DE102012008699B4 - Method for increasing the measuring range of a device for non-contact measurement of a distance - Google Patents

Abstract

The invention relates to an inductive position measuring system for the proportional and contactless detection of linear and rotary motion processes as well as their forms of superposition. The types of superimposition require a system that is integrated in the volume model of the distance to be measured and that is implemented without mechanical coupling to the measurement task. A ferritic pot core with a coil, open on one side, serves as the measuring element, with which the distance to any metallic target can be detected. First of all, embodiments are described with which the pot core coil L located in an electrical oscillating system (30) is operated with an ideally high quality and, as a consequence, maximum range. In addition, the introduction of a regulated energy return (50) into the resonant circuit is proposed, which allows a further and adjustable increase in the resonant circuit quality. The measuring range of such a sensor, which works on the inductive principle, can thus be increased again considerably.

Description

The invention relates to an inductive measuring system for the proportional and non-contact detection of linear and rotary motion processes as well as their overlay forms. In particular, such sensors are used in motor vehicle technology as well as in industrial automation technology. Here, there is a need for both quasi-static and dynamic measurement of distances, paths and angles. Frequently occur due to the mechanical design in the application overlay types of these forms of motion, so that in the distance-volume model of the measuring section integrating system is necessary, which should be carried out in consequence and beyond in a preferred manner without mechanical coupling to the measurement task.

In the mentioned fields of application, but especially in automotive engineering, very flexible sensor concepts are needed that can be adapted as easily as possible to different housing designs and application situations and, moreover, are small and lightweight. Furthermore, a coordinated development of transducers for the most varied forms of motion for the dedicated detection of relevant measured variables is required. Technical requirements arise from the areas of transmissions, clutches, chassis, brake systems and the brake, clutch and accelerator pedals. The sensory acquisition of the associated position values is or becomes necessary as the systems are to be increasingly electrically controlled and regulated or even electrified completely.

In automation technology, proximity switches, which are based on different physical-technical principles, dominate. The devices consistently operate switching functions with fixed switching distances and hystereses. However, with the increasing electronification in the control technology and the associated increasingly demanding control processes for the exact sequence control of the plant and automation components, increasingly sensor signals with a proportional position detection character are required.

Thus, a displacement and angle measuring system is proposed which has metrological properties which are about two to three times higher than comparable systems and, moreover, opens up application areas which can not yet be covered by today's systems.

State of the art

In motion and angle sensors, the following technical-physical principles of action have prevailed for applications in motor vehicle and industrial automation technology: inductively coil-based, in which case the eddy-current principle and the PLCD principle (permanent-magnetic linear contactless displacement sensor) apply are called, capacitive, optical, ultrasound and HALL effect. Automation technology mainly uses these principles to build non-contact position and limit switches. In the motor vehicle, the HALL effect as a non-contact switch was widely used, today also as a linear sensor in weg- and angle-proportional measurement tasks in use. Inductive and PLCD sensors are widely used in vehicles, capacitive and optical are hardly used here. Ultrasound is used as a remote field measuring system for the PDC (Park Distance Control) application.

For the sake of completeness, it should be mentioned that resistance-potentiometric principles as path-but in particular angle sensors served numerous application segments, but they are no longer in the focus of research and development and no longer represent the current state of the art.

The widely used principles of HALL effect and PLCD basically require a magnet which is movable along the measuring path for their control and should therefore be dispensed with for further evaluation.

In the following, the focus is now placed on the inductive principle, since the invention is based thereon.

An inductively operating linear sensor is in DE 20 2008 008 477 described, in which the measuring and evaluation element is housed as a stator in a rod-shaped housing. An annular index element moves along the rod-shaped housing and thus generates its position information relative to the measuring element.

Also inductive and detecting a linear path can be found in DE 10 2006 055 409 a configuration of two transmitting and one receiving coil, which are housed along the path to be measured in a sensor housing and realized as conductor tracks on a printed circuit board. For determining the position and consequently measuring path detection, two resonator circuits are used, which move parallel and longitudinally to the surface of the coil arrangement.

The concept of printed or etched trace coils on printed circuit boards to implement the inductive principle as a rotary encoder engages DE 101 56 238 in which in each case one excitation and two receiving coils are proposed, all of which are arranged in a circle and along the measuring angle as printed conductors on a printed circuit board. For determining the position and consequently measuring angle detection, two resonator circuits are used, which follow the angle of rotation parallel to the surface of the coil arrangement. Such a configuration can also be found in DE 199 41 464 , wherein a special embodiment of the receiving coil geometries is proposed here to increase the accuracy of measurement. The sensor should be used in particular for torsion on waves. The fundamentally identical mode of action is also used in DE 197 38 841 Here, a dual structure of two systems for mutual synchronization and fulfillment of the redundancy condition is proposed. Continue sets DE 197 38 836 also on the principle of coupled circuit board coils, wherein the coupling element is designed as a short-circuit line in periodically repeated loop structure. DE 197 38 834 describes a structure consisting of an exciter and three receiver coil configurations with a movable coupling element, the application aims at a 'soft' oscillator circuit for driving the exciter coil from.

Such PCB-based coils can also be used for the construction of inductive linear sensors, which in DE 100 26 019 is realized with an exciter and a plurality of receiving coils, which are arranged along the path to be measured in the sensor housing. For determining the position and consequently measuring path detection, a resonator circuit is used, which moves in parallel and longitudinally to the surface of the coil arrangement. In particular, it is also taken into account here that in the absence of a resonator, the coil output voltage becomes zero. In DE 199 20 190 an optimization is proposed such that a special configuration of the resonator circuit is provided with which inhomogeneities in the coupling magnetic fields can be minimized.

Inductive proximity switches can also be realized with said printed circuit board-based coils DE 10 2006 053 023 as a proposal for the arrangement of three coils, a Hauptsende-, a compensation and a receiving coil and their corresponding signal loading with the aim of maximum immunity describes. DE 103 50 733 shows an arrangement of a transmitting, a compensation and two receiving coils with the aim of increased switching distance. In DE 103 62 165 an arrangement of a transmitting and a plurality of interconnected receiving coils is described, which overlap the transmitting coil. In the switching distance, the magnetic flux induced in the receiver coils is just zero. Likewise describes DE 103 18 350 a proposal of a special geometry for a transmitting and a receiving coil and their corresponding positioning to each other. In the switching distance, the magnetic flux induced in the receiver coil is just zero. Ultimately, in DE 100 57 773 a transmitting and two receiving coils constructed mechanically and locally separate, wherein the applied conductor tracks on printed circuit boards receiving coils in turn consist of a plurality of individual coil sections.

Another class of realization concepts includes ferrite half-shell cores as an inductive sensor element, but these are consistently constructed as inductive proximity switches and not as Linearweggeber. However, they are mentioned for the sake of completeness, since the present invention utilizes the use of a ferrite coil as a sensor element.

So describes DE 10 2004 006 901 an oscillator circuit for the ferrite coil with an oscillator supply, which is equipped with a temperature sensor consisting of four diodes. The temperature sensing intervenes in such a way in the oscillator voltage supply that a temperature-compensated switching threshold and a compensated release distance associated therewith can be achieved.

Further inductive proximity switches: DE 43 05 385 - Automatic detection of the output circuit in either two- or three-wire operation. After detection, the device automatically adjusts to the load state connected by the user and operates the sensor permanently in this configuration. DE 41 41 264 - LC resonant circuit with measuring coil L in the upper and a controllable current or voltage source in the lower half branch of a full bridge circuit. The current or voltage source is readjusted by the bridge voltage amplifier depending on the impedance conditions at the measuring coil so that the bridge voltage always results in zero. DE 41 07 457 - Sensor coil on an amorphous metal body, which flooded this with the excited magnetic field. If one now approaches a magnetic trigger (permanent magnet) to the coil, the amorphous metal body is rapidly driven into magnetic saturation. This has the consequence that the amorphous metal body - provided with a correspondingly steep hysteresis curve - already undergoes a definite change of state for itself and thus alone causes a switching behavior, as a consequence, the subsequent circuit can be kept very simple. In addition, can be with the concept achieve a good temperature stability. DE 41 02 542 - LC oscillator, wherein the coil does not function here as a measuring coil but is used only to generate the basic oscillation frequency. To measure the arrangement contains two more sensor coils, which are connected in opposite directions and are designed so that the differential AC voltage at the desired switching distance is zero. The arrangement serves to achieve the largest possible switching distance.

From the analysis of the prior art, the following technical segmentation in inductive sensors becomes clear:
Ferrite cores are used in the design of semi-open pot core as sensor elements only in inductive proximity switches. These devices are designed as non-contact switches and used accordingly. Embodiments with an analog path characteristic as a transmission characteristic exist only in a few, under-represented product segments, so that they can be weighted as uninteresting for further consideration. In addition, these are by no means suitable for the intended applications in the automotive industry.

Modern position encoders for path or angle detection work consistently without contact and are consequently constructed either on the inductive principle or as a subcategory of these inductively-based on a coil plate. Pure inductive sensors have long coils with one, two or three wire windings, in which a metallic or ferromagnetic core or plunger armature moves. If one works here with a single winding, the damping principle is exploited, in the case of two windings it is the mode of operation of the so-called differential transformer and three windings are necessary for the construction of an LVDT (Linear Variable Differential Transducer). As a design feature is common to all these sensors that they can be used only as Linearweggeber due to the preferred straight construction of the coil.

Furthermore, their operation is limited to call contactless, since the core must always immerse in the coil, as a result of which a certain local proximity between coil and core is required and consequently no real effect is achieved over distance. Applications find such structured products exclusively in special measuring tasks of automation technology, they are consistently also relatively large, heavy and expensive.

Inductive sensors with printed circuit board coils represent the latest state of the art and can be realized both as Linearweg- or rotary encoder. Here, the coils are designed in the form of printed conductors on printed circuit boards, which then no more coil or wire wound body are required. The PCB coils send and receive high frequency alternating fields. Opposite the circuit board is a designated as a target or resonator counterpart whose position corresponds to the measuring position. This in turn can be designed as a simple circuit board with some short circuit conductors. If you now move the target parallel to the circuit board, so you can determine the position of the target relative to the circuit board with appropriate design of the PCB coil geometry.

Corresponding conductor track geometries and target constructions in circular form allow the construction of encoders, in longitudinal form corresponding to the structure of linear encoders.

Here, however, as a restriction, as mentioned above, these arrangements can not manage without near-field coupling between the coils and the targets. Thus, although the close range between the coil and the target can be constructed without mechanical contact, the actual measuring path can not be overcome in non-contact form or without mechanical auxiliary parts or actuators.

task

The invention makes it its mission to use the advantages of magnetic field generation with a ferrite coil for inductive displacement measurement for the implementation in a completely non-contact displacement and angle sensor and perform by appropriate, circuit, coil and construction measures so that not only the basic function of a non-contact distance measurement over distance is optimal, but also the technical properties in each case corresponding parameters of other principles of action are far above the comparable parameters of these respective known principles of action.

In addition, the invention has for its object to provide a reliable working device which detects the distance to a preferably axially moved to the ferrite coil surface object to be measured. However, due to a suitable arrangement between the coil and the object to be measured, rotational movements can be detected with the same accuracy due to the integration behavior caused by the magnetic field properties in the distance-volume model of the measuring section. The measured object will be referred to as a target hereinafter.

Furthermore, the sensor should be characterized by extremely small size and low weight. A relatively simple construction and the possibilities of integration of the sensor electronics, which may be multi-stage, allow a priori low costs. The system should be hermetically sealed complete, so that a whole range of environmental requirements such as mechanical robustness and impermeability to contamination and foreign substance inputs of any kind, such as solids, dusts, liquids or oils, can be accommodated. Also, the problem of icing, especially in the case of outdoor applications such as in the chassis of a motor vehicle, can be solved here sustainable.

Furthermore, the electronic and electro-mechanical structure should be kept so that a high immunity to electrical - conducted as well as irradiated - interference can be achieved. Conversely, the electrical voltages and currents in the system must be interpreted as low and as low frequency as possible, so that the arrangement emits no or very little electrical interference to the environment.

The solution of these objects is achieved according to the invention with the features specified in the claims 1 to 8. Preferred embodiments are subject of the dependent claims.

Preferably, a ferritic, one-sided semi-open shell core is used for the sensor coil, since this design is designed from its geometrical design forth already on a particularly favorable, local distribution of the radiated between the sensor surface and object to be measured (= target) magnetic field. These designs are also very easy to produce. The shell core requires a wound coil for its operation, the winding and material parameters of which have to be optimized with regard to the optimal functioning of the sensor. The inductance of the shell core coil will be referred to below as L. Since it is an AC oscillating system, a suitable sized capacitor is still required. This will be referred to below with C.

Here, the basic electrical structure of the resonant circuit - serial, parallel or as a split system - initially of secondary importance. It is important, and this is known for the design of such systems, that the electrical quality of the interconnection as high as possible to drive, since this has a favorable effect on the achievable range of the sensor. In turn, other parameters, such as electronic stability criteria and system bandwidth, must be taken into account.

Preferably, the target consists of a flat metal surface, the exact shape of which does not matter for the function. The inductive principle is able to detect any form of metallic objects. Here, there are only differences in the measurement sensitivity with respect to the type of metal. For example, iron can be detected better than stainless steel, which in turn is better than brass, which in turn is better aluminum and this better than copper. Non-ferromagnetic materials, such as plastics, organic materials or most fluids, do not affect the system, so are not measured in their presence and are therefore not disturbing.

In order to avoid complex simulation processes, which would have to include the complex magnetic field distribution in front of the coil and the eddy current effects in the metallic target, for the design of the resonant circuit including the radiated magnetic field to optimize the resonant circuit quality, the following methodology should preferably be used: The coil has the inductance L, the winding resistance is RI. The electrical losses around the resonant circuit are composed of two components: 1) The radiation of the magnetic field requires energy. This can be represented as field loss resistance and transformed to the terminal side of the coil. 2) Since the coil is necessarily operated in a real circuit environment, one has here only a finite ideal source for feeding the electrical energy as well as only a finely high-impedance signal tap for signal recovery. Both loss effects 1) and 2) can be represented in the equivalent circuit as a parallel resistor to the coil. This is called Rv. The capacitor used has the capacity C.

First of all, all parameters are to be regarded as variables. However, it is the case that the two variables L and RI are linked with each other and thus can not be set arbitrarily separately. One must therefore determine these under other optimization criteria. In the interpretation go: The desired inductance and the desired current consumption. Both quantities in turn are related to parameters such as vibration frequency, ferrite core material properties and winding design. Likewise, Rv on the magnetic field side is influenced by the shell core geometry and, on the electrical side, by the circuit design; here, after selecting certain configurations for the coil, fixed values are obtained. On the side of the electronic circuit, however, these can be optimized as will be described below. The aim is to set all these parameters as high-impedance as possible in order to obtain the lowest possible damping of the resonant circuit. For example, the transformed field loss resistance may be determined for a given configuration by a few relatively simple measurements.

Now only the resonant circuit capacitance C remains as a variable. In the following, one calculates the damping factor d of the configuration formed by L, RI, C and Rv, differentiates it to C (local minimum) and then obtains a calculation equation for C. For a simple parallel resonant circuit, this is for example: C = L / (RI × Rv). This has initially quality-optimized the design of the resonant circuit components. Other methods are of course conceivable. However, the goal is always to obtain the highest possible inertial quality of the vibration system.

The component of the loss resistance Rv influenced by the electronic circuit depends on how the resonant circuit is electrically driven and the resonant circuit signal containing the path information of the sensor is electrically tapped. In the following, only the use of a parallel resonant circuit of L and C will be discussed, since the circuit technology used has been tuned and optimized to this end. For such a resonant circuit, the control would be optimal over an ideal AC power source, since this couples infinite high impedance. For the signal extraction in turn, an infinite high-impedance amplifier input resistance would be the best case. Both can not be achieved in reality. However, assuming as a condition that all connected, electrical impedances are about six orders of magnitude (ie factor 10 ⁶ ) above the resonant circuit impedance in the case of resonance, one comes very close to the ideal conditions of a free-running resonant circuit.

Furthermore, there is a requirement to always operate the resonant circuit independently of the damping situation generated by the target, since the quality values are naturally highest here.

All these requirements can be achieved preferably by connecting the resonant circuit in a free-running oscillator such that the first terminal <A1> in a correspondingly implemented in the oscillator control loop, although only finely high-impedance is driven, the second terminal <A2> but by the Inverted control circuit is supplied with energy so that the losses caused by the non-ideal control can be exactly compensated again. The control of the second terminal <A2> can be done by an AC voltage source, ie low impedance. This leaves only the inaccuracy of the controlled system as a loss feature left, which can be kept very small according to the above requirement.

For the signal tap and at the same time as an input amplifier of the control loop with the oscillator, preferably an operational amplifier can be used whose input impedance is the required six orders of magnitude above the resonance impedance of the resonant circuit. The resonant circuit should be dimensioned so that its own maximum impedance - the impedance without a damping effect by a target - in the order of 1 Mohm. This requires an amplifier input resistance of about 1 TOhm, which can be achieved with commercial operational amplifiers without much effort and expense.

The circuit, consisting of the oscillator with the control circuit for resonant circuit operation, should be free and self-starting under all conditions and also always run on the resonant frequency of the resonant circuit. The self-oscillation condition is known to be difficult at high resonant circuit attenuations, ie small measuring distances, since all useful signal amplitudes in the system become very small. Here it must be ensured that it does not come to a so-called 'tearing' the vibration, as a result, the sensor characteristic is no longer closed continuously and steadily - so jump-free. The back-channel controlled system in the control loop fulfills the stated conditions. This picks up the oscillator output voltage Uo and returns it as a synchronizing signal for driving the L-C resonant circuit back into the oscillator or the resonant circuit. Among other things, a signal conditioning of Uo is carried out in this return path, z. B. a sine-square conversion.

Taking into account the criteria listed above, ie optimization of the L-RI-C configuration of the ferrite coil and optimal resonant circuit, it has now been created a system that represents the ideal configuration for the quality-optimized operation of the sensor coil. This achieves optimal range results with respect to the measuring path of the system, which are now mainly limited by the caused by the resonant circuit, internal losses.

To overcome these, one must make further considerations. For the construction of the magnetic field, which is needed between the sensor coil and the target for path measurement, naturally energy is always consumed. The resonant circuit damping produced as a result is several orders of magnitude greater than the damping effects produced by internal losses in the ferrite material and in the coil winding. The immanent magnetic field losses can now be compensated preferably on the electrical side, since the system is fully reversible, that is to say effects in the magnetic field are formed on the electrical side and vice versa. Due to the fact that the circuit properties described above are already ideal with the Avoiding resonant circuit coupling, additional measures must be taken, with which the energy consumed in the magnetic field can be compensated. For this purpose is preferably - active and circuitry controlled - additional auxiliary energy fed back into the resonant circuit.

This can be done, for example, by picking up the signal at the second terminal of the resonant circuit <A2> or the signal Uo. This is easily and effortlessly possible, since here each AC sources are and you are here low impedance. Now, a small additional current is generated by an appropriate circuit technology, which is controlled as a phase-correct superposition signal in the resonant circuit (terminal <A1>). Since this is again in-phase condition, the requirements for the high resistance are given only conditionally. One will choose an impedance value for the current source which is approximately at the resonance impedance of the resonant circuit (so-called power adjustment). The additional energy of this energy return compensates for the internal field losses, thus increasing the quality of the system, thus contributing to an increase in the measuring range of the sensor.

The oscillator output voltage Uo is available as an alternating voltage signal which oscillates in the natural frequency of the resonant circuit and contains in its amplitude the distance information of the sensor. This signal will not be usable for almost all applications, so it must be reworked accordingly. It is preferable to use a microcontroller (uC), irrespective of the technical field of application, to perform various signal processing tasks. Here, Uo can be converted into a DC voltage in a synchronous rectifier, for example, and fed to an ADC input of the μC (ADC = Analog-to-Digital Converter). Another possibility is a phase-locked, but not every period sampling of Uo and measurement in the ADC, which saves a synchronous rectification and must be designed only according to the sampling theorem to ensure the system bandwidth. A synchronous clock is obtained, for example, from the output signal of the return path controlled system.

Furthermore, an NTC thermistor can be used to measure the sensor and ambient temperatures. Its values are preferably also read into the μC. The uC has algorithms for temperature compensation of the system and, if necessary, for the linearization of characteristics, if this is necessary, depending on the characteristics of the transfer function of the sensor in the application.

The output signal of the sensor can also be processed in the uC via the software and made available at the output. Common: Analog voltage or current levels, PWM (Pulse Width Modulation) and various bus systems, such as LIN (Local Interconnection Network), Local Area Network (LAN), SEN (Single Edge Nibble Transmission) or PSI (Peripheral Component Interconnect) ,

embodiment

In the following the invention will be explained in more detail with reference to the drawings.

Show it:

1 : the relationship between oscillator output voltage Uo and measuring distance x of the system according to the invention, parameter is the resonant circuit quality Q.

2 : the relationship between resonant circuit impedance Z and angular frequency ω of the system according to the invention, parameter is the measuring distance x.

3 : The block diagram of the electronic interconnection for the sensor with the necessary to achieve the required system properties circuit blocks.

4 : A schematic representation of the measuring paths x between three shell core coils and targets for comparing various inductive systems.

5 : the transfer characteristics of in 4 illustrated configurations.

In 1 First, the relationship between the oscillator output voltage Uo and the measuring distance x is plotted. In this case, the oscillator circuit used or the exact resonant circuit circuit is initially largely irrelevant. If one uses an oscillator whose output voltage Uo follows the resonance impedance of the resonant circuit, then this curve corresponds to the resonant circuit impedance curve over the measuring path x. Since the values of the resonant circuit elements L, RI and C are fixed, the course of the further also corresponds to the course of the transformed field resistance. The shape of the curve follows exactly a tanhyp function. For large values of x - corresponding to the state, no target '- the function reaches its maximum, the resonant circuit runs here virtually without external damping.

The area for small x, however, is much more demanding in the realization: for one Position sensor system can be acceptable only a characteristic that is continuously closed and continuous, so shows a jump-free course. This in turn means that the oscillator has to exhibit non-disruptive rocker characteristics under all operating conditions. For x = 0, that is, in the worst case on the pot core resting target, the oscillation amplitudes in the oscillator can be very small due to the high attenuation values. The functional condition can be set here to 0.1 ... 1 ‰ of the final value. Assuming that in a system supplied with 5 V, the sinusoidal oscillator output voltage Uo = 2 Vss, then 0.2 ... 2 mVss of signal regeneration must be subjected here to obtain a permanent feedback for a self-oscillating system.

As a parameter is in the curves of 1 the influence of quality Q is plotted. The solid line represents a low inertial arrangement, the dashed line describes a middle and the dotted line a high quality. The term inertial quality here stands for the respective basic parameter, ie for a system without external damping (without target, x → ∞). As the resonant circuit quality increases, the travel curves become ever flatter. This means in consequence that the achievable measuring path, otherwise exactly the same configuration of the shell core coil, increases. It is irrelevant which parameters affect the quality, that is to say electrically, by choice of material or magnetic field properties.

2 shows curves of the LC resonant circuit impedance Z over the angular frequency ω. The parameter here is the measuring path x, ie the distance between the coil and the target. As the measuring path increases, the quality Q of the system, which is basically defined as Q = ω ₀ / B (resonance frequency ω _0, -3 dB bandwidth B), also increases. Since the overall circuit is designed as a self-oscillator, the resonance condition is fulfilled at all times, that is, the resonant circuit impedance is always at the resonant frequency ω ₀ and thus always at the maximum. With decreasing x, ie increasing damping, the curves shift slightly towards lower values of ω ₀ due to a relatively spontaneous and non-linear increase in the inductance L, but the effect is in the range of less than ppm and can also be neglected always readjusted to resonance mode.

On the one hand, the quality of the system, which is variable over the measuring path x, is the actual signaling effect; on the other hand, a special feature in the system properties is shown here: The frequency bandwidth behaves in a control-dependent manner, ie depending on the currently measured displacement x. In terms of frequency response, the sensor thus behaves like a first-order variable low-pass filter, with its -3 dB cut-off frequency at the respective measuring distance x corresponding to the -3 dB bandwidth of the quality curve.

In the 2 Curves shown are for a particular configuration of the circuit and resonant circuit parameters. With a different design, in particular the increase in quality due to the implementation of the energy return, the curves accordingly shift towards higher values and lower bandwidths. The resonant frequency ω ₀ is always given by the LC combination and remains under all circumstances only dependent on this.

In 3 is a preferred circuit configuration 10 reproduced to obtain the system properties described as a block diagram. The oscillator 20 is with its input terminals 21 and 22 with the clamps 31 and 32 of the LC resonant circuit 30 connected, the current waveforms are indicated by the current arrows. Here it is completely irrelevant whether 30 is constructed as a series or parallel resonant circuit or a plurality of divided coils. clamp 21 serves as a signal tap for the oscillating circuit signal and is the controlling variable in the entire, later control processes. The sinusoidal oscillator output voltage Uo is connected to terminal 23 and in amplitude contains the path information Uo (x) to be measured.

To dampen the resonant circuit as weak as possible through the wiring, the input impedance is connected to terminal 21 by about six orders of magnitude (ie a factor of 10 ⁶⁾ via the resonant circuit impedance at resonance. Typically, the latter will be in the range 0.1 ... 1 MOhm, so that at terminal 21 about 0.1 ... 1 TOhm must be achieved. This is easy to implement with conventional operational amplifiers.

clamp 22 represents an alternating voltage source or sink, the temporal signal conforming to the waveform at terminal 23 runs. Circuit topology can 22 also with 23 be equated. As the resonant circuit 30 in the case of resonance with all loss effects is a purely ohmic resistance, the signals at <A1> and <A2>, of low running time effects in the oscillator 20 apart, exactly inverted, ie phase-shifted by 180 °. Thus, the control condition for a permanent and always operated in resonance frequency oscillation can be ensured at the same time optimally damping-free Schwingkreisbeschaltung at any time.

To expand the system now to a permanent and constantly excited self-oscillator, is a feedback channel controlled system 40 introduced. Their input impedance at terminal 41 is irrelevant, the output impedance at terminal 42 should like follows to be highly resistive: through the active-inverse operation of the resonant circuit 30 through the oscillator 20 is a disturbing damping by the interconnection of terminal 42 at terminal 21 Although reduced by the factor of the control gain, by the real conditions you have here, however, limitations. As low has been proven when the output impedance to 42 is about the factor 2 ... 3 above the undamped resonant circuit resonance impedance (without target). Then it can be almost ideal and connected by the circuits 20 and 40 undisturbed operation of the resonator can be assumed.

The return channel route 40 Simultaneously serves the signal conditioning of Uo. In fully damped operation of the resonant circuit 30 , ie at x = 0, the AC amplitudes are only a few mV. These are through the circuit 40 so prepared that at terminal 42 at any time a constant alternating current amplitude for feeding back into the resonant circuit 30 is available.

Due to the 180 ° phase condition between <A1> and <A2>, the signal must also be on 41 and 42 be exactly 180 ° out of phase. Minor run-time effects, as they occur in the implementation in a real circuit technology, are allowed.

By the described interpretation of the corresponding circuit parameters in the oscillator 20 and return channel 40 becomes the resonant circuit 30 now initially operated in the condition of optimum quality. A further increase would not be feasible by whatever measures are taken to optimize the resonant circuit parameters and the connected circuit technology. The losses are defined by the energy consumption in the magnetic field and the damping effects in the target and the oscillating circuit elements.

To still achieve a further increase in the quality of the system, is due to the energy return 50 More energy in the right way in the resonator 30 fed, that the losses occurring therein further minimized, the quality of the system increased and thus the range of the sensor can be significantly improved.

In addition, the energy return takes effect 50 via its input terminal 51 the signal Uo at the output terminal 23 of the oscillator 20 It conditions and conditions it as a constant and displacement proportional auxiliary current through the output terminal 52 in the resonator input 31 is fed. By this measure, the electrical quality of 30 further increased, which at the same time leads to an increase in the Sensormeßwegs. The output impedance of the output side power source to terminal 52 should ideally be of the order of magnitude of the undamped resonant circuit resonance impedance (without target).

Due to the possibility of individual adjustment of the amount of energy returned, the measuring range of the sensor is now no longer quasi-static, solely by the mentioned optimization parameters for coil and circuit, but can be set dynamically and, above all, purely electrically within relatively wide limits. In the 2 As a result, resonant curves shown are no longer dependent solely on the measuring distance x for a given coil-circuit configuration, but also on the inertial quality of the system set by the energy feedback. Since the quality Q and the system bandwidth B are directly linked with one another, the dynamic behavior of the sensor can also be influenced beyond that.

4 shows a comparison of achievable by different systems Meßreichweiten. Here are with 61 . 62 , and 63 each half-side open shell core coils shown, consisting of respective ferrite cores 64 . 65 and 66 exist, in which the sensor coils 67 . 68 and 69 are embedded. The shell cores radiate their magnetic fields 71 . 72 and 73 off to the open side. 71 . 72 and 73 are shown only schematically simplified. The corresponding targets are with 81 . 82 and 83 denotes their diameter or their edge length (square metal piece) corresponds to the pot core diameter D, so a standard measurement configuration. The targets 81 . 82 and 83 are drawn in such a way that they are located at the respective maximum measuring range position.

The upper configuration ( 61 . 71 . 81 ) represents a typical inductive proximity switch as described in detail in the prior art. The switching distances of such devices are about one third of the pot core diameter D, denoted here sn, sn = 0.3 D. The average configuration ( 62 . 72 . 82 ) stands for a proportional measuring system, which is based on a possible low-damping operation of the resonator 30 is designed. This corresponds to a circuit 20 . 30 and 40 according to 3 , The maximum measuring path xm1 here corresponds approximately to the pot core diameter D, xm1 = D.

The lower configuration ( 63 . 73 . 83 ) is created by adding the energy feedback 50 into the sensor circuit 10 , Here, the maximum measuring path xm2 is now set to about twice the pot core diameter D, xm2 = 2 D. Since the amount of energy returned is electrically adjustable, other values can also be set depending on the application, for example xm2 = 1.5 D or xm2 = 2.5 D. A technical limit is reached where higher-order stability criteria in the circuit can no longer be safely met.

The to the arrangements of the 4 belonging transmission characteristics of the sensors are in 5 summarized. The dashed curve shows the switching characteristic of the proximity switch 61 . 71 . 81 with sn = 0.3 D. Of course, these switching functions always have a switching hysteresis in practice, which for reasons of simplification is not shown here. The solid line represents the arrangement 62 . 72 . 82 with xm1 = D, the dotted line ultimately for the arrangement 63 . 73 . 83 with xm2 = 2 D. It can be seen on the solid line the already large, proportional measuring path, which can be realized by an approximately ideal operation of the resonator, namely largely externally undamped, while the dotted line the further improvement, by introducing a controlled energy recovery can be achieved.

Claims (10)

Device for non-contact measurement of the distance (x) to a free-moving metallic counterpart ( 81 . 82 . 83 ), which as a physical-electrical measuring means a corresponding magnetic field emitting, half-open shell pot ( 61 . 62 . 63 ) of a resonant circuit ( 30 ) and which at any time of a circuit configuration ( 10 ) is operated in the resonant frequency, by the resonant circuit comprising the half-open shell core ( 30 ), the circuit configuration ( 10 ) an oscillator ( 20 ) and a return channel ( 40 ), Constitutes characterized in that the circuit configuration ( 10 ) a controlled energy return ( 50 ) and that by means of the controlled energy return ( 50 ) maximizing the quality of the vibration system ( 10 . 30 ), consisting of the resonant circuit ( 30 ) and an associated circuit configuration ( 10 ), which in consequence causes an enlargement of the measuring range (x) of the device. Device according to claim 1, characterized in that the circuit configuration ( 10 ) basically the, the half-side open shell core containing resonant circuit ( 30 ) so electrically controls that the lowest possible damping of the same by the control and as a result, a maximum system quality of the arrangement can be achieved. Device according to claim 1 or 2, characterized in that the circuit configuration ( 10 ) basically the, the half-side open shell core containing resonant circuit ( 30 ) so electrically tapped that the lowest possible attenuation of the same by the signal pickup and consequently a maximum system quality of the arrangement can be achieved. Device according to one of claims 2 to 3, characterized in that the amount of recirculated energy is adjustable and that by the energy return ( 50 ) the electrical quality of the resonant circuit ( 30 ) and thus the quality of the overall system ( 10 . 30 ), beyond the measures of claims 2 and 3, can be increased in such a way that it is an adjustable and significant measure above the quality, which by the configuration only from oscillator ( 20 ) and return channel ( 40 ) would be achievable. Device according to one of claims 1 to 4, characterized in that the half-open shell pot ( 61 . 62 . 63 ) with the inductance (L) in conjunction with the resonant circuit capacitor (C) a parallel resonant circuit, a series resonant circuit or any electrically resonant system is formed, which is capable of emitting a measuring magnetic field capable. Device according to one of claims 1 to 3, characterized in that the output ( 42 ) of the return channel controlled system ( 40 ) the electrical behavior of an AC source with a source impedance of 2 to 3 times the unattenuated resonance impedance of the resonant circuit ( 30 ), which operates at the resonant frequency. Device according to one of claims 1 to 4, characterized in that the output ( 52 ) of the energy return ( 50 ) the electrical behavior of an AC source with a source impedance on the order of the undamped resonant impedance of the resonant circuit 30 corresponds, which works in the resonant frequency. Device according to one of claims 1 to 7, characterized in that by means of a more or less plane-parallel target in the axial and parallel direction to the coil surface, a linear movement shape is detected. Device according to one of claims 1 to 7, characterized in that in the axial direction to the coil surface, a rotational, ie rotating movement form is detected by the rotational movement is converted by means of a running over the rotation angle, suitable target geometry contour in a linear movement. Device according to one of claims 1 to 7, characterized in that in the axial direction to the coil surface a rotatory, ie rotating movement shape is detected by the tilt angle of the rotational movement is laterally outside the escape direction of the emitted magnetic field and by means of more or less plane-parallel target a mixed form of linear and tilting movement is generated.

Images