Field
The present application relates to the field of inductive measurement of
displacement of one mechanical part with respect to another. The term inductive
measurement denotes herein the measurement of alternating electromagnetic fields, by
means of electrical coils. More specifically but not restrictively, the present application
relates to the technical sub-field of eddy-current 5 sensors, wherein an electromagnetic
field generated by an inductor is established differently according to the presence and
the arrangement of movable (with respect to the inductor) conductive parts in the
vicinity of the inductor. Such electromagnetic phenomena become exploitable for
instrumentation purposes when certain electrical frequencies of the electromagnetic
10 field adopt sufficiently large values, this concept of largeness being determined by a
plurality of parameters such as the geometric dimensions of the conductive parts, the
electrical and magnetic properties thereof, the temperature thereof, etc. The term
displacement measurement denotes herein the estimation of information relating to the
position, speed, acceleration or any other characteristic quantity of the displacements of
15 the conductive part with respect to the inductor or to the inductor reference frame. As
displacements, equally well angular (rotation about an axis), linear (translation along an
axis) displacements, or any combination of such displacements with one another or
along separate axes are taken into consideration. More particularly but not restrictively,
the present application relates to the technical sub-fields of inductive position sensors,
20 inductive speed sensors and/or inductive acceleration sensors.
Description of the prior art
2
An inductive displacement sensor typically comprises a transducer (for example
rigidly connected to a measurement reference frame, also known as a frame), and a
target (for example rigidly connected to a movable mechanical part with respect to the
measurement reference frame). The target is placed away from the transducer, and is not
in contact (either mechanically 5 or electrically) with the transducer (contactless
measurement). The transducer includes a primary winding, or inductor, suitable for
producing an alternating electromagnetic field, and at least one secondary winding at
the terminals whereof an alternating voltage is induced, also referred to as electromotive
force or EMF, in the presence of the electromagnetic field produced by the primary
10 winding. The target is a partially or fully conductive element, also referred to as a
coupling armature, the presence and/or movement whereof in front of the transducer
modifies the coupling between the primary winding and the secondary winding. It
should be noted that the effect of the target on the coupling between the primary
winding and the secondary winding is dependent on the position of the target with
15 respect to the transducer, but also on the speed thereof with respect to the transducer.
The electromagnetic field distribution is thus formed spatially according to the
position and the relative displacement of the target with respect to the transducer.
During a displacement of the mechanical part, the spatial distribution of the
electromagnetic field changes, and therefore the EMF induced in the secondary winding
20 also changes. The analysis of the EMF induced, at the terminals of the secondary
winding, by the electromagnetic field produced by the primary winding, makes it
possible to estimate the position and/or the displacement of the target with respect to the
secondary winding of the transducer. More particularly but not restrictively, the
temporal variations of the EMF amplitude at the terminals of the secondary winding
25 make it possible to estimate the position, speed and/or acceleration of the target with
respect to the transducer.
It is specified that, herein and hereinafter in the present application, the term
electromotive force range amplitude at the terminals of the secondary winding refers to
the instantaneous value adopted by a limited frequency content signal, for example in a
30 frequency band between -Δf and +Δf about the excitation frequency (i.e. the frequency
of the alternating voltage applied at the terminals of the primary winding), where Δf
3
could for example adopt a value between 100Hz and 100kHz, carrying the information
or a part of the characteristic information of the mechanical displacement. This signal is
contained in the electromotive force, modulated at the excitation frequency and/or the
harmonics thereof. It can be obtained by means of a frequency transposition and
filtering method, and more 5 specifically by means of base band transposition and
filtering. A preferred example of such a method consists of carrying out synchronous
demodulation of the (modulated) electromotive force using a synchronous excitation
frequency signal, and wherein the electrical phase has been chosen to meet specific
criteria, for example to maximize the signal obtained at the demodulation output. An
10 alternative method consists of computing the modulus of the signal after synchronous
demodulation, which involves the advantage and disadvantage of not setting an
electrical demodulation phase. It is also specified that the amplitude of the
electromagnetic force is a preferred measurement quantity for the implementation of a
displacement measurement with the sensors according to the invention, but that it is in
15 no way exclusive from further electrical measurement quantities such as the phase,
frequency, or the electrical power at the secondary winding when a load of finite value
is connected to the terminals of the secondary winding (load adaptation).
Examples of inductive displacement sensors, and more particularly of eddycurrent
position sensors have been described in the patent EP0182085.
20 However, known inductive displacement sensors involve various drawbacks. In
particular, known sensors are relatively sensitive to assembly inaccuracies
(misalignment, inclination and/or target/transducer distance), as well as the presence of
conductive parts in the vicinity of the measurement zone, which poses problems for
industrial use. Problems associated with the lack of linearity of the sensor response may
25 also arise. Furthermore, the precision and robustness of the estimation of the position
and/or displacement of the target in known sensors would merit being improved.
Moreover, it would be desirable to be able to increase the extent of the measurement
range of some types of known sensors. In addition, one drawback of known sensors is
that they are relatively fragile, which poses problems in some types of application,
30 particularly in an industrial environment.
4
It would be desirable to be able to have inductive displacement sensors
remedying all or some of the drawbacks of known sensors.
Summary
As such, one e 5 mbodiment envisages a target for an inductive displacement
sensor, comprising a plurality of conductive patterns distributed along a zone having a
dimension Dtot in a direction, said patterns being defined by the overlay of at least a first
set of elementary periodic patterns having a period approximately equal to Dtot/N,
including N first elementary conductive patterns of a dimension approximately equal to
10 Dtot/2N in said direction, regularly distributed along said zone, and of a second set of
elementary periodic patterns having a period approximately equal to Dtot/(N+r),
including N+r second elementary patterns of a dimension approximately equal to
Dtot/2(N+r) in said direction, regularly distributed along said zone, where N is an integer
greater than or equal to 2 and r is a positive integer, different to zero and less than or
15 equal to N-1, wherein first and second elementary conductive patterns overlap at least
partially.
According to one embodiment, the first and second elementary conductive
patterns have respectively the shape of portions of overlaid first and second strips
parallel with said direction.
20 According to one embodiment, the first and second strips are of approximately
identical widths.
According to one embodiment, the first and second strips are of separate widths,
the first strip being at least two times wider than the second strip.
According to one embodiment, N is an even number.
25 According to one embodiment, said patterns are defined by the overlay of the
first and second sets of periodic elementary patterns, and of a third set of periodic
elementary patterns having a period approximately equal to Dtot/(N+r), comprising N+r
third elementary patterns of a dimension approximately equal to Dtot/2(N+r) in said
direction, regularly distributed along said zone with an offset of
30 approximately Dtot/2(N+r) with respect to the elementary patterns of the second set of
5
periodic patterns, first and third elementary conductive patterns overlapping at least
partially.
According to one embodiment, the first, second and third elementary patterns
have respectively the shape of portions of first, second and third strips parallel with said
5 direction, the first and second strips, on one hand, and the first and third strips, on the
other, being overlaid, and the second and third strips being approximately of the same
width less than the width of the first strip.
According to one embodiment, the direction is a circular direction.
According to one embodiment, the dimension Dtot is an angular dimension equal
10 to 360°.
According to one embodiment, r is equal to 1.
A further embodiment envisages a transducer for an inductive displacement
sensor, comprising: a primary winding; a first set of at least two secondary windings
each comprising N first turns of the same winding direction or 2N first turns of
15 alternating winding directions, regularly distributed along a zone having a
dimension Dtot in a direction, each first turn having a dimension in said direction
approximately equal to Dtot/2N; and a second set of at least two secondary windings
each comprising N+r second turns of the same winding direction or 2(N+r) second turns
of alternating winding directions, regularly distributed along said zone, each second turn
20 having a dimension in said direction approximately equal to Dtot/2(N+r), where N is an
integer greater than or equal to 2 and r is a positive integer, different to zero and less
than or equal to N-1, wherein first and second turns overlap at least partially.
According to one embodiment, the first and second turns have respectively the
shape of portions of overlaid first and second strips parallel with said direction.
25 According to one embodiment, the transducer further comprises a third set of at
least two secondary windings each comprising N+r third turns of the same winding
directions or 2(N+r) turns of alternating winding directions, regularly distributed along
said zone with an offset of approximately Dtot/2(N+r) with respect to the second set,
each third turn having a dimension approximately equal to Dtot/2(N+r) in said direction,
30 and first and second turns overlapping at least partially.
6
According to one embodiment, the first, second and third turns have respectively
the shape of portions of first, second and third strips parallel with said direction, the first
and second strips being overlaid and the first and third strips being overlaid.
According to one embodiment, the second and third secondary windings are
5 connected in series.
According to one embodiment, the serial connection point of the second and
third secondary windings is connected to an electrical connection terminal.
A further embodiment envisages an inductive displacement sensor, comprising a
transducer of the type mentioned above, and a target of the type mentioned above.
10
Brief description of the drawings
These features and advantages, along with others, will be described in detail in
the following description of particular embodiments given non-restrictively with
reference to the attached figures wherein:
15 figures 1A and 1B are respectively a front view and a profile view schematically
representing an example of an inductive angular displacement sensor;
figure 2 is a diagram schematically illustrating the operation of the sensor in
figure 1;
figures 3A and 3B are front views schematically representing a transducer and a
20 target of a further example of an inductive angular displacement sensor;
figure 4 is a diagram schematically illustrating the operation of the sensor in
figures 3A and 3B;
figure 5 is a front view schematically representing a transducer of a further
example of an inductive angular displacement sensor;
25 figure 6 is a diagram schematically illustrating the operation of the sensor in
figure 5;
figure 7 is a front view schematically representing a transducer of a further
example of an inductive angular displacement sensor;
figure 8 is a diagram schematically illustrating the operation of the sensor in
30 figure 7;
7
figure 9A is a diagram representing the expected theoretical progression of
output signals of an inductive angular displacement sensor;
figure 9B is a diagram representing the actual progression, typically obtained in
practice, of the output signals of an inductive angular displacement sensor;
figure 10 is 5 a diagram representing, for a plurality of distinct target-transducer
distances, the progression of an output signal of an inductive angular displacement
sensor;
figure 11 is a diagram representing the progression, according to the targettransducer
distance, of the linearity error of an output signal of an inductive angular
10 displacement sensor;
figures 12A to 12D are cross-sectional views schematically illustrating four
examples of embodiments of an inductive angular displacement sensor;
figure 13A is a diagram representing, for the four examples of sensors in figures
12A to 12D, the progression, according to the target-transducer, of the linearity error of
15 an output signal of the sensor;
figure 13B is a diagram representing the progression of the optimal targettransducer
distance in terms of linearity according to a parameter of an example of an
inductive displacement sensor;
figure 13C is a diagram representing the progression of the optimal target20
transducer distance in terms of linearity according to a parameter of a further example
of an inductive displacement sensor;
figure 14 is a front view representing an example of a field confinement part of
an example of an embodiment of an inductive angular displacement sensor;
figure 15 is a front view representing a further example of a field confinement
25 part of an example of an embodiment of an inductive angular displacement sensor;
figures 16A and 16B are front views schematically representing two examples of
embodiments of a target of an inductive angular displacement sensor;
figure 17 is a diagram representing the progression, in an inductive angular
displacement sensor, of the optimal target-distances in terms of linearity, according to a
30 shape parameter of a pattern of the target;
8
figure 18A is a front view schematically and partially representing three
examples of embodiments of a target of an inductive angular displacement sensor;
figure 18B is a front view schematically and partially representing an example of
an embodiment of a secondary winding of a transducer suitable for operating in
5 cooperation with the targets in figure 18A;
figure 19 is a diagram representing the progression, in an inductive angular
displacement sensor, of the optimal target-transducer distance in terms of linearity,
according to a further shape parameter of a pattern of the target;
figure 20A is a front view schematically representing an example of a transducer
10 of an inductive angular displacement sensor;
figure 20B is a front view schematically representing an example of a transducer
of an inductive linear displacement sensor;
figure 20C is a front view schematically representing an example of an
embodiment of a transducer of an inductive angular displacement sensor;
15 figure 20D is a front view schematically representing an example of an
embodiment of a transducer of an inductive linear displacement sensor;
figure 20E is a small-signal electrical representation of the behavior of the
transducer in figure 20D;
figures 21A and 21B are front views schematically representing an example of
20 an embodiment of a transducer of an inductive angular displacement sensor;
figures 22A and 22B are front views schematically representing a further
example of an embodiment of a transducer of an inductive angular displacement sensor;
figure 23 is a front view schematically representing a target of an example of
inductive angular displacement sensor;
25 figure 24 is a diagram schematically representing the progression of
measurement signals of the sensor in figure 23;
figure 25 is a front view schematically representing a target of an example of an
embodiment of an inductive angular displacement sensor;
figure 26 is a front view schematically representing a target of an alternative
30 embodiment of an inductive angular displacement sensor;
9
figures 27A to 27C are front views schematically representing a further
alternative embodiment of an inductive angular displacement sensor;
figure 28 is a perspective view representing an example of an embodiment of an
inductive angular displacement sensor target; and
figure 29 is a perspective 5 view representing a further example of an embodiment
of an inductive angular displacement sensor target.
Detailed description
For the purposes of clarity, the same elements have been denoted with the same
10 references in the various figures and, furthermore, the various figures are not plotted to
scale. Moreover, hereinafter in the description, unless specified otherwise, the terms
"approximately", "substantially", "around", "of the order of", "practically", etc., mean
"within 20% and preferably within 5%", or "within 5° and preferably within 2°" when
they relate to angular distances, and directional references such as "vertical",
15 "horizontal", "lateral", "below", "above", "top", "bottom", etc., apply to device oriented
in the manner illustrated in the corresponding views, it being understood that, in
practice, these devices may be oriented differently.
Particular focus is placed on the inductive measurement of displacement of one
mechanical part with respect to another. The term inductive measurement denotes
20 herein the measurement of alternating electromagnetic fields, by means of electrical
coils. More specifically but not restrictively, the present application relates to the
technical sub-field of eddy-current sensors, wherein an electromagnetic field generated
by an inductor is established differently according to the presence and the arrangement
of movable (with respect to the inductor) conductive parts in the vicinity of the inductor.
25 Such electromagnetic phenomena become exploitable for instrumentation purposes
when certain electrical frequencies of the electromagnetic field adopt sufficiently large
values, this concept of largeness being determined by a plurality of parameters such as
the geometric dimensions of the conductive parts, the electrical and magnetic properties
thereof, the temperature thereof, etc. The term displacement measurement denotes
30 herein the estimation of information relating to the position, speed, acceleration or any
other characteristic quantity of the displacements of the conductive part with respect to
10
the inductor or to the inductor reference frame. As displacements, equally well angular
(rotation about an axis), linear (translation along an axis) displacements, or any
combination of such displacements with one another or along separate axes are taken
into consideration. More particularly but not restrictively, the present application relates
to the technical sub-field 5 s of inductive position sensors, inductive speed sensors and/or
inductive acceleration sensors.
An inductive displacement sensor typically comprises a transducer (for example
rigidly connected to a measurement reference frame, also known as a frame), and a
target (for example rigidly connected to a movable mechanical part with respect to the
10 measurement reference frame). The target is placed away from the transducer, and is not
in contact (either mechanically or electrically) with the transducer (contactless
measurement). The transducer includes a primary winding, or inductor, suitable for
producing an alternating electromagnetic field, and at least one secondary winding at
the terminals whereof an alternating voltage is induced, also referred to as electromotive
15 force or EMF, in the presence of the electromagnetic field produced by the primary
winding. The target is a partially or fully conductive element, also referred to as a
coupling armature, the presence and/or movement whereof in front of the transducer
modifies the coupling between the primary winding and the secondary winding. It
should be noted that the effect of the target on the coupling between the primary
20 winding and the secondary winding is dependent on the position of the target with
respect to the transducer, but also on the speed thereof with respect to the transducer.
The electromagnetic field distribution is thus formed spatially according to the
position and the relative displacement of the target with respect to the transducer.
During a displacement of the mechanical part, the spatial distribution of the
25 electromagnetic field changes, and therefore the EMF induced in the secondary winding
also changes. The analysis of the EMF induced, at the terminals of the secondary
winding, by the electromagnetic field produced by the primary winding, makes it
possible to estimate the position and/or the displacement of the target with respect to the
secondary winding of the transducer. More particularly but not restrictively, the
30 temporal variations of the EMF amplitude at the terminals of the secondary winding
11
make it possible to estimate the position, speed and/or acceleration of the target with
respect to the transducer.
It is specified that, herein and hereinafter in the present application, the term
electromotive force range amplitude at the terminals of the secondary winding refers to
the instantaneous value adopted by 5 a limited frequency content signal, for example in a
frequency band between -Δf and +Δf about the excitation frequency (i.e. the frequency
of the alternating voltage applied at the terminals of the primary winding), where Δf
could for example adopt a value between 100Hz and 100kHz, carrying the information
or a part of the characteristic information of the mechanical displacement. This signal is
10 contained in the electromotive force, modulated at the excitation frequency and/or the
harmonics thereof. It can be obtained by means of a frequency transposition and
filtering method, and more specifically by means of base band transposition and
filtering. A preferred example of such a method consists of carrying out synchronous
demodulation of the (modulated) electromotive force using a synchronous excitation
15 frequency signal, and wherein the electrical phase has been chosen to meet specific
criteria, for example to maximize the signal obtained at the demodulation output. An
alternative method consists of computing the modulus of the signal after synchronous
demodulation, which involves the advantage and disadvantage of not setting an
electrical demodulation phase. It is also specified that the amplitude of the
20 electromagnetic force is a preferred measurement quantity for the implementation of a
displacement measurement with the sensors according to the invention, but that it is in
no way exclusive from further electrical measurement quantities such as the phase,
frequency, or the electrical power at the secondary winding when a load of finite value
is connected to the terminals of the secondary winding (load adaptation).
25 Examples of inductive displacement sensors, and more particularly of eddycurrent
position sensors have been described in the patent EP0182085.
However, known inductive displacement sensors involve various drawbacks. In
particular, known sensors are relatively sensitive to assembly inaccuracies
(misalignment, inclination and/or target/transducer distance), as well as the presence of
30 conductive parts in the vicinity of the measurement zone, which poses problems for
industrial use. Problems associated with the lack of linearity of the sensor response may
12
also arise. Furthermore, the precision and robustness of the estimation of the position
and/or displacement of the target in known sensors would merit being improved.
Moreover, it would be desirable to be able to increase the extent of the measurement
range of some types of known sensors. In addition, one drawback of known sensors is
that they are relatively fragile, which 5 poses problems in some types of application,
particularly in an industrial environment.
It would be desirable to be able to have inductive displacement sensors
remedying all or some of the drawbacks of known sensors.
Very particular focus is placed herein on angular displacement sensors, and
10 more specifically on angular displacement sensors having an approximately planar
general shape, for example sensors having a general disk shape, sensors having a
circular annular strip shape having an angular aperture less than or equal to 360°. It will
be understood however upon reading the following that all the examples of
embodiments, embodiments and alternative embodiments described in the present
15 application can be adapted to further types of inductive displacement sensors, for
example inductive linear displacement sensors of the type described in the patent
EP0182085 mentioned above. The adaptation of the examples of embodiments
described in the present application to further types of inductive displacement sensors is
within the grasp of those skilled in the art and therefore will not be detailed hereinafter.
20 By way of an illustrative but non-restrictive example, the inductive sensors
described in the present application and illustrated in the figures have characteristic
dimensions (diameter for angular sensors and width for linear sensors) between 5 mm
and 200 mm, and preferably between 40 mm and 50 mm.
Figures 1A and 1B are respectively a front view and a profile view schematically
25 representing an example of a planar type inductive angular position sensor 100, having
a general disk shape.
The sensor 100 comprises a transducer 110 including a primary conductive
winding 101 and a secondary conductive winding 103. In figure 1B, the primary and
secondary windings of the transducer 110 have not been detailed. Preferably, the
30 primary winding 101 comprises two approximately circular, concentric and coplanar
conductive turns or loops 101a and 101b, of opposite winding directions and separate
13
radii. Each turn 101a, 101b of the primary winding 101 comprises at least one
revolution, preferably a plurality of revolutions. The turns 101a and 101b are preferably
connected in series so as to be traversed by currents of the same intensity but in
opposite flow directions, but can optionally be connected in parallel so as to see the
same voltage at the terminals 5 thereof (applied preferably such that the current flow
direction in the two turns are opposite). An advantage of the example of primary
winding arrangement in figure 1 is that it makes it possible to produce a substantially
uniform excitation field in the annular strip situated between the two turns, and
substantially zero outside this strip. Alternatively, the primary winding 101 can include
10 a single turn (with one or a plurality of revolutions). More generally, the primary
winding 101 can include one or a plurality of concentric turns (with one or a plurality of
revolutions each) arranged so as to generate an electromagnetic field in the
measurement zone of the transducer. The embodiments described are not restricted to
these particular arrangements of the primary winding.
15 In the example represented, the secondary winding 103 consists of a conductive
turn or loop arranged spatially in the shape of a circular annular strip situated between
the turns 101a and 101b. The winding 103 is for example situated approximately in the
same plane as the turns 101a and 101b, or in a substantially parallel plane.
In this example, in a front view, the turn 103 substantially follows the contour of
20 an angular sector having an angular aperture  of the annular strip defined by the
turns 101a and 101b. The turn 103 particularly comprises radial portions and orthoradial
portions of the contour of the annular strip portion. Such a winding enables an
angular position measurement over a range of °. In the example represented, the
angular aperture  of the turn 103 is approximately equal to 30°. The embodiments
25 described are however not restricted to this particular case. Alternatively, the angle 
can adopt any value between 0 and 180°. The turn 103 preferably comprises a single
revolution but can optionally comprise a plurality of revolutions. The primary 101 and
secondary 103 windings are for example arranged in and on the same dielectric
substrate (not shown) in the form of a wafer of some micrometers to some millimeters
30 in thickness, for example a PCB ("Printed Circuit Board") type substrate.
14
The sensor 100 further comprises a target 111 comprising a conductive
pattern 107, situated at a distance different to zero from the transducer and suitable for
moving with respect to the transducer. In figure 1A, only the conductive part 107 of the
target has been represented. In this example, the conductive pattern 107 of the
target 111 has substantially the sa 5 me shape as the annular strip portion defined by the
pattern of the turn 103 of the transducer. The target is rotatably mounted about an axis Z
orthogonal to the plane of the transducer passing through the center of the turns 101a
and 101b, such that, when the target rotates by an angle 2 about the axis Z, the
conductive pattern 107 (having the angular aperture ), covers approximately entirely
10 and then uncovers approximately entirely the surface of the annular strip defined by the
turn of the secondary winding 103 of the transducer. By way of non-restrictive example,
the target can consist of a plate made of a dielectric material, for example in the shape
of a disk, wherein one face oriented towards the transducer is partially coated with a
layer of a conductive material, optionally magnetic, for example a metal layer, for
15 example a layer of iron, steel, aluminum, copper, etc., forming the conductive
pattern 107. Alternatively, the target can consist solely of a portion of metal plate cut to
the shape of the conductive pattern 107, mounted by any suitable means so as to be able
to move in rotation with respect to the transducer above the portion of annular strip
defined by the turns 101a and 101b.
20 The operation of the sensor 100 in figures 1A and 1B will now be described with
reference to figure 2 which represents the progression of the amplitude of the
electromotive force V at the terminals of the secondary winding 103 of the sensor
according to the angular position  of the target 111 with respect to the transducer 110.
In operation, the flow of an alternating current IP is applied by electrical means
25 in the primary winding 101. The flow of the current IP in the winding 101 produces an
electromagnetic field B having, in the absence of a target, a substantially symmetrical
distribution by revolution in the circular annular strip traversed by the secondary
winding 103. By way of non-restrictive example, the frequency of the alternative
excitation current IP applied in the primary winding is between 500 kHz and 50 MHz
30 (for example 4 MHz). The amplitude of the current IP is for example between 0.1 mA
and 100 mA (for example 2 mA). In the absence of a target 111, or, more generally,
15
when the conductive pattern 107 of the target does not cover the secondary winding 103,
the secondary winding 103 supplies between the ends thereof an alternating EMF V,
having a frequency substantially equal to the excitation frequency of the primary
winding, and having an amplitude in principle different to zero. When the conductive
pattern 107 of 5 the target 111 covers all or part of the secondary winding 103, the spatial
electromagnetic field distribution in the vicinity of the turn 103 varies according to the
arrangement and the displacement of the surface portion of the conductive pattern 107
situated facing the turn 103. A further formulation consists of considering that, under
the effect of the magnetic excitation generated by the flow of the current IP in the
10 primary winding, eddy currents appear in the conductive pattern 107, inducing a
modification of the spatial distribution of the electromagnetic field according to the
arrangement and the displacement of the surface portion of the pattern 107 situated
facing the turn 103. These changes or variations of the spatial distribution of the
electromagnetic field according to the arrangement and the displacement of the surface
15 portion of the pattern 107 situated facing the turn 103, are conveyed, by induction, by
variations or changes in the amplitude V of the voltage range at the terminals of the
secondary winding, according to the arrangement and the displacement of the surface
portion of the pattern 107 situated facing the turn 103.
It is considered by way of non-restrictive illustrative example that the target can
20 move in rotation about the axis Z with respect to the transducer, in a range of angular
positions from =-° to =°. It is considered arbitrarily that the position =-°
corresponds to the arrangement represented in figure 1A, wherein the conductive
pattern 107 does not conceal the turn 103, but has, viewed from above, a radial edge
adjoined to a radial edge of the turn 103. As such, for the angular positions  ranging
25 from -° to 0°, the surface area of the portion of the conductive pattern 107 situated
facing the turn 103 increases when the angular position  increases, and, for the angular
positions  ranging from 0° to °, the surface area of the portion of the conductive
pattern 107 facing the turn 103 decreases when the angular position  increases. Outside
the range of angular positions ranging from =-° to =°, the surface area of the
30 portion of the conductive pattern 107 facing the turn of the secondary winding 103 is
16
zero, and the position and/or the displacement of the target 111 with respect to the
transducer cannot be measured.
The amplitude V of the range of the voltage measured at the terminals of a
secondary winding of an inductive displacement sensor is theoretically proportional to
the area of the portion of surface area of the conductive 5 pattern of the target situated
facing the secondary winding. As such, as seen in figure 2, for the angular positions 
ranging from -° to 0°, the signal V decreases when the angular position  increases,
changing from a high value Vmax for  = -° to a low value Vmin for =0°, and for the
angular positions  ranging from 0° to °, the signal V increases when the angular
10 position  increases, changing from the low value Vmin for  = 0° to the high value Vmax
for =°. The signal V is thus theoretically a triangular signal varying linearly between
Vmin and Vmax over the angular range ranging from -° to °. It will be seen hereinafter
that, in practice, the signal V has non-linearity zones and consequently tends to have a
sinusoidal shape.
15 As such, in the range of angular positions from =-° to =0°, or in the range of
angular positions from =0° to =°, the measurement of the amplitude V of the range
of the electromotive force at the terminals of the secondary winding 103 makes it
possible to determine the angular position  of the target with respect to the transducer.
Although the value of the signal V varies according to the angular position  of the
20 target in the two angular position ranges mentioned above, the measurement of the
signal V does not make it possible to discriminate the position values of the range from
-° to 0° from the position values of the range from 0° to ° (non-surjective
measurement). The extent of the range of angular positions that could actually be
measured by the sensor 100 is thus approximately equal to °, provided that the angle 
25 does not exceed 180°.
Figures 3A and 3B are front views schematically representing a further example
of an inductive angular position sensor having a general disk shape. This sensor
comprises a transducer 112 represented in figure 3A and a target 114 represented in
figure 3B. The target 114 in figure 3B differs from the target 111 in figure 1A
30 essentially by the conductive pattern thereof. In particular, the target 114 in figure 3B
17
differs from the target 111 in figure 1A in that it no longer comprises a single
conductive pattern 107, but a set of N conductive patterns 117i rigidly connected to the
target, and suitable for moving with respect to the transducer, N being an integer greater
than or equal to 2 and i being an integer ranging from 1 to N. The transducer 112 in
figure 3A differs from 5 the transducer 110 in figure 1A essentially by the shape of the
secondary winding 113 thereof. In particular, the secondary winding 113 of the
transducer 112 in figure 3A no longer comprises a single conductive turn, but a set of N
turns 113i. The target 114 in figure 3B is intended to be rotatably mounted with respect
to the transducer 112 in figure 3A, similarly or identically to that described with
10 reference to figures 1A and 1B.
In this example, in a front view, the set of conductive patterns 117i and the set of
turns 113i, consist of the repetition by revolution of N substantially identical patterns,
respectively 117i and 113i. The repetition by revolution of these patterns is performed
with a spatial frequency of 2α, i.e. each pattern having an angular aperture substantially
15 equal to α° is spaced from the closest neighbor thereof by a portion of empty circular
annular strip of ortho-radial range substantially equal to α°.
For sensors wherein the general shape is a closed circular annular strip, i.e.
having an angular aperture equal to 360°, the value of the angular aperture α of the
patterns is chosen preferably such that α = 360°/2N, in order to ensure a whole number
20 of pattern repetitions per revolution (over 360°). In the example in figures 3A and 3B, N
= 6 and α = 30°.
In other words, the transducer in figure 3A comprises a secondary winding 113
comprising N loops or turns 113i in series. Each turn 113i has a shape of a circular band
strip sector, of the same type as the turn 103 in figure 1A, and has an angular dimension
25 approximately equal to =360°/2N (i.e. =30° in this example). The N turns 113i are
regularly distributed along the 360° of the circular annular strip approximately defined
by the turns 101a and 101b of the primary winding 101, i.e. two consecutive turns 113i
of the secondary winding are separated by an annular strip portion having an angle
approximately equal to .
30 The target in figure 3B comprises N conductive patterns 117i. Each pattern 117i
has a shape of a circular band strip sector, of the same type as the conductive
18
pattern 107 in figure 1, and has an angular dimension approximately equal to
=360°/2N. The N conductive patterns 117i are regularly distributed along an annular
strip of the target intended to be positioned facing the annular strip of the transducer
containing the turns 113i.
Hereinafter in the present application, 5 the term multi-pole sensor shall refer to
the sensors of the type described with reference to figures 3A and 3B, N denoting the
number of poles of the sensor. In the example in figure 1A, if α adopts the value 180°,
reference is made to a sensor with one pole pair. More particularly, the term multi-pole
sensor shall denote a sensor wherein an elementary conductive pattern is regularly
10 repeated at least twice on the target along a parallel direction with a degree of freedom
of displacement of the target with respect to the transducer (i.e. along an ortho-radial
direction in an angular sensor of the type described above).
By analogy with the electrical period of an electric motor with a plurality of pole
pair, reference shall now be made to the angular aperture between two adjacent
15 patterns 117i, and to the angular aperture between two adjacent patterns 113i, as being
the electrical period of the sensor. In the specific case of the sensor in figures 3A and 3B,
for which the conductive patterns have an angular aperture α° and the hollows between
these patterns also have an angular aperture α°, the electrical period is equal to 2α°, and,
conversely, the angular aperture of a conductive pattern is equal to an electrical half20
period of the sensor, which is a preferred but not exclusive case. By design, for sensors
wherein the general shape is a closed circular annular strip, an electrical period is
preferably a sub-multiple of 360°, since α = 360°/2N. Under these terms, a multi-pole
inductive sensor has a measurement range of α°, equal to half the electrical period
thereof of 2α°. In the example in figure 1A, if α adopts the value 180°, the electrical
25 period is equal to 360°, and the measurement range is approximately equal to half the
electrical period, i.e. 180°. In the example in figures 3A and 3B for which α = 30°, the
electrical period is 2α = 60°, and the measurement range is approximately equal to half
the electrical period i.e. α = 30°.
Figure 4 is a diagram representing the progression of the amplitude V of the
30 range of the electromotive force at the terminals of the secondary winding 113 of the
19
sensor in figures 3A and 3B according to the angular position  of the target with
respect to the transducer.
As seen in figure 4, when the angular position  of the target with respect to the
transducer varies from 0° to 360°, the signal V varies periodically between a high
value Vmax and a low value Vmin, with 5 an angular period of variation approximately
equal to the electrical period 2 of the sensor.
The amplitude of the range of angular positions  suitable for being measured by
the sensor in figures 3A and 3B is approximately equal to half of the electrical period,
i.e. °.
10 One advantage of the sensor in figures 3A and 3B with respect to the sensor in
figures 1A and 1B is that the greater number of patterns distributed on the target and on
the transducer enables a distributed measurement on an extended measurement zone,
wherein each pattern contributes locally and by design to the generation of an overall
electromotive force, this electromotive force being more immune to positioning errors
15 of the target with respect to the transducer than in the sensor in figures 1A and 1B,
wherein the measurement made is a local measurement made using a single set of
patterns 107-103. This robustness of the measurement is especially great as the
number N of pairs of poles of the sensor increases.
Figure 5 illustrates an alternative embodiment of the sensor in figures 3A and 3B.
20 In figure 5, only the transducer of the sensor has been shown, the target being identical
to that in figure 3B.
The transducer of the sensor in figure 5 comprises the same elements as the
transducer in figure 3A, and further comprises a second secondary winding 113'
comprising N loops or turns 113i' in series. For the purposes of clarity, the connections
25 between the different loops 113i of the winding 113 and the connections between the
different loops 113i' of the winding 113' have not been shown in figure 5. The secondary
winding 113' (represented as a dashed line) is substantially identical to the secondary
winding 113 (represented as a solid line), and is arranged in the same annular strip of
the transducer as the secondary winding 113, with an angular offset corresponding to a
30 quarter of the electrical period of the sensor, i.e. approximately equal to /2, with
respect to the secondary winding 113.
20
Figure 6 is a diagram representing the progression of the amplitude V (as a solid
line) of the range of the electromotive force at the terminals of the secondary
winding 113 of the sensor in figure 5, and the progression of the amplitude V' (as a
dashed line) of the range of the electromotive force at the terminals of the secondary
winding 113' 5 of the sensor in figure 5, according to the angular position  of the target
with respect to the transducer.
As seen in figure 6, when the angular position  of the target with respect to the
transducers varies from 0° to 360°, the signals V and V' vary periodically between a
high value Vmax and a low value Vmin, with a variation period equal to the electrical
10 period of the sensor, i.e. approximately equal to 2° in this example, and with an
angular offset with respect to one another substantially equal to one quarter of the
electrical period of the sensor, i.e. approximately /2° in this example.
One advantage of the transducer in figure 5 with respect to the transducer in
figure 3A is that it makes it possible to extend the range of angular positions  suitable
15 for being measured by the sensors up to approximately an entire electrical period (i.e.
2°), instead of a half-period (i.e. °) in the example in figures 3A and 3B.
Figure 7 illustrates a further alternative embodiment of the sensor in figures 3A
and 3B. In figure 7, only the transducer of the sensor has been shown, the target being
identical to that in figure 3B.
20 The transducer of the sensor in figure 7 differs from the transducer in figure 3A
essentially by the shape of the secondary winding thereof. The transducer of the sensor
in figure 7 comprises a secondary winding 123 comprising 2N loops or turns of
alternating winding directions, interconnected in series. In other words, the secondary
winding 123 comprises 2N patterns of electrical circuits or turns, each being connected
25 to the closest neighbor thereof in anti-series. More particularly, the winding 123
comprises N turns 123i+ of the same winding direction, substantially identical to the N
turns 113i of the transducer in figure 3A, and further comprises N turns 123i- of opposite
winding direction, each turn 123i- being arranged between two consecutive turns 123i+,
and each turn 123i- having a shape of a circular annular strip sector, of the same type as
30 the turns 123i+. For the purposes of clarity, the connections between the turns 123i+ and
123i- of the winding 123 have not been shown in figure 7, and the two winding
21
directions have been represented schematically by a + sign for the turns 123i+ and by a –
sign for the turns 123i-.
More specifically, in the example in figure 7, the angular aperture α of each
turn 123i+ and 123i- has been chosen strictly less than an electrical half-period so as to
5 enable a more legible graphic representation. In practice, the angular aperture α of each
turn 123i+ and 123i- can approximate an electrical half-period with a lower value, with
an exact value, or with a greater value. In the specific case where the angular aperture
equals exactly one electrical half-period, which is a preferred but non-exclusive
example of an embodiment, the sum of the angular apertures of the N turns 123i+ and of
10 the angular apertures of the N turns 123i- equals 360°, or in other words, the constituent
radial tracks of two adjacent turns 123i+ and 123i- share the same spatial coordinates in a
reference frame {R, θ} (not shown) directed by the axis Z and having as a center the
center of the sensor. This obviously does not mean however that these tracks are merged
and that the turns 123i+ and 123i- are short-circuited, insofar as the tracks can be
15 positioned on two separate planes along the axis Z.
The spatial repetition period between two adjacent turns 123i+, and the spatial
repetition period between two adjacent turns 123i-, are kept equal to one electrical
period of the sensor regardless of the angular aperture α of the turns 123i+ and 123i-. A
preferred but non-restrictive example of use of such a set of turns having an angular
20 aperture different from an electrical half-period of the sensor consists of distributing the
turns 123i+ and 123i- regularly ortho-radially as illustrated in figure 7.
Figure 8 is a diagram representing the progression of the amplitude V of the
range of the electromotive force at the terminals of the secondary winding 123 of the
sensor in figure 7 according to the angular position  of the target with respect to the
25 transducer.
As seen in figure 8, when the angular position  of the target with respect to the
transducer varies from 0° to 360°, the signal V varies periodically between a high
value Vmax and a low value Vmin, with an angular period of variation approximately
equal to one electrical period.
30 One advantage of the transducer in figure 7 with respect to the transducer in
figure 3A is that the amplitude V is approximately centered around 0 volts (Vmin≈-Vmax).
22
More generally, the use of a spatially differential measurement, such as that which is for
example described with reference to figure 7, makes it possible to a low mean
amplitude V with respect to the values Vmin et Vmax. This simplifies the processing of
the measurement for the purposes of estimating the displacement, and in particular
5 reduces the influence of drift and parasitic disturbances.
Indeed, some variations of the amplitude V associated with parasitic effects, i.e.
not originating from the displacement of the target, are merely conveyed by a gain
variation in the case of the sensor in figure 7, whereas they are conveyed both by a gain
variation and an offset variation in the case of the sensor in figure 3A. This is for
10 example the case when the coupling coefficient between the primary, the target and the
secondary varies due to a parasitic variation of the target-transducer distance. This is
furthermore the case when the amplitude of the excitation current varies, for example in
the case of parasitic fluctuation of the power supply voltage, or in the event of drift of
the electrical properties of the primary winding, for example according to the
15 temperature and the relative distance of the transducer and the target.
Moreover, in the example in figure 7, the coupling of the secondary winding
with external fields not carrying information on the displacement of the target, is
considerably reduced due to the spatially differential nature of the measurement. This is
particular the case for the portion of the electromagnetic field generated by the primary
20 which induces the constant portion (independent of the target position) of the amplitude
of the EMF, but also for any external electromagnetic interference exhibiting a
substantially uniform distribution in the vicinity of the secondary winding 123.
The alternative embodiment in figure 7 can be combined with the alternative
embodiment in figure 5 in order to obtain two signals of amplitude V and V' angularly
25 offset by a quarter of an electrical period and centered on approximately 0 volts.
It should be noted that the fact that the amplitude V of the range of the EMF is
approximately centered on 0 volts does not necessarily mean that the modulated
electromotive force verifies said properties before the implementation of a method of
frequency transposition and filtering. Generally, the electromotive force (modulated)
30 has a mean value different to zero, either due to voluntary referencing of one of the two
terminals of the secondary winding at a defined electric potential (electrical mass for
23
example), or due to referencing by capacitive coupling of the mean potential thereof to
the potential of the environment (for example the mechanical mass) in the case of a
high-impedance measurement at the secondary winding. This illustrative example
applied to the mean value of the electromotive force is also applicable to any frequency
component of the electr 5 ical signal, regardless of the origin thereof, which is situated
outside a frequency band of interest -f to +f about the modulation frequency, or, in
other words, which is situated outside a frequency band of interest -f to +f about the
zero frequency following the frequency transposition method.
10 First aspect
Figure 9A is a diagram representing the expected theoretical progression of the
signals of amplitude V and V' according to the angular position , in an inductive sensor
of the type described above combining the embodiment options in figures 5 (two
secondaries spatially offset by one quarter of an electrical period) and 7 (each secondary
15 comprises 2N turns of alternating winding directions). As seen in figure 9A, the
expected theoretical amplitudes V and V' are triangular periodic signals having a period
equal to the electrical period of the sensor, varying linearly between the values Vmin and
Vmax, with an angular offset of one quarter of an electrical period with respect to one
another. Indeed, in theory, as indicated in the patent EP0182085 mentioned above
20 (column 12, lines 22 to 57), the amplitude of the range of the voltage measured at the
terminals of secondary winding of an inductive sensor is proportional to the area of the
portion of surface of conductive patterns of the target situated facing this secondary
winding. However, in the examples of embodiments described above, the portion of
conductive surface of the target situated facing the electrical circuit patterns or turns of
25 the secondary winding varies linearly with the angular position , for the patterns 123i+
and for the patterns 123i- in figure 7. Therefore, the signals V and V' should vary
linearly by portions according to the position .
The inventors observed, however, that in practice the variation of the signals V
and V' according to the position  generally has wide non-linear zones in an electrical
30 period of the sensor. More specifically, in practice, the variation of the signals V and V'
according to the position  indeed has two substantially linear zones of reduced range in
24
an electrical period of the sensor, these zones being approximately centered on the zero
crossings of the amplitudes V and V', but, between these linear zones, saturated and de
facto less linear zones are inserted, these zones being approximately centered on
extrema of the amplitudes V and V'.
The low linearity 5 of the amplitudes V and V' according to the position  involves
drawbacks. In particular, by way of non-restrictive example, having ranges of reduced
linearity does not make it possible to benefit fully from the signal processing methods
described in the patents FR2914126 and FR2891362.
Figure 9B is a diagram representing the actual progression, typically obtained in
10 practice, of the signals V and V' according to the angular position  in an inductive
sensor of the type described above. As seen in figure 9B, the signals V and V' only vary
linearly in portions of reduced angular range L of the measurement range of the sensor,
referred to as linearity ranges. By way of example, each linearity range L has a range
between 20% and 90% of the electrical half-period of the sensor (equal to ° in the
15 example shown). The linearity range αL is for example defined as being the maximum
angular range, substantially centered on the mean value of the amplitude V, for which it
is possible to find a linear approximation VL to the amplitude V, such that the
difference EL between the linear approximation VL and the amplitude V is less than a
threshold EL0, the threshold EL0 being for example defined as a percentage of the
20 extrema of the amplitude V, for example in a range of values between 0.01% and 10%
of the extrema of the amplitude V according to the degree of linearity sought for the
sensor. In other words, the linearity range αL is the maximum angular range whereon the
amplitude V varies substantially linearly with the position of the target with respect to
the transducer, within one maximum approximation of set value EL0. In practice, it is
25 generally sought to do the opposite, i.e. evaluate the maximum linearity error ELM over
a given angular range αL, for example but not restrictively the angular range whereon it
is sought to make the measurement. Also, a further manner to assess the linearity of a
sensor is that of assessing the linearity error ELM, defined as the maximum difference
between the amplitude V and the linear approximation VL for a given range αL.
30 Preferably but not restrictively, the linearity range sought for a sensor with two
secondary windings is at least 50% of an electrical half-period, for example between 50%
25
and 80% of an electrical half-period when the displacements to be measured are rapid
and the observation of a plurality of samples of the amplitude requires going beyond 50%
of an electrical half-period. In a further preferred example, the linearity range sought for
a sensor with three secondary windings is at least 33% of an electrical half-period, for
example between 33% a 5 nd 50% of an electrical half-period when the displacements to
be measured are rapid. Hereinafter, unless specified otherwise and without being
considered to be an exclusive choice, the description will be limited to presenting a
sensor with two secondary windings, and the purposes of legibility, the description will
be limited to presenting the linearity error over a sought linearity range of 50% of an
10 electrical half-period, without explicitly mentioning these terms, and referring to the
linearity error defined under these terms by merely mentioning the linearity error EL.
The inventors particularly observed that, for a given target-transducer distance
(and for a given range αL), the linearity error EL generally increases as the number N of
poles of the sensor increases.
15 However, this restriction does not indicate industrial use of an inductive sensor
insofar as such a use generally requires a high number of poles, typically N=6, to ensure
a robust measurement as stated above.
It would be desirable to be able to have inductive displacement sensors, and
particularly multi-pole sensors, having a lower linearity error (or broader linearity
20 ranges) than existing sensors, in order in particular to facilitate the processing of the
amplitudes supplied by the sensor. By way of non-restrictive example, extending the
linearity ranges can make it possible to benefit from the signal processing methods
described in the patents FR2914126 and FR2891362.
According to a first aspect, it is sought, in an inductive displacement sensor, and
25 particularly (but not only) a multi-pole sensor, for example sensor with two pole pairs or
more and preferably a sensor with six pole pairs or more, to reduce the linearity error EL
over a given angular range αL, for example over a range αL extending over half an
electrical half-period of the sensor for a sensor with two secondary windings, or over a
range αL extending over one third of an electrical half-period for a sensor with three
30 secondary windings. It can also be taken into consideration that it is sought to increase
the extent of the linearity range of the sensor, i.e. the extent of the position range,
26
included in the measurement range of the sensor, wherein the amplitude of the range of
the electromotive force at the terminals of a secondary winding of the sensor varies
approximately linearly according to the angular position  of the target with respect to
the transducer.
The studies conducted by the i 5 nventors demonstrated that the extents of the
linearity range of an inductive sensor is dependent on the target-transducer distance d,
sometimes referred to as air gap, i.e. the distance between the median plane of the
secondary winding(s) of the transducer, and the conductive patterns of the target. By
way of example, the target-transducer distance d is defined as being the distance
10 between the median plane of the secondary winding(s) of the transducer and the surface
of the conductive patterns of the target oriented towards the transducer.
Figure 10 is a diagram representing, for a plurality of separate target-transducer
distances in an inductive sensor of the type described above (for example of the type
described with reference to figure 7, where N=6 pole pairs), the progression of the
15 amplitude V of the range of the electromotive force measured at the terminals of a
secondary winding of the transducer according to the angular position  of the target.
The curve V1 represents the progression of the amplitude V for a target-transducer
distance d1, the curve V2 represents the progression of the amplitude V for a targettransducer
distance d2 less than d1, and the curve V3 represents the progression of the
20 amplitude V for a target-transducer distance d3 less than d2. The line Vl1, in dotted-line
format, represents the linear approximation of the amplitude V1, the line Vl2, in dottedline
format, represents the linear approximation of the amplitude V2, and the line Vl3,
in dotted-line format, represents the linear approximation of the amplitude V3. As seen
in figure 10, the signal V has, at the distance d3, a maximum amplitude greater than the
25 maximum amplitude obtained at the distances d2 and d1. On the other hand, the
linearity error EL2 of the amplitude V, at the distance d2, is less than the linearity
errors EL1 and EL3 of the amplitude V at the distances d1 and d3 respectively.
Figure 11 is a diagram representing the progression, according to the targettransducer
distance, of the linearity error EL of the amplitude V of the range of the
30 electromotive force measured at the terminals of a secondary winding of the transducer
of an inductive displacement sensor, for example a sensor of the type described with
27
reference to figure 7 (where N=6 pole pairs). In this example, the linearity error EL
corresponds, in a given range of angular positions  extending for example over half of
the electrical period of the sensor (over a monotone portion of the EMF), at the
maximum difference (in absolute value) between a linear approximation of the response
of the sensor and the actual respo 5 nse measured. As seen in figure 11, there is an optimal
target-transducer distance dopt for which the linearity error EL passes through a
minimum. More generally, the inventors observed that a minimum linearity error can be
observed in all the types of inductive displacement sensor, regardless of the number of
pole pairs in particular. This minimum value is achieved for an optimal target10
transducer distance which is dependent on the configuration of the sensor (and
particularly on the number of pole pairs). It is thus theoretically possible to obtain a
linear response regardless of the inductive sensor. The term theoretically denotes that,
when the number of pole pairs N is particularly high, the distance dopt becomes
extremely small so as no longer to be measurable in practice during the limited
15 precision and constraints of use of suitable measuring instruments.
According to a first embodiment, an inductive displacement sensor is envisaged
wherein the target-transducer distance d is between 0.8 and 1.5 times the distance dopt
for which the linearity error of the amplitude measured by the sensor is minimal. It
should be noted that this optimal distance can easily be determined using tests, for
20 example by plotting curves of the type represented in figure 11.
The inventors observed however that, in practice, for some sensors, and
particularly sensors having a high number N of pole pairs, typically greater than or
equal to three and more particularly for N greater than or equal to six, the optimal
target-transducer distance in terms of linearity can be relatively small, for example less
25 than 0.2 mm, which can pose problems for some types of measurement, particularly in
industrial environments wherein such distances are unacceptable, particularly due to
manufacturing, assembly and use tolerances.
Moreover, the inventors observed that the optimal target-transducer distance in
terms of linearity is dependent on a plurality of further parameters, including geometric
30 parameters of the sensor such as the outer diameter of the transducer and/or the target.
More particularly, the inventors observed that when the diameter of the sensors
28
increases, the optimal target-transducer distance increases and can adopt a relatively
high value, for example greater than 1 mm, which can pose problems for some types of
measurements, particularly in industrial environments wherein it is sought to ensure a
somewhat compact design.
In the case where the optimal 5 target-transducer distance in terms of linearity is
incompatible (excessively high or excessively low) with the measurement environment,
it is possible to envisage positioning at the closest possible target-transducer distance to
the optimal distance in the environmental constraint limits, and correcting the nonlinearity
by applying mathematical processing (post-processing) of the measurement
10 signal. The inventors observed however that, in practice, this solution has limitations in
terms of precision and robustness, and is not satisfactory particularly for the
implementation of the signal processing methods described in the patents FR2914126
and FR2891362.
A first solution proposed by the inventors and illustrated by figures 12A to 12D,
15 13A to 13C, 14 and 15, is that of adding to the sensor an additional electromagnetic
field confinement part, placed at a specific distance from the primary winding of the
transducer, chosen so as to increase the target-transducer distance significantly in terms
of linearity.
Figures 12A to 12D are cross-sectional views schematically illustrating four
20 examples of embodiments of an inductive angular displacement sensor.
In the example in figure 12A, the sensor comprises a transducer 201 and a
target 203, arranged at a target-transducer distance d (d being in this example the
distance between the median plane of the secondary winding(s) of the transducer and
the plane of the surface of the conductive patterns of the target oriented towards the
25 transducer), and does not comprise an additional field confinement part.
In the example in figure 12B, the sensor comprises a transducer 201 and a
target 203, arranged at a target-transducer distance d, and further comprises an
additional field confinement part 205 made of a conductive material, for example made
of the same material as the conductive patterns of the target, or of any other conductive
30 material, optionally magnetic, such as iron, steel, aluminum, copper, etc. In this
example, the part 205 is arranged on the side of the target 203 opposite the
29
transducer 201 (i.e. the target 203 is situated between the transducer 201 and the
part 205), the surface of the part 205 oriented towards the target 203 being preferably
approximately parallel with the median plane of the transducer, and therefore also
approximately parallel with the median plane of the target (subject to assembly
imprecision). The field confinement part 5 205 is preferably periodic along a parallel
direction with a degree of freedom of displacement of the sensor, i.e. periodic by
revolution (about an axis which is approximately the axis of symmetry of the target) in
the case of an angular position sensor, the spatial period of the conductive patterns of
the confinement part being preferably separate from that of the conductive patterns of
10 the target. By way of illustrative but non-restrictive example, the part 205 is symmetric
by revolution. The part 205 is arranged at a part-transducer distance l, defined in this
example as being the distance between the median plane of the primary winding(s) of
the transducer, and the plane of the surface of the conductive pattern(s) of the part
oriented towards the transducer. The part 205 is preferably rigidly connected to the
15 target, i.e. movable with respect to the transducer when the position of the target with
respect to the transducer changes.
In the example in figure 12C, the sensor comprises a transducer 201 and a
target 203, arranged at a target-transducer distance d, and further comprises an
additional field confinement part 205', for example identical or similar to the part 205 in
20 figure 12B. The part 205' is preferably periodic by revolution, and for example
symmetric by revolution, about an axis of symmetry which is approximately the axis of
symmetry of the primary winding of the transducer. In this example, the part 205' is
placed on the side of the transducer 201 opposite the target 203 (i.e. the transducer 201
is situated between the target 203 and the part 205'). The part 205' is arranged at a part25
transducer distance l'. By way of example, the distance l' is defined as being the distance
between the median plane of the primary winding(s) of the transducer, and the plane of
the surface of the conductive patterns of the part oriented towards the transducer. The
part 205' is preferably rigidly connected to the transducer, i.e. fixed with respect to the
transducer when the position of the target with respect to the transducer changes.
30 In the example in figure 12D, the sensor comprises a transducer 201 and a
target 203, arranged at a target-transducer distance d, a first field confinement part 205
30
(for example identical or similar to the part 205 in figure 12B) arranged on the side of
the transducer 201 opposite the target 203, at a distance l from the transducer, and a
second field confinement part 205' (for example identical or similar to the part 205' in
figure 12C), arranged on the side of the target 203 opposite the transducer 201, at a
distance l' from 5 the transducer (i.e. the transducer 201 and the target 203 are situated
between the parts 205 and 205').
The parts 205 and/or 205' can be electrically connected or not, point-wise or in a
spatially distributed manner, to other elements of the sensor. In particular, the part 205
can be electrically connected to one or a plurality of conductive patterns of the target,
10 and the part 205' can be electrically connected to an electrical potential available on the
transducer, for example at a point of a secondary winding, at a point of the primary
winding, or to the ground of the transducer.
Figure 13A is a diagram including four curves ELA, ELB, ELC and ELD
representing respectively, for the four examples of a sensor in figures 12A to 12D, the
15 progression of the linearity error EL of the sensor according to the target-transducer
distance. Each of the curves ELA, ELB, ELC and ELD is of the same type as the curve in
figure 11, i.e. it passes via a linearity error value for a certain optimal target-transducer
distance, doptA, doptB, doptC and doptD, respectively. As seen in figure 13A, the
distance doptA is less than the distance doptB which is in turn less than the distance doptC
20 which is in turn less than the distance doptD. The tests conducted by the inventors
demonstrated that adding one or a plurality of additional field confinement parts can
increase the optimal target-transducer distance in terms of linearity of an inductive
displacement sensor from several tens of millimeters to several millimeters.
The positioning along the axis Z of the additional field confinement part(s), and
25 more specifically the distance between this or these parts and the primary winding of the
transducer, has an influence on the effectiveness of the increase in the optimal targettransducer
distance in terms of linearity resulting from adding this or these parts. There
is therefore a (some) optimal distance(s) lopt and/or lopt' between the primary winding
and the additional field confinement part(s), such that the optimal target-transducer
30 distance dopt is increased to attain a value between 0.65 and 1.25 times the distance d at
which it is sought to have the sensor operate, this sought value possibly being but not
31
restrictively between 0.5 and 1.5 mm, which is a range of values compatible with
various industrial applications.
Figure 13B is a diagram representing the progression, for an inductive angular
displacement sensor of the type described above, of the optimal target-transducer
distance dopt in terms of linearity, according to the 5 ratio of the part-primary distance dpipr'
over the target-primary distance dcpr, in the case of addition of the additional field
confinement part 205' as represented in figure 12C or 12D. As seen in figure 13B, the
optimal target-transducer distance in terms of linearity increases as the ratio dpipr'/dcpr
decreases.
10 Figure 13C is a diagram representing the progression, for an inductive angular
displacement sensor of the type described above, of the optimal target-transducer
distance dopt in terms of linearity, according to the ratio of the part-primary distance dpipr
over the target-primary distance dcpr, in the case of addition of the additional field
confinement part 205 as represented in figure 12B or 12D. As seen in figure 13C, the
15 optimal target-transducer distance in terms of linearity increases as the ratio dpipr'/dcpr
decreases.
In other words, if the transducer is considered as an assembly wherein the
constituent layers are not differentiated, the optimal target-transducer distance dopt can
be said to increase as the ratio l/d (respectively l'/d) decreases.
20 Under these conditions, an illustrative but non-restrictive example of positioning
of the additional field confinement parts in figure 12D, is that of placing:
- the upper part 205' at a distance from the primary winding approximately
between 0.5 and 2 times the distance separating the primary winding and the surface
area of the conductive patterns of the target;
25 - the lower part 205 at a distance from the primary winding approximately
between 1.3 and 3 times the distance separating the primary winding and the surface
area of the conductive patterns of the target.
As such, for a given sensor configuration, the ratio dpipr/dcpr and/or the
ratio dpipr'/dcpr can be chosen such that the distance dopt is compatible with the constraints
30 of the application, for example either greater than or equal to 0.3 mm, for example
32
between 0.3 and 10 mm, and preferably between 0.5 and 1.5 mm, particularly for a
sensor including a high number N of pole pairs, for example N4 and preferably N≥6.
It should be noted that the abovementioned choice of distance between the field
confinement part and the transducer is generally not optimal in terms of signal level
5 supplied by the secondary winding(s) of the transducer. Indeed, at this distance, the
conductive part 205/205' causes a non-negligible reduction in the level of the signals V
and V' supplied by the transducer. It should be noted in particular that in the prior art of
inductive angular displacement measurement, it is accepted to separate the conductive
parts liable to modify the spatial distribution of the electromagnetic field which is
10 established in the presence of only the primary, secondary and target elements as much
as possible. This dimensioning criterion applies in particular in the case of electrostatic
screens (or shielding screens), which, when provided, are arranged at distances along
the axis Z much greater than the distances envisaged in the embodiments described, so
as not to attenuate the wanted signal level measured at the secondary excessively.
15 However, the embodiments proposed define a compromise which may be
appropriate in applications for which the linearity is important, and particularly in
applications wherein it is sought to implement signal processing methods of the type
described in the patents FR2914126 and FR2891362 mentioned above.
Figures 14 and 15 are front views representing examples of field confinement
20 parts 205 liable to be used in an inductive displacement sensor of the type described
above (the parts 205' of the sensors mentioned above can have similar or identical
configurations). In the example in figure 14, the part 205 is a mere disk made of a
conductive material (for example metal) having a diameter for example greater than or
equal to the outer diameter of the target. Alternatively (not shown), the disk can be
25 drilled at the center thereof, for example with a hole less than or equal to the inner
diameters of the conductive patterns of the target. In the example in figure 15, the
part 205 is a disk of the same diameter but having cohesive radial striations or slots with
the patterns of the target, suitable for obtaining a Moiré type structure effect with the
target suitable for amplifying the influence of the part 205 on the field distribution at a
30 secondary winding of the transducer. The embodiments described are, however, not
restricted to these two particular examples.
33
A second solution for modifying the optimal target-transducer distance in terms
of linearity, suitable for use in addition or as an alternative to adding a conductive field
confinement part, is illustrated by figures 16A, 16B and 17.
Figures 16A and 16B illustrate two examples of embodiments of an inductive
angular position sensor. In figures 16A and 5 16B, only the target of the sensor has been
represented. The arrangement of the transducer, and particularly of the primary winding
thereof or the secondary windings thereof, is consistent with the arrangement of the
target, and can easily be deduced from the shape of the target on reading the above. In
this example, the target of the sensor in figure 16A is similar or identical to the target in
10 figure 3B. The target of the sensor in figure 16B also comprises N conductive
patterns 137i in the shape of an annular strip sector of angular aperture  approximately
equal to one electrical half-period (for example 360°/2N), the N patterns 137i being
regularly distributed along an annular strip described by the target. The target in
figure 16B differs from the target in figure 16A in that the conductive patterns 137i have
15 different radial dimensions (less in the example shown) from the radial dimensions of
the conductive patterns 117i of the target in figure 16A. More particularly, in this
example, the annular strip determining the shape of the conductive patterns 137i has an
external radius Rext substantially identical to that of the annular strip determining the
shape of the patterns 117i, but has an internal radius Rint less than that of the annular
20 strip of the conductive patterns 117i.
The inventors observed, as illustrated by figure 17, that, for a given number of
pole pairs, the optimal target-transducer distance dopt in terms of linearity of the
response of the sensor, varies according to the ratio Rint/Rext between the internal radius
and the external radius of the annular strip wherein the conductive patterns of the target
25 are situated, and consequently wherein the turns of the secondary winding(s) of the
sensor are situated. It should be noted that the embodiment of figure 16B, which
consists of varying the ratio Rint/Rext by modifying the internal radius Rint of the
conductive patterns of the target, is in no way exclusive of further embodiments suitable
for varying the ratio Rint/Rext by modifying either the external radius Rext, or both radii
30 in combination.
34
Figure 17 is a diagram representing the progression, for an inductive angular
displacement sensor of the type described above, of the optimal target-transducer
distance dopt in terms of linearity, according to the ratio Rint/Rext. As seen in figure 17,
the optimal target-transducer in terms of linearity increases as the ratio Rint/Rext
increases. As such, for a given sensor configuration, 5 the ratio Rint/Rext can be chosen
such that the distance dopt is compatible with the constraints of the application, for
example either greater than or equal to 0.3 mm, for example between 0.3 and 10 mm,
and preferably between 0.5 and 1.5 mm, particularly for a sensor including a high
number N of pole pairs, for example N4 and preferably N≥6.
10 In electromagnetic terms, it would appear that the modifications made to the
internal and/or external radii of the target have the effect of modifying the conductive
pattern shape ratio, and in particular modifying the contribution of the radial edges with
respect to the contribution of the ortho-radial edges, this ratio of the contributions being
a determining factor of the optimal target-transducer distance in terms of linearity dopt.
15 When the ratio Rint/Rext between the internal radius and the external radius of the target
increases, the annular strip portion constituting a conductive pattern is compressed
along the radial direction, inducing a reduction in the contribution of the radial edges to
the overall field distribution measured by the secondary, conveyed at the secondary
output signal by an increase in the optimal target-transducer distance in terms of
20 linearity. The solution described therefore consists of modifying the spatial distribution
of the electromagnetic field, and more particularly the ratio of the radial contributions
with respect to the ortho-radial contributions, so as to adjust the optimal targettransducer
distance in terms of linearity dopt so that it is compatible with the constraints
of the application.
25 In the sensor in figure 16B, when the internal radius Rint and/or the external
radius Rext of the target in figure 16B change, the internal and external radii of the
associated transducer change preferably substantially in the same proportions, so as to
maximize the signal level received by the secondary. By maximizing the signal level at
the secondary output, reference is more specifically made to maximizing the slope at the
30 origin of the signal rather than maximizing the values adopted by the signal extrema for
some positions.
35
For a given set of internal Rint and external Rext target radii, the signal received
by the secondary of the associated transducer is maximum when the annular strip
defining the patterns of the target and the annular strip defining the patterns of the
secondary are substantially overlaid, or in other words, when the external and
respectively 5 internal ortho-radial edges of the target and the external and respectively
internal ortho-radial branches of the secondary are overlaid.
It should be noted that for a given sensor size (and particularly for an upper
external radius limit and a lower internal radius limit), increasing the ratio Rint/Rext
amounts to decreasing the surface area of the conductive patterns of the target, which
10 induces a decrease in the amplitude of the variations of the sensor output signal level
according to the position of the target with respect to the transducer. As such, in the
prior art of inductive angular displacement measurement, the internal diameter and the
external diameter of the annular strip wherein the conductive patterns of the target are
situated, and consequently wherein the turns of the secondary winding(s) of the sensor
15 are situated, are designed so as to occupy the maximum surface area available in the
given size, the size being generally restricted by the internal aperture and the external
diameter of the substrate and/or the casing wherein the sensor is integrated, or by the
external diameter of the shaft about which the sensor is fitted and by the internal
diameter of the interface parts between which the sensor is housed.
20 Nevertheless, the proposed solution consisting of modifying the ratio Rint/Rext
defines a compromise which may be appropriate in applications for which linearity is
important.
A third solution for modifying the optimal target-transducer distance in terms of
linearity, suitable for use in addition or as an alternative to adding an additional field
25 confinement part, and/or to modifying the ratio Rint/Rext, is illustrated by les figures 18A,
18B and 19.
This third solution follows the same logic as the solution that has just been
described, in that it consists of modifying the shape factor of the conductive patterns of
the target and/or the corresponding secondary winding turns, and particularly of
30 modifying the ratio between the radial dimension and the ortho-radial dimension of the
36
patterns of the target and/or secondary winding turns, so as to adapt the optimal targettransducer
distance in terms of linearity to the constraints of the application.
Figure 18A illustrates three examples of embodiments of an angular position
sensor of the type described above. In figure 18A, only one conductive pattern of the
target, designated re 5 spectively by the references 117i for the first example (solid line),
117i' for the second example (dashed line), and 117i'' for the third example (dotted line),
has been shown. In each example, the target is obtained by regularly repeating the
conductive pattern represented along a circular annular strip. The internal and external
radii of the patterns 117i, 117i', and 117i'' are substantially identical, but the patterns 117i,
10 117i', and 117i'' differ from one another by the angular dimensions thereof. More
particularly, in this example, the angular aperture of the pattern 117i' is approximately
equal to one electrical half-period (for example 360°/2N), as described above, the
angular aperture of the pattern 117i'' is greater than one electrical half-period of a
value 1, for example between 0% and 50% of an electrical half-period, and the
15 angular aperture of the pattern 117i is less than 360°/2N of a value 2, for example
between 0% and 50% of an electrical half-period.
As for the embodiment of the solution in figures 16A, 16B and 17, the
arrangement of the secondary of the transducer is preferentially consistent with the
arrangement of the conductive patterns of the target, i.e. the angular aperture of the
20 secondary patterns adapted to the patterns 117i' of the target is substantially equal to one
electrical half-period (for example 360°/2N), the angular aperture of the secondary
patterns adapted to the patterns 117i'' of the target is greater than 360°/2N of a value
substantially equal to 1, and the angular aperture of the secondary patterns adapted to
the target 117i is less than 360°/2N of a value substantially equal to 2. In practice,
25 when the angular aperture of the secondary patterns adopts a value greater than an
electrical half-period of the sensor, it can be envisaged, in order to provide electrical
insulation between adjacent turn tracks, to modify the shape of the tracks in at least one
metallization plane, and/or to increase the number of metallization planes. A further
embodiment option can consist of limiting the maximum angular aperture of the
30 patterns of the secondaries to substantially one electrical half-period, and only varying
the angular aperture of the patterns of the target (of the values 1 or 2). In this case,
37
the angular aperture of the secondary winding patterns of the transducer is not strictly
consistent with the angular aperture of the patterns of the target.
The inventors observed that the optimal target-transducer distance dopt in terms
of linearity of the response of the sensor, varies according the angular deviation 
5 between the angular aperture chosen for the patterns of the target and the secondary, and
the nominal angular aperture α equal to one electrical half-period of the sensor.
Figure 19 is a diagram representing the progression, for a given multi-pole
angular displacement sensor of the type described above and illustrated in figures 18A
and 18B, of the optimal target-transducer distance dopt in terms of linearity, according to
10 the value . As seen in figure 19, the optimal target-transducer distance in terms of
linearity decreases as the value  increases for negative values, and conversely
increases as the value  increases for positive values. As such, for a given sensor
configuration, the angular aperture of the conductive patterns of the target can be
modified by a value  with respect to the nominal value  (equal to one electrical half15
period, for example 360°/2N), the value  being chosen such that the distance dopt is
compatible with the constraints of the application, for example either greater than or
equal to 0.3 mm, for example between 0.3 and 10 mm, and preferably between 0.5 and
1.5 mm, particularly for a sensor including a high number N of pole pairs, for example
N≥4 and preferably N≥6.
20 Solutions have been described for reducing the linearity error (or increasing the
extent of the linearity range) of the response of an inductive displacement sensor, as
well as for modifying, i.e. increasing or reducing according the initial situation, the
target-transducer distance for which an inductive displacement sensor has or approaches
optimal characteristics in terms of linearity.
25 It should be noted that if the linearity error remains nonetheless excessively high
(or if the extent of the linearity range obtained remains insufficient), one or a plurality
of additional secondary windings, spatially offset (by a substantially equal angular
offset between one another), may be added, so as to reduce the extent of the minimal
linearity range required for proper reconstruction of the information in respect of
30 positioning and/or displacement of the target, in combination with the application of the
solutions described above. By way of illustrative example, in the sensor in figure 5,
38
instead of providing two identical secondary windings spatially offset by a quarter of an
electrical period, it is possible to envisage three identical secondary winding spatially
offset by a sixth of the electrical period of the sensor.
Furthermore, it should be noted that the solutions described above can be
adapted 5 to inductive linear displacement sensors, for example by "unwinding" the
patterns in circular strip form described above in order to convert same into patterns in
rectilinear strip form.
Moreover, it should be noted that the solutions described above can be adapted
to inductive angular displacement sensors wherein the transducer has an angular
10 aperture less than 360°, for example less than 180° in order to enable assembly "from
the side" of the transducer about a rotary shaft, rather than a "through" assembly. In this
case, the angular aperture of the target can have a value of 360°, independent of the
angular aperture of the transducers, or adopt a value less than 360°, corresponding for
example to the angular displacement range of the application.
15
Second aspect
The inventors further observed that, in practice, independently of the issue of
linearity, existing inductive displacement sensors, and particularly multi-pole sensors,
are sensitive to various disturbances induced by coupling effect. Such disturbances
20 occur for example at the transduction zone, i.e. directly at the secondary of the
transducer, and furthermore at the electrical connection zone between the secondary of
the transducer and a functional conditioning block of the electronic means. These
disturbances particularly feature the coupling of electromagnetic disturbances from
outside the sensor (i.e. not generated by the primary winding), direct inductive coupling
25 of the primary winding with the secondary winding (i.e. the proportion of inductive
coupling remaining constant regardless of the position of the target), and/or capacitive
coupling between the primary winding and the secondary winding. These disturbances
can cause undesirable fluctuations of the sensor output signal(s) and sensor output
signal interpretation errors.
39
It would be desirable to be able to avail of inductive displacement sensors, and
particularly multi-pole sensors, less sensitive to parasitic disturbances and/or less
subject to parasitic couplings than existing sensors.
As such, according to a second aspect, it is sought to reduce the sensitivity to
disturbances and parasitic coupling effects of multi-5 pole inductive displacement sensors,
and more particularly of the sensors of the type described with reference to figure 7, i.e.
wherein the secondary winding(s) each comprise 2N turns of alternating winding
directions, N being the number of pole pairs of the sensor. For this, the inventors
propose a particular arrangement of the secondary winding(s) of the sensor, which will
10 be described hereinafter.
Figures 20A and 20C schematically illustrate two examples of embodiments of
an inductive angular displacement sensor, of angular aperture 360°, consisting of N=6
pole pairs, and making a spatially differential measurement (for example as described
with reference to figure 7). In figures 20A and 20C, only one secondary 213 of each
15 sensor has been shown, the embodiment of the primary winding, the target, and,
optionally, one or a plurality of secondary windings spatially offset with respect to the
winding 213, being within the grasp of those skilled in the art on the basis of the
explanations of the present description. In this example, the secondary of the sensor in
figure 20A and the secondary of the sensor in figure 20C are similar or identical to the
20 secondary in figure 7, except that the electrical connections between the turns are shown.
The secondary in figure 20A shows a first method for interconnecting the turns,
whereby the entire angular aperture of the annular strip whereon the secondary extends
is traversed a first time, for example in the trigonometric direction in the figure, and the
entire annular strip is traversed a second time, this time in the clockwise direction, so as
25 to approach the electrical terminal end E2 towards the electrical starting end E1, and
thereby close the measurement circuit. The secondary in figure 20C shows a second
method for interconnecting the turns, whereby a first half of the angular aperture of the
annular strip whereon the secondary extends is first traversed, for example in the
trigonometric direction in the figure, then the return path is traversed in the clockwise
30 direction so as to approach the input end E1, then the other half of the angular aperture
of the annular strip whereon the secondary extends is traversed, retaining the clockwise
40
direction of rotation, and then the return path is traversed in the trigonometric direction
so as to approach the electrical terminal end E2 towards the electrical starting end E1,
and thereby close the measurement circuit as for the secondary in figure 20A.
Figures 20B and 20D are front views schematically representing an example of
an embodiment of a transducer of 5 an inductive linear displacement transducer. The
sensors in figures 20B and 20D are sensors wherein a target (not shown) comprising N
conductive patterns is suitable for moving in translation along a rectilinear direction x
with respect to the transducer. The sensor in figure 20B is for example of the same type
as the sensor in figure 20A, adapted to a linear configuration, which essentially consists
10 of "unwinding" the circular annular strips of the sensor in figure 20A and replacing the
conductive patterns and turns in the shape of an annular strip sector, by conductive
patterns and turns having a general rectangular or square shape. The sensor in
figure 20D is for example of the same type as the sensor in figure 20C, adapted to a
linear configuration. In figures 20B and 20D, only one secondary winding 213 of each
15 sensor has been shown, the embodiment of the target, the primary winding, and,
optionally, one or a plurality of additional secondary windings spatially offset with
respect to the winding 213 being within the grasp of those skilled in the art on the basis
of the explanations in the present description. By way of example and unlike the
primary winding of the angular sensors in figures 20A and 20C, an example of primary
20 winding obtained when the set of two concentric turns 101a and 101b for example
described for the sensor in figure 1A is "unwound", consists for example of a single turn
for a linear sensor as described in figures 20B and 20D, optionally consisting of a
plurality of revolutions. The turn of the primary winding has for example a general
rectangular shape, of dimension along y similar to the dimension along y of the
25 conductive patterns of the target and/or of the turns of the secondary as described above,
and of dimension along x greater than the dimension along x of the conductive patterns
of the target and/or of the turns of the secondary, so that the contribution to the overall
electromagnetic field distribution, created at the primary branches oriented along y and
which are situated at both ends along x of the primary, are relatively attenuated in the
30 vicinity of the secondary branches oriented along y and which are situated at both ends
along x of the secondary. In particular, for a transducer with a single secondary winding,
41
the range along x of the primary will be greater than the range along x of the secondary,
and preferably but not exclusively, greater by at least one electrical half-period of the
sensor, distributed equally (at one quarter of an electrical period) at each end of the
sensor. As a general rule, a preferred example of embodiment of the primary winding of
an inductive 5 linear displacement sensor is a turn having a rectangular general shape and
a range greater than the overall range of the set of secondaries, for example but not
exclusively, greater by at least one electrical half-period of the sensor, distributed
equally (at one quarter of an electrical period) at each end of the sensor.
In the examples in figures 20B and 20D, the sensors comprise N=6 pole pairs.
10 However, the embodiments described are not restricted to this particular case.
In the example of the sensor in figure 20D, the secondary winding 213 extends
into a zone having a dimension Dtot parallel with the degree of freedom of the sensor, i.e.
parallel with the direction x of displacement of the target with respect to the transducer.
The winding 213 comprises 2N loops or turns of alternating winding directions
15 electrically connected in series between the ends E1 and E2 thereof. More particularly,
the winding 213 comprises N loops or turns 213i+ having the same first winding
direction, and N loops or turns 213i- having the same second winding direction opposite
the first direction, each turn 213i+ or 213i- having a dimension along the direction x
approximately equal to an electrical half-period of the sensor (i.e. for example
20 approximately equal to Dtot/2N), and the turns 213i- and 213i+ being juxtaposed in pairs
in alternation along the zone of dimension Dtot of the secondary winding.
According to a second embodiment, the secondary winding consists of:
a first coiled conductive section 213A forming N half-turns of alternating
directions, extending between a first end E1 of the winding, situated approximately at
25 the midpoint of the distance Dtot along which the winding 213 extends parallel with the
direction x, and a first intermediate point A of the winding, situated at a first end of the
distance Dtot;
a second coiled conductive section 213B forming N half-turns of alternating
directions, complementary to the N half-turns of the section 213A, extending between
30 the point A and a second intermediate point M of the winding, approximately at the
midpoint of the distance Dtot;
42
a third coiled conductive section 213C forming N half-turns of alternating
directions, extending between the point M and a third intermediate point B of the
winding, situated at the second end of the distance Dtot; and
a fourth coiled conductive section 213D forming N half-turns of alternating
directions, complementary 5 to the N half-turns of the section 213C, extending between
the point B and a second end E2 of the winding, situated approximately at the midpoint
of the distance Dtot, in the vicinity of the first end E1 of the winding.
More particularly, in the example shown, in the left part of the winding (in the
orientation of the figure), the section 213A comprises N U-shaped half-turns wherein
10 the vertical branches are oriented in opposite directions along a direction y
approximately normal to the direction x, and the section 213B comprises N U-shaped
half-turns wherein the vertical branches are oriented alternately in opposite directions
along the direction y. Each U-shaped half-turn of the section 213A has the vertical
branches thereof approximately aligned with the vertical branches of a U-shaped half15
turn of opposite orientation of the section 213B. The sections 213C and 213D are
arranged according to a similar arrangement in the right part of the winding. As such, in
this example, the portions of the winding 213 orthogonal to the direction of
displacement x are traversed twice and twice only by the wire or track of the winding
(except for the two orthogonal end portions of the winding situated at both ends of the
20 distance Dtot, which, in this example, are traversed a single time - this exception does
not arise however in the case of an angular sensor of angular aperture of 360°, wherein
all the radial portions of the primary winding can be traversed twice and twice only by
the wire or track of the winding), and the portions of the winding 213 parallel with the
direction of displacement x are traversed once and once only by the wire or track of the
25 winding.
In terms of path travelled by the constituent electrical circuit of the patterns of
the secondary winding, the embodiment of the solution in figure 20D corresponds to the
embodiment of a solution of the type described with reference to figure 20B, and by
linear-angular transposition also corresponds to the embodiment of the solutions in
30 figures 20A and 20C. On the other hand, the sequence whereby this path is travelled
differs between the transducer in figure 20D (and by transposition the transducer in
43
figure 20C), and the transducer in figure 20B (and by transposition the transducer in
figure 20A). In particular, the arrangement described with reference to figures 20D and
20A is designed so as to show an intermediate connection point M between the ends E1
and E2.
The winding 213 can be provided, 5 in addition to the connection terminals PE1
and PE2 at the ends E1 and E2 thereof, with a third access terminal PM connected to the
midpoint M of the winding.
In the case of multi-pole sensors comprising an even number N of pole pairs,
and as represented in figure 20D, the secondary winding has as many turns 213i+
10 (referred to as positive) on the right as turns 213i+ on the left (N/2 on each side), and
consequently the secondary winding has as many turns 213i- (referred to as negative) on
the right as turns 213i- on the left (N/2 on each side).
One advantage of the secondary winding arrangement in figure 20D when the
number of pole pairs adopts an even value, lies in that the induction is substantially
15 identical, to the nearest sign, regardless of the position of the target with respect to the
transducer, on the two portions E1-M and E2-M on either side of the midpoint, while
enabling the three connections E1, E2 and M to be situated adjacent to one another.
This preferred embodiment wherein the number of pole pairs adopts even values
is in no way exclusive of other embodiments. Alternatively, if the number N of pole
20 pairs is high, the choice of an odd number N is perfectly acceptable insofar as the error
of signal symmetry between the portion E1-M and the portion E2-M varies as an
inverse function of N.
The inventors observed that when the sensor is embodied according to the
second embodiment, if the midpoint M of the winding is referenced at a given electric
25 potential of the differential measurement means, for example at a constant potential
centered on the voltage measurement range of the measurement means, the common
mode component contained in the electrical signal present at the terminals of the
dipole E1-E2, which does not carry useful information on the position and the
displacement of the target with respect to the transducer, is low with respect to the
30 differential mode component contained in the same electrical signal present at the
terminals of the dipole E1-E2, the differential mode component carrying however, the
44
useful information on the displacement of the target with respect to the transducer. The
arrangement of the sensor in figures 20C and 20D suitable for positioning the
midpoint M in the immediate vicinity of the ends E1 and E2 therefore has a definite
advantage, for example with respect to the arrangement of the sensor in figure 20B
wherein the midpoint M is removed from the 5 ends E1 and E2, and more generally with
respect to the arrangement of the sensors in figures 20A and 20B wherein the values E1-
M and E2-M are dependent on the position of the target with respect to the transducer,
or, in other words, with respect to the sensor arrangements wherein the ratio of the
common mode component over the differential mode component at the terminals of a
10 secondary winding is not low and varies significantly with the position of the target
with respect to the transducer.
In particular, one advantage of the sensors described in figures 20C and 20D
when the midpoint M is suitably connected to the measurement means, lies in high
immunity of the two electric potentials at the ends E1 and E2, to the component of the
15 electromagnetic excitation field (primary) which does not vary with the position,
whereas the spatially differential nature of the measurement of the sensor in figure 7
only guarantees immunity on the difference in potential at the ends E1 and E2.
In addition to the immunity to the "direct" field emitted by the primary (internal
source of the system), the sensors in figures 20C and 20D also offer increased immunity
20 to electromagnetic and/or electrostatic disturbances emitted by an external source at the
transduction zone and wherein the spatial distribution is relatively homogeneous, or
more generally increased immunity to any form of electromagnetic and/or electrostatic
disturbance with respect to sensors as described in figures 20A and 20B.
Examples of practical examples of increased immunity to external disturbances
25 in the transduction zone are for example the reduction of constraints on electronic
measurement means protections, such as voltage surge protections, and/or the relaxation
of design constraints on electrical signal conditioning systems, such as the common
mode rejection rate of differential amplifiers.
It should be noted that adapting an inductive sensor to apply a midpoint
30 according to the second embodiment can give rise to an increase in the number of
interfaces of the conditioning circuit (for example the number of tabs of an integrated
45
circuit). It should particularly be noted that according to the prior art of inductive
measurement, it tends to be conventional to minimize the number of physical interfaces
by substituting same with electronic or digital processing. However, this second
embodiment makes it possible to achieve a relatively simple electronic solution, of high
immunity 5 and measurement robustness levels than with known solutions.
Figure 20E is a "small signal" electrical representation of the effective induction
phenomena VM1 and VM2, i.e. of the signals carrying the information or a portion of the
information on the position and/or displacement of the target with respect to the
transducer, and parasitic induction phenomena VP, VP', et VP'' at the connection wires
10 between the terminals E1, E2 and M of the transducer, and the terminals PE1, PE2 and
PM for example connected at the external electrical means. In this figure, and inasmuch
as the wires connected from E1, E2 and M follow each other in close succession, the
common mode disturbances VP, VP', and VP'' are substantially equal and are
substantially offset in the measurements VPE1 (made at the terminals of the dipole PM15
PE1) and VPE2 (made at the terminals of the dipole PM-PE2) on one hand, and in the
measurement VPE1PE2 made at the terminals of the dipole PE1-PE2 on the other. Once
the potential of the terminal PM set to a known value VREF, the signals measured at the
terminals of the tripole (PE1, PE2, PM) become extremely immunized to external
electromagnetic interference in the connection zone between the terminals of the
20 transducer (E1, E2, M) and the connection terminals for the external electrical
means (PE1, PE2, PM), firstly by limiting the risks of overvoltage on the inputs of the
electrical means (the signal levels remain within the range of the conditioning means,
and the measurement is unconditionally valid), and secondly by relaxing requirements
on the common mode rejection rate of the differential measurement VPE1PE2 (the
25 measurement error introduced by the disturbances is low). For example, it is possible to
apply to the terminal PM a reference voltage of the conditioning block, or half the
power supply range of the conditioning block, or the electronic ground, without these
embodiments being exclusive of other embodiments such as for example the connection
of the terminal PM or M directly to a potential of the transducer such as the ground.
30 A representative signal of the position of the target with respect to the transducer
is thereby obtained, particularly robust to disturbances and/or to parasitic coupling
46
effects, whether they occur at the transduction zone or the connection zone between the
transducer and external electrical means, and whether they are inductive in nature as
shown in the electrical diagram in figure 20E, or capacitive in nature with the electrical
environment of the transducer and/or the primary winding or in particular the portions
5 close to the hot spot of the primary (high voltage).
Moreover, in the case where the transducer comprises a plurality of spatially
offset secondary windings (for example as described with reference to figure 5), the
various windings can be arranged in and/or on various overlaid support layers each
comprising one a plurality of metallization levels. This configuration, though
10 satisfactory for many applications, can however pose problems in respect of robustness
and precision. Indeed, as a result, the median planes of the various secondary windings
are situated at slightly different distances from the primary winding and the target. This
results in particular, firstly, in a difference in transduction gain, and therefore a
difference in output signal level of the various secondary windings, and secondly in
15 different linearity characteristics between a plurality of secondary windings of the same
transducer.
To solve this problem, it is preferably envisaged, as illustrated by figures 21A,
21B, 22A and 22B by way of non-restrictive example, to distribute the various
secondary windings of the transducer into two metallization levels, for example in the
20 same support layer with two metallization levels, such that, for each winding, the length
of track or wire of the winding arranged in the first metallization level is approximately
equal to the length of track or wire of the winding arranged in the second metallization
level. Preferably, a sustained alternation of the changes of metallization plane is
envisaged, such that a secondary track cannot travel, on the same plane, a distance (for
25 example an angular aperture in the case of an angular sensor) greater than an electrical
half-period. In a preferred embodiment, the metallization plane transition zones are
located such that there is a symmetric and/or anti-symmetric relationship between most
of the track portions arranged on the first metallization level, and most of the track
portions arranged on the second metallization level, as illustrated in figures 21A, 21B,
30 22A and 22B.
47
As such, the median planes of the various secondary windings are merged and
correspond to a virtual intermediate plane situated between the first and second
metallization levels. This gives each electromotive force induced at the terminals of
each secondary, a response according to the position of the target substantially identical
in terms of amplitude and 5 linearity, to that of the electromotive forces induced at the
terminals of the other secondaries.
It should be noted that the examples of embodiments shown in figures 21A, 21B,
22A and 22B correspond to sensors of angular range Dtot=360°, i.e. wherein the angular
range occupied by each secondary has an angular aperture substantially equal to a
10 complete revolution. These examples are exclusive of alternative embodiments
involving sensors of angular aperture strictly less than 360°, for example less than 180°
in order to enable assembly "from the side" of the transducer about a rotary shaft, rather
than a "through" assembly of the sensor about said shaft in the case of a sensor of
angular aperture 360° as described in figures 21A, 21B, 22A and 22B for example.
15 Under these conditions, it is reiterated moreover that the angular aperture of the target
can alternatively retain a value of 360° independently of the angular aperture adopted by
the secondary/secondaries of the transducer, or adopt a value less than 360° and for
example adapted to the angular displacement range of the application.
Figures 21A and 21B are front views schematically representing an example of
20 an embodiment of a transducer with two secondary windings 223 (hollow line) and 223'
(solid line) spatially offset by a quarter of an electrical period, for an inductive angular
displacement sensor. In the example shown, the number N of pole pairs of the sensor is
equal to 6, and each secondary winding 223, 223' comprises 2N=12 loops or turns. The
embodiments described are not however restricted to this specific case. In this example,
25 the two secondary windings 223 and 223' are formed in and on the same substrate with
two metallization levels M1 and M2 connected by conductive vias (schematically
represented by circles). For each winding, the length of track formed in the level M1 is
approximately equal to the length of track formed in level M2. Figure 21A is a front
view of the metallization level M1, and figure 21B is a front view of the metallization
30 level M2. The patterns of the level M1 are substantially found on the basis of the
48
patterns of the level M2 by antisymmetry with respect to an intermediate plane between
the median planes of the levels M1 and M2.
The windings 223 and 223' each have, viewed from above, an arrangement of
the type described with reference to figure 20C (i.e. an arrangement of the type
described with refer 5 ence to figure 20D adapted to an angular configuration, the coiling
principle described with reference to figure 20D applying in a similar manner, the
distance Dtot no longer being a linear distance but now being an angular distance, equal
to 360°).
As such, the winding 223 comprises:
10 a first curved coiled conductive section 223A forming N half-turns of alternating
directions, extending along a first circular annular half-strip in the example shown
between a first end E1 of the winding 223, situated approximately at the midpoint of the
distance Dtot (for example in the vicinity – i.e. within 5° and preferably with 2° - of an
angular position to which the value 0° is arbitrarily assigned), and an intermediate
15 point A of the winding, situated at a first end of the distance Dtot (for example in the
vicinity of the angle 180°);
a second curved coiled conductive section 223B forming N half-turns of
alternating directions, complementary to the N half-turns of the section 223A, extending
along the first annular half-strip between the point A and a second intermediate point M
20 of the winding, situated approximately at the midpoint of the distance Dtot (for example
in the vicinity of the angle 0°);
a third curved coiled conductive section 223C forming N half-turns of
alternating directions, extending along a second annular half-strip complementary with
the first half-strip between the point M and a third intermediate point B of the winding,
25 situated at an opposite end of the distance Dtot (for example in the vicinity of the angle -
180°); and
a fourth curved coiled conductive section 223D forming N half-turns of
alternating directions, complementary to the N half-turns of the section 223C, extending
along the second annular half-strip between the point B and a second end E2 of the
30 winding, situated approximately at the midpoint of the distance Dtot (in this example in
the vicinity of the angle 0°).
49
As seen in figures 21A and 21B, in this (non-restrictive) example, the portions of
the winding 223 orthogonal to the direction of displacement of the target with respect to
the sensor, i.e. the radial branches of the winding, are traversed twice and twice only by
the wire or track of the winding, and the portions of the winding 223 parallel with the
direction of displacement 5 of the target with respect to the sensor, i.e. the ortho-radial
branches of the winding, are traversed once and once only by the wire or track of the
winding.
More particularly, in this example: the radial portions positioned at angles offset
by 0° modulo an electrical half-period, with respect to the angle characterizing the
10 end E1, are traversed twice and twice only by the wire or track of the winding 223; the
radial portions positioned at angles offset by a quarter of an electrical period modulo an
electrical half-period, with respect to the angle characterizing the end E1, are traversed
twice and twice only by the wire or track of the winding 223'; and the ortho-radial
portions are traversed once and once only by the wire or track of the winding 223, and
15 once and once only by the wire or track of the winding 223'.
This embodiment makes it possible to contain over two planes and only two
metallization planes, two secondaries as described in the preceding solutions, i.e.
without making any concession on the overall shape of the patterns of each secondary.
It should be noted that the embodiments shown in figures 21A, 21B, 22A and 22B
20 implement two secondaries arranged over two metallization planes, but are in no way
exclusive of further embodiments such as an embodiment implementing for example
three secondaries arranged over three metallization planes.
In this example, each of the U-shaped half-turns of each of the sections 223A,
223B, 223C and 223D of the winding 223 (hollowed line) has approximately half of the
25 length thereof in the metallization level M1 and the other half of the length thereof in
the metallization level M2. A change of level occurs approximately every L/2 meters of
conductive track, where L denotes the length of a turn of the winding, consisting of the
connection in series of two complementary U-shaped half-turns. In the example shown,
the level change points of the winding are situated at the midpoints of the ortho-radial
30 branches (or horizontal branches) of the U shapes forming the half-turns. However, the
embodiments described are not restricted to this specific case. In figures 21A and 21B,
50
the numbers ranging from c1 to c28 denote, in the order of travel between the
terminals E1 and E2, different portions of the winding 223.
The secondary winding 223' (solid line) is arranged in the levels M1 and M2
according to an arrangement substantially identical to that of the winding 223, but with
an angular offset of approximately 5 a quarter of an electrical period (i.e. 15° in this
example) with respect to the winding 223.
It should be noted that in the structure in figures 21A and 21B, the connection
tracks to the ends E1 and E2 of the winding 223 can for example be situated
respectively in the metallization levels M1 and M2, and be overlaid on one another.
10 This makes it possible to minimize the parasitic coupling difference on each of these
branches with any source of external induction (connection track to the primary,
external electromagnetic disturbance, etc.). An access track to the midpoint M of the
winding can be situated in a third metallization level (not shown), overlaid on the access
tracks to the terminals E1 and E2 which are situated in the metallization levels M1
15 and/or M2, to be situated in one of the metallization levels M1 and M2, slightly offset
with respect to the access tracks to the terminals E1 and E2. A similar arrangement of
the access tracks to the corresponding terminals E1', E2' and M' of the winding can be
envisaged for the winding 223'. More generally, regardless of the arrangement of the
access tracks, so as to increase the immunity to electromagnetic disturbances between
20 the transduction zone (secondary) and the access and/or connection terminals to the
signal conditioning means, it is preferably sought to keep the paths from the ends E1
and E2 as close as possible (for example overlaid in PCB technology), and to a lesser
degree position the path from the intermediate point M relatively close to the paths from
the ends E1 and E2.
25 It should be noted further that, in the example in figures 21A and 21B, besides
the vias making changes in metallization levels of the windings 223 and 223', and the
conductive tracks travelling in each metallization level for field pickup purposes, vias or
conductive filling chips, with no electrical connection function between field pickup
tracks, have been regularly distributed along the windings 223 and 223'. These
30 conductive filling patterns have the role of symmetrizing the conductive structure of the
transducer, so as to periodize the influence thereof on the spatial distribution of the field,
51
and more particularly of minimizing the field distribution singularities which would be
conveyed by a variation of the secondary output signal according to the position. The
addition of these conductive filling patterns is however optional. In particular, if the vias
making the changes of metallization level have small dimensions with respect to the
skin thickness, the operating frequency, the 5 constituent material thereof, it is possible to
envisage not adding the conductive chips and in particular not performing the drilling
thereof, which can reduce the cost of the device.
Figures 22A and 22B are front views schematically representing an alternative
embodiment of a transducer of the type described with reference to figures 21A and 21B.
10 This alternative embodiment differs from the example in figures 21A and 21B in that, in
the example in figures 22A and 22B, the changes of metallization level are more
numerous than in the example in figures 21A and 21B. As such, in the example in
figures 22A and 22B, instead of a change of metallization level every L/2 meters of
conductive track of the secondary winding, where L is the length of a turn of the
15 winding, it is envisaged to make k changes of metallization level every L/2 meters of
track, where k is an integer greater than or equal to 2. The number k can be chosen
accounting for the internal and external radii of the transducer. By way of nonrestrictive
example, for given sensor sizes and when the changes of level are made only
in the ortho-radial portions of the turns, k can be chosen as great as it possible to place
20 adjacent (for example equi-distributed) vias over other-radial portions without these vias
short-circuiting. For the purposes of simplification, figures 22A and 22B show an
example of an embodiment for a sensor with N=2 pole pairs, wherein the transducer
comprises 2 secondary windings 233 (dashed line) and 233' (solid line) angularly offset
by a quarter of the electrical period of the sensor (i.e. 360°/4N = 45° in this example).
25 The alternative embodiment in figures 22A and 22B is however compatible with
sensors comprising a greater number of pole pairs. As in the example in figures 21A
and 21B, conductive filling patterns with no electrical connection function can be
envisaged to symmetries the structure further.
30 Third aspect
52
In the examples of embodiments of multi-pole sensors hitherto described, for a
given dimension Dtot of a secondary winding of the transducer parallel with the degree
of freedom of the target with respect to the sensor, and for a given number N of pole
pairs, the maximum extent of the range of positions suitable for being detected by the
sensor is ap 5 proximately one electrical half-period (for example Dtot/2N i.e. 360°/2N in
the case of an angular sensor) if the sensor comprises a single secondary winding, and
can increase to approximately one electrical period (for example Dtot/N i.e. 360°/N in
the case of an angular sensor) if the sensor comprises more than one secondary winding,
for example if it comprises two identical secondary windings spatially offset by a
10 quarter of an electrical period (for example Dtot/4N, i.e. 360°/4N in the case of an
angular sensor), or if it comprises three identical secondary windings spatially offset by
a sixth of an electrical period (for example Dtot/6N, i.e. 360°/6N in the case of an
angular sensor). In any case, the multi-pole angular displacement sensors of the type
described above do not make it possible to make displacement measurements over a
15 complete revolution (360°) absolutely, i.e. without using displacement log
memorization methods, and/or methods for referencing the position at start-up and/or
during the operation of the sensor. This remark is valid regardless of the number N of
pole pairs greater than or equal to 2, and can be more problematic when the number N is
high, for example N≥4 and preferably N≥6. The inductive linear displacement sensors
20 described above have the same limitations and do not make it possible to make a
measurement over the complete range of Dtot absolutely.
According to a third aspect, it is sought to embody an inductive displacement
sensor such that, for a given number N of pole pairs, for a given dimension Dtot of the
secondary winding(s) of the transducer parallel with the degree of freedom of the sensor,
25 the sensor is suitable for detecting the position of the target with respect to the
transducer substantially over the entire range Dtot of the transducer. In particular, in the
case of an angular position sensor, it is sought to embody a sensor suitable for detecting
the position of the target with respect to the transducer over a complete revolution, i.e.
over an angular range of approximately 360°, even when the number N of the pole pairs
30 of the sensor is high, for example N≥4 and preferably N≥6.
53
Figure 23 is a front view schematically representing an example of a multi-pole
inductive angular displacement sensor. In figure 23, only the target of the sensor has
been shown.
The target of the sensor in figure 23 comprises, as in the example in figure 3B, N
conductive patterns 117i (N=6 in the example 5 shown) regularly distributed along the
360° of a first circular annular strip 118 of the target. Each conductive pattern 117i has
the shape of a portion or of a sector of the first annular strip 118, of angular aperture N
approximately equal to Dtot/2N = 360°/2N, two consecutive patterns 117i being
separated by a sector of the first annular strip 118, substantially of the same angular
10 aperture N. The target of the sensor in figure 23 further comprises N+1 conductive
patterns 119j, where j is an integer ranging from 1 to N+1, regularly distributed along
the 360° of a second circular annular strip 120 of the target, concentric with the first
strip 118 and not overlaid with the first strip 118. In the example shown, the second
annular strip has an internal radius greater than the external radius of the first annular
15 strip. Each conductive pattern 119j has the shape of a sector of the second annular
strip 120, of angular aperture N+1 approximately equal Dtot/2(N+1) = 360°/2(N+1), two
consecutive conductive patterns 119j being separated by a sector of the second annular
strip 120, substantially having the same angle N+1.
The transducer (not shown for simplification purposes) of the sensor in figure 23
20 corresponds with the target shown, i.e. it comprises:
one or a plurality of primary windings suitable for producing a magnetic
excitation in first and second circular annular strips of the transducer substantially
identical to the first and second annular strips 118 and 120 of the target, intended to be
positioned respectively facing the first and second annular strips 118 and 120 of the
25 target;
at least first and second secondary windings of electrical period Dtot/N (for
example 360°/N in the example of an angular sensor), each comprising N turns of the
same winding direction, in the shape of sectors of angular aperture N of the first
annular strip of the transducer, regularly distributed along the first annular strip of the
30 transducer, or, alternatively, comprising 2N turns of alternating winding directions in
54
the shape of sectors of angular aperture N of the first annular strip of the transducer,
regularly distributed along the first annular strip of the transducer; and
at least third and fourth secondary windings of electrical period Dtot/(N+1) (for
example 360°/(N+1)), each comprising N+1 turns of the same winding direction in the
shape of sectors of angular aperture N+1 of the 5 second annular strip of the transducer,
regularly distributed along the second annular strip of the transducer, or, alternatively,
comprising 2(N+1) turns of alternating winding directions in the shape of the sectors of
angular aperture N+1 of the second annular strip of the transducer, regularly distributed
along the second annular strip of the transducer.
10 Preferably, in the first annular strip, the second secondary winding of electrical
period Dtot/N is substantially identical to the first winding and spatially offset by a
quarter of an electrical period (Dtot/4N) with respect to the first winding, and, in the
second annular strip, the fourth secondary winding of electrical period Dtot/(N+1) is
substantially identical to the third winding and spatially offset by a quarter of an
15 electrical period (Dtot/4(N+1)) with respect to the second winding. More generally, the
transducer can comprise, in the first annular strip, a plurality of secondary windings of
electrical period Dtot/N, substantially identical to the first winding and spatially offset
with respect to one another by a certain electrical period percentage, and, in the second
annular strip, a plurality of secondary windings of electrical period Dtot/(N+1),
20 substantially identical to the third winding and spatially offset with respect to one
another by a certain electrical period percentage.
The operation of the sensor in figure 23 will now be described with reference to
figure 24. The (non-restrictive) case is taken into consideration where the transducer of
the sensor comprises, in the first annular strip of the transducer, a first pair of identical
25 secondary windings of electrical period 360°/2N, spatially offset by a quarter of an
electrical period, and, in the second pair of identical secondary windings of electrical
period 360°/2(N+1), spatially offset by a quarter of an electrical period. As described
above, this sensor is suitable for supplying two sets of two separate electromotive forces,
from which it is possible to construct an estimation of position respectively over a
30 position range equal to 360°/2N and over a position range equal to 360°/2(N+1).
55
Figure 24 is a diagram representing the progression, according to the position of
the target with respect to the transducer, of the estimation N (solid line) of the position
obtained using the electromotive forces measured at the terminals of the first pair of
secondary windings, and of the estimation N+1 (dashed lines) of the position obtained
using the electromotive forces measured at 5 the terminals of the second pair of secondary
windings of the transducer.
As seen in figure 24, when the angular position  of the target with respect to the
transducer varies from 0° to 360°, the position estimation signal N varies periodically
between a low value substantially equal to 0 and a high value substantially equal to 1
10 (the position estimations are standardized herein for the purpose of simplification, the
embodiments described not being restricted to this particular case), with a variation
period equal to the electrical period of the first pair of secondary windings, i.e. equal to
360°/N = 60° for N = 6. Furthermore, the position estimation signal N+1 varies
periodically between the low 0 and high values 1, with a variation period equal to the
15 electrical period of the second pair of secondary windings, i.e. equal to 360°/N+151.4°
for N = 6.
By combining the levels of the position estimation signals N and N+1, two
separate measurement scales are obtained over a complete sensor revolution, i.e. two
different splits of the same range of 360°. The principle of a vernier applied to these two
20 angular measurement scale, i.e. the construction of the difference N+1-N between the
two standardized position estimations N+1 and N, is suitable for estimating the position
and/or the displacement of the target relative to the transducer over the entire
distance Dtot = 360° (i.e. over a complete revolution).
More particularly, one of the position estimation signals, for example the
25 signal N, can be used to provide "refined" target displacement information in N angular
ranges restricted to the electrical period 360°/N, and the difference N+1-N between the
other position estimation signal (the signal N+1 in this example) and this signal can be
used to provide rough absolute information of the position of the target over a complete
revolution. Under these conditions, the rough absolute information makes it possible to
56
adapt the refined but angularly restricted information, so as to carry out an absolute and
refined displacement estimation over 360°.
One advantage of the sensor in figure 23 is that it makes it possible to benefit to
a certain degree of the advantages of multi-pole sensors, particularly in terms of
robustness to positioning 5 errors, while being suitable for providing measurements over
an extended position range with respect to the multi-pole sensors of the type described
above.
As a general rule, it should be noted that the embodiment described above can be
adapted to two signals N1 and N2, N1 and N2 being different integers not necessarily
10 exhibiting a unitary difference. Under these conditions, a sensor characterized by N1
and N2 = N1+2, exhibiting a similar arrangement to the arrangement of the sensor in
figure 23, is suitable for extending the absolute measurement over a range Dtot/N = 180°.
As a general rule, a sensor characterized by N1 and N2 = N1+r, where r is a positive
integer strictly less than N1, makes it possible under certain conditions to extend the
15 absolute measurement over a range Dtot/k = 360°/r.
In this general case, r is obviously a strictly positive integer, i.e. different to zero
(or greater than or equal to 1), such that N2 is greater than or equal to N1+1. If r was not
strictly positive, N2 could be equal to N1 if r=0, and the two signals N1 and N2 would
be identical (not distinct) and would not be suitable for estimating the absolute position
20 by the difference between the two standardized position estimations, as explained above.
Furthermore, r is an integer less than or equal to N1-1, such that N2 is less than
or equal to 2N1-1. If N2 could be equal to 2N1, the difference between the two
standardized position estimations, as described above, would furnish information
similar to that furnished solely by the first set of patterns (corresponding to N1), and
25 would not be suitable for better estimating the absolute difference by the two
standardized position estimations. As such, once r is less than or equal to N1-1, the
embodiments and advantages of the invention are applicable.
In practice, r has a preferably low value, for example r is equal to 1 as described
above and illustrated in figure 23. This makes it possible to make an absolute
30 measurement over the greatest range, 360°. In some applications, it can be preferable to
choose a value of r equal to 2 (absolute measurement over 180° when Dtot = 360°), or
57
choose a value of r equal to 3 (absolute measurement over 120° when Dtot = 360°), or
choose a value of r equal to 4 (absolute measurement over 90° when Dtot = 360°), or
choose a value of r equal to 5 (absolute measurement over 72° when Dtot = 360°), etc.
The sensor in figure 23 poses a number of problems, however. In particular, the
size of the sensor is increased w 5 ith respect to a sensor of the type described above.
Indeed, in the example in figure 23, the "effective" transducer surface area for making a
measurement is that of a circular annular strip approximately two times greater in width
than that of the "effective" annular strip of a transducer of the type described with
reference to figure 3A. Similarly, the "effective" target surface area for making a
10 measurement is that of an annular strip approximately two times greater in width than
that of the "effective" annular strip of the type described with reference to figure 3B.
Furthermore, the embodiment of the primary is more complex than in the preceding
embodiments if it is sought to excite in a relatively uniform manner each of the annular
strips of scale N and N+1 of the sensor. In practice, it can be necessary to use three sets
15 of separate turns to embody the primary excitation winding.
Figure 25 is a front view schematically representing an example of an
embodiment of an inductive displacement sensor. The sensor in figure 25 is a multi-pole
sensor with two measurement scales N and N+1, operating according to the principle of
a vernier as described with reference to figures 23 and 24. In figure 25, only the target
20 of the sensor has been shown.
The target of the sensor in figure 25 comprises a plurality of separated
conductive patterns 127i, distributed along the 360° of a circular annular strip 130 of the
target. As seen in figure 25, the set of patterns formed by the conductive patterns 127i is
not periodic. The various conductive patterns 127i have the shape of angular sectors, of
25 different angular apertures, of the annular strip 130 of the target, and are in principle
irregularly distributed along the annular strip 130.
The set of patterns formed by the conductive patterns 127i over the annular
strip 130 of the target corresponds to the (virtual) overlay of the first and second sets of
periodic conductive patterns of respective periodicities 360°/N and 360°/(N+1). The
30 first set of patterns comprises N elementary patterns 129j (solid lines) regularly
distributed along the annular strip 130 of the target, each elementary pattern 129j having
58
the shape of a sector of the annular strip 130, of angular aperture approximately equal to
360°/2N. The second set of patterns comprises N+1 elementary patterns 131k (dashed
lines), regularly distributed along the annular strip 130, each elementary pattern 131k
having the shape of a sector of the annular strip 130, of angular aperture approximately
equal to 3 5 60°/2(N+1). In other words, the surface areas of conductive patterns of the
target in figure 25 correspond to the total or the combination of the surface areas of the
conductive patterns of a first target of the type described with reference to figure 3B, of
electrical period 360/N, and of a second similar target, having the same internal and
external radii as the first target, but having an electrical period 360°/(N+1).
10 The transducer (not shown for the purpose of simplification) of the sensor in
figure 25 is for example suitable for the conductive patterns of the target in a similar
manner to that described with reference to the example in figure 23. In particular, it
comprises for example:
at least one primary winding suitable for producing an approximately uniform
15 magnetic excitation in a circular annular strip of the transducer substantially identical to
the circular annular strip 130 of the target, intended to be positioned facing the annular
strip 130 of the target;
at least first and second secondary windings of periodicity 360°/N, spatially
offset by a fraction of an electrical period, extending along the circular annular strip of
20 the transducer; and
at least third and fourth secondary windings of periodicity 360°/(N+1), spatially
offset by a fraction of an electrical period, extending along the same annular strip of the
transducer.
The inventors observed that, although the conductive patterns of electrical
25 period 360°/N and 360°/(N+1) of the target overlap and short-circuit one another, and
consequently the target comprises conductive patterns 127i irregularly distributed over a
complete revolution of 360°, these patterns having residual angular apertures which
may be different to the periodic angular apertures of the patterns of the sets of
secondary windings of the transducer, the sensor in figure 25 is suitable for making,
30 with very good performances, displacement measurements over the entire distance Dtot
59
(i.e. over a complete revolution) using a vernier type reading method similar or identical
to the method described with reference to figures 23 and 24.
One advantage of the sensor in figure 25 is that, due to the overlay of the
patterns of respective electrical periods 360°/N and 360°/(N+1), the size of the sensor
can be reduced with respect to a c 5 onfiguration of the type described with reference to
figure 23. Furthermore, a single primary winding, for example of the type described
with reference to figure 3A, suffices to generate a sufficiently uniform magnetic
excitation for proper operation of the sensor.
Figure 26 is a front view schematically representing an alternative embodiment
10 of the sensor in figure 25. In figure 26, only the target of the sensor has been shown.
The target of the sensor in figure 26 comprises a plurality of separated
conductive patterns 137i, distributed along the 360° of a first circular annular strip 138
or wide strip of the target.
The set of patterns formed by the patterns 137i on the annular strip 138 of the
15 target corresponds to the overlay of the first and second sets of periodic patterns of
respective electrical periods 360°/N and 360°/(N+1). The first set of patterns comprises
N elementary conductive patterns 139j (solid lines) regularly distributed along the first
annular strip 138 of the target, each elementary pattern 139j having the shape of a sector
of the first annular strip 138 of the target, of angular aperture approximately equal to
20 one electrical half-period 360°/2N. The second set of patterns comprises N+1
elementary patterns 141k (dashed lines), regularly distributed along the second circular
annular strip 142 or narrow strip of the target, concentric with the annular strip 138 and
included in the annular strip 138, i.e. having an internal radius greater than the internal
radius of the first annular strip, and/or an external radius less than the external radius of
25 the annular strip 138. Each elementary pattern 141k has the shape of an annular sector of
the annular strip 142 of the target, of angular aperture approximately equal to
360°/2(N+1). The width (radial dimension) of the second annular strip 142 of the target
is preferably markedly less than the (radial) width of the first annular strip 138 of the
target, for example two to twenty times less than the width of the first annular strip (the
30 wide strip).
60
The transducer (not shown for the purpose of simplification) of the sensor in
figure 26 is for example suitable for the conductive patterns of the target in a similar
manner to that described with reference to the examples in figures 23 and 25. In
particular, it comprises for example:
at least one 5 primary winding suitable for producing an approximately uniform
magnetic excitation in a first circular annular strip of the transducer (wide strip)
substantially identical to the first annular strip 138 of the target, intended to be
positioned facing the first annular strip of the target;
at least first and second secondary windings of periodicity 360°/N, spatially
10 offset by a fraction of an electrical period, extending along the first circular annular strip
of the transducer (the wide strip); and
at least third and fourth secondary windings of periodicity 360°/(N+1), spatially
offset by a fraction of an electrical period, arranged along a second circular annular strip
of the transducer (narrow strip), substantially identical to the second annular strip 142 of
15 the target and intended to be positioned facing the annular strip 142 of the target.
The operation of the sensor in figure 26 is similar to that of the sensor in figure
25. Preferably, in the sensor in figure 26, the secondary winding(s) making the "refined"
measurement as described above, are windings wherein the turns have the shape of
annular sectors of the widest annular strip of the transducer (substantially identical to
20 the annular strip 138 of the target). By the concept of refined measurement, it is
understood that priority is given to design work to provide the measurement made by
the secondaries of the wide strip with performance and robustness, optionally and to a
certain degree, at the expense of the performance and robustness of the measurement
made by the secondaries of the narrow strip.
25 An additional advantage of the sensor in figure 26 with respect to the sensor in
figure 25 is that it is more robust to positioning errors between the target and the
transducer than the sensor in figure 25. In particular, the measurement obtained at the
terminals of the secondary windings of the wide strip (preferably associated with the
refined measurement) is more robust to positioning errors between the target and the
30 transducer than the sensor in figure 25. Indeed, in the sensor in figure 26, reducing the
surface area of one of the measurement scales with respect to the other makes it possible
61
reduce to a certain degree, the coupling created by the patterns of the narrow strip on the
patterns of the wide strip at the target, particularly with respect to the target in figure 25
for which the mutual influence of one set of patterns on the other is substantially
equivalent and very strong. It is thereby possible to increase the robustness of one of the
5 sets of secondaries to positioning errors.
It should be noted that, in the example shown, the mean radius of the second
circular annular strip of the sensor (the narrow strip) is approximately equal to the mean
radius of the first circular annular strip of the target (the wide strip). This configuration
is advantageous as it makes it possible to remove in a substantially equivalent manner
10 the effects of the internal and external ortho-radial portions of the conductive patterns.
The embodiments described are however not restricted to this particular configuration.
Figures 27A to 27C are front views schematically representing a further
alternative embodiment of the sensor in figure 25. More particularly, figure 27A is a
front view of the target, figure 27B is a front view of a portion of the transducer, and
15 figure 27C is a front view of a further portion of the transducer. In practice, the two
portions of the transducer represented separately in figures 27B and 27C for illustration
purposes, are rigidly connected and overlaid concentrically in a single transducer,
without the breakdown of the constituent elements of said transducer in these two
figures foreseeing a particular distribution over a plurality of metallization levels.
20 The target of the sensor in figures 27A to 27C comprises a plurality of separated
conductive patterns 147i, distributed along the 360° of a first circular annular strip 148
or wide strip of the target.
The set of patterns formed by the conductive patterns 147i on the first annular
strip 148 corresponds to the overlay of a first set of periodic patterns of electrical
25 period 360°/N, and of second and third sets of periodic patterns of electrical
periods 360°/(N+1). The first set of patterns comprises N elementary conductive
patterns 149j (solid lines) regularly distributed along the annular strip 148 of the target
(wide strip), each elementary pattern 149j having the shape of a sector of the strip 148,
of angular aperture approximately equal to 360°/2N.

CLAIMS
1. Target for an inductive displacement sensor, comprising a plurality of conductive
patterns (127i; 137i; 147i) distributed along a zone having a dimension Dtot in a direction,
said patterns (127i; 137i; 147i) being defined by the overlay of at least a first set of
elementary periodic patterns having 5 a period approximately equal to Dtot/N, including N
first elementary conductive patterns (129j; 139j; 149j) of a dimension approximately
equal to Dtot/2N in said direction, regularly distributed along said zone, and of a second
set of elementary periodic patterns having a period approximately equal to Dtot/(N+r),
including N+r second elementary patterns (131k; 141k; 151k) of a dimension
10 approximately equal to Dtot/2(N+r) in said direction, regularly distributed along said
zone, where N is an integer greater than or equal to 2 and r is a positive integer,
different to zero and less than or equal to N-1, wherein first and second elementary
conductive patterns overlap at least partially.
15 2. Target according to claim 1, wherein the first (129j; 139j; 149j) and second (131k;
141k; 151k) elementary conductive patterns have respectively the shape of portions of
overlaid first and second strips (130; 138, 142; 148, 152, 154) parallel with said
direction.
20 3. Target according to claim 2, wherein the first and second strips (130) are of
approximately identical widths.
4. Target according to claim 2, wherein the first and second strips (138, 142; 148, 152,
154) are of separate widths, the first strip being at least two times wider than the second
25 strip.
5. Target according to any one of claims 1 to 4, wherein N is an even number.
6. Target according to any one of claims 1 to 5, wherein said patterns (147i) are defined
30 by the overlay of the first and second sets of periodic elementary patterns, and of a third
set of periodic elementary patterns having a period approximately equal to Dtot/(N+r),
75
comprising N+r third elementary patterns (153k) of a dimension approximately equal to
Dtot/2(N+r) in said direction, regularly distributed along said zone with an offset of
approximately Dtot/2(N+r) with respect to the elementary patterns (151k) of the second
set of periodic patterns, first (149j) and third (153k) elementary conductive patterns
5 overlapping at least partially.
7. Target according to claim 6, wherein the first (149j), second (151k) and third (153k)
elementary patterns have respectively the shape of portions of first, second and third
strips parallel with said direction, the first and second strips, on one hand, and the first
10 and third strips, on the other, being overlaid, and the second and third strips being
approximately of the same width less than the width of the first strip.
8. Target according to any one of claims 1 to 7, wherein said direction is a circular
direction.
15
9. Target according to claim 8, wherein said dimension Dtot is an angular dimension
equal to 360°.
10. Target according to any one of claims 1 to 9, wherein r is equal to 1.
20
11. Transducer for an inductive displacement sensor, comprising:
a primary winding (211);
a first set of at least two secondary windings (243) each comprising N first turns of the
same winding direction or 2N first turns of alternating winding directions, regularly
25 distributed along a zone having a dimension Dtot in a direction, each first turn having a
dimension in said direction approximately equal to Dtot/2N; and
a second set of at least two secondary windings (253) each comprising N+r second turns
of the same winding direction or 2(N+r) second turns of alternating winding directions,
regularly distributed along said zone, each second turn having a dimension in said
30 direction approximately equal to Dtot/2(N+r), where N is an integer greater than or equal
to 2 and r is a positive integer, different to zero and less than or equal to N-1,
wherein first and second turns overlap at least partially.
76
12. Transducer according to claim 11, wherein the first and second turns have
respectively the shape of portions of overlaid first and second strips parallel with said
direction.
5
13. Transducer according to claim 11 or 12, further comprising a third set of at least two
secondary windings (255) each comprising N+r third turns of the same winding
directions or 2(N+r) turns of alternating winding directions, regularly distributed along
said zone with an offset of approximately Dtot/2(N+r) with respect to the second set,
10 each third turn having a dimension approximately equal to Dtot/2(N+r) in said direction,
and first and second turns overlapping at least partially.
14. Transducer according to claim 13, wherein the first, second and third turns have
respectively the shape of portions of first, second and third strips parallel with said
15 direction, the first and second strips being overlaid and the first and third strips being
overlaid.
15. Transducer according to claim 13 or 14, wherein the second (253) and third (255)
secondary windings are connected in series.
20
16. Transducer according to claim 15, wherein a serial connection point of the
second (253) and third (255) secondary windings is connected to an electrical
connection terminal.
25 17. Inductive displacement sensor, comprising a transducer according to any one of
claims 11 to 16 and a target according to any one of claims 1 to 10.