EP1769289B1 - Method and device for the highly accurate digital measurement of an analogue signal - Google Patents

Abstract

The invention relates to a method for the highly accurate digital measurement of an analogue signal and to a corresponding device. A measuring clock pulse (f<SUB>M</SUB>) is generated from a clock pulse signal (f<SUB>T</SUB>). In a first calibration step, a capacitor (C) is charged with a constant current (I<SUB>K</SUB>) during the entire period of the clock pulse signal (f<SUB>T</SUB>). The charge voltage is maintained and is converted into a digital comparison number. In a subsequent measuring step, a first portion (T<SUB>1</SUB>) of the duration of the analogue signal is determined as an appropriately selected, whole-numbered multiple (m) of the duration (T<SUB>0</SUB>) of the period of the clock pulse signal (f<SUB>T</SUB>). The capacitor (C) is charged with the same constant current (I<SUB>K</SUB>) from the end of the first portion (T<SUB>1</SUB>) until the occurrence on the specified flank of the received signal (f<SUB>E</SUB>). The charge voltage is maintained and converted into a digital measuring number. The measuring number is divided by the comparison number, which permits the duration of a second portion (T<SUB>2</SUB>) of the analogue signal to be calculated as a fraction of a measuring clock pulse period. The measured value of the analogue signal is calculated from the first portion (T<SUB>1</SUB>) and the second portion (T<SUB>2</SUB>).

Description

The invention relates to methods according to the preamble of claim 1 for highly accurate digital measurement of the transit time of an ultrasonic signal through a fluid. It further relates to devices according to the preamble of claim 7 for carrying out the method.

From the DE 38 34 938 C is a circuit arrangement for digital detection of analog information, in particular the time interval of two successive states of at least one signal known. The circuit arrangement comprises an integration capacitor, which is charged during a charging phase via a parallel circuit of a first and a second resistor to a voltage representing the analog information. Upon completion of the charging phase, a first switch connected in series with the first resistor and controlled by a control device interrupts the flow of current through the first resistor, so that the integration capacitor is only charged further via the second resistor during the subsequent charge change phase. This further charge lasts until the capacitor voltage reaches a predetermined threshold monitored by a comparator. Since the second resistor has a greater resistance than the parallel connection of the first and second resistors, the charging time constant during the second phase is greater than the charging time constant during the first phase. During the second phase, a counter counts periodic clock pulses of a reference clock signal. The counting result of the counter is read out by an evaluation device after completion of the charge change phase. A disadvantage of this circuit arrangement is that the shorter the period of time to be measured, the shorter the shorter the second charging phase, so that long measurement times are counted more inaccurately than short measurement times. In addition, the integration capacitor is not charged linearly via resistors but according to an e-function.

From the EP 0 662 650 B a device for measuring small time intervals is known. In this case, a pulse packet composed of a number of successive individual pulses is used for the measurement in a measurement cycle. The individual pulses, which have an unknown width, are added together to form a so-called registration sum, which is just greater than a predetermined one Registration interval. The registration sum is multiplied by a stretch factor at a sampling time. The sampling time itself is measured by means of sampling pulses. Finally, the average width of the individual pulses is determined as the time interval to be measured from the ratio of the sampling time to the product of the minimum number of input pulses and the expansion factor. After a predetermined number of measurement cycles, a calibration cycle is inserted for calibration of the expansion factor. In this case, calibration pulses of known width are used instead of the individual pulses of unknown width. The disclosed circuit arrangement uses an integration capacitor which is alternately discharged and charged during successive individual pulses by means of two constant current sources until the capacitor charging voltage falls below a threshold value starting from a maximum value.

A disadvantage of this circuit arrangement is the relatively high circuit complexity, due to the use of two constant current sources, two comparators and two reference voltage sources.

In the US 4,966,150 Another method for determining the variable transit time of an ultrasonic signal is described. This method serves to calculate the distance between an ultrasonic transmitter and receiver on the one hand and a mobile wall on the other hand. The known method defines a time window within which the ultrasonic signal is detected. This time window has a fixed width and is shifted by successive approximation of measuring time period to measuring time period. For time measurement, a digital counter is used.

From the US 5 101 306 is an integrating analog-to-digital converter known, with the help of an analog voltage can be converted into a digital value.

The present invention has for its object to provide a method for high-precision, digital measurement of the transit time of an ultrasonic signal by a fluid, which with minimal effort and in the shortest possible time the delivers the desired measurement result, the resolution remaining proportional to the measurement duration.

This object is achieved by a method having the features of claim 1.

Thanks to the present method, it is possible to resolve measuring times with accuracy in the pico-seconds range with clock frequencies which operate in the single-digit megahertz range. Another advantage is that the majority of the required hardware components such as clock generator, clock divider, counter, analog-to-digital converter and reference voltage source are standard in commercially available microcontrollers. In addition, the microcontroller can statistically analyze and evaluate a series of consecutive measurements, thereby further improving the measurement accuracy.

Another advantage of the method according to the invention is that one or more measurements are preceded by a calibration. This means that gradual changes in the circuit components, in particular the integration capacitor or the constant current source, have no influence on the measurement accuracy.

A further increase in measurement accuracy in conjunction with a reduction in power consumption is achieved when the charge of the capacitor is started only after a certain number of clock signal periods. This specific number can either be fixed if it is known in which time interval the variable to be measured varies, or it can be determined adaptively with the aid of the microcontroller from a series of successive measurements.

As mentioned above, the measuring method according to the invention finds an excellent application in the measurement of the transit time of an ultrasonic signal through a fluid, for example in the consumption measurement. For this purpose, an ultrasonic signal is derived from the clock signal, which is sent via a measuring path between an ultrasonic transmitting transducer and an ultrasonic receiving transducer. The signal recorded at the receiving transducer is compared in the comparator with the reference voltage, wherein the occurrence of a signal, preferably a certain edge, at the output of the comparator means the end of the time interval to be measured.

The present invention is also based on the object, a circuit arrangement for high-precision digital measurement of the transit time of an ultrasonic signal by a fluid, in particular for the consumption measurement z. B. of drinking water, indicate. This object is achieved by a circuit arrangement having the features of the claim 7th

Advantage of this circuit arrangement is that the vast majority of the required hardware in standard microcontrollers standard exists. Only comparator, constant current source, integration capacitor and changeover switch must be connected externally. The clock generation, the reference voltage generation as well as all counting and arithmetic operations are done by the microcontroller. This also applies to the statistical evaluation of measurement series. Likewise, individual correction factors for consumption measurement can be stored in the microcontroller.

Fig. 1

eine Schaltungsanordnung zur Messung der Laufzeit eines Ultraschallsignals und

Fig. 2

die zugehörigen Puls-Zeit-Diagramme.

Reference to the drawing, the invention will be explained in more detail in the form of an embodiment. Show it

Fig. 1

a circuit arrangement for measuring the transit time of an ultrasonic signal and

Fig. 2

the associated pulse-time diagrams.

Fig. 1 shows purely schematically a circuit arrangement for measuring the transit time of an ultrasonic signal through a fluid. For this purpose, a clock signal f _{T is} generated in a clock generator 2. This clock signal is divided down in a frequency divider 3 into a measurement signal f _M. A changeover switch 13 switches the measuring signal f _M as an ultrasonic transmission signal US1 10 to an ultrasonic transmitting transducer after traversing a measuring section 11, the ultrasonic signal from an ultrasonic receiving transducer 12 is received and as a reception signal US2 via the changeover switch 13 to the one input of a Conducted comparator 14. In the comparator 14, the ultrasonic received signal US2 is compared with a reference voltage U _{ref provided} by a reference voltage generator 4. Furthermore, a pulse counter 5 is provided, which counts the clock pulses f _T , starting with the emission of the ultrasonic pulses US1. This counter 5 outputs after the time period T ₁ of a certain, suitable for the propagation time measurement number of clock pulses f m _T a switching signal to the gate 15 °.

Furthermore, the circuit includes an integration capacitor C, which is charged by a constant _current source 16 with the constant current I _K , as soon as a switch 17 is closed. The switch 17 is controlled by a gate 15, which starts the charging of the capacitor C with the switching signal from the counter 5 and stops with the next edge of the output signal f _{E of} the comparator 14. This means that at the beginning of the charging time T _2, the switch 17 is closed and reopened at the end of the charging time T ₂ .

The charging voltage of the capacitor C is determined by means of a sample-and-hold circuit 6 held. The output voltage of the sample-and-hold circuit is converted in an A / D converter into a digital number that corresponds to the measuring time.

The present circuit makes it possible to charge the integration capacitor C to a relatively high charging voltage within very short measuring times in the nanosecond range. The charging voltage itself can then be determined subsequently with high resolution, as a result of which an increase in the time resolution down to the pico-seconds range is possible.

Once the charging voltage of the capacitor C is determined, a second switch 18 is closed and the capacitor C discharged.

Since the values of the circuit components required for the measurement can change during the often several years of operation, a calibration step is possibly preceded by the measuring step. For this purpose, the capacitor C is charged during a full period of the clock frequency f _T. Thanks to this calibration, it is known in what ratio the charging voltage of the capacitor C is to the period of the clock frequency f _T. By comparing the capacitor charging voltage during the immediately following measurement process with the value determined at a preceding calibration step, the measurement time can be calculated as a fraction of a period of the clock frequency f _T.

Claims (8)

1. A method for the highly accurate, digital measurement of the transit time of an ultrasonic signal (US1) through a fluid, at least comprising the method steps:

   - generating a clock pulse signal (f_T),

   - the ultrasonic signal (US1) is transmitted via a measurement path (11) between a transmitting transducer (10) and a receiving transducer (12),

   - detecting the ultrasonic signal as an analogue signal (US2),

   - generating an output signal (f_E) as soon as the analogue signal exceeds a reference value (U_ref),

   characterised by the features:

   - a measuring clock pulse (f_M) is generated from the clock pulse signal (f_T) by division with the quotients 2ⁿ (n = 1,2,3...),

   - the ultrasonic signal (US) is derived from the measuring clock pulse (f_M),

   - in a first calibrating step:

   -- during a complete period of the clock pulse signal (f_T) a capacitor (C) is charged with a constant current (I_K), subsequently

   -- the charging voltage of the capacitor (C) is sampled and held, subsequently

   -- the charging voltage of the capacitor (C) is converted into a digital comparative number,

   - in a succeeding measuring step:

   -- starting with the emitting of the ultrasonic signal (US), a first duration (T₁) is waited, which corresponds to the whole-number multiple (m) of the period of the clock pulse signal (f_T), subsequently

   -- from the end of the first duration (T₁) until the occurrence of a specified, preferably the next, flank of the output signal (f_E) of the capacitor (C) is charged with the constant current (I_K), subsequently

   -- the charging voltage of the capacitor (C) is sampled and held, subsequently

   -- the charging voltage of the capacitor (C) is converted into a digital measuring number,

   - by division of the digital measuring number by the digital comparative number a second duration (T₂) is calculated as a fraction of a period of the measuring clock pulse (f_M),

   - and the transit time of an ultrasonic signal is determined from the first duration (T₁) and the second duration (T₂).
2. A method according to Claim 1, characterised by the feature:

   - the measured quantity is determined from the sum T₁ + T₂.
3. A method according to Claim 1 or 2, characterised by the feature:

   - the measured quantity is determined from the difference T₁ - T₂.
4. A method according to Claim 1, characterised by the feature:

   - a sample-and-hold circuit (6) is used to hold the charging voltage of the capacitor (C).
5. A method according to Claim 1 or 2, characterised by the feature:

   - an A-D converter (7) is used to convert the capacitor voltage into a digital number.
6. A method according to any one of Claims 1 to 5 for measuring the transit time of an ultrasonic signal (US1) through a fluid,
   characterised by the features:

   - the analogue signal (US2) is compared with the reference voltage (U_ref) in a comparator (14) .
7. A circuit arrangement for the highly accurate, digital measurement of the transit time of an ultrasonic signal (US) through a fluid, at least comprising

   - a transmitting transducer (10),

   - an ultrasonic measurement path (11),

   - a receiving transducer (12) for generating an analogue signal (US2),

   - a pulse generator (2) for generating a clock pulse signal (f_T),

   - a reference voltage (U_ref),

   - a comparator (14) for generating an output signal (f_E) by comparing the analogue signal (US2) with the reference voltage (U_ref),

   - a capacitor (C),

   - a device (16,17) for charging the capacitor (C),

   - a device (18) for discharging the capacitor (C),

   - a device (6,7) for measuring the capacitor voltage,

   - and a computing and control logic (1),

   characterised by:

   - a frequency divider (3) which generates from the clock pulse signal (f_T) a measuring pulse (f_M) and the ultrasonic signal (US),

   - a counter (5) which, starting from the emitting of the ultrasonic signal (US), generates a first duration (T₁) which corresponds to the whole-number multiple (m) of the period of the clock pulse signal (f_T), and which generates a control signal at the end of the first duration (T₁),

   - a constant-current source (16),

   - a gate (15) which applies the constant-current source (16)

   -- in a first calibrating step:

   during a complete period of the clock pulse signal (f_T) to the capacitor (C) and charges it with a constant current (I_K), and

   -- in a succeeding measuring step:

   applies with the control signal from the counter (5) to the capacitor (C), and subsequently

   disconnects the constant current source (16), with the occurrence of a specified, preferably the next, flank of the output signal (f_E), at the comparator (14) from the capacitor (C),

   - the device (6,7) for measuring the capacitor voltage, comprising:

   -- a sample-and-hold circuit (6) for sampling and holding the capacitor charging voltage,

   -- and an analogue-digital converter (7) which converts the held capacitor charging voltage either in the calibrating step into a digital comparative number or in the measuring step into a digital measuring number,

   - the computing and control logic (1) calculates:

   -- by division of the digital measuring number by the digital comparative number a second duration (T₂) as a fraction of a period of the clock pulse signal (f_T),

   -- and from the first duration (T₁) and the second (T₂) the transit time of the ultrasonic signal.
8. A circuit arrangement according to Claim 7, further comprising according to the invention:

   - a change-over switch (13) which switches the ultrasonic transducer (10;12) alternately as a transmitting transducer and a receiving transducer.

Images