WO2008032008A1 - Position sensor - Google Patents

Position sensor Download PDF

Info

Publication number
WO2008032008A1
WO2008032008A1 PCT/GB2006/003423 GB2006003423W WO2008032008A1 WO 2008032008 A1 WO2008032008 A1 WO 2008032008A1 GB 2006003423 W GB2006003423 W GB 2006003423W WO 2008032008 A1 WO2008032008 A1 WO 2008032008A1
Authority
WO
WIPO (PCT)
Prior art keywords
excitation
sensor
winding
signal
film
Prior art date
Application number
PCT/GB2006/003423
Other languages
French (fr)
Inventor
Victor Evgenievich Zhitomirsky
Original Assignee
Sagentia Limited
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Sagentia Limited filed Critical Sagentia Limited
Priority to PCT/GB2006/003423 priority Critical patent/WO2008032008A1/en
Publication of WO2008032008A1 publication Critical patent/WO2008032008A1/en

Links

Classifications

    • GPHYSICS
    • G01MEASURING; TESTING
    • G01DMEASURING NOT SPECIALLY ADAPTED FOR A SPECIFIC VARIABLE; ARRANGEMENTS FOR MEASURING TWO OR MORE VARIABLES NOT COVERED IN A SINGLE OTHER SUBCLASS; TARIFF METERING APPARATUS; MEASURING OR TESTING NOT OTHERWISE PROVIDED FOR
    • G01D5/00Mechanical means for transferring the output of a sensing member; Means for converting the output of a sensing member to another variable where the form or nature of the sensing member does not constrain the means for converting; Transducers not specially adapted for a specific variable
    • G01D5/12Mechanical means for transferring the output of a sensing member; Means for converting the output of a sensing member to another variable where the form or nature of the sensing member does not constrain the means for converting; Transducers not specially adapted for a specific variable using electric or magnetic means
    • G01D5/14Mechanical means for transferring the output of a sensing member; Means for converting the output of a sensing member to another variable where the form or nature of the sensing member does not constrain the means for converting; Transducers not specially adapted for a specific variable using electric or magnetic means influencing the magnitude of a current or voltage
    • G01D5/20Mechanical means for transferring the output of a sensing member; Means for converting the output of a sensing member to another variable where the form or nature of the sensing member does not constrain the means for converting; Transducers not specially adapted for a specific variable using electric or magnetic means influencing the magnitude of a current or voltage by varying inductance, e.g. by a movable armature
    • G01D5/204Mechanical means for transferring the output of a sensing member; Means for converting the output of a sensing member to another variable where the form or nature of the sensing member does not constrain the means for converting; Transducers not specially adapted for a specific variable using electric or magnetic means influencing the magnitude of a current or voltage by varying inductance, e.g. by a movable armature by influencing the mutual induction between two or more coils

Definitions

  • the present invention relates to position sensors or encoders and to apparatus and circuits for use in such sensors or encoders.
  • the invention has particular but not exclusive relevance to non-contact rotary and linear position encoders.
  • Some embodiments of the invention are suitable for use in relatively small systems, which operate at relatively high temperatures and in which there may be magnetic and electromagnetic interference.
  • Inductive position encoders are well known in the art and typically comprise a movable member, whose position is related to the machine about which position or motion information is desired, and a stationary member which is inductively coupled to the moving member.
  • the stationary member typically includes a number of detectors which provide electrical output signals which can be processed to provide an indication of the position, direction, speed and/or acceleration of the movable member and hence for those of the related machine.
  • Some of these inductive position encoders employ an AC magnetic field generator mounted on the movable member and one or more sensor windings mounted on the stationary member.
  • the magnetic field generator and the sensor windings are arranged so that the magnetic coupling between them varies with the position of the movable member relative to the stationary member. As a result, an output signal is obtained from each sensor winding which continuously varies with the position of the movable member.
  • Hall Effect detectors Another type of magnetic position sensor that is known uses permanent magnets and Hall Effect detectors.
  • the problem with such Hall Effect detectors is that they are point magnetic field detectors which can only detect the presence or absence of the permanent magnet. Therefore, such Hall Effect systems do not provide a "continuous" output signal which varies with the position of the movable object.
  • inductive position sensor is a Flux Gate sensor which employs a film of soft magnetisable material.
  • an AC excitation magnetic field is applied along the plane of the soft magnetisable material causing the material to be driven into and out of magnetic saturation.
  • a DC magnet is usually provided on the movable member, the magnetic field of which interacts with the saturable magnetic element to cause some of the AC in-plane flux to be expelled from the material adjacent the magnet.
  • a number of sensor windings are provided adjacent the material for detecting the flux that is expelled from the saturable element (which depends on the position of the magnet), to output a signal which depends on the position of the magnet relative to the sensor windings.
  • Flux Gate sensors As a result of the driving of the saturable magnetic element into and out of saturation, the output signals that are generated in such Flux Gate sensors are at twice the frequency of the AC excitation frequency.
  • the applicant's earlier international application WO 97/14935 describes such a Flux Gate position sensor.
  • the problem with such Flux Gate sensors is that they are relatively bulky as they employ a relatively large ferromagnetic core around which the excitation and sensor windings are typically wound. They also tend not to be as rugged as smaller more integrated sensor technologies such as the Hall Effect systems discussed above.
  • the applicant has proposed an alternative type of inductive position encoder, which is described in WO 2005/085763.
  • This encoder is similar in structure to the Flux Gate encoder in that it uses a film of magnetisable material, a magnet and excitation and sensor windings.
  • title excitation circuit is arranged so as not to drive the film of magnetisable material into and out of saturation near the sensor winding.
  • the sensor winding is arranged to sense the change in mutual coupling between the excitation winding and the sensor winding, which varies with the position of an in-homogeneity in the film of magnetisable material caused by the magnetic field generated by the magnet.
  • the present application describes various improvements to the encoder and electronics described in WO '763. Some of these improvements are specific to the type of inductive encoder described in WO'763, whilst others are more generally applicable to other types of encoders.
  • the invention provides a position encoder comprising: a magnetic field generator for generating a magnetic field that varies with position along a measurement path; first and second sensor transducers each comprising: i) an excitation winding and a sensor winding; and ii) a film of magnetisable material located adjacent said excitation and sensor windings and located, in use, within said positionally varying magnetic field to cause the film to have a positionally varying magnetisation state along the measurement path; wherein the film of magnetisable material of said first transducer is positioned, in use, closer to said magnetic field generator than the film of magnetisable material of said second transducer; wherein the excitation winding and the sensor winding of each transducer are arranged relative to the corresponding film of magnetisable material so that a mutual electromagnetic coupling between them varies in dependence upon the positionally varying magnetisation state of the corresponding film of magnetisable material, so that when the excitation winding is energised with an excitation signal, a sensor signal is
  • the excitation windings of said transducers may be formed from a single common winding or two separate windings may be provided.
  • the sensor windings of said transducers may be provided as separate windings or they may be formed by a single winding that is provided in common to the two transducers.
  • connections to the excitation or the sensor windings are arranged so that the differencing circuitry adds the signals obtained from the sensor windings of the first and second transducers.
  • excitation circuitry that generates an excitation signal having an excitation frequency and the processing circuit processes the difference signal which is substantially at said excitation frequency, to determine said value indicative of the relative position between the first and second relatively movable members.
  • the excitation and the sensor windings are arranged relative to each other so that in the absence of the magnetic field generator, there is substantially no electromagnetic coupling between them.
  • Figure 1 comprising Figures IA, IB and 1C, illustrates the main components of an inductive position sensor used to sense the position of a piston within a cylinder;
  • Figure 2 schematically illustrates the layout of a set of windings that forms part of the sensor shown in Figure 1 ;
  • Figure 3 A is a plot illustrating the way in which the peak amplitude of a signal generated in a sin sensor winding (shown in Figure 3B) varies with the position of the piston within the cylinder;
  • Figure 4A is a plot illustrating the way in "which the, peak amplitude of a signal generated in a cos sensor winding (shown in Figure 4B) varies with the position of the piston within the cylinder;
  • Figure 5A is a plot illustrating the locus of points obtained by plotting the peak amplitude of the signal induced in the sin sensor winding against the peak amplitude of the signal induced in the cos sensor winding as the piston moves from one end of the cylinder to the other end;
  • Figure 5B is a phase plot illustrating the way in which a phase angle obtained from measured sin and cos signals varies with the position of the piston within the cylinder;
  • Figure 6 is a block diagram illustrating the main components of exemplary excitation and detection circuitry that forms part of the sensor shown in Figure 1 ;
  • Figure 7 is a block diagram illustrating the main components of a ratiometric calculator forming part of the detection circuitry shown in Figure 6;
  • Figure 8 is a block diagram illustrating the main components of a phase detector forming part of the ratiometric calculator shown in Figure 7;
  • Figures 9A and 9B illustrate the way in which a magnet positioned adjacent to the piston and cylinder assembly shown in Figure 1 can introduce an error into the phase measurements, which can lead to errors in the determined position of the piston within the cylinder:
  • Figure 10 schematically illustrates the way in which two sets of excitation and sensor windings can be used to overcome the problem associated with adjacent magnets
  • Figure 11 is a block diagram illustrating the main components of exemplary excitation and detection circuitry that can be used with the excitation and sensor windings shown in Figure 10;
  • Figure 12 comprising Figures 12A and 12B, illustrates the operation of an algorithm controller used to control the operation of multiplexing switches that form part of the detection circuitry shown in Figure 11;
  • Figure 13 is a schematic diagram illustrating an alternative arrangement of excitation and sensor windings that can be used to reduce the effect of adjacent magnets on the position sensor;
  • Figure 14 is a block diagram illustrating the main components of exemplary excitation and detection circuitry that can be used with the excitation and sensor windings shown in Figure 13;
  • Figure 15 which comprises Figures 15 A, 15B and 15 C, illustrates the operation of an algorithm controller that forms part of the detection circuitry shown in Figure 14;
  • Figure 16 which . comprises Figure 16A,..lj5B and 16C, illustrates an alternative sensor design which mitigates the problem associated with adjacent magnets which employs two sets of spaced excitation and sensor windings;
  • Figure 17 schematically illustrates the way in which the two sets of excitation and sensor windings of the sensor illustrated in Figure 16 maybe connected together;
  • Figure 18 schematically illustrates an alternative way in which the two sets of excitation and sensor windings illustrated in Figure 16 may be connected together;
  • Figure 19 is a block diagram illustrating an alternative form of excitation and detection circuitry that can be used with the sensors illustrated in Figures 2, 17 and 18;
  • Figure 20 is a block diagram illustrating an alternative form of excitation and detection circuitry that can be used with the sensors illustrated in Figures 2, 17 and 18;
  • Figure 21 is an electrical equivalent circuit illustrating an alternative way of connecting the windings of the sensors shown in Figures 2, 17 and 18 to the excitation and detection circuitry in which the windings previously connected to the excitation circuitry are connected to the detection circuitry and the windings previously connected to the detection circuitry are connected to the excitation circuitry;
  • Figure 22 is a block diagram illustrating excitation and detection circuitry that can be used to drive the windings in the inverted manner illustrated in Figure 21;
  • Figure 23 is a block diagram illustrating an alternative arrangement of the excitation and detection circuitry that can be used to drive the windings in the inverted manner illustrated in Figure 21 ;
  • Figure 24 is a block diagram illustrating the excitation and detection circuitry that can be used in a further alternative for driving the windings in the inverted manner illustrated in Figure 21;
  • Figure 25 is a block diagram illustrating the excitation and detection circuitry that can be used in a further alternative for driving the windings in the inverted manner illustrated in Figure 21;
  • Figure 26 is a block diagram illustrating the excitation and detection circuitry that can be used in a further alternative for driving the windings in the inverted manner illustrated in Figure 21;
  • Figure 27 is a block diagram illustrating the excitation and detection circuitry that can be used in a further alternative for driving the windings in the inverted manner illustrated in Figure 21;
  • Figure 28 is a block diagram illustrating the excitation and detection circuitry that can be used in a further alternative for driving the windings in the inverted manner illustrated in Figure 21;
  • Figure 29 illustrates pulse trains generated by the excitation circuitry shown in Figure 28;
  • Figure 30 is a block diagram illustrating an alternative form of excitation and detection circuitry that can be used with the sensors illustrated in Figures 2, 17 and 18;
  • Figure 31 is a signal diagram illustrating the signals generated by a PROM forming part of the circuit shown in Figure 30;
  • Figure 32 is a signal diagram illustrating in more detail the way in which the detection circuitry shown in Figure 30 operates to generate a signal whose phase varies with the position being measured;
  • Figure 33 is a circuit diagram of circuitry used to filter and amplify signals obtained from mixing switches forming part of the circuit shown in Figure 30;
  • Figure 34 is a block diagram illustrating an alternative form of excitation and detection circuitry that can be used with the sensors illustrated in Figures 2, 17 and 18;
  • Figure 35 is a signal diagram illustrating the signals generated by a PROM forming part of the circuit shown in Figure 34;
  • Figure 36 is a block diagram illustrating the excitation and detection circuitry that can be used in a further alternative for driving the windings in the inverted manner illustrated in Figure 21;
  • Figure 37 is a signal diagram illustrating the signals generated by a PROM forming part of the circuit shown in Figure 36.
  • Figure 38 is a block diagram illustrating the excitation and detection circuitry that can be used in a further alternative for driving the windings in the inverted manner illustrated in Figure 21.
  • Figures IA, IB and 1C schematically illustrate the form and operation of an inductive position sensor 1 and how it is used to sense the position of a piston within a cylinder. More details of the position sensor and the way in which it works can be found in the applicant's earlier PCT patent application WO 2005/085763, the contents of which are incorporated herein by reference. For ease of understanding the following sections, a brief description of the sensor illustrated in Figure 1 will now be given.
  • the piston and cylinder assembly comprises an elongate cylinder 2 having a central bore 3 in which a piston 4 is reciprocally mounted.
  • the piston 4 includes a piston head 5 whose outer diameter corresponds to the inner diameter of the cylinder bore 3.
  • the piston head 5 includes a DC magnet 6 oriented with its north and south poles being aligned horizontally in the Figure (i.e. in the Z direction).
  • the cylinder 2 is encased within a metal casing 7, part of which is cut away to show the cylinder 2 and the piston 4.
  • a printed circuit board (PCB) 8 is attached to the left hand side (in the Figure) of the cylinder casing 7 and carries a number of conductive tracks 9 (shown in Figure IA) which are connected together to define a number of excitation and sensor windings (not shown) which are connected to excitation and detection electronics 10 via a connection interface 20.
  • a film 11 of magnetisable material is provided on the outer surface (i.e. furthest away from the cylinder 2) of the PCB 8, although it may be provided on the inner surface of the PCB 8 instead of or in addition to being on the outer surface.
  • the film 11 of magnetisable material is Niso%Fe2o%, which is a soft ferromagnetic material having a high initial and maximum permeability (100,000 to 400,000), a low coercivity and a relatively low saturation field (lA/m) and polarisation (0.7T to 0.9T).
  • the film 11 has a thickness in the Z direction of less than 0.2mm and has approximately the same width and length as the PCB 8.
  • the excitation and sensor windings carried by the PCB 8 are geometrically arranged on the PCB 8 so that, in the absence of the piston 4, there is substantially no electromagnetic (inductive) coupling between them.
  • substantially no signal is induced in the (or each) sensor winding.
  • the film 11 of magnetisable material is substantially homogenous so that, in the absence of the magnet 6, its presence adjacent to the excitation and sensor windings does not affect the balance between the windings.
  • the magnet 6 within the piston head 5 creates a positionally varying DC magnetic field, which penetrates through the metallic cylinder casing 7 and the PCB 8 and interacts with the portion of the film 11 that is next to the magnet 6.
  • the interaction of the DC magnetic field with this portion of the film 11 creates an imbalance between the excitation and sensor windings such that when the AC excitation current is applied to the excitation winding, AC signals are generated in the (or each) sensor winding which vary with the position of the piston 4 within the cylinder 2. Therefore, by suitable processing of the signals generated in the (or each) sensor winding, the electronics 10 determines and outputs (e.g. on a display) a value indicative of the position of the piston 4 within the cylinder 2.
  • Figure IA schematically shows the field lines of the DC magnetic field generated by the magnet 6.
  • This magnetic field has two components, one which is directed perpendicular to the film 11 of magnetisable material and the other which is directed "in the plane" of the film 11 of magnetisable material.
  • the component that is perpendicular to the film 11 of magnetisable material does not interact with the film 11.
  • the other "in plane” component does interact with the film 11 and, depending on its magnitude, can magnetically saturate the part of the film 11 located adjacent the magnet 6. Therefore, at the spot 12 where the DC magnetic field generated by the DC magnet 6 is perpendicular to the film 11, no saturation of the film 11 occurs.
  • this spot 12 is located where the magnetic axis of the DC magnet 6 intersects with the film 11 of magnetisable material.
  • the magnetic field generated by the DC magnet includes in-plane components that saturate the magnetisable material within the portion 16 immediately adjacent the magnet 6. The size of the portion 16 depends on the distance between the magnet 6 and the film 11 of magnetisable material and the strength of the DC magnetic field.
  • Figure IB is a plot showing the way in which the strength of the in-plane component (H) of the DC magnetic field varies along the X direction through the centre of the spot 12 (which corresponds to the origin of the plot).
  • the in-plane component (H) is zero in the centre of the spot 12 and is negative on the left hand side of the origin and is positive on the right hand side of the origin.
  • the horizontal dashed lines shown in the plot correspond to the strength of the magnetic field required to saturate the film 11 of magnetisable material.
  • the in-plane component (H) of the magnetic field is strong enough to saturate the film 11 of magnetisable material.
  • Figure IB shows the way in which the magnetic flux density (B) within the film 11 of magnetisable material varies along the X direction taken through the centre of the spot 12.
  • the magnetic flux density within the film 11 is constant where the film 11 is saturated.
  • the in-plane component (H) of the magnetic field is not strong enough to saturate the film 11.
  • the third plot shown in Figure IB illustrates the way in which the AC relative permeability ( ⁇ ) of the film 11 varies along the X direction through the centre of the spot 12.
  • the relative permeability is approximately one where the film 11 has become MIy saturated.
  • the film 11 of magnetisable material is not saturated and because of the large rate of change of the in-plane magnetic field (H), the relative permeability within the spot 12 increases rapidly.
  • the maximum permeability within the spot 12 depends on the material properties of the film 11. With the magnetisable material used in this system, the maximum relative permeability is approximately 5000.
  • the spot 12 and the surrounding region of saturated film 11 will sometimes be referred to hereinafter as the in-homogeneity spot 12.
  • FIG 2 schematically illustrates the conductor tracks on the sensor PCB 8 that form the excitation winding 13, a "sin” sensor winding 14 and a "cos” sensor winding 15, all of which extend along the length of the PCB 8.
  • the PCB 8 is a two layer PCB, with the conductor tracks on the upper layer (closest to the film 11) being shown in solid lines and with the conductor tracks on the lower layer (closest to the piston 4) being shown in dashed lines.
  • the sensor windings 14 and 15 have a pitch (L x ) that approximately corresponds to the range over which the piston 4 can move within the cylinder 2.
  • the ends of the excitation winding 13 and the ends of the sensor windings 14 and 15 are connected to the excitation and processing electronics 10 (which are also mounted on the PCB 8) via the connection interface 20.
  • the sin sensor winding 14 is formed from two turns of conductor that effectively define two sets of loops which are connected together in the opposite sense in a figure of eight configuration.
  • EMFs induced in the first set of loops by a common background magnetic field will oppose the EMFs induced in the second set of loops by the same common background magnetic field.
  • the cos sensor winding 15 is effectively formed by shifting the sin sensor winding 14 by a quarter of the pitch (L x ) along the X direction.
  • the cos sensor winding 15 effectively defines three sets of loops, with the loops of the first and third sets being wound in the same direction as each other but opposite to the winding direction of the loops of the second (middle) set.
  • the sizes of the loops forming the cos sensor winding 15 are chosen so that (in the absence of the magnet 6) the EMFs induced in the first and third sets of loops by the excitation magnetic field cancel with the EMFs induced in the second set of loops by the excitation magnetic field.
  • the excitation winding 13 is wound around the outside of the sensor windings 14 and 15 and is arranged so that (in the absence of the magnet 6) when an AC excitation current is applied across its ends, an AC excitation magnetic field is generated which extends predominantly in the Z direction shown in Figure 1.
  • This excitation magnetic field will be substantially symmetric along an axis which is parallel to the Y axis and which passes through the middle of the excitation winding.
  • This axis of symmetry is also an axis of symmetry for the sensor windings 14 and 15.
  • the signal generated in each sensor winding 14 and 15 will vary with the position of the piston 4 within the cylinder 2.
  • the variation of the signal generated in each sensor winding 14 and 15 is approximately sinusoidal.
  • the signal generated in each sensor winding 14 and 15 is an AC signal at the same frequency as the excitation signal, whose peak amplitude varies approximately sinusoidally with the position of the piston 4 within the cylinder 2.
  • the main signals generated in each sensor winding are at the same frequency as the excitation frequency because, the direction and magnitude of the AC excitation magnetic field (generated by the AC current flowing in the excitation winding 31) is such that it does not saturate a substantial area of the film 11 of magnetisable material in the vicinity of the sensor windings. If the excitation current is strong enough to enforce changing magnetisation of substantial areas of the film 11 at the excitation frequency, then it causes two unwanted effects. First, the measured signals will vary with the orientation of the magnet 6, such that reversing the magnetisation of the magnet 6 will change the readings of the sensor due to the inherent substantial magnetisation caused by the excitation currents.
  • FIG 3A schematically illustrates the way in which the peak amplitude of the signal generated *in the r sin' sensor winding 14 (shown in Figure 3B) varies with the position of the piston 4 within the cylinder 2.
  • the variation in the peak amplitude is approximately sinusoidal (except at the ends of the winding) and hence the sensor winding 14 is referred to as the "sin" sensor winding 14.
  • the period of the sinusoidal variation corresponds approximately to the pitch (L x ) of the sensor winding 14.
  • Figure 4A schematically illustrates the way in which the peak amplitude of the signal induced in sensor winding 15 (shown in Figure 4B) varies with the position of the piston 4 within the cylinder 2.
  • the variation is in phase quadrature (90° out of phase) to the variation obtained from sensor winding 14, which is why the sensor winding 15 is referred to as the "cos" sensor winding.
  • the quadrature nature of the variation between the signals output from the sensor windings 14 and 15 is obtained because sensor winding 15 is effectively shifted along the X direction by a quarter of the pitch L x relative to sensor winding 14.
  • plots shown in Figures 3A and 4A are approximate in that the peak amplitudes of the output signals do not vary exactly sinusoidally with the position of the piston 4 within the cylinder 2. This is an approximation to the actual variation, which will depend upon edge effects, positions of via holes on the PCB 8 and other effects that introduce non-linearities into the system.
  • two sensor windings 14 and 15 are provided in 5 order to be able to uniquely encode the position of the piston 4 within the cylinder 2 along the entire pitch (L x ) of the sensor windings 14 and 15.
  • the measurement range of the piston 4 within the cylinder 2 is limited, for example to between the thirty percent and the seventy percent points on the plot shown in Figure 3 A, then only one sensor winding (winding 14) would be required to determine the absolute
  • the piston 4 when the piston 4 can move over the entire pitch of the sensor windings 14 and 15, at least two sensor windings are required in order to overcome the phase ambiguity common with sinusoidal signals. For example, if the peak amplitude of the signal generated in sensor winding 14 is determined to be 0.5, then, as shown in Figure 3 A, the position may correspond to approximately ten
  • Figure 5 A illustrates the locus 27 obtained by plotting the peak amplitude of the EMF
  • the plot 27 is substantially circular except for when the piston 4 is at each end of the cylinder 2. Therefore, as shown by plot 28 in Figure 5B, except in these end regions, the phase angle ⁇ varies linearly with the position of the piston 4 in the
  • the excitation and detection electronics 10 determines the position of the piston 4 within the cylinder 2 by taking an inverse arctangent function of the ratio of the peak amplitudes of the signals induced in sensor windings 14 and 15. The use of such a ratiometric calculation is preferred as it makes the system less sensitive to variations in the amplitude of the
  • Figure 2 schematically illustrates the way in which the excitation winding 13 and the 35 sensor windings 14 and 15 are connected to the excitation and detection circuitry 10.
  • the excitation winding 13 is connected to the excitation circuitry via terminals A and B of the connection interface 20 and the sensor windings 14 and 15 are connected to the detection circuitry at terminals 1, 2, 3 and 4 of the connection interface 20. 40
  • FIG. 6 is a block diagram illustrating one form of the excitation circuitry 24 and the detection circuitry 26 that can be used.
  • the excitation circuitry 24 includes a signal generator 30 which is operable to generate an AC drive voltage 31 which is applied to terminal A of the connection interface 20.
  • the excitation circuitry also 45 includes an inverter 32 which inverts the AC voltage 31 to generate an inverted AC voltage 33, which is applied to terminal B of the connection interface 20.
  • symmetrical sinusoidal excitation signals are applied to both ends of the excitation winding 13.
  • the AC voltages 31 and 33 can have an amplitude in the range 0.1 V to 10V and a frequency between 10OkHz and 10MHz.
  • the inventor has found that by driving both ends of the excitation winding 13 with symmetrical excitation voltages 31 and 33, the sensor is able to operate at much higher frequencies.
  • the inventor has found that driving only one end of the excitation winding 13 with an excitation voltage results in a loss of performance at about 500kHz. This is because, at higher frequencies and without symmetrical drive voltages, current flowing in the excitation winding 13 capacitively couples into the film 11 of magnetisable material, which changes the current distribution in the sensor windings, which in turn affects the operation of the sensor.
  • the sensor is able to operate at frequencies up to approximately 10MHz.
  • the advantage of operating at higher frequencies is that greater signal levels can be obtained from the sensor windings than when operating at lower frequencies.
  • the signals obtained at terminals 1 and 2 of the connection interface 20 are applied to a first differential amplifier 34 " and the signals obtained from terminals 3 and 4 of the connection interface 20 (which are connected to the sin sensor winding 14) are applied to a second differential amplifier 34 ⁇ 2 .
  • the differential amplifiers amplify the signals obtained from the sensor windings and remove any common mode signals that might arise.
  • the signals induced in the sensor windings 14 and 15 are AC signals at the same frequency as the excitation signal and having a peak amplitude that varies sinusoidally and cosinusoidally with the position of the piston 4 within the cylinder 2. Therefore, the signals output by the differential amplifiers 34 " and 34 " can be represented (approximated) by the following equations:
  • A is an unknown amplitude term that depends on the sensor design
  • d is the position of the in-homogeneity spot 12 along the length (L x ) of the sensor board (corresponding to the position of the piston 4 in the cylinder 2); and
  • f is the excitation frequency of the AC signals 31 and 33.
  • the signals output from the differential amplifiers 34 are applied to a respective mixer 36 “1 and 36 “2 where they are mixed with the excitation signal 31.
  • the outputs from the mixers 36 are then input to a respective low pass filter 38 "1 and 38 “2 which operate to remove the high frequency products of the mixing and to leave the following DC values which vary with the position of the in-homogeneity spot 12:
  • the signals output from the low pass filters 38 are applied to a ratiometric calculator 40 which determines the position (d) of the in-homogeneity spot 12 using the following arctangent function:
  • the ratiometric calculator 40 uses analogue processing techniques to generate a value that continuously varies with the value of d.
  • Figure 7 is a block diagram illustrating the way in which the ratiometric calculator 40 performs the calculation.
  • the ratiometric calculator 40 includes a low frequency generator 48 that generates a low frequency (of the order of a few kHz) periodic signal. This low frequency periodic signal is applied directly to a first amplitude modulator 50 "1 and a 90 degree phase shifted version of the low frequency periodic signal is applied to a second amplitude modulator 50 "2 .
  • the 90 degree phase shift is obtained by passing the low frequency periodic signal though a 90 degree phase shifter 52.
  • the amplitude modulator 50 " amplitude modulates the received low frequency periodic signal with the voltage (Vcos) obtained from the low pass filter 38 "1 ; and the amplitude modulator 50 "2 amplitude modulates the phase shifted low frequency periodic signal with the voltage (V s i n ) obtained from the low pass filter 38 ⁇ .
  • the two amplitude modulated signals output from the amplitude modulators 50 are then added together by an adder 54.
  • the output from the adder 54 can therefore be represented (approximated) by:
  • f 2 is the frequency of the low frequency periodic signal
  • the output from the adder 54 includes a sinusoidal signal at the low frequency (f 2 ) whose phase is proportional to the ratiometric arctangent function defined in equation 5 given above. Therefore, the ratiometric calculator 40 calculates equation 5 by using a phase detector 56 to detect the phase of this signal output from the adder 54 relative to the low frequency signal generated by the low frequency generator 48. The output of the phase detector 56 will therefore be a value that continuously changes with the position of the in-homogeneity spot 12 relative to the PCB
  • Figure 8 is a block diagram illustrating in more detail the components of the phase detector 56 used in this design.
  • the AC signal 55 from the adder 54 is input to a first comparator 57 "1 where it is compared with a threshold voltage.
  • the AC signal 58 from the low frequency generator 48 is input to a second comparator 57 "2 where it is also compared with the same threshold value.
  • the square wave signals 59 "1 and 59 ⁇ 2 output by the comparators 57 are shifted in time relative to each other by a value ⁇ t corresponding to the phase shift between AC signal 55 and AC signal 58.
  • a logic circuit 63 which generates a pulse width modulated sequence of pulses 65 whose pulse width corresponds to the timing difference ⁇ t.
  • this pulse width modulated signal 65 is input to a PWM/DC converter 67 which converts this pulse train into a DC value which monotonically varies with the value of ⁇ t and hence with the phase shift between the AC signal 55 and the AC signal 58.
  • the output from the phase detector 56 is input to a level signal display 44, which uses the DC value output by the phase detector 56 to generate a display corresponding to the position of the piston 4 within the cylinder 2.
  • Figure 9 which comprises Figures 9A and 9B illustrate this problem.
  • Figure 9A shows the piston and cylinder assembly of Figure 1 C and a second magnet 73 which can also move relative to the sensor PCB 8. If the second magnet 73 is close enough to the film 11 of magnetic material on the sensor PCB 8, then the magnetic field generated by the second magnet ' 73 will also create an in-homogeneity spot in the film 11 which will disturb the balance between the excitation and sensor windings carried by the sensor PCB 8. As the second magnet 73 is likely to be much further away from the film 11 of magnetisable material than the magnet 6 (typically upto 10 times further away), the disturbance caused by the magnet 73 will be significantly smaller than the disturbance caused by the magnet 6. This is illustrated in the locus plot 27 shown in Figure 9B which is similar to the plot shown in Figure 5A, but which includes a disturbance represented by the curve 75 caused by the magnet 73 which can introduce an error ( ⁇ ) into the measured phase angle.
  • error
  • Figure 10 schematically illustrates the form of the excitation and sensor windings used in a second design of the sensor board 8 which is designed to mitigate the problem of interfering magnetic fields discussed above.
  • excitation and sensor windings are provided in the regions corresponding to each end of the cylinder 2.
  • excitation winding 13 ⁇ 2 and sensor windings 14 "2 and 15 “2 are positioned at one end of the cylinder 2 and a second excitation winding 13 "3 and a second pair of sensor windings 14 "3 and 15 “3 are provided at the other end of the cylinder 2.
  • the two excitation windings 13 "2 and 13 "3 can be connected in series so that no modifications have to be made to the excitation circuitry 24.
  • the two sets of sensor windings 14 and 15 are connected to respective detection circuitry 26 "2 and 26 "3 .
  • These detection circuits 26 may operate independently and signal an appropriate controller (not shown) when they detect the presence (and position) of the piston 4.
  • the signals from these two sets of sensor windings 14 and 15 may be multiplexed through a common detection channel, as illustrated in Figure 11.
  • the excitation circuitry 24 is the same as the excitation circuitry shown in Figure 6 and the detection circuitry 26 is similar to that shown in Figure 6, except that multiplexing switches 76 "1 and 76 " are provided which are arranged either to pass the signals from sensor windings 14 ⁇ 2 and 15 ⁇ 2 to the differential amplifiers 34 or to pass the signals obtained from sensor windings 14 ⁇ 3 and 15 ⁇ 3 to the differential amplifiers 34.
  • an algorithm controller 78 is used to control the switching of the multiplexing switches 76 and, based on the measurements obtained from the two sets of sensor windings 14 and 15, to determine the approximate position of the piston 4 within the cylinder 2 for output on the position signal display 44.
  • the remaining components of the detection circuitry 26 are the same as used in the circuit shown in Figure 6 and will not, therefore, be described again.
  • Figure 12A illustrates the way in which the phase angle ( ⁇ ) calculated by the ratiometric calculator 40 varies with the position of the piston 4 in the cylinder 2, when the signals from the two sets of sensor windings 14 and 15 are processed.
  • plot 28 ⁇ 2 illustrates the way in which the calculated phase angle varies with the position of the piston 4 based on the measurements obtained from sensor windings 14 " and 15 " ;
  • plot 28 " illustrates the way in which the calculated phase angle varies with the position of the piston 4 based on the measurements obtained from sensor windings 14 "3 and 15 “3 .
  • Figure 12B is a flow chart illustrating the algorithm performed by the algorithm controller 78 to calculate the position of the piston 4.
  • the algorithm controller 78 controls the multiplexing switches 76 so that the signals from sensor windings 14 " and 15 " are passed through to the differential amplifiers 34.
  • the algorithm controller 78 determines a first position measurement (x) using the phase angle obtained from the ratiometric calculator 40 in response to the processing of the signals obtained from sensor windings 14 "2 and 15 "2 .
  • the algorithm controller 78 determines if the thus calculated position (x) is less than L 2 . If it is, then the algorithm controller 78 determines the output position measurement (x ) as the thus determined position (x).
  • step s6 the algorithm controller 78 determines that the first position measurement (x) is not less than L 2 then the processing proceedds to step s8, where the algorithm controller 78 controls the multiplexing switches 76 so that the signals from sensor windings 14 " and 15 " are passed through to the differential amplifiers 34. Then, in step slO, the algorithm controller 78 determines a second position measurement (x ) using the phase angle obtained from the ratiometric calculator 40 in response to the processing of the signals obtained from sensor windings 14 "3 and 15 "3 . In step sl2, the algorithm controller 78 determines if the thus calculated position (x ) is greater than zero (in which case the piston 4 is within the operating range L 3 ).
  • the algorithm controller 78 determines the output position measurement (x ) as being the second position measurement (x ) plus (Li-L 3 ). Otherwise the piston 4 is somewhere between the two ends and outputs the position (x ) as being halfway along the length of the cylinder 2.
  • FIG. 13 schematically illustrates the windings of a third design of the sensor PCB 8 which is formed using a combination of the windings used in the first design and those used in the second design.
  • a single excitation winding 13 is provided together with three sets of sin and cos sensor windings (14 “1 and 15 “1 ; 14 “2 and 15 ⁇ 2 ; 14 ⁇ 3 and 15 “3 ).
  • the three sets of sin and c nsor windings 14 and 15 may each be connected to respective detection circuitry 26 “2 and 26 "3 .
  • the signals from the three sets of sensor windings 14 and 15 may be multiplexed into a common detection channel using multiplexing switches 76 "1 , 76 " and 76 " .
  • an algorithm controller 78 is provided to control the switching of the multiplexing switches 76 and to determine the position of the piston 4 based on the measurements obtained from the ratiometric calculator 40, which position information it outputs for display on the position signal display 44.
  • the algorithm controller 78 determines the actual position of the piston 4 using a weighted combination of the positions calculated from these two sets of sensor measurements. The way in which the algorithm controller 78 performs this weighted combination will now be explained with reference to Figure 15.
  • Figure 15A illustrates the way in which the phase angle ( ⁇ ) calculated by the ratiometric calculator 40 varies with the position of the piston 4 in the cylinder 2, when the signals from the three sets of sensor windings 14 and 15 are processed.
  • plot 28 "1 illustrates the way in which the calculated phase angle varies with the position of the piston 4 based on the measurements obtained from sensor windings 14 " and 15 "1 ;
  • plot 28 "2 illustrates the way in which the calculated phase angle varies with the position of the piston 4 based on the measurements obtained from sensor windings 14 "2 and 15 “2 ;
  • plot 28 "3 illustrates the way in which the calculated phase angle varies with the position of the piston 4 based on the measurements obtained from sensor windings 14 "3 and 15 “3 .
  • the algorithm controller 78 when the piston 4 is within the operating range L 2 , the algorithm controller 78 performs a weighted combination of the measurements obtained from plots 28 " and 28 " . Similarly, when the piston 4 is within the measurement range L 3 , the algorithm controller 78 calculates a weighted combination of the measurements obtained from plots 28 "1 and 28 " .
  • the particular weighting functions used in this embodiment are illustrated graphically in Figure 15B.
  • Figurel5C is a flow chart illustrating the algorithm performed by the algorithm controller 78 to calculate the position of the piston 4.
  • the algorithm controller 78 controls the multiplexing switches 76 so that the signals from sensor windings 14 "1 and 15 "1 are passed through to the differential amplifiers 34.
  • the algorithm controller 78 determines a first position measurement (x) using the phase angle obtained from the ratiometric calculator 40 in response to the processing of the signals obtained from sensor windings 14 "1 and 15 "1 .
  • the algorithm controller 78 determines if the thus calculated position (x) is less than L2 plus a threshold ( ⁇ ).
  • the algorithm controller 78 controls the multiplexing switches 76 so that the signals obtained from sensor windings 14 "2 and 15 "2 are passed through to the differential amplifiers 34.
  • the algorithm controller 78 determines a new position measurement (x ) using the phase measurements obtained by the ratiometric calculator 40 using the signals from sensor windings 14 "2 and 15 "2 .
  • the algorithm controller 78 subtracts the measurement range L 2 from the thus obtained position measurements so that they can be applied to the weighting function illustrated in Figure "15B.
  • the algorithm controller 78 determines the position of the piston (x ) using the following weighted combination:
  • step s5 the algorithm controller 78 determines that the determined position (x) is greater than L 2 plus the threshold ( ⁇ ) then the processing proceeds to step si 5 where the algorithm controller 78 determines if the position (x) is greater than (Li-L 3 )- ⁇ . If it is, then the piston 4 may be within the operating range L 3 and therefore, the processing proceeds to steps sl7, sl9 and s21 where a similar processing to steps s7, s9 and sll is performed but in respect of measurements obtained from sensor windings 14 "3 and 15 ⁇ 3 . The algorithm controller 78 then calculates in step sl3 the weighted combination of the determined measurements as before.
  • step sl5 the algorithm controller 78 determines that the determined position (x) is not greater than (Li -L 3 )- ⁇ , then the piston 4 is somewhere in the middle of the cylinder 2 and therefore the algorithm controller 78 uses the determined position (x) as the output 5 position (x ** ) which it outputs to the position signal display 44.
  • Figure 16 (which comprises Figures 16A, 16B and 16C) illustrates an alternative sensor design that mitigates the problem associated with interfering magnetic fields from0 adjacent magnets.
  • two sensor boards (transducers) 8 "1 and 8 “2 are provided which are spaced apart from each other relative to the piston 4 using a spacer block 80.
  • the film of magnetisable material carried by sensor board 8 " will be between zero and 20mm from the magnet 6 and the film of magnetisable material carried by sensor board 8 " will be between 2mm5 and 25mm from the magnet 6.
  • the excitation and sensor windings (not shown) carried by the two sensor boards 8 “ and 8 " are arranged to generate signals which both vary in the same way with the position of the piston 4 within the cylinder 2. Since the sensor board 8 "1 is located further away from the magnet 6 than the sensor board 8 “2 , the signals obtained from sensor board 8 " will be smaller in magnitude than the signals obtained from sensor board 8 "2 . However, as the interfering magnet 73 is located at a much greater distance from the sensor boards 8 (than the magnet 6 in the piston 4), it will have substantially the same interfering effect on each of the sensor boards 8.
  • Figure 16B illustrates the locus 27 ⁇ 2 of points obtained by plotting the measured values V S m and V CO s obtained from the sensor windings mounted on sensor board 8 "2 and similarly the locus 27 "1 of points obtained from the sensor windings
  • the locus of points 27 "2 has a larger diameter ⁇ than the locus of points 27 "1 because the signals obtained from the sensor windings mounted on the sensor board 8 " will be much larger than the signals obtained from the sensor windings on sensor board 8 " .
  • Figure 16B also illustrates by the curves 75 " and 75 " the effect that the interfering magnet 73 can have on the signals obtained from the respective sensor boards.
  • Figure 16C illustrates the locus of points 27 " obtained by subtracting the signals obtained from the two sensor boards 8 "1 and 8 "2 .
  • the electronics 10 used in the sensor design shown in Figure 16 may be arranged to perform the subtraction of the signals obtained from the two sensor boards. However, performing the subtraction in the electronics
  • the excitation and sensor windings mounted on the two sensor boards 8 "1 and 8 “2 are preferably connected together in such a way that the signals from the two sin sensor windings 14 are subtracted from each other and the signals from the two cos sensor windings 15 are subtracted from each other.
  • Figure 17 illustrated one way in which this can be achieved.
  • Figure 17 schematically illustrates the excitation winding 13 "1 and the sensor windings 14 "1 and 15 “1 mounted on the sensor board 8 "1 and the excitation winding 13 "2 and the sensor windings 14 "2 and 15 '2 mounted on the sensor board 8 "2 .
  • the excitation and sensor windings mounted on the two sensor boards 8 "1 and 8 “2 are connected together on the spacer block 80.
  • sensor winding 14 "1 is connected in series with sensor winding 14 ⁇ 2 so that the signals induced in the two sensor windings are added together.
  • sensor winding 15 "1 is connected in series with sensor winding 15 " so that the signals induced in the two sensor windings add together.
  • the excitation windings 13 "1 and 13 ⁇ 2 are connected in series so that the direction of current flow in the two excitation windings 13 is in the opposite direction.
  • the AC magnetic field generated by the excitation winding 13 "1 will be of opposite polarity to the AC excitation magnetic field generated by the excitation winding 13 " .
  • the signals induced in the sensor windings 14 "1 and 15 “1 will be of opposite polarity to the signals induced in sensor windings 14 "2 and 15 “2 . Consequently, adding the signals induced in sensor windings 15 “1 and 15 “ and adding the signals induced in sensor windings 14 " and 14 “2 has the desired result of subtracting (or differencing) the signals from these two pairs of sensor windings.
  • Figure 18 illustrates an alternative way in which the same result can be achieved.
  • the excitation windings 13 “1 and 13 “2 are connected in series so that the excitation magnetic fields generated by these windings will have the same polarity, but the respective pairs of sensor windings (14 “1 and 14 “2 ; and 15 “1 and 15 “ ) are connected together so that the current flow is in the opposite direction from each other.
  • the signals induced in sensor winding 14 " will subtract from the signals induced in sensor winding 14 "2 and the signals induced in sensor winding 15 "1 will subtract from the signals induced in sensor winding 15 " .
  • excitation and detection circuitry illustrated in Figure 6 can be used to energise the excitation windings 13 and to process the signals obtained from the sensor windings 14 and 15 of the sensor designs illustrated in Figure 17 and 18.
  • excitation and detection circuits that can be used with the first and fourth sensor designs described above.
  • excitation and detection circuits could also be modified so that they can operate with the second and third sensor designs described above.
  • modification may simply involve the addition of suitable multiplexing circuitry between the detection circuit 26 and the connection interface 20 and the provision of a suitable controller for controlling the multiplexing of the signals into the detection circuit.
  • FIG 19 is a block diagram illustrating alternative excitation and detection circuitry that can be used. Components that are the same as those used in the circuitry shown in Figure 6 have been given the same reference numerals.
  • the AC signals output from the differential amplifiers 34 are demodulated by mixers 36 and then remodulated onto a low frequency carrier in the ratiometric calculator 40.
  • this demodulation arid then remodulation is omitted.
  • the AC signal output from the differential amplifier 34 "2 is applied to a 90 degree phase shifter 52.
  • the phase shifted signal is then added with the AC signal output from the differential amplifier 34 '1 in the adder 54.
  • the output from the adder 54 can therefore be represented by equation 7 above, except replacing the low frequency value ⁇ % with the frequency of the excitation signals 31 and 33. Consequently, the arctangent function defined above in equation 5 can be evaluated by measuring the phase of the signal output from the adder 54 relative to the phase of the excitation signal 31. As shown in Figure 19, the phase detector 56 makes this phase measurement.
  • the remaining components of the excitation and detection circuitry shown in Figure 19 are the same as those shown in Figure 6 and will not, therefore, be described again.
  • circuitry shown in Figure 19 is a simplified version of the circuitry shown in Figure 6, the circuitry shown in Figure 19 is not actually preferred as it requires the phase shifter 52, the adder 54 and the phase detector 56 to operate at the higher excitation frequency rather than at the lower frequency of the low frequency generator 48. As a result, more complex and more expensive circuitry has to be used for these components.
  • Figure 20 is a block diagram illustrating alternative excitation circuitry 24 and detection circuitry 26 that can be used.
  • the sensor can operate over a range of frequencies.
  • a pulse generator 60 that generates square form pulses of excitation signal (having a pulse width of the order of several hundred nanoseconds) with sharp rising and falling fronts (having widths of the order of several nanoseconds).
  • Fast fronts of the excitation pulses correspond to the equivalent frequency of tens of MHz.
  • the pulse repetition frequency that is used is less important as it only affectsrihe overall signal levels that will be processed by the detection circuitry 26. Therefore, in order to minimise power consumption of the sensor, the pulse repetition frequency should be set to the minimum necessary to obtain the required level of signal to noise ratio for the measurements to be made.
  • the pulsed excitation signal 31 generated by the pulse generator 60 is applied to terminal A of the connection interface 20 and is applied to an inverter 32 for generating an inverted version 33 of the pulsed excitation signal which is applied to terminal B of the connection interface 20.
  • terminals 1 and 2 of the connection interface 20 are connected to a reference capacitor 64 "1 through switches 62 “1 and 62 “2 , which are opened and closed in synchronism with the pulses of the excitation signal 31 and connected so that the signal induced in the sensor winding connected to terminals 1 and 2 is alternately connected to opposite plates of the reference capacitor 64 " .
  • terminals 3 and 4 of the connection interface 20 are connected to a second reference capacitor 64 "2 through switches 62 "3 and 62 , which are also switched in synchronism with the pulses of the excitation signal 31.
  • voltages are accumulated on the reference capacitors 64 which will vary slowly over time reflecting the movement of the in-homogeneity spot 12 along the length of the sensor PCB 8.
  • the switches 62 therefore act to demodulate the signals received from the sensor windings so that the above described DC voltages (V COs and V S in) accumulate on the reference capacitors 64.
  • the voltages accumulated on the reference capacitors 64 are amplified by differential amplifiers 66 ⁇ and 66 ⁇ 2 respectively and then applied to the ratiometric calculator 40 as before for processing in the same way.
  • the reference capacitors 64 and the amplifiers 66 together form respective low pass filters 68 "1 and 68 '2 which remove any high frequency components caused by the switching of the switches 62.
  • the detection circuitry 26 can use relatively cheap low frequency electronic components to amplify and process the signals obtained from the sensor windings. Excitation And Detection Circuiti ⁇ - 4
  • the excitation signals were applied to the two ends of the excitation winding 13 shown in Figure 2 and the signals obtained from the sin and cos sensor windings 14 and 15 were connected to the detection circuitry 26.
  • the roles of these windings can be reversed so that the sin and cos windings 14 and 15 are connected to the excitation circuitry 24 and the winding 13 is connected to the detection circuitry 26. This is illustrated schematically in the electrical equivalent circuit shown in Figure 21.
  • FIG 22 is a block diagram illustrating the excitation circuitry 24 and the detection circuitry 26 that can be used in such an "inverted” system.
  • the excitation circuitry 24 includes two signal generators 30 "1 and 30 “2 , which generate AC excitation signals.
  • the AC excitation signal 31 "1 generated by the generator 30 " has a lower frequency than the frequency of the AC excitation signal 31 "2 generated by the generator 30 " .
  • the excitation circuitry 24 also includes two inverters 32 "1 and 32 “2 , which invert the AC excitation signals 31 output from the respective generators 30. -As .
  • the AC excitation signal 31 "1 is applied to terminal 1 of the connection interface 20 and the inverted excitation signal 33 "1 is applied to terminal 2 of the connection interface 20.
  • the higher frequency excitation signal 31 "2 is applied to terminal 3 of the connection interface 20 and the inverted excitation signal 33 "2 is applied to terminal 4 of the connection interface 20.
  • the connections made in the connection interface 20 are the same as the connections made in the connection interface 20 illustrated in Figure 2. Therefore, the lower frequency excitation signals are applied to the cos winding 15 and the higher frequency excitation signals are applied to the sin winding 14.
  • the detection circuitry 26 As shown in Figure 22, this is connected to terminals A and B of the connection interface 20. Therefore, signals induced in winding 13 are received at terminals A and B. As shown in Figure 22, the signals from these terminals are applied to a differential amplifier 34, which subtracts the signals obtained at these two terminals to remove any common mode signal induced in the winding 13. As before, as a result of the symmetrical arrangement of the windings 13, 14 and 15 substantially no signals will be output from the differential amplifier 34 in the absence of an in-homogeneity spot 12 in the film 11 of the sensor. However, when such an in- homogeneity spot 12 exists, signals will be obtained from the differential amplifier 34, which vary with the position of the in-homogeneity spot 12 along the length of the PCB
  • the signal output from the differential amplifier 34 can be represented (approximated) by:
  • Vu 34 A cos cos 2 ⁇ f 2 t (9)
  • the signal output from the differential amplifier 34 is applied to two mixers 36 “ and 36 " where it is multiplied by a respective one of the excitation signals 31 " and 31 " .
  • the outputs from the mixers 36 are then filtered by a respective low pass filter 38 "1 and 38 “2 to generate the DC signal values given above in equations 3 and 4, which are applied to the ratiometric calculator 40 as before.
  • FIG. 23 is a block diagram illustrating the form of the excitation and detection circuitry that is used in this alternative sensor design.
  • a single pulse generator 60 is provided that generates a train -of * voltage pulses 61, each pulse having a similar pulse shape and duration as the pulses generated by the pulse generator 60 described with reference to Figure 20.
  • the pulse train 61 is applied to switches 81 "1 and 81 “2 and to a counter 83 which is used to control the switches 81 to generate time shifted pulse trains 31 " and 31 " .
  • pulse train 31 "1 is applied to an inverter 32 “1 to generate an inverted pulse train 33 "1
  • the pulse train 31 " is applied to an inverter 32 "2 to generate the inverted pulse train 33 "2 .
  • pulse train 31 "1 is applied to terminal 1 of the connection interface 20; pulse train 33 “1 is applied to terminal 2 of the connection interface 20; pulse train 31 "2 is applied to terminal 3 of the connection interface 20; and pulse train 33 "2 is applied to terminal 4 of the connection interface 20.
  • terminal A of the connection interface 20 is connected to switch 62 “1 and to switch 62 “3 and terminal B of the connection interface 20 is connected to switch 62 “2 and switch 62 .
  • Switches 62 "1 and 62 " are opened and closed in synchronism with the pulses of the pulse train 31 "1 and the switches 62 " and 62 are opened and closed in synchronism with the pulses of the time shifted pulse train 31 "2 . Therefore, during the times at which pulses of the excitation voltage are applied to terminals 1 and 2 of the connection interface 20, the signals induced in the winding attached to terminals A and B of the connection interface 20 are applied to the differential amplifier 34 "1 and then to the mixer 36 "1 .
  • the mixer 36 "1 operates to demodulate the received signals, which are low pass filtered to regenerate the above described DC voltage V CO s. Similarly, during the 5 times at which voltage pulses are applied to terminals 3 and 4 of the connection interface 20, the signals induced in the sensor winding connected to terminals A and B of the connection interface 20 are passed through the differential amplifier 34 "2 to the mixer 36 "2 .
  • the mixer 36 "2 operates to demodulate the received signals by multiplying them with the excitation signal 31 " and the results are then low pass filtered by the low pass 10 filter 38 "2 to generate the above described DC voltage V S i n .
  • the remaining processing of the detection circuitry 26 is the same as the processing performed in the circuitry described with reference to Figure 6 and will not, therefore, be described again.
  • Figure 24 is a block diagram illustrating alternative excitation and detection circuitry 10 that can be used when the windings shown in Figure 2 are connected in the inverted manner illustrated in Figure 21.
  • the excitation circuitry includes a signal generator 30, which generates an AC excitation signal 31 "1 .
  • the excitation circuitry 24 also includes a 90 degree phase shifter 52, which applies a 90 degree phase shift to the excitation signal 31 "1 to generate a phase shifted excitation signal 31 "2 .
  • This phase shifted excitation signal 31 " is also applied to an inverter 32 "2 to generate an inverted and phase shifted excitation signal 33 " .
  • the excitation is also applied to an inverter 32 "2 to generate an inverted and phase shifted excitation signal 33 " .
  • a differential amplifier 34 of the detection circuitry 26 is 35 connected to terminals A and B of the connection interface 20.
  • the output from the differential amplifier 34 can be represented by equations 6 and 7 given above, hi other words the output from the differential amplifier 34 includes a sinusoidal signal whose phase varies with the ratiometric arctangent function given in equation 5 above.
  • the 40 output from the differential amplifier 34 is applied to a phase detector 56, which measures the phase of this signal relative to the phase of the AC excitation signal 31 "1 .
  • the remaining processing carried out by the detection circuitry 26 is the same as before and will not, therefore, be described again.
  • Figure 25 is a block diagram illustrating a further example of the excitation and detection circuitry that can be used when the windings shown in Figure 2 are connected in the inverted manner illustrated in Figure 21.
  • pulse width modulated pulse trains are used as excitation signals.
  • the excitation circuitry 24 includes a pulse generator 60 that is arranged to generate a high frequency periodic pulse train 61, the pulses of which have relatively fast rising and falling fronts (of the order of several nanoseconds) and a relatively long pulse period (of the order of several hundred nanoseconds).
  • the excitation circuitry 24 also includes a low frequency generator 48 that is arranged to generate a low frequency modulation signal 49 having a frequency of the order of several kHz.
  • the pulse train 61 generated by the pulse generator 60 and the low frequency modulation signal 49 are applied to a modulator 82 "1 which uses the low frequency modulation signal 49 to pulse width modulate the pulse train 61, to generate a pulse width modulated pulse train 84.
  • the pulse width modulated pulse train 84 is applied to terminal 1 of the connection interface 20 and is also applied to an inverter 32 " .
  • the inverted pulse train generated by the inverter 32 "1 is then applied to terminal 2 of the connection interface 20.
  • the low frequency modulation signal 49 generated by the low frequency generator 48 is also applied to a 90 degree phase shifter 52 that applies a 90 degree phase shift to the modulation signal 49.
  • the 90 degree phase shifted version of the modulation signal 49 is then applied to a pulse width modulator 82 "2 where it is used to pulse width modulate the pulse train 61 generated by the pulse generator 60.
  • Figure 25 illustrates the pulse width modulated signal 86 output from the modulator 82 "2 .
  • the pulse width modulated signal 86 is applied to terminal 3 of the connection interface 20 and is also applied to a second inverter 32 ⁇ 2 .
  • the inverted version of the pulse width modulated signal 86 is then applied to terminal 4 of the connection interface 20.
  • terminals A and B of the connection interface 20 are connected to switches 62 "1 and 62 ⁇ 2 which are opened and closed in synchronism with the pulses of the pulse train 61.
  • the switches 62 operate to down convert the high frequency component of the excitation signals (the component corresponding to the high frequency pulse train 61) to leave the lower frequency modulation signal component which is itself modulated by Vcos and V S in as before.
  • the.,signals-passmg-from the switches 62 are filtered by the low pass filter 68 (formed by the reference capacitor 64 and the differential amplifier 66) to remove high frequency components separated by the switching action of the switches 62, to leave a signal of the form given above in equation 7, whose frequency corresponds to that of the low frequency modulation signal 49 and whose phase varies with the ratiometric arctangent function defined in equation 5 above.
  • this sinusoidal signal output from the low pass filter 68 is applied to a phase detector 56 which measures the phase of the sinusoidal signal relative to the phase of the low frequency modulation signal 49 generated by the low frequency generator 48.
  • the remaining processing of the detection circuitry 26 is the same as before and will not, therefore, be described again.
  • Figure 26 is a block diagram illustrating a further alternative arrangement of the excitation and detection circuitry that can be used with the windings shown in Figure 2 when connected in the inverted manner illustrated in Figure 21.
  • the excitation circuitry 24 is similar to the excitation circuitry 24 used in the circuitry shown in Figure 25, except that the low frequency modulation signal 49 is used to amplitude modulate the pulses of the pulse train 61 generated by the pulse generator 60. This is achieved by mixing the low frequency modulation signal 49 with the pulse train 61 using the mixer 36 '1 and by modulating the 90 degree phase shifted version of the modulation signal with the pulse train 61 using the mixer 36 ' .
  • the remaining components of the excitation circuitry 24 are the same and will not, therefore, be described again.
  • the detection circuitry 26 is the same as the detection circuitry 26 shown in Figure 25. Therefore, a further description of the detection circuitry 26 will not be given.
  • Figure 27 is a block diagram illustrating a further alternative arrangement of the excitation and detection circuitry that can be used with the electrodes shown in Figure 2 when connected in the inverted manner illustrated in Figure 21.
  • the circuitry shown in Figure 27 is substantially the same as the circuitry shown in Figure 26 except that a different phase detection technique is used to detect the phase of the signal output by the filter and amplifier circuit 68.
  • the signal output from the filter and amplifier circuit 68 is mixed in a mixer 36 with a phase shifted version of the low frequency modulation signal 49 obtained by passing the low frequency modulation signal 49 through a variable phase shifter circuit 70.
  • the signal output from the mixer 36 is then filtered by a low pass filter 38 which removes AC signal components to leave a DC value that is input to the phase detector 72.
  • the phase detector 72 then varies the amount of phase shift applied by the phase shifter circuit 70 to determine the phase shift that minimises or maximises the DC value received from the filter 38. From this determined amount of phase shift, the phase detector 72 determines the value of the above described ratiometric function given in equation 5.
  • phase detector 72 is operable to minimise the DC signal obtained from the mixer 38, then 90 degrees is added to the determined phase to give the value of the ratiometric arctangent function of equation 5, whereas if the phase detector 72 is operable to maximise the DC signal obtained from the mixer 38, then the determined phase corresponds to the value of the ratiometric arctangent .functioj ⁇ ,,, of: equatioB55r- ⁇ he remaining components excitation and detection circuitry illustrated in Figure 27 are the same as those illustrated in Figure 26 and will not, therefore, be described again.
  • the detection circuitry 26 includes a phase detector 56 for measuring the phase of a filtered signal.
  • the amplification, filtering and phase detection circuitry used to process the signals obtained from terminals A and B of the connection interface 20 will introduce an uncontrolled phase shift into the signals.
  • a particular source of such error is related to the actual detection of the phase in the phase detector 56.
  • the crossing of a predetermined threshold level is typically measured in the phase detector 56 by using a comparator.
  • threshold-crossing detection will introduce a finite offset in the value of the output of the phase detector 56.
  • the actual value at the output of the phase detector 56 can be determined as a mean arithmetic value of the two measurements. In such case the output of the phase detector 56 will become independent of the level of the threshold voltage used in the comparator.
  • Figure 28 illustrates a more general approach that can be used to remove the uncontrolled offset in the phase of the measured signal simultaneously with an uncontrolled offset introduced in the measurement circuit of the phase detector 56.
  • the excitation circuitry 24 and the detection circuitry 26 are based on the circuitry shown in Figure 26.
  • the excitation circuitry 24 includes circuitry that allows the phase of the pulse trains applied to terminals 3 and 4 of the connection interface 20 to be inverted.
  • the excitation circuitry 24 includes switches 94 "1 and 94 "2 that operate to apply, during a first time interval, the non-inverted pulse train to terminal 3 and the inverted pulse train to terminal 4 and, during a second time interval, to apply the inverted pulse train to terminal 3 and the non-inverted pulse train to terminal 4.
  • Figure 29 illustrates in more detail the voltages labelled V 1 , V 2 , V 3 and V 4 applied to the four terminals 1, 2, 3 and 4 of the connection interface 20 during these two intervals (labelled Interval 1 and Interval 2 to the left and right of the line 77).
  • the phase of the pulse trains V 3 and V 4 changes by 180 degrees.
  • the phase detector 56 is arranged to detect the phase of the signal output by the low pass filter 68 during Interval 1 to calculate a first phase delay or time delay ti.
  • An interval controller 88 then switches over the position of the switches 94 "1 and 94 " and informs the phase detector 56 accordingly.
  • the phase detector 56 then re- measures the phase of the received signal during Interval 2, to calculate a second phase delay t 2 .
  • the phase detector 56 then calculates a phase (or time) difference, t 0 , as the phase ti-t 2 .
  • Figure 30 is a block diagram illustrating a further alternative arrangement of the excitation and the detection circuitry that can be used with the windings shown in Figure 2 when connected in the non-inverted manner as per the first sensor design described above, hi this design, the excitation signals and the signals used to control the detection are stored in a programmable read-only memory (PROM) 80 and read out of this memory in response to pulses generated by a pulse generator 60.
  • the PROM outputs a high frequency periodic pulse train 31 "1 from terminal OA * and an inverted high frequency periodic pulse train 31 " from terminal OB which are applied to terminals A and B of the connection interface 20.
  • the PROM outputs a high frequency periodic pulse train 31 "1 from terminal OA * and an inverted high frequency periodic pulse train 31 " from terminal OB which are applied to terminals A and B of the connection interface 20.
  • a low frequency amplitude control signal which it outputs from terminal 02 ; a mixer/sign control signal which it outputs from terminal 03 ; and a sin/cos multiplexer signal which it outputs from terminal 04 .
  • the sin/cos multiplexer signal is used to control the multiplexing of the signals from the cos sensor winding 15 (obtained from terminals 1 and 2 of the connection interface 20) and the signals from the sin sensor winding 14 (obtained from terminals 3 and 4 of the connection interface 20) using the multiplexer switches 76 "1 and 76 "2 .
  • the multiplexed signal output from the multiplexing switches 16 and an inverted version of the multiplexed signal are applied to mixing switches 62 "1 and 62 "2 respectively.
  • the mixing switches 62 "1 and 62 ⁇ 2 effectively demodulate the multiplexed signals which are then re-modulated onto a low frequency carrier signal by the low frequency amplitude control signal via the switches 62 " and 61 4 .
  • the signals from the sin and cos sensor windings 14 and 15 are then combined, filtered and amplified by the circuitry 97 to leave a signal of the form given above in equation 7, whose frequency corresponds to that of the low frequency amplitude control signal and whose phase varies with the ratiometric arctangent function defined in equation 5 above.
  • the phase of this signal is then measured by the phase detector 56 as before.
  • Figure 31 is a signal diagram illustrating the form of the signals generated by the PROM 80. hi particular, Figure 31 illustrates the high frequency excitation signals 31 output from terminals OA * and OB * of the PROM 80. Figure 31 also illustrates the sin/cos multiplexer signal output from terminal 04 of the prom 80. As shown, the multiplexer signal operates at a frequency of 1/16* that of the excitation signals 31.
  • the signals from the sin sensor winding 14 are passed through to the mixing switches 62; and during the even numbered windows W2, W4 etc, the signals obtained from the cos sensor winding 15 (obtained from terminals 1 and 2 of the connection interface 20) are passed through to the mixing switches 62.
  • the mixer/sin control signal is a square wave signal having the same frequency as that of the excitation signals 31. Therefore, the mixing switches 62 "1 and 62 ⁇ 2 operate to demodulate the multiplexed signals. The demodulated signals are then either connected to ground or accumulated onto a reference capacitor 64 "1 or 64 ⁇ 2 for
  • the low frequency amplitude control signal effectively amplitude modulates the demodulated signals from the sin and cos windings 14 and 15 on to low frequency carrier signals (having a frequency 1/576 that of the excitation signals) that are 90 degrees out of phase with each other.
  • This is illustrated in Figure 32 which shows the amplitude modulated signals 101 and 103 obtained from the mixing switch 62 " . Similar, but 180 degree phase shifted signals will be obtained from mixing switch 62 .
  • Signal 101 is obtained when the signal from the sin sensor winding 14 is passed through the multiplexing switches 76 and signal 103 is obtained when the signals obtained from the cos sensor winding are passed through the multiplexing switches 76.
  • both signals 101 and 103 are accumulated on the reference capacitor 64, the actual signal that is accumulated will represent the sum of these two signals, as illustrated by signal 105.
  • the filtering and amplification circuitry 97 effectively filters this signal 105 to produce a smooth sinusoidal signal 107 whose phase varies with the arctangent function defined in equation 5 above. This sinusoidal signal 107 is then passed to the phase detector 56 which measures the phase of this signal relative to the reference signal supplied by the PROM 80 from terminal 01 .
  • the low frequency amplitude control signal is a binary signal that controls the opening and closing of switches 62 "3 and 62 A .
  • the pulse duration of the low frequency amplitude control signal is cyclically varied from a minimum duration corresponding to one period of the excitation signal (such as is applied during windows Wl and W3) to a maximum duration corresponding to 16 periods of the excitation signal (such as is applied during window W2).
  • the amount of signal accumulated on the reference capacitor 64 will also cyclically vary, with the peak amplitude of the accumulated signal depending on the level of the demodulated signals obtained from mixer switch 62 " and hence on V CO s and V S i n .
  • the switches 62 "3 and 62 ⁇ 4 are simple switches that do not have a reverse polarity, in order to achieve the change in polarity of the amplitude modulation, the phase of the mixing signals applied to the mixer switches 62 " and 62 "2 are shifted by 180 degrees at the appropriate timings in order to change the sign of the re-modulated signal. This is illustrated in Figures 31 and 32 around the zero crossing for the signal 101.
  • Figure 32 illustrates the form of the signal 105 accumulated on one of the reference capacitors 64.
  • the signal accumulated on the other reference capacitor 64 will have the same form but opposite polarity.
  • a description will now be given (with reference to Figure 33) of the way in which the amplification and filtering circuitry 97 operates to combine these signals to generate the signal 107 which is output to the phase detector 56.
  • the reference capacitors 64 "1 and 64 "2 are connected to ground.
  • the high frequency components of the input currents Ii and h are shorted to ground thrpugh a .smalLimpedance, whereas low frequencyvariatio ⁇ s " (such as are caused by the low frequency modulation of the demodulated signals) marked as Ii and I 2 , are further amplified and combined by the circuitry 97 to generate the output signal 93.
  • the output current (Zi) from the circuitry 97 will become proportional to the difference in the currents Ii and I 2 , and as a result, any common mode signals will be rejected.
  • the desired signals that vary with the position being measured
  • the two reference capacitors 64 " *and 64 "2 are of opposite polarity and will therefore be added together by the circuitry 97. Consequently, the output signal can be approximated by equation 7 given above, with the frequency f 2 corresponding to half the repetition frequency of the low frequency amplitude control signal.
  • Excitation And Detector Circuitry - 12 Figure 34 is a block diagram illustrating a further alternative arrangement of the excitation and detection circuitry that can be used with the windings shown in Figure 2 when connected in the non-inverted manner as in the first sensor design described above.
  • the excitation and detection circuitry shown in Figure 34 is based on the excitation and detection circuitry shown in Figure 30.
  • the main difference is that the sign control and the low frequency amplitude control are effectively combined with the original excitation signals 31 to form a new set of excitation signals 31 "1 and 31 "2 that are output from the PROM 80 from terminals OA and OB .
  • these signals are effectively combined with the excitation signals so that during the periods that the switches 62 ⁇ 3 and 02 "4 in the Figure 30 circuit were connected to ground, no excitation signals are applied to the excitation winding attached to terminals A and B of the connection interface 20.
  • Figure 35 is a signal diagram illustrating the form of the signals generated by the PROM 80 in Figure 34. As shown, during window
  • the phase of each excitation pulse applied during window W3 is 180 degrees shifted compared to the phase of the excitation pulses applied during window Wl. Therefore, as those skilled in the art will appreciate, since the sign control is also achieved by varying the excitation signals 31, the mixer signal applied to the mixing switches 62 “1 and 62 " can be a constant square wave signal at the same frequency as the frequency of the excitation pulses (as shown in Figure 35).
  • the remaining processing circuitry illustrated in Figure 34 is the same as the processing circuitry of Figure 30 and a further description will therefore be omitted.
  • Figure 36 is a block diagram illustrating a further alternative arrangement of the excitation and detection circuitry that can be used with the winding shown in Figure 2 when connected in the inverted manner illustrated in Figure 21.
  • the circuitry illustrated in Figure 36 is based on the circuitry illustrated in Figure 30.
  • the PROM 80 generates four excitation signals 31 "1 , 31 ⁇ 2 , 31 "3 , and 31 "4 which it outputs from terminals 05 * , 06 * , 07 * and 08 * .
  • These excitation signals 31 are applied to a respective one of -the terminals numbered 1 to 4 of the connection interface 20 which are connected to the windings 14 and 15, as shown in Figure 21.
  • the PROM 80 is arranged to generate the excitation signals so that excitation signals 31 "3 and 31 "4 are applied to terminals 3 and 4 of the connection interface 20 during the odd numbered windows Wl, W3 etc and so that excitation signals 31 "1 and 31 "2 are applied to terminals 1 and 2 of the connection interface 20 during the even numbered windows W2, W4 etc. hi this way, the excitation signals are applied to the windings 14 and 15 in a time division multiplexed manner.
  • the circuitry is the same as the detection circuitry illustrated in Figure 30, except there is no need for the multiplexing switches 76 due to the multiplexing of the excitation signals. A further description of the operation of the detection circuitry will therefore be omitted.
  • Figure 38 is a block diagram illustrating a modification of the excitation circuitry shown in Figure 36.
  • the PROM 80 outputs excitation signals 31 “1 and 31 “2 .
  • the DC components of these pulse trains are removed by the capacitors 96 “1 and 96 “2 to generate pulse trains 98 “1 and 98 “2 , which are 180 degrees out of phase with each other.
  • these pulse trains 98 are applied to terminals 1 to 4 of the connection interface 20 through a set of diodes 99 “ , 99 “ , 99 “ and 94 “ ⁇
  • diodes 99 “1 and 99 “2 when the pulse train 98 “1 is positive and the pulse train 98 “2 is negative, excitation current will flow through diode 99 “ into terminal 1 of the connection interface 20 . and the current will flow from terminal 2 of the connection interface 20 through diode 99 “2 back to ground.
  • pulse train 98 "1 is negative and pulse train 98 "2 is positive, no current is applied to terminals 1 and 2 of the connection interface 20.
  • the detection circuitry is the same as before, and therefore a further description will be omitted.
  • the DC magnet 6 was oriented with its magnetic axis parallel with the Z direction. This results in the position encoder being sensitive to rotational movement of the piston 4 about its axis. This is not a problem where the piston 4 is"arranged*so thaHt cannot rotate within the cylinder 2, However, in most piston and cylinder assemblies, the piston is free to rotate within the cylinder. Therefore, in a preferred embodiment, the magnet 6 is a ring magnet which is attached around the piston shaft so that the magnetic axis of the magnet is directed in the X direction shown in Figure 1. With such an arrangement, the magnetic field generated by the magnet will be the same whether or not the piston 4 rotates within the cylinder 2.
  • the DC magnetic field was generated by the DC magnet 6.
  • the DC magnetic field may be generated by using an electromagnet or by applying a DC current to an appropriately oriented coil.
  • the movable object e.g. the piston
  • the AC magnetic field generator which generates a positionally varying magnetic field. Provided this positionally varying magnetic field interacts with the film 25 of magnetic material so that its magnetisation state varies along the measurement path, a signal will be generated in the sensor winding when the excitation winding is energised with the excitation signal.
  • the AC magnetic field generator mounted on the moving object preferably generates a low frequency signal that can penetrate through metallic walls. Additionally, in such an embodiment, the signal generated in the sensor winding will be at an intermodulation frequency defined by the difference in frequency between the two AC magnetic fields.
  • the AC magnetic field generator may be an active device such as a powered coil or a passive device, such as a resonator that is energised by the excitation of the excitation windings.
  • the magnetisable material that was used was a soft ferromagnetic material having high initial and maximum permeability and a low coercivity.
  • other magnetisable materials may be used.
  • amorphous alloys such as VITROVAC from Vacuumschmelze or Nanno-crystalline alloys such as VITROPERM from Vacuumschmelze, silicon iron alloys with three percent silicon, pure iron, nickel iron alloys, cobalt iron alloys etc.
  • the sensor windings were each formed from two turns of conductor. As those skilled in the art will appreciate, the use of two turn sensor windings is not essential. Any number of turns may be provided. Preferably, as many turns as possible are provided in the space allowed by the dimensions of the PCB 8 as this maximises signal levels obtained from the sensor windings.
  • the excitation winding included a single turn of conductor.
  • the number of turns of conductor for the excitation winding and for the sensor windings can be varied in order to vary the reactive impedance of the windings to match the impedance of the appropriate output or input of the excitation and detection electronics.
  • the position encoder was used to determine the position of a piston within a cylinder.
  • the position sensor described above may be used in a number of different applications. For example, it can be used in shock absorbers, damping cylinders, syringes, machine tool applications, etc.
  • the applicants earlier International Application WO2005/085763 illustrates- a -number-of- such different applications and sensor arrangements (including " rotary sensor arrangements) to which the improvements described herein can be applied.
  • the position of the piston determined by the detection electronics was displayed on a display.
  • the position information may be provided to another computer system for controlling another part of a system.
  • the determined position may be supplied to an engine management unit which can use the position information to control the timing of ignition of the fuel mixture within the piston and cylinder assembly.
  • the excitation current applied to the excitation winding had a frequency of approximately 10MHz. As those skilled in the art will appreciate, it is not essential to use such an excitation frequency. Excitation frequencies between 10kHz and 1 OMHz are preferably used.
  • the sensor windings were formed in a figure of eight configuration. As those skilled in the art will appreciate, it is not essential to form the sensor windings in such a figure of eight configuration.
  • the only requirement of the sensor windings is that they are able to detect a magnetic field which positionally varies along the measurement direction. This can be achieved by a single sensor winding positioned at a position along the measurement path. Alternatively, it can be achieved using a sensor winding which geometrically varies along the measurement path. This geometrical variation may be its shape along the measurement path or its dimensions such as the thickness of the conductor forming the sensor winding or the number of turns of the sensor winding etc.
  • the sensor windings and the film of magnetic material have been fixed (stationary) and the magnet was mounted in the movable member.
  • the position encoder described above will operate where the magnet is stationary and where the sensor windings move relative to the magnet.
  • the sensor windings and/or the excitation windings may move in addition to the magnet. All that is necessary is that there is relative movement between the magnet and at least one of the sensor winding and the excitation winding.
  • the magnetic material may move with the movable object or it may move independently of the object. However, since the film of magnetic material is substantially homogenous, movement of the film relative to the sensor windings will not affect the operation of the position encoder.
  • the excitation and sensor windings were formed as conductor tracks on a printed circuit board.
  • the excitation and sensor windings may be formed using any conductive material, such as conductive inks which can be printed on an appropriate substrate or conductive wire wound in the appropriate manner.
  • two separate printed circuit boards may be provided, one carrying the excitation winding and the other carrying the or each sensor winding.
  • the excitation signal applied to the excitation winding was an AC signal at a particular frequency.
  • the excitation signal may be any cyclically varying signal.
  • phase quadrature sensor windings were provided. As those skilled in the art will appreciate, it is not essential to use sensor windings that are in phase quadrature. For example, instead of using the cos sensor winding, a second sensor winding phase shifted by an eighth of the pitch along the measurement path may be used. However, as those skilled in the art will appreciate, the use of phase quadrature sensor windings is preferred as this simplifies the processing to be performed by the detection circuitry to determine the position of the movable member. Additionally, as those skilled in the art will appreciate, in embodiments that use sensor windings that provide signal levels that vary substantially sinusoidally with the position of the movable member, it is not essential to only use two sensor windings.
  • three or four sensor windings may be provided each separated along the measurement path by an appropriate distance (or angle in the case of a rotary position encoder).
  • an AC current was applied to the excitation winding having a peak amplitude of approximately 5OmA.
  • the magnitude of the excitation current is preferably chosen depending on the position and layout of the excitation winding relative to the film of magnetisable material so that the excitation magnetic field generated by the excitation winding will be strong enough to produce a reasonable signal to noise ratio in the measurement electronics. Therefore, appropriate excitation current strengths may vary from 0.01mA to 1OA.
  • the film of magnetic material had a thickness of less than 0.2mm.
  • films of various thicknesses may be used.
  • the thickness of the material is between 20microns to 2mm as such films are readily available.
  • the film of magnetic material and the PCB had approximately the same width and extent along the measurement direction.
  • the width and length of the film of magnetisable material is greater than the width and length of the PCB, as this minimises edge effects associated with the edge of the film of magnetisable material (which can alter the balance between the excitation and sensor windings).
  • the film of magnetisable material was initially unsaturated and the DC magnet created an in-homogeneity spot in the film. The position of this in- homogeneity spot was then detected by detecting the change in the mutual inductance between an excitation winding and a sensor winding.
  • the film of magnetisable material does not have to be initially in an unsaturated state.
  • a strong background (or bias) magnetic field may be provided near the film which saturates the entire film. Such a fully saturated film is still homogenous and will not alter the mutual coupling between the excitation and sensor windings.
  • the DC magnet carried by the moving member is strong enough' to counter the effects of the saturating field to create an in-homogeneity region in the film, a similar imbalance will be created between the excitation and sensor windings. This can then be detected in the manner described above in the first embodiment. As those skilled in the art will appreciate, the position of this in-homogeneity region will not correspond to the position where the DC magnetic field is perpendicular to the film, but where the DC field is strong enough to counter the effect of the bias field.
  • two excitation signals of different frequencies where simultaneously applied, one to each of the sin and cos windings 14 and 15.
  • the two excitation signals may be applied one after the other, in which case only one mixer 36 and one low pass filter 38 are required in the detection circuitry 26. Further, if the excitation signals are applied to the two windings 14 and 15 at different times, then the same excitation frequency can be applied to each pair.
  • the windings were arranged to provide signals that varied (approximately) in a sinusoidal manner with the position to be detected.
  • different shaped windings may be provided that provide a substantially linear variation with position along the length of the sensor head.
  • two excitation windings were provided at each end of the cylinder. These two excitation windings were connected in series and energised by a single excitation circuit. As those skilled in the art will appreciate, the two excitation windings may be individually connected to separate excitation circuitry which drives the respective excitation windings accordingly.
  • the single excitation winding may be mounted on one of the sensor boards (transducers) 8 "1 or 8 "2 or it may be mounted on a separate board, for example, positioned between the sensor boards 8 "1 and 8 "2 .
  • a single sensor winding may be provided in common to both transducers. Again, this common sensor winding may be provided on either of the sensor boards 8 "1 or 8 “2 or on another sensor board, for example, positioned between sensor boards 8 "1 and 8 "2 .
  • transducers in the form of printed circuits boards which carried excitation and sensor windings and a film of magnetisable material were used.
  • other types of sensing transducers can be used.
  • sensing transducers can be used that define the excitation and sensor windings by conductive ink printed on a suitable substrate or by appropriately shaped wire bonded onto an appropriate substrate. Further, in order to maximise signal levels obtained from the sensor windings, the magnetic film is preferably provided on each side of the sensing transducer substrate.. , — — . , r
  • the excitation and the detection circuits included various electronic hardware circuits, hi an alternative embodiment, a programmable circuit (processor) controlled by software stored in a memory may implement these circuits.
  • the software may be provided in any appropriate form and in any computer language. It may be supplied as a signal or stored on a computer readable medium such as a CD ROM.

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Measurement Of Length, Angles, Or The Like Using Electric Or Magnetic Means (AREA)
  • Transmission And Conversion Of Sensor Element Output (AREA)

Abstract

A position encoder is provided for indicating the relative position between first and second relatively movable members. One of the members carries a magnetic field generator which generates a magnetic field whose magnitude and direction vary with position. In one embodiment, the other member carries an excitation winding, one or more sensor windings and a film of magnetisable material. The arrangement is such that the positionally varying magnetic field interacts with the film to change the mutual coupling between the excitation winding and the/or each sensor winding. Excitation and processing circuitry is provided for energising the excitation winding and for processing the signals generated in the sensor windings, to determine a value indicative of the relative position between the first and second relatively movable members.

Description

POSITION SENSOR
The present invention relates to position sensors or encoders and to apparatus and circuits for use in such sensors or encoders. The invention has particular but not exclusive relevance to non-contact rotary and linear position encoders. Some embodiments of the invention are suitable for use in relatively small systems, which operate at relatively high temperatures and in which there may be magnetic and electromagnetic interference.
Other embodiments are suitable for use in systems having a large measurement range and which require a sensor head that is relatively insensitive to mechanical misalignments, dirt, grease and the like.
Inductive position encoders are well known in the art and typically comprise a movable member, whose position is related to the machine about which position or motion information is desired, and a stationary member which is inductively coupled to the moving member. The stationary member typically includes a number of detectors which provide electrical output signals which can be processed to provide an indication of the position, direction, speed and/or acceleration of the movable member and hence for those of the related machine.
Some of these inductive position encoders employ an AC magnetic field generator mounted on the movable member and one or more sensor windings mounted on the stationary member. The magnetic field generator and the sensor windings are arranged so that the magnetic coupling between them varies with the position of the movable member relative to the stationary member. As a result, an output signal is obtained from each sensor winding which continuously varies with the position of the movable member.
The main limitation of this type of inductive position encoder is that the technique can only work if the space between the magnetic field generator and the sensor windings is free from metallic walls which cannot be penetrated by the AC magnetic field. This therefore limits the applications to which such inductive position sensors can be used: "^"
Another type of magnetic position sensor that is known uses permanent magnets and Hall Effect detectors. The problem with such Hall Effect detectors is that they are point magnetic field detectors which can only detect the presence or absence of the permanent magnet. Therefore, such Hall Effect systems do not provide a "continuous" output signal which varies with the position of the movable object.
Another type of inductive position sensor is a Flux Gate sensor which employs a film of soft magnetisable material. With this type of sensor, an AC excitation magnetic field is applied along the plane of the soft magnetisable material causing the material to be driven into and out of magnetic saturation. A DC magnet is usually provided on the movable member, the magnetic field of which interacts with the saturable magnetic element to cause some of the AC in-plane flux to be expelled from the material adjacent the magnet. A number of sensor windings are provided adjacent the material for detecting the flux that is expelled from the saturable element (which depends on the position of the magnet), to output a signal which depends on the position of the magnet relative to the sensor windings. As a result of the driving of the saturable magnetic element into and out of saturation, the output signals that are generated in such Flux Gate sensors are at twice the frequency of the AC excitation frequency. The applicant's earlier international application WO 97/14935 describes such a Flux Gate position sensor. The problem with such Flux Gate sensors is that they are relatively bulky as they employ a relatively large ferromagnetic core around which the excitation and sensor windings are typically wound. They also tend not to be as rugged as smaller more integrated sensor technologies such as the Hall Effect systems discussed above.
The applicant has proposed an alternative type of inductive position encoder, which is described in WO 2005/085763. This encoder is similar in structure to the Flux Gate encoder in that it uses a film of magnetisable material, a magnet and excitation and sensor windings. However, unlike the Flux Gate encoder, title excitation circuit is arranged so as not to drive the film of magnetisable material into and out of saturation near the sensor winding. Instead the sensor winding is arranged to sense the change in mutual coupling between the excitation winding and the sensor winding, which varies with the position of an in-homogeneity in the film of magnetisable material caused by the magnetic field generated by the magnet. The present application describes various improvements to the encoder and electronics described in WO '763. Some of these improvements are specific to the type of inductive encoder described in WO'763, whilst others are more generally applicable to other types of encoders.
According to one aspect, the invention provides a position encoder comprising: a magnetic field generator for generating a magnetic field that varies with position along a measurement path; first and second sensor transducers each comprising: i) an excitation winding and a sensor winding; and ii) a film of magnetisable material located adjacent said excitation and sensor windings and located, in use, within said positionally varying magnetic field to cause the film to have a positionally varying magnetisation state along the measurement path; wherein the film of magnetisable material of said first transducer is positioned, in use, closer to said magnetic field generator than the film of magnetisable material of said second transducer; wherein the excitation winding and the sensor winding of each transducer are arranged relative to the corresponding film of magnetisable material so that a mutual electromagnetic coupling between them varies in dependence upon the positionally varying magnetisation state of the corresponding film of magnetisable material, so that when the excitation winding is energised with an excitation signal, a sensor signal is generated in the sensor winding that varies with the relative position between said first and second members; differencing circuitry operable to combine the sensor signals generated in said sensor windings to form a difference signal representing the difference in electromagnetic coupling between the excitation and sensor windings of the first and second transducers; and a processing circuit operable to process said difference signal to determine a value indicative of the relative position of said first and second relatively moveable members.
The excitation windings of said transducers may be formed from a single common winding or two separate windings may be provided. Similarly, the sensor windings of said transducers may be provided as separate windings or they may be formed by a single winding that is provided in common to the two transducers.
In one embodiment the connections to the excitation or the sensor windings are arranged so that the differencing circuitry adds the signals obtained from the sensor windings of the first and second transducers.
In a preferred embodiment excitation circuitry is provided that generates an excitation signal having an excitation frequency and the processing circuit processes the difference signal which is substantially at said excitation frequency, to determine said value indicative of the relative position between the first and second relatively movable members. In one embodiment, the excitation and the sensor windings are arranged relative to each other so that in the absence of the magnetic field generator, there is substantially no electromagnetic coupling between them.
Other aspects of the invention relate to numerous novel excitation and detection circuits that are described below.
These and various other features and aspects of the invention will become apparent from the following detailed description of embodiments which are described with reference to the accompanying figures in which:
Figure 1, comprising Figures IA, IB and 1C, illustrates the main components of an inductive position sensor used to sense the position of a piston within a cylinder;
Figure 2 schematically illustrates the layout of a set of windings that forms part of the sensor shown in Figure 1 ;
Figure 3 A is a plot illustrating the way in which the peak amplitude of a signal generated in a sin sensor winding (shown in Figure 3B) varies with the position of the piston within the cylinder;
Figure 4A is a plot illustrating the way in "which the, peak amplitude of a signal generated in a cos sensor winding (shown in Figure 4B) varies with the position of the piston within the cylinder;
Figure 5A is a plot illustrating the locus of points obtained by plotting the peak amplitude of the signal induced in the sin sensor winding against the peak amplitude of the signal induced in the cos sensor winding as the piston moves from one end of the cylinder to the other end;
Figure 5B is a phase plot illustrating the way in which a phase angle obtained from measured sin and cos signals varies with the position of the piston within the cylinder;
Figure 6 is a block diagram illustrating the main components of exemplary excitation and detection circuitry that forms part of the sensor shown in Figure 1 ;
Figure 7 is a block diagram illustrating the main components of a ratiometric calculator forming part of the detection circuitry shown in Figure 6; Figure 8 is a block diagram illustrating the main components of a phase detector forming part of the ratiometric calculator shown in Figure 7;
Figures 9A and 9B illustrate the way in which a magnet positioned adjacent to the piston and cylinder assembly shown in Figure 1 can introduce an error into the phase measurements, which can lead to errors in the determined position of the piston within the cylinder:
Figure 10 schematically illustrates the way in which two sets of excitation and sensor windings can be used to overcome the problem associated with adjacent magnets;
Figure 11 is a block diagram illustrating the main components of exemplary excitation and detection circuitry that can be used with the excitation and sensor windings shown in Figure 10;
Figure 12, comprising Figures 12A and 12B, illustrates the operation of an algorithm controller used to control the operation of multiplexing switches that form part of the detection circuitry shown in Figure 11;
Figure 13 is a schematic diagram illustrating an alternative arrangement of excitation and sensor windings that can be used to reduce the effect of adjacent magnets on the position sensor;
Figure 14 is a block diagram illustrating the main components of exemplary excitation and detection circuitry that can be used with the excitation and sensor windings shown in Figure 13;
Figure 15, which comprises Figures 15 A, 15B and 15 C, illustrates the operation of an algorithm controller that forms part of the detection circuitry shown in Figure 14;
Figure 16, -which. comprises Figure 16A,..lj5B and 16C, illustrates an alternative sensor design which mitigates the problem associated with adjacent magnets which employs two sets of spaced excitation and sensor windings;
Figure 17 schematically illustrates the way in which the two sets of excitation and sensor windings of the sensor illustrated in Figure 16 maybe connected together;
Figure 18 schematically illustrates an alternative way in which the two sets of excitation and sensor windings illustrated in Figure 16 may be connected together;
Figure 19 is a block diagram illustrating an alternative form of excitation and detection circuitry that can be used with the sensors illustrated in Figures 2, 17 and 18;
Figure 20 is a block diagram illustrating an alternative form of excitation and detection circuitry that can be used with the sensors illustrated in Figures 2, 17 and 18;
Figure 21 is an electrical equivalent circuit illustrating an alternative way of connecting the windings of the sensors shown in Figures 2, 17 and 18 to the excitation and detection circuitry in which the windings previously connected to the excitation circuitry are connected to the detection circuitry and the windings previously connected to the detection circuitry are connected to the excitation circuitry;
Figure 22 is a block diagram illustrating excitation and detection circuitry that can be used to drive the windings in the inverted manner illustrated in Figure 21;
Figure 23 is a block diagram illustrating an alternative arrangement of the excitation and detection circuitry that can be used to drive the windings in the inverted manner illustrated in Figure 21 ;
Figure 24 is a block diagram illustrating the excitation and detection circuitry that can be used in a further alternative for driving the windings in the inverted manner illustrated in Figure 21;
Figure 25 is a block diagram illustrating the excitation and detection circuitry that can be used in a further alternative for driving the windings in the inverted manner illustrated in Figure 21;
Figure 26 is a block diagram illustrating the excitation and detection circuitry that can be used in a further alternative for driving the windings in the inverted manner illustrated in Figure 21;
Figure 27 is a block diagram illustrating the excitation and detection circuitry that can be used in a further alternative for driving the windings in the inverted manner illustrated in Figure 21;
Figure 28 is a block diagram illustrating the excitation and detection circuitry that can be used in a further alternative for driving the windings in the inverted manner illustrated in Figure 21;
Figure 29 illustrates pulse trains generated by the excitation circuitry shown in Figure 28;
Figure 30 is a block diagram illustrating an alternative form of excitation and detection circuitry that can be used with the sensors illustrated in Figures 2, 17 and 18;
Figure 31 is a signal diagram illustrating the signals generated by a PROM forming part of the circuit shown in Figure 30;
Figure 32 is a signal diagram illustrating in more detail the way in which the detection circuitry shown in Figure 30 operates to generate a signal whose phase varies with the position being measured;
Figure 33 is a circuit diagram of circuitry used to filter and amplify signals obtained from mixing switches forming part of the circuit shown in Figure 30;
Figure 34 is a block diagram illustrating an alternative form of excitation and detection circuitry that can be used with the sensors illustrated in Figures 2, 17 and 18; Figure 35 is a signal diagram illustrating the signals generated by a PROM forming part of the circuit shown in Figure 34;
Figure 36 is a block diagram illustrating the excitation and detection circuitry that can be used in a further alternative for driving the windings in the inverted manner illustrated in Figure 21;
Figure 37 is a signal diagram illustrating the signals generated by a PROM forming part of the circuit shown in Figure 36; and
Figure 38 is a block diagram illustrating the excitation and detection circuitry that can be used in a further alternative for driving the windings in the inverted manner illustrated in Figure 21.
Overview
Figures IA, IB and 1C schematically illustrate the form and operation of an inductive position sensor 1 and how it is used to sense the position of a piston within a cylinder. More details of the position sensor and the way in which it works can be found in the applicant's earlier PCT patent application WO 2005/085763, the contents of which are incorporated herein by reference. For ease of understanding the following sections, a brief description of the sensor illustrated in Figure 1 will now be given.
As shown in Figure 1C, the piston and cylinder assembly comprises an elongate cylinder 2 having a central bore 3 in which a piston 4 is reciprocally mounted. The piston 4 includes a piston head 5 whose outer diameter corresponds to the inner diameter of the cylinder bore 3. As shown, the piston head 5 includes a DC magnet 6 oriented with its north and south poles being aligned horizontally in the Figure (i.e. in the Z direction).
The cylinder 2 is encased within a metal casing 7, part of which is cut away to show the cylinder 2 and the piston 4.
A printed circuit board (PCB) 8 is attached to the left hand side (in the Figure) of the cylinder casing 7 and carries a number of conductive tracks 9 (shown in Figure IA) which are connected together to define a number of excitation and sensor windings (not shown) which are connected to excitation and detection electronics 10 via a connection interface 20. A film 11 of magnetisable material is provided on the outer surface (i.e. furthest away from the cylinder 2) of the PCB 8, although it may be provided on the inner surface of the PCB 8 instead of or in addition to being on the outer surface. In this system, the film 11 of magnetisable material is Niso%Fe2o%, which is a soft ferromagnetic material having a high initial and maximum permeability (100,000 to 400,000), a low coercivity and a relatively low saturation field (lA/m) and polarisation (0.7T to 0.9T). Preferably the film 11 has a thickness in the Z direction of less than 0.2mm and has approximately the same width and length as the PCB 8.
The excitation and sensor windings carried by the PCB 8 are geometrically arranged on the PCB 8 so that, in the absence of the piston 4, there is substantially no electromagnetic (inductive) coupling between them. In other words, in the absence of the piston 4, when an AC excitation current is applied to the excitation winding, substantially no signal is induced in the (or each) sensor winding. Additionally, the film 11 of magnetisable material is substantially homogenous so that, in the absence of the magnet 6, its presence adjacent to the excitation and sensor windings does not affect the balance between the windings. However, when the piston 4 is present, the magnet 6 within the piston head 5 creates a positionally varying DC magnetic field, which penetrates through the metallic cylinder casing 7 and the PCB 8 and interacts with the portion of the film 11 that is next to the magnet 6. The interaction of the DC magnetic field with this portion of the film 11 creates an imbalance between the excitation and sensor windings such that when the AC excitation current is applied to the excitation winding, AC signals are generated in the (or each) sensor winding which vary with the position of the piston 4 within the cylinder 2. Therefore, by suitable processing of the signals generated in the (or each) sensor winding, the electronics 10 determines and outputs (e.g. on a display) a value indicative of the position of the piston 4 within the cylinder 2.
Figure IA schematically shows the field lines of the DC magnetic field generated by the magnet 6. This magnetic field has two components, one which is directed perpendicular to the film 11 of magnetisable material and the other which is directed "in the plane" of the film 11 of magnetisable material. The component that is perpendicular to the film 11 of magnetisable material does not interact with the film 11. The other "in plane" component does interact with the film 11 and, depending on its magnitude, can magnetically saturate the part of the film 11 located adjacent the magnet 6. Therefore, at the spot 12 where the DC magnetic field generated by the DC magnet 6 is perpendicular to the film 11, no saturation of the film 11 occurs. With the orientation of the magnet 6 shown in Figure 1C, this spot 12 is located where the magnetic axis of the DC magnet 6 intersects with the film 11 of magnetisable material. However, surrounding the spot 12 of unsaturated film, the magnetic field generated by the DC magnet includes in-plane components that saturate the magnetisable material within the portion 16 immediately adjacent the magnet 6. The size of the portion 16 depends on the distance between the magnet 6 and the film 11 of magnetisable material and the strength of the DC magnetic field.
Figure IB is a plot showing the way in which the strength of the in-plane component (H) of the DC magnetic field varies along the X direction through the centre of the spot 12 (which corresponds to the origin of the plot). As shown, the in-plane component (H) is zero in the centre of the spot 12 and is negative on the left hand side of the origin and is positive on the right hand side of the origin. The horizontal dashed lines shown in the plot correspond to the strength of the magnetic field required to saturate the film 11 of magnetisable material. As shown, on either side of the spot 12 the in-plane component (H) of the magnetic field is strong enough to saturate the film 11 of magnetisable material. This is illustrated in the second plot shown in Figure IB, which shows the way in which the magnetic flux density (B) within the film 11 of magnetisable material varies along the X direction taken through the centre of the spot 12. As shown, on either side of the spot 12, the magnetic flux density within the film 11 is constant where the film 11 is saturated. However, between the vertical dashed lines the in-plane component (H) of the magnetic field is not strong enough to saturate the film 11.
The third plot shown in Figure IB illustrates the way in which the AC relative permeability (μ) of the film 11 varies along the X direction through the centre of the spot 12. As shown, on either side of the spot 12, the relative permeability is approximately one where the film 11 has become MIy saturated. However, within the spot 12 between the vertical dashed lines, the film 11 of magnetisable material is not saturated and because of the large rate of change of the in-plane magnetic field (H), the relative permeability within the spot 12 increases rapidly. The maximum permeability within the spot 12 depends on the material properties of the film 11. With the magnetisable material used in this system, the maximum relative permeability is approximately 5000. As there is a rapid change of permeability of the film 11 within the spot 12, the spot 12 and the surrounding region of saturated film 11 will sometimes be referred to hereinafter as the in-homogeneity spot 12.
Sensor Design - 1
Figure 2 schematically illustrates the conductor tracks on the sensor PCB 8 that form the excitation winding 13, a "sin" sensor winding 14 and a "cos" sensor winding 15, all of which extend along the length of the PCB 8. The PCB 8 is a two layer PCB, with the conductor tracks on the upper layer (closest to the film 11) being shown in solid lines and with the conductor tracks on the lower layer (closest to the piston 4) being shown in dashed lines. The sensor windings 14 and 15 have a pitch (Lx) that approximately corresponds to the range over which the piston 4 can move within the cylinder 2. As shown in Figure 2, the ends of the excitation winding 13 and the ends of the sensor windings 14 and 15 are connected to the excitation and processing electronics 10 (which are also mounted on the PCB 8) via the connection interface 20.
As shown, the sin sensor winding 14 is formed from two turns of conductor that effectively define two sets of loops which are connected together in the opposite sense in a figure of eight configuration. As a result, EMFs induced in the first set of loops by a common background magnetic field will oppose the EMFs induced in the second set of loops by the same common background magnetic field. As those skilled in the art will appreciate, the cos sensor winding 15 is effectively formed by shifting the sin sensor winding 14 by a quarter of the pitch (Lx) along the X direction. As shown, the cos sensor winding 15 effectively defines three sets of loops, with the loops of the first and third sets being wound in the same direction as each other but opposite to the winding direction of the loops of the second (middle) set. As with the sin sensor winding 14, the sizes of the loops forming the cos sensor winding 15 are chosen so that (in the absence of the magnet 6) the EMFs induced in the first and third sets of loops by the excitation magnetic field cancel with the EMFs induced in the second set of loops by the excitation magnetic field.
The excitation winding 13 is wound around the outside of the sensor windings 14 and 15 and is arranged so that (in the absence of the magnet 6) when an AC excitation current is applied across its ends, an AC excitation magnetic field is generated which extends predominantly in the Z direction shown in Figure 1. This excitation magnetic field will be substantially symmetric along an axis which is parallel to the Y axis and which passes through the middle of the excitation winding. This axis of symmetry is also an axis of symmetry for the sensor windings 14 and 15. Therefore, as a result of the figure of eight arrangement of the sensor windings 14 and 15 and as a result of the common symmetry between the excitation winding 13 and the sensor windings 14 and 15, there is minimal direct inductive coupling between the excitation winding 13 and the sensor windings 14 and 15. However, as mentioned above, when the piston 4 is in the cylinder 2 the DC magnet 6 creates a positionally varying magnetic field that interacts with the film of magnetisable material to create an imbalance between the excitation winding 13 and the sensor windings 14 and 15. As a result of the geometrical variation of the sensor windings 14 and 15 along the X direction, the imbalance that is created by the magnet 6 and the film 11 depends upon the position of the magnet 6 along the X direction relative to the sensor windings 14 and 15. Therefore the signal generated in each sensor winding 14 and 15 will vary with the position of the piston 4 within the cylinder 2. As a result of the figure of eight configuration of the sensor windings 14 and 15, the variation of the signal generated in each sensor winding 14 and 15 is approximately sinusoidal. In fact, with an AC excitation signal, the signal generated in each sensor winding 14 and 15 is an AC signal at the same frequency as the excitation signal, whose peak amplitude varies approximately sinusoidally with the position of the piston 4 within the cylinder 2. The main signals generated in each sensor winding are at the same frequency as the excitation frequency because, the direction and magnitude of the AC excitation magnetic field (generated by the AC current flowing in the excitation winding 31) is such that it does not saturate a substantial area of the film 11 of magnetisable material in the vicinity of the sensor windings. If the excitation current is strong enough to enforce changing magnetisation of substantial areas of the film 11 at the excitation frequency, then it causes two unwanted effects. First, the measured signals will vary with the orientation of the magnet 6, such that reversing the magnetisation of the magnet 6 will change the readings of the sensor due to the inherent substantial magnetisation caused by the excitation currents. Secondly, in the presence of external magnetic fields, magnetisation and demagnetisation of the film 11 due to the excitation current becomes significantly non-linear. Thus substantial second harmonic generation will occur in the film reducing the signal to noise ratio of the desired signal (which is at the same frequency as the excitation frequency).
Figure 3A schematically illustrates the way in which the peak amplitude of the signal generated *in ther sin' sensor winding 14 (shown in Figure 3B) varies with the position of the piston 4 within the cylinder 2. As shown, the variation in the peak amplitude is approximately sinusoidal (except at the ends of the winding) and hence the sensor winding 14 is referred to as the "sin" sensor winding 14. The period of the sinusoidal variation corresponds approximately to the pitch (Lx) of the sensor winding 14.
Figure 4A schematically illustrates the way in which the peak amplitude of the signal induced in sensor winding 15 (shown in Figure 4B) varies with the position of the piston 4 within the cylinder 2. As shown, the variation is in phase quadrature (90° out of phase) to the variation obtained from sensor winding 14, which is why the sensor winding 15 is referred to as the "cos" sensor winding. As those skilled in the art will appreciate, the quadrature nature of the variation between the signals output from the sensor windings 14 and 15 is obtained because sensor winding 15 is effectively shifted along the X direction by a quarter of the pitch Lx relative to sensor winding 14.
As those skilled in the art will appreciate, the plots shown in Figures 3A and 4A are approximate in that the peak amplitudes of the output signals do not vary exactly sinusoidally with the position of the piston 4 within the cylinder 2. This is an approximation to the actual variation, which will depend upon edge effects, positions of via holes on the PCB 8 and other effects that introduce non-linearities into the system.
As those skilled in the art will appreciate, two sensor windings 14 and 15 are provided in 5 order to be able to uniquely encode the position of the piston 4 within the cylinder 2 along the entire pitch (Lx) of the sensor windings 14 and 15. However, if the measurement range of the piston 4 within the cylinder 2 is limited, for example to between the thirty percent and the seventy percent points on the plot shown in Figure 3 A, then only one sensor winding (winding 14) would be required to determine the absolute
10 position of the piston 4 within the cylinder 2. However, when the piston 4 can move over the entire pitch of the sensor windings 14 and 15, at least two sensor windings are required in order to overcome the phase ambiguity common with sinusoidal signals. For example, if the peak amplitude of the signal generated in sensor winding 14 is determined to be 0.5, then, as shown in Figure 3 A, the position may correspond to approximately ten
15 percent or approximately forty-two percent of the range of movement of the piston 4 within the cylinder 2. This ambiguity can be resolved using the peak amplitude of the signal generated in the cos sensor winding 15.
Figure 5 A illustrates the locus 27 obtained by plotting the peak amplitude of the EMF
20 induced in the sin sensor winding 14 against the peak amplitude of the EMF induced in the cos sensor winding 15 as the piston 4 moves from one end of the cylinder 2 to the other. As shown, the plot 27 is substantially circular except for when the piston 4 is at each end of the cylinder 2. Therefore, as shown by plot 28 in Figure 5B, except in these end regions, the phase angle φ varies linearly with the position of the piston 4 in the
25 cylinder 2. Consequently, as will be described in more detail below, the excitation and detection electronics 10 determines the position of the piston 4 within the cylinder 2 by taking an inverse arctangent function of the ratio of the peak amplitudes of the signals induced in sensor windings 14 and 15. The use of such a ratiometric calculation is preferred as it makes the system less sensitive to variations in the amplitude of the
30 excitation current and variations in the distance between the magnet 6 and the film 11 of
• -. -magnetisable material. .. ' . - - ,~ * v «~
Excitation And Detection Circuitry - 1
Figure 2 schematically illustrates the way in which the excitation winding 13 and the 35 sensor windings 14 and 15 are connected to the excitation and detection circuitry 10. As illustrated in Figure 2, the excitation winding 13 is connected to the excitation circuitry via terminals A and B of the connection interface 20 and the sensor windings 14 and 15 are connected to the detection circuitry at terminals 1, 2, 3 and 4 of the connection interface 20. 40
Figure 6 is a block diagram illustrating one form of the excitation circuitry 24 and the detection circuitry 26 that can be used. As shown, the excitation circuitry 24 includes a signal generator 30 which is operable to generate an AC drive voltage 31 which is applied to terminal A of the connection interface 20. The excitation circuitry also 45 includes an inverter 32 which inverts the AC voltage 31 to generate an inverted AC voltage 33, which is applied to terminal B of the connection interface 20. As a result, symmetrical sinusoidal excitation signals are applied to both ends of the excitation winding 13. The AC voltages 31 and 33 can have an amplitude in the range 0.1 V to 10V and a frequency between 10OkHz and 10MHz. The inventor has found that by driving both ends of the excitation winding 13 with symmetrical excitation voltages 31 and 33, the sensor is able to operate at much higher frequencies. In particular, the inventor has found that driving only one end of the excitation winding 13 with an excitation voltage results in a loss of performance at about 500kHz. This is because, at higher frequencies and without symmetrical drive voltages, current flowing in the excitation winding 13 capacitively couples into the film 11 of magnetisable material, which changes the current distribution in the sensor windings, which in turn affects the operation of the sensor.
However, as mentioned above, by applying symmetrical drive voltages 31 and 33 to both ends of the excitation winding 13 the sensor is able to operate at frequencies up to approximately 10MHz. The advantage of operating at higher frequencies is that greater signal levels can be obtained from the sensor windings than when operating at lower frequencies.
As shown in Figure 3, the signals obtained at terminals 1 and 2 of the connection interface 20 (which are connected to the cos sensor winding 15) are applied to a first differential amplifier 34" and the signals obtained from terminals 3 and 4 of the connection interface 20 (which are connected to the sin sensor winding 14) are applied to a second differential amplifier 34~2. The differential amplifiers amplify the signals obtained from the sensor windings and remove any common mode signals that might arise. As discussed above, the signals induced in the sensor windings 14 and 15 are AC signals at the same frequency as the excitation signal and having a peak amplitude that varies sinusoidally and cosinusoidally with the position of the piston 4 within the cylinder 2. Therefore, the signals output by the differential amplifiers 34" and 34" can be represented (approximated) by the following equations:
Figure imgf000012_0001
where A is an unknown amplitude term that depends on the sensor design, d is the position of the in-homogeneity spot 12 along the length (Lx) of the sensor board (corresponding to the position of the piston 4 in the cylinder 2); and f is the excitation frequency of the AC signals 31 and 33.
As shown in Figure 6, the signals output from the differential amplifiers 34 are applied to a respective mixer 36"1 and 36"2 where they are mixed with the excitation signal 31. The outputs from the mixers 36 are then input to a respective low pass filter 38"1 and 38"2 which operate to remove the high frequency products of the mixing and to leave the following DC values which vary with the position of the in-homogeneity spot 12:
2nd
^cos = Λ cos (3)
Figure imgf000013_0001
As shown, the signals output from the low pass filters 38 are applied to a ratiometric calculator 40 which determines the position (d) of the in-homogeneity spot 12 using the following arctangent function:
d = arctan " sin X
2π (5)
The above calculation may be performed using a microprocessor and the voltage levels (Vcos and Vsin) obtained from the low pass filters 38. However, in this case, the ratiometric calculator 40 uses analogue processing techniques to generate a value that continuously varies with the value of d. Figure 7 is a block diagram illustrating the way in which the ratiometric calculator 40 performs the calculation. As shown, the ratiometric calculator 40 includes a low frequency generator 48 that generates a low frequency (of the order of a few kHz) periodic signal. This low frequency periodic signal is applied directly to a first amplitude modulator 50"1 and a 90 degree phase shifted version of the low frequency periodic signal is applied to a second amplitude modulator 50"2. As shown in Figure 7, the 90 degree phase shift is obtained by passing the low frequency periodic signal though a 90 degree phase shifter 52. The amplitude modulator 50" amplitude modulates the received low frequency periodic signal with the voltage (Vcos) obtained from the low pass filter 38"1; and the amplitude modulator 50"2 amplitude modulates the phase shifted low frequency periodic signal with the voltage (Vsin) obtained from the low pass filter 38~ . The two amplitude modulated signals output from the amplitude modulators 50 are then added together by an adder 54. The output from the adder 54 can therefore be represented (approximated) by:
Fcos sin 2^" 2t + Fsin cos 2^2t (6)
where f2 is the frequency of the low frequency periodic signal, which can be rewritten as:
Figure imgf000013_0002
As can be seen from equation 7 above, the output from the adder 54 includes a sinusoidal signal at the low frequency (f2) whose phase is proportional to the ratiometric arctangent function defined in equation 5 given above. Therefore, the ratiometric calculator 40 calculates equation 5 by using a phase detector 56 to detect the phase of this signal output from the adder 54 relative to the low frequency signal generated by the low frequency generator 48. The output of the phase detector 56 will therefore be a value that continuously changes with the position of the in-homogeneity spot 12 relative to the PCB
8. Figure 8 is a block diagram illustrating in more detail the components of the phase detector 56 used in this design. As shown, the AC signal 55 from the adder 54 is input to a first comparator 57"1 where it is compared with a threshold voltage. Similarly, the AC signal 58 from the low frequency generator 48 is input to a second comparator 57"2 where it is also compared with the same threshold value. As shown, the square wave signals 59"1 and 59~2 output by the comparators 57 are shifted in time relative to each other by a value Δt corresponding to the phase shift between AC signal 55 and AC signal 58. These two square wave signals are then input to a logic circuit 63 which generates a pulse width modulated sequence of pulses 65 whose pulse width corresponds to the timing difference Δt. As shown in Figure 8, this pulse width modulated signal 65 is input to a PWM/DC converter 67 which converts this pulse train into a DC value which monotonically varies with the value of Δt and hence with the phase shift between the AC signal 55 and the AC signal 58.
Returning now to Figure 6, as shown, the output from the phase detector 56 is input to a level signal display 44, which uses the DC value output by the phase detector 56 to generate a display corresponding to the position of the piston 4 within the cylinder 2.
Interfering Magnetic Fields As those skilled in the art will appreciate, one of the advantages of the sensor design described above is that it is insensitive to constant (or homogonous) magnetic fields in its vicinity. However, a problem arises with the sensor design described above if there is a second magnet that moves in the vicinity of the sensor PCB 8. Such a situation might arise where, for example, a plurality of piston and cylinder assemblies are mounted side by side, with each having a DC magnet mounted within the piston of each assembly.
Figure 9, which comprises Figures 9A and 9B illustrate this problem. In particular, Figure 9A shows the piston and cylinder assembly of Figure 1 C and a second magnet 73 which can also move relative to the sensor PCB 8. If the second magnet 73 is close enough to the film 11 of magnetic material on the sensor PCB 8, then the magnetic field generated by the second magnet' 73 will also create an in-homogeneity spot in the film 11 which will disturb the balance between the excitation and sensor windings carried by the sensor PCB 8. As the second magnet 73 is likely to be much further away from the film 11 of magnetisable material than the magnet 6 (typically upto 10 times further away), the disturbance caused by the magnet 73 will be significantly smaller than the disturbance caused by the magnet 6. This is illustrated in the locus plot 27 shown in Figure 9B which is similar to the plot shown in Figure 5A, but which includes a disturbance represented by the curve 75 caused by the magnet 73 which can introduce an error (Δφ) into the measured phase angle.
Sensor Design - 2
Figure 10 schematically illustrates the form of the excitation and sensor windings used in a second design of the sensor board 8 which is designed to mitigate the problem of interfering magnetic fields discussed above. In a piston and cylinder assembly, it is usually only important to know when the piston 4 is approaching either end of the cylinder bore 3. Therefore, in this alternative design, excitation and sensor windings are provided in the regions corresponding to each end of the cylinder 2. In particular, excitation winding 13~2 and sensor windings 14"2 and 15"2 are positioned at one end of the cylinder 2 and a second excitation winding 13"3 and a second pair of sensor windings 14"3 and 15"3 are provided at the other end of the cylinder 2. As illustrated in Figure 10, the two excitation windings 13"2 and 13"3 can be connected in series so that no modifications have to be made to the excitation circuitry 24. However, as illustrated in Figure 10, the two sets of sensor windings 14 and 15 are connected to respective detection circuitry 26"2 and 26"3. These detection circuits 26 may operate independently and signal an appropriate controller (not shown) when they detect the presence (and position) of the piston 4. Alternatively, as the piston 4 can only be in one place at any one time, the signals from these two sets of sensor windings 14 and 15 may be multiplexed through a common detection channel, as illustrated in Figure 11. As shown, the excitation circuitry 24 is the same as the excitation circuitry shown in Figure 6 and the detection circuitry 26 is similar to that shown in Figure 6, except that multiplexing switches 76"1 and 76" are provided which are arranged either to pass the signals from sensor windings 14~2 and 15~2 to the differential amplifiers 34 or to pass the signals obtained from sensor windings 14~3 and 15~3 to the differential amplifiers 34. As shown in Figure 11, in this system, an algorithm controller 78 is used to control the switching of the multiplexing switches 76 and, based on the measurements obtained from the two sets of sensor windings 14 and 15, to determine the approximate position of the piston 4 within the cylinder 2 for output on the position signal display 44. The remaining components of the detection circuitry 26 are the same as used in the circuit shown in Figure 6 and will not, therefore, be described again.
Figure 12A illustrates the way in which the phase angle (φ) calculated by the ratiometric calculator 40 varies with the position of the piston 4 in the cylinder 2, when the signals from the two sets of sensor windings 14 and 15 are processed. In particular, plot 28~2 illustrates the way in which the calculated phase angle varies with the position of the piston 4 based on the measurements obtained from sensor windings 14" and 15" ; and plot 28" illustrates the way in which the calculated phase angle varies with the position of the piston 4 based on the measurements obtained from sensor windings 14"3 and 15"3. As illustrated in Figure 12A, when the piston 4 is located within the range L2 (which corresponds to the measurement, range of the sensor windings 14"2 and 15"2) signals that vary with the position of the piston 4 are obtained from sensor windings 14"2 and 15"2. Similarly, when the piston 4 is within the range L3 (corresponding to the measurement range of the sensor windings 14"3 and 15"3) signals that vary with the position of the piston 4 are obtained from sensor windings 14"3 and 15"3. However, when the piston 4 is located between the measurement ranges L2 and L3, the signals obtained from the two sets of sensor windings do not vary with the position of the piston 4.
Figure 12B is a flow chart illustrating the algorithm performed by the algorithm controller 78 to calculate the position of the piston 4. As shown, in step s2, the algorithm controller 78 controls the multiplexing switches 76 so that the signals from sensor windings 14" and 15" are passed through to the differential amplifiers 34. Then, in step s4, the algorithm controller 78 determines a first position measurement (x) using the phase angle obtained from the ratiometric calculator 40 in response to the processing of the signals obtained from sensor windings 14"2 and 15"2. In step s6, the algorithm controller 78 determines if the thus calculated position (x) is less than L2. If it is, then the algorithm controller 78 determines the output position measurement (x ) as the thus determined position (x). If at step s6 the algorithm controller 78 determines that the first position measurement (x) is not less than L2 then the processing procedds to step s8, where the algorithm controller 78 controls the multiplexing switches 76 so that the signals from sensor windings 14" and 15" are passed through to the differential amplifiers 34. Then, in step slO, the algorithm controller 78 determines a second position measurement (x ) using the phase angle obtained from the ratiometric calculator 40 in response to the processing of the signals obtained from sensor windings 14"3 and 15"3. In step sl2, the algorithm controller 78 determines if the thus calculated position (x ) is greater than zero (in which case the piston 4 is within the operating range L3). If it is, then the algorithm controller 78 determines the output position measurement (x ) as being the second position measurement (x ) plus (Li-L3). Otherwise the piston 4 is somewhere between the two ends and outputs the position (x ) as being halfway along the length of the cylinder 2.
Sensor Design - 3 The problem with the design illustrated in Figure 11 is that no position information is provided when the piston 4 is in a central area of the cylinder 2. Figure 13 schematically illustrates the windings of a third design of the sensor PCB 8 which is formed using a combination of the windings used in the first design and those used in the second design.
As shown, a single excitation winding 13 is provided together with three sets of sin and cos sensor windings (14"1 and 15"1; 14"2 and 15~2; 14~3 and 15"3). As illustrated in Figure 13, the three sets of sin and c nsor windings 14 and 15 may each be connected to respective detection circuitry
Figure imgf000016_0001
26"2 and 26"3. Alternatively, and as illustrated in Figure 14, the signals from the three sets of sensor windings 14 and 15 may be multiplexed into a common detection channel using multiplexing switches 76"1, 76" and 76" . As in the second sensor design, an algorithm controller 78 is provided to control the switching of the multiplexing switches 76 and to determine the position of the piston 4 based on the measurements obtained from the ratiometric calculator 40, which position information it outputs for display on the position signal display 44.
As those skilled in the art will appreciate, with this design, when the piston 4 is located at either, end of the cylinder 2, signals that vary with the position of the "ptstόif "4" will be obtained from two of the sets of sensor windings. Therefore, in this embodiment, the algorithm controller 78 determines the actual position of the piston 4 using a weighted combination of the positions calculated from these two sets of sensor measurements. The way in which the algorithm controller 78 performs this weighted combination will now be explained with reference to Figure 15.
Figure 15A illustrates the way in which the phase angle (φ) calculated by the ratiometric calculator 40 varies with the position of the piston 4 in the cylinder 2, when the signals from the three sets of sensor windings 14 and 15 are processed. In particular, plot 28"1 illustrates the way in which the calculated phase angle varies with the position of the piston 4 based on the measurements obtained from sensor windings 14" and 15"1; plot 28"2 illustrates the way in which the calculated phase angle varies with the position of the piston 4 based on the measurements obtained from sensor windings 14"2 and 15"2; and plot 28"3 illustrates the way in which the calculated phase angle varies with the position of the piston 4 based on the measurements obtained from sensor windings 14"3 and 15"3. As illustrated in Figure 15 A, when the piston 4 is located within the range L2 (which corresponds to the measurement range of the sensor windings 14~2 and 15"2) signals that vary with the position of the piston 4 are obtained from sensor windings 14"1 and 15"1 and from sensor windings 14"2 and 15"2. Similarly, when the piston 4 is within the range L3 (corresponding to the measurement range of the sensor windings 14"3 and 15~3) signals that vary with the position of the piston 4 are obtained from sensor windings 14"1 and 15"1 and from the sensor windings 14"3 and 15~3. However, when the piston 4 is located between the measurement ranges L2 and L3, only the signals obtained from sensor windings 14"1 and 15"2 vary with the position of the piston 4.
Therefore, in this embodiment, when the piston 4 is within the operating range L2, the algorithm controller 78 performs a weighted combination of the measurements obtained from plots 28" and 28" . Similarly, when the piston 4 is within the measurement range L3, the algorithm controller 78 calculates a weighted combination of the measurements obtained from plots 28"1 and 28" . The particular weighting functions used in this embodiment are illustrated graphically in Figure 15B.
Figurel5C is a flow chart illustrating the algorithm performed by the algorithm controller 78 to calculate the position of the piston 4. As shown, in step si, the algorithm controller 78 controls the multiplexing switches 76 so that the signals from sensor windings 14"1 and 15"1 are passed through to the differential amplifiers 34. Then, in step s3, the algorithm controller 78 determines a first position measurement (x) using the phase angle obtained from the ratiometric calculator 40 in response to the processing of the signals obtained from sensor windings 14"1 and 15"1. In step s5, the algorithm controller 78 determines if the thus calculated position (x) is less than L2 plus a threshold (Δ). If it is, then the piston 4 may be within the operating range L2 and therefore, the algorithm controller 78 controls the multiplexing switches 76 so that the signals obtained from sensor windings 14"2 and 15"2 are passed through to the differential amplifiers 34. In step s8, the algorithm controller 78 determines a new position measurement (x ) using the phase measurements obtained by the ratiometric calculator 40 using the signals from sensor windings 14"2 and 15"2. In step si 1, the algorithm controller 78 subtracts the measurement range L2 from the thus obtained position measurements so that they can be applied to the weighting function illustrated in Figure "15B. Finally, in step S 13, the algorithm controller 78 determines the position of the piston (x ) using the following weighted combination:
Figure imgf000017_0001
If at step s5, the algorithm controller 78 determines that the determined position (x) is greater than L2 plus the threshold (Δ) then the processing proceeds to step si 5 where the algorithm controller 78 determines if the position (x) is greater than (Li-L3)-Δ. If it is, then the piston 4 may be within the operating range L3 and therefore, the processing proceeds to steps sl7, sl9 and s21 where a similar processing to steps s7, s9 and sll is performed but in respect of measurements obtained from sensor windings 14"3 and 15~3. The algorithm controller 78 then calculates in step sl3 the weighted combination of the determined measurements as before. If at step sl5, the algorithm controller 78 determines that the determined position (x) is not greater than (Li -L3)-Δ, then the piston 4 is somewhere in the middle of the cylinder 2 and therefore the algorithm controller 78 uses the determined position (x) as the output 5 position (x**) which it outputs to the position signal display 44.
Sensor Design - 4
Figure 16 (which comprises Figures 16A, 16B and 16C) illustrates an alternative sensor design that mitigates the problem associated with interfering magnetic fields from0 adjacent magnets. In particular, as illustrated in Figure 16 A, in this sensor design, two sensor boards (transducers) 8"1 and 8"2 are provided which are spaced apart from each other relative to the piston 4 using a spacer block 80. Typically the film of magnetisable material carried by sensor board 8" will be between zero and 20mm from the magnet 6 and the film of magnetisable material carried by sensor board 8" will be between 2mm5 and 25mm from the magnet 6. In this design, the excitation and sensor windings (not shown) carried by the two sensor boards 8" and 8" are arranged to generate signals which both vary in the same way with the position of the piston 4 within the cylinder 2. Since the sensor board 8"1 is located further away from the magnet 6 than the sensor board 8"2, the signals obtained from sensor board 8" will be smaller in magnitude than the signals obtained from sensor board 8"2. However, as the interfering magnet 73 is located at a much greater distance from the sensor boards 8 (than the magnet 6 in the piston 4), it will have substantially the same interfering effect on each of the sensor boards 8. Consequently, by subtracting the signals obtained from the two sensor boards 8, signals can be obtained which vary with the position of the piston 4 within the cylinder 2 and in which the interfering signal caused by the magnet 73 is cancelled (or at least significantly reduced). This is illustrated by the locus plots shown in Figures 16B and 16C.
In particular, Figure 16B illustrates the locus 27~2 of points obtained by plotting the measured values VSm and VCOs obtained from the sensor windings mounted on sensor board 8"2 and similarly the locus 27"1 of points obtained from the sensor windings
,-r -•■ mounted on the sensor board 8"',..,.As shown, the locus of points 27"2 has a larger diameter < than the locus of points 27"1 because the signals obtained from the sensor windings mounted on the sensor board 8" will be much larger than the signals obtained from the sensor windings on sensor board 8" . Figure 16B also illustrates by the curves 75" and 75" the effect that the interfering magnet 73 can have on the signals obtained from the respective sensor boards. Figure 16C illustrates the locus of points 27" obtained by subtracting the signals obtained from the two sensor boards 8"1 and 8"2. As shown, although the locus of points 27"3 is smaller in magnitude than the locus 27"1, the interfering effect caused by the magnet 73 results in a much smaller possible phase error (Δφ3) than the possible phase errors on the individual measurements from the two sensor boards 8"1 and 8"2 (i.e. Δ φi and Δ φ2).
As those skilled in the art will appreciate, the electronics 10 used in the sensor design shown in Figure 16 may be arranged to perform the subtraction of the signals obtained from the two sensor boards. However, performing the subtraction in the electronics
. simply increases the cost associated with the electronics 10. Therefore, in the preferred design, the excitation and sensor windings mounted on the two sensor boards 8"1 and 8"2 are preferably connected together in such a way that the signals from the two sin sensor windings 14 are subtracted from each other and the signals from the two cos sensor windings 15 are subtracted from each other. Figure 17 illustrated one way in which this can be achieved. In particular, Figure 17 schematically illustrates the excitation winding 13"1 and the sensor windings 14"1 and 15"1 mounted on the sensor board 8"1 and the excitation winding 13"2 and the sensor windings 14"2 and 15'2 mounted on the sensor board 8"2. As shown in Figure 17, the excitation and sensor windings mounted on the two sensor boards 8"1 and 8"2 are connected together on the spacer block 80. As can be seen from the direction of the arrows on the windings (which represent a notional direction of current flow), sensor winding 14"1 is connected in series with sensor winding 14~2 so that the signals induced in the two sensor windings are added together. Similarly, sensor winding 15"1 is connected in series with sensor winding 15" so that the signals induced in the two sensor windings add together. However, the excitation windings 13"1 and 13~2 are connected in series so that the direction of current flow in the two excitation windings 13 is in the opposite direction. As a result, the AC magnetic field generated by the excitation winding 13"1 will be of opposite polarity to the AC excitation magnetic field generated by the excitation winding 13" . As a result, the signals induced in the sensor windings 14"1 and 15"1 will be of opposite polarity to the signals induced in sensor windings 14"2 and 15"2. Consequently, adding the signals induced in sensor windings 15"1 and 15" and adding the signals induced in sensor windings 14" and 14"2 has the desired result of subtracting (or differencing) the signals from these two pairs of sensor windings.
Figure 18 illustrates an alternative way in which the same result can be achieved. In this case, the excitation windings 13"1 and 13"2 are connected in series so that the excitation magnetic fields generated by these windings will have the same polarity, but the respective pairs of sensor windings (14"1 and 14"2; and 15"1 and 15" ) are connected together so that the current flow is in the opposite direction from each other. As a result the signals induced in sensor winding 14" will subtract from the signals induced in sensor winding 14"2 and the signals induced in sensor winding 15"1 will subtract from the signals induced in sensor winding 15" .
As those skilled in the, art will appreciate, the excitation and detection circuitry illustrated in Figure 6 can be used to energise the excitation windings 13 and to process the signals obtained from the sensor windings 14 and 15 of the sensor designs illustrated in Figure 17 and 18.
A description will now be given of various different types of excitation and detection circuits that can be used with the first and fourth sensor designs described above. As those skilled in the art will appreciate, the following excitation and detection circuits could also be modified so that they can operate with the second and third sensor designs described above. Such modification may simply involve the addition of suitable multiplexing circuitry between the detection circuit 26 and the connection interface 20 and the provision of a suitable controller for controlling the multiplexing of the signals into the detection circuit.
Excitation And Detection Circuitiγ - 2
Figure 19 is a block diagram illustrating alternative excitation and detection circuitry that can be used. Components that are the same as those used in the circuitry shown in Figure 6 have been given the same reference numerals. As described above, in the circuit shown in Figure 6, the AC signals output from the differential amplifiers 34 are demodulated by mixers 36 and then remodulated onto a low frequency carrier in the ratiometric calculator 40. In the alternative circuitry illustrated in Figure 19, this demodulation arid then remodulation is omitted. Instead, the AC signal output from the differential amplifier 34"2 is applied to a 90 degree phase shifter 52. The phase shifted signal is then added with the AC signal output from the differential amplifier 34'1 in the adder 54. The output from the adder 54 can therefore be represented by equation 7 above, except replacing the low frequency value ϊ% with the frequency of the excitation signals 31 and 33. Consequently, the arctangent function defined above in equation 5 can be evaluated by measuring the phase of the signal output from the adder 54 relative to the phase of the excitation signal 31. As shown in Figure 19, the phase detector 56 makes this phase measurement. The remaining components of the excitation and detection circuitry shown in Figure 19 are the same as those shown in Figure 6 and will not, therefore, be described again.
Although it appears that the circuitry shown in Figure 19 is a simplified version of the circuitry shown in Figure 6, the circuitry shown in Figure 19 is not actually preferred as it requires the phase shifter 52, the adder 54 and the phase detector 56 to operate at the higher excitation frequency rather than at the lower frequency of the low frequency generator 48. As a result, more complex and more expensive circuitry has to be used for these components.
Excitation And Detection Circuitry - 3
Figure 20 is a block diagram illustrating alternative excitation circuitry 24 and detection circuitry 26 that can be used. As discussed above, the sensor can operate over a range of frequencies. However, instead of using AC excitation signals 31 and 33, it is possible to use a pulse generator 60 that generates square form pulses of excitation signal (having a pulse width of the order of several hundred nanoseconds) with sharp rising and falling fronts (having widths of the order of several nanoseconds). Fast fronts of the excitation pulses correspond to the equivalent frequency of tens of MHz. The pulse repetition frequency that is used is less important as it only affectsrihe overall signal levels that will be processed by the detection circuitry 26. Therefore, in order to minimise power consumption of the sensor, the pulse repetition frequency should be set to the minimum necessary to obtain the required level of signal to noise ratio for the measurements to be made.
As shown in Figure 20, the pulsed excitation signal 31 generated by the pulse generator 60 is applied to terminal A of the connection interface 20 and is applied to an inverter 32 for generating an inverted version 33 of the pulsed excitation signal which is applied to terminal B of the connection interface 20.
As shown in Figure 20, terminals 1 and 2 of the connection interface 20 are connected to a reference capacitor 64"1 through switches 62"1 and 62"2, which are opened and closed in synchronism with the pulses of the excitation signal 31 and connected so that the signal induced in the sensor winding connected to terminals 1 and 2 is alternately connected to opposite plates of the reference capacitor 64" . Similarly, terminals 3 and 4 of the connection interface 20 are connected to a second reference capacitor 64"2 through switches 62"3 and 62 , which are also switched in synchronism with the pulses of the excitation signal 31. As a result, voltages are accumulated on the reference capacitors 64 which will vary slowly over time reflecting the movement of the in-homogeneity spot 12 along the length of the sensor PCB 8. The switches 62 therefore act to demodulate the signals received from the sensor windings so that the above described DC voltages (VCOs and VSin) accumulate on the reference capacitors 64. As shown, the voltages accumulated on the reference capacitors 64 are amplified by differential amplifiers 66Λ and 66~2 respectively and then applied to the ratiometric calculator 40 as before for processing in the same way. The reference capacitors 64 and the amplifiers 66 together form respective low pass filters 68"1 and 68'2 which remove any high frequency components caused by the switching of the switches 62. As a result of the use of the pulsed excitation signals 31 and 33 and the use of the switches 62, the detection circuitry 26 can use relatively cheap low frequency electronic components to amplify and process the signals obtained from the sensor windings. Excitation And Detection Circuitiγ - 4
In the systems described above, the excitation signals were applied to the two ends of the excitation winding 13 shown in Figure 2 and the signals obtained from the sin and cos sensor windings 14 and 15 were connected to the detection circuitry 26. In an alternative sensor design the roles of these windings can be reversed so that the sin and cos windings 14 and 15 are connected to the excitation circuitry 24 and the winding 13 is connected to the detection circuitry 26. This is illustrated schematically in the electrical equivalent circuit shown in Figure 21.
Figure 22 is a block diagram illustrating the excitation circuitry 24 and the detection circuitry 26 that can be used in such an "inverted" system. As shown, the excitation circuitry 24 includes two signal generators 30"1 and 30"2, which generate AC excitation signals. As illustrated in Figure 22, the AC excitation signal 31"1 generated by the generator 30" has a lower frequency than the frequency of the AC excitation signal 31"2 generated by the generator 30" . The excitation circuitry 24 also includes two inverters 32"1 and 32"2, which invert the AC excitation signals 31 output from the respective generators 30. -As . shown -in-rFigure 2-2,,~the AC excitation signal 31"1 is applied to terminal 1 of the connection interface 20 and the inverted excitation signal 33"1 is applied to terminal 2 of the connection interface 20. Similarly, the higher frequency excitation signal 31"2 is applied to terminal 3 of the connection interface 20 and the inverted excitation signal 33"2 is applied to terminal 4 of the connection interface 20. hi this sensor design, the connections made in the connection interface 20 are the same as the connections made in the connection interface 20 illustrated in Figure 2. Therefore, the lower frequency excitation signals are applied to the cos winding 15 and the higher frequency excitation signals are applied to the sin winding 14.
With regard to the detection circuitry 26, as shown in Figure 22, this is connected to terminals A and B of the connection interface 20. Therefore, signals induced in winding 13 are received at terminals A and B. As shown in Figure 22, the signals from these terminals are applied to a differential amplifier 34, which subtracts the signals obtained at these two terminals to remove any common mode signal induced in the winding 13. As before, as a result of the symmetrical arrangement of the windings 13, 14 and 15 substantially no signals will be output from the differential amplifier 34 in the absence of an in-homogeneity spot 12 in the film 11 of the sensor. However, when such an in- homogeneity spot 12 exists, signals will be obtained from the differential amplifier 34, which vary with the position of the in-homogeneity spot 12 along the length of the PCB
8. As a result of the application of two different excitation signals (at different excitation frequencies), one to each of the two windings 14 and 15, the signal output from the differential amplifier 34 can be represented (approximated) by:
Vu 34 = A cos cos 2πf2t (9)
Figure imgf000022_0001
where fi is the frequency of the AC signal generated by the generator 30"1 and f2 is the frequency of the AC signal generated by the generator 30"2.
As shown in Figure 22, in this embodiment, the signal output from the differential amplifier 34 is applied to two mixers 36" and 36" where it is multiplied by a respective one of the excitation signals 31" and 31" . The outputs from the mixers 36 are then filtered by a respective low pass filter 38"1 and 38"2 to generate the DC signal values given above in equations 3 and 4, which are applied to the ratiometric calculator 40 as before.
The remaining processing of the detection circuitry 26 is the same as in the circuitry described with reference to Figure 6 and will not, therefore, be described again.
Excitation And Detection Circuitry - 5
In the sensor design described with reference to Figure 22, the sensor effectively employs frequency division multiplexing techniques to allow the detection circuitry to be able to separate out the sine and cosine signals. An alternative technique, which uses time division multiplexing to allow the detection circuitry to separate the sine and cosine signals, will now be described. Figure 23 is a block diagram illustrating the form of the excitation and detection circuitry that is used in this alternative sensor design. As shown, in this ,embqdimenty,.a single pulse generator 60 is provided that generates a train -of * voltage pulses 61, each pulse having a similar pulse shape and duration as the pulses generated by the pulse generator 60 described with reference to Figure 20. As shown in Figure 23, the pulse train 61 is applied to switches 81"1 and 81"2 and to a counter 83 which is used to control the switches 81 to generate time shifted pulse trains 31" and 31" . As shown, pulse train 31"1 is applied to an inverter 32"1 to generate an inverted pulse train 33"1 and the pulse train 31" is applied to an inverter 32"2 to generate the inverted pulse train 33"2. As shown in Figure 23, pulse train 31"1 is applied to terminal 1 of the connection interface 20; pulse train 33"1 is applied to terminal 2 of the connection interface 20; pulse train 31"2 is applied to terminal 3 of the connection interface 20; and pulse train 33"2 is applied to terminal 4 of the connection interface 20.
As shown in Figure 23, terminal A of the connection interface 20 is connected to switch 62"1 and to switch 62"3 and terminal B of the connection interface 20 is connected to switch 62"2 and switch 62 . Switches 62"1 and 62" are opened and closed in synchronism with the pulses of the pulse train 31"1 and the switches 62" and 62 are opened and closed in synchronism with the pulses of the time shifted pulse train 31"2. Therefore, during the times at which pulses of the excitation voltage are applied to terminals 1 and 2 of the connection interface 20, the signals induced in the winding attached to terminals A and B of the connection interface 20 are applied to the differential amplifier 34"1 and then to the mixer 36"1. The mixer 36"1 operates to demodulate the received signals, which are low pass filtered to regenerate the above described DC voltage VCOs. Similarly, during the 5 times at which voltage pulses are applied to terminals 3 and 4 of the connection interface 20, the signals induced in the sensor winding connected to terminals A and B of the connection interface 20 are passed through the differential amplifier 34"2 to the mixer 36"2. The mixer 36"2 operates to demodulate the received signals by multiplying them with the excitation signal 31" and the results are then low pass filtered by the low pass 10 filter 38"2 to generate the above described DC voltage VSin. The remaining processing of the detection circuitry 26 is the same as the processing performed in the circuitry described with reference to Figure 6 and will not, therefore, be described again.
15 Excitation And Detection Circuitry - 6
Figure 24 is a block diagram illustrating alternative excitation and detection circuitry 10 that can be used when the windings shown in Figure 2 are connected in the inverted manner illustrated in Figure 21. As shown, in this embodiment, the excitation circuitry includes a signal generator 30, which generates an AC excitation signal 31"1. A first
20 inverter 32"1 is provided for providing an inverted excitation signal 33"1. The excitation circuitry 24 also includes a 90 degree phase shifter 52, which applies a 90 degree phase shift to the excitation signal 31"1 to generate a phase shifted excitation signal 31"2. This phase shifted excitation signal 31" is also applied to an inverter 32"2 to generate an inverted and phase shifted excitation signal 33" . As shown in Figure 24, the excitation
25 signal 31"1 is applied to terminal 1 of the connection interface 20; the inverted excitation signal 33"1 is applied to terminal 2 of the connection interface 20; the phase shifted excitation signal 31"2 is applied to terminal 3 of the connection interface 20; and the inverted and phase shifted excitation signal 33"2 is applied to terminal 4 of the connection interface 20. The excitation circuitry 24 illustrated in Figure 24 is therefore equivalent to
30 the excitation circuitry shown in Figure 23, except the pulsed excitation signals are ■":<■*"" replaced with AC excitation signals that are applied at the same time to the "excitation"" windings.
As shown in Figure 24, a differential amplifier 34 of the detection circuitry 26 is 35 connected to terminals A and B of the connection interface 20. As a result of the use of the common receive winding 13, the output from the differential amplifier 34 can be represented by equations 6 and 7 given above, hi other words the output from the differential amplifier 34 includes a sinusoidal signal whose phase varies with the ratiometric arctangent function given in equation 5 above. In this embodiment the 40 output from the differential amplifier 34 is applied to a phase detector 56, which measures the phase of this signal relative to the phase of the AC excitation signal 31"1. The remaining processing carried out by the detection circuitry 26 is the same as before and will not, therefore, be described again.
45 Excitation And Detection Circuitry - 7
Figure 25 is a block diagram illustrating a further example of the excitation and detection circuitry that can be used when the windings shown in Figure 2 are connected in the inverted manner illustrated in Figure 21. In this case pulse width modulated pulse trains are used as excitation signals. As shown, the excitation circuitry 24 includes a pulse generator 60 that is arranged to generate a high frequency periodic pulse train 61, the pulses of which have relatively fast rising and falling fronts (of the order of several nanoseconds) and a relatively long pulse period (of the order of several hundred nanoseconds). The excitation circuitry 24 also includes a low frequency generator 48 that is arranged to generate a low frequency modulation signal 49 having a frequency of the order of several kHz. As shown in Figure 25, the pulse train 61 generated by the pulse generator 60 and the low frequency modulation signal 49 are applied to a modulator 82"1 which uses the low frequency modulation signal 49 to pulse width modulate the pulse train 61, to generate a pulse width modulated pulse train 84. As shown, the pulse width modulated pulse train 84 is applied to terminal 1 of the connection interface 20 and is also applied to an inverter 32" . The inverted pulse train generated by the inverter 32"1 is then applied to terminal 2 of the connection interface 20.
As shown in Figure 25, the low frequency modulation signal 49 generated by the low frequency generator 48 is also applied to a 90 degree phase shifter 52 that applies a 90 degree phase shift to the modulation signal 49. As shown in Figure 25, the 90 degree phase shifted version of the modulation signal 49 is then applied to a pulse width modulator 82"2 where it is used to pulse width modulate the pulse train 61 generated by the pulse generator 60. Figure 25 illustrates the pulse width modulated signal 86 output from the modulator 82"2. As shown, the pulse width modulated signal 86 is applied to terminal 3 of the connection interface 20 and is also applied to a second inverter 32~2. The inverted version of the pulse width modulated signal 86 is then applied to terminal 4 of the connection interface 20.
As shown in Figure 25, terminals A and B of the connection interface 20 are connected to switches 62"1 and 62~2 which are opened and closed in synchronism with the pulses of the pulse train 61. The switches 62 operate to down convert the high frequency component of the excitation signals (the component corresponding to the high frequency pulse train 61) to leave the lower frequency modulation signal component which is itself modulated by Vcos and VSin as before. As shown in.Figure 25,,the.,signals-passmg-from the switches 62 are filtered by the low pass filter 68 (formed by the reference capacitor 64 and the differential amplifier 66) to remove high frequency components separated by the switching action of the switches 62, to leave a signal of the form given above in equation 7, whose frequency corresponds to that of the low frequency modulation signal 49 and whose phase varies with the ratiometric arctangent function defined in equation 5 above. As shown in Figure 25, this sinusoidal signal output from the low pass filter 68 is applied to a phase detector 56 which measures the phase of the sinusoidal signal relative to the phase of the low frequency modulation signal 49 generated by the low frequency generator 48. The remaining processing of the detection circuitry 26 is the same as before and will not, therefore, be described again.
Excitation And Detection Circuitry - 8
Figure 26 is a block diagram illustrating a further alternative arrangement of the excitation and detection circuitry that can be used with the windings shown in Figure 2 when connected in the inverted manner illustrated in Figure 21. In this case, the excitation circuitry 24 is similar to the excitation circuitry 24 used in the circuitry shown in Figure 25, except that the low frequency modulation signal 49 is used to amplitude modulate the pulses of the pulse train 61 generated by the pulse generator 60. This is achieved by mixing the low frequency modulation signal 49 with the pulse train 61 using the mixer 36'1 and by modulating the 90 degree phase shifted version of the modulation signal with the pulse train 61 using the mixer 36' . The remaining components of the excitation circuitry 24 are the same and will not, therefore, be described again.
As shown in Figure 26, the detection circuitry 26 is the same as the detection circuitry 26 shown in Figure 25. Therefore, a further description of the detection circuitry 26 will not be given.
Excitation And Detection Circuitry - 9
Figure 27 is a block diagram illustrating a further alternative arrangement of the excitation and detection circuitry that can be used with the electrodes shown in Figure 2 when connected in the inverted manner illustrated in Figure 21. The circuitry shown in Figure 27 is substantially the same as the circuitry shown in Figure 26 except that a different phase detection technique is used to detect the phase of the signal output by the filter and amplifier circuit 68. In particular, in this circuit, the signal output from the filter and amplifier circuit 68 is mixed in a mixer 36 with a phase shifted version of the low frequency modulation signal 49 obtained by passing the low frequency modulation signal 49 through a variable phase shifter circuit 70. The signal output from the mixer 36 is then filtered by a low pass filter 38 which removes AC signal components to leave a DC value that is input to the phase detector 72. The phase detector 72 then varies the amount of phase shift applied by the phase shifter circuit 70 to determine the phase shift that minimises or maximises the DC value received from the filter 38. From this determined amount of phase shift, the phase detector 72 determines the value of the above described ratiometric function given in equation 5. In particular, if the phase detector 72 is operable to minimise the DC signal obtained from the mixer 38, then 90 degrees is added to the determined phase to give the value of the ratiometric arctangent function of equation 5, whereas if the phase detector 72 is operable to maximise the DC signal obtained from the mixer 38, then the determined phase corresponds to the value of the ratiometric arctangent .functiojα,,, of: equatioB55r-^Φhe remaining components excitation and detection circuitry illustrated in Figure 27 are the same as those illustrated in Figure 26 and will not, therefore, be described again.
Excitation And Detection Circuitry - 10
In a number of the circuits described above, the detection circuitry 26 includes a phase detector 56 for measuring the phase of a filtered signal. As those skilled in the art will appreciate, the amplification, filtering and phase detection circuitry used to process the signals obtained from terminals A and B of the connection interface 20 will introduce an uncontrolled phase shift into the signals. A particular source of such error is related to the actual detection of the phase in the phase detector 56. As mentioned above during the description of Figure 8, instead of the zero crossing, the crossing of a predetermined threshold level is typically measured in the phase detector 56 by using a comparator. Depending on the ratio between the actual amplitude of the measured signals to the threshold voltage level, threshold-crossing detection will introduce a finite offset in the value of the output of the phase detector 56. If the threshold crossing event is performed on both rising and the falling fronts of the measured signal, the actual value at the output of the phase detector 56 can be determined as a mean arithmetic value of the two measurements. In such case the output of the phase detector 56 will become independent of the level of the threshold voltage used in the comparator. However, Figure 28 illustrates a more general approach that can be used to remove the uncontrolled offset in the phase of the measured signal simultaneously with an uncontrolled offset introduced in the measurement circuit of the phase detector 56.
The excitation circuitry 24 and the detection circuitry 26 are based on the circuitry shown in Figure 26. The main difference is that the excitation circuitry 24 includes circuitry that allows the phase of the pulse trains applied to terminals 3 and 4 of the connection interface 20 to be inverted. In particular, the excitation circuitry 24 includes switches 94"1 and 94"2 that operate to apply, during a first time interval, the non-inverted pulse train to terminal 3 and the inverted pulse train to terminal 4 and, during a second time interval, to apply the inverted pulse train to terminal 3 and the non-inverted pulse train to terminal 4. Figure 29 illustrates in more detail the voltages labelled V1, V2, V3 and V4 applied to the four terminals 1, 2, 3 and 4 of the connection interface 20 during these two intervals (labelled Interval 1 and Interval 2 to the left and right of the line 77). As shown in Figure 29, at the time corresponding to the line 77, the phase of the pulse trains V3 and V4 changes by 180 degrees.
On the detection side, the phase detector 56 is arranged to detect the phase of the signal output by the low pass filter 68 during Interval 1 to calculate a first phase delay or time delay ti. An interval controller 88 then switches over the position of the switches 94"1 and 94" and informs the phase detector 56 accordingly. The phase detector 56 then re- measures the phase of the received signal during Interval 2, to calculate a second phase delay t2. The phase detector 56 then calculates a phase (or time) difference, t0, as the phase ti-t2. Therefore, any common phase offset introduced by the detection circuitry 26 will be removed by this subtraction operation, leaving just a phase term that is proportional to the arctangent function of equation 5 above (which is not removed due to the inverted phase of the excitation voltages V3 and V4). The rest of the processing carried out by the detection circuitry 26 is the same as before and will not, therefore, be described again.. -~"-''-ra..c <~ -■ —-—- _ . ...... „„„,..,.. ...
Excitation And Detection Circuitry - 11
Figure 30 is a block diagram illustrating a further alternative arrangement of the excitation and the detection circuitry that can be used with the windings shown in Figure 2 when connected in the non-inverted manner as per the first sensor design described above, hi this design, the excitation signals and the signals used to control the detection are stored in a programmable read-only memory (PROM) 80 and read out of this memory in response to pulses generated by a pulse generator 60. As illustrated in Figure 30, the PROM outputs a high frequency periodic pulse train 31"1 from terminal OA* and an inverted high frequency periodic pulse train 31" from terminal OB which are applied to terminals A and B of the connection interface 20. For the detection side of the processing, the PROM
80 generates a reference signal for the phase detector 56 which it outputs from terminal
01 ; a low frequency amplitude control signal which it outputs from terminal 02 ; a mixer/sign control signal which it outputs from terminal 03 ; and a sin/cos multiplexer signal which it outputs from terminal 04 .
As shown in Figure 30, the sin/cos multiplexer signal is used to control the multiplexing of the signals from the cos sensor winding 15 (obtained from terminals 1 and 2 of the connection interface 20) and the signals from the sin sensor winding 14 (obtained from terminals 3 and 4 of the connection interface 20) using the multiplexer switches 76"1 and 76"2. The multiplexed signal output from the multiplexing switches 16 and an inverted version of the multiplexed signal are applied to mixing switches 62"1 and 62"2 respectively. As will be described in more detail below, the mixing switches 62"1 and 62~2 effectively demodulate the multiplexed signals which are then re-modulated onto a low frequency carrier signal by the low frequency amplitude control signal via the switches 62" and 614. The signals from the sin and cos sensor windings 14 and 15 are then combined, filtered and amplified by the circuitry 97 to leave a signal of the form given above in equation 7, whose frequency corresponds to that of the low frequency amplitude control signal and whose phase varies with the ratiometric arctangent function defined in equation 5 above. The phase of this signal is then measured by the phase detector 56 as before.
Figure 31 is a signal diagram illustrating the form of the signals generated by the PROM 80. hi particular, Figure 31 illustrates the high frequency excitation signals 31 output from terminals OA* and OB* of the PROM 80. Figure 31 also illustrates the sin/cos multiplexer signal output from terminal 04 of the prom 80. As shown, the multiplexer signal operates at a frequency of 1/16* that of the excitation signals 31. During the odd numbered windows Wl, W3, etc, the signals from the sin sensor winding 14 (obtained from terminals 3 and 4 of the connection interface 20) are passed through to the mixing switches 62; and during the even numbered windows W2, W4 etc, the signals obtained from the cos sensor winding 15 (obtained from terminals 1 and 2 of the connection interface 20) are passed through to the mixing switches 62.
As illustrated in Figure 31, the mixer/sin control signal is a square wave signal having the same frequency as that of the excitation signals 31. Therefore, the mixing switches 62"1 and 62~2 operate to demodulate the multiplexed signals. The demodulated signals are then either connected to ground or accumulated onto a reference capacitor 64"1 or 64~2 for
- a variable time period controlled by the low frequency amplitude control signal using the switches 62~3 or 6T4. As will become apparent from the following discussion, the low frequency amplitude control signal effectively amplitude modulates the demodulated signals from the sin and cos windings 14 and 15 on to low frequency carrier signals (having a frequency 1/576 that of the excitation signals) that are 90 degrees out of phase with each other. This is illustrated in Figure 32 which shows the amplitude modulated signals 101 and 103 obtained from the mixing switch 62" . Similar, but 180 degree phase shifted signals will be obtained from mixing switch 62 . Signal 101 is obtained when the signal from the sin sensor winding 14 is passed through the multiplexing switches 76 and signal 103 is obtained when the signals obtained from the cos sensor winding are passed through the multiplexing switches 76. As both signals 101 and 103 are accumulated on the reference capacitor 64, the actual signal that is accumulated will represent the sum of these two signals, as illustrated by signal 105. The filtering and amplification circuitry 97 effectively filters this signal 105 to produce a smooth sinusoidal signal 107 whose phase varies with the arctangent function defined in equation 5 above. This sinusoidal signal 107 is then passed to the phase detector 56 which measures the phase of this signal relative to the reference signal supplied by the PROM 80 from terminal 01 . The way in which the low frequency amplitude control signal generates the amplitude modulated signals 101 and 103 will now be described in more detail. As can be seen from Figure 31, the low frequency amplitude control signal is a binary signal that controls the opening and closing of switches 62"3 and 62A. In order to achieve the amplitude modulation, the pulse duration of the low frequency amplitude control signal is cyclically varied from a minimum duration corresponding to one period of the excitation signal (such as is applied during windows Wl and W3) to a maximum duration corresponding to 16 periods of the excitation signal (such as is applied during window W2). As a result, the amount of signal accumulated on the reference capacitor 64 will also cyclically vary, with the peak amplitude of the accumulated signal depending on the level of the demodulated signals obtained from mixer switch 62" and hence on VCOs and VSin. However, as the switches 62"3 and 62~4 are simple switches that do not have a reverse polarity, in order to achieve the change in polarity of the amplitude modulation, the phase of the mixing signals applied to the mixer switches 62" and 62"2 are shifted by 180 degrees at the appropriate timings in order to change the sign of the re-modulated signal. This is illustrated in Figures 31 and 32 around the zero crossing for the signal 101. In particular, as can be seen from Figure 32, between the time windows Wl and W3, the signal 101 passes through zero and, as shown in Figure 31, the phase of the mixing signal applied during window Wl is 180 degrees out of phase with the phase of the mixing signal applied during window W3. A similar change in the phase of the mixer/sign control signal is performed when the signal 103 passes through zero.
Figure 32 illustrates the form of the signal 105 accumulated on one of the reference capacitors 64. As mentioned above, the signal accumulated on the other reference capacitor 64 will have the same form but opposite polarity. A description will now be given (with reference to Figure 33) of the way in which the amplification and filtering circuitry 97 operates to combine these signals to generate the signal 107 which is output to the phase detector 56. As shown in Figure 33, the reference capacitors 64"1 and 64"2 are connected to ground. Thus, the high frequency components of the input currents Ii and h are shorted to ground thrpugh a .smalLimpedance, whereas low frequencyvariatioήs "(such as are caused by the low frequency modulation of the demodulated signals) marked as Ii and I2, are further amplified and combined by the circuitry 97 to generate the output signal 93. m the particular case where the two resistances R2 and R3 of the circuitry 97 are equal, the output current (Zi) from the circuitry 97 will become proportional to the difference in the currents Ii and I2, and as a result, any common mode signals will be rejected. However, as mentioned above, the desired signals (that vary with the position being measured) applied to the two reference capacitors 64"*and 64"2 are of opposite polarity and will therefore be added together by the circuitry 97. Consequently, the output signal can be approximated by equation 7 given above, with the frequency f2 corresponding to half the repetition frequency of the low frequency amplitude control signal.
Excitation And Detector Circuitry - 12 Figure 34 is a block diagram illustrating a further alternative arrangement of the excitation and detection circuitry that can be used with the windings shown in Figure 2 when connected in the non-inverted manner as in the first sensor design described above.
The excitation and detection circuitry shown in Figure 34 is based on the excitation and detection circuitry shown in Figure 30. The main difference is that the sign control and the low frequency amplitude control are effectively combined with the original excitation signals 31 to form a new set of excitation signals 31"1 and 31"2 that are output from the PROM 80 from terminals OA and OB . In particular, these signals are effectively combined with the excitation signals so that during the periods that the switches 62~3 and 02"4 in the Figure 30 circuit were connected to ground, no excitation signals are applied to the excitation winding attached to terminals A and B of the connection interface 20.
This is illustrated more clearly in Figure 35, which is a signal diagram illustrating the form of the signals generated by the PROM 80 in Figure 34. As shown, during window
Wl, a single pulse of each excitation signal 31 is output from the PROM 80. Similarly, during window W3, a single pulse of each excitation signal 31, is output from the PROM
80. In order to achieve the change of sign required for the low frequency modulation, the phase of each excitation pulse applied during window W3 is 180 degrees shifted compared to the phase of the excitation pulses applied during window Wl. Therefore, as those skilled in the art will appreciate, since the sign control is also achieved by varying the excitation signals 31, the mixer signal applied to the mixing switches 62"1 and 62" can be a constant square wave signal at the same frequency as the frequency of the excitation pulses (as shown in Figure 35). The remaining processing circuitry illustrated in Figure 34 is the same as the processing circuitry of Figure 30 and a further description will therefore be omitted.
Excitation And Detection Circuitry - 13
Figure 36 is a block diagram illustrating a further alternative arrangement of the excitation and detection circuitry that can be used with the winding shown in Figure 2 when connected in the inverted manner illustrated in Figure 21. The circuitry illustrated in Figure 36 is based on the circuitry illustrated in Figure 30.
As shown in Figure 36, in this design the PROM 80 generates four excitation signals 31"1, 31~2, 31"3, and 31"4 which it outputs from terminals 05*, 06*, 07* and 08*. These excitation signals 31 are applied to a respective one of -the terminals numbered 1 to 4 of the connection interface 20 which are connected to the windings 14 and 15, as shown in Figure 21. In this design, as illustrated in Figure 37, the PROM 80 is arranged to generate the excitation signals so that excitation signals 31"3 and 31"4 are applied to terminals 3 and 4 of the connection interface 20 during the odd numbered windows Wl, W3 etc and so that excitation signals 31"1 and 31"2 are applied to terminals 1 and 2 of the connection interface 20 during the even numbered windows W2, W4 etc. hi this way, the excitation signals are applied to the windings 14 and 15 in a time division multiplexed manner.
On the detection side, the circuitry is the same as the detection circuitry illustrated in Figure 30, except there is no need for the multiplexing switches 76 due to the multiplexing of the excitation signals. A further description of the operation of the detection circuitry will therefore be omitted.
Excitation And Detection Circuitry - 14
Figure 38 is a block diagram illustrating a modification of the excitation circuitry shown in Figure 36. As shown, in this design, the PROM 80 outputs excitation signals 31"1 and 31"2. The DC components of these pulse trains are removed by the capacitors 96"1 and 96"2 to generate pulse trains 98"1 and 98"2, which are 180 degrees out of phase with each other. As shown in Figure 38, these pulse trains 98 are applied to terminals 1 to 4 of the connection interface 20 through a set of diodes 99" , 99" , 99" and 94"\ As a result of the diodes 99"1 and 99"2, when the pulse train 98"1 is positive and the pulse train 98"2 is negative, excitation current will flow through diode 99" into terminal 1 of the connection interface 20. and the current will flow from terminal 2 of the connection interface 20 through diode 99"2 back to ground. However, when pulse train 98"1 is negative and pulse train 98"2 is positive, no current is applied to terminals 1 and 2 of the connection interface 20. Instead, a current will flow through diode 99 into terminal 4 of the connection interface 20 and will return from terminal 3 of the connection interface 20 through diode 99"3 back to ground. Thus pulses of excitation signal will be applied sequentially to the winding connected to terminals 1 and 2 of the connection interface 20 and then to the winding connected to terminals 3 and 4 of the connection interface 20, in a similar manner to the way in which the excitation signals are applied to the excitation circuitry shown in Figure 36.
The detection circuitry is the same as before, and therefore a further description will be omitted.
Modifications And Other Alternatives
A number of sensor designs and circuits have been described above with reference to the accompanying Figures. As those skilled in the art will appreciate a number of modifications and variations can be made to these systems. A number of these modifications and variations will now be described for illustration only.
In the first sensor design described above, the DC magnet 6 was oriented with its magnetic axis parallel with the Z direction. This results in the position encoder being sensitive to rotational movement of the piston 4 about its axis. This is not a problem where the piston 4 is"arranged*so thaHt cannot rotate within the cylinder 2, However, in most piston and cylinder assemblies, the piston is free to rotate within the cylinder. Therefore, in a preferred embodiment, the magnet 6 is a ring magnet which is attached around the piston shaft so that the magnetic axis of the magnet is directed in the X direction shown in Figure 1. With such an arrangement, the magnetic field generated by the magnet will be the same whether or not the piston 4 rotates within the cylinder 2.
hi the above sensor designs, the DC magnetic field was generated by the DC magnet 6. As those skilled in the art will appreciate, the DC magnetic field may be generated by using an electromagnet or by applying a DC current to an appropriately oriented coil. Additionally, it is not essential to use a DC magnet within the position encoder, hi particular, the movable object (e.g. the piston) may carry an AC magnetic field generator which generates a positionally varying magnetic field. Provided this positionally varying magnetic field interacts with the film 25 of magnetic material so that its magnetisation state varies along the measurement path, a signal will be generated in the sensor winding when the excitation winding is energised with the excitation signal. In such an embodiment, the AC magnetic field generator mounted on the moving object preferably generates a low frequency signal that can penetrate through metallic walls. Additionally, in such an embodiment, the signal generated in the sensor winding will be at an intermodulation frequency defined by the difference in frequency between the two AC magnetic fields. The AC magnetic field generator may be an active device such as a powered coil or a passive device, such as a resonator that is energised by the excitation of the excitation windings.
In the first sensor design, the magnetisable material that was used was a soft ferromagnetic material having high initial and maximum permeability and a low coercivity. As those skilled in the art will appreciate, other magnetisable materials may be used. For example amorphous alloys such as VITROVAC from Vacuumschmelze or Nanno-crystalline alloys such as VITROPERM from Vacuumschmelze, silicon iron alloys with three percent silicon, pure iron, nickel iron alloys, cobalt iron alloys etc.
In the first sensor design described above, the sensor windings were each formed from two turns of conductor. As those skilled in the art will appreciate, the use of two turn sensor windings is not essential. Any number of turns may be provided. Preferably, as many turns as possible are provided in the space allowed by the dimensions of the PCB 8 as this maximises signal levels obtained from the sensor windings.
Similarly, in the first sensor design described above, the excitation winding included a single turn of conductor. As those skilled in the art will appreciate, the number of turns of conductor for the excitation winding and for the sensor windings can be varied in order to vary the reactive impedance of the windings to match the impedance of the appropriate output or input of the excitation and detection electronics.
In the first sensor design described above, the position encoder was used to determine the position of a piston within a cylinder. As those skilled in the art will appreciate, the position sensor described above may be used in a number of different applications. For example, it can be used in shock absorbers, damping cylinders, syringes, machine tool applications, etc. The applicants earlier International Application WO2005/085763 illustrates- a -number-of- such different applications and sensor arrangements (including " rotary sensor arrangements) to which the improvements described herein can be applied.
In the above sensor designs, the position of the piston determined by the detection electronics was displayed on a display. As those skilled in the art will appreciate, in alternative embodiments, the position information may be provided to another computer system for controlling another part of a system. For example, where the piston and cylinder assembly forms part of an engine, the determined position may be supplied to an engine management unit which can use the position information to control the timing of ignition of the fuel mixture within the piston and cylinder assembly.
In the above sensor design, the excitation current applied to the excitation winding had a frequency of approximately 10MHz. As those skilled in the art will appreciate, it is not essential to use such an excitation frequency. Excitation frequencies between 10kHz and 1 OMHz are preferably used.
In the first sensor design described above, the sensor windings were formed in a figure of eight configuration. As those skilled in the art will appreciate, it is not essential to form the sensor windings in such a figure of eight configuration. The only requirement of the sensor windings is that they are able to detect a magnetic field which positionally varies along the measurement direction. This can be achieved by a single sensor winding positioned at a position along the measurement path. Alternatively, it can be achieved using a sensor winding which geometrically varies along the measurement path. This geometrical variation may be its shape along the measurement path or its dimensions such as the thickness of the conductor forming the sensor winding or the number of turns of the sensor winding etc.
hi all of the sensor designs described above, the sensor windings and the film of magnetic material have been fixed (stationary) and the magnet was mounted in the movable member. As those skilled in the art will appreciate, the position encoder described above will operate where the magnet is stationary and where the sensor windings move relative to the magnet. Additionally, the sensor windings and/or the excitation windings may move in addition to the magnet. All that is necessary is that there is relative movement between the magnet and at least one of the sensor winding and the excitation winding. Similarly, it is not essential for the film of magnetic material to be fixed relative to the excitation and sensor windings. The magnetic material may move with the movable object or it may move independently of the object. However, since the film of magnetic material is substantially homogenous, movement of the film relative to the sensor windings will not affect the operation of the position encoder.
In the above sensor designs, the excitation and sensor windings were formed as conductor tracks on a printed circuit board. As those skilled in the art will appreciate the excitation and sensor windings may be formed using any conductive material, such as conductive inks which can be printed on an appropriate substrate or conductive wire wound in the appropriate manner. Additionally, it is not essential for the excitation winding and the sensor winding to be mounted on the same member. For example, two separate printed circuit boards may be provided, one carrying the excitation winding and the other carrying the or each sensor winding.
In the first sensor design described above, the excitation signal applied to the excitation winding was an AC signal at a particular frequency. As those skilled in the art will appreciate, it is not essential for the excitation signal to be AC. For example, the excitation signal may be any cyclically varying signal.
In the first sensor design described above, two phase quadrature sensor windings were provided. As those skilled in the art will appreciate, it is not essential to use sensor windings that are in phase quadrature. For example, instead of using the cos sensor winding, a second sensor winding phase shifted by an eighth of the pitch along the measurement path may be used. However, as those skilled in the art will appreciate, the use of phase quadrature sensor windings is preferred as this simplifies the processing to be performed by the detection circuitry to determine the position of the movable member. Additionally, as those skilled in the art will appreciate, in embodiments that use sensor windings that provide signal levels that vary substantially sinusoidally with the position of the movable member, it is not essential to only use two sensor windings. For example, three or four sensor windings may be provided each separated along the measurement path by an appropriate distance (or angle in the case of a rotary position encoder). In the first sensor design described above, an AC current was applied to the excitation winding having a peak amplitude of approximately 5OmA. As those skilled in the art will appreciate, it is not essential to apply an excitation current with this peak amplitude. The magnitude of the excitation current is preferably chosen depending on the position and layout of the excitation winding relative to the film of magnetisable material so that the excitation magnetic field generated by the excitation winding will be strong enough to produce a reasonable signal to noise ratio in the measurement electronics. Therefore, appropriate excitation current strengths may vary from 0.01mA to 1OA.
In the first sensor design described above, the film of magnetic material had a thickness of less than 0.2mm. As those skilled in the art will appreciate, films of various thicknesses may be used. Preferably, the thickness of the material is between 20microns to 2mm as such films are readily available.
In the sensor designs described above, the film of magnetic material and the PCB had approximately the same width and extent along the measurement direction. In a preferred embodiment, the width and length of the film of magnetisable material is greater than the width and length of the PCB, as this minimises edge effects associated with the edge of the film of magnetisable material (which can alter the balance between the excitation and sensor windings).
In the above sensor designs, the film of magnetisable material was initially unsaturated and the DC magnet created an in-homogeneity spot in the film. The position of this in- homogeneity spot was then detected by detecting the change in the mutual inductance between an excitation winding and a sensor winding. As those skilled in the art will appreciate, the film of magnetisable material does not have to be initially in an unsaturated state. A strong background (or bias) magnetic field may be provided near the film which saturates the entire film. Such a fully saturated film is still homogenous and will not alter the mutual coupling between the excitation and sensor windings. Provided the DC magnet carried by the moving member is strong enough' to counter the effects of the saturating field to create an in-homogeneity region in the film, a similar imbalance will be created between the excitation and sensor windings. This can then be detected in the manner described above in the first embodiment. As those skilled in the art will appreciate, the position of this in-homogeneity region will not correspond to the position where the DC magnetic field is perpendicular to the film, but where the DC field is strong enough to counter the effect of the bias field.
In the inverted excitation and detection arrangement described with reference to Figure 22, two excitation signals of different frequencies where simultaneously applied, one to each of the sin and cos windings 14 and 15. In an alternative embodiment, the two excitation signals may be applied one after the other, in which case only one mixer 36 and one low pass filter 38 are required in the detection circuitry 26. Further, if the excitation signals are applied to the two windings 14 and 15 at different times, then the same excitation frequency can be applied to each pair.
In all of the sensor designs described above, the windings were arranged to provide signals that varied (approximately) in a sinusoidal manner with the position to be detected. As those skilled in the art will appreciate, different shaped windings may be provided that provide a substantially linear variation with position along the length of the sensor head. In this case, the detection circuitry can be arranged to perform a ratiometric function like R =(Vi-2 -V3-4) I (V1-2 + V3-4 ).
In the sensor design described with reference to Figure 10, two excitation windings were provided at each end of the cylinder. These two excitation windings were connected in series and energised by a single excitation circuit. As those skilled in the art will appreciate, the two excitation windings may be individually connected to separate excitation circuitry which drives the respective excitation windings accordingly.
hi the sensor design described with reference to Figures 16, 17 and 18, two sensing transducers were provided, each having an excitation winding and two sensor windings. As those skilled in the art will appreciate, it is possible to provide a single excitation winding that is used to provide the excitation magnetic field for both sensors. In this case, the single excitation winding may be mounted on one of the sensor boards (transducers) 8"1 or 8"2 or it may be mounted on a separate board, for example, positioned between the sensor boards 8"1 and 8"2. Likewise, when the windings are driven in the inverted manner (described with reference to Figure 21) a single sensor winding may be provided in common to both transducers. Again, this common sensor winding may be provided on either of the sensor boards 8"1 or 8"2 or on another sensor board, for example, positioned between sensor boards 8"1 and 8"2.
hi the above sensor designs, transducers in the form of printed circuits boards which carried excitation and sensor windings and a film of magnetisable material were used. As those skilled in the art will appreciate, other types of sensing transducers can be used.
For example sensing transducers can be used that define the excitation and sensor windings by conductive ink printed on a suitable substrate or by appropriately shaped wire bonded onto an appropriate substrate. Further, in order to maximise signal levels obtained from the sensor windings, the magnetic film is preferably provided on each side of the sensing transducer substrate.. ,— — . , r
In the above embodiments, the excitation and the detection circuits included various electronic hardware circuits, hi an alternative embodiment, a programmable circuit (processor) controlled by software stored in a memory may implement these circuits. The software may be provided in any appropriate form and in any computer language. It may be supplied as a signal or stored on a computer readable medium such as a CD ROM.
In the above embodiments, complimentary excitation signals were applied to each end of the or each excitation winding. As discussed above in the first sensor design, this is found to be advantageous as it allows the system to operate at higher excitation frequencies. However, as those skilled in the art will appreciate, it is not essential to apply excitation current to the excitation winding in this way. Instead, one end of the/or each excitation winding may be connected to a reference potential such as ground, with the other end being connected to receive the excitation signal.
As those skilled in the art will appreciate, a number of excitation and detection circuits have been described above. Whilst the circuits are preferably used with the specific sensor heads described in the present application, they can be used with other sensor designs.

Claims

Claims
1. A position encoder comprising: first and second members which are relatively movable along a measurment path; a magnetic field generator carried by the second member and operable to generate a magnetic field that varies with position along the measurement path; first and second sensor transducers each comprising: i) an excitation winding and a sensor winding, at least one of which is carried by the first member; and ii) a film of magnetisable material located adjacent said excitation and sensor windings and located, in use, within said positionally varying magnetic field to cause the film to have a positionally varying magnetisation state along the measurement path; wherein the film of magnetisable material of said first transducer is positioned, in use, closer to said magnetic field generator than the film of magnetisable material of said second transducer; wherein the excitation winding and the sensor winding of each transducer are arranged relative to the corresponding film of magnetisable material so that a mutual electromagnetic coupling between them varies in dependence upon the positionally varying magnetisation state of the corresponding film of magnetisable material, so that when the excitation winding is energised with an excitation signal, a sensor signal is generated in the sensor winding that varies with the relative position between said first and second members; differencing circuitry operable to combine the sensor signals generated in said sensor windings to form a difference signal representing the difference in electromagnetic coupling between the excitation and sensor windings of the first and second transducers, which difference signal varies with the relative position between said first and second members; and a processing circuit operable to process said difference signal to determine a value indicative of the relative position of said first and second relatively moveable members.
2. A position encoder according to claim 1, wherein the excitation winding of said first and second transducers is formed as the same conductor coil that is provided in common to said first and second transducers.
3. A position encoder according to claim 1, wherein the sensor winding of said first and second transducers is formed as the same conductor coil that is provided in common to said first and second transducers.
4. A position encoder according to claim 1, wherein said excitation winding of said first transducer is connected in series with the excitation winding of said second transducer.
5. A position encoder according to any preceding claim, wherein the excitation winding of said first transducer is operable, when energised, to generate an electromagnetic field that is substantially in-phase with an electromagnetic field generated, when energised, by the excitation winding of the second transducer.
6. A position encoder according to claim 5, wherein said differencing circuitry is operable to subtract the signals obtained from the sensor windings of the first and second transducers.
7. A position encoder according any of claims 1 to 4, wherein the excitation winding of said first transducer is connected in series with the excitation winding of said second transducer such that an electromagnetic field generated by the excitation winding of the first transducer is substantially 180 degrees out of phase with an electromagnetic field generated by the excitation winding of the second transducer.
8. A position encoder according to claim 7, wherein said said differencing circuitry is operable to add the signals obtained from the sensor windings of the first and second transducers.
9. A position encoder according to any preceding claim, wherein said differencing circuitry comprises connecting conductors that connect the sensor winding of the first transducer to the sensor winding of the second transducer.
10. A position encoder according to any preceding claim, wherein the film of magnetisable material of said first transducer is spaced from said magnetic field generator by a distance of between zero and 20mm and wherein the film of magnetisable material of said second transducer is spaced from said magnetic field generator by a distance of between 2mm and 25mm.
11. A position encoder according to any preceding claim, further comprising excitation circuitry operable to generate an excitation signal for energising the excitation winding of said first and second transducers.
12. A position encoder according to claim 11, wherein the excitation electromagnetic field generated by the excitation winding of each transducer comprises a first component which is orthogonal to the surface of the corresponding film and a second component which is parallel to the surface of the corresponding film and wherein the excitation winding and the excitation circuitry are arranged so that the magnitude of said second component is insufficient to drive significant portions of the corresponding film into and out of saturation in the vicinity of the corresponding sensor winding.
13. A position encoder according to claim 11 or 12, wherein the excitation winding of each transducer is arranged relative to the corresponding film so that said excitation electromagnetic field is substantially perpendicular to the film along the measurement direction.
14. A position encoder according to any of claims 11 to 13, wherein said excitation circuitry is operable to generate an excitation signal having an excitation frequency and wherein said processing circuit is operable to process a difference signal which is substantially at said excitation frequency, to determine said value indicative of the relative position between the first and second relatively movable members.
15. A position encoder according to claim 14, wherein said processing circuit is operable to combine said difference signal with a signal having the same frequency as said excitation frequency.
16. A position endcoder according to claim 15, wherein said processing circuit is operable to mix said difference signal with a signal having the same frequency as said excitation frequency.
17. A position encoder according to any of claims 11 to 16, wherein said excitation circuitry is operable: i) to generate first and second cyclically varying excitation signals, the second excitation signal being the inverse of said first excitation signal; and ii) to apply said first excitation signal to one end of the excitation winding of said transducers and said second excitation signal to the other end of the excitation winding of said transducers.
18. A position encoder according to claim 17, wherein said excitation circuitry is operable to generate AC excitation signals.
19. A position encoder according to claim 17, wherein said excitation circuitry is operable to generate excitation signals that comprise sequences of voltage pulses.
20. A position detector according to claim 19, wherein said excitation circuitry is operable to generate excitation signals comprising a sequence of modulated voltage pulses.
21. A position detector according to claim 20, wherein said excitation circuitry is operable to amplitude modulate or pulse width modulate said sequence of voltage pulses using a modulation signal having a lower frequency than the pulse repetition frequency of the pulses and wherein said processing circuit is operable to detect modulations of said difference signal that vary with the relative position of said first and second members.
22. A position encoder according to any preceding claim, wherein said magnetic field generated by said magnetic field generator creates an in-homogeneity spot in each said film, the position of which varies with the relative position between the first and second relatively movable members and wherein the excitation and sensor windings of said first and second transducers are arranged so that the mutual electromagnetic coupling between them varies in dependence upon the position of the in-homogeneity spot in the corresponding film.
23. A position encoder according to claim 22, wherein each in-homogeneity spot comprises an unsaturated region of the magnetisable material surrounded by a saturated region of the magnetisable material.
24. A position encoder according to claim 22 or 23, wherein the in-homogeneity spot is created at a position in each film where the magnetic field generated by said magnetic field generator is substantially perpendicular to the film of magnetisable material.
25. A position encoder according to any preceding claim, wherein each transducer comprises first and second sensor windings that are separated along said measurement path and which are arranged so that when the corresponding excitation winding is energised, a respective sensor signal is generated in each sensor winding that varies with the relative position between said first and second members; wherein said differencing circuitry is operable: i) to combine the sensor signals generated in the first sensor winding of each transducer to form a first difference signal representing the difference in electromagnetic coupling between the excitation winding and the first sensor winding of the first and second transducers; and ii) to combine the sensor signals generated in the second sensor winding of each transducer to form a second difference signal representing the difference in electromagnetic coupling between the excitation winding and the second sensor winding of the first and second transducers; and wherein said processing circuit is. operable to process said first and second difference signals to determine the value of a ratiometric function, which value is indicative of the relative position between the first and second relatively movable members.
26. A position encoder according to any of claims 1 to 14, wherein each transducer comprises first and second excitation windings operable to generate, when energised, an excitation electromagnetic field; wherein each excitation winding and the sensor winding of each transducer are arranged relative to the corresponding film of magnetisable material so that a mutual electromagnetic coupling between them varies in dependence upon the positionally varying magnetisation state of the corresponding film of magnetisable material, so that when the first excitation winding of each transducer is energised with an excitation signal, a first sensor signal is generated in the sensor winding of each transducer that varies with the relative position between said first and second members and so that when the second excitation winding of each transducer is energised with an excitation signal, a second sensor signal is generated in the sensor winding of each transducer that varies with the relative position between said first and second members; wherein said differencing "circuitry" is operable: i) to combine the first sensor signal generated in the sensor winding of each transducer to form a first difference signal representing the difference in electromagnetic coupling between the first excitation winding and the sensor winding of the first and second transducers; and ii) to combine the second sensor signal generated in the sensor winding of each transducer to form a second difference signal representing the difference in electromagnetic coupling between the second excitation winding and the sensor winding of the first and second transducers; and wherein said processing circuit is operable to process said first and second difference signals to determine the value of a ratiometric function, which value is indicative of the relative position between the first and second relatively movable members.
27. A position encoder according to claim 25 or 26, wherein said processing circuit is operable to calculate a ratiometric function using said first and second difference signals to determine said value.
28. A position encoder according to claim 25 or 26, wherein said processing circuit is operable to combine said first and second difference signals to generate a signal whose phase varies with the value of said ratiometric function and is operable to determine the value of said ratiometric function by measuring the phase of said signal.
29. A position encoder according to any preceding claim, wherein said at least one winding of each transducer which is carried by said first member is arranged along said measurement path in a geometrically varying manner.
30. A position encoder according to claim 29, wherein said winding carried by said first member geometrically varies along the measurement path so that said sensor signal generated in said sensor winding of each transducer varies substantially sinusoidally with the relative position between said first and second relatively movable members.
31. A position encoder according to any preceding claim, wherein said magnetic field generator is operable to generate a magnetic field having a magnetic axis which lies at an angle to said first and second films.
32. A position encoder according to claim 31, wherein said magnetic field generator is operable to generate a magnetic field having an axis which is substantially perpendicular to said first and second films.
33. A position encoder according to any preceding claim, wherein said magnetic field generator is operable to generate a DC magnetic field.
34. A position encoder according to any preceding claim, wherein said winding of each transducer that is carried by said first member comprises at least two loops of conductor which extend along the measurement direction and which are connected in series in a figure of eight arrangement.
35. A position encoder according to any preceding claim, wherein each transducer comprises a~prurality of sensor windings each of which is provided adjacent to a different portion of the corresponding film of magnetisable material.
36. A position encoder according to any preceding claim, wherein said film of magnetisable material of each transducer has a high permeability and a low coercivity.
37. A position encoder according to any preceding claim, wherein said film of magnetisable material of each transducer comprises at least one of: pure iron, nickel iron alloy, cobalt iron alloy, an amorphous alloy, nano-crystalline alloy or a silicon iron.
38. A position encoder according to any preceding claim, wherein said measurement path is linear.
39. A position encoder according to any of claims 1 to 37, wherein said measurement path is circular.
40. A position encoder according to any preceding claim, wherein said first and second films lie in substantially parallel planes.
41. A position encoder according to any preceding claim, wherein said first and second films are substantially planar.
5 42» A position encoder according to any preceding claim, wherein the excitation winding and the sensor winding of each transducer are arranged so that, in the absence of said magnetic field generator, there is substantially no electromagnetic coupling between them.
10 43. A position encoder according to claim 42, wherein the excitation winding and the sensor winding of each transducer lie in substantially the same plane or in substantially parallel planes.
44. A sensor according to any preceding claim, wherein each transducer comprises a 15 printed circuit board carrying conductive tracks that define said excitation and sensor windings and on at least one side of which said film of magnetisable material is carried.
45. A position encoder comprising: first and second members which are relatively movable along a measurment path;
20 a magnetic field generator carried by the second member and operable to generate a magnetic field which varies with position along the measurement path; a first excitation winding and a first sensor winding at least one of which is carried by the first member; a second excitation winding and a second sensor winding at least one of which is 5 carried by the first member; a first film of magnetisable material located adjacent said first excitation and sensor windings and located, in use, within said positionally varying magnetic field at a first distance from said magnetic field generator, to cause the first film to have a positionally varying magnetisation state along the measurement path;
30 a second film of magnetisable material located adjacent said second excitation
■ -ir and sensor windings andjocated, in use, within said positionally varying magnetic field-at' a second greater distance from said magnetic field generator to cause the second film to have a positionally varying magnetisation state along the measurement path; wherein the first excitation winding and the first sensor winding are arranged 5 relative to said first film so that a mutual electromagnetic coupling between them varies in dependence upon the positionally varying magnetisation state of said first film of magnetisable material, so that when said first excitation winding is energised with an excitation signal, a sensor signal is generated in said first sensor winding that varies with the relative position between said first and second members; 0 wherein the second excitation winding and the second sensor winding are arranged relative to said second film so that a mutual electromagnetic coupling between them varies in dependence upon the positionally varying magnetisation state of said second film of magnetisable material, so that when said second excitation winding is energised with an excitation signal, a sensor signal is generated in said second sensor 5 winding that varies with the relative position between said first and second members; a differencing circuitry operable to combine the sensor signals generated in said sensor windings to form a difference signal representing the difference in electromagnetic coupling between the first excitation winding and the first sensor winding and the second excitation winding and the second sensor winding, which difference signal varies with the relative position between said first and second members; and a processing circuit operable to process said difference signal to determine a value indicative of the relative position of said first and second relatively moveable members.
46. A position encoder comprising: first and second members which are relatively movable along a measurment path; a magnetic field generator carried by the second member and operable to generate a magnetic field that varies with position along the measurement path; first and second transducers each comprising: i) an excitation winding and a sensor winding, at least one of which is carried by the first member; and ii) a film of magnetisable material located adjacent said excitation and sensor windings and located, in use, within said positionally varying magnetic field to cause the film to have a positionally varying magnetisation state along the measurement path; wherein the film of magnetisable material of said first transducer is positioned, in use, closer to said magnetic field generator than the film of magnetisable material of said second transducer; wherein the excitation winding and the sensor winding of each transducer are arranged relative to the corresponding film of magnetisable material so that a mutual electromagnetic coupling between them varies in dependence upon the positionally varying magnetisation state of the corresponding film of magnetisable material, so that when the excitation winding is energised with an excitation signal, a sensor signal is generated in the sensor winding that varies with the relative position between said first and second members; means for combining the sensor signals generated in said sensor windings to form a difference signal representing the difference in electromagnetic coupling between the excitation and sensor windings of the first and second transducers, which difference signal varies with the relative position between said first and second members; and means for processing said difference signal to determine a value indicative of the relative position of said first and second relativelymoveable members.
47. A position encoder comprising: first and second members which are relatively movable along a measurment path between a start region and an end region; a first excitation winding and a first sensor winding positioned adjacent said start region and at least one of which is carried by the first member; a second excitation winding and a second sensor winding positioned adjacent said end region and at least one of which is carried by the first member; a magnetic field generator carried by the second member and operable to generate a magnetic field which varies with position along the measurement path; at least one film of magnetisable material located adjacent said start and end regions and arranged so that when the film is located within said positionally varying magnetic field the film has a positionally varying magnetisation state along the measurement path; wherein the first excitation winding and the first sensor winding are arranged relative to said film so that a mutual electromagnetic coupling between them varies in dependence upon the positionally varying magnetisation state of said film of magnetisable material, so that when said first excitation winding is energised with an excitation signal and said first and second members are in the start region, a sensor signal is generated in said first sensor winding that varies with the relative position between said first and second members; wherein the second excitation winding and the second sensor winding are arranged relative to said film so that a mutual electromagnetic coupling between them varies in dependence upon the positionally varying magnetisation state of said film of magnetisable material, so that when said second excitation winding is energised with an excitation signal and said first and second members are in the end region, a sensor signal is generated in said second sensor winding that varies with the relative position between said first and second members; and a processing circuit operable to process the sensor signals generated in the sensor windings in response to the energisation of said excitation windings, to determine the relative position of said first and second members in said start or end region.
48. A position encoder according to claim 47, further comprising a third excitation winding and a third sensor winding which are provided extending along a measurement range between said start and end regions and wherein said processing circuit is operabe to use the signals from said third sensor winding to determine the relative position of said first and second members when they are not in said start or end region.
49. A position encoder according to claim 48, wherein when said first and second members are in said start region, said processing circuit is operable to combine measurements obtained using the signals from said first sensor winding and said third sensor winding.
50. A position encoder according to claim 48, wherein when said first and second members are in said end region, said processing circuit is operable to combine measurements obtained using the signals from said second sensor winding and said third sensor winding.
51. A position encoder according to claim 49 or 50, wherein said processing circuit is operable to perform a weighted combination of the measurements obtained from said sensor windings.
52. A position encoder according to any of claims 47 to 51, wherein said first and second excitation windings are connected in series.
53. A position encoder according to any of claims 47 to 52, further comprising the features of any of claims 1 to 39.
54. A position encoder comprising: first and second members which are relatively movable along a measurment path; an excitation winding and a sensor winding, at least one of which is carried by the first member; a magnetic field generator carried by the second member and operable to generate a magnetic field which varies with position along the measurement path; a film of magnetisable material which is located, in use, within said positionally varying magnetic field to cause the film to have a positionally varying magnetisation state along the measurement path; wherein the excitation and sensor windings are arranged relative to said film so that a mutual electromagnetic coupling between them varies in dependence upon the positionally varying magnetisation state of said film of magnetisable material, so that when said excitation winding is energised with an excitation signal, a sensor signal is generated in said sensor winding that varies with the relative position between said first and second members; excitation circuitry operable: i) to generate first and second cyclically varying excitation signals, the second excitation signal being the inverse of said first excitation signal; and ii) to apply said first excitation signal to one end of said excitation winding and said second excitation signal to the other end of said excitation winding to thereby energise the excitation winding with symmetric excitation signals; and a processing circuit operable to process the sensor signal generated in the sensor winding in response to the energisation of said excitation winding, to determine a value indicative of the relative position between the first and second relatively movable members.
55. A position encoder according to claim 54, further comprising the features of any of claims 1 to 52.
56. A method of determining the relative position of first and second relatively movable members, the method comprising the steps of: providing a position encoder according to any preceding claim; energising the or each excitation winding with an excitation signal; and processing the sensor signal induced in the or each sensor winding which varies in dependence upon the relative position of the first and second members, to determine a value indicative of the relative position between the first and second relatively movable members.
57. Excitation and/or1 detection circuitry substantially as described herein above with reference to or as shown in Figures 6 to 8, 11, 12, 14, 15, 19, 20 and 22 to 38.
PCT/GB2006/003423 2006-09-14 2006-09-14 Position sensor WO2008032008A1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/GB2006/003423 WO2008032008A1 (en) 2006-09-14 2006-09-14 Position sensor

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/GB2006/003423 WO2008032008A1 (en) 2006-09-14 2006-09-14 Position sensor

Publications (1)

Publication Number Publication Date
WO2008032008A1 true WO2008032008A1 (en) 2008-03-20

Family

ID=38109609

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/GB2006/003423 WO2008032008A1 (en) 2006-09-14 2006-09-14 Position sensor

Country Status (1)

Country Link
WO (1) WO2008032008A1 (en)

Cited By (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
GB2517152A (en) * 2013-08-12 2015-02-18 Gde Technology Ltd Position sensor
WO2015140080A1 (en) * 2014-03-17 2015-09-24 Tt Electronic Technology Limited Position sensing apparatus and method
DE102014213221A1 (en) * 2014-07-08 2016-01-14 Continental Teves Ag & Co. Ohg Displacement measurement based on eddy currents and a shield canceling donor element
DE102015116429A1 (en) 2015-09-29 2017-03-30 Hella Kgaa Hueck & Co. Electronic unit with sensor arrangement
US9709614B2 (en) 2008-10-15 2017-07-18 Azoteq (Pty) Ltd Parasitic capacitance cancellation in capacitive measurement
US10527457B2 (en) 2015-02-27 2020-01-07 Azoteq (Pty) Ltd Inductance sensing
US20200132511A1 (en) * 2014-10-28 2020-04-30 Horst Siedle Gmbh & Co. Kg Position sensor, position measuring device and method for the operation thereof
CN111562618A (en) * 2019-02-13 2020-08-21 巴鲁夫公司 Inductive sensor and method of operating the same
CN112504104A (en) * 2020-11-20 2021-03-16 苏州纳芯微电子股份有限公司 Position sensing circuit based on inductance principle, sensor and position measuring method
US20220128381A1 (en) * 2020-10-22 2022-04-28 Renesas Electronics America Inc. Position sensor system
US11703359B2 (en) 2019-05-14 2023-07-18 Kyocera Avx Components (Werne) Gmbh Inductive position sensing apparatus including a screening layer and method for the same
US12013264B2 (en) 2018-10-16 2024-06-18 Kyocera Avx Components (Werne) Gmbh Position sensing apparatus and method

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE3914787A1 (en) * 1989-05-05 1990-11-08 Hermann Michael Dipl Phys Inductive position sensor measuring linear or rotary movement - has magnetisable core locally saturated by external magnetic field to obtain virtual air gap
WO2005085763A2 (en) * 2004-03-01 2005-09-15 Scientific Generics Limited Position sensor

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE3914787A1 (en) * 1989-05-05 1990-11-08 Hermann Michael Dipl Phys Inductive position sensor measuring linear or rotary movement - has magnetisable core locally saturated by external magnetic field to obtain virtual air gap
WO2005085763A2 (en) * 2004-03-01 2005-09-15 Scientific Generics Limited Position sensor

Cited By (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9709614B2 (en) 2008-10-15 2017-07-18 Azoteq (Pty) Ltd Parasitic capacitance cancellation in capacitive measurement
US10564013B2 (en) 2013-08-12 2020-02-18 Gde Technology Ltd Position sensor
CN105659056A (en) * 2013-08-12 2016-06-08 Gde技术公司 Position sensor
GB2517152A (en) * 2013-08-12 2015-02-18 Gde Technology Ltd Position sensor
GB2521482A (en) * 2013-08-12 2015-06-24 Gde Technology Ltd Position sensor
WO2015140080A1 (en) * 2014-03-17 2015-09-24 Tt Electronic Technology Limited Position sensing apparatus and method
DE102014213221A1 (en) * 2014-07-08 2016-01-14 Continental Teves Ag & Co. Ohg Displacement measurement based on eddy currents and a shield canceling donor element
US20200132511A1 (en) * 2014-10-28 2020-04-30 Horst Siedle Gmbh & Co. Kg Position sensor, position measuring device and method for the operation thereof
US10527457B2 (en) 2015-02-27 2020-01-07 Azoteq (Pty) Ltd Inductance sensing
DE102015116429A1 (en) 2015-09-29 2017-03-30 Hella Kgaa Hueck & Co. Electronic unit with sensor arrangement
US12013264B2 (en) 2018-10-16 2024-06-18 Kyocera Avx Components (Werne) Gmbh Position sensing apparatus and method
CN111562618A (en) * 2019-02-13 2020-08-21 巴鲁夫公司 Inductive sensor and method of operating the same
US11703359B2 (en) 2019-05-14 2023-07-18 Kyocera Avx Components (Werne) Gmbh Inductive position sensing apparatus including a screening layer and method for the same
US20220128381A1 (en) * 2020-10-22 2022-04-28 Renesas Electronics America Inc. Position sensor system
CN112504104A (en) * 2020-11-20 2021-03-16 苏州纳芯微电子股份有限公司 Position sensing circuit based on inductance principle, sensor and position measuring method
CN112504104B (en) * 2020-11-20 2022-09-09 苏州纳芯微电子股份有限公司 Position sensing circuit based on inductance principle, sensor and position measuring method

Similar Documents

Publication Publication Date Title
WO2008032008A1 (en) Position sensor
EP2265901B1 (en) Processing circuitry
US7868609B2 (en) Position sensor
AU2012101855A4 (en) Coil arrangement
JPS62204118A (en) Magnetically detecting device for position or speed
JPS63279131A (en) Torque detector
EP2955487A1 (en) Position detecting device
WO2006016144A2 (en) Sensing apparatus and method
JP2004191050A (en) Encoder motion detection device
EP2193383B1 (en) Magnetoresistive sensor device and method of processing a signal thereof
EP0434089B1 (en) Method and device for detection of and protection against the effect of static magnetic fields on magnetoelastic force transducer
US5841275A (en) Magnetic measuring system having a particularly oriented auxiliary field
US11512982B2 (en) Electromagnetic measuring system for detecting length and angle on the basis of the magnetoimpedance effect
EP2896125B1 (en) Method and apparatus to determine the position in an electrical machine
US9791300B2 (en) Inductive position-sensing
CN102216796A (en) Modulation of input signals for a sensor apparatus
JP3920896B2 (en) Linear position detector
JPH06307889A (en) Position detecting device
CN111492206A (en) Electromagnetic measuring system for distance or angle measurement based on the magnetoresistive effect
Jeng et al. Harmonic response of GMR sensor under sinusoidal excitation field
RU2672978C1 (en) Method for detecting defects in a long-dimensional ferromagnetic object
SU913228A1 (en) Device for magnetic noise checking of ferromagnetic materials
SU1270116A1 (en) Method of eddy current check
SU1527521A1 (en) Device for measuring characteristics of ship power plant
JPH03296615A (en) Detecting apparatus of position of moving body

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 06779435

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 06779435

Country of ref document: EP

Kind code of ref document: A1