Information
-
Patent Grant
-
6298709
-
Patent Number
6,298,709
-
Date Filed
Friday, June 2, 200024 years ago
-
Date Issued
Tuesday, October 9, 200123 years ago
-
Inventors
-
Original Assignees
-
Examiners
Agents
-
CPC
-
US Classifications
Field of Search
US
- 073 137
- 073 138
- 073 102
- 073 108
- 073 116
- 073 157
-
International Classifications
-
Abstract
A device for the determination of a measured variable, a motion variable in particular, is described. At its output, the device signals a malfunction via a fixed voltage level at the output. The device includes a self-test module that functions concurrently with a measurement operation without external triggering and interrupting or interfering with the actual measurement.
Description
FIELD OF THE INVENTION
The present invention relates to a sensor device.
BACKGROUND OF THE INVENTION
A sensor device with a testing module that checks the sensor device for a possible malfunction is described in German Patent No. 44 47. This testing module must be activated externally by actuating a test input. In doing so, the sensor is detuned in the testing phase by an interference signal. The interference signal corresponds to a specific value of a measured variable to be tested.
SUMMARY OF THE INVENTION
In contrast, the device according to the present invention has the advantage that it represents an inherently reliable sensor that continuously carries out a self-test and accordingly identifies malfunctions in the mechanical and/or electronic systems without detuning the sensor and thereby influencing the measurement function. At the same time, no external triggering and no external additional circuitry is necessary; a three-pin terminal connection (voltage supply+signal output) can be implemented. As a result, there exists downward compatibility with already existing sensors that do not have a self-test module and have such a connection. These existing sensors can be advantageously replaced by the device according to the invention.
BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1
shows a first embodiment according to the present invention.
FIG. 2
shows frequency relationships of the first embodiment shown in FIG.
1
.
FIG. 3
shows a second embodiment according to the present invention.
FIG. 4
shows frequency relationships of the second embodiment shown in FIG.
3
.
DETAILED DESCRIPTION
FIG. 1
shows a measurement and analysis unit
10
having a measurement input
9
and an outset
11
. A self-test module
15
is electrically connected to measurement and analysis unit
10
. Measurement and analysis unit
10
has a measuring element
1
, which is connected to input
9
via a summing point
3
. Measuring element
1
has a sensor transfer function
2
, which possesses the value K
1
. The output of the measuring element is connected to an analysis circuit
4
with an analysis transfer function K
2
. The output of analysis circuit
4
is in turn electrically connected to an amplifier
5
which has gain K
out
and an upper cutoff frequency f
2
. The output of amplifier
5
is connected to output
11
. Self-test module
15
has a frequency-selective detection means in the form of a synchronous demodulator
6
, which is connected to output
11
. Synchronous demodulator
6
has a demodulator
7
and an I-controller
8
. The I-controller
8
has an upper cutoff frequency f
1
. Moreover, an oscillator
18
is provided that feeds a signal having a test frequency ft into the demodulator via a frequency divider
19
. Demodulator
7
supplies its output signal to I-controller
8
, which in turn is electrically connected via a reverse signal lead
16
to an interrupt logic circuit
17
having a window comparator. Interrupt logic circuit
17
, in turn, is electrically connected to amplifier
5
of measurement and analysis unit
10
. In addition to interrupt logic circuit
17
, the output of the synchronous demodulator is connected to an electrostatic voltage-force converter
12
. Voltage-force converter
12
supplies a compensation signal to summing point
3
. In the exemplary embodiment, the measuring element is, for example, an acceleration sensor. An inertial force proportional to an acceleration to be measured is present at an input
9
of the measuring element
1
. Self-test module
15
has a test amplitude generator, which is not shown in
FIG. 1
that relays a test amplitude
14
to an electrostatic test voltage-force converter
13
, which feeds a test voltage into summing point
3
. Voltage-force converter
12
and test voltage-force converter
13
obtain the signal frequency from frequency divider
19
. Oscillator
18
supplies a signal with an oscillator frequency fosz to analysis circuit
4
, voltage-force converter
12
and test voltage-force converter
13
.
Measuring element
1
, designed as a capacitive acceleration sensor, detects an inertial force proportional to an acceleration via a deflection of a movable element of the measuring element. The effect of the inertial force on measuring element
1
is represented by input
9
in FIG.
1
. Value K
1
of the sensor transfer function is the quotient of the change in position of the movable element and the effective inertial force. The change in position of the movable element is analyzed in analysis circuit
4
and converted into a voltage change. The standard for this is analysis transfer function K
2
, which is proportional to the deflection of the movable element. In this connection, the movable element is a part of two capacitances, the capacitance change of which is converted into a voltage signal in the analysis circuit as a result of the deflection of the movable element. In a first refinement of the first exemplary embodiment, the analysis takes place by switched capacitor technology (SC technology) in a sample and hold circuit. The scanning frequency with which the analysis circuit “scans” the change in capacitance is, for example, the fourth fractional part of the oscillator frequency of oscillator
18
. The voltage signal of analysis circuit
4
is amplified with gain K
out
in a downstream amplifier
5
; the amplified voltage signal can be picked off at output
11
. The oscillator frequency is, for example, 200 kHz, the scanning frequency fa is accordingly 50 kHz. A frequency trace with the frequencies relevant to the device according to
FIG. 1
is shown in FIG.
2
. The acceleration values to be measured range, for example, from a frequency of 0 to 50 Hz, the maximum useful frequency being identified as fn in
FIG. 2
, i.e., the maximum frequency of such acceleration changes over time that are still to be detected with the measuring element. Measurement and analysis unit
10
represents an open loop since no feedback to the amplifier is provided in the sequence from the measuring element via the analysis circuit for the analysis of the acceleration signal to be measured. In this connection, the upper cutoff frequency f
2
of amplifier
5
must be greater than the maximum useful frequency fn (see
FIG. 2
in which the frequency relationships
28
are shown in the open loop arrangement). f
2
must also be greater than a test frequency ft, which is, for example, 400 Hz and is thus in turn greater than the maximum useful frequency fn. The 3-dB limit frequency of the measuring element is, for example, greater than the upper cutoff frequency of amplifier
5
(see
FIG. 2
in which a value of 1 kHz is specified as an example for the 3-dB limit frequency, the value being entered to the right of the upper cutoff frequency). Test frequency ft is the frequency of the test signal that is superimposed on the measured variable with electrostatic test voltage-force converter
13
. The test amplitude generator supplies a test amplitude that the test voltage-force converter uses to superimpose a test voltage signal with the test frequency, which is greater than the maximum useful frequency, on the measured variable by superimposing an additional force on the measured variable, i.e., the inertial force. Since upper cutoff frequency f
2
of amplifier
5
is greater than the test frequency, the test signal, which is converted into a voltage signal, is transmitted by amplifier
5
and can be detected by synchronous demodulator
6
. Demodulator
7
demodulates the output signal of amplifier
5
in a frequency-selective manner with test frequency ft. The I-controller functions as a low pass filter with a cutoff frequency f
1
. Its output signal is a voltage signal used to activate electrostatic voltage-force converter
12
. This voltage-force converter
12
superimposes a compensation signal on the measured variable in order to compensate—as described above—the supplied test signal. Voltage-force converter
12
as well as test voltage-force converter
13
are executed as circuit components that are present at electrodes provided in the capacitive acceleration sensor in order to bring about a suitable deflection or compensation deflection of the movable element of the capacitive acceleration sensor corresponding to the voltage signals. If measurement and analysis unit
10
is in proper order, the test signal is completely compensated by the compensation signal and a direct voltage is present at the output of I-controller
8
, the direct voltage corresponding to test amplitude
14
. If measurement and analysis unit
10
is not in proper order because either the mechanical measuring element is damaged or the electronic system does not function properly, the signal component with test frequency ft is amplified at output
11
. This amplification takes place via the loop formed by synchronous demodulator
6
and voltage-force converter
12
, the loop representing a negative feedback loop for test frequency signal components for the open loop of measurement and analysis unit
10
. Since I-controller
8
functions as an integrator, it integrates a signal amplitude present for an extended time with test frequency ft at output
11
; thus an extreme value (maximum or minimum voltage value) is present in this case at the output of I-controller
8
, the extreme value activating the interrupt logic circuit via reverse signal lead
16
. A window comparator assigned to interrupt logic circuit
17
registers that the voltage value of the reverse signal lead is not in a certain window which means that measurement and analysis unit
10
is not functioning properly. Accordingly, interrupt logic circuit
17
disconnects amplifier
5
or assigns a fixedly defined voltage value to output
11
of measurement and analysis unit
10
via the is electrical connection between interrupt logic circuit
17
and amplifier
5
. The voltage value is recognized as a fault signal by the external electronic system connected at output
11
. If measurement and analysis unit
10
functions properly, a voltage value is present at the output of the I-controller This voltage value corresponds to test amplitude
14
generated by the test amplitude generator. Accordingly, the window comparator of interrupt logic circuit
17
is designed in such a way that it recognizes a malfunction of measurement and analysis unit
10
as soon as the voltage value present at the output of the I-controller is not within a value range approximating test amplitude
14
, which is defined by the window comparator. Cutoff frequency f
1
of I-controller
8
, also named robustness frequency, defines the robustness of the self-test control loop formed by the open loop of measurement and analysis unit
10
and the feedback loop with regard to the useful signals occurring at test frequency ft. I-controller
8
only transmits voltage components in an integrative manner that occur with the test frequency at output
11
. Cutoff frequency f
1
is as low as possible in order to preclude upward integration of acceleration force signal components with the test frequency. With a given low cutoff frequency f
1
(0<f
1
<fn), such an upward integration is precluded since the acceleration forces to be measured customarily are within the useful band provided by maximum useful frequency fn. Thus acceleration forces with 400 Hz signal components are never present at the measuring element long enough and with a large enough magnitude to become noticeable at the output of I-controller
8
. From the above remarks, it is thus evident that the self-test carried out by self-test module
15
is superimposed on the original operation that is defined by the measured value recording performed by the open loop of the measurement and analysis unit and has no effect on the recording of the measured value because the frequency components of the variable to be measured in the frequency range of interest, which is specified by maximum useful frequency fn, are below test frequency ft and the test signal is compensated by the self-test control loop. Thus, it is a frequency-selective monitoring method without influencing the function of the measurement and analysis unit and external triggering. A test signal is no longer present at sensor output
11
apart from a very slight system deviation, i.e., the self-test is not visible to the outside. Only in the event of a malfunction is output
11
assigned a defined voltage level. The analysis of the capacitance change by the analysis circuit and the application of electrostatic forces via the voltage-force converter and test voltage-force converter
12
and
13
, respectively, take place by a time-division multiplex method, i.e., advantage is taken of the fact that the mechanical measuring element is slow-acting in relation to the time period corresponding to the scanning frequency. A condition for the time-division multiplex method is that the scanning frequency fa, which amounts, for example, to the fourth fractional part, i.e., 50 kHz, of oscillator frequency fosz, is large in comparison to the 3-dB limit frequency of the measuring element. The 3-dB limit frequency of the measuring element determines the frequency range of signal components that may be measured reliably (thus, for example, acceleration force components with frequencies up to 1 KHz in the embodiment). Oscillator
18
, which via a frequency divider
19
, supplies a signal with the test frequency that is used in demodulator
7
for demodulation and in the voltage-force converters and test voltage-force converters to generate a test and compensation signal, respectively. The oscillator
18
also triggers analysis circuit
4
and voltage-force converters
12
and
13
. Test voltage-force converter
13
is triggered in a first step analysis circuit
4
is triggered in a second step and a compensation signal is applied by voltage-force converter
12
in a third step. The measured value scanning, i.e., the activation of analysis circuit
4
by oscillator
18
, takes place at the named scanning frequency, thus for example, 50 KHz. The fact that the scanning frequency is high compared to the 3-dB limit frequency of the measuring element makes it possible for the regular recording of measured values to take place directly after application of the test signal, “although” amplifier
5
does not filter out signal components with the test frequency since the measuring element is too slow-acting to react to an applied test signal in the cycle of the scanning frequency. Only after many periods of the scanning frequency is a possibly anomalous operating state evident at output
11
, i.e., a signal component with the test frequency significantly different from zero. As an alternative to the time-division multiplex method of analysis, a carrier frequency method of analysis may also be used. In this case, the connection between oscillator
18
and test voltage-force converters and voltage-force converters
13
and
12
, respectively, is eliminated in FIG.
1
. The connection between oscillator
18
and analysis circuit
4
is retained since in this case, i.e., the second refinement of the first exemplary embodiment, the oscillator supplies the carrier frequency for analysis circuit
4
upon which the useful signal is modulated (amplitude modulation). In this case, the analysis and test signal can be simultaneous, although the test signal is applied and the measured value is picked off by test voltage converter
13
and analysis circuit
4
, respectively, at the same pick-off of the movable element of the mechanical measuring element, since as a consequence of the carrier frequency analysis, test signal applications and analysis do not mutually influence each other. In this connection, it must be observed that in
FIG. 1
, the summing point schematically represents the superimposition of the physical measured variable, test signal and compensation signal. For purposes of circuitry, the test signal application as well as the compensation signal application and analysis are implemented via the same connections to mechanical measuring element
1
.
Synchronous demodulator
6
may be replaced by a bandpass filter with downstream amplitude analysis, the bandpass frequency being equal to the test frequency.
In an advantageous manner, analysis circuit
4
, amplifier
5
, self-test module
15
and oscillator
18
may be integrated on a chip. The exemplary embodiment is also not limited to measurements of inertial force and acceleration but rather it is generally applicable to time-dependent measured variables. Correspondingly, converters
12
and
13
, respectively, must be designed accordingly in order to bring about a suitable voltage conversion.
FIG. 3
shows a second exemplary embodiment of a measurement and analysis unit
10
with a correspondingly modified self-test module
15
. In this second exemplary embodiment, measurement and analysis unit
10
represents a closed loop, in contrast to the open loop of the first exemplary embodiment according to FIG.
1
. Identical or similar circuit components are provided with the same reference symbols and are not described again. Analysis circuit
4
is connected to a pulse-width modulator
34
that activates an electrostatic signal voltage-force converter
35
(signal converter function Kr). At summing point
3
, this signal voltage-force converter in turn exercises a force on mechanical measuring element
1
simultaneously with the external inertial force. The closed circuit arrangement of mechanical measuring element
1
, analysis circuit
4
, pulse-width modulator
34
and signal voltage-force converter
35
results in a closed loop arrangement. The physical measured variable present at the input, i.e. the useful signal proportional to the inertial force, is picked off at the output of pulse-width modulator
34
and output at output
11
after amplification in amplifier
5
.
Apart from a small system deviation, the mechanical movable element of a measuring element
1
in the form of an acceleration sensor always remains in the same position; the electrical variable for compensation of the deflection of the mechanical element as a result of the effective inertial force simultaneously is the electrical signal which th can be picked off at output
11
, and proportional to the inertial force. The pulse-width modulator activates the signal voltage-force converter with a signal having the frequency of oscillator
18
, thus for example 200 KHz. The pulse width of this signal is varied with the output signal of analysis circuit
4
. The oscillator frequency is also large in relation to all other characteristic frequencies of the arrangement according to
FIG. 3
(see
FIG. 4
in which frequency relationships
29
are shown in the closed loop arrangement). In comparison to
FIG. 1
, an additional summing point
36
is provided in the second exemplary embodiment according to FIG.
3
. The additional summing point is located between the output of measuring element
1
and the input of analysis circuit
4
. In this embodiment, sensor transfer function
2
has a value K
1
that corresponds to the quotient of the resulting output voltage of measuring element
1
and the associated physical measured variable, for example, an acceleration. In normal operation when all circuit components are in proper order, this test signal is also adjusted by this closed loop arrangement, just as the useful signal, which in contrast to the test signal having a test frequency of 1 KHz, for example, only has frequencies of interest from 0 to, for example, 50 Hz as a maximum useful frequency fn (FIG.
4
). Only a voltage proportional to the physical variable to be measured is present at output
11
of measurement and analysis unit
10
since amplifier
5
connected upstream of output
11
has a cutoff frequency f
2
which corresponds to maximum useful frequency fn (FIG.
4
); i.e., signal components with test frequency ft present at the output of the pulse-width modulator are filtered out. Between measuring element
1
and summing point
36
, demodulator
7
picks off the voltage present at the output of the measuring element. In normal operation, this voltage will correspond entirely with the inverted test signal so that a signal with the test frequency is not present at the output of summing point
36
. Synchronous demodulator
6
in this case does not activate interrupt logic circuit
17
; output
11
is isolated and delivers voltage signals proportional to the measured variable. If, however, the loop arrangement is separated, the measuring element is mechanically damaged, or another electronic component of measurement and analysis circuit
10
is out of proper service condition, the signal component at the output of mechanical measuring element
1
is different from the test frequency of the test amplitude; the synchronous demodulator detects this via a signal located at the output of lowpass filter
8
that is different from the test signal and if this value is outside a specific tolerance range, amplifier
5
is “switched off” via interrupt logic circuit
17
in that output
11
is assigned a fixed voltage level.
Claims
- 1. A device for determining a time-dependent measured variable within a useful band of the time-dependent measured variable, the device comprising:a measurement circuit including a measuring element and an analysis arrangement, the measuring element supplying a measurement signal, the analysis arrangement obtaining an output voltage from the measurement signal, the output voltage representing the time-dependent measured variable, the measurement signal being present at an output of the device; and a self-test module for superimposing a test signal with a test frequency above the useful band on the time-dependent measured variable, the test signal being superimposed concurrently with the determination of the time-dependent measured variable, the self-test module including a frequency-selective detection device, a voltage-force converter, and an interrupt logic circuit, the frequency-selective detection device picking off a reaction of the measurement circuit to the test signal at the output, the voltage-force converter being capable of activation via an output of the frequency-selective detection device so that a signal for compensating the test signal may be superimposed on the time-dependent measured variable, the interrupt logic circuit being capable of activation from the output of the frequency-selective detection device with a window comparator, the interrupt logic circuit detecting whether the reaction of the measurement circuit is outside of a value range, the interrupt logic circuit assigning a fixed voltage level to the output of the measurement circuit when the reaction of the measurement circuit is outside of the value range.
- 2. The device according to claim 1 wherein:the detection device includes a demodulator and at least one of an I-controller and a lowpass filter.
- 3. The device according to claim 1 wherein:the detection device includes a bandpass filter and a downstream amplitude analysis circuit, the bandpass filter having the test frequency as a filter frequency.
- 4. The device according to claim 1 further comprising:an oscillator having an oscillator frequency, the oscillator being capable of generating a signal via a frequency divider using the test frequency.
- 5. The device according to claim 1 wherein:an input of the measurement circuit, the measuring element, the analysis arrangement, and the output of the measurement circuit form an open loop.
- 6. The device according to claim 1, wherein:the analysis arrangement includes an analysis circuit and a downstream amplifier, the analysis circuit generating a voltage correlated with the time-dependent measured variable, the downstream amplifier having a cutoff frequency above the test frequency.
- 7. The device according to claim 1 wherein: self-test module further includes a test voltage-force converter used to superimpose the test signal on the time-dependent measured variable, an amplitude of the test signal being determined by a test amplitude generated by a test amplitude generator.
- 8. The device according to claim 4, wherein:the analysis arrangement, the voltage-force converter, and a test voltage-force converter may be triggered via the oscillator, a scanning rate of the analysis arrangement being greater than a 3-dB limit frequency of the measuring element, the analysis arrangement, the voltage-force converter, and the test voltage-force converter being alternately triggerable in succession.
- 9. The device according to claim 4, whereinthe analysis arrangement, the oscillator, and the self-test module are situated on a chip.
Priority Claims (1)
Number |
Date |
Country |
Kind |
197 39 903 |
Sep 1997 |
DE |
|
PCT Information
Filing Document |
Filing Date |
Country |
Kind |
102e Date |
371c Date |
PCT/DE98/02484 |
|
WO |
00 |
6/2/2000 |
6/2/2000 |
Publishing Document |
Publishing Date |
Country |
Kind |
WO99/13345 |
3/18/1999 |
WO |
A |
US Referenced Citations (7)
Foreign Referenced Citations (4)
Number |
Date |
Country |
38 09 299 |
Sep 1989 |
DE |
44 43 941 |
Jun 1996 |
DE |
44 47 005 |
Jul 1996 |
DE |
88 01242 |
Feb 1988 |
WO |