The invention relates to a circuit for measuring distances, of the general type disclosed in DE 42 25 968 A1 and corresponding U.S. Pat. No. 5,629,619. These prior patents disclose a circuit which is used to measure distances in a noncontacting manner by means of a measuring coil. The circuit is an inductive distance sensor, which is operated with relatively low frequencies in a range from 1 kHz to 10 kHz. To increase the accuracy of the measurement, it is necessary to determine the temperature influence on the measured value. In this known circuit, this occurs by means of a discrete circuit, which determines the dc voltage component with a circuit that is excited by ac voltages. The circuit has two inputs, which are activated in phase opposition by two input signals that are generated by a signal source. The operational amplifiers subsequent to the inputs operate with their resistors as voltage/current transformers, with the current being coupled into the measuring coil from both sides.
In the normal measuring operation, the circuit is activated with two ac voltages in phase opposition. To determine the temperature behavior, a dc offset voltage signal, i.e. dc component, is superposed upon the ac voltages. Since the circuit requires that the supplied current be equal at both ends of the measuring coil, different voltages adjust via the resistors, which are associated with the measuring coil, and additionally with respectively one operational amplifier circuit. These different voltages are caused by the offset of the ac voltages as well as the temperature-dependent components. The temperature-dependent output voltage, i.e. dc voltage, is determined with a further operational amplifier.
If one applies the superposition principle, one will see that the lower input of the circuit exhibits a low-pass behavior, and the upper input of the circuit a band-pass behavior. For ideal input signals in phase opposition, the total transfer function is thus a low-pass function, which is smoothed by a further capacitance. The low pass develops by the difference of a high-pass and an amplification path that is matched with it.
Since the offset in a first approximation is inversely proportionate to the temperature
it is thus possible to determine the temperature, and to correct the temperature-caused measuring errors. However, these measurements are only seldom inserted into the normal measurements with the mere ac voltage input signals. During the measurement with the mere ac voltage input signals, it is also possible to determine a dc voltage component, which is used to determine and correct the temperature drift of the measuring coil.
The known circuit is problematic in particular to the extent that the occurring time constants of the filtration are very great, and that because of its setup, the known circuit is made relatively large and, therefore, unsuitable for an application in which only a very small space is available for the circuit.
It is therefore an object of the present invention to describe both a circuit and a method for measuring distances of the initially described kind, which enable a use of the circuit also when the space available for the circuit is small.
In accordance with the invention, the foregoing object is accomplished by a circuit for measuring distances, which comprises at least two inputs, at least one measuring coil, and at least one signal source for generating at least two input signals. The input signals are applied to the inputs of the measuring coil. The input signals are also applied to at least one, preferably timed SC network, and used for generating a measuring signal and/or an output signal that is dependent on a temperature influence.
Furthermore, with respect to a method for measuring distances, the foregoing object is accomplished by a method which utilizes the circuit as described above and wherein the input signals are applied to at least one, preferably timed SC network, and used for generating a measuring signal and/or an output signal that is dependent on the temperature influence.
In accordance with the invention, it has been found that departing from the practice of the art, it is necessary to achieve a miniaturization of the circuits of the prior art for purposes of enabling a large number of possible uses of the circuit. This is accomplished in that the circuit is configured such that it can be set up as an integrated circuit, and that integration into an ASIC is made possible with a comparable transfer function of the known circuit. This is realized by the use of an SC network, i.e. a switched capacitor network, that exhibits an excellent matching behavior, and permits a particularly simple integration and with that a miniaturization of the circuit, thereby making the circuit applicable even in a very small space and thus universally applicable, while keeping the price of the circuit relatively low.
In this case, the high-pass circuit is replaced with an equivalent, passive double-resistive reference network, which comprises a voltage divider and an inductance that is parallel connected to one of the resistors of the voltage divider. The transfer function of this filter is that of a first-degree high-pass
The transfer function includes one zero point at p=0 and a pole at p=−R1R2/(R1+R2)1/L, with p being the general complex frequency variable.
By means of known methods, it is now possible to prepare a corresponding wave flow diagram. A three-gate parallel adapter is used to adapt to one another the different wave impedances of the three components of the reference network. The wave flow diagram of a resistive voltage source is on the left side of the three-gate parallel adapter, the wave flow diagram of the inductance is in the center above, and the wave flow diagram of the terminating resistance is on the right side. Since wave filters are time-discrete, one will define instead of the complex frequency variable p, a new frequency variable ψ with
where T=1/F is the sampling period, and F the sampling frequency. For purely imaginary frequencies p becomes jω and thus
In this case, the adapter equations can be set up as follows:
b
2
=b
3
−a
2 (6)
with incident voltages waves ai and emergent voltage waves bi, where i=1, 2, 3. The output voltage results from
In addition, an inversion of the signal in the wave flow diagram of the inductance is to be realized
b2*=−b2 (8)
If parasitic currents are not to be admitted, it will be possible to amplify positively delayed or negatively undelayed, or to integrate. With this technique, different ways of realizing switched-capacitor wave filters, i.e. SC filters, are known.
Within the scope of a very excellent determination of the temperature influence, it is possible that at least two input signals are essentially unipolar and/or opposite in phase. Preferably, the input signals are essentially square-wave voltages, since they permit generating input signals in phase opposition in a particularly simple manner.
To reduce low frequencies in the input signals, the input signals could be adapted for filtering by means of at least one filter in a proportionate way, and/or by means of a high-pass.
With respect to a particularly simple configuration, the SC network could comprise at least one SC amplifier. This would permit a particularly simple setup of the circuit.
In a further advantageous configuration, it would be possible to realize a first SC amplifier as a positive delayed SC amplifier, and/or to multiply the two inputs respectively with one factor. In a realization of the SC amplifier as a positive delayed SC amplifier, this would reduce parasitic currents.
To further avoid parasitic currents, a second SC amplifier could be realized as a positive delayed SC amplifier, and/or delay at least one of the input signals, preferably unamplified, by one half period of the clock frequency.
The SC network could comprise at least one SC integrator. With respect to avoiding parasitic currents, the SC integrator could be realized as a negative undelayed integrator, and/or have an amplification of 1, and/or be lossy.
In a particularly simple manner, the output of the SC integrator could be applied to a second input of the first SC amplifier.
With respect to a particularly simple determination of the temperature influence, an SC adder could be used to add the outputs of the first SC amplifier and the second SC amplifier. This would permit tapping at the output of the SC adder, a temperature-dependent output signal, which could be used for compensating the temperature influence.
The output of the first SC amplifier could be applied to the inputs of the SC integrator and/or the SC adder. In addition or as an alternative, the output of the second SC amplifier could be applied to a second input of the SC adder.
Furthermore, the SC network could also comprise at least one SC amplifier, and/or at least one SC integrator, and/or at least one SC adder.
To reduce parasitic currents, it would be possible to realize the first and/or the second SC amplifier and/or the SC adder negatively undelayed. In addition or as an alternative, the SC integrator could be realized positively delayed. This would permit inverting the output signal.
Within the scope of an advantageous configuration, the SC network could comprise at least one SC amplifier, and/or at least one SC integrator, and/or at least one SC difference amplifier.
In a further advantageous manner, it would be possible to store in the SC integrator at least one of the input signals preferably multiplied by a factor. A further factor could be again erasable by a capacitance of the SC integrator respectively from the result in each sampling period.
In an advantageous manner, an SC amplifier could be realized as a positive delayed SC amplifier, and/or delay at least one of the input signals unamplified and/or by one half period of the clock frequency.
To determine the temperature influence, the output of the SC amplifier and of the SC integrator could be adapted for subtraction by means of an SC difference amplifier, and/or for delay by one half period of the clock frequency.
In a particularly simple manner, the output of the SC amplifier could be applied to the second input of the SC integrator. With that, the output signal could have a delay of one clock period.
The method of the invention could be used in particular for operating a circuit as described above. The method is advantageous in that a circuit operated by this method is simple to integrate because of its excellent matching behavior.
There exist various possibilities of improving and further developing the teaching of the present invention in an advantageous manner. To this end, one may refer on the one hand to the claims of this application and on the other hand to the following description with reference to the drawing of preferred embodiments of the circuit according to the invention and the method of the invention for measuring distances. In conjunction with the detailed description of the circuit and the method of the invention with reference to the drawing, also generally preferred improvements and further developments of the teaching are described in greater detail.
In the drawings:
The known circuit for measuring distances is provided as a discrete circuit and comprises two inputs 1, 2, as well as a measuring coil 3. A signal source (not shown) permits generating two input signals epos and eneg. The inputs 1, 2 are activated by means of the input signals epos and eneg. The input signals epos and eneg are applied in a preprocessed form to the inputs 1, 2 of the measuring coil 3.
The known circuit that is excited by ac voltage makes it possible to determine by measuring a dc component, which is proportional to the temperature. Operational amplifiers that follow the inputs 1, 2 form with their resistors a voltage/current transformer. The current is coupled into the measuring coil 3 from both sides. In the normal operation, i.e., measuring operation, the ac signals epos and eneg shown
As is very clearly shown in
If one applies the superposition principle, one will find that an operational amplifier 5 associated with the input 1, exhibits a low-pass behavior in connection with resistors R2 and R3 and capacitance C2, and that an operational amplifier 6 associated with the input 2 shows a band-pass behavior in connection with resistors R1, R3 and capacitances C1 and C2. For input signals in an ideal phase opposition, the total transfer function is a low-pass function, which is smoothed by capacitance C2. The low-pass forms from the difference between a high-pass and an amplification path that is matched with it.
Since in a first approximation, the offset is dependent on the temperature in an inversely proportionate way
it is thus possible to determine the temperature and to correct temperature-necessitated effects.
with a zero point at p=0 and a pole at p=−R1, R2/(R1+R2)1/L, with p again being the complex frequency variable.
It is now possible to set up a corresponding wave flow diagram, as shown in
where T=1/F is the sampling period and F the sampling frequency. For purely imaginary frequencies, p becomes jω, and with that ψ becomes
In this case, the adapter equations that are to be computed, can be set up as follows:
b
2
=b
3
−a
2 (14)
with the incident voltage waves ai and the emergent voltage waves bi, where i=1, 2, 3. The output voltage results from
In addition, the inversion of the signal is to be realized in the wave flow diagram of the inductance
b2*=−b2 (16)
With this technique, different kinds of realization of switched-capacitor wave filters, or SC filters, are described.
Since according to equation 15, the output voltage results as an arithmetic mean value from the incident and the reflected voltage wave, the 0 Db level is reached at the output, since the division with the factor Two is not performed. The two coefficients of equation 13 are reproduced as capacitance ratios in the signal paths. An undelayed negative SC integrator 9 forms the realization of both equation 13 and equation 15. By coupling the output signal of the SC integrator 9 into the SC amplifier 8 in phase Φ, the feedback loop is closed. The transfer function of this high-pass circuit is shown in
A circuit of the present invention for measuring distances is shown in
A part of the circuit corresponds to the high-pass circuit of
The input signal eneg is shifted toward the output with a positive delay by half a clock period. Furthermore, the SC network comprises an SC adder 11, which is used to add the two input signals, i.e. the output signals of the SC amplifiers 8 and 10. The SC network is a positive, delaying SC circuit, which has a total delay of one clock period. Should this be too high, it would also be possible to realize the input amplifier and the output amplifier negatively undelayed. It will then be necessary to realize the SC integrator for a positive delay. In this instance the output signal is inverted.
For inverting the signal, b2* can be derived as follows:
b2*=a2−γ2a2−γ1a1=a2(1−γ2)−γ1a1 (17)
The output voltage of the high-pass is the voltage via the resistor R2, or the voltage via the inductance L, since both elements are parallel connected. Contrary to the resistor R2, at which the incident wave always equals zero, the voltage of the inductance is defined by
Accordingly, the output voltage results as difference of the incident wave a2 and the negative reflected wave b2 divided by 2. Without the division, one obtains again a maximum level of 0 dB. For this reason, the signal can again be added to the input signal eneg.
A circuit that realizes this, is shown in
Since for the described application, the limit frequency of the high-pass is always clearly in a frequency range from 0 to one fourth of the sampling frequency, γ2 is always smaller than 1. Since the output value at the SC integrator 13 should always be positive, one multiplies in the circuit the input signal epos positively delayed, with γ1. The output voltage of the high-pass is generated by means of the SC difference amplifier 14. The input capacitance for the lossy integrator 13 is used to generate the difference a2−b2. To this end, the difference amplifier 14 is initialized with Φ, and the output signal U is thus inverted. As previously described with reference to
The transfer function of the circuit of
As regards further details, the general description is herewith incorporated by reference for purposes of avoiding repetitions.
Finally, it should be expressly remarked that the above-described embodiments are used only to explain the claimed teaching, without however limiting it to these embodiments.
Number | Date | Country | Kind |
---|---|---|---|
101 46 287 | Sep 2001 | DE | national |
The present application is a continuation of international application PCT/DE02/03507, filed Sep. 19, 2002, and which designates the U.S. The disclosure of the referenced application is incorporated herein by reference.
Number | Name | Date | Kind |
---|---|---|---|
4731591 | Weigand | Mar 1988 | A |
5148166 | Ribner | Sep 1992 | A |
5223830 | Romes | Jun 1993 | A |
5281860 | Krenik et al. | Jan 1994 | A |
5589778 | Ono et al. | Dec 1996 | A |
5629619 | Mednikov | May 1997 | A |
5712563 | Kawagoe et al. | Jan 1998 | A |
5729163 | McCleary et al. | Mar 1998 | A |
6073043 | Schneider | Jun 2000 | A |
6462536 | Mednikov et al. | Oct 2002 | B1 |
6639414 | Lien | Oct 2003 | B1 |
Number | Date | Country |
---|---|---|
42 25 968 | Feb 1994 | DE |
Number | Date | Country | |
---|---|---|---|
20040201376 A1 | Oct 2004 | US |
Number | Date | Country | |
---|---|---|---|
Parent | PCT/DE02/03507 | Sep 2002 | US |
Child | 10803298 | US |