DE102013114427A1 - Arrangement and method for determining a flow-related measured variable - Google Patents

Abstract

Arrangement for determining a flow-related process variable of a measuring medium by at least one first measuring tube, in particular for determining the flow rate of a measuring medium and / or the volume flow of the measuring medium, comprising A) at least the first measuring tube with a measuring tube axis, B) at least two coil systems diametrically opposed to the measuring tube for forming a magnetic field, C) a first pair of measuring electrodes, with two measuring electrodes diametrically opposite each other on the measuring tube, which form a first connecting axis, which runs substantially perpendicular to the magnetic field, and D) at least one second pair of measuring electrodes with two diametrically diametrically opposed on the first or on one second measuring tube opposite measuring electrodes, which form a second connection axis, which is arranged substantially along the Meßrohrachse parallel to the first connection axis; E) a control and / or regulating unit, for generating a magnetic field of alternating polarity through the coil systems, F) a circuit arrangement comprising an evaluation unit for determining a flow rate and / or the volume flow of a measuring medium, characterized in that the measuring electrode pairs connected to the evaluation unit are that a tapping of measuring voltages from the first and from the second pair of measuring electrodes takes place and wherein the circuit arrangement is designed to carry out the following method steps: I. Conductive tapping of measuring voltages in the form of signal voltages of each of the measuring electrode pairs; II.I determination of a measured value with regard to the phase position and / or the phase size of at least the interference component of the measured voltages or II.II specification of a reference value for determining the phase position and / or the phase magnitude of at least the interference component of the measured voltages; III. Signal processing of the measurement voltages as a function of a determined or predetermined phase offset, wherein the determination of an uncorrected measurement signal from the measurement voltages, wherein in step III.I a correction of the measurement signal with the measured value determined in step II.I or with the in step II.II given reference value; and IV output the corrected measurement signal.

Description

The present invention relates to an arrangement for determining a flow-related process variable according to the preamble of claim 1 and a method according to the preamble of claim 12.

Magnetic-inductive flowmeters are already known which have measuring electrode pairs arranged one behind the other along the measuring tube. However, these arrangements are selected for various reasons, which differ substantially from each other and in particular from the present invention.

The DE 10 2006 014 679 A1 discloses a measuring tube with a plurality of measuring electrode pairs connected in series with each other.

The DE 29 500 84 A1 serves the purpose of preventing a reduced measuring voltage, provided that the medium acts as a load outside of the magnetic field. For this purpose, the outer pairs of measuring electrodes are operated with the potential of the middle measuring electrode pairs.

The DE 2 226 356 has permanent magnets for generating a magnetic field.

The JP 54139756 discloses a filtering of the analog measurement signal for AC / DC components by means of a high-pass or bandpass filter. Such circuits can produce a measurement error due to parasitic effects in components.

The JP 56135118 discloses a multiple electrode arrangement for the purpose of improving the linearity of the meter. In this case, several measuring electrodes are outside the actual magnetic field.

The JP 50 66 138 describes a capacitive tap of a measuring signal at a measuring electrode. The capacitive tap is a different method from the conductive tap - because the capacitive tap, for example, requires a much larger minimum electrode area for the tap and is sensitive to changes in the strength or capacitance of the electrically insulating lining of the measuring tube in the signal pickup area

The US 5398553 discloses a multiple electrode arrangement which detects the occurrence of eddy currents and serves for self-calibration.

The EP 0 305 609 discloses a multi-electrode arrangement for eliminating the influence of variable flow profiles on the measurement result.

Based on the aforementioned prior art, it is now an object of the present invention to improve the measurement performance.

The present invention achieves this object by the features of independent claim 1 and independent claim 12. Advantageous embodiments of the claims are the subject of the dependent claims.

• A) zumindest das erste Messrohr mit einer Messrohrachse,
• B) zumindest zwei sich diametral am Messrohr gegenüberstehende Spulensysteme zur Ausbildung eines Magnetfeldes,
• C) ein erstes Messelektrodenpaar, mit zwei sich diametral am Messrohr gegenüberstehenden Messelektroden, welche eine erste Verbindungsachse ausbilden, welche im Wesentlichen senkrecht zum Magnetfeld verläuft, und
• D) zumindest ein zweites Messelektrodenpaar mit zwei sich diametral am ersten oder an einem zweiten Messrohr gegenüberstehenden Messelektroden, welche eine zweite Verbindungsachse ausbilden, welche im Wesentlichen parallel zur ersten Verbindungsachse und entlang der Messrohrachse verschoben angeordnet ist;
• E) eine Steuer- und/oder Regeleinheit, zur Erzeugung eines Magnetfeldes wechselnder Polarität durch die Spulensysteme,
• F) eine Schaltungsanordnung umfassend eine Auswerteeinheit zur Ermittlung einer Durchflussgeschwindigkeit und/oder des Volumendurchflusses eines Messmediums,

dadurch gekennzeichnet, dass
die Messelektrodenpaare derart mit der Auswerteeinheit verbunden sind, dass ein Abgriff von Messspannungen vom ersten und vom zweiten Messelektrodenpaar erfolgt und wobei die Schaltungsanordnung ausgebildet ist, folgende Verfahrensschritte auszuführen:

• I. Konduktiver Abgriff von Messspannungen in Form von Signalspannungen eines jeden der Messelektrodenpaare;
• II.I Bestimmung der Phasenlage zumindest des Störanteils der Messspannungen oder einer von der Phasenlage abhängigen Größe oder
• II.II Vorgabe eines Richtwertes zur Ermittlung der Phasenlage und/oder der Phasengröße zumindest des Störanteils der Messspannungen;
• III. Signalverarbeitung der Messspannungen in Abhängigkeit von einem ermittelten oder vorgegebenen Phasenversatz, wobei die Ermittlung eines unkorrigierten Messsignals aus den Messspannungen erfolgt, wobei III.I eine Korrektur des Messsignals mit Verwendung der in Schritt II.I ermittelten Phasenlage oder der von der Phasenlage abhängigen Größe oder mit dem in Schritt II.II vorgegebenen Richtwert erfolgt; und
• IV Ausgabe des korrigierten Messsignals.

An arrangement according to the invention for determining a flow-related process variable of a measuring medium by at least one first measuring tube, in particular for determining the flow rate of a measuring medium and / or the volume flow of the measuring medium comprises

• A) at least the first measuring tube with a measuring tube axis,
• B) at least two coil systems which are diametrically opposed to the measuring tube and form a magnetic field,
• C) a first measuring electrode pair, with two diametrically opposite to the measuring tube measuring electrodes, which form a first connection axis, which is substantially perpendicular to the magnetic field, and
• D) at least one second measuring electrode pair with two diametrically opposed to the first or on a second measuring tube measuring electrodes, which form a second connection axis, which is arranged substantially parallel to the first connection axis and along the measuring tube axis;
• E) a control and / or regulating unit for generating a magnetic field of alternating polarity through the coil systems,
• F) a circuit arrangement comprising an evaluation unit for determining a flow rate and / or the volume flow rate of a measuring medium,

characterized in that
the measuring electrode pairs are connected to the evaluation unit such that a tapping off of measuring voltages takes place from the first and from the second measuring electrode pair and wherein the circuit arrangement is designed to execute the following method steps:

• I. Conductive tapping of measurement voltages in the form of signal voltages of each of the pairs of measuring electrodes;
• II.I Determination of the phase angle of at least the interference component of the measurement voltages or a quantity dependent on the phase position or
• II.II specification of a guideline value for determining the phase position and / or the phase size of at least the noise component of the measured voltages;
• III. Signal processing of the measured voltages as a function of a determined or predetermined phase offset, wherein the determination of an uncorrected measurement signal from the measured voltages, wherein III.I a correction of the measurement signal with the use of the determined in step II.I phase or phase-dependent magnitude or with the benchmark given in step II.II; and
• IV Output of the corrected measuring signal.

The arrangement may be configured with a plurality of separate flow meters, so that the second pair of measuring electrodes is associated with a second flow meter or in a preferred compact design as a flow meter with a plurality of measuring electrode pairs.

The reference value mentioned in step II.II can be, for example, a fixed propagation speed of the disturbance or a fixed time difference between the arrival at different electrode axes of electrode spacing.

The signal processing of the measurement voltages in step III is carried out using the arrangement according to the invention in dependence on a determined or predetermined phase offset, for example by adapting the effective measurement window to a region in which the interference to be eliminated is detectable on all the measuring electrode pairs considered and / or by a time offset the signal voltages detected at different measuring electrode pairs, followed by suitable superimposition of the optionally time-shifted signal voltages at different pairs of measuring electrodes within a defined effective measuring window with the aim of extinguishing the interference.

The arrangement according to the invention makes it possible in particular to compensate for disturbances, in particular solid-state-caused disturbances

Advantageous embodiments of the invention are the subject of the dependent claims.

To reduce the measuring electrode distance, it is advantageous if the arrangement is realized in a flow meter with at least two measuring electrode pairs.

For reasons of redundancy and to increase the accuracy of measurement and to cover a larger range of propagation velocities of the perturbation, it is advantageous if the arrangement has a third pair of measuring electrodes with two diametrically opposed measuring electrodes which form a third connecting axis, which is substantially parallel to the measuring tube axis first connection axis is arranged. When using two electrode axes, the disturbance at the next field clock has already moved past the second electrode axis at a certain speed, then a third electrode axis further away is advantageous. On the other hand, below another speed, the time of a measurement period does not suffice for reaching the second electrode axis, then a third electrode axis between the first and second is advantageous.

It is advantageous if the second and / or the third measuring electrode pair has measuring electrodes which differ in their geometric shape and / or in their material composition from the first pair of measuring electrodes.

• • Verstärkung und/oder Impedanzwandlung einer Messspannung jedes einzelnen Messelektrodenpaares oder einer gemeinsamen Messspannung durch parallelen Abgriff mehrerer Messelektrodenpaare gemeinsam;
• • Offsetkorrektur und ggf. weitere Verstärkung der Messspannung eines jeden einzelnen Messelektrodenpaares oder der gemeinsamen Messspannung der Messelektrodenpaare
• • Analog-Digitalwandlung der Messspannung eines jeden einzelnen Messelektrodenpaares oder der gemeinsamen Messspannung der Messelektrodenpaare; und
• • Ermittlung einer Durchflussgeschwindigkeit und/oder eines Volumendurchflusses anhand der störabstandsoptimierten Messspannung oder Messspannungen,

wobei bei der Offset-korrektur und ggf. Verstärkung der Messspannung, sowohl Wechselspannungs- und Gleichspannungsanteile der Messspannungen miteinander addiert werden und wobei durch die Schaltungsanordnung eine Optimierung des Störabstandes der Messspannungen eines jeden einzelnen Messelektrodenpaares oder der gemeinsamen Messspannungen aller Messelektrodenpaare erfolgt.The circuit arrangement can advantageously carry out the following further method steps:

• Amplification and / or impedance transformation of a measuring voltage of each individual measuring electrode pair or of a common measuring voltage by parallel tapping of several measuring electrode pairs in common;
• • Offset correction and, if necessary, further amplification of the measuring voltage of each individual measuring electrode pair or the common measuring voltage of the measuring electrode pairs
• • Analog-to-digital conversion of the measuring voltage of each individual measuring electrode pair or of the common measuring voltage of the measuring electrode pairs; and
• Determination of a flow velocity and / or a volume flow on the basis of the noise margin-optimized measuring voltage or measuring voltages,

wherein in the offset correction and possibly amplification of the measuring voltage, both AC voltage and DC voltage components of the measuring voltages are added together and wherein the circuit arrangement is carried out optimizing the signal to noise ratio of the measuring voltages of each individual pair of measuring electrodes or the common measuring voltages of all pairs of measuring electrodes.

It is advantageous for the phase position to be determined by cross-correlation or by iterative calculation on the basis of a plurality of preceding individual measured values of the measurement signal.

Alternatively, if the process is known, the phase position can be specified as a parameter.

It is advantageous if the information obtained in the signal processing of the phase shift or the change in the phase position is used to adjust the excitation of the magnetic field.

Several cases can be distinguished. In the simplest case (a) one starts from a flow fluctuating only slightly by a constant mean value. The excitation can then be selected so that the interference between two measurement windows on average just covers the distance between the electrode axis. This represents the case of a constant time interval between two field changes adapted to the typical flow (referred to as "measuring period" in the following). Should flow change in the next case (b) considered, e.g. During a multistage filling, but only with slight fluctuations relative to a periodic flow profile averaged over several periods, a periodically variable (ie variable) adaptation of the measuring period can take place. Cases (a) and (b) assume that the flow rate is known and predictable to some degree, so that a targeted adjustment of the measurement period can be made.

In many cases, however, there is no information about the flow in advance. With a constant measurement period and constant propagation speed, it could then happen that the disturbance always arrives just outside the measurement window (integration period) at the second electrode pair, either too early or too late. In this case, according to variant c), the measuring period can advantageously be changed in time, for example shortened or extended by more than 5%, in particular more than 20% shortened or extended, in particular shortened or extended by more than 50%. The temporal course of the change can take place according to an explicitly predetermined fixed pattern or contain random values as a proportion or external variables (for example, time, temperature, flow) to determine the change. In the case of the flow here is meant no adaptation, but only an inclusion of this variable for variable adaptation in the sense of a control, no regulation. The variable change of the measurement period described in this section avoids the disturbance constantly arriving outside the measurement window.

In the case of the method described in the last section, the measuring period will typically not be adapted to the disturbance with a greater percentage of, for example, more than 5%, in particular more than 20%, in particular more than 50% of the measurement periods in the operation of a device. For the computational effort for adaptation is relatively low. If there is no knowledge of the flow in advance and nevertheless the disturbance occurs as completely as possible in the measurement window in the second electrode pair, the method according to the invention proposes in a variant d) after each determination of the phase shift and thus with knowledge of the time offset between Measuring window and passing the fault shorten the measurement period after the immediately following field change accordingly or extend, so that the disturbance at the next measurement window can be detected as completely as possible. In the case of very rapid changes in the flow, it may be advantageous to operate the device with a higher field change frequency, thus to detect the flow of the flow with a higher temporal resolution and to detect the fault even in measurement periods that lie further in the future. For example, suppose a disturbance needs a standard measurement period between the two electrode axes. Now the measurement period is reduced to half. Then a flow change can already be detected at the next measurement period, but the disturbance has not yet arrived at the second electrode pair. From the detected change in flow rate, the measurement period of the next but one field clock can now be adjusted so that the disturbance arrives in the measurement window on the second pair of electrodes. Another possibility is to use flow information before the actual measurement window to adjust the measurement period. Although this information has a higher error, since the magnetic field in front of the measurement window is not yet stable, this does not increase the error of the flow determination, since it is not used directly for flow calculation but only for adaptation of the integration window. Yet another possibility is to continuously calculate a current flow value during a much higher sampling (advantageously at least a factor of 10, in particular a factor of 100 higher than the field frequency) during the measurement window. This flow rate is associated with a higher uncertainty due to the low averaging time and the failure to compensate for the disturbance, but this does not increase the flow determination error since it is not used directly for flow calculation but only to adjust the measurement period or integration window.

The object of the invention will be explained in more detail below with reference to an exemplary embodiment with the aid of drawings.

They show:

1 schematic structure of a magnetic-inductive flowmeter;

2a -F possible circuit arrangements for tapping the measuring voltages at the measuring electrodes of the flowmeter;

3 schematic representation of various electrode designs;

4 Correction at ideal measuring period;

5 Adaptation if the measuring period is too short;

6 Flowchart for the inventive method;

7 Correction results from experimental experiment; and

8th Measured value fluctuations compared to averaging the measured values;

9 Diagram for the position of correlated sections of the interference signal at different speeds; and

10 Diagram 10 as a qualitative representation of the filter effect.

The basic structure and the measuring principle of a magnetic-inductive flowmeter are known per se. According to Faraday's law of induction, a voltage is induced in a conductor moving in a magnetic field. In the magnetic-inductive measuring principle, the fluid medium corresponds to the moving conductor. A magnetic field with alternating polarity and during the integration window as far as possible with constant and always the same strength is generated by two field coils on both sides of a measuring tube. Perpendicular to this are located on the tube inner wall of the measuring tube, two measuring electrodes, which tap the voltage generated when flowing through the medium. The induced voltage is proportional to the flow velocity and thus to the volume flow. The magnetic field built up by the field coils is generated by a clocked DC alternating polarity. This ensures a stable zero point and makes the measurement insensitive to influences by multiphase substances, inhomogeneities in the liquid or low conductivity. Magnetic-inductive flowmeters with coil arrangements with more than two field coils and other geometrical arrangements are known.

The quality of a measurement by means of a magnetic-inductive flowmeter can be characterized among other things by the mean measurement deviation and the measured value dispersion. The measured value scattering depends, for example, on the type of medium and the flow rate of the medium.

For example, at a nominally constant flow, the standard deviation of successive measurements should be as low as possible. Averaging or otherwise filtering successive measurements reduces the spread of measurements, but at the expense of the bandwidth of the measurement.

An optimization of the signal-to-noise ratio - that is to say the ratio of the useful signal and the interfering signal component which makes up each measuring voltage - leads to a reduction of the measured value dispersion and / or to an increase in the bandwidth of the measurement. This improvement of the signal-to-noise ratio is the aim of the present invention. For example, it is possible to either average the same length while achieving a lower spread with the same bandwidth. Or it is averaged shorter in another example and thus achieved a higher bandwidth with the same dispersion. Or it is averaged in a third example, that both reduces the spread and increases the bandwidth.

It is particularly advantageous to improve the S / N ratio in low-conductivity media or even in media with heavy particle loading, in which this disorder occurs to a greater extent and for which other solutions have been established, for example in the filling. For example, increasing the field strength with the disadvantage of higher energy consumption or avoidance of solids in the flow meter, for example according to EP 2 272 790 B1 ,

The solution for optimizing the signal-to-noise ratio of a measuring voltage consists in a redundant tapping of the measuring voltage by means of several measuring electrode pairs.

The connection axes of the respective measuring electrode pairs are shifted parallel along the measuring tube axis. In this case, all measuring electrode pairs are in the range of a single magnetic field, which is generated by coil systems correlated with one another in terms of their time control. Of course, this does not apply if the electrode axes are distributed over several consecutively arranged flowmeters.

1 shows a schematic structure of a magnetic-inductive flowmeter, as well as in the DE 10 2013 103 211.7 already described. This flowmeter has a measuring tube 1 which, for example, as a plastic measuring tube or as a steel measuring tube is present, which is lined with an insulating plastic layer, the so-called liner.

The measuring tube 1 has at least two, preferably three or more measuring electrode pairs 2.1 . 2.2 . 2.3 which are arranged on a common plane, the measuring electrode plane. The measuring electrode plane is arranged perpendicular to the coil connecting axis and the measuring tube axis runs in the measuring electrode plane. The connecting axes of the measuring electrode pairs are perpendicular to the measuring tube axis, parallel to each other and along the measuring tube axis by different distance parallel to the measuring tube 1 arranged.

The distance between the measuring electrodes is variably selectable. However, it is limited by the effective extent of the magnetic field. This magnetic field is through at least two coil systems 5.1 and 5.2 generated, wherein the magnetic field is substantially perpendicular to the measuring electrode axis.

The measuring electrode pairs 2.1 . 2.2 and 2.3 transmitted on signal paths 3 the determined voltage (which is dependent on the flow velocity of the measuring medium) to an evaluation unit 4 , It is understood that the transmission is not dependent on the flow velocity, but only the determined voltage itself. Also part of the magnetic-inductive flowmeter is a so-called coil current control unit 6 which the coil systems 5.1 and 5.2 drives and energizes.

The measuring electrodes can be connected to the evaluation unit in different ways. These circuit diagrams of the signal processing of the tapped off at the measuring electrodes voltage are in the embodiments 2a - 2f shown. Here are the variants of the 2d F preferred embodiments of the invention. 2d is a particularly preferred embodiment of the invention.

In the 2a - 2f Each interconnected circuit elements are shown.

In 2a - 2f become the incoming voltage signals of the measuring electrode pairs 2.1 - 2.3 via a differential amplifier 10 to an instrument amplifier 11 , a so-called adder, forwarded. The instrumentation amplifier also allows impedance conversion.

The instrumentation amplifier is equipped with an analogue to digital converter 12 interconnected and optionally or additionally with a multiplexer.

Finally, the signal is sent to an evaluation unit 13 , For example, passed a digital evaluation in the form of a microcontroller.

2 B represents a second less preferred embodiment of an arrangement of circuit elements. In this case, each of the incoming signals by means of a separate differential amplifier 10 strengthened. The gain can be weighted so that each individual type of measuring electrode is different in the total value.

2c represents a third less preferred embodiment of an arrangement of circuit elements. In this case, each incoming signal is amplified by means of a separate differential amplifier and instrumentation amplifier and converted by a common A / D converter. Also in this case, the gain of the signals can be weighted.

The weighted gain may be dependent upon the expected signal quality at the different electrodes, depending on the application, as set forth below. The measuring arrangement may in fact have a plurality of different electrodes known per se. Various measuring electrodes used in the field of electromagnetic flow measurement are in 3 shown schematically. The measuring electrode 5a is a standard measuring electrode, also called mushroom head electrode. It is characterized by a dome-shaped surface in the direction of the measuring medium, which causes a slight influence on the flow of the medium.

The measuring electrode 5b is a pointed electrode. This surface of this measuring electrode runs in the direction of the measuring medium pointed. Due to the tapered electrode shape, a coating formation which impairs the measurement is reduced or even prevented in the case of typical scale-forming media in the region of the electrodes.

The measuring electrode 5c is a so-called neck electrode. Their surface protrudes a few millimeters into the medium. This is done by spacers for Meßrohrwandung. The neck electrode is used for measuring media with low flow rates or flow rates.

The measuring electrode 5d is referred to as ironing electrode. It has two pin-like end portions, which protrude starting from the electrode body into the measuring medium. This ironing electrode allows the measurement of solids loaded liquids, such as fruit mash, and liquids with a high proportion of gas and corresponding gas bubbles.

The measuring electrode 5e is a tungsten carbide electrode, which sits flat on the Meßrohrwandung and is used in particular for the measurement of abrasive fluids, such as abrasive sludge.

With larger proportions of solids in the medium, such as coarse sludge, it can cause vibration or damage to a rigid electrode. As an alternative, a brush electrode is particularly suitable 5f whose electrode head is flexible.

In addition, other types of electrodes are known, for example, pin electrodes, sintered electrodes or alternating electrodes which can be used in the subject matter of the invention.

According to the invention it is possible by the arrangement of measuring electrodes to connect several types of measuring electrodes, for example two or more types of the aforementioned measuring electrodes in a magnetic-inductive flowmeter, so that the flow of different media can be determined with the same flow meter. Depending on requirements, one of the signals can be amplified. Are for example in 1 the measuring electrodes 2.1 and 2.2 Mushroom head electrodes and a third measuring electrode 2.3 a tip electrode and the medium is milk, which can form a fatty coating on the electrodes, it is advisable to weight or amplify the signal of the tip electrode against the signal of the mushroom head electrodes stronger.

2d In this case, each measurement voltage is amplified separately from the other measurement voltages by a differential amplifier and an instrument amplifier, offset corrected and each converted by an analog / digital converter. The converted digital signals are then supplied to the microcontroller, which optionally carries out a weighting of these signals. Each individual information obtained from the signal is fed to the microcontroller via a separate path of circuit elements, whereby a separate evaluation of each individual signal can take place without information being lost by previous addition.

2e shows an arrangement of circuit elements, which allows two signals are added by means of a single instrumentation amplifier, while another measurement voltage is supplied via an independent path to circuit elements independently of the other two signals of the evaluation unit. This measurement signal can be available as a comparison value or, due to another measuring electrode geometry, enable interference compensation (eg of the disturbance factor of the air bubbles or solid components) of the two added signals.

2f shows one too 2e analog arrangement of circuit elements with the difference that the signals of the two instrumentation amplifiers 11 through a single A / D converter 12 being transformed.

The arrangement may enable tapping and addition of the analog AC and DC portions of the measurement voltage to reduce the occurrence of measurement errors.

The purpose of the invention is to optimize the signal-to-noise ratio of the measurement signal. The signal-to-noise ratio is generally referred to as the ratio of the mean power of the useful signal to the mean noise power of the interference signal.

Like from the 2a - 2f can be seen, the tapping of the signal voltage of each pair of measuring electrodes takes place parallel to each other. A different superposition of the signal voltages takes place by means of the inventive arrangement of circuit elements, the so-called amplifier electronics.

As a result, depending on the characteristic of the interference to be suppressed, a plurality of redundant signal paths-or the corresponding circuit elements of a signal path-can be combined in a circuit element. This means a high degree of flexibility of the measuring system depending on the requirements of the measuring medium.

The redundancy of the signal paths can also be used to check the functional safety. This may be, for example, for an arrangement of the circuit elements as in 2d such that individual measuring voltages of measuring electrodes are compared with one another by the evaluation unit and a failure is detected if one of the three measured values deviates significantly from the other two measured values. This may indicate a defect in the electrode or a defect in one of the circuit elements.

By optimizing the signal-to-noise ratio, the measuring range of the magnetic-inductive measuring device according to the invention can also be extended to media which have a low conductivity in comparison to the typical specification for IDM and therefore due to high interference signals after DE 10 2013 103 211.7 cause a higher measurement uncertainty. The same applies to particle-laden media, which usually cause larger interference signals.

The invention is based on the assumption that at least two pairs of electrodes are used.

To generate the magnetic field, the coil systems are operated in a conventional manner by a clocked DC alternating polarity.

The detection of the measured voltages is carried out in the present invention over the entire time measuring range. It is u.a. also measured in the period during the switching of the polarity of the magnetic field. By switching the polarity of the DC component of the noise voltage is suppressed. A DC interference voltage can occur, for example, because, depending on the nature of the electrode and the measuring medium, an electrochemical potential is formed on the surface of the electrode in contact with the medium. This creates a voltage difference between the measuring electrodes, which can be up to 1.2V and more. By switching the polarity, however, capacitive and inductive noise signals are generated with the same polarity as the useful signal and therefore difficult to separate from each other. Typically, therefore, the measurement is waited until these interference signals caused by the field change have decayed, or at least in each measurement period are the same irrespective of the application, so that a computational compensation can take place if required.

The measuring voltages contain DC components and AC components. These are always present in the analog range of the measuring voltages.

In order to separate these analog components from each other, analog compensation circuits for filtering were used for this purpose. At the output of the DC voltage component of the measuring voltage was zero. High-pass or band-pass filters were also used to filter the analog measurement voltage.

However, such an analog compensation circuit can be a cause of measured value errors. According to the invention, therefore, there is no filtering of the DC components of the measuring voltage as long as the measuring voltage is analog. As a result, the analog path requires a larger range of voltages that it can measure, so that no voltage values that are too high are cut off, resulting in no distortion of the mean value of the signals.

After the analog / digital conversion, the voltage components can be filtered.

In 4 the offset correction is displayed at the ideal propagation speed of the disturbance. In this case, the disturbance is unchanged in its shape in the time interval 1 (ZI 1) on electrode pair 1 (upper schematic course of the electrode difference signal) and in the time interval 2 (ZI 2) on electrode pair 2 (lower schematic course of the electrode difference signal). Relative to the respective measurement window ZI i with i ε {1, 2} (integration time), the disturbance has the same position or phase position.

4 represents the special case, in which a complete suppression of the disturbance succeeds by forming the difference of the measurement signals in the time interval ZI1 and time interval ZI2. The disturbance continues with the flow rate of the medium. The mean velocity of the perturbation is just so great in the example that the distance between the electrode axes is covered within a measurement period, defined as the time between two magnetic field switches. The fault thus appears in phase with respect to the measuring window (solid line) or in the integration time in both measuring periods or field clocks. Since the magnetic field has changed its polarity when it reaches the electrode axis of the second measuring electrode pair, the electrode voltage is charged with the opposite sign as in the previous field clock.

As in the example of 4 the disturbance, in contrast to the useful signal in the second electrode pair in the second field clock, is identical to the first field clock in the first electrode pair, it is completely eliminated when subtracting or averaging with different signs.

Starting from the in 4 However, it can also happen that the disturbance progresses more slowly and partially arrives at the second electrode pair only after the end of the next measurement period. This case will be in 5 presented and discussed in more detail below. A disorder with the behavior as it is in 5 can be represented by the subject of the DE 10 2013 103 211.7 Although partially corrected during the measurement, the method according to the invention provides, however, in a further development of the aforementioned publication a significantly better compensation of the measurement error.

In particular, the method described below is intended to make use of the correlation of the interference signals and to eliminate them so that more stable measured values result.

In the 4 . 5 and 5a and 5b are two measuring voltages 101 and 102 which are connected to two (as in 1 represented) successively arranged measuring electrode axes were shown. The respective interference component of the measurement signal was designated A1 and A2.

The diagram of 5 shows an embodiment for the case in which the propagation speed of the disturbance proceeds somewhat slower than in the 4 illustrated special case. The fault requires a slightly longer time than a measuring period to cover the distance between the two electrode axes. This temporal difference now leads to the fault having a phase shift relative to the integration time in the second field clock at a first field clock. In the example, a part of the disturbance does not arrive until after the end of the integration time, is thus not detected and can not be used to compensate for this section of the disturbance on electrode pair 1 at the earlier time. Partial compensation remains possible as long as at least one subsection of the disturbance is present during the integration time for electrode pair 1, even during the integration time for electrode pair 2. This method with a constant measuring window with a phase shift is in 5a shown.

5b In this case, the measurement window on both pairs of electrodes, or the ZI 2 and ZI 1, is reduced to the area in which the correlated interference could be detected on both electrode axes during the integration time. In this variant, the correction in the entire integration range is possible, with a reduction of the measurement window is necessary. As a result, the noise increases due to the lower integration period. However, the advantages of the disturbance compensation predominate, especially in the case of large disturbance amplitudes, which reach more than 10%, in particular more than 100%, in particular more than 500% of the signal amplitude, as long as the integration time is greater than 10%, in particular greater than 50%, in particular greater than 90% of the original duration is.

The part of the disturbance not detected at a fixed field clock and a non-optimal propagation velocity of the disturbance can also be detected by adjusting the field clock. In this case, the time duration between two field switches with the same integration time (each at the end of each field period) is extended so that the (unchanged drawn) fault with electrode pair 1 and 2 has the same phase relative to the integration window again with the situation as in Figure 4, only with a slightly different time scale. As a result, an almost complete extinction of the fault is also possible.

This makes it possible to achieve the best possible suppression of the disturbance also for migration speeds of the disturbance, which deviate from the special case in which the phase offset of the disturbance corresponds to exactly one measurement period or zero within a measurement period.

It is to be accepted that by extending the time interval, within which real flow changes can be detected, since a flow value is calculated per measurement period. The sampling rate of the flow measurement is therefore lower.

The great advantage of the method is that it allows to achieve the best suppression of the interference signals not only punctually at certain flow velocities (special case in diagram of the 4 ), but also at deviating flow velocities (such as in the diagram of 5 ).

The method allows the above-described measuring device to record information about parameterization from the outside or to obtain from the measurement signals it detects itself and to adapt its signal processing and optionally also the time profile of the field excitation based on this information for different flow velocities such that the best possible suppression of the interference signals is reached.

The block diagram of 6 shows the operation of the method. Dotted and dashed lines represent optional process steps.

In step I, one signal voltage each 101 and 102 taken from the example, two measuring electrode pairs of the magnetic-inductive flowmeter and transmitted to a signal processing unit. Before that, there is always a magnetic field excitation, which can be an excitation after step V or a standard excitation. The signal voltages are digitized to measurement voltages in the signal processing unit, if appropriate after corresponding analog preprocessing. The measured voltages are voltage values that were acquired over a specific time interval.

In an optional intermediate step I.I, a zero point adjustment takes place. This can be done in the signal processing unit. The zero point adjustment can advantageously reduce the measurement deviation, especially at low speeds.

In step II.I or II.II, a determination or predefinition of the phase position or the phase offset of the interference component A ₁ and A _{2 of} the two measured voltages takes place 101 and 102 at the respective medium and the respective flow velocity relative to each other. Alternatively, a variable dependent on the phase position or on the phase offset can be determined. A phase position can be determined during a measurement period, for example, while the dependent variable can be determined over several measurement periods.

The tapped measuring voltages basically consist of a useful signal and interference. The phase can be determined, for example, by summing the signals on electrode pair 1 and electrode pair 2 for successive measurement periods with reversed sign magnetic field, whereby the useful signal is largely eliminated and the interferences on electrode pair 1 and electrode pair 2 remain. These temporal courses of the disturbances at the two electrode pairs are now shifted against each other by different time periods, multiplied and integrated the product. This process is known as cross-correlation. The time offset at which the integral becomes maximum corresponds to the phase shift of the disturbances on electrode pair 1 and 2 relative to one another.

In the iterative calculation, first a flow or a flow velocity is determined, which has not been corrected on the basis of the disturbance compensation. In the following, it is assumed that the disturbance propagates substantially at the flow rate. From this uncorrected value, a phase offset is determined with known electrode spacing and known flow rate, which occurs between the first and the second measuring electrode axis. Now, the measured voltages determined at the measuring electrode axes can be shifted relative to one another by the phase which was previously determined on the basis of the uncorrected flow value. This achieves a first corrected, but in comparison to the cross-correlation less optimal-corrected flow value. If further optimization is desired, the first corrected flow value can be used again to calculate the phase shift. This more precisely determined phase shift allows a calculation of a further optimized second corrected flow measurement value. The iteration loops can be repeated as needed. The convergence of the phase shift can be used as evaluation criterion and, if necessary, termination criterion for the iteration.

Alternatively or in addition to determination, the expected phase shift can also be specified as a parameter in the case of an approximation value for a propagation velocity. For this purpose, a parameter specification in the optional step II.II. For example, in the case of a bottling plant, a rough guideline value is often already known for which flow velocities will prevail when the valve is open. However, beyond the guideline, a more accurate flow measurement for this range is desired. For this purpose, the approximate guideline value can be specified and used as a parameter for determining the exact speed. Additionally or alternatively, during or after the determination of the phase shift in step II.I or II.II, a determination of a decision value based on predetermined decision parameters or criteria also takes place. If little or no correlation is observed, a signal correction by a subsequent step III.I can not take place, but conventional evaluation methods VI are used. Other decision criteria may be e.g. be too small or too large a phase shift.

After the step II of determining or specifying the phase, a signal correction follows in step III. In this step, as a function of the previously determined phase in a substep III.I, a mathematical cancellation of the correlated interference in the measurement signal takes place on the basis of the phase offset or a value proportional to the phase offset.

In step IV, the output of a corrected measurement signal takes place.

There are two sets of measurement signals, each of pair of electrodes 1 and 2. In both sentences, one (of course there may be several) interference occurs at different times. Due to the correlation of the interference signal at the two pairs of electrodes, it is possible to identify this. This can be done a correction of the two measurement signals. And then the "normal" signal processing takes place.

Optionally, after the signal processing in a step V, the excitation can be adjusted. This can be done on the basis of predetermined parameters. The adaptation of the excitation is carried out to a constant preset value. Preferably, however, the excitement can also be adjusted variably.

In this case, the information obtained in the signal processing of the phase shift or the change in the phase position can be advantageously used to adjust the excitation of the magnetic field.

Several cases can be distinguished. In the simplest case (variant a) one starts from a flow that fluctuates only slightly by a constant mean value. The excitation can then be chosen so that the interference between two measurement windows on average just covers the distance between the electrode axes. This represents the case of a constant time interval adapted to the typical flow between two field changes (called in the further measuring period). Should the flow change, eg during a multi-stage filling, but only with small fluctuations relative to a periodic averaged over several periods Flow rate, a periodically variable (ie variable) adjustment of the measurement period can be carried out (variant b). Cases (a) and (b) assume that the flow rate is known and predictable to some degree, so that a targeted adjustment of the measurement period can be made.

In many cases, however, there is no information about the flow in advance. It could then occur with a constant measuring period and constant propagation speed that the disturbance always arrives just outside the measuring window (integration period) on the second pair of electrodes, either too early or too late. In this case, according to variant c), the measuring period can advantageously be changed in time, for example shortened or extended by more than 5%, in particular more than 20% shortened or extended, in particular shortened or extended by more than 50% (variant c). The temporal course of the change can take place according to an explicitly predetermined fixed pattern or contain random values as a proportion or external variables (for example, time, temperature, flow) to determine the change. In the case of flow, here only an inclusion of this variable for variable adaptation in the sense of a control is meant, not in the sense of a regulation. The variable change of the measurement period described in this section avoids the interference constantly arriving outside the measurement window.

In the method described in the last section, the measuring period will typically not be adapted to the disturbance at a greater percentage of, for example, more than 5%, in particular more than 20%, in particular more than 50% of the measurement periods in the operation of a device. For the computational effort for adaptation is relatively low. If there is no knowledge about the flow in advance and nevertheless the disturbance occurs as completely as possible in the measurement window in the second electrode pair, the method according to the invention proposes, after each determination of the phase shift and thus with knowledge of the time offset between the measurement window and passing the disturbance to shorten or lengthen the measuring period accordingly after the immediately following field change, so that the disturbance can be detected as completely as possible at the next measuring window (variant d). In the case of very rapid changes in the flow, it may be advantageous to operate the device with a higher field change frequency, thus to detect the flow of the flow with a higher temporal resolution and to detect the fault even in measurement periods that lie further in the future. For example, suppose a disturbance needs a standard measurement period between the two electrode axes. Now the measurement period is reduced to half. Then a flow change can already be detected at the next measurement period, but the disturbance has not yet arrived at the second electrode pair. From the detected change in flow rate, the measurement period of the next but one field clock can now be adjusted so that the disturbance arrives in the measurement window on the second pair of electrodes. Another possibility is to use flow information before the actual measurement window to adjust the measurement period. Although this information has a higher error, since the magnetic field in front of the measurement window is not yet stable, this does not increase the error of the flow determination, since it is not used directly for flow calculation but only for adaptation of the integration window. Yet another possibility is to continuously calculate a current flow value during a much higher sampling (advantageously at least a factor of 10, in particular a factor of 100 higher than the field frequency) during the measurement window. This flow rate is associated with a higher uncertainty due to the low averaging time and the failure to compensate for the disturbance, but this does not increase the flow determination error since it is not used directly for flow calculation but only to adjust the measurement period or integration window.

• (1) Phasenbestimmung online a. Kreuzkorrelation b. Iterative Berechnung des Durchflusses und Bestimmung der Phase
• (2) Angepasste Erregung (Schritt V) a. Konstant, auf typischen Durchfluss angepasst b. Variabel, auf typisches Durchflussprofil angepasst c. Variabel mit fixem Muster, so dass vermieden wird, dass die Erregerfrequenz dauerhaft dem schlechtesten Fall entspricht d. Variabel, adaptiert auf frühere Durchflussmessung

In summary, there are several possible procedures for the phase determination and the matched excitation of the coil systems:

• (1) Phase determination online a. Cross correlation b. Iterative calculation of the flow and determination of the phase
• (2) Adjusted arousal (step V) a. Constant, adjusted to typical flow rate b. Variable, adapted to typical flow profile c. Variable with fixed pattern, so that it is avoided that the exciter frequency permanently corresponds to the worst case d. Variable, adapted to earlier flow measurement

Based on 7 can be shown how great the potential for improvement of the process. The example builds on interference signals that were synchronously measured with a sensor with several electrode axes. The medium contained pieces of fruit which caused a significant interference signal. It is therefore about the measurement of a solids-containing medium. For these interfering signals, a generated measuring signal was added, which corresponds to a constant flow in similar height as in the recording of the interference signals.

In 7 is the tapped as a measured signal voltage applied over a time axis. The The upper and the lower measuring signals M1 and M2 each represent the measuring signal at the electrode axes. The middle measuring signal M3 shows the corrected value, which was corrected according to the aforementioned method.

The situation in this example is similar to that used to explain the procedure above.

7 shows the two signals at the individual electrode axes. These distorted signals correlate very strongly and have a phase offset corresponding to the time it takes for the medium or the disturbance to flow from one electrode axis to the next. This phase shift has already been determined by means of cross-correlation and the signals are shifted relative to one another by this phase. By averaging the two signals, the noise component falls to a large extent away. This results in the middle much more stable signal. In this example, there is a small time difference between the phase offset and the measurement period, which explains the "steps" in the polarity reversal.

The diagram in 8th shows how much the scattering of the measured values by the method (see 6 ) can be reduced. It shows two traces. The measurement curve with the increased fluctuations characterizes the scatter of the uncorrected measurement curves M1 or M2 of the 7 , The second measurement curve characterizes the scatter of the corrected measurement curve M3. As can be seen by comparison of the two measurement curves, the scatter of the measured values in this case decreased to one fifth, which means a reduction of 80%. The signal-to-noise ratio increased fivefold, which represents an increase of 400%.

Of course, the disturbance-related scatters are strongly dependent on the medium, the type of disturbances, the flow rate and the switching frequency.

The diagrams of 9 and 10 show further cases for different propagation velocities of the disorder. 9 represents a structure with three electrode axes and how the disturbed measurement voltage is tapped at the individual electrode axes as a function of the flow rates. The solid lines describe the measuring interval in which the measuring voltages can be tapped. Further dotted areas in which the field switching takes place were not used for calculating the measured value and can not be used due to the disruptive effect of the field switching. The encircled symbols + and - represent the polarity of the magnetic field. Compensation of the interfering signal presupposes that in the intervals with correlated, interference is measured on an electrode axis with positive field polarity and measured on at least one electrode axis with negative field polarity. The 9 shows three cases A, B, C, in which at least one positive and negative polarity is applied to the electrode axes.

Due to the variable excitation, as they were described in particular in the variants c) -d) to step V, the setting always succeeds so that a Meßsignalabgriff is avoided with only positive or only negative polarities.

In 9 the position of correlated portions of the interfering signal relative to excitation at different velocities is shown

In 9 the situation at the individual electrode axes over time is analogous to 4 and 5 ,

In 10 the gain G of the respectively selected filtering is plotted over the propagation speed of the disturbance. It concerns with 10 Therefore, assuming that the perturbation propagates at the same rate as the mean flow rate, a qualitative representation of the filtering effect as a function of the flow rate. In 10 is qualitatively illustrated how the method described compared to the simple averaging depending on the nature of the suppression of the disturbance for a larger speed range allows.

Measurement curve L1 represents a simple averaging. Measurement curve L2 represents a stepwise adaptation to the phase with a fixed integration time. Measurement curve L3 represents a stepwise adaptation to the phase with a fixed integration time, depending on the propagation velocity of the disturbance over several measurement periods Measurement curves L2 and L3, the further electrode pair and its measuring voltage is used for compensation. Measurement curve L4 represents a continuous adjustment to the phase with variable integration time. The measurement curve L5 represents a continuous adjustment to the phase with variable integration time and adaptive adjustment of the natural frequency. For the measurement curves L4 and L5 compensation takes place taking into account the position and duration of the integration interval ,

v1 sets the highest explicitly considered flow velocity of the medium through the in 1 The interference needs between the first and third electrode axis significantly less than a measurement period, more precisely, a fault that occurs at the end of the measurement window in field clock 1 on pair of electrodes 1 just at the beginning of the measurement window in field clock on electrode pair 3 are detected. At this limiting speed, the correlated interference can still be compensated in the overlap area (assuming the temporal constancy in the local coordinate system), since it falls into the integration window in both field clocks. At higher speeds than v1, the disturbance is eliminated to an even lesser extent and finally no longer by the compensation method according to the invention, since it has already passed the electrode pair 3 before the beginning of the next field clock. In contrast to special case v2, the integration time is reduced compared to the standard.

The limit velocity v2 <v1 represents the special case for electrode pair 1 and 3 shown in diagram 4 and described above instead of 1 and 2. The disturbance can be completely eliminated under the assumptions described above and taking advantage of the total integration time using the electrode pairs 1 and 3 because the occurrence in field clocks 1 and 2 has no phase shift. Alternatively, it is also possible to compensate with the electrode pairs 1 and 2 and / or 2 and 3, which are only half as far apart, if the integration time is shortened accordingly (analogously to v1 for electrode pair 1 and 3).

At the limit velocity v3 <v2, the disturbance requires exactly two measurement periods from the first to the third electrode pair and a measurement period to the second electrode pair lying in the middle. The compensation can be carried out in measurement period 2 between the second and first electrode pair and in measurement period 3 between the third and second electrode pair (analogously to v2 for electrode pair 1 and 3).

At the limit speed v4 <v3, compensation over the entire integration range is possible if the next but one field clock at electrode axis 3 is used to compensate for the disturbance at electrode axis 1 (and vice versa), or compensation in a partial area of the integration window is possible when considering successive field clocks on electrode axis 1 and 2 and / or electrode axis 2 and 3. Depending on the duration of the preservation of the correlation, the reference to the next but one or next field clock may prove to be advantageous. With knowledge of the correlation duration, it is also possible to provide the compensations with the next or the next field clock with correspondingly different weighting.

QUOTES INCLUDE IN THE DESCRIPTION

This list of the documents listed by the applicant has been generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Cited patent literature

• DE 102006014679 A1 [0003]
• DE 2950084 A1 [0004]
• DE 2226356 [0005]
• JP 54139756 [0006]
• JP 56135118 [0007]
• JP 5066138 [0008]
• US 5398553 [0009]
• EP 0305609 [0010]
• EP 2272790 B1 [0045]
• DE 102013103211 [0048, 0075, 0086]

Claims (13)

Arrangement for determining a flow-related process variable of a measuring medium by at least one first measuring tube, in particular for determining the flow rate of a measuring medium and / or the volume flow of the measuring medium, comprising A) at least the first measuring tube with a measuring tube axis, B) at least two coil systems diametrically opposed to the measuring tube for forming a magnetic field, C) a first pair of measuring electrodes, with two measuring electrodes diametrically opposite each other on the measuring tube, which form a first connecting axis, which runs substantially perpendicular to the magnetic field, and D) at least one second pair of measuring electrodes with two diametrically diametrically opposed on the first or on one second measuring tube opposite measuring electrodes, which form a second connection axis, which is arranged substantially along the Meßrohrachse shifted parallel to the first connection axis; E) a control and / or regulating unit, for generating a magnetic field of alternating polarity through the coil systems, F) a circuit arrangement comprising an evaluation unit for determining a flow rate and / or the volume flow of a measuring medium, characterized in that the measuring electrode pairs connected to the evaluation unit are that a tapping of measuring voltages from the first and from the second pair of measuring electrodes takes place and wherein the circuit arrangement is designed to carry out the following method steps: I. Conductive tapping of measuring voltages in the form of signal voltages of each of the measuring electrode pairs; II.I determination of the phase position of at least the interference component of the measurement signals at different pairs of measuring electrodes or one of the phase-dependent size or II.II specification of a reference value for determining the phase position and / or the phase size of at least the noise component of the measured voltages at different pairs of measuring electrodes relative to each other; III. Signal processing of the measured voltages as a function of a determined or predetermined phase offset, wherein the determination of an uncorrected measurement signal from the measured voltages, wherein III.I a correction of the measurement signal with the determined in step II.I phase or the phase-dependent size or with the benchmark given in step II.II; and IV output the corrected measurement signal. Arrangement according to claim 1, characterized in that the circuit arrangement is designed to carry out a step V, in particular following step III or IV, wherein step V an adjustment of the time course of the magnetic field generation in dependence of the predetermined or determined in the process phase of the interference different electrode axes relative to each other. Arrangement according to claim 1 or 2, characterized in that the arrangement is realized in a flow meter with at least two measuring electrode pairs. Arrangement according to one of the preceding claims, characterized in that the arrangement, in particular the magnetic-inductive flowmeter, a third measuring electrode pair having two diametrically opposed measuring electrodes, which form a third connecting axis, which is arranged substantially along the measuring tube axis parallel to the first connecting axis , Arrangement according to one of the preceding claims, characterized in that the second and / or the third measuring electrode pair has measuring electrodes which differ in their geometric shape and / or in their material composition from the first measuring electrode pair. Arrangement according to one of the preceding claims, characterized in that the circuit arrangement carries out the following further method steps: • amplification and / or impedance conversion of a measuring voltage of each individual measuring electrode pair • offset correction and possibly further amplification of the measuring voltage of each individual measuring electrode pair • analog-to-digital conversion of the measuring voltage each individual pair of measuring electrodes; and determining a flow rate and / or a volume flow on the basis of the noise margin optimized measurement voltage or voltage, wherein the offset correction and possibly amplification of the measurement voltage, both AC and DC voltage components of the measured voltages are added together and wherein the circuit arrangement, an optimization of the signal to noise ratio Measuring voltages of each individual pair of measuring electrodes takes place. Arrangement according to one of the preceding claims, characterized in that the Determining the phase position and / or the phase size by cross-correlation based on a plurality of previous individual measured values of the measurement signal. Arrangement according to one of the preceding claims, characterized in that the determination of the phase position and / or the phase size is carried out by iterative calculation on the basis of a plurality of preceding individual measured values of the measuring signal. Arrangement according to one of the preceding claims, characterized in that the arrangement has two operating modes, wherein a first operating mode comprises a conventional measuring signal processing method VI and wherein a second operating mode comprises a measuring signal processing with the steps I-III.I, wherein during or after the determination of a Measured value after step IV or after determination of the phase position and / or the phase size of at least the interference component of the measured voltages after step II.I a decision value is determined based on which is selected between the two operating modes. Arrangement according to one of the preceding claims, characterized in that the guide value is derived from an expected propagation speed of the disturbance and the known axis distance. Arrangement according to one of the preceding claims, characterized in that in step V, an adaptation of the excitation of the magnet system in dependence on the phase position determined according to step II.I or the predetermined reference value of the phase position after step II.II. Method for determining a flow-related process variable of a measuring medium by at least one first measuring tube, in particular for determining the flow rate of a measuring medium and / or the volume flow of the measured medium by means of an arrangement comprising at least a first magnetic-inductive flow meter with a first measuring electrode pair and at least one second electromagnetic flowmeter with a second measuring electrode pair or the first magnetic-inductive flowmeter, which has at least one second measuring electrode pair the method being characterized by the following steps: I. Conductive tapping of measurement voltages in the form of signal voltages of each of the pairs of measuring electrodes; II. Determination of the phase position and / or the phase size of the measuring voltages for the detection and compensation of measurement errors caused by disturbances, in particular by solids, in the measuring medium. A method according to claim 12, characterized in that the method comprises a further step in which an adaptation of the measuring period or the switching frequency of the magnetic field in dependence of the determined or predetermined phase position of the interference signals at different electrode axes relative to each other.

Images