EP2433098A1

EP2433098A1 - A method and a system for determining the angular position of a rotary element, and a rolling bearing including such a system

Info

Publication number: EP2433098A1
Application number: EP10720909A
Authority: EP
Inventors: Stéphane MOISY; Alexis Gatesoupe
Original assignee: SKF AB
Current assignee: SKF AB
Priority date: 2009-05-18
Filing date: 2010-05-17
Publication date: 2012-03-28
Also published as: FR2945626B1; WO2010133546A1; FR2945626A1

Abstract

A magnetic ring is used that is constrained to rotate with a rotary element that is provided with pairs of poles, and sensors, each suitable for issuing an electric signal (U₁ (t)...U_N (t) ) that is representative of a magnetic field. The method comprises the steps that consist in verifying (101, 102) each of the signals issued by the sensors and of determining whether at least one of the sensors is faulty. If at least one sensor is then determined as being faulty, the number of faulty sensors is counted (106). If the number of faulty sensors is equal to one, a replacement signal (U_drep(t₁)) is calculated (108) that is equivalent to the electric signal that would normally have been issued by the faulty sensor, and the previously calculated replacement signal (U_drep(t₁)) is used instead of the electric signal normally issued by the faulty sensor in order to calculate (105) a signal (T (t₁) ) that is representative of the angular position of a rotary element relative to the non-rotary element. The replacement signal is calculated as the difference between the previously-stored sum (S (t) ) of the electric signals as delivered by the set of sensors (C_i) prior to detecting failure of the sensor (C_d), and the sum of the electric signals as issued by the other sensors after detecting failure of said sensor.

Description

A METHOD AND A SYSTEM FOR DETERMINING THE ANGULAR POSITION OF A ROTARY ELEMENT, AND A ROLLING BEARING INCLUDING SUCH A SYSTEM

The invention relates to a method of determining the angular position of a rotary element, such as a ring of a ball bearing, or the equivalent.

WO-A-2007/077389 discloses using Hall effect sensors regularly distributed around a magnetic ring to supply sinusoidal type electric signals that enable the angular position of a rotary element to be determined. In order to improve the accuracy of such a system, it is known to use a number of Hall effect sensors that is greater than three, thus making it possible to combine the signals supplied by the respective cells for the purpose of determining the angular position of the rotary element. Increasing the number of cells in such a system significantly increases the risk that one of the cells will be faulty, thus making the system non-operational. In other words, when one of the Hall effect cells is out of operation, the detector system breaks down and can no longer perform its function.

US-A-2005/0189938 discloses reconstructing the signals from faulty sensors on the basis of a bell curve. That approach is not accurate, and in practice it cannot be used on its own.

Those are the drawbacks that the invention seeks more particularly to remedy by proposing a method of determining the angular position of a rotary element that enables the failure of a sensor, such as a Hall effect sensor, to be accommodated.

To this end, the invention relates to a method of determining the angular position of a rotary element rotating relative to a non-rotary element, wherein a magnetic ring is used that is constrained to rotate with the rotary element and that is provided with P pairs of poles, where P is greater than or equal to one, and N sensors each suitable for issuing an electric signal representative of a magnetic field, with N greater than or equal to two. The method comprises the steps consisting in: a) verifying each of the signals issued by the sensors, and determining whether at least one of the sensors is faulty; b) if at least one sensor is determined as being faulty during step a) , counting the number of sensors that are faulty; c) if the number of faulty sensors is equal to one, calculating a replacement signal equivalent to the electric signal that would normally be issued by the faulty sensor; and d) using the replacement signal calculated in step c) as a replacement for the electric signal that would normally be issued by the faulty sensor in order to calculate a signal representative of the angular position of the rotary element relative to the non- rotary element. The method is characterized in that it includes an additional step that is implemented if no sensor is considered as being faulty during step a) , the additional step consisting in e) storing in a memory a value corresponding to the sum of the signals issued by the sensors; and in that, during step c) , the replacement signal is calculated as being equal to the difference between the sum of the electric signals delivered by the set of sensors prior to detecting failure of the sensor and the sum of the electric signals issued by the other sensors after detecting failure of the sensor.

By means of the invention, a degraded mode of operation is provided when one of the sensors is faulty, with the signal that would normally have been issued by that sensor being replaced with a calculated replacement signal. The invention thus enables the system for determining the angular position of a rotary element to continue to be used, even though one of its sensors is no longer functioning or is no longer functioning correctly.

According to aspects of the invention that are advantageous but not essential, such a method may incorporate one or more of the following characteristics:

- If the number of faulty sensors is greater than or equal to two, an error message is delivered.

- During step c) , the replacement signal is calculated at an instant t₁ using the equation: where, for i lying in the range 1 to N:

• O_i is the offset from zero of the electric signal issued by a sensor C₁;

• A_s is equal to - with a equal to b equal to A₁cosφ_i , and A_i equal to the amplitude of the electric signal issued by a sensor

• is equal to arctan and arctan and • U_i (t₁) is the electric signal issued by the sensor C_i at instant t₁.

^■ In a variant, during step c) , the replacement signal is calculated at an instant ti using the equation: where, for lying in the range 1 to N:

O₁ is equal to the offset relative to zero of the electric signal issued by a sensor C₁; and • U_i (t₁) is the electric signal issued by the sensor C_i at the instant t₁. - During step a) , it is considered that at least one sensor is faulty if the sum of the electric signal issued by the N sensors crosses a threshold value, upwards or downwards . • During step a) , a sensor is considered as being faulty if the electric signal issued by said sensor crosses a threshold value, upwards or downwards, and/or if the electricity consumption of said sensor varies in predetermined manner. The invention also provides a system for determining the angular position of a rotary element that is rotary relative to a non-rotary element by implementing a method as described above, the system comprising a magnetic ring constrained to rotate with the rotary element and provided with P pair(s) of poles, where P is greater than or equal to one, N sensors each suitable for issuing an electric signal representative of a magnetic field, with N greater than or equal to two, and a processor unit for processing signals issued by the various sensors. The system is characterized in that said processor unit is suitable for calculating a replacement signal for a faulty sensor by calculating the difference between the sum of the electric signals delivered by the set of sensors before detecting failure of one of the sensors and the sum of the electric signals issued by the other sensors after detecting the failure of said sensor, and for using said replacement signal instead of the electric signal that would normally have been issued by the faulty sensor, in order to calculate a signal representative of the angular position of the rotary element relative to the non-rotary element.

Advantageously, the sensors used are Hall effect cells .

Finally, the invention provides a rolling bearing comprising a stationary ring and a rotary ring, together with a detector system as mentioned above. The invention can be better understood and other advantages thereof appear more clearly in the light of the following description of implementations and embodiments of a method and a system in accordance with the principle of the invention, given solely by way of example and made with reference to the drawings, in which:

• Figure 1 is a diagram showing the principle of a system in accordance with the invention implementing a method in accordance with the invention;

• Figure 2 is a block diagram of the method in accordance with the invention; and

• Figure 3 is a fragmentary schematic showing the principle of a second system in accordance with the invention.

The system 2 shown in Figure 1 comprises a magnetic ring 4 having two poles, namely a north pole N and a south pole S. The ring 4 rotates about an axis X₄ perpendicular to the plane of Figure 1. Five Hall effect cells C_i to C₅ are regularly distributed around the axis X₄ and around the ring 4, and each of them delivers an analog electric signal in the form of a voltage that varies as a function of time, which signal is delivered to a processor unit 8. For representing a natural integer in the range 1 to 5, the voltage delivered by sensor C_i is written U_i (t) . In normal operation of the system 2, at an instant the signals U_i (t) are conditioned and combined using an approach analogous to that mentioned in WO-A-2007/077389 , the content of which is incorporated in the present application by reference, in order to create two signals at a phase difference of 90° electrical, enabling the angular position of the ring 4 relative to a stationary structure supporting the cells C_i to be calculated. In practice, the ring 4 may be mounted on the rotary ring of a ball bearing, while the cells C_i are supported by a stationary portion of the bearing, e.g. its non- rotary ring. To clarify the drawing, the above-mentioned bearing portions are not shown in Figure 1.

The signals U_i (t) are sent to the unit 8 at a frequency that is a function of the speed of rotation expected of the ring 4, e.g. once every 1 to

10 milliseconds. The angular position of the ring 4 about the axis X₄ may be identified by an angle θ between a straight line D₄ passing via the two interfaces between the north and south poles of the ring 4, and a straight line D_z that is drawn vertically in Figure 1 and that intersects the axis X₄. This angle θ varies as a function of time, and its value as a function of time is written θ(t) .

The unit 8 is designed to issue an electric signal that varies as a function of time and that is representative of the instantaneous value of the angle θ. The value of this signal as a function of time is written T(t) .

The unit 8 comprises a module 82 for conditioning the signals U_i (t) and a module 84 for calculating the value of the signal T(t) as a function of time. The calculation performed by the module 84 is based on the signals as conditioned in the module 82. In particular, the module 82 serves to transform the analog signals constituted by the voltages U_i (t) into digital signals that are suitable for processing by a computer incorporated in the module 84.

There follows a description of the operation of a system such as the system 2 having N Hall effect cells or sensors, where N is a natural integer greater than or equal to three, the system being shown in part in Figure 3. In the particular example shown in Figure 1, N is equal to five. The number of pairs of poles of the magnetic ring of the system shown in Figure 3 is written P. In the particular example shown in Figure 1, the number P is equal to one. In operation of the system shown in Figure 3, the various signals U_i (t) are verified in a first step 101 in order to determine whether one or more of the sensors C₁ of the system is/are faulty. This verification step 101 may be performed by monitoring the sum of the voltages U_i (t) coming from the various sensors C₁. In normal operation, this sum may be considered as being substantially constant, ignoring tolerances. Given the normal value for the sum of the voltages U_i (t), in particular because of prior use of the system 2, it is possible to set a high threshold value and a low threshold value, e.g. respectively equal to the normal value +5% and to the normal value -5%. Under such circumstances, the system is considered as operating normally, i.e. without any of the sensors C_i being faulty, providing the sum S(t) of the voltages U_i (t) delivered by the sensors C_i for lying in the range 1 to N remains between the high and low threshold values .

During a step 102, it is determined whether all of the signals U_i (t) are correct. If they are, the method moves on to a step 103 in which the value of the sum S(t) is stored in a memory for subsequent use as a basis for comparison. Here likewise, the method moves on to a step 104 in which the signals U_i (t) are conditioned in the module 82, and then the value of the signal T(t) is determined in a step 105 within the module 84 by calculating its sine and its cosine. This corresponds to the system 2 operating without fault.

Consider the configuration in which N sensors are distributed regularly around a ring having P poles.

Under such circumstances, and taking the position of a first sensor C_i as a reference position, the various sensors have angular positions about the axis X₄ that satisfy the relationship: where i_ is a natural integer in the range 1 to N, k is a relative integer, and C is a real constant.

When at least one of the sensors C_i is detected in step 102 as being faulty, additional verification is performed in step 106 to determine the number of faulty sensors .

This number of faulty sensors is determined by verifying the operation of each sensor, by comparing the value of its output signal with the threshold values or by monitoring the electricity consumption of the sensor. When a sensor is completely out of operation, it delivers a zero output voltage and no longer consumes any electricity. In other types of breakdown, it may deliver an output voltage that is well above the value that it delivers in normal operation. It is possible in step 106 to count the number of sensors C_i that are faulty. If this number is greater than or equal to two, then the method moves on to a step 107 in which the processor unit 8 delivers an error message, possibly together with a warning signal. Under such circumstances, the system 2 as a whole is considered as being incapable of performing its function.

When only one sensor is considered as being faulty at an instant t_1# e.g. the sensor C_d, where is a natural integer lying in the range 1 to N, the method moves on to a step 108 in which a replacement signal U_drep(t₁) is created for use in the module 84 as a replacement for the signal U^t₁) from the faulty sensor C_d.

Each signal U_i (t₁) for lying in the range 1 to N may be expressed in the form: where is defined as above, CO is equal to the angular frequency, and O₁ is equal to the offset of the signal U_i (t₁) relative to the value zero, this offset being equal to 2.5 volts for example for a Hall effect sensor, while A₁ is the sinusoidal amplitude of the signal U_i (t) about the value 0, . At any instant the sum S(t) of the voltages from the N sensors of the device 2 is expressed as follows:

By developing this expression, the sum S(t) may be expressed as a function of time in the form:

with :

and :

In the particular circumstance where the elements C₁ are identical, it may be considered that the offset values O_i and the amplitude values A₁ are the same for _i lying in the range 1 to N. The sum S(t) may be simplified as follows:

Consider one electrical period, given that one mechanical period (one rotation of the rotary element) makes up P electrical periods, then:

Wi th : Thus :

Whence :

It is assumed that each cell delivers a signal of identical amplitude A_i. Thus:

Whence : Furthermore:

Whence :

However at all instants

Proving that thus amounts to proving that For this purpose, consideration is given to the following integral which by definition is zero over one period:

By making Riemann integrals discrete at a constant pitch, with the signal being quantized into N equal portions corresponding to the phase offset of 2π/N of the N cells, it is possible to write: however :

The same reasoning applies to:

Thus :

However : Thus :

With It is thus shown that : When the sensor C_d is detected as being faulty, at instant t₁ the unit 8 calculates a posteriori a replacement signal U_drep(t₁) that varies as a function of time and that has the value: where S(t) is known since it has previously been stored in memory in step 103.

With the notations used above, U_drep(t₁) may be expressed as follows:

N

5

In the particular circumstance when the values of O_i and A_i are identical for all of the sensors of the system 2, the sum S(t) is constant and the value of U_drep(t₁) may be expressed using the following equation: i o

It is thus possible in the system 2 to create a "virtual" cell or sensor that is considered as generating a signal comparable to that which ought to be delivered by the sensor C_d, thus enabling the modules 82 and 84 to

15 process the signals as in normal operation, even though the signal C_d is faulty. The system 2 thus operates correctly in a degraded mode, even though one of the sensors C_i is out of operation.

As can be seen in Figure 3, the signals issued by

20 the cells C_i are verified by the module 86, as explained with reference to steps 101, 102, and 106, and the value S(t) is stored by said module in a memory 88, during normal operation of the system 2. The values of the sum S(t) stored in the memory 88 correspond to the values of

25 said sum for one revolution period of the ring 4 when all of the sensors C_i are operating normally.

The memory 88 may store the prerecorded values in analog form, e.g. using resistances, or in digital form. When only one sensors C_d is found to be faulty in

30 step 106, then a virtual cell C'_d is created by the module 86 during step 108, by accessing the memory 88 in which prerecorded values for the above-mentioned sum S(t) are stored, and by calculating the virtual output signal for each instant t₁ when the sensor C_d is faulty, as explained above.

The signals to are supplied to the modules 82 and 84 so that they can function normally, implementing steps 104 and 105. In particular, the module 84 can calculate, at each instant t_1# the value of the signal T¹U₁) on the basis of the conditioned signals supplied by the module 82, including the signal The invention thus makes it possible to create a signal that corresponds to the signal that would have been issued by the faulty sensor C_d at an instant t₁ had it not been faulty, with this being based on the signals issued by the other sensors and on the sum S(t) of said signals, as previously determined at a time when the sensor was functioning correctly.

The replacement signal may be calculated in analog or digital manner during step 108. The subtraction operation may be performed by analog means or by a microcomputer.

Claims

1. A method of determining the angular position (θ) of a rotary element (4) rotating relative to a non-rotary element, wherein a magnetic ring (4) is used that is constrained to rotate with the rotary element and that is provided with P pairs of poles, where P is greater than or equal to one, and N sensors (C_i) each suitable for issuing an electric signal U_i (t) representative of a magnetic field, with N greater than or equal to two, the method comprising the steps consisting in: a) verifying (101, 102) each of the signals issued by the sensors, and determining whether at least one of the sensors is faulty; b) if at least one sensor (C_d) is determined as being faulty during step a), counting (106) the number of sensors that are faulty; c) if the number of faulty sensors is equal to one, calculating (108) a replacement signal (U_drep(t₁)) equivalent to the electric signal that would normally be issued by the faulty sensor (C_d) ; and d) using the replacement signal (U_drep(t₁)) calculated in step c) as a replacement for the electric signal that would normally be issued by the faulty sensor (C_d) in order to calculate a signal representative of the angular position of the rotary element (4) relative to the non-rotary element; the method being characterized in that it includes an additional step that is implemented if no sensor is considered as being faulty during step a) , the additional step consisting in: e) storing (103) in a memory (88) a value corresponding to the sum (S (t) ) of the signals (U_i (t)) issued by the sensors (C₁); and in that, during step c) , the replacement signal

(U_drep(t₁)) is calculated as being equal to the difference between the sum (S (t) ) of the electric signals delivered by the set of sensors (C₁) prior to detecting failure of the sensor (C_d) and the sum of the electric signals issued by the other sensors after detecting failure of the sensor.

2. A method according to claim 1, characterized in that if the number of faulty sensors is greater than or equal to two, an error message is delivered (107) .

3. A method according to any preceding claim, characterized in that during step c) , the replacement signal is calculated at an instant t₁ using the equation: where, for _i lying in the range 1 to N: • O₁ is the offset from zero of the electric signal issued by a sensor C₁;

A_s is equal to , with a. equal to V Lu A₁cosφ_i ,

b equal to A₁cosφ_i, and A₁ equal to the amplitude of the electric signal issued by a sensor C₁;

• φ_s is equal to arctan and arctan and

• U_i (t₁) is the electric signal issued by the sensor C_i at instant t₁.

4. A method according to claim 3, characterized in that during step c) , the replacement signal is calculated at an instant t. using the equation: where, for lying in the range 1 to N: is equal to the offset relative to zero of the electric signal issued by a sensor and • U_i (t₁) is the electric signal issued by the sensor C_i at the instant t₁.

5. A method according to any preceding claim, characterized in that during step a) , it is considered that at least one sensor is faulty if the sum (S (t) ) of the electric signal issued by the N sensors crosses a threshold value, upwards or downwards.

6. A method according to any preceding claim, characterized in that during step a) , a sensor (C_d) is considered as being faulty if the electric signal issued by said sensor crosses a threshold value, upwards or downwards, and/or if the electricity consumption of said sensor varies in predetermined manner.

7. A system (2) for determining the angular position of a rotary element that is rotary relative to a non-rotary element by implementing a method according to any preceding claim, the system comprising a magnetic ring (4) constrained to rotate with the rotary element and provided with P pair(s) of poles, where P is greater than or equal to one, N sensors (C₁) each suitable for issuing an electric signal representative of a magnetic field, with N greater than or equal to two, and a processor unit (8) for processing signals (U_i (t) ) issued by the various sensors, said processor unit being suitable for calculating a replacement signal (U_drep(t₁)) for a faulty sensor (C_d) by calculating the difference between the sum (S (t) ) of the electric signals delivered by the set of sensors (C₁) before detecting failure of one of the sensors (C_d) and the sum of the electric signals issued by the other sensors after detecting the failure of said sensor, and for using said replacement signal (at 84) instead of the electric signal that would normally have been issued by the faulty sensor, in order to calculate a signal (T (t) ) representative of the angular position (θ(t)) of the rotary element relative to the non-rotary element.

8. A system according to claim 7, characterized in that the sensors (C_i) are Hall effect cells.

9. A rolling bearing comprising a stationary ring and a rotary ring (4), the bearing being characterized in that it includes a system (2) according to claim 7 or claim 8.