Patent Application
20090102460
This application claims priority to UK Patent Application No. 0714663.2, filed on Jul. 27, 2007, which is hereby incorporated by reference in its entirety for all purposes. This application also claims priority to UK Patent Application No. 0716477.5, filed on Aug. 23, 2007, which is hereby incorporated by reference in its entirety for all purposes.
The present invention relates to position sensors operable to determine the position of a probe relative to a scale and in particular to a method of operating such sensors.
Position sensors are utilized in a number of applications including as angular position sensors for rotors such as switch elements, shafts or similar or as linear position sensor for tooling machines, elevators etc. In rotary applications, a sensing probe is usually fixed in position adjacent to a rotating scale although the alternative arrangement is potentially possible. In linear applications, a sensing probe is provided adjacent to a linear scale, and either the probe is movable relative to a fixed scale or the scale is movable relative to a fixed probe.
In a typical such position sensor, one or more sensing elements will be provided in the probe. The sensing elements are operable to sense properties of the scale that vary with relative position either angular or linear as appropriate. Typically, the probe will comprise a pair of sensing elements that are offset in position and/or orientation allowing a determination of position by comparing the different sensing element outputs.
In one example, scale may comprise a magnetized rotor and the sensing elements may comprise magnetic field sensing elements. By determining the outputs of each magnetic field sensing element an indication of the local magnetic field orientation can be obtained and thus an indication of the orientation of the rotor can be obtained. In other embodiments where the scale may comprise a toothed paramagnetic or ferromagnetic wheel and the probe may comprise one or more magnetic sensing elements back biased with a magnet. In such cases, the passage of a tooth may be determined by a variation in magnetic field magnitude and the orientation of the wheel by maintaining a count of tooth passing events.
Whilst each of these methods are effective for determining probe position, they are each vulnerable to errors due to DC offsets in the sensing elements and they may require a division operation between the output values of at least two sensor elements.
It is an object of the present invention to provide a new method for determining the position of a probe relative to a scale.
According to a first aspect of the present invention there is provided a method of determining the position of a probe relative to a scale wherein the probe contains a sensor of the type comprising one or more pairs of orthogonal sensing elements operable to measure properties varying locally with the position of the probe relative to the scale, the method comprising the steps of: applying a periodically varying bias to each sensing element; determining the sum of and/or the difference between the outputs of pairs of sensing elements to provide a sum output signal and/or a difference output signal for each pair of sensing elements; and determining one of: the phase difference between the sum and difference output signals; the phase difference between the sum output signal and the applied bias; or the phase difference between the difference output signal and the applied bias for each pair of sensing elements to thereby determine the position of the probe relative to the scale.
In this manner, the position of a probe relative to a scale may be determined in a way that is not affected by a systematic bias in the output level of each sensing element.
The probe may be fixed and the scale moveable or the scale may be fixed and the probe movable. Relative movement may be rotational or linear. In rotational embodiments the scale may be or may be incorporated into a rotor such as a shaft, a switch element, a disc, a rotary encoder or similar. In linear embodiments the scale may be or may be incorporated into a track, a rod or similar.
The scale may provide a single scale period of property variation or may provide multiple repeating scale periods of property variations. In the case wherein the scale has a single scale period the method may involve directly determining the absolute position of the probe relative to the scale. In the case wherein the scale has multiple scale periods, the method may involve determining the position of the probe relative to a particular scale period. In such cases, the absolute position of the probe may be determined by maintaining a count of successive scale periods.
In the context of the present invention a pair of orthogonal sensing elements should be understood to comprise a pair of sensing elements orientated or positioned such that the pair of sensing elements obtains output signals that are substantially 90° out of phase as a result of the position of the probe relative to the scale. The method may be applied to a sensor comprising a single pair of orthogonal sensing elements. In an alternative embodiment, the method may be applied to a sensor comprising multiple pairs of orthogonal sensing elements. If multiple pairs of orthogonal sensing elements are provided each pair may be orientated or positioned orthogonally or may be orientated or positioned at another or at a desired offset.
The output signal from each sensing element may be amplified using suitable amplification means, if required or desired. Additionally or alternatively, the sum and/or difference signals may be amplified by suitable amplification means, if required or desired. Suitable amplification means may be a differential amplifier. Additionally or alternatively, the amplified signal may be filtered using a band pass filter after amplification.
Preferably, the periodic bias applied to the sensing elements in the or each pair is 90° out of phase. The periodic bias may be any form of periodic signal including a square or rectangular waveform. Most preferably, the sinusoidal bias applied to each sensing element is a temporal sinusoidal bias having a common timebase ω. For example, the bias applied to one sensing element may be have the form sin(ωt) whilst the bias applied to the other sensing element may have the form cos(ωt). If the phase difference between the sum and difference signals is determined, the output phase difference shifts by two periods in phase over a full period of scale variation. The position of the probe within a given period of scale variation thus may be calculated by halving the phase difference angle between the sum output and the difference output. If only one of the sum or difference signals is used, the position of the probe within a given period of scale variation may be calculated directly from the phase difference angle between the sum or difference signal and either applied bias signal. In some embodiments, both methods may be applied: the phase difference angle between the sum or difference signal and either applied bias signal used to determine the probe position and the phase difference angle between the sum output and the difference output being used to improve the resolution of the determined probe position.
In any such embodiment, the phase difference may be determined by use of a phase comparator. The output of the phase comparator may be a pulse width modulated signal wherein the duty period varies in proportion to the determined phase difference. Preferably, the duty cycle varies between zero and full width over a full scale period. Such a signal may be converted to an analogue output by low pass filtering. Alternatively, such a signal may be converted to a digital signal by starting a counter upon each rising pulse and stopping said counter on each falling pulse. Preferably, the counter is clocked by a signal having the same timebase as the applied bias. The output value is thus independent of any timebase drift.
The properties sensed by the sensing element may be a local magnetic field or local radiation intensity, in particular local visual radiation intensity.
In the event that the monitored property is a magnetic field, the sensing elements are magnetic field sensing elements and either: the scale is magnetized or has a magnet attached thereto; or the probe is magnetized or has a magnet attached thereto. The magnetic field sensing elements may be Hall elements. Alternatively, the magnetic field sensing elements may be magnetoresistive elements such as anisotropic magneto resistors (AMR), giant magneto resistors (GMR) or tunneling magneto resistors (GMR). In such magnetic sensing embodiments, the applied periodic bias may be a sinusoidal bias current.
The magnetic field variation provided by the scale may comprise variation in magnetic field orientation or variation in magnetic field magnitude. The variation may comprise a single period of variation or may comprise multiple periods of variation.
In embodiments having multiple periods of magnetic field orientation variation, the scale may comprise alternating bands of magnetic material of opposite polarity. A pair of such band may thus define a scale period. The probe may comprise either a pair of orthogonally orientated magnetic sensing elements or a pair of magnetic sensing elements having a parallel orientation and a lateral separation equal to half the width of each band. In such cases the determined phase difference may be directly related to the position of the probe within a particular scale period.
In embodiments having multiple periods of magnetic field magnitude variation, the scale may comprise a plurality of regularly spaced projecting teeth. The probe may comprise either a pair of magnetic sensing elements having a parallel orientation and a lateral separation equal to half the width of each tooth. A tooth and an adjacent gap may thus define a scale period. The probe may be provided with one or more back biasing magnets. In such cases the determined phase difference may be directly related to the position of the probe within a particular scale period.
In such embodiments, the scale may be linear or curved. If curved, the scale may comprise an arc, a ring, a wheel, a helical or spiral form.
In embodiments having a single period of magnetic variation, the scale may comprise a magnetic rotor having a magnetic axis substantially perpendicular to its axis of rotation. A full rotation may thus define the single scale period. The probe may comprise a pair of orthogonally orientated magnetic sensing elements. The axis of rotation may be substantially orthogonal to the orientations of both of the magnetic sensing elements. In such cases the determined phase difference may be directly related to the orientation of the rotor relative to the probe.
In the event that the monitored property is radiation intensity, probe comprises a pair of radiation sensing elements and the scale comprises a pair of encoder tracks, each track having sinusoidally varying reflection or transmission properties which are mutually 90° out of phase. Each radiation sensing element may be operable to detect only light reflected or transmitted by one of the tracks. One or more light sources are additionally provided for illuminating the encoder tracks. In such cases: if the reflection property of the tracks varies sinusoidally, the light source is provided on the same side of the encoder tracks as the sensing elements; and if the transmission property of the tracks varies sinusoidally, the light source is provided on the opposite side of the encoder tracks as the sensing elements. In such embodiments, the applied periodic bias may be a sinusoidal variation in the intensity of the light sources or a sinusoidal bias applied to the output of the sensing elements.
The encoder tracks may comprise a single period of sinusoidal variation along their length. In such embodiments, the phase difference may be directly indicative of position. Alternatively, the encoder tracks may each comprise multiple periods of sinusoidal variation along their lengths. In such embodiments, the phase difference may be indicative of position within a particular scale period.
The encoder tracks may be linear or curved. If curved, the tracks may trace an arc, a ring, a helical or spiral form. The tracks may be provided on a planar, linear, arcuate, annular, circular, helical or spiral substrate as required.
According to a second aspect of the present invention there is provided a position sensor for sensing the position of a probe relative to a scale comprising: one or more pairs of sensing elements; bias generating means operable to apply a periodic bias to each sensing element; one or both of summing means for calculating a sum output signal for each pair of sensing elements the sum output signal being the sum of the outputs of the pair of sensing elements and difference means for calculating a difference output signal for each pair of sensing elements the difference output signal being the difference between the outputs of the pair of sensing elements; and phase comparator means for determining one of: the phase difference between the sum and difference output signals; the phase difference between the sum output signal and the applied bias; or the phase difference between the difference output signal and the applied bias for each pair of sensing elements to thereby determine the position of the probe relative to the scale.
The apparatus of the second aspect of the present invention may incorporate any or all features of the method of the first aspect of the present invention as desired or as appropriate.
The sensing elements may be operable to sense a local magnetic field. In such embodiments, the sensing elements may be magnetic field sensing elements such as Hall elements. Alternatively, the sensing elements may be radiation sensing elements.
The sensor may comprise an amplifier for amplifying the output signal. The sensor may comprise a demodulator for demodulating the output signal before amplification. The sensor may comprise a band pass filter for filtering the output signal after amplification.
According to a third aspect of the present invention there is provided a method of removing inherent offset from an output signal of a Hall element of the type having a pair of biasing contacts (designated supply negative SN and supply positive SP) and a pair of output contacts (designated output negative ON and output positive OP), the method comprising the steps of: commutating connection to the contacts such that each of the contacts is operable as each of the pair of bias contacts SN, SP and operable as each of the output contacts ON, OP in a regular cycle; and applying a suitable filter to the obtained output such that the element due to the inherent offset signal is substantially removed but the element due to the true Hall output signal is substantially retained.
By commutating the contacts in this manner, different properties are induced into the element of the obtained output due to the Hall signal as compared with the that due to the inherent offset.
Preferably, the designated bias contacts are driven by a periodically varying bias current which may be a sinusoidally varying bias current.
The obtained output signal may be amplified by use of a suitable amplification means, such as a differential amplifier. The applied filter may be operable to remove offset introduced by the amplification means in addition to offset inherent to the Hall element.
In one configuration, the commutation might be arranged such that each contact is operable as each of as each of the pair of bias contacts SN, SP and as each of the output contacts ON, OP in turn in a four step cycle. In one example, the contacts may change function in a unidirectional rotating cycle around the Hall element. In such a cycle, each contact may be operable in sequence as SN, OP, SP, ON in turn. In such a sequence each contact would be one step out of phase with the adjacent contacts. In such a cycle, the inherent Hall offset varies in sign from phase to phase.
In an alternative example, the bias contacts and the output contacts may change function according to contra-rotational cycles. In such a contra-rotational cycle one opposing pair of contacts may each be operable in sequence as SN, ON, SP, OP in turn and the other opposing pair of contacts may be operable in sequence as ON, SN, OP, SP. In such a sequence each pair of opposed contacts would be two steps out of phase with each other. In such a cycle, the inherent Hall offset maintains a constant sign between phases.
The commutation cycle may be controlled by a switching means in response to a chopping signal. In such an embodiment, the filter may comprise a band pass filter. The upper cutoff frequency of said band pass filter is preferably less than the frequency of the chopping signal. This enables noise from the chopping signal to be substantially eliminated and also, if a unidirectional commutation cycle is used, the inherent Hall offset to be substantially eliminated. Preferably, the lower cutoff frequency is greater than the frequency of the sinusoidal bias current. This enables the substantial elimination of DC offsets inherent in the signal such as an inherent amplification offset and also, if a contra-rotational commutation cycle is used, the substantial elimination of the inherent Hall offset.
Alternatively, the commutation cycle may be controlled by a modulator in response to an input modulation signal. In such an embodiment, the filter means may comprise a demodulator operable in response to a signal substantially identical to the input modulation signal.
According to a fourth aspect of the present invention there is provided an offset removal circuit for removing inherent offset from an output signal of a Hall element of the type having a pair of biasing contacts (designated supply negative SN and supply positive SP) and a pair of output contacts (designated output negative ON and output positive OP), the circuit comprising: commutation means operable to apply a commutation cycle to the contacts; means for applying a bias current to the currently operable bias contacts; means for receiving an obtained output signal from the currently operable output contacts; and filter means for applying a filter to the obtained output signal such that such that the element due to the inherent offset signal is substantially removed but the element due to the true Hall output signal is substantially retained.
The offset removal circuit of the fourth aspect of the present invention may incorporate any or all features of the third aspect of the present invention, as desired or as appropriate.
The method of the third aspect of the present invention can be applied to the Hall element signals analyzed in the method of the first aspect of the present invention. Similarly, an offset removal circuit according to the fourth aspect of the present invention may be incorporated into a magnetic field orientation sensor of the second aspect of the present invention.
According to a fifth aspect of the present invention there is provided a method of determining the orientation of a magnetic field using a sensor of the type comprising one or more pairs of Hall elements operable to measure substantially orthogonal field components, the method comprising the steps of: applying a sinusoidally varying bias current to each Hall element; determining the sum of and/or the difference between the outputs of pairs of orthogonal Hall elements to provide a sum output signal and/or a difference output signal for each orthogonal pair of Hall elements; and determining one of: the phase difference the sum and difference output signals; the phase difference between the sum output signal and the applied bias current; or the phase difference between the difference output signal and the applied bias current for each orthogonal pair of Hall elements to thereby determine the orientation of the field relative to said pair of orthogonal Hall elements.
According to a sixth aspect of the present invention there is provided a magnetic field orientation sensor comprising: one or more pairs of substantially orthogonal Hall elements; bias current generating means operable to apply a sinusoidal bias current to each Hall element; one or both of summing means for calculating a sum output signal for each pair of orthogonal Hall elements the sum output signal being the sum of the outputs of the pair of Hall elements and difference means for calculating a difference output signal for each pair of orthogonal Hall elements the difference output signal being the difference between the outputs of the pair of Hall elements; and phase comparator means for determining one of: the phase difference between the sum and difference output signals; the phase difference between the sum output signal and the applied bias current; or the phase difference between the difference output signal and the applied bias current for each pair of orthogonal Hall elements to thereby determine the orientation of the magnetic field relative to said pair of orthogonal Hall elements.
So that the invention may be more clearly understood one embodiment will now be described further below, with reference to the accompanying drawings, in which:
a shows a schematic diagram of a position sensor according to the present invention implemented as a magnetic linear position sensor;
b shows a schematic diagram of a position sensor according to the present invention implemented as an alternative embodiment of a magnetic linear position sensor;
c shows a schematic diagram of a position sensor according to the present invention implemented as a magnetic rotational position sensor;
d shows a schematic diagram of a position sensor according to the present invention implemented as an alternative embodiment of a magnetic rotational position sensor;
e shows a schematic diagram of a position sensor according to the present invention implemented as a further alternative embodiment of a magnetic rotational position sensor;
a shows schematically the variation in phase difference over a rotation period, a pole period or a tooth period as calculated using the circuitry of
b shows schematically the variation in phase difference over a rotation period, a pole period or a tooth period as calculated using the circuitry of
Turning now to
In use, a periodic bias current Ix, Iy is applied to each of the Hall elements X, Y, Ix and Iy being 90° out of phase. For simplicity of explanation in the described embodiments the periodic bias is considered to have a sinusoidal form. Other periodic forms may however be used including rectangular waveforms. In such cases some additional filtering may be required.
The resultant outputs Vx, Vy of Hall elements X and Y are fed to summing means 101 and Difference means 102. The summing means 101 is operable to output a summed signal Vsum equal to the instantaneous sum of Vx and Vy. The difference means 102 is operable to output a difference signal Vdiff equal to the instantaneous difference between Vx and Vy. The outputs Vsum and Vdiff are fed into a phase comparator 103, which is operable to determine the phase difference between them. The phase angle between Vsum and Vdiff provides an indication of the orientation of the magnetic field component lying parallel to the x-y plane, as is explained in greater detail below.
The magnetic field vector component parallel to the x-y plane may be described in terms of its magnitude B and its orientation angle relative to the x-axis, α. This being the case, the output voltage Vx of Hall element X is proportional to Bx=B*cos(α). Additionally, the output voltage Vy of Hall element Y is proportional to By=B*sin(α). Furthermore, both Vx and Vy are proportional to the biasing current Ix, Iy applied to Hall elements X and Y respectively. In particular this proportionality is described by the magnetic sensitivity S. Typically, if the two Hall elements X, Y are like Hall elements, then assuming that they have equal magnetic sensitivity (Sx=Sy=S) is fair. Accordingly, Vx and Vy are given by:
Vx=S*B*cos(α)*Ix
Vy=S*B*sin(α)*Iy
In a conventional sensor, wherein field orientation is determined by considering the magnitudes of Vx and Vy the bias currents Ix and Iy are held equal and constant and thus are not directly relevant to the calculation of the field orientation. However, in the present invention, the bias currents Ix and Iy are varied sinusoidally over time (t) with a 90° phase difference therebetween, such that:
Ix=I*sin(ωt)
Iy=I*cos(ωt)
As such, the Hall element X, Y outputs will be given by the following identities:
Vx=S*B*cos(α)*I*sin(ωt)
Vy=S*B*sin(α)*I*cos(ωt)
At the next stage of processing, the sum and difference output signals Vsum and Vdiff are generated by summing means 101 and difference means 102. These signals are given by the identities below:
Vsum=Vx+Vy=S*B*I*(cos(α)*sin(ωt)+sin(α)*cos(ωt))
Vdiff=Vx−Vy=S*B*I*(cos(α)*sin(ωt)−sin(α)*cos(ωt))
Using the trigonometric identity:
sin(A±B)=sin(A)*cos(B)±cos(A)*sin(B)
the above identities can be re-expressed as
Vsum=Vx+Vy=S*B*I*sin(ωt+α)
Vdiff=Vx−Vy=S*B*I*sin(ωt−α)
It is therefore clear that, where the field maintains a steady orientation, that both Vsum and Vdiff are sinusoidal signals with period ω that have a relative phase angle of 2a, Vsum being shifted forward by α and Vdiff being shifted back by α. Accordingly, the field orientation angle a can be determined by first using phase comparator 103 to determine the phase difference between Vsum and Vdiff; and then having the determined phase difference.
Monitoring the orientation angle of rotor 115 in this manner can help to eliminate systematic bias that may occur if the magnitude of the bias current applied to one Hall element consistently exceeds that applied to the other or if the sensitivity of the Hall elements is not equal.
An alternative embodiment is shown in
Vsum=Vx+Vy=S*B*I*sin(ωt+α)
Vref=cos(ωt)=sin(ωt+90°)
If the relative phase angle determined by phase comparator 103 is Pref, then solving for α leaves
α=Pref+90°
In this embodiment, it is thus possible to calculate the field orientation angle a directly from Vsum without also calculating Vdiff. The skilled man will of course appreciate that Vref could be taken from the bias current input to Hall element X (in which case Vref=sin(ωt) and α=Pref). Additionally or alternatively, Vdiff could be used in place of Vsum with either bias current signal used as Vref. In such circumstances α=−Pref−90 for Vref=cos(ωt) and α=−Pref for Vref=sin(ωt).
The use of the method of
It is clear to the skilled man that the above techniques may be applied to sensors having Hall elements operable to measure magnetic field components along three mutually orthogonal axes x, y, z. In such circumstances, it is necessary to generate a Vsum and Vdiff for at least two of the planes x-y, y-z and z-x in order to determine the overall orientation of the field. It is also possible to apply the method to sensors wherein two or more Hall elements are provided to measure magnetic field components in each direction.
The method of the present invention may also be applied in a linear position sensor 100 of the form shown in
Close to the scale 120 the field perpendicular to the scale is more or less rectangular N-S-N-S, however, with distance the higher harmonics within rectangular shape signals decay faster than the fundamental wave, and the field shape approaches a sinusoidal signal. When the probe 121 is aligned with a pole, Vy is a maximum or minimum value (depending upon whether it is a north (N) or south (S) pole) and Vx is substantially zero. When the probe 121 is between a N and S pole, the Vy is substantially zero and Vx is a minimum or maximum value (depending on the orientation N-to-S or S-to-N). Consequently, assuming a suitable separation between the probe 121 and the scale 120, the outputs Vx, Vy of the Hall elements X, Y with respect to position comprise a pair of approximately sinusoidal signals with a mutual phase difference of 90°.
In order to determine the linear position of the probe 121 relative to the scale 120, as in the embodiments of
As is illustrated in
The method of the present invention may also be applied in a linear position sensor 100 of the form shown in
Close to the scale 120 the field perpendicular to the scale is more or less rectangular N-S-N-S, however, with distance the higher harmonics within rectangular shape signals decay faster than the fundamental wave, and the field shape approaches a sinusoidal signal. The Hall elements X, Y are orientated parallel to each other and at mutual separation d equal to w/4. Consequently, assuming a suitable separation between the probe 122 and the scale 120, the outputs Vx, Vy of the Hall elements X, Y with respect to probe 122 position comprise a pair of substantially sinusoidal signals with a mutual phase difference of 90°.
In order to determine the linear position of the probe 122 relative to the scale 120, as in the embodiments of
As is illustrated in
The method of the present invention may also be applied in a rotary position sensor 100 of the form shown in
Close to the scale 123 the field perpendicular to the scale is more or less rectangular N-S-N-S, however, with distance the higher harmonics within rectangular shape signals decay faster than the fundamental wave, and the field shape approaches a sinusoidal signal. When the probe 121 is aligned with a pole, Vy is a maximum or minimum value (depending upon whether it is a north (N) or south (S) pole) and Vx is substantially zero. When the probe 121 is between a N and S pole, the Vy is substantially zero and Vx is a minimum or maximum value (depending on the orientation N-to-S or S-to-N). Consequently, assuming a suitable separation between the probe 121 and the scale 123, the outputs Vx, Vy of the Hall elements X, Y with respect to position comprise a pair of approximately sinusoidal signals with a mutual phase difference of 90°.
In order to determine the angular position of the probe 121 relative to the scale 123, as in the embodiments of
As is illustrated in
The method of the present invention may also be applied in a rotary position sensor 100 of the form shown in
Close to the scale 123 the field perpendicular to the scale is more or less rectangular N-S-N-S, however, with distance the higher harmonics within rectangular shape signals decay faster than the fundamental wave, and the field shape approaches a sinusoidal signal. The Hall elements X, Y are orientated parallel to each other and at separation d equal to w/4. Consequently, assuming a suitable separation between probe 122 and scale 123, the outputs Vx, Vy of the Hall elements X, Y with respect to probe 122 position comprise a pair of substantially sinusoidal signals with a mutual phase difference of 90°.
In order to determine the angular position of the probe 122 relative to the scale 123, as in the embodiments of
As is illustrated in
The method of the present invention may also be applied in a rotary position sensor 100 of the form shown in
Close to the scale 123 the field perpendicular to the scale is more or less rectangular N-S-N-S, however, with distance the higher harmonics within rectangular shape signals decay faster than the fundamental wave, and the field shape approaches a sinusoidal signal. The Hall elements X, Y are orientated parallel to each other and at separation d equal to T/2. Consequently, assuming a suitable separation between probe 122 and scale 124, the outputs Vx, Vy of the Hall elements X, Y with respect to probe 122 position comprise a pair of substantially sinusoidal signals with a mutual phase difference of 90°.
In order to determine the angular position of the probe 122 relative to the scale 124, as in the embodiments of
As is illustrated in
The method of the present invention may also be applied in a sensor 200 of the form shown in
The track x has transmission characteristics such that the intensity of light Ax transmitted through track x as a function of angle α is Ax=A*cos(α), wherein A is the intensity of light incident upon the track x from light source 212. Additionally, the track y has transmission characteristics such that the intensity of light Ay transmitted through track y as a function of angle α is Ay=A*sin(α), wherein A is the intensity of light incident upon the track y from light source 212. The output voltages Vx, Vy of sensing elements X, Y are thus proportional to cos(α) and sin(α) respectively. If the sensing elements X, Y are biased using a sinusoidal signal, the method of the present invention can be used to determine the orientation angle a of the disc 210.
In a sensor 200, biasing can be achieved by directly biasing the output signal with a biasing signal of the form sin(ωt) for sensing element X and cos(ωt) for sensing element Y or by varying the intensity light of emitted by the light source 212 to be incident on each track according to a bias signal of the form sin(ωt) for track x and cos(ωt) for track y. As in the embodiments of
In a sensor 200, biasing can be achieved by directly biasing the output signal with a biasing signal of the form sin(ωt) for sensing element X and cos(ωt) for sensing element Y or by varying the intensity light of emitted by the light source 212 to be incident on each track according to a bias signal of the form sin(ωt) for track x and cos(ωt) for track y. As in the embodiments of
As in the embodiment of
In alternative variants of the sensor 200, the probe 221 may rotate whilst the disc 210 remains fixed. The encoder tracks x and y may have more than one period of sinusoidal variation in transmission or reflectance properties along their length. In such cases, by analogy with the embodiments of
Turning now to
The circuitry for carrying out this commutation is shown in
The amplifier 106 has an inherent DC offset, which can be considered as being added to the idealized output of amplifier 106 when providing the amplified output signal Vampx. In order to eliminate the Hall element X offset and the amplifier 106 offset, a band pass filter 107 is provided. The band pass filter 107 is operable to average the alternating Hall offset voltage. In order to achieve this, the upper cutoff frequency must be less than the frequency of chopping signal 105. Additionally, the band pass filter 107 is operable to eliminate the DC amplifier offset. In order to achieve this, the lower cutoff frequency must be less than the Hall bias modulation frequency ω.
If the band pass filter is applied correctly, the signal Voutx will an amplified signal from Hall element X, which does not contain any inherent offset due to either the Hall element X or the amplifier 106. The signal Voutx can then be used in the methods described in relation to
An alternative implementation of commutation to remove inherent offsets from a Hall element X output signal is illustrated by
The circuitry for carrying out this commutation is shown in
As in the previous embodiment, the amplifier 106 has an inherent DC offset, which can be considered as being added to the idealized output of amplifier 106 when providing the amplified output signal Vampx. In this case, both the Hall element offset and the amplifier offset are DC signals and accordingly, a demodulator 110 is provided. The demodulator 110 receives modulation input signal 109 as a reference signal for the demodulation operation.
If the demodulation of Vampx is applied successfully Vampx (sample1−sample2+sample3−sample4 etc) an output signal Voutx is obtained comprising the amplified Hall voltage without the inherent offset voltage from Hall element X or amplifier 106. The signal Voutx can then be used in the methods described in relation to
It is of course clear such commutation schemes outlined in relation to
It is of course to be understood that the invention is not to be restricted to the details of the above embodiment which is described by way of example only.
| Number | Date | Country | Kind |
|---|---|---|---|
| 0714663.2 | Jul 2007 | GB | national |
| 0716477.5 | Aug 2007 | GB | national |
