CA2048382C - Interface circuits for electromagnetic position transducers - Google Patents

Abstract

Interface circuits for electromagnetic position transducers are provided to reduce the effects of capacitive coupling occurring in the conductive cabling between the position transducers and the associated interfaces located remotely therefrom. Capacitive coupling of excitation signals to the transducer output signals result in error signals appearing as cyclic error in measured positions. The magnitude of the coupled signals are substantially reduced by providing excitation signals symmetrical with respect to ground. The magnitude of the coupled signals are also reduced by presenting equal impedances relative to ground to the output signal return and signal paths. By reducing the effects of capacitively coupled signals, conducting cables for the excitation signals, and the transducer output signals may be incorporated in a single shielded cable.

Description

-20483~
INTERFACE CIRCUITS FOR ELE~ N~IC POSITION TRANSDUCERS
Ba~ hu, uu--d of the Invention I. Field of the Invention This invention relates generally to apparatus for position meaiuL~ ~. In particular, this invention relates to interface circuits used with ele~ n.ot; ~ position trAn~ rs .
II. Description of the Prior Art Position tr~n~=dl]c~rs which are of interest herein include resolvers and slider and scale systems producing AC
output signals in response to AC excitation signals wherein a phase shift between the excitation signals and the output signals is introduced by the relative position of a trAncdllo.or armature and stator. The position of the armature relative to the stator is measured by detecting this phase difference. Two alternative methods are hnown for detecting the phase difference: a phase discrimination technique wherein the excitation signals are applied to pairs of windings arranged in quadrature, and the position induced phase shift is detected by phase comparison of the output signal with a reference from which the excitation signals are derived; and, an amplitude technique wherein the output signals are produced by the quadrature windings and the position induced phase shift is detected from the ratio of the instantaneous magnitudes of the output signals.
Fig. la illustrates an arrangement used with the amplitude technique employing a resolver to measure position of a moveable member of, for example, machine tools, robots or other position controlled ~ nt. The resolver 10 includes a rotor 12 having an armature coil 14, and a stator having stator coils 16 and 18. The rotor 12 is rotated relative to the stator by, for example, a motor 28. The -204838~

tr~nc~ r~r 10 is located remotely from a control device 20 wherein a drive amplifier 22 produces an AC excitation signal applied to the armature coil 14. Output signal6 appearing at the stator coils 16 and 18 are L~:~ULIled to differential amplifiers 24 and 26 located in control 20. As shown, the return side of the drive amplifier output is grounded and the receiving amplifiers 24 and 26 present unmatched ;Tr~ ncF,c to the signal and return paths because of the input resistor ne~,Lh~ . Conducting cables 30, 32, and 34, typically twisted pairs, provide connection of excitation and output signals between the interface circuits of control 20 and the resolver 10.
Fig. lb illustrates an arrangement used with the phase discrimination technique employing a resolver to measure position of a moveable member. In this arrangement excitation signals are ~L.,duced by drive amplifiers 23 and 25 and applied to the resolver stator coils 17 and 19. An output signal appears at resolver armature coil 13 and is ~e~uLlled to differential amplifier 21 in control 19. The excitation signals are derived from a single reference signal and are phased displaced one from the other by 7r/2 radians .
Fig. 2 illustrates capacitive coupling between an excitation signal cable and an output signal cable which will exist as a result of proximity of the conducting cables 30, 32 and 34 of Figs. la and lb. In Fig. 2 capacitors Cl, C2, C3, and C4 represent lumped values of the coupling capacitances distributed over the lengths of the conducting cables; source SD represents the source of excitation signals; and, load LD represents the load i --~nre presented to an output signal. Inductive coupling of the rotor and stator windings is intentionally omitted to simplify the analysis of the capacitive GOIlrl ;n~ in the 204838;~

cfm~ t i n~ cables . It will be appreciated from Fig. 2 that by virtue of the grounded return paths only capacitance C1 contributes an error .- -nt to the output signal appearing across the load.
The voltage error, ~ in the output signals arising from capacitive collrl ;n~ as shown in Fig. 2 has a magnitude equal to the excitation signal magnitude and is phase displaced ~r/2 radians therefrom. The current due to this error, - nt is added algebraically to the output signal current magnitude, resulting in a position error repeated over the range of position measured by the resolver. Such errors are referred to as "cyclic errors. "
It is common practice to provide individual shields for each conducting cable, such as shields 31, 33, and 35 to reduce or eliminate capacitive coupling between the excitation and output signal cables. The cost of such shielding significantly increases the material and labor costs associated with the installation of such cables.
S ry of the Irlvention It is an object of the present invention to provide apparatus for position measurement by ele~;LL, gnPtic position tr~nc~ rs having inherent elimination and reduction of capacitively coupled signals between excitation and output signal conducting cables.
It is a further object of the present invention to provide apparatus for position measurement by el~:LL, 7nl~tic position tr~nC~Ilc~rs having excitation signals which are ~y ~Lical with respect to ground, thereby providing inherent reduction of a first mode of capacitively coupled signals between excitation and output signal conducting cables.
It is a further object of the present invention to provide apparatus for position mea:,uL~ - L by ~:04838.~

electromagnetic position trAncr~llrprs hàving h~ 1 ~nred signal and return paths for the output signals thereby providing inherent reduction of a second mode of capacitively coupled signals between the excitation and output signal conducting cables.
Other objects and advantage6 of the present invention shall become apparent from the A~ nying drawings and the following description thereof.
In accordance with the aforesaid objects the present invention provides an excitation signal source for use with ele.;~L , ^tic position trAnqfhlrc~rs producing an AC
excitation signal symmetrical with respect to ground. A
first AC signal is inverted to produce a second AC signal and the excitation signal is taken as the difference between the first and second AC signals. Further in auuuLdc,l~ce with the aforesaid objects, a receiver for use with elecLL ^jn,~tic position tr;~nq~ r-rs is provided having matched; ~ nr_q for signal and return lines of an output signal .
Brief Description of the Drawi n~^q Figs. la and lb illustrate prior art circuits for position mea:.uL~ L using a resolver.
Fig 2 illustrates capacitive r~ ; n~ between excitation and output signal conducting cables.
Fig. 3 illustrates circuits of the present invention as applied to a resolver.
Figs. 4a, 4b and 4c are ec~uivalent circuits illustrating the effects of the circuits of Fig. 3 on the capacitive coupling illustrated by Fig. 2.
Fig 5 illustrates circuits of the present invention as applied to a slider and scale system.

Z0483~3~
DetAiled DescriPtisrl of the Preferred r ~ ~ir-~lt To illustrate the invention, ele~LL , ~ic position tr~n~ c~r interface circuits of a motor control device developed for (~in~innlti Milacron Inc., the A~ n~ of the present invention shall be described in detail. While the interface circuits to be described constitute a preferred 1,, it is not the intention of applicants to limit the scope of the invention to the details thereof.
Referring to Fig. 3, a resolver 40, ~-chAnicAlly coupled to motor 50, is shown remotely located from motor controller 60. None of the details of motor controller 60 pertaining to control of motor 50 are pertinent to the present invention and these details shall not be described herein. Motor 50, under control of controller 60, effects rotation of rotor 42 relative to a resolver stator. Stator coils 46 and 48 are fixed relative to the resolver stator and produce AC output signals in response to an AC
excitation signal impres6ed on rotor coil 44. The output signals El and E2 are ~,~pL~ ed as functions of the excitation signal and the relative angular position of the rotor 42 and stator as follows:
Pl = E * sine(e) * sine(Wt) F2 = E * cosine (e) * sine(Wt) Where:
E = magnitude of the excitation signal e = relative angular position of the resolver rotor and stator in radians W = the freguency of the excitation signal in radians t = time 204838;~

Although shown directly applied to the rotor coil 44, the excitation signal may in fact be inductively coupled thereto from the stator in brushless resolvers. In either case, the signal and return paths of the excitation signal are provided by twisted pair conductor cable 52. The signal and return paths for the output signals Fl and F2 are provided in, respectively, twisted pair c~n~ tor cables 54 and 56. By virtue of the symmetrical excitation signal and the symmetrical i - ~nre of the output signal receivers provided by the circuitry of the present invention, capacitive coupling between the excitation and output signals is inherently eliminated and all of the twisted pair conductor cables may be advantageously enclosed by a single shield 58.
Continuing with reference to Fig. 3, it is seen that the excitation signal Vl is taken across the outputs of amplifiers 62 and 64. Amplifier 62 receives a sinusoidal AC
signal Sl of constant frequency W derived from a square wave. The output of amplifier 62 is inverted by amplifier 64. Series inductors 66 and 68 are provided to reduce the possibility of high frequency oscillation appearing at the outputs of amplifiers 62 and 64 when connected to cables presenting relatively high capacitive loads. Gain setting resistors Rl and R2 are of equal value within a moderately close tolerance as may be readily achieved, for example, using 196 - u~.e--Ls. The excitation signal Vl is ay ' ical about ground due to the inversion of the output of amplifier 62 by amplifier 64.
The effect of providing an excitation signal :,y 1 . ical with respect to ground in combination with the prior art receiving circuits is illustrated by the equivalent circuit of Fig. 4a. With the return path grounded at the receiver, the signals coupled by -~04~38Z

capacitances C4 and C3 do not appear in the output signal and these capacitances do not appear in the equivalent circuit. The ~ ; n; n~ capacitances Cl and C2 couple the excitation signal to the output signal producing an error , , ~nt having a magnitude expressed as a function of the excitation signal magnitude as follows:
VCA = E*(Cl - C2)/(Cl + C2) Continuing with reference to Fig. 3, output signals Fl and F2 are received by differential amplifiers 70 and 72 which amplify the potential difference appearing acrQss the signal and return paths of the twisted pair conductor cables 56 and 54. The use of differential amplifiers provides high rejection of noise signals common to the amplifier inputs.
Gain det~m;n;n~ ,_ Ls R4, R7, and R6, R9 are chosen to have equal values within a ,ent tolerance of 0.1% to facilitate analogue to digital conversion with an accuracy of 12 binary digits. Gain ~i~t~rm;n;n~ nts R8 and R3 of amplifier 10 and resistors R10 and R5 of amplifier 72 are also chosen to be equal within a ~ , ~nt tolerance of 0. 1%.
To balance the apparent; ,-' nc~S presented to the signal and return paths at the inputs of motor controller 60, ; ,~ n-~e matching resistors Rll and R12 are connected between the return path input and ground at respectively, amplifier 70 and amplifier 72. The resistor R11 has a value equal to half the product of the sum of the values of resistors R7 and R8 multiplied by the ratio of R7 to R8 and the resistor R12 has a value equal to half the product of the sum of the values of resistors R9 and R10 and the ratio of Rg to R10.
The effect of providing matched impedances to the signal and return lines in combination with an un:~y ' ical excitation signal of the prior art is illustrated by the -204~38:~

esluivalent circuit of Fig. 4b. In this circuit one side of the source is grounded. The contribution to the error ^^t of the output signal from capacitances C2 and C4 of Fig. 2 15 therefor null and these capacitances do not appear in the eyuivalent circuit. The magnitude of the error , L of the output signal contributed by the ~ ; n 1 n.^J capacitances Cl and C3 is expressed as a function o~ the excitation signal magnitude as follows:
Vc~ = E*(Cl - C3)/(Cl + C3) The ' ;nPi effect of providing an excitation signal ~iy LLical with respect to ground and matched i ^-lAn~-Pc Of the signal and return lines at the output signal receivers is illustrated by the ecluivalent circuit of Fig. 4c. The magnitude of the error , L of the output signals contributed by the capacitive ~~ ; n~ is expressed as a function of the excitation signal magnitude E as follows:
Vcc = E*(Cl + C4 - C2 -- C3)/(C1 + C2 + C3 + C4) From the equivalent circuits of Figs. 4a, 4b, and 4c and the analysis of the magnitude of the error ~ , ^^t associated with the resultant capacitive coupling, it is apparent that the error component is eliminated in cases where Cl = C2 = C3 = C4. In cases where these capacitances are not equal, the magnitude of the error ~ , o~t is substantially reduced, permitting the elimination of the individual shields intended to otherwise reduce the effects of capacitive coupling between the excitation signal and output signal conducting cables. It is also apparent that substantial reduction of the error , -- L magnitude is achieved by use of either an excitation signal 2.y LL ical with respect to ground or matched imrP~n~^Pq in the signal and return lines of the output signal.
Fig. 5 shows the use of a symmetrical excitation signal and b~ 1 ~ncPfl ~ , - 'Anre receivers for the output signals as 20483~3~

applied to a slider and scale measuring system 80. A scale excitation signal output by scale amplifier 88 is applied to a scale 82. Scale 82 has a formed conductor /l~ofin;ng pole segments SPI having a pitch I. Slider output signals are induced in slider formed conductors 85 and 86 by the scale excitation signal. The slider formed conductors 85 and 86 define pole segments APSI and APCI having the same pitch I
as the scale pole segments. The slider pole segments are arranged relative to one another 50 as to be spatially separated by I/4. The slider output signals are transmitted to line amplifiers 89 proximate the slider 84 via conducting cables 118 and 120. Line amplifiers 89 produce AC output signals which are transmitted to control 90 via C~ rt;
cables 112 and 114. An excitation signal generated at control 90 is transmitted to scale amplifier 88 proximate the scale 82 via conducting cable 110. Scale amplifier 88 produces the scale excitation signal conducted by cable 116 to scale 82.
Relative position of the slider 84 and scale 82 may be det~rm;n~d from the slider output signals using the amplitude technique described for use with resolvers.
Because of the low impedance of the interface between the slider and scale system and of the amplifiers 88 and 89, capacitive coupling between the scale excitation signal and the slider output signals does not generally give rise to an appreciable error ~ , t in the measured position.
Conversely, the interface between the control 90 and the amplifiers 88 and 89 is susceptible to the same capacitive effects al;cc~lccPd with reference to Fig. 2. Therefor, the use of an excitation signal ~y ical with respect to ground and matched ; , a_ ncpc: in the signal and return lines of the output signals provide the same advantages as previously ,al i ccllccPd thereby permitting the inclusion of 20gi~3382 cables 110, 112 and 114 within a single shield 130. The circuits of control 90 used in this application are the same as shown in Fig. 3.
While the invention has been rl i ~clos~d by reference to the preferred: ' o~ " and the preferred ' ~
have been described in considerable detail, it is not the intention of IrPl; c~nts to limit the invention to such details. Rather, it is intended that the scope of the invention be defined by the ~rpon~lod claims and all equivalents thereto.

Claims (13)

1. In an apparatus for position measurement wherein conducting cables convey AC excitation signals and AC
transducer output signals between an electromagnetic position transducer and associated interface circuits remote therefrom, the excitation siganls beimg generated from an excitation signal generator having a signal output and a grounded return input, the improvement comprising transducer output signal receiving means providing signal and return paths for the transducer output signals which present equal impedances relative to ground to the signal and return paths of the transducer output signals.

2. The apparatus of Claim 1 wherein the transducer output signal receiving means further comprises differential amplifying means having first and second inputs and an output, the first and second inputs presenting equal impedances to ground to signal and return paths of the AC
transducer output signals.

3. In an apparatus for position measurement wherein conducting cables convey AC excitation signals and AC
transducer output signals between an electromagnetic position transducer and associated interface circuits remote therefrom, the interface circuits including transducer output signal receivers presenting unequal impedances relative to ground to the sigan and return paths of the transducer output signals, the improvement comprising excitation signal producing means producing AC
excitation signals symmetrical with respect to ground.

4. The apparatus of Claim 3 wherein the excitation signal producing means further comprises:
a. a first amplifying means for producing a first AC
signal; and b. a second amplifying means for inverting the first AC
signal to produce a second AC signal anti-phase relative the first AC signal, the excitation signal being taken as the potential difference between the first and second AC signals.

5. In an apparatus for position measurement wherein conducting cables convey AC excitation signals and AC
transducer output signals between an electromagnetic position transducer and associated interface circuits remote therefrom, the improvement comprising:
a. excitation signal producing means producing AC
excitation signals symmetrical with respect to ground; and b. transducer output signal receiving means providing signal and return paths for the output signals which present equal impedances to ground to the signal and return paths of the transducer ouptut signals.

6. The apparatus of Claim 5 wherein the excitation signal producing means further comprises:
a. a first amplifying means for producing a first AC
signal; and b. a second amplifying means for inverting the first AC
signal to produce a second AC signal anti-phase relative the first AC signal, the excitation signal being taken as the potential difference between the first and second AC signals.

7. The apparatus of Claim 5 wherein the output signal receiving means comprises differential amplifying means having first and second inputs and an output, the first and second inputs presenting equal impedances to ground to signal and return paths of the AC transducer output signals.

8. An apparatus for position measurement comprising:
a. an electromagnetic transducer having an armature and a stator for resolving an AC excitation signal into first and second AC output signals having a predetermined phase difference relative to one another and having a common phase difference from the excitation signal determined by the relative position of the armature and stator; and b. an output signal receiving means including differential amplifying means having first and second inputs and an output, the first and second inputs presenting equal impedances to signal and return paths of the first and second AC output signals.

9. The apparatus of Claim 1 further comprising an AC
excitation signal producing means providing an AC
excitation signal being symmetrical as measured between signal and return paths for the excitation signal.

10. The apparatus of claim 2 wherein the AC excitation signal producing means further comprises:
a. a first amplifying means for producing a first AC
signal; and b. a second amplifying means for inverting the first AC
signal to produce a second AC signal anti-phase relative the first AC signal, the excitation signal being taken as the potential difference between the first and second AC signals.

11. An apparatus for position measurement comprising:
a. an electromagnetic transducer having an armature and a stator for resolving an AC excitation signal into first and second AC output signals having a predetermined phase difference relative to one another and having a common phase difference from the excitation signal determined by the relative position of the armature and stator;
b. an AC excitation signal producing means providing an AC excitation signal being symmetrical as measured between signal and return paths for the excitation signal.

12. The apparatus of Claim 4 further comprising an output signal receiving means including differential amplifying means having first and second inputs and an output, the first and second inputs presenting equal impedances to signal and return paths of the first and second AC output signals.

13. The apparatus of Claim 4 wherein the AC excitation signal producing means further comprises:
a. a first amplifying means for producing a first AC
signal; and b. a second amplifying means for inverting the first AC
signal to produce a second AC signal anti-phase relative the first AC signal, the excitation signal being taken as the potential difference between the first and second AC signals.