DE102015215352A1 - Sensor system with microcontroller and minimized peripherals - Google Patents

Abstract

The invention relates to a device for measuring a measured variable, wherein a first inductance is supplemented by a capacitance to a parallel resonant circuit and excited by a microcontroller. A measuring inductance coupled to the first inductance is measured by the microcontroller. This makes it possible to deduce a measured variable, with only a few components being required beyond a microcontroller.

Description

The invention relates to a device for measuring a measured variable. In particular, it may be an inductive sensor.

Inductive sensors are typically based on a change in one or more characteristics of a system of one or more inductive components by a measurand. Such inductive components can be called, for example coil, winding or inductance.

• – Selbstinduktivität L, auch kurz Induktivität genannt,
• – Verlustwiderstand R, der sich aus einem ohmschen Widerstand der Wicklung und anderen Verlustbeiträgen zusammensetzt,
• – komplexe Impedanz Z = jωL + R mit der imaginären Einheit j und der Kreisfrequenz ω,
• – Verlustwinkel δ = arctan(Re{Z}/Im{Z}),
• – sowie, insbesondere im Falle der magnetischen Kopplung zwischen mehreren Bauelementen, Gegeninduktivität M. Die Gegeninduktivität M kann u.a. als induzierte Spannung in einem Leiter als Reaktion auf einen bekannten Strom in einem anderen Leiter indirekt gemessen werden.

As parameters are in particular:

• - Self-inductance L, also called inductor for short,
• Loss resistance R, which is composed of a resistance of the winding and other loss contributions,
• Complex impedance Z = jωL + R with the imaginary unit j and the angular frequency ω,
• Loss angle δ = arctan (Re {Z} / In {Z}),
• - And, in particular in the case of magnetic coupling between multiple components, mutual inductance M. The mutual inductance M can be measured indirectly as induced voltage in a conductor in response to a known current in another conductor, inter alia.

Measurands that cause the change of the characteristic values can be, among other things, position or length, angle, force, pressure or torque. As an example, a position sensor for the brake pedal of an automobile may be cited.

For inductive sensors exist in the prior art, in particular two main circuit approaches to make an electrical measurement of the characteristics:
On the one hand, this is a resonant system: The inductive sensor with its variable characteristic, usually the inductance L, is part of the frequency-determining network of an oscillator. The oscillator always oscillates with its natural frequency, whose most important influencing factor is L. The measurement of L is thus attributed to a frequency measurement, which can easily be made, for example, by counting the periods or zero crossings of the oscillator oscillation.

On the other hand, this is a lock-in amplifier (also phase-sensitive rectifier, synchronous demodulator or carrier frequency amplifier): The inductive sensor receives a stimulus with a fixed frequency (current or voltage). A signal processing circuit measures the other electrical quantity at the impedance (voltage or current). The signal processing corresponds to a narrowband filtering of this size around the frequency of the stimulus with subsequent determination of the complex amplitude and quotient formation with the stimulus for determining the characteristic value. These functions can be implemented either with analog electronics or largely with the means of digital signal processing and software.

Both approaches have different disadvantages.

The resonant system has limitations in the design of the inductive system because only one oscillation per oscillator is possible. Multiple signals can only be obtained with multiple independent oscillators and inductive systems, which significantly increases the cost of using ratiometric or differential measurement sensors. Moreover, the inductive system always has frequency dependencies, i. it can only be optimally designed for one frequency; the frequency range of the oscillator is always a compromise. Cross-sensitivities can falsify the measurement result by changing the oscillation frequency because, for example, the inductance L is influenced by a further frequency-dependent variable in addition to the sensitivity to the measured variable. Finally, the difference between the maximum and the minimum count result of the frequency measurement must exceed a minimum value, so that the respective requirements with regard to measurement accuracy and resolution can be achieved. Depending on the frequency, a minimum measuring time is required, which may not be available under certain circumstances.

The lock-in amplifier, on the other hand, operates at a constant frequency, but also requires a stimulus at that frequency. The frequency of these forced vibrations is arbitrary, but due to the frequency dependence of the inductive system, this is in contradiction to operation, i.e., resonance. Therefore, the following advantages of the resonance can not be exploited: The inductive system, operated as a resonator, already constitutes a filter in that it can reach a particularly high amplitude in its natural frequency, which facilitates the measurement. Disturbances whose frequency deviates significantly from this frequency are suppressed by the filter effect. Moreover, in resonance, the power requirement of the inductive system for sustaining the vibration is lowest when all other parameters remain the same. Given the power of the stimulus, a particularly high amplitude is thus possible. Of course, these two advantages represent the same facts, once from a measurement point of view and once from a stimulus point of view.

It is therefore an object of the invention to provide an alternative, compared to the prior art, In particular, provide improved apparatus for measuring a measured variable.

This is inventively achieved by a device according to claim 1. Advantageous embodiments can be taken, for example, the dependent claims. The content of the claims is made by express reference to the content of the description.

The invention relates to a device for measuring a measured variable. The device has a first inductance. It has a measuring inductance which is designed to sense the measured variable and which is coupled to the first inductance. In addition, it has a capacity which is connected to the first inductance to a parallel resonant circuit.

The device further comprises an electronic control unit. The electronic control unit is connected directly to the parallel resonant circuit and is configured to oscillate the parallel resonant circuit with an excitation frequency derived from a clock of the electronic control unit. In addition, the electronic control unit is connected directly to the measuring inductance and is configured to measure a value indicating the measured variable at the measuring inductance.

By means of the device according to the invention, the measuring inductance can be measured only by using an electronic control unit in a resonant circuit. This saves energy because the resonant circuit oscillates on its own and as a rule only lost energy is tracked. So there is a forced oscillation. This allows exploiting the benefits of resonance and is free of the above-mentioned limitations of a resonant system. The use of an electronic control unit to implement the functions saves costs and simplifies the system.

The excitation of the resonant circuit can also be called a stimulus. The excitation frequency can also be called the stimulus frequency. The derivation of a clock can be done in particular by the derivation of an internal, but also by an external clock. This means in particular that the excitation frequency is independent of a resonant frequency of the parallel resonant circuit.

In particular, the device may be constructed so that it has no external to the electronic control unit amplifier. This also saves costs. Typically, the provision of a direct connection excludes a respective amplifier between electronic control unit and component such as inductance or capacitance. A respective amplifier can be integrated in particular in the electronic control unit.

• 1. Die Spannungsspitzen und damit die Gefahr für die Portpins verschwinden.
• 2. Der Strom durch die Induktivitäten steigt deutlich an, daher wird mehr magnetischer Fluss im induktiven System erzeugt, gleichbedeutend mit einem stärkeren Signal.
• 3. Der Strom durch die Induktivitäten wird nahezu sinusförmig, was unter Gesichtspunkten der elektromagnetischen Verträglichkeit günstig und für die nachgeschaltete Signalverarbeitung vorteilhaft ist.
• 4. Der Strom in den Zuleitungen der Schwingkreise sinkt, weil die Impedanz des Schwingkreises in der Nähe der Resonanz besonders hoch ist. Dadurch wird ermöglicht, den Stimulus mit wenigen Portpins bzw. Ausgangsstufen geringer Strombelastbarkeit zu erzeugen.

The effect of a resonant circuit in the vicinity of the resonant frequency causes four advantages according to the invention even without resonant oscillation:

• 1. The voltage peaks and thus the danger for the port pins disappear.
• 2. The current through the inductors increases significantly, therefore, more magnetic flux is generated in the inductive system, equivalent to a stronger signal.
• 3. The current through the inductors is almost sinusoidal, which is favorable from the viewpoint of electromagnetic compatibility and advantageous for the downstream signal processing.
• 4. The current in the supply lines of the resonant circuits decreases because the impedance of the resonant circuit in the vicinity of the resonance is particularly high. This makes it possible to generate the stimulus with few Portpins or output stages low current carrying capacity.

To protect the port pins from voltages outside the allowable range, the complementary stimulus is advantageous because it causes oscillation about the average of the supply voltage. In this case, an amplitude of likewise half the supply voltage is possible or permissible; depending on the specification of the microcontroller also beyond, for example, up to half the supply voltage +0.5 V. The voltages at the port pins are typically no longer rectangular voltages because of the internal resistance of the output stages. A stimulus that works unilaterally with port pins against reference potential or supply voltage typically offers no comparable possibilities, because any stimulus (AC voltage) that is added to these voltages is already outside the permissible range at low amplitudes.

An electronic control unit can be understood as meaning, for example, a microprocessor or a microcontroller. Under a microcontroller is in particular a device to understand that in addition to the functions of a microprocessor and read-write memory (RAM), at least one kind of read only memory (ROM, EPROM, EEPROM, Flash, etc.), analog-digital Converter (ADC) and / or timer and port blocks integrated on a chip contains. Such a device typically does not require external memory for operation and may execute a program without external circuitry, make measurements with the ADC, and make digital signals via port pins which are determined by software or by the timer blocks. For this purpose, the microcontroller typically receives only a voltage and clock supply and a reset pulse at power. In modern components, a clock oscillator and a reset logic can also be integrated, so that typically only the supply voltage is supplied from the outside.

However, the electronic control unit may be, for example, an assembly of a microprocessor with external circuitry such as memory and clock supply, an application specific integrated circuit (ASIC), a programmable logic controller (PLC) or an integrated circuit (IC).

For measuring, an analog-to-digital converter of the electronic control unit or of the microcontroller is typically used. For most microcontrollers, the ADC is connected through a multiplexer to a number of port pins where the voltage can be measured. The number of port pins used as input and the sequence in which the inputs to the ADC are turned on are typically configurable. The port pins connected to the multiplexer are thus available as inputs for performing a number of independent measurements.

It is assumed that the measurement results are usually not only to remain in the electronic control unit or in the microcontroller, but are passed on to other systems, for example by a serial interface. For this purpose, suitable components may be present.

An integrated circuit (IC) may also be considered or defined as a microcontroller containing the above-mentioned circuit functions. It is not necessary that the IC is also marketed by the manufacturer as a microcontroller or as a standard module is freely available for sale. In this sense, in particular an ASIC ("Application Specific Integrated Circuit", in particular an IC developed for a customer application), which fulfills all technical criteria, is considered as a microcontroller.

• – Selbstinduktivität L, auch kurz Induktivität genannt,
• – Verlustwiderstand R, der sich aus einem ohmschen Widerstand der Wicklung und anderen Verlustbeiträgen zusammensetzt,
• – komplexe Impedanz Z = jωL + R mit der imaginären Einheit j und der Kreisfrequenz ω,
• – Verlustwinkel δ = arctan(Re{Z}/Im{Z}),
• – sowie, insbesondere im Falle einer magnetischen Kopplung zwischen mehreren Bauelementen, Gegeninduktivität M. Die Gegeninduktivität M kann u.a. als induzierte Spannung in einem Leiter als Reaktion auf einen bekannten Strom in einem anderen Leiter indirekt gemessen werden.

In particular, the following values can be considered as the measured value:

• - Self-inductance L, also called inductor for short,
• Loss resistance R, which is composed of a resistance of the winding and other loss contributions,
• Complex impedance Z = jωL + R with the imaginary unit j and the angular frequency ω,
• Loss angle δ = arctan (Re {Z} / In {Z}),
• - And, in particular in the case of a magnetic coupling between a plurality of components, mutual inductance M. The mutual inductance M can be measured indirectly, inter alia, as induced voltage in a conductor in response to a known current in another conductor.

From the value indicating the measured variable can typically be concluded on the measured variable, for example, the measured variable can be calculated or looked up in a table.

According to a preferred embodiment, it is provided that the parallel resonant circuit is connected to a first port pin and to a second port pin of the electronic control unit. Cleverly, the second port pin is supplied with a clock signal inverted to the first port pin.

In this way, it can be achieved that an oscillation of the parallel resonant circuit does not take place around a reference potential, in particular ground, or about the supply voltage, but by half the supply voltage or at least approximately in half the supply voltage. This avoids a risk to the electronic control unit or the port pins or their wiring.

The excitation frequency can be achieved in particular by dividing down or assuming a system clock at the port pin. This allows the provision of a signal clocked at the excitation frequency at the port pin using common functionality of a control unit such as a microcontroller.

In particular, the stimulus may be derived from a system clock of a microcontroller or other electronic control unit. For this a timer or counter can be used. This typically divides the clock in an integer ratio down to the frequency of the stimulus. In some cases, a direct use of the system clock is possible. The divided clock is preferably connected to the output stages of two or more port pins, with one port pin or a group of port pins being driven by the clock itself, while a second port pin or a second group of port pins is driven by the inverted clock. Depending on the architecture of a microcontroller, it may be necessary to use a separate timer for each port pin, because each timer can only be connected to a specific port pin without the possibility of branching the timer signal. This circumstance is usually not a problem, because modern microcontroller have many timers and usually only a few port pins are needed, a minimum of two.

Since the frequency of the stimulus can be selected by setting the timer, this frequency can optionally be changed during operation of the system by software.

According to one embodiment, a respective resistor, in particular an ohmic resistor, can be connected between a number of the port pins and an inductance, that is, for example, the first inductance or the measuring inductance. This, too, is still understood as a direct connection. In particular, a current can be limited with the resistor, which avoids damage due to overloading.

The parallel resonant circuit may be connected in parallel to the first port pin with a number of further first port pins, which are switched synchronously to the first port pin. It can also be connected in parallel to the second port pin with a number of further second port pins, which are switched synchronously to the second port pin. Thus, the respective maximum current loads of the port pins can be added, so that a total of a higher maximum current can be achieved than when using only one port pin.

In this context, a synchronous circuit is to be understood in particular as meaning that the same potential is fundamentally applied to the respective port pins, that is, it is also switched over at the same time.

The voltages at the port pins serve as a stimulus in particular. There are two main reasons, optionally instead of a single port pin being clocked, as well as a single port pin supplied with the inverted clock, each using a group of port pins. On the one hand, a number of port pins, which have too low an ampacity, can provide, by parallel connection, a number of times the amount of current, as already mentioned above. Thus, inductive systems can be supplied with a stimulus whose impedance would otherwise be too low. On the other hand, the inductive system can provide several inductors to be supplied with stimulus. In this case, the use of one timer per port pin can optionally also make it possible to supply the individual inductances with stimuli of different frequency or phase angle.

It should therefore be mentioned that it is also possible to connect further inductors to further port pins.

Portpins to which the parallel resonant circuit is connected preferably have a push-pull output stage or a tristate output stage. This may in particular mean that each one switching element (bipolar or MOS transistor) is present for a low-resistance connection to the voltages of the two logic levels. Such embodiments have proved to be advantageous for the present application.

According to a preferred embodiment, it is provided that the excitation frequency differs by a maximum of 25%, preferably a maximum of 20%, more preferably a maximum of 15%, even more preferably a maximum of 10% from a resonant frequency of the parallel resonant circuit. This achieves an advantageous excitation. Operation at resonant frequency is typically not intended, but can normally be tolerated as a limiting case.

The circuit of the stimulus, in particular also the measuring circuit described herein, scales well and can be operated with excitation frequencies from the kHz range up to the MHz range, without changing the requirements for the electronic control unit or the microcontroller, e.g. preferably from 5 kHz to 5 MHz. For this purpose, typically only the parameters of the inductive system and the capacitor of the resonant circuit are adjusted. The lower limit of 5 kHz, in particular, causes the inductances to be made sufficiently small and to have the least possible influence on the system to be measured. In particular, the upper limit causes the parasitic capacitances of the port pins and the connected outer network, together with the internal resistances of the output stages, not to form effective low-pass filters. However, it should be understood that frequencies outside the specified range can also be used, since maximum tolerable influences always depend on the actual conditions.

Within the desired stimulus frequency range, the upper range, e.g. from 500 kHz to 5 MHz, for many measurement tasks, especially in the automobile, preferably suitable. The inductive system is then very small and, since the required inductances amount to only a few microhenrys, can also be integrated in the form of planar coils in a compact and cost-effective manner on a printed circuit board. If the inductive system is realized in the form of planar coils, then a stimulus frequency in the upper region is particularly preferred, because otherwise the impedances in the inductive system typically become too small to be accessible to precise measurement without great effort.

It is preferably provided that the measuring inductance is connected to a first pole on a port pin of the electronic control unit, and connected to a second pole with a potential which at least approximately corresponds to half the supply voltage of the electronic control unit.

An at least approximately half supply voltage may in this case be understood in particular to be a value which has a deviation of ± 10% or ± 5% from half the supply voltage. It can also be used exactly half the supply voltage. A deviation from this exact value does not lead to a measurement error, but may possibly only worsen the achievable resolution.

The potential, which corresponds at least approximately to half the supply voltage of the electronic control unit, can be generated, for example, by means of a voltage divider.

The potential, which corresponds at least approximately to half the supply voltage of the electronic control unit, can also be generated by means of a smoothing capacitor, the smoothing capacitor having a first pole connected to a reference potential or to the supply voltage, and wherein a second pole of the smoothing capacitor via a resistor a port pin of the electronic control unit is connected, which is acted upon by a pulsed signal, in particular with a signal with a predeterminable duty cycle.

The resistor can be designed in particular for achieving a small residual ripple, wherein it preferably has a resistance value of over 100 kΩ.

The embodiments described above are based on the recognition that the voltages on the port pins may typically exceed only minimal the limits given by the supply voltage. For the secondary windings of a measuring transformer, this means that they are preferably not simply connected to reference potential or VDD on one side and connected to the port pin with the other terminal, because in such a case, typically already small amplitudes would lead out of the supply voltage range. Instead, the indicated potential of approximately half the supply voltage is preferably used in order to allow maximum amplitudes.

This can be solved with a simple voltage divider, which corresponds to a simple design. It can also be used as specified an RC element. The capacitor is connected on one side to the supply voltage or reference potential, the other side is connected to a terminal of a secondary winding of the measuring transformer. The resistor is also connected at this node, the value of which is typically in the range of more than 100 kΩ. The other side of the resistor is connected to a port pin. Typically, the output stage changes in a timer-controlled manner between the logic levels and possibly also the high-impedance state of a tristate output stage, in order to keep the capacitor voltage approximately at half the supply voltage. It is particularly advantageous if timers of the microcontroller can form a pulse width modulator, because then only the pulse width modulation (PWM) ratio must be set for the correction of parameter changes. A central unit is then burdened only slightly with this task. The resistance value in the RC element should preferably be high so that the voltage at the half supply voltage has only a small residual ripple. Due to the filtering effect of the measurement circuit, the effect of residual ripple can be minimized by keeping the signal frequency to be measured by an analog-to-digital converter (ADC) a sufficient distance from the PWM frequency.

In particular, the RC element offers the possibility of achieving a further improvement in the metrological properties of the circuit. The residual ripple on the capacitor can be targeted as a so-called "dither" (English "tremble") can be used to increase the resolution of the measurement. In this case, an auxiliary signal of low amplitude (the residual ripple) is added to a useful signal. By multiple sampling of the sum signal then the resolution in the measurement of the useful signal can be increased. The method is known as dithering. It should be mentioned that the dithering can also be provided otherwise, that is to say without an RC element, for example.

The filter function can be done instead of the simple RC element by other passive low passes.

Preferably, the electronic control unit is configured to initially charge the smoothing capacitor to the at least approximately half supply voltage when switching on, while port pins to which the parallel resonant circuit is connected have the same logic level, and only then to start an excitation of the parallel resonant circuit.

For the benefit of this approach, the following is stated.

The use of inductors causes both the port pins that generate the stimulus and the port pins that serve the measurement, the danger of overvoltages that can destroy a microcontroller. activities to protect the microcontroller, such as a series-connected resistor refer to the steady state. However, surges can occur when the system is turned on, even if the circuit is designed so that there is no danger in the steady state. In order to avoid damage, a switch-on procedure designed for protection is advantageous if the use of protective components should be avoided especially for the switch-on process. It is a process that can be performed by the central unit of the electronic control unit or the microcontroller by program. Once the steady state has been reached, typically no separate measures are necessary and measurement operation can begin.

The switch-on procedure preferably starts with the measuring circuit, during which the stimulus remains switched off. All existing RC elements or passive low-pass filters, which serve to generate a voltage of approximately half the supply voltage, are preferably pre-charged so that they can fulfill their purpose. If the stimulus is activated before reaching approximately half the supply voltage at the respective node, the induced voltages may violate the allowable voltage range. In this context, it is troublesome that the resistance of the RC elements must be high as described above, because this can take a long time to precharge, which causes an undesirable delay between the switching on of the supply voltage and the readiness of the sensor system. This can be avoided by charging the capacitances of the RC elements instead by the inductors and the corresponding port pins, which are also connected. For this purpose, it may be necessary to provide current limiting resistors on these port pins, if there are no output stages or configurations of output stages with which a current limit can be made in the microcontroller, such as switchable integrated pull-up resistors.

As long as the precharging of the RC elements is underway, the two terminals of the electronic control unit or microcontroller providing the stimulus should have the same logic level so that no current flows through the corresponding inductance. The inductance is very low impedance for DC and equals a short circuit that would destroy these terminals. Another advantage is the high-impedance state of tristate outputs. Finally, when the stimulus is started, it should preferably be activated almost simultaneously at both terminals in order to keep the inevitable short circuit at the first level change of the first terminal as short as possible. The program is preferably to be written such that the delay between the activation times of the outputs is deterministic by blocking other activities, e.g. no interrupt is allowed.

A signal generated by the measuring inductance is preferably detected with an analog-to-digital converter of the electronic control unit, wherein its characteristic values, in particular amplitude, phase, real part and / or imaginary part are determined taking into account aliasing, preferably at the excitation frequency.

It is advantageously provided that a frequency component at an evaluation frequency is determined from a signal detected in the electronic control unit, wherein the evaluation frequency is in particular the excitation frequency or an alias of the excitation frequency.

For such statements, the following remarks are given.

If the selected stimulus frequency is so high that the Nyquist frequency of the ADC is lower, even at the maximum possible sampling rate, subsampling can be used purposefully. The signal of the stimulus appears by aliasing at a frequency below the respective Nyquist frequency at the output of the ADC; Aliasing is therefore not a disturbance or an exclusion criterion, but desirable. From a telecommunications point of view, the stimulus is an amplitude-modulated carrier, because if the measured variable changes, an amplitude modulation (AM) of the stimulus with the measured variable occurs. From this, the width of the AM sidebands and the requirement that the sampling rate for a correct function of the system according to the invention should preferably be at least four times as high as the bandwidth in the measured variable, so that the bands do not overlap by aliasing, because the AM signal is twice the bandwidth of the signal of the measurand. Since the signal of the stimulus itself is very narrow - the coil current is typically almost sinusoidal due to the addition of the inductive system to the tuned circuit - and the bands at low frequency measurements can be made slightly narrower than the Nyquist bandwidth of the analog-to-digital converter (ADC) , no separate measures are necessary to prevent unwanted overlapping of bands by aliasing and thus a distortion of the signal.

Since the measuring inductance or, more generally, the inductances at which the voltages to be measured are tapped are typically low-impedance, noise is not a significant factor Only the low ohmic resistance of the inductors is a source of noise. For the same reason, broadband sources of interference in the vicinity of the sensor typically must have very large amplitudes in order to significantly influence the measurement: the low ohmic resistance acts for capacitively or inductively coupled Sources of interference almost like a short circuit, because almost all the noise voltage at the coupling impedance drops. Only narrowband interferers achieve amplitudes that are relevant to the measurement circuit in most applications. Therefore, also narrowband filtering of the voltages to be measured is useful.

Such filtering is typically done digitally by using a series of samples of the ADC to isolate the stimulus frequency in the input-frequency mixture. For this purpose, an implementation of the DFT (Discrete Fourier Transformation) can be used, whose algorithm is executed by a central unit of the electronic control unit or the microcontroller. Particularly preferred is the use of a Goertzel filter or Goertzel algorithm, which can numerically provide the amount and phase of a single spectral line of the DFT particularly efficient. As the frequency of this spectral line is preferably the frequency of the stimulus, ie the excitation frequency, or that alias frequency, which arises by sub-sampling the excitation frequency to choose.

According to one embodiment, the measuring inductance is galvanically coupled to the first inductance. According to a further embodiment, the measuring inductance is magnetically coupled to the first inductor. The measuring inductance can in particular be coupled to the first inductance by forming a measuring transformer. Such embodiments have proven to be advantageous for typical applications.

The first inductance, the measuring inductance and / or the capacitance are preferably components with a respective tolerance of between 1% and 10%, preferably of 1%, or of less than 1%. The tolerance typically specifies the design-related maximum deviation from a nominal value. Correspondingly small tolerances in the present case facilitate the conception of the device, in particular in the context of mass production, since with a low tolerance the maximum deviation of the resonant frequency of the parallel resonant circuit is correspondingly low.

• – der Parallelschwingkreis eine maximale Güte aufweist, welche durch Maximierung eines Werts von Vt·Vt/V0 erhalten wird,
• – wobei Vt ein Verhältnis aus Spulenstrom und Zuleitungsstrom bei maximaler Abweichung der Kapazität und der ersten Induktivität von ihren jeweiligen Werten bei der Resonanzfrequenz des Parallelschwingkreises bezeichnet, und
• – wobei V0 ein Verhältnis aus Spulenstrom und Zuleitungsstrom bei jeweiligen Werten von Kapazität und erster Induktivität bei der Resonanzfrequenz des Parallelschwingkreises bezeichnet.

According to a preferred embodiment it is provided that

• - the parallel resonant circuit has a maximum quality, which is obtained by maximizing a value of Vt · Vt / V0,
• Wherein Vt denotes a ratio of coil current and supply current at maximum deviation of the capacitance and the first inductance from their respective values at the resonant frequency of the parallel resonant circuit, and
• - where V0 denotes a ratio of coil current and supply current at respective values of capacitance and first inductance at the resonant frequency of the parallel resonant circuit.

The maximum quality can be limited in particular by interconnecting a resistor in the parallel resonant circuit.

The maximum deviation is typically specified by the tolerance, that is to say the maximum deviation from the nominal value already mentioned above, of the respective component.

For quality and the aforementioned advantageous embodiment, some embodiments are given below.

Passive components of electrical engineering, such as coils and capacitors, are usually offered with tolerances of their characteristic values, which are typically between 1% and 10%. Even lower values than 1% usually lead to very high costs for the components. On the other hand, tolerances of more than 10% make it difficult to interpret circuits in a meaningful way. If a resonant circuit is constructed from components subject to tolerances with the characteristic values L and C (capacitance), the oscillation formula for the resonant frequency ω ₀ = 1 / √ (LC) shows that the tolerance of the resonant frequency ω ₀ corresponds to the geometric mean √ (LC) corresponds to the tolerances of the components. The tolerance of the resonant frequency of resonant circuits, which are constructed with the usual components, therefore also moves in the range of 1% to 10%. If a resonant circuit is designed so that its nominal resonant frequency coincides with the stimulus frequency, ie the excitation frequency, then it must be assumed that the real resonant frequency deviates by ± 1% to ± 10%.

An important feature of resonant circuits is their quality Q. It is a measure of the losses that occur due to the ohmic resistance in the circuit and for the decay of a free vibration. The higher the quality, the lower the losses and the slower the decay.

1 shows a family of curves related to the effect of the quality on the benefits of the addition of the inductive system to the resonant circuit on the stimulus, so the excitation by the excitation circuit shows. Plotted is the ratio of the current I _L through the excited inductance to the current I _{St of} the stimulus. Without supplementing the inductive system with the resonant circuit, this ratio would always be one, because there is no further current path. Since the current I _L corresponds to the useful current because the magnetic flux is generated by him, while the current I _St corresponds to the effort, the applied quotient means an evaluation factor for the benefit. Values greater than one correspond to a profit. This benefit is plotted over the normalized to the resonant circuit frequency ω ₀ angular frequency ω. The coulter parameter is the quality; the value at ω ₀ in each curve corresponds to the value of the quality, ie the curves represent the quality Q with the values 8; 6; 4; 2; 1; 1 / √2 (top to bottom).

Considering the benefits of supplementing the inductive system with the resonant circuit at frequencies that deviate by ± 10% from the resonant circuit frequency ω ₀ , for example, it can be seen that a high benefit can also be achieved for these frequencies, not only at the nominal frequency , However, the higher the quality, the higher the ratio of the benefit at the resonance frequency ω ₀ and the utility at a frequency deviating by ± 10%. Reason is the well-known connection that the bandwidth of the resonant peak becomes increasingly narrower as the quality increases. For the interpretation of the device described herein, there is therefore the following conflict of objectives: On the one hand, the quality should be as high as possible for the highest possible quality-related benefit; On the other hand, a high ratio of utility in the center (resonance) and at the margins of the tolerance causes a reduction in usability, as follows: Any conventional measuring system, whether analog or digital, has a maximum input amplitude, which typically must not be exceeded. Within this maximum, the relative resolution to the input signal is proportional because the smallest distinguishable step of the input signal is constant in the input signal range. If the input signal range is fully utilized at resonance, then it is off 1 resulting benefits also fully usable. However, at the margins of the tolerance range, the benefit is reduced by the factor that results from the ratio of the utility in the center (resonance) and at the margins of the tolerance range.

The goal is an optimal design of the sensor system within the scope of the design. First, it is desirable to keep the tolerance of the component characteristics L and C, or their geometric mean √ (LC), as low as possible. It is helpful that in recent years capacitors with tolerances of 1% and better available and cheaper have become. The main importance for the optimization but comes to the optimal choice of quality. In general, high quality values can typically be used advantageously only in combination with small tolerances of the term √ (LC). The design methodology described below is therefore based on the idea of establishing a tolerance for √ (LC) for the respective application due to the availability of suitable, tightly tolerated capacitors and the design of an inductive system, from which the optimum quality can then be calculated.

With the tolerance of √ (LC), the tolerance of the angular frequency ω is also determined by the oscillation formula. Out 1 Now the value for I _L / I _St can be read for ω ₀ (resonance) and ω _T (angular frequency given by the tolerance, the lower frequency limit can be selected, the upper one or both). This can also be done numerically; the calculation of 1 From the supply current I _St and the coil current I _L for a parallel resonant circuit given quality is elementary electrical engineering and is assumed here. Characterized the sizes are V0 = I _L (ω ₀₎ / I _St (ω ₀₎ and Vt = I _L (ω _T) / I _St (ω _T) now known. For a frequency ω _{T the} following applies: The fraction with which the input voltage range of the measuring system is exhausted is given by the term Vt / V0. At this frequency, the benefit obtained by adding the inductive signal to the resonant circuit is equal to Vt. Thus, the factor is benefited by the whole system by the addition of the inductive signal to the resonant circuit, the product of two terms: V t ² / V0. By elemental transformations of the underlying equations this factor can be calculated as a function of ω ₀ , ω _T and Q. The optimization of a system then consists in maximizing Vt ² / V0, yielding a result for Q as a function of ω ₀ and ω _T results. By further elementary mathematical transformations, it is also possible to apply Q as a function of ω ₀ and ω _T. From such a representation Q can be read directly, thus saving the repeated insertion of values for Q into the calculation for maximum search. In both cases, the results for Q are always maximum values, ie for a correspondingly designed system with ω ₀ and ω _T , Q should never exceed the calculated values, because otherwise the underlying input signal range of the measuring system is violated. If the calculated value of Q is not exceeded, no violation occurs; the input signal range of the measuring system is simply not exhausted. This is important for the design of the resonant circuit, since a too high Q quality can always be reduced in a simple manner by adding an ohmic resistance; However, it may be difficult to constructively increase a low Q given by the properties of the inductive system. The reason is the parasitic nature of the loss resistance in inductive components, which is particularly in Sensor inductance is often high because its windings are expanded and the magnetic circuits are open.

• 1. Auslegung des induktiven Systems (weitgehend außerhalb des Themas dieser Anmeldung)
• 2. Festlegung der Arbeitsfrequenz des induktiven Systems (weitgehend außerhalb des Themas dieser Anmeldung, da von der Auslegung bestimmt)
• 3. Auswahl eines Kondensators mit möglichst geringer Toleranz (dies kann insbesondere nach kommerziellen Kriterien erfolgen)
• 4. Berechnung des geometrischen Mittels der Toleranz der Bauelementwerte L und C
• 5. Bestimmung der Frequenz ω_T aus (4.)
• 6. Wiederholte Berechnung und Maximierung von Vt²/V0; Ergebnis ist die maximal zulässige Güte Q mit den gewählten Parametern
• 7. Ist die Güte des Schwingkreises aus L, C und seinem konstruktiv bedingten Verlustwiderstand kleiner als die maximal zulässige Güte Q? Wenn ja, fertig.
• 8. Wenn nein, Einfügen eines Widerstandes in den Schwingkreis, der die Güte auf den maximal zulässigen Wert herabsetzt.

The overall result is the following advantageous procedure or the following method, summarized from the above explanations:

• 1. Interpretation of the inductive system (largely outside the scope of this application)
• 2. Determination of the working frequency of the inductive system (largely outside the scope of this application, as determined by the interpretation)
• 3. Selection of a capacitor with the lowest possible tolerance (this can be done in particular according to commercial criteria)
• 4. Calculation of the geometric mean of the tolerance of the component values L and C
• 5. Determination of the frequency ω _T (4.)
• 6. Repeated calculation and maximization of Vt ² / V0; The result is the maximum permissible quality Q with the selected parameters
• 7. Is the quality of the resonant circuit of L, C and its design-related loss resistance smaller than the maximum allowable Q? If yes, done.
• 8. If no, inserting a resistor into the resonant circuit, which reduces the quality to the maximum permissible value.

It should be understood that the method steps just described as a whole or in any sub-combination can represent an independent invention aspect.

From these statements, in particular, the above-described procedure for determining the quality is derived.

The measured quantity can be sensed, for example, by changing a position of a magnetic core in the measuring inductance. This can change the value of the inductance of the measuring inductance.

The measured variable can also be sensed by changing a distance between the measuring inductance and the first inductance. This typically corresponds to a design as a measuring transformer, wherein the magnetic coupling is changed.

In addition, the measurand may be sensed, for example, by varying a position of a magnetic and conductive, a nonmagnetic and conductive, or a magnetic and nonconductive element adjacent to the sensing inductance and the first inductance. This allows adaptation to different tasks and circumstances. The element may be arranged, for example, between or next to the first inductance and the measuring inductance. By changing its position, the coupling between the first inductance and the measuring inductance is changed.

A magnetic and conductive element may be made of steel, for example. A non-magnetic and conductive element may be made of aluminum, for example. A magnetic and non-conductive element may be made of ferrite, for example.

The measuring inductance can in particular be designed to sense a measured variable in the form of a position, a length, an angle, a force, a pressure and / or a torque. This corresponds to typical usage scenarios.

• – Selbstinduktivität bzw. Induktivität,
• – Verlustwiderstand,
• – komplexe Impedanz,
• – Verlustwinkel,
• – Gegeninduktivität zur ersten Induktivität.

The electronic control unit is preferably designed to measure one or more of the following characteristic values over the measuring inductance:

• - self-inductance or inductance,
• - loss resistance,
• - complex impedance,
• - loss angle,
• - Mutual inductance to the first inductance.

From such characteristics can typically be concluded on the measured variable.

According to a development it is provided that the device has two, three or more than three measuring inductances. These may for example be designed individually or together movable. In particular, they can be influenced by the same measured variable, so that a compensation of disturbance variables, in particular the temperature, is possible.

In particular, the device may have a first measuring inductance and a second measuring inductance, wherein the first measuring inductance is arranged on a first longitudinal end of the first inductance and the second measuring inductance is arranged on a second longitudinal end of the first inductance. Such an embodiment has proven to be advantageous for typical applications.

The device according to the invention may comprise, for example, an oscillator, an optional amplifier, an inductive system, a resonant circuit capacitor, an optional measuring amplifier and a measuring system, in particular the measuring device. The amplifier amplifies the signal of the oscillator, it can be omitted in particular if the output of the oscillator already provides a sufficiently strong signal. This signal or that of the amplifier form the stimulus for the inductive system, which is supplemented by the capacitor to the resonant circuit. In particular, the above-mentioned first inductance and the already mentioned measuring inductance can be regarded as an inductive system. If the output signal is too weak for the measuring system, an additional measuring amplifier can be used Otherwise, in particular, the measuring system can be connected directly to the inductive system, for example by tapping or becoming one or more electrical variables on the inductive system. The measuring system can be embodied, for example, as a lock-in amplifier (or phase-sensitive rectifier, synchronous demodulator or carrier frequency amplifier) implemented in analog or digital technology and possibly software and can determine one or more characteristic values of the inductive system with the aid of the oscillator signal as the reference signal.

Preferably, the excitation frequency is adjustable. In particular, it may be adjustable by software. This allows a modulation of the excitation frequency or the stimulus. Alternatively, the excitation frequency can also be fixed.

The electronic control unit may be configured or configured in particular as a lock-in amplifier for the measurement described herein. This can also be referred to as a phase-sensitive rectifier, synchronous demodulator or carrier frequency amplifier and has proved to be advantageous for the application relevant here.

The described design of an inductive system and a combination of the components described herein achieves that the elevation of the measurement-relevant magnitudes, as is characteristic of resonance, in conjunction with a signal processing path known from the lock-in amplifier is, can be connected. As a result, the power requirement of an inductive sensor system is reduced and / or its measurement resolution is increased. For the practical design of inductive sensor systems with open magnetic circuits, typical values in the range of 3 to 8 can advantageously be used or expected for Vt ² / V0.

Further features and advantages will be apparent to those skilled in the art from the embodiments described below with reference to the accompanying drawings.

1 shows exemplary courses of a quality, which has already been discussed earlier in this application.

2 shows a device according to a first embodiment of the invention.

3 shows a device according to a second embodiment of the invention.

1 was mentioned and explained earlier in the text.

2 shows a device according to a first embodiment of the invention with an electronic control unit in the form of a microcontroller MK. An inductive system is designed as a measuring transformer from the first inductance LP as well as three measuring inductances, namely a first measuring inductance LS1, a second measuring inductance LS2 and a third measuring inductance LS3. A stimulus is provided by port pins P1 and P2 of the microcontroller MK, which are driven by internal timers with an intended excitation frequency or stimulus frequency.

Parallel to the first Induktiviät LP a capacitor CP is connected, resulting in a parallel resonant circuit P. The oscillations of the resonant circuit P are continuously maintained by alternating voltages at P1 and P2, so that a vibration of constant amplitude results.

The current through the first inductance LP causes a magnetic flux, which also detects the measuring inductances LS1, LS2, LS3. The magnetic coupling is symbolized by the arrow. The percentage of the magnetic flux of the first inductance LP, which also passes through the measuring inductances LS1, LS2 and LS3, depends on the measured variable. In the present embodiment, the distance from the first inductance LP and the measuring inductances LS1, LS2 and LS3 may change to measure this.

Alternatively, however, it would also be possible, for example, to have a soft magnetic core or an electrically conductive body between the first inductance LP and the measuring inductances LS1, LS2 and LS3 whose movement is measured. Any arrangement that causes a parameter change of the inductive system is basically possible.

Basically, the magnetic flux in the measuring inductances LS1, LS2 and LS3 induces voltages which are fed to port pins P4, P5, P6 of the microcontroller MK. An advantage of using three inductors LS1, LS2 and LS3 instead of a single one is the ability to obtain more data to measure the measurand. Values of the measuring inductances LS1, LS2 and LS3 therefore depend on the measured variable differently, so that any measuring errors at one of the measuring inductances LS1, LS2, LS3 can be corrected by taking into account the measured values at all measuring inductances LS1, LS2 and LS3. In particular, a compensation of the typically most important disturbance variable, namely the temperature, can be undertaken. Alternatively, however, several measured variables can also be detected with one system.

The voltages at the port pins P4, P5, P6 are alternately supplied via an integrated multiplexer to an analog-to-digital converter (ADC) in the microcontroller MK. The ADC records each of these voltages several times. From the plurality of samples obtained by the ADC for each voltage, the amplitude and phase at the stimulus frequency are determined. Further steps of digital signal processing and software can be made to finally determine the measured value for the measured variable from the amplitude and phase values.

As shown, the sensing inductors LS1, LS2, LS3 are connected at their respective poles, which face the poles connected to the port pins P4, P5, P6, to a common potential defined by a smoothing capacitor C1. The smoothing capacitor C1 is connected at its pole, which defines this potential, via a resistor R1 to a port pin P3 of the microcontroller MK. During operation, the microcontroller MK applies a pulse-width-modulated signal to this port pin P3 in order to maintain an average potential on the smoothing capacitor C1. This potential corresponds approximately to half the supply voltage of the microcontroller MK.

It should be explicitly noted that apart from the already mentioned components microcontroller MK, first inductance LP, measuring inductances LS1, LS2, LS3, capacitance CP, smoothing capacitor C1 and resistor R1, there are no further components in the device according to the embodiment shown. In particular, all passive components are connected directly to the microcontroller MK. The provision of additional active components such as amplifiers is thus dispensed with, which minimizes costs.

3 shows a device according to a second embodiment of the invention. Also, a microcontroller MK is provided. The differences in wiring compared to the device according to the first embodiment will be explained below.

In the device according to the second embodiment, all inductors are galvanically coupled. A parallel resonant circuit P is formed by a first inductance L1, a measuring inductance L2 and a capacitance CP. The measuring inductance L2 is sensitive to the measured variable and changes, while the first inductance L1 has a constant value. In the connecting lines to the port pins P1, P2 are resistors R2 and R3, which protect the port pins P1, P2 from excessive currents. Between the first inductance L1 and the measuring inductance L2, a voltage is tapped, which oscillates with the stimulus frequency and whose amplitude and phase depend on the measured variable. Regarding further aspects be on 2 directed.

The claims belonging to the application do not constitute a waiver of the achievement of further protection.

If, in the course of the procedure, it turns out that a feature or a group of features is not absolutely necessary, it is already desired on the applicant side to formulate at least one independent claim which no longer has the feature or the group of features. This may, for example, be a subcombination of a claim present at the filing date or a subcombination of a claim limited by further features of a claim present at the filing date. Such newly formulated claims or feature combinations are to be understood as covered by the disclosure of this application.

It should also be noted that embodiments, features and variants of the invention, which are described in the various embodiments or embodiments and / or shown in the figures, can be combined with each other as desired. Single or multiple features are arbitrarily interchangeable. Resulting combinations of features are to be understood as covered by the disclosure of this application.

Recoveries in dependent claims are not to be understood as a waiver of obtaining independent, objective protection for the features of the dependent claims. These features can also be combined as desired with other features.

Features that are disclosed only in the specification or features that are disclosed in the specification or in a claim only in conjunction with other features may, in principle, be of independent significance to the invention. They can therefore also be included individually in claims to distinguish them from the prior art.

Claims (15)

Device for measuring a measured variable, comprising - a first inductance (LP, L1), - a measuring inductance (LS1, LS2, LS3, L2), which is designed to sense the measured variable and which is coupled to the first inductance (LP, L1) , - a capacitor (CP), which is connected to the first inductor (LP, L1) to a parallel resonant circuit (P), and - An electronic control unit (MK), - wherein the electronic control unit (MK) is connected directly to the parallel resonant circuit (P) and configured to oscillate the parallel resonant circuit (P) with an excitation frequency, which is derived from a clock of the electronic control unit , and - wherein the electronic control unit (MK) is connected directly to the measuring inductance (LS1, LS2, LS3, L2) and is configured to measure a value indicating the measured variable at the measuring inductance (LS1, LS2, LS3, L2). Device according to claim 1, characterized in that - the parallel resonant circuit (P) is connected to a first port pin and to a second port pin of the electronic control unit (MK), the second port pin being supplied with a clock inverted to the first port pin. Apparatus according to claim 2, characterized in that - the parallel resonant circuit (P) - is connected parallel to the first port pin with a number of other first port pins, which are switched synchronously to the first port pin, and / or - parallel to the second port pin with a number of further second Portpins is connected, which are switched synchronously to the second port pin. Device according to one of the preceding claims, characterized in that - Portpins, to which the parallel resonant circuit (P) is connected, have a push-pull output stage or a tristate output stage. Device according to one of the preceding claims, characterized in that - the excitation frequency differs by a maximum of 25%, preferably at most 20%, more preferably at most 15%, even more preferably at most 10% of a resonant frequency of the parallel resonant circuit (P). Device according to one of the preceding claims, characterized in that - the measuring inductance (LS1, LS2, LS3, L2) is connected with a first pole to a port pin of the electronic control unit (MK), and is connected to a second pole with a potential, which corresponds at least approximately to half the supply voltage of the electronic control unit (MK). Apparatus according to claim 6, characterized in that - the potential which corresponds at least approximately to half the supply voltage of the electronic control unit (MK) is generated by means of a voltage divider. Device according to one of claims 6 or 7, characterized in that - the potential which corresponds at least approximately to half the supply voltage of the electronic control unit (MK), by means of a smoothing capacitor (C1), - wherein the smoothing capacitor (C1) with a first Pol is connected to a reference potential or to the supply voltage, - and wherein a second pole of the smoothing capacitor (C1) via a resistor (R1) to a port pin of the electronic control unit (MK) is connected, which with a pulsed signal, in particular with a signal with specifiable duty cycle, is applied. Apparatus according to claim 8, characterized in that - the electronic control unit (MK) is configured to initially charge the smoothing capacitor to the at least approximately half supply voltage when turning on, while port pins to which the parallel resonant circuit (P) is connected, have the same logic level , and only then to start an excitation of the parallel resonant circuit (P). Device according to one of the preceding claims, characterized in that - one of the measuring inductance (LS1, LS2, LS3, L2) generated signal with an analog-to-digital converter of the electronic control unit (MK) and taking into account aliasing of its characteristics, in particular Amplitude, phase, real part and / or imaginary part are determined, preferably at the excitation frequency. Device according to one of the preceding claims, characterized in that - from a in the electronic control unit (MK) detected signal, a frequency component is determined at an evaluation frequency, - wherein the evaluation frequency is in particular the excitation frequency or an alias of the excitation frequency. Device according to one of the preceding claims, characterized in that - The measuring inductance (LS1, LS2, LS3, L2) is galvanically or magnetically coupled to the first inductance (LP, L1). Device according to one of the preceding claims, characterized in that - the parallel resonant circuit (P) has a maximum quality, which is obtained by maximizing a value of Vt · Vt / V0, - where Vt is a ratio of coil current and supply current with maximum deviation of the capacitance (CP) and the first inductance (LP, L1) of their respective values at the resonant frequency of the parallel resonant circuit (P), and - where V0 is a ratio of coil current and supply current at respective values of capacitance (CP) and first inductance (LP, L1) at the resonant frequency of the parallel resonant circuit (P). Device according to one of the preceding claims, characterized in that the measured variable - by changing a position of a magnetic core in the measuring inductance (LS1, LS2, LS3, L2), - by changing a distance between the measuring inductance (LS1, LS2, LS3, L2) and first inductance (LP, L1), and / or - by changing a position of a - magnetic and conductive, - non-magnetic and conductive, or - magnetic and non-conductive element adjacent to the measuring inductance (LS1, LS2, LS3, L2) and the first Inductance (LP, L1) is sensed. Device according to one of the preceding claims, characterized in that - the device has two, three or more than three measuring inductances (LS1, LS2, LS3, L2).

Images