U.S. patent application number 09/457670 was filed with the patent office on 2001-10-18 for capacitance position transducer.
Invention is credited to MOORE, COLIN, WILSON, EDWARD.
Application Number | 20010030544 09/457670 |
Document ID | / |
Family ID | 10835874 |
Filed Date | 2001-10-18 |
United States Patent
Application |
20010030544 |
Kind Code |
A1 |
WILSON, EDWARD ; et
al. |
October 18, 2001 |
CAPACITANCE POSITION TRANSDUCER
Abstract
A non-contacting capacitive position transducer comprises a
stator substrate carrying two electrically conducting inverted
wedge regions 14,16 whose width varies inversely in a sensing
direction. A moveable pick-off 20 is capacitively coupled to both
wedges. The wedges 14,16 are driven with respective distinguishable
time varying periodic waveforms, e.g. a sine wave and a cosine
wave, and the pick-off voltage at 20 processed to determine the
position of the pick-off in the sensing direction. Various
configurations of transducer are described, including those of
linear, cylindrical and disc form. The pick-off voltage is
preferably transferred from the pick-off 20 back to the stator by
capacitively coupling the pick-off to a suitably screened pick-off
track 36 on the stator.
Inventors: |
WILSON, EDWARD; (LANCASHIRE,
GB) ; MOORE, COLIN; (LANCASHIRE, GB) |
Correspondence
Address: |
CUSHMAN DARBY & CUSHMAN LLP
1100 NEW YORK AVENUE N W
NINTH FLOOR EAST TOWER
WASHINGTON
DC
200053918
|
Family ID: |
10835874 |
Appl. No.: |
09/457670 |
Filed: |
December 9, 1999 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
09457670 |
Dec 9, 1999 |
|
|
|
PCT/GB99/02390 |
Jul 22, 1999 |
|
|
|
Current U.S.
Class: |
324/658 |
Current CPC
Class: |
G01D 5/2412
20130101 |
Class at
Publication: |
324/658 |
International
Class: |
G01R 027/26 |
Foreign Application Data
Date |
Code |
Application Number |
Jul 22, 1998 |
GB |
9815826.4 |
Claims
1. A capacitive position transducer comprising: a relatively fixed
substrate including two spaced electrically conductive regions; a
mover element comprising a common electrical pick-off means
capacitively coupled to both of said electrically conductive
regions and mounted for movement relative to said substrate in a
sensing direction, the transverse extent of at least one of said
electrically conductive regions varying with the position of said
mover element in the sensing direction, drive means for supplying
respective different periodic time varying voltages to said
electrically conductive regions, and processing means for
processing the electrical signal received by said pick-off means to
determine the position of said mover element, wherein said drive
means applies respective different sinusoidal voltages to said
electrically conductive regions, said sinusoidal voltages having a
relative phase difference.
2. A capacitive position transducer according to claim 1, wherein
said processing means determines at least one of the phase and
magnitude of the electrical signal received by said pick-off means,
to determine the relative position of said mover element.
3. A capacitive position transducer according to claim 1 or claim
2, wherein both of said electrically conductive regions vary in
transverse extent in said sensing direction.
4. A capacitive position transducer according to claim 3, wherein
said electrically conductive regions vary linearly and inversely in
transverse extent in said sensing direction.
5. A capacitive position transducer according to claim 3, wherein
said electrically conductive regions each vary with a profile
selected in accordance with said periodic time-varying voltages at
least partially to compensate for non-linearities in the signal
processed by said processing means.
6. A capacitive position transducer according to any of the
preceding claims, wherein said pick-off means is connected directly
to said processing means.
7. A capacitive position transducer according to any of claims 1 to
5, wherein said pick-off means is capacitively coupled to a rail
means or track means extending adjacent the path of movement of
said mover element.
8. A capacitive position transducer according to any of the
preceding claims wherein said substrate is generally flat
planar.
9. A capacitive position transducer according to any of the
preceding claims, wherein said mover element is mounted for linear
movement in a straight line relative to said substrate.
10. A capacitive position transducer according to any of claims 1
to 8, wherein said mover element is mounted for rotary movement
about a rotary axis.
11. A capacitive position transducer according to any of the
preceding claims, wherein said electrically conductive regions are
disposed side by side on one face of said substrate together to
define a shape of generally constant combined width in the
direction transverse to the sensing direction.
12. A capacitive position transducer according to any of claims 1
to 7, wherein said substrate comprises a cylindrical or part
cylindrical surface and said mover element is disposed adjacent
said cylindrical or part-cylindrical surface and mounted for
movement about an axis generally coaxial with the principal axis of
said cylindrical or part-cylindrical surface.
13. A capacitive position transducer according to any of the
preceding claims, wherein said sinusoidal voltages are of
substantially equal amplitude.
14. A capacitive position transducer according to claim 13, wherein
said sinusoidal voltages have a relative phase difference of
90.degree., to provide respective sine and cosine drive
waveforms.
15. A capacitive position transducer according to claim 1, wherein
said drive means applies to the electrically conductive regions
respective sinusoidal waveforms of the form V.sin(.omega.t) and
V.cos(.omega.t), and the relative position of the mover element is
determined from the signal V.sub.o at said pick-off, wherein
V.sub.o=K.V.Sin (.omega.t+.theta.); .theta.=arctan
(.alpha./1-.alpha.); K={square root}{square root over (
)}(.alpha..sup.2+(1-.alpha.).sup.2) and .alpha. is the distance of
the mover element from the highest capacitance part of the
electrically conducting region driven by said V.Sin(.omega.t)
waveform expressed as a proportion of the maximum range of sensing
movement of said mover element.
16. A capacitive position transducer according to claim 15, wherein
said processing means is operable to multiply said output voltage
V.sub.o at said pick-off means with a D.C.-nulling periodic voltage
at the same frequency as said output voltage and to adjust the
relative phase of said D.C.-nulling periodic time varying voltage
to null the D.C. term of said product, and thereby to determine the
relative position of said mover element.
17. A capacitive position transducer according to claim 16, wherein
said processing means includes generating means for producing said
D.C.-nulling periodic voltage, said generating means including
means for multiplying the time varying voltages (V.sin(.omega.t);
V.cos (.omega.t)) supplied to said electrically conductive regions
with respective inversely related voltages ((V.sub.ref-V.sub.con)
and V.sub.con), means for summing the products of said multiplying
means and phase-shifting the sum to obtain said D.C.-nulling
periodic voltage.
18. A capacitive position transducer according to claim 17, wherein
said processing means includes integrator means for receiving the
product of said pick-off output voltage V.sub.o with the
D.C.-nulling voltage, and adjusting said inversely related voltages
((V.sub.ref-V.sub.con) and V.sub.con) to null said D.C. term.
19. A capacitive position transducer comprising: a relatively fixed
substrate including two spaced electrically conductive regions; a
mover element comprising a common electrical pick-off means
capacitively coupled to both of said electrically conductive
regions, and mounted for movement in a sensing direction, the
electrically conductive regions being arranged such that at least
one of the respective capacitances between said pick-off means and
the electrically conductive regions varies with the position of aid
mover element in said sensing direction, drive means for applying
respective voltages to each of said electrically conductive
regions, and position determining means for monitoring the position
of said mover element relative to said substrate, wherein said
drive means supplies respective different sinusoidal voltages to
said electrically conductive regions, said sinusoidal voltages
having a relative phase difference.
20. A method of position detection which comprises providing a
relatively fixed substrate including two spaced electrically
conducting regions and a pick-off means capacitively coupled to
both of said regions, said pick-off means being connected to a
mover element for movement in a sensing direction, at least one of
respective capacitances between the pick-off and the electrically
conductive regions varying with position in said sensing direction,
the method further comprising applying respective different
sinusoidal voltages to said electrically conductive regions, said
sinusoidal voltages having a relative phase difference, and
monitoring at least one of said capacitances thereby to determine
the position of said mover element.
21. A capacitive position transducer comprising: a relatively fixed
substrate including two spaced electrically conductive regions
varying linearly and inversely in transverse extent; a mover
element comprising a common electrical pick-off means capacitively
coupled to both of said electrically conductive regions and mounted
for movement relative to said substrate in a sensing direction, the
transverse extent of at least one of said electrically conductive
regions varying with the position of said mover element in the
sensing direction, drive means for supplying respective different
periodic time varying voltages to said electrical conductive
regions, and processing means operable to multiply an output
voltage at said pick-off means with a DC nulling periodic voltage
at the same frequency as said output voltage and to adjust the
relative phases of said DC nulling periodic time varying voltage to
null the DC term of said product, and thereby to determine the
relative position of said mover element.
22. A capacitive position transducer according to claim 21, wherein
said drive means applies to the electrically conductive regions
respective sinusoidal waveforms of the form V.sin(.omega.t) and
V.cos(.omega.t), and the relative position of the mover element is
determined from the signal V.sub.o at said pick-off, wherein
V.sub.o=K.V.Sin (.omega.t+.theta.); .theta.=arctan
(.alpha./1-.alpha.); K={square root}{square root over (
)}(.alpha..sup.2+(1-.alpha.).sup.2) and a is the distance of the
mover element from the highest capacitance part of the electrically
conducting region driven by said V.Sin(.omega.t) waveform expressed
as a proportion of the maximum range of sensing movement of said
mover element.
23. A capacitive transducer according to claim 21 or claim 22,
wherein said processing means includes generating means for
producing said D.C.-nulling periodic voltage, said generating means
including means for multiplying the time varying voltages
(V.sin(.omega.t); V.cos (.omega.t)) supplied to said electrically
conductive regions with respective inversely related voltages
((V.sub.ref-V.sub.con) and V.sub.con), means for summing the
products of said multiplying means and phase-shifting the sum to
obtain said D.C.-nulling periodic voltage.
24. A capacitive transducer according to claim 23, wherein said
processing means includes integrator means for receiving the
product of said pick-off output voltage V.sub.o with the
D.C.-nulling voltage, and adjusting said inversely related voltages
((V.sub.ref-V.sub.con) and V.sub.con) to null said D.C. term.
25. A capacitive position transducer substantially as hereinbefore
described with reference to, and as illustrated in, any of the
accompanying drawings.
26. A method of position detection substantially as hereinbefore
described with reference to any of the accompanying drawings.
Description
[0001] This invention relates to capacitive position transducers
and associated methods for position detection. The invention is
particulary concerned with linear and rotary position transducers
but is not limited to such transducers.
[0002] There is frequently a requirement in modern control systems
for a positional transducer which provides a high resolution,
absolute output, that is one in which the output indication is a
unique expression of position. In applications where the accuracy
requirements do not preclude their use, e.g. positional servo
systems, potentiometers have traditionally been used, providing a
readily available, cost effective solution where applicable.
[0003] However potentiometers do have several shortcomings. There
is an inherent wear-out mechanism between the wiper and the
resistive element which will ultimately result in failure of the
device. They are prone to creating wiper noise, particularly under
high rates of movement, and noise performance tends to deteriorate
with life; this is a particular problem in high gain servo systems
where the noise can interfere with correct system operation.
Frequently the need is to monitor a linear motion and, whilst some
linear travel potentiometers are available, these are almost
invariably commercial parts unsuitable for anything other than
consumer type applications. The application of a rotary
potentiometer to such a requirement would therefore necessitate
some kind of motional translation--a rack and pinion for
example.
[0004] There is therefore a need for a potentiometer which obviates
at least some of the above shortcomings. We have therefore designed
a new form of transducer which does not require electrical physical
contact between the mover and the stator and which provides an
absolute output with high resolution, with an accuracy at least
comparable with that of conventional servo-grade
potentiometers.
[0005] Accordingly, in one aspect, this invention provides a
capacitive position transducer comprising:
[0006] a relatively fixed substrate including two spaced
electrically conductive regions;
[0007] a mover element comprising a common electrical pick-off
means capacitively coupled to both of said electrically conductive
regions and mounted for movement relative to said substrate in a
sensing direction, the transverse extent of at least one of said
electrically conductive regions varying with the position of said
mover element in the sensing direction,
[0008] drive means for supplying respective different periodic time
varying voltages to said electrically conductive regions, and
[0009] processing means for processing the electrical signal
received by said pick-off means to determine the position of said
mover element,
[0010] wherein said drive means applies respective different
sinusoidal voltages to said electrically conductive regions, said
sinusoidal voltages having a relative phase difference.
[0011] Preferably, the processing means determines at least one of
the phase and magnitude of the electrical signal received by the
pick-off means, to determine the relative position of the mover
element.
[0012] In another aspect, this invention provides a capacitive
position transducer comprising:
[0013] a relatively fixed substrate including two spaced
electrically conductive regions;
[0014] a mover element comprising a common electrical pick-off
means capacitively coupled to both of said electrically conductive
regions, and mounted for movement in a sensing direction, the
electrically conductive regions being arranged such that at least
one of the respective capacitances between said pick-off means and
the electrically conductive regions varies with the position of aid
mover element in said sensing direction, and
[0015] drive means for applying respective voltages to each of said
electrically conductive regions, and
[0016] position determining means for monitoring the position of
said mover element relative to said substrate,
[0017] wherein said drive means supplies respective different
sinusoidal voltages to said electrically conductive regions, said
sinusoidal voltages having a relative phase difference.
[0018] In yet another aspect, this invention provides a method of
position detection which comprises providing a relatively fixed
substrate including two spaced electrically conducting regions and
a pick-off means capacitively coupled to both of said regions, said
pick-off being connected to a mover element for movement in a
sensing direction, at least one of the respective capacitances
between the pick-off and the electrically conductive regions
varying with position in said sensing direction, the method further
comprising monitoring at least one of said capacitances thereby to
determine the position of said mover element, and
[0019] applying respective different sinusoidal voltages to said
electrically conductive regions, said sinusoidal voltages having a
relative phase difference.
[0020] Preferably, both of said electrically conducting regions
vary in transverse extent in said sensing direction whereby the
capacitance between each electrically conductive region and the
pick-off means varies in said sensing direction.
[0021] Preferably, said electrically conductive regions vary
linearly and inversely in transverse extent in said sensing
direction. Thus, in one arrangement the electrically conductive
regions may be arranged as two triangles in inverted relationship
together defining a generally rectangular plan shape.
[0022] Alternatively, the electrically conductive regions may each
vary in said sensing direction with a profile selected in
accordance with the drive voltages to said electrically conducting
regions, at least partially to compensate for non-linearities in
the output signal.
[0023] The pick-off means may be connected directly to said
processing means e.g. via an electrically conducting wire or the
like. Alternatively, to avoid the need to provide a moveable
electrically conducting element such as a wire or track, the
pick-off means may be capacitively coupled to a track means which
extends alongside the path of movement of said mover element, with
suitable screening, so that the output signal may be taken from the
track means.
[0024] The substrate may take any of a number of different forms.
For example, the substrate may be generally flat or planar. Here
the mover element may be mounted for linear movement in a straight
line relative to said substrate, analogous to a conventional linear
potentiometer. Alternatively, said mover element may be mounted for
rotary movement about a rotary axis and the substrate may be of
generally disc or annular form. In this instance the transducer is
akin to a rotary potentiometer.
[0025] Alternatively, said substrate may comprise a cylindrical or
part-cylindrical surface and said mover element may be disposed
adjacent said cylindrical a part-cylindrical surface and mounted
for movement about an axis generally coaxial with the principal
axis of said cylindrical or part-cylindrical surface.
[0026] The drive means may apply various forms of drive voltages to
the electrically conductive regions such that the combined signal
at the electrical pick-off means may be processed to determine the
position of the mover element. Thus the drive means may apply
respective different sinusoidal voltages. The sinusoidal voltages
are preferably of substantially equal amplitude. Conveniently, said
sinusoidal voltages have a relative phase difference of 90.degree.,
effectively to provide respective sin and cosine waveforms. However
other phase differences may also be used.
[0027] The processing means preferably determines at least one of
the relative phase and magnitude of the electrical signal received
by said pick-off means, to determine the position of said mover
element relative to a fixed datum position.
[0028] Where the electrically conductive regions vary linearly and
inversely in axial extent in said sensing direction, the drive
means may apply respective sinusoidal waveforms of the form
V.Sin(.omega.t) and V.Cos(.omega.t) and the relative position of
the mover element may be determined from the signal V.sub.o at said
pick-off, wherein
V.sub.o=K.V.Sin (.omega.t+.theta.);
.theta.=arctan (.alpha./1-.alpha.);
K={square root}{square root over (
)}(.alpha..sup.2+(1-.alpha.);
[0029] and .alpha. is the distance of the mover element from the
highest capacitance part of the electrically conducting region
driven by said V.Sin(.omega.t) waveform expressed as a proportion
of the maximum range of sensing movement of said mover element.
[0030] It will be noted that .alpha. and .omega. vary non-linearly;
this may be overcome by modifying the profiles of the electrically
conductive regions to reduce or remove the non-linearity so that
the phase difference (.omega.) varies linearly with the position
(.alpha.) of the mover element.
[0031] Alternatively, the non linearity may be overcome by signal
processing. Thus said processing means may be operable to multiply
said output voltage V.sub.o at said pick-off means with a
D.C.-nulling periodic time varying voltage at the same frequency as
said output voltage, and adjusting the relative phase of said
D.C.-nulling periodic time varying voltage to null the D.C. term of
said product, and thereby determine the position of said mover
element relative to a fixed datum.
[0032] Preferably said D.C.-nulling periodic voltage is generated
by multiplying the drive voltages applied to said electrically
conducting regions. Thus said processing means may include
generating means for producing said D.C.-nulling periodic voltage,
said generating means including means for multiplying the time
varying voltages (V.Sin(.omega.t); V.Cos(.omega.t)) supplied to
said electrically conductive regions with respective inversely
related voltages, (V.sub.ref-V.sub.con) and (V.sub.con), means for
summing the product of said multiplying means and phase shifting
the sum to obtain said D.C.-nulling periodic voltage.
[0033] Preferably said processing means includes integrator means
for receiving the product of said pick-off output voltage V.sub.o
with the D.C.-nulling voltage V'.sub.o, and adjusting said
inversely related voltages, (V.sub.ref-V.sub.con) and (V.sub.con)
to null said D.C. term.
[0034] In yet another aspect, there is provided a capacitive
position transducer comprising:
[0035] a relatively fixed substrate including two spaced
electrically conductive regions varying lineally and inversely in
transverse extent;
[0036] a mover element comprising a common electrical pick-off
means capacitively coupled to both of said electrically conductive
regions and mounted for movement relative to said substrate in a
sensing direction, the transverse extent of at least one of said
electrically conductive regions varying with the position of said
mover element in the sensing direction,
[0037] drive means for supplying respective different periodic time
varying voltages to said electrical conductive regions, and
[0038] processing means operable to multiply an output voltage at
said pick-off means with a DC nulling periodic voltage at the same
frequency as said output voltage and to adjust the relative phases
of said DC nulling periodic time varying voltage to null the DC
term of said product, and thereby to determine the relative
position of said mover element.
[0039] Whilst the invention has been described above, it extends to
any inventive combination of the features set out above or in the
following description.
[0040] The invention may be performed in various ways and, by way
of example only a specific embodiment and various modifications
thereof now will be described in detail, reference being made to
the accompanying drawings in which:
[0041] FIG. 1 is a schematic view of a first non-contacting
capacitive transducer in accordance with the invention;
[0042] FIG. 2 shows an equivalent electrical circuit of the
embodiment of FIG. 1 and its output circuit, and FIG. 3 is a
diagram of an electrical circuit for processing the output received
by the pick-off to provide an output signal which varies
substantially linearly with the position of the mover element.
[0043] Referring initially to FIG. 1, a first embodiment of
position transducer 10 comprises a stator substrate 12 of flat
planar form on which are deposited two triangular sections or
tracks 14, 16 of copper, spaced by an insulating gap 18. A mover
element carrying a pick-off 20 extends across the width of the
substrate and is spaced above the surface thereof by a small air
gap such that the pick-off is capacitively coupled to both the
triangular sections 14 and 16. The two triangular sections of
copper 14, 16 are driven with alternating voltage waveforms, one
(14) with V.Sin(.omega.t) and the other (16) with
V.Cos(.omega.t).
[0044] Referring to FIG. 2, the output from the pick-off 20 is fed
to a high input impedance voltage buffer 22. The values of C1 and
C2 are proportional to the areas of the tracks 14, 16 under the
pick-off 20, so that with the pick-off at either end, one capacitor
will be maximised whilst the other is reduced to zero. Thus at the
Sin end,
[0045] C1=Cmax, C2=0, and V.sub.o=V.Sin(.omega.t)
[0046] and at the Cos end,
[0047] C1=0, C2=Cmax and V.sub.o=V.Cos(.omega.t)
[0048] At some point in between the output voltage, V.sub.o, is
given by:
V.sub.o=V.Sin (.omega.t)/(1+C1/C2)+V.Cos(.omega.t)/(1+C2/C1)
[0049] Now C1=Cmax. (1-.alpha.) and C2=Cmax..alpha., (where .alpha.
is the ratio of the pick-off's distance from the sin end divided by
the overall length of the tracks (14, 16), and substituting for C1
and C2 gives:
V.sub.o=V.Sin(.omega.t).(1-.alpha.)+V.Cos(.omega.t)..alpha.
(Equation 1)
[0050] Now
V.Sin(.omega.t+.theta.)=V.Sin(.omega.t).Cos(.theta.)+V.Cos(.ome-
ga.t).Sin(.theta.)
[0051] If (1-.alpha.).ident.Cos (.theta.) and
.alpha..ident.Sin(.theta.), then:
V.sub.o=K.V.Sin(.omega.+.theta.) (Equation 2)
[0052] where
.theta.=arcrtan (.alpha./(1-.alpha.)) (Equation 3)
K=(1-.alpha.)/Cos(.theta.)=.alpha./Sin(.theta.) (Equation 4)
[0053] and, by vector analysis
K={square root}{square root over (
)}(.alpha..sup.2+(1-.alpha.).sup.2)
[0054] It can be seen from Equations 2, 3, and 4 that the output
voltage, V.sub.o, will be a Sin wave of varying phase and
amplitude. At the mid point, (.alpha.=0.5), V.sub.o will be
1/{square root}2 of the value at either end, and .theta. will be
45.degree.. Between the mid point and either end however the
relationship between .theta. and .alpha. becomes non-linear. Thus,
at a quarter distance from the Sin end, (.alpha.=0.25),
.theta.=arctan (1/3)=18.435.degree., and not 22.5.degree. as would
be the case were the relationship linear.
[0055] This non-linearity may be removed by modifying the profile
of the facing edges of tracks 14 and 16 but we describe below an
alternative approach which multiplies the output voltage, V.sub.o
by a cosine term at the same frequency to obtain a D.C. term.
Thus:
2.Sin(A).Cos(B)=Sin(A+B)+SIN (A-B),
[0056] substituting A=(.omega.t+.theta.) and B=(.omega.t+.phi.),
the following expression results:
2.sin
(.omega.t+.theta.).Cos(.omega.t+.phi.)=Sin(2.omega.t+.theta.+.phi.)+-
Sin(.theta.-.phi.) (Equation 5)
[0057] Sin(2.omega.t+.theta.+.phi.) is an alternating term at twice
the modulation frequency (2.omega.t) and can be removed with a
low-pass filter. The remaining term Sin(.theta.-.phi.) is a D.C.
term which becomes zero when .theta.=.phi..
[0058] Therefore, if we multiply the Sin(.omega.t+.theta.) term
from the output voltage of Equation (2) by Cos(.omega.t+.theta.)
and adjust the result for zero D.C., then .theta.=.phi.. Knowing
this, a value for .phi. can be obtained so that .theta. can be
determined; a (the proportional distance of the slider along the
track) is determined from Equation (3), to give a read out of the
position of the slider.
[0059] Referring to FIG. 3, the terms Cos(.omega.t+.phi.) may be
derived as follows:
[0060] Two multipliers 24, 26 are used to multiply the
V.Sin(.omega.t) and V.Cos(.omega.t) waveforms by
(V.sub.ref-V.sub.con) and (V.sub.con) respectively, where V.sub.ref
is a reference voltage and V.sub.con is a control voltage derived
from the output of an integrator 28 with the range
0.fwdarw.V.sub.ref. The outputs of the multipliers 24, 26, are
added in a summing amplifier 30 and the output is phase-shifted by
a feedback capacitor 32.
[0061] Were it not for the phase shift, the output of the summing
amplifier 20 would be of the form:
V.sub.o'=K'.((V.sub.ref-V.sub.con).V.Sin(.omega.t))+(V.sub.con.V.Cos(.omeg-
a.t))
[0062] Where K' is the gain through the summing amplifier. This
reduces to
V.sub.o'=K".(V.Sin(.omega.t).(1-.beta.)+Cos (.omega.t)..beta.)
(Equation 6)
[0063] where .beta.=V.sub.con/V.sub.ref and K" =K'.V.sub.ref
[0064] The similarity to Equation (1) will be noted.
[0065] This in turn reduces to
V.sub.o=K'".V.Sin(.omega.t+.phi.) (Equation 7)
[0066] where
.phi.=arctan (.beta./(1-.beta.)) (Equation 8)
K'"=K".{square root}{square root over (
)}(.beta..sup.2+(1-.beta.).sup.2)
[0067] By introducing the 90.degree. phase shift in the summing
amplifier Equation (7) becomes:
V.sub.o'=K'".V.Cos(.omega.t+.phi.) (Equation 9)
[0068] which is the required term for Equation (5).
[0069] A third multiplier 33 takes these two terms and multiplies
them to generate the terms to the right of the equality in Equation
(5) and the output of this multiplier consists of an A.C. term at
2..omega. plus a D.C. term. The alternating component is removed by
a low-pass filter 34 and the integrator 28 adjusts V.sub.con until
the D.C. term becomes zero.
[0070] Because the forms of Equations (1) and (6) are the same, the
non-linearity that exists in the relationship between .alpha. and
.theta. is exactly cancelled by the same relationship between
.beta. and .phi. and .beta. is therefore linearly related to
.alpha..
[0071] Furthermore, since it is .beta. and not V.sub.con which is
important, V.sub.con will be scaled by V.sub.ref, thereby giving an
output which is potentiometric in nature. Throughout the circuit
actual signal values are unimportant, so long as signal levels are
sufficient to maintain adequate signal to noise, and it is purely
the relative phases of the signal which matters.
[0072] Furthermore it can be shown that, as opposed to excitation
waveforms of Sin(.omega.t) and Cos(.omega.t), (which is actually
the same as Sin(.omega.t+90.degree.)), Sin(.omega.t) and
Sin(.omega.t+.epsilon.) could be used, where E is some phase angle
other than 90.degree., with no change in performance other than, in
the limit, a degradation of signal to noise. (E tending to zero for
example).
[0073] This means that the circuit will be tolerant of variations
in the phase/frequency of the drive waveforms and variations of
pick-off voltage due to discrepancies in the size of the air gap.
Likewise, so long as the relative amplitudes of the two drive
waveforms remain constant, the actual levels are not important.
[0074] In the embodiment described above with reference to FIGS. 1,
2, and 3, the voltage from the pick-off 20 is passed by a trailing
wire to the voltage buffer amplifier 22. In particular applications
this may not be desirable and so in an alternative arrangement
shown in dotted lines in FIG. 1, a third track or rail 36 is
provided alongside the substrate 12 and the pick-off 20 is extended
at 38 so that it is capacitively coupled to the track 36, thus
allowing the signal to be picked off from a static position. To
prevent the track 36 "seeing" the signal on the adjacent track
section 14, suitable screening is provided including the provision
of an electrical barrier 40 between the track 36 and the track
14.
[0075] As a further modification, instead of being flat planar, the
copper tracks, 14, 16 can be made circular, either in the form of a
cylinder or a disc, of proportions suitable for embodiment in a
traditional potentiometer housing. It should be noted that the
tracks need not be separated by a straight diagonal gap but instead
the gap may be of sinusoidal form, so that the output voltage from
the buffer amplifier 22 is linearly related to the position of the
pick-off. In this instance, the position is determined by observing
the value of V.sub.con relative to V.sub.ref in a similar way in
which the position indicated in a conventional contacting
potentiometer is measured by the wiper voltage relative to the
voltage across the whole of the track.
[0076] It should also be appreciated that several similar
transducers may be stacked together one above the other with a
common actuator, e.g. a common axle or rotor shaft to which an
appropriate number of pick-off rotors are attached.
[0077] In each of these embodiments, the excitation voltage for the
potentiometer would be used to power the electronics, incorporated
inside a housing in the form of a low-power analogue ASIC
(application specific integrated circuit), and the output would be
a D.C. voltage which behaved exactly like a wiper voltage, ranging
from zero to reference voltage, but without the attendant
potentiometer problems outlined at the start.
[0078] Alternatively, a 5-terminal approach could be adopted
whereby two additional terminals provide power to the electronics
whilst the other three terminals simulate a potentiometer, two of
the three being used for the potentiometer reference and the third
being the wiper output voltage.
[0079] Whilst in the above embodiment the drive waveforms have been
sine waves, tests have shown that triangular waveforms give results
comparable to those using sine waves. Also waveforms which are not
pure sine waves, i.e. those containing some harmonic content or
distortion do not within quite wide limits, appear to make any
significant difference to the results obtained, provided they
contain incremental phase information, from which a position
dependent phase difference may be obtained.
[0080] Referring now to FIGS. 4 to 7, a further embodiment of
non-contacting capacitative transducer will now described. In this
embodiment, a pattern of copper tracks making up the wedges 14,16
and the pick-off strip 40 are provided on a thin, flexible,
insulating substrate 42, for example by etching, plating,
deposition, or other suitable accurate photographic method or the
like. The substrate 42 is then folded round to form a cylinder as
shown in FIG. 5. A centre shaft 44 carrying a first wedge pick-off
46 and a second cylindrical pick-off 48 electrically connected
thereto, is mounted coaxially within the cylinder for rotation
about the cylindrical or sensing axis. As seen more clearly in
FIGS. 6 and 7, the wedge pick-off 46 is of axial length equivalent
to the maximum width of each wedge 14,16. The cylindrical pick-off
48 is capacitively coupled to the pick-off strip 40. The substrate
42 is housed within a metal cylindrical housing 50 which includes a
disc-shaped screen 52 which screens the upper and lower sections of
the substrate. The electric connections to the wedges 14, 16 and
the pick-off strip 40 are made through the housing walls, with
suitable layers of screening to prevent coupling between the wedges
and the pick-off strip 40.
[0081] Referring now to FIGS. 8 and 9, in this embodiment the
wedges 14 and 16 are formed on a first substrate disc 54 with the
gap 18 therebetween describing a spiral, and a radial insulating
gap 55.
[0082] The first substrate disc 54 is mounted on one side of a
support disc 56 which includes a metallic screen element 58 in its
mid region. The lower surface of the support disc 56 carries a
second substrate 60 which is a plane pick-off disc. A pick-off
shaft 52 is coaxially mounted with respect to the support disc 56
and carries a radially extending phase pick-off 64 and a coupling
disc 66 electrically connected to the pick-off 64 through the shaft
62.
[0083] Electrical connections are made to the wedges 16, 18 and the
pick-off substrate 60 via the support disc 56. As previously, the
rotary position of the shaft 62 determines the proportions of the
widths of the wedges 16 and 18 capacitively coupled to the phase
pick-off 64. The wedges 16 and 18 are driven by respective
sinusoidal voltages or other suitable periodic time-varying
waveforms, and the resultant voltage picked up by the pick-off 64
processed to determine the angular position of the shaft.
* * * * *