DE102011087494B4 - Capacitive sensor for position or motion detection - Google Patents

Abstract

Capacitive rotary encoder with a high-frequency generator (1) for generating a transmission signal (S), a first stator (2) with a transmission electrode (21), a rotor (3) as an angle encoder with a rotor electrode (31, 32), a second stator (4) with pairs of receiving electrodes (41), the distribution of the received displacement currents being influenced by the angular position of the rotor (3), with switching mixers (5) controlled by the transmission signal (S), integrating current-voltage converters (6), a digital evaluation unit (7) and / or analog evaluation units (8) and an actuator (9) for influencing the amplitude and / or the frequency of the high-frequency generator (1), characterized in that pairs of receiving electrodes (41) alternate with integrating current-voltage converters (6) via switching mixers (5) are connected and whose outputs are linked to the digital evaluation unit (7) and / or to analog evaluation units (8).

Description

The invention relates to a capacitive sensor for position or motion detection according to the features of the preamble of claim 1. Furthermore, a method for operating a capacitive sensor according to the invention is claimed.

Capacitive sensors for position or motion detection, in particular rotary encoders or rotary encoders, are mainly used in automation technology but also in vehicle construction as rotary angle sensors in order to detect angular positions during rotary movements. Examples are the steering angle detection in motor vehicles and the position detection of rotating components of machines. A distinction is made between incremental rotary encoders and absolute measuring sensors.
Both types of sensors are manufactured and sold in various designs by the applicant, among others.
Capacitive sensors are valued primarily because of their simple design, the associated low manufacturing costs, their low energy requirements and their insensitivity to magnetic fields. Disadvantages are their drift behavior, in particular the temperature drift and their susceptibility to electromagnetic radiation because of the electrodes acting as receiving antennas.
Such sensor arrangements usually have a plurality of mutually offset transmitting electrodes, usually arranged on concentric rings, and a common, often shielded, receiving electrode. A movable element referred to as a flag, in particular a rotor, with an electrically conductive structure influences the amplitude of the serially transmitted signals of the transmission electrodes as a function of the angle of rotation.
The electrically conductive structure on the flag neither has to be grounded nor electrically connected to the sensor electronics. It only shortens the effective plate spacing.

From the US 6 876 209 B2 and the US 2010/0 222 967 A1 absolute rotary encoders are known which are particularly suitable for automotive applications for detecting the angular position of a rotatable body, such as a steering column of a motor vehicle. The US 6 876 209 B2 has a variable capacitor, the transmitter and
The receiver electrode area is dimensioned in such a way that an angular movement of the rotor triggers a change in capacitance. The sensor includes means for measuring the induced charge on the receiver electrodes, whereby the charges indicate the angular positions of the rotatable body.
Furthermore, from the DE 10 2007 058 707 A1 a capacitance sensor known.
The EP 1 251 338 A1 shows a capacitive sensor and a method for measuring four transmitter electrodes arranged concentrically in two rings, each semicircular around the axis of rotation. The semicircular transmission electrodes overlap by 90 °. Between this immovable arrangement and a likewise immovable receiving electrode there is a flag (rotor) which can be rotated by at least 360 ° and which changes the capacity distribution between the four transmitting electrodes and the receiving electrode depending on the angle of rotation. The four transmitter electrodes are connected one after the other to a square wave generator by a control unit. A measuring amplifier and then an analog-digital converter are connected to the receiving electrodes. The serial processing of the measurement is viewed as unfavorable, as it is extremely disadvantageous with a higher number of electrodes (magnifying glass function). In addition to reducing the switching frequency, all fluctuations in the transmission signal, but also the drift of the receiving electronics, are included in the measurement result.
The WO 00/63 653 A2 ( DE 600 16 395 T2 ) shows various capacitive rotation angle sensors that are used to detect the position of a rotating shaft.
They comprise either a multi-part transmitter and a single-part receiver or a single-part transmitter and a sectored receiver. A known coarse and fine pattern of the electrodes is intended to increase the angular resolution.
In a first version in 9 there are at least 4 transmitter electrodes, which are excited by a four-phase alternating current source with a 10kHz square wave signal. A charge amplifier designed as a current-voltage converter amplifies the displacement currents dependent on the angle of rotation and forms the weighted sum signal of the four excitation voltages. An active rectifier controlled by the alternating current source generates two signals offset from one another by 90 °, which are referred to as the sine and cosine functions. However, their specific course depends on the shape of the electrodes. In a second version in 10 In this document, the transmission signal of a one-piece transmission electrode is recorded by four circular-ring sector-shaped reception electrodes (receiver quadrant plates) and fed to current-voltage converters. Differential amplifiers connected in pairs to their outputs generate the above-mentioned sine and cosine signals, which are offset from one another by 90 °, depending on the angle of rotation. The problem is the carrier frequency of only 10 kHz because it is relatively close to the signal frequencies of the speed measurement. In the case of the finely structured rotors, in particular, this requires high-order low-pass filters to smooth the received signal.
At higher carrier frequencies, current-voltage converters would have a correspondingly high bandwidth needed. In addition, there is a risk of capacitive cross-talk between the four receiving channels from which the sine and cosine signals are obtained.
In serial processing after 9 follow up the amplitude fluctuations of the transmitted signal and during the parallel processing 10 the tolerances of the current-voltage transformers in the measurement result.

The control of the transmission signal required for high-precision measurement is implemented in the DE 10 2007 038 225 A1 shown. The reference signal required for this is obtained with the aid of an additional reference electrode. Of course, this additional electrode requires a sensor surface and therefore space on the angle encoder disks, which is not even available in some applications. One way out is the certainly disadvantageous reduction in size of the receiving surface effective for the actual angle measurement.

The object of the invention is to specify a capacitive sensor that is particularly suitable as a rotary encoder, which is immune to interference and has high measurement accuracy with small external dimensions.

This object is achieved according to the features of claim 1. The subclaims relate to the advantageous embodiment of the invention. Furthermore, a method for operating a capacitive sensor according to the invention is specified.

The main idea of the invention is to carry out the phase-synchronous rectification and the difference formation directly on the receiving electrodes. For this purpose, the receiving electrodes are connected in pairs with active rectifiers or switching mixers. The difference signal arises at their anti-parallel connected outputs.
The analog multiplexers used for this cannot falsify the difference signal, because practically no voltage drops across their track resistance of 2.5 ohms.
The current measurement known from the above-mentioned prior art by virtual grounding of the receiving electrodes is therefore retained. However, the current-voltage conversion takes place advantageously only after the rectification or mixing and the difference formation. In this way, the downstream current-voltage converters only have to process direct current or the comparatively low-frequency signals resulting from the rotary movement. Their tolerances, especially those of the resistors, are less included in the measurement result. The signal-to-noise ratio can be improved by increasing the operating frequency without increasing the demands on the bandwidth of the receiving channels.
A reduction of the effective electrode areas is also possible in this way. In this way, the hollow shaft diameter can be increased or the size can be reduced. Because the voltage at the inputs of the current-voltage converters is zero, there is practically no crosstalk between the lines or electrodes of the receiving channels.

A second idea of the invention consists in regulating the amplitude of the received signal by changing the transmission frequency. The transmitter can work with constant amplitude, for example with digital logic signals such as transistor-transistor logic (TTL).

A third idea of the invention consists in obtaining the reference signal required for amplitude control not in the known manner with a reference electrode, but computationally by summing received signals, in particular from the sine signal and the cosine signal. This saves valuable sensor space.

• 1 zeigt die Prinzipschaltung eines erfindungsgemäßen kapazitiven Sensors.
• 2 zeigt ein etwas detailliertes Ausführungsbeispiel für einen kapazitiven Drehgeber
• 3 zeigt eine Schaltung zur Gewinnung des Differenzsignals der Empfangselektroden.
• 4 und 5 zeigen die Erzeugung eines Referenzsignals aus den Sensorsignalen

The invention is explained in more detail below with reference to the drawing

• 1 shows the basic circuit of a capacitive sensor according to the invention.
• 2 shows a somewhat detailed exemplary embodiment for a capacitive rotary encoder
• 3 shows a circuit for obtaining the difference signal of the receiving electrodes.
• 4th and 5 show the generation of a reference signal from the sensor signals

The 1 shows a capacitive sensor for position or motion detection. A high frequency generator 1 (VCO) generates a transmission signal S. that is used simultaneously as a clock signal T serves. The transmission signal delayed by two inverters S. becomes a differential capacitor 10 fed from a first stator 2 , a mechanical measuring element 3 and a second stator 4th consists.
The differential capacitor 10 has a transmitting electrode 21st and two receiving electrodes 41 on. The capacitance ratio of the differential capacitor 10 is from the measuring element 3 (Flag) influenced by local shortening of the effective plate spacing (air gap).

The pair of receiving electrodes 41 is with one of the high frequency generator 1 (VCO) controlled switching mixer 5 connected. A difference signal arises here FROM , which is the difference in the displacement currents in the receiving electrodes 41 represents.
With the measuring element at rest 3 is a direct current signal, which in the middle position with two identical partial capacitances of the differential capacitor 10 becomes zero.
That on the mixer tap 5 The resulting current signal is first smoothed in order to free it from residues of the high-frequency transmission signal and then converted into a voltage. This is done with the help of the integrating current-voltage converter 6 from a low pass 6a and the actual current-voltage converter 6b consists.
This voltage is a measure of that of the measuring element 3 dependent capacitance ratio in the differential capacitor 10 . After digitization and further processing in the microcontroller 7th it can be output as a position or switching signal.
The microcontroller 7th is via an actuator 9 with the high frequency generator 1 connected. The oscillator frequency of the VCO can be determined via a pulse-width modulated signal (PWM) from the microcontroller 7th to be controlled. Using a resistor network, which is also connected to the microcontroller, frequency spreading is also possible, which serves to improve the immunity to interference and to reduce the radiated interference.

The 2 shows a capacitive rotary encoder according to the invention in a simplified representation. The high frequency generator 1 , a voltage controlled oscillator (VCO), generates a square wave signal of 2 - 4 MHz. As operating voltage 5 Volt is selected, a TTL signal is generated. This TTL signal controls the switching mixer 5 and feeds one to all receiving electrodes 41 effective transmitting electrode 21st on the first stator 2 .
To the switching times of the switching mixer 5 to compensate and, moreover, to give them a time advantage over the switching edges of the transmission signal, the transmission signal S. delayed by two inverters of the type 74HC14 by approx. 2 x 12 nanoseconds. This is how switching mixers switch 5 when de-energized.
As a switch mixer 5 monolithic analog multiplexers of the type ADG749 from Analog Devices were used here. They have a typical forward resistance Ron = 2 ohms. Switching in the de-energized state avoids the temperature drift of the analog multiplexer and thus reduces the influence of the switching processes on the measurement result. Dispensing with preamplifiers, ie low-resistance switching without prior amplification, avoids the influence of resistance tolerances when forming the difference. On the second stator 4th are receiving electrodes A. to H Arranged in pairs so that they are separated from the rotor electrodes 31 and 32 Form influenceable differential capacitors. The receiving electrode pairs 41 are about switching mixer 5 with the integrating current-voltage converters 6 connected. The integrating current-voltage converters 6 are only shown symbolically. They have a low-pass filter and a correspondingly wired operational amplifier. The input of the operational amplifier is virtually at ground potential so that a current measurement can take place.
The outputs of the integrating current-voltage converters 6 are with the microcontroller 7th connected. Here the signals are digitized and processed further.

The rotor 3 preferably consists of a 1.5 mm thick FR4 plate. The material FR4 itself is not conductive, but has circular ring sectors that are conductive on both sides and are preferably of the same size and are connected to one another via vias. They serve as rotor electrodes 31 and 32 because it determines the effective plate distance (air gap) between the transmitter electrode 21st and the receiving electrode pairs 41 shorten. The total capacitance of the individual differential capacitors remains constant. The rotary movement only changes the ratio of the two partial capacitances of the differential capacitors. It should be noted here that the illustration with four rotor shafts was chosen for better understanding. A common rotor shaft is preferably present.

As can be seen, the electrode pairs AB and CD are on the stator 4th arranged offset by 90 ° to each other. Two signals offset by 90 ° are obtained, which are referred to as the sine and cosine of the angle of rotation, although their shape depends on the shape of the electrode. Based on the values FROM and CD one can clearly determine where the semicircular inner rotor electrode is 31 on which the rotor 3 is located. However, the accuracy is not yet sufficient. Therefore, these electrode pairs are only used to divide into 8 circular sectors. The outer circular ring sectors E. , F. , G , H on the stator 4th serve for more precise position determination (magnifying glass). They are from the rotor electrodes 32 influenced in the manner described. This arrangement is clear within 45 °. This two-stage angle determination using the coarse and magnifying glass signals is well known and is not the subject of the invention.

That on the receiving electrodes 41 pending current signal is from the mixer switches 5 phase-synchronously rectified and smoothed by the low-pass filter. The multiplexer outputs connected in anti-parallel ensure that the difference is formed. The current signal is then sent to the integrating current-voltage converter 6 amplified and converted into tension.

The evaluation can be done completely in a digital evaluation unit 7th , preferably in a microcontroller (µC). A preprocessing or intermediate processing in one or more analog Evaluation units 8th but can be an advantage. As shown, besides the current angular position of the rotor 3 A zero pulse can also be generated for each revolution. This is done by NAND linking signals D1 to D7. This link can take place in the microcontroller7, but also advantageously in a NAND gate with 8 inputs of the 74HC30 type. This relieves the load on the microcontroller and the arrangement is suitable as a tachometer for high speeds.
In addition, as in the 4th shown that the additions, subtractions and comparison operations required to evaluate the sensor signals, in particular the arithmetic operations required to determine a reference signal from the magnifying glass signals, are efficiently carried out with operational amplifiers or comparators.

The electronic circuit can be made largely temperature-compensated. That leaves those of the mechanics, like tumbling of the rotor 3 , and to consider amplitude fluctuations caused by the media, such as water vapor and condensation (condensation).
A direct amplitude control is not possible or not sufficient because the transmission signal is fixed at TTL level.

Since the differential capacitors represent frequency-dependent resistances, the amplitude of the received signal can advantageously be based on the frequency of the high-frequency generator designed as a voltage-controlled oscillator (VCO) 1 be managed.

In order to save the known reference electrode and reference receiver and thus gain sensor area, according to the invention, the signals from the outer circular ring sectors are used E. to H (Magnifying glass) a reference signal is generated and used by the microcontroller 7th output as a pulse-width modulated signal (PWM signal). More on this in the 4th and 5 . So that this signal can be used as an actual controlled variable, it is first integrated, then in one in the actuator 9 compared with a setpoint and via a PID controller as a control voltage for the high-frequency generator 1 (VCO) used.
Too high a controlled variable (amplitude) is corrected by lowering the working frequency and too low a controlled variable by increasing the working frequency.
With PN x marked modulation inputs on the VCO can be used for frequency spreading. This improves the device's active and passive immunity to interference.

Because the mixer tap 5 directly with the pairs of receiving electrodes 41 are connected, rectification or mixing with the transmit signal and the difference formation take place simultaneously in one step. The integrating voltage converters 6 no longer have to process the working frequency of the VCO, but only the pulses resulting from the rotary movement. In this way, the working frequency can be increased compared to the known technical solutions without tightening the requirements for the gain bandwidth product of the operational amplifier.
Working frequencies of 10 MHz allow measuring capacities in the picofarad range without the need for high-frequency current-voltage converters 6 required are.
This saves costs and allows smaller designs and / or higher measurement accuracy.

By generating a reference signal from the measurement signals and thereby dispensing with the known reference electrode, the rotor surface is also free.
In this way the rotor can be made smaller. This results in structural advantages, for example the possibility of enlarging the hollow shaft. The reference recipient can be completely dispensed with. This again saves costs and PCB space.
The reference obtained from the useful signals can of course also be used for computational correction of the measured values in the microcontroller 7th be used.

The 3 shows a circuit for obtaining the difference signals FROM and CD from the four inner electrodes for rough position determination. The evaluation of the external electrodes E. to H (Sensor magnifier) can be done in the same way. As already mentioned, these signals are referred to in the literature as sine or cosine signals, although their shape depends on the geometry of the electrodes. These are triangle signals. The high frequency generator 1 (VCO) feeds the transmission electrode of the differential capacitor 10 with slightly compared to the clock signal T delayed transmission pulses S. .
The two receiving electrodes A and B are each connected to an analog multiplexer. The ADG749 from Analog Devices shown here can process signals up to around 10 MHz with a typical turn-on time of 20 ns and a turn-off time of only 6 ns. Diode ring mixers are required for even higher working frequencies.
The ADG749 analog multiplexers work as switch mixers 5 . They are wired in such a way that electrodes A and B alternately with half the operating voltage or with the inverting input of the same current-voltage converter 6 are connected. While electrode A is subjected to half the operating voltage, electrode B is connected to a 10 nF capacitor and vice versa. So the displacement currents are the Electrodes collected in the two 10 nF smoothing capacitors.
Half the operating voltage defines the zero potential. Due to the in-phase control of the analog multiplexer through the clock T becomes the current-voltage converters 6 A direct current signal is supplied, which is converted into the output voltage U (via a negative feedback resistor of 100 kΩ FROM ) is converted. The current-voltage converter 6 is wired in such a way that it has low-pass properties. That is why it can also be called an integrating current-voltage converter. So the signal becomes the switching mixer 6 additionally smoothed. The negative feedback capacitance of 1 nF is used for stabilization.

At this point, the inventive idea of performing the phase-synchronous rectification without temperature-dependent diode rectifiers and the difference formation directly on the receiving electrodes without resistors subject to tolerances becomes clear.
The smoothing capacitors are constantly discharged against virtual ground potential.

• Ein erster Komparator vergleicht das Signal E-F mit der halben Betriebsspannung. Liegt es darüber, schaltet der Ausgang des ersten Analogmultiplexer ADG749 auf halber Betriebsspannung und der angeschlossene erste Operationsverstärker wird zum Inverter (invertierender Verstärker). Es entsteht ein negatives Ausgangssignal, das nach einer weiteren Invertierung einen positiven Beitrag zum Referenzsignal liefert.
• Liegt das Signal E-F unterhalb der halben Betriebsspannung, schaltet der Ausgang des ersten Komparators auf Betriebsspannung und der Analogmultiplexer ADC749 schaltet um. Nun ist der positive Eingang des Operationsverstärkers frei und er wird zum Folger (nichtinvertierenden Verstärker). Die Beschaltung wurde hier so gewählt, dass der Verstärkungsfaktor in beiden Fällen 1 beträgt. Das ist allerdings nicht erfindungswesentlich. So liefert auch die andere Halbwelle einen positiven Beitrag zum Referenzsignal. Auf diese Weise wird das durch die Differenzbildung E-F entstandene Dreieckssignal „hochgeklappt“, so dass ein mit der halben Betriebsspannung vorgespanntes Dreieckssignal der doppelten Frequenz entsteht. Die verbleibenden Lücken werden durch das um 90° phasenverschobene auf die gleiche Weise verarbeitete „Sinussinussignal“ G-H ausgefüllt, so dass eine von der Drehbewegung unabhängige, die Empfangsamplitude repräsentierende Gleichspannung entsteht, die erfindungsgemäß vorteilhaft als Referenzsignal verwendet werden kann. Weiterhin wird die zur groben Positionsbestimmung mit 8 Sektoren erforderliche logische Verknüpfung der beiden Differenzsignale A-B und C-D, der Grobsignale A-B und C-D gezeigt, die natürlich auch im Mikrocontroller 7 erfolgen kann. So wird er aber entlastet.

The 4th shows further circuit details. The reference signal is obtained on the left. For this purpose, the amounts of the signals running symmetrically to half the operating voltage (Ub / 2) are used EF and GH generated by the sensor magnifier ("cosine" and "sine") and added to half the operating voltage. This is done with the help of an optional inverter in the following way:

• A first comparator compares the signal EF with half the operating voltage. If it is higher, the output of the first analog multiplexer ADG749 switches to half the operating voltage and the connected first operational amplifier becomes an inverter (inverting amplifier). A negative output signal is created which, after a further inversion, makes a positive contribution to the reference signal.
• Is the signal EF below half the operating voltage, the output of the first comparator switches to operating voltage and the ADC749 analog multiplexer switches over. Now the positive input of the operational amplifier is free and it becomes the follower (non-inverting amplifier). The wiring was chosen so that the gain factor in both cases 1 amounts. However, this is not essential to the invention. The other half-wave thus also makes a positive contribution to the reference signal. This is how it is done by subtracting EF The resulting triangular signal is "folded up", so that a triangular signal of double the frequency is created with half the operating voltage. The remaining gaps are processed in the same way by the "sinusoidal signal", which is phase-shifted by 90 ° GH filled in, so that a direct voltage which is independent of the rotary movement and represents the reception amplitude is produced, which can advantageously be used as a reference signal according to the invention. Furthermore, the logical combination of the two difference signals required for rough position determination with 8 sectors FROM and CD , the coarse signals FROM and CD shown, of course, in the microcontroller 7th can be done. But that way he is relieved.

The temporal resolution of the arrangement shown is sufficient for medium speeds. A logical combination of the digital signals D1 to D7 from all 7 comparators in a NAND gate delivers a narrow zero pulse per revolution. This can also be done through software.
At very high speeds, an exact angle determination is no longer necessary, but one zero pulse per revolution is sufficient. By using the NAND gate, the demands on the operating speed of the microcontroller can be reduced.

The 5 illustrates the generation of the reference signal. The two differential signals labeled “sin” and “cos” offset by 90 ° GH and EF (Magnifying glass signals) are used by the 4th The optional inverters shown are mirrored on the central axis (2.5 V), ie "folded up" and then added up. As you can see, the areas marked “a + b” in both channels and thus duplicated fill the gaps between the areas “a” and “b” so that a direct voltage, the reference voltage Uref, is independent of the current rotor position. It can also be seen that a triangular signal shape is particularly favorable here. So it results for angle w between 0 ° and 90 ° the simple relationship: w = 90 ° * ("sin" (w) -Ub / 2) / (Uref-Ub / 2).
However, other signal forms, especially sinusoidal signals, should not be excluded.

The present invention relates to a capacitive sensor for position and / or movement detection, in particular a rotary encoder with a differential capacitor 10 . He points two through a measuring element 3 partial capacities that can be influenced in opposite directions. These consist of a high frequency generator 1 fed common transmitter electrode 21st and at least two receiving electrodes 41 . The receiving electrodes 41 are each directly connected to a switching mixer 5 connected. The mixer tap 5 is an integrating current-voltage converter 6 downstream. It also serves as a low-pass filter to smooth the switching mixer 5 resulting difference signal. The current-voltage converter 6 converts the displacement current of the receiving electrode 41 in a Voltage signal around. That from the integrating current-voltage converter 6 The output signal represents the difference in the partial capacitance of the differential capacitor 10 and is therefore a measure of the position of the measuring element 3 .

A preferred embodiment of the invention relates to a capacitive rotary encoder with a high-frequency generator 1 for generating a transmission signal S. whose frequency and amplitude are essential for the measurement result. The broadcast signal S. is fed to a differential capacitor on its first stator 2 the transmitter electrode 21st is located. On a second stator 4th are pairs of receiving electrodes 41 arranged such that a rotor electrode 31 on a rotor 3 , which serves as an angle sensor, the distribution of the displacement currents of the receiving electrode pairs 41 depending on the angular position of the rotor 3 influenced in opposite directions.
The receiving electrode pairs 41 are in sync with the broadcast signal S. acting switch mixer 5 connected. An integrating current-voltage converter 6 converts the received current into a voltage. This becomes a digital evaluation unit 7th and advantageously also analog evaluation units 8th fed. An actuator 9 serves to influence the amplitude or the frequency of the high frequency generator 1 in a control loop for controlling the reception amplitude.
The receiving electrode pairs 41 are via the switching mixer 5 alternating with the integrating current-voltage converters 6 connected. Here, with the rectification, there is also a subtraction of the two from the relevant electrode pair 41 received displacement streams. The outputs of the integrating current-voltage converters 6 are with the digital evaluation unit 7th and / or with analog evaluation units 8th connected. The analog evaluation units 8th enable simple arithmetic operations to be carried out quickly to evaluate the sensor signals. The evaluation initially includes a rough position determination based on the signals from electrodes A to D, which already provides a clear angular position. The electrodes E. to H serve for fine evaluation (magnifying glass) for more precise position determination.

In an advantageous embodiment of the capacitive rotary encoder according to the invention, the angular position of the rotor 3 influenced dielectric displacement currents in the receiving electrode pairs 41 instead of a reference electrode as the main component of the manipulated variable for the actuator 9 used to control the reception amplitude.
In a special embodiment, the reception amplitude is regulated by changing the frequency of the transmission signal S. . This is possible because the measuring capacitances have frequency-dependent resistances. The main advantage is that the transmission amplitude can be kept constant. The transmitting electrode is advantageously controlled with a logic level, for example a signal compatible with the transistor-transistor logic (TTL). So can the mixer tap 5 (controlled rectifier) parallel to the transmitter electrode 21st with the broadcast signal S. be controlled.
Furthermore, it is advantageous to reduce the transmission signals by a defined period of time compared to the switchover signal for the switching mixer 5 to delay. The switching times of two logic gates connected one behind the other in the transmission line can be used for this purpose ( 2 ).
In a further embodiment of the invention, a zero signal can be generated for each revolution. This can be done by NAND linking signals D1 to D7 in the microcontroller 7th happen. It is advantageous to use a 74HC30 type NAND gate with 8 inputs, the output of which emits a signal with every full revolution.

The method according to the invention includes the operation of the described arrangement. A capacitive sensor that has a differential capacitor with a transmitting electrode 21st and at least one pair of receiving electrodes 41 has, is applied with alternating voltage in the range from 1 to 100 MHz.
The capacitance of the differential capacitor is taken from the measuring element 3 influenced in opposite directions. The measuring element is preferably shortened 3 the effective plate spacing.
The receiving electrode pairs are used to measure the dielectric displacement currents 41 with a switching mixer 5 connected. A switching mixer operating in phase synchronism with the transmission signal is advantageously used here. Its output current is smoothed and measured with low resistance. To do this, it becomes a current-voltage converter at virtual ground potential 6 fed. Because the input of the integrating current-voltage converter 6 is connected to ground in terms of alternating current, the receiving electrodes are also virtually connected to ground. This is how a real current measurement takes place. Because the input voltage is zero because of the virtual grounding, no capacitive crosstalk can take place.
Because the mixer tap 5 directly with the receiving electrode pairs 41 is connected, the rectification / mixing with the transmission signal and the difference formation takes place in a common step. The integrating voltage converter 6 low-frequency current pulses are supplied. It does not have to be suitable for high frequencies.
In an advantageous embodiment of the method according to the invention, the amplitude of the received signal is regulated by changing the transmission frequency. The advantage is that the transmit amplitude can remain constant at the TTL level.
In a further embodiment of the method according to the invention, a reference signal for amplitude control is obtained by summing the received signals or the inverted received signals. In particular, the sine and cosine signals of the receiving electrodes which are larger in area are used for this purpose, offset from one another by 90 ° E. to H evaluated with the magnifying glass. That with an analog evaluation unit 8th and / or in the microcontroller 7th The reference signal generated can be used for amplitude regulation or for evaluating the received signal, or can be transferred to a higher-level control unit.

List of reference symbols

1

High frequency generator, voltage controlled oscillator (VCO)

2

First stator

3

Rotor (flag), measuring element

4th

Second stator

5

Switching mixer (controlled rectifier)

6th

Integrating current-voltage converter

6a

Integrator

6b

Current-voltage converter

7th

Digital evaluation unit (microcontroller)

8th

Analog evaluation unit (comparator, differential amplifier)

9

Actuator

10

Differential capacitor

21st

Transmitting electrode

31

Inner rotor electrode (coarse)

32

Outer rotor electrode (fine)

41

Pair of receiving electrodes

A to H

Receiving electrodes

PN x

Modulation input or output

R.

Control signal

S.

Transmission signal

T

Clock signal

TP

Low pass

w

angle

Claims (8)

Capacitive rotary encoder with a high-frequency generator (1) for generating a transmission signal (S), a first stator (2) with a transmission electrode (21), a rotor (3) as an angle encoder with a rotor electrode (31, 32), a second stator (4) with pairs of receiving electrodes (41), the distribution of the received displacement currents being influenced by the angular position of the rotor (3), with switching mixers (5) controlled by the transmission signal (S), integrating current-voltage converters (6), a digital evaluation unit (7) and / or analog evaluation units (8) and an actuator (9) for influencing the amplitude and / or the frequency of the high-frequency generator (1), characterized in that pairs of receiving electrodes (41) alternate with integrating current-voltage converters (6) via switching mixers (5) are connected and whose outputs are linked to the digital evaluation unit (7) and / or to analog evaluation units (8). Capacitive encoder after Claim 1 , characterized in that the dielectric displacement currents in the receiving electrode pairs (41) influenced by the angular position of the rotor (3) are part of the manipulated variable of the actuator (9). Capacitive encoder after Claim 2 , characterized in that the reception amplitude is regulated by changing the frequency of the transmission signal (S). Capacitive encoder after Claim 1 or 2 , characterized in that the transmission signal (S) is delayed by a defined period of time compared to the switching signal for the switching mixer (5). Capacitive rotary encoder according to one of the preceding claims, characterized in that a zero signal is generated after each revolution. Method for operating a capacitive sensor according to one of the preceding claims, characterized by the following method steps: charging of a differential capacitor (10) which has a transmitting electrode (21) and a pair of receiving electrodes (41) and which acts as a capacitive transducer with alternating voltage of 1 to 100 MHz. - Rectification of the dielectric displacement currents arising in the receiving electrode pairs (41) with a switching mixer (6) connected directly to the receiving electrodes (41) with simultaneous formation of a differential signal. Smoothing of the output current. - Current measurement with a current-voltage converter (6), the input of which is virtually at ground potential. - Amplification and evaluation of the signal generated by the current-voltage converter (6), the amplitude of the received signal being regulated by changing the frequency of the transmission signal (S), the transmission amplitude remaining constant and the transmitter being able to work with constant amplitude with digital logic signals. Procedure according to Claim 6 , characterized in that the actual controlled variable for the amplitude control is obtained by summation from received signals. Procedure according to Claim 7 , characterized in that a reference signal is obtained by summing received signals and inverted received signals.

Images