency of the square signal is therefore the double of the frequency of COS and SI N signals. This permits to increase the accuracy of sensor unit 2 compared to a simple discretization of COS and SIN.

The generation of square signal Ur2 is similar, except the fact that it is offset by 45° with respect to Ur1 . Ur1 and Ur2 are therefore shifted along the horizontal time axis on figure 4. The value of XOR2 equals 1 between 45° and 135° and between 225° and 315°. This value equals 0 between 315° and 45° and between 135° and 225°.

On figure 4, the scales of the vertical axis of the chart are different for SIN and COS signals on a first hand, and for Ur1 and Ur2 signals on the second hand. Ur1 and Ur2 are vertically shifted for the clarity of the figure. The low portions of signals Ur1 and Ur2 correspond to values approximately equal to 0, while the high portions of Ur1 and Ur2 correspond to values approximately equal to 1.

Advantageously, the comparisons performed by comparators C 1 to C4 and the operations performed by XOR gates XOR1 and XOR2 are realized permanently.

A second embodiment of the invention is represented on figure 2. In this embodiment, elements similar to the first embodiment have the same references and work in the same way. This embodiment differs from the first embodiment by the fact that it comprises hysteresis resistors which act on input signals to produce a counter reaction.

Processing system S comprises a resistor R5 connected between the SIN source and the positive input of comparator C1 . Similar resistors R6 and R8 are respectively connected between the COS source and the positive input of comparator C2, and the COS source and the positive input of comparator C3. A resistor R7 is connected between the S I N source and the negative input of comparator C3, in order to equilibrate the negative and positive inputs of comparator C3. Such a resistor R7 is optional. Resistors R5, R6 and R8 permit to produce a resistance on the positive input signals of comparators C1 to C3, in order to allow the hysteresis effect, thanks to hysteresis resistors R9 to R12. A similar resistance is produced by resistors R3 and R4 on the positive input signal of comparator C4. Resistors R9 to R12 are respectively connected between the respective positive input and the output of comparators C1 , C2, C3 and C4.

Hysteresis resistors R9 to R12 slightly shift transition points of comparators C1 to C4 counter-clockwise when encoder 20 is rotating counter-clockwise, and oppositely slightly shift transition points of comparators C1 to C4 clockwise when encoder 20 is rotating clockwise. This allows increased output stability, especially when processing system S is likely to oscillate around such a transition point.

Square signals Ur1 and Ur2 of both embodiments can be used, thanks to further electronical processings, to determine the rotation speed and direction of rotation of rotatable element 6 with respect to fixed element 8. More precisely, the transitional portions of square signals Ur1 and Ur2 comprised between their high and low values are used.

According to a non-shown embodiment of the invention, processing system S may comprise more than four comparators and more than two XOR gates.

Claims

1 . - System (S) for processing electric signals delivered by a sensor unit (2), said electric signals comprising a sine electrical voltage (SIN) and a cosine electrical voltage (COS), wherein it comprises at least four comparators (C1 , C2, C3, C4) adapted to compare a first parameter to a second parameter, said parameters being:

a mean output electrical voltage (Urn) of the sensor unit (2) and the voltage of the sine electrical signal (SIN) for a first comparator (C1 );

said mean output voltage (Urn) and the voltage of the cosine electrical signal (COS) for a second comparator (C2);

the voltages of the sine signal (SIN) and the cosine signal (COS) for a third comparator (C3);

said mean output voltage (Urn) and the average of the voltages of the sine (SIN) and cosine (COS) signals for a fourth comparator (C4),

wherein the system further comprises a first XOR gate (XOR1 ), having in inputs the output signals (U1 , U2) of the first and second comparators (C1 , C2), and a second XOR gate (XOR2) having in inputs the outputs signals (U3, U4) of the third and fourth comparators (C3, C4),

and wherein the output signals (Ur1 , Ur2) of the system (S) are the output signals of the first and second XOR gates (XOR1 , XOR2).

2. - System according to claim 1 , wherein it comprises resistors (R3, R4, R5, R6, R8) connected between the sine and cosine inputs of the comparators (C1 - C4) and the positive inputs of the comparators, and wherein it further comprises resistors (R9, R10, R1 1 , R12) connected between the positive inputs and the outputs of the comparators.

3. - System according to any preceding claim, wherein it comprises a voltage generator (G1 ) adapted to deliver a voltage equal to the mean output voltage (Urn) of the sensor unit (2), and wherein said voltage generator (G1 ) is connected to the negative inputs of the first, second and fourth comparators (C1 , C2, C4).

4. - System according to any preceding claim, wherein the comparators (C1 -C4) are operational amplifiers.

5. - System according to any preceding claim, wherein the output value (U1 -U4) of each comparator (C1 -C4) is 1 when the value of the second parameter is superior or equal to the value of the first parameter.

6. - Sensor unit (2) for sensing the angular position of a rotatable element (6) with respect to a fixed element (8), comprising a processing system (S) according to one of the previous claims.

7. - Instrumented bearing (2, 4) comprising a sensor unit (2) according to claim 6.

8. - Method for processing electric signals delivered by a sensor unit (2), said electric signals comprising a sine electrical voltage (SIN) and a cosine electrical voltage (COS), wherein the method comprises the following steps:

- a) compare a mean output electrical voltage (Urn) of the sensor unit (2) to the voltage of the sine electrical signal (SIN);

- b) compare said mean output voltage (U rn) to the voltage of the cosine electrical signal (COS);

- c) compare the voltages of the sine and cosine voltages (SI N, COS) to each other;

- d) compare said mean output voltage (Urn) to the average of the voltages of the sine and cosine signals (SIN, COS),

- e) on the basis of the comparisons made at steps a) to d), generate signals (U1 ,

U2, U3, U4) having the values 1 or 0 depending on the result of the comparison,

- f) generate a square signal (Ur1 ) having the value 1 when the values of the signals (U1 , U2) generated at step e) on the basis of the comparisons made at steps a) and b) are different, and the value 0 in the other cases;

- g) generate a square signal (Ur2) having the value 1 when the values of the signals (U3, U4) generated at step e) on the basis of the comparisons made at steps c) and d) are different, and the value 0 in the other cases.

9. - Method according to claim 8, wherein steps a) to g) are performed permanently.

10. - Method according to any of claims 8 and 9, wherein it comprises a further step h) consisting in adding a hysteresis effect in the comparisons made at steps a) to d).