METHOD FOR DESIGNING A MAGNETIC-INDUCTIVE FLOWMETER

DE502019013782D1Active Publication Date: 2025-08-28ENDRESS HAUSER FLOWTEC AG
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
DE502019013782
Authority / Receiving Office
DE · DE
Patent Type
Patents
Current Assignee / Owner
Priority Date
2018-04-06
Filing Date
2019-03-07
Publication Date
2025-08-28
Estimated Expiration
2039-03-07

AI Technical Summary

Technical Problem

Magnetic-inductive flowmeters are sensitive to rotationally asymmetric flow profiles, leading to measurement errors, especially in large diameter pipes, and existing solutions either introduce pressure loss or are limited in applicability.

Method used

A magnetic-inductive flowmeter design with at least two pairs of electrodes, where the angles α and β are optimized to minimize measurement errors, allowing for electrodes to be short-circuited, eliminating the need for voltage weighting and reducing errors to less than 0.5% in asymmetric flow conditions.

Benefits of technology

The optimized electrode angles and short-circuiting design significantly reduce measurement errors in rotationally asymmetric flow profiles, achieving high accuracy and simplifying the evaluation process for flow velocity and volume flow determination.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader
Need to check novelty before this filing date? Find Prior Art

Description

[0001] The present invention relates to a method for designing a magnetic-inductive flowmeter comprising at least two pairs of electrodes for measuring the flow velocity or volume flow of a medium with a rotationally asymmetric flow profile. Magnetic-inductive flowmeters are used to determine the flow velocity and volume flow of a medium in a measuring tube. A magnetic-inductive flowmeter consists of a magnet system that generates a magnetic field perpendicular to the flow direction of the medium. Individual or multiple coils are typically used for this purpose. To create a predominantly homogeneous magnetic field, additional pole pieces are shaped and attached so that the magnetic field lines run essentially perpendicular to the measuring tube axis across the entire pipe cross-section.A pair of electrodes attached to the outer surface of the measuring tube measures an electrical voltage perpendicular to the flow direction and the magnetic field. This voltage is generated when a conductive medium flows in the direction of flow while a magnetic field is applied. Since the measured voltage depends on the velocity of the flowing medium according to Faraday's law of induction, the flow velocity u and, with the addition of a known pipe cross-section, the volumetric flow rate V can be determined from the voltage.

[0002] Magnetic-inductive flowmeters are sensitive to the flow profile of the medium. Depending on the piping system and the measuring device, measurement errors of several percent can occur. Therefore, a straight pipe, whose length corresponds to at least five to ten times the nominal diameter of the measuring pipe, is typically installed on the inlet-side face. However, there are known applications in which this minimum distance, the so-called inlet section, cannot be maintained. This is the case, for example, when a piping system is located in a very confined space. One solution is provided by the invention disclosed in DE 10 2014113408 A1, in which a narrowing of the pipe diameter leads to flow conditioning, thereby minimizing the influence of the flow profile, allowing the use of a 0-DN inlet section.The disadvantage of this design, however, is that while it can achieve lower sensitivity to rotationally asymmetric flow profiles, it also results in a pressure loss. Furthermore, this design is limited to pipe systems with a diameter of less than 350 mm.

[0003] The sensitivity of the flow measurement to a rotationally asymmetric flow profile depends on the geometry of the measuring tube and the electrodes. Therefore, the influences of the tube and electrode geometry must be taken into account for the correct description of the velocity-dependent induced voltage. These two influences are mathematically represented by a weighting function. GF described.

[0004] The influence of geometry on flow can best be illustrated by the following relationship: U x = ∫ V v x ′ GF x ′ , x dV where for the determination of the voltage U ( x), the flow velocity v(x') and the weight function GF ( x',x) be integrated over the volume of the measuring tube. The weight function GF based on GF(x',x) = B × VG(x',x), with the magnetic field B ( x' ) and a Green's function G, which is given by the electrical boundary conditions. The goal of an optimization procedure is to optimize the geometry of the structure so that in the entire flow profile ∇ × GF = 0. However, this is not possible for a pipe with a single point-shaped electrode pair. One possible solution is to adapt the electrode shape. However, this is impractical and creates new difficulties. Another solution is to use multiple electrode pairs.

[0005] For example, CN 101294832 A discloses a magnetic-inductive flowmeter comprising two pairs of electrodes arranged axially symmetrically in a pipe cross-section to minimize the influence of the flow profile on the determination of the volume flow. The two electrode axes defined by the respective electrode pairs span an angle of approximately 40° in the cross-section of the measuring tube. Another embodiment is shown in DE 10 2015113390 A1, in which a second and third pair of electrodes are arranged on defined electrode axes, which are arranged at an angle of less than or equal to ±45° relative to a first electrode axis oriented perpendicular to the magnetic field.

[0006] EP 0878694 A1 and US Pat. No. 6,094,992 A1 each also disclose a magnetic-inductive flowmeter that, based on the prior art, achieves an improvement in measurement accuracy with errors of less than 1% by using two additional electrode pairs, whose electrode axes each form an angle of approximately 45° to the electrode axis of the conventional electrode pair and the measuring tube axis. This is achieved in particular by individually recording and weighting the potential differences present at the electrodes.

[0007] However, these designs have the disadvantage that, while the measurement accuracy is optimized for small diameters, they do not achieve the desired reduction in measurement errors for commercially available measuring tubes with large nominal diameters. Another disadvantage is that a weighting factor must be considered for each electrode pair, and it is not immediately clear how this should be selected depending on the pipe system or the rotationally asymmetric flow profile.

[0008] Based on the prior art described, the present invention is based on the object of providing a magnetic-inductive flow meter which minimizes the influences of a rotationally asymmetric flow profile when determining the flow velocity and the volume flow.

[0009] The object is achieved according to the invention by the method according to independent claim 1.

[0010] A corresponding magnetic-inductive flowmeter for measuring the flow velocity u or volume flow V̇ of a medium comprises a measuring tube for guiding the medium in a longitudinal direction defined by a measuring tube axis, wherein the measuring tube has an inlet-side end face and an outlet-side end face, which delimit the measuring tube in the longitudinal direction, at least one magnetic field generating device positioned in the direction of a cross section of the measuring tube for generating a magnetic field in the medium that is substantially perpendicular to the longitudinal direction, wherein the magnetic field generating device is characterized by a segment that is adjacent to the measuring tube or has a predetermined minimum distance and couples the magnetic field into the medium, wherein the segment in the cross section of the measuring tube surrounds the measuring tube at a maximum circular arc angle βencompasses, an electrode system with at least two pairs of electrodes which are designed to detect a voltage induced in the medium perpendicular to the magnetic field and to the longitudinal direction, wherein a vertical measuring tube longitudinal plane divides the measuring tube into a first side and a second side, wherein a first electrode of the electrode pair is located on the first side of the measuring tube, wherein a second electrode of the electrode pair is located on the second side, wherein an angle α in the cross-section of the measuring tube, a minimal circular sector is formed in which the electrodes located on each side of the measuring tube are distributed, characterized in that for the angle α it applies that 30° ≤ α ≤ 60° and that for the angle β it applies that 70° ≤ β ≤ 80°.

[0011] The adjustment of the angles α and βThis results in a flow velocity measurement error being minimized in a test measurement with a single disturbance, where the disturbance is generated by an orifice plate or a 90° pipe bend. According to the invention, a pair of angles α and β is determined, the maximum measurement error of which is minimal for all test measurements performed.

[0012] Advantageous embodiments of the invention are the subject of the subclaims.

[0013] According to a further development, insensitivity to a rotationally asymmetric flow profile is given when the Reynolds number of the medium in the measuring tube is greater than or equal to 10,000, in particular greater than or equal to 50,000 and preferably greater than or equal to 100,000.

[0014] According to a further embodiment, the flow meter has three pairs of electrodes.

[0015] The electrode pairs do not necessarily have to be arranged diametrically. The at least two electrode pairs are galvanically or capacitively coupled to the measuring medium. According to a further embodiment, at least two electrodes, in particular all electrodes located on one side of the measuring tube relative to the vertical longitudinal plane of the measuring tube, are short-circuited.

[0016] The technical success of this embodiment of the invention is that it was found that by adjusting the angles α and βSampling the individual potential differences with the addition of empirically determined weighting factors is no longer necessary, and the applied voltage across all electrodes deviates by less than 0.5% from a measured value determined based on a fully developed flow profile in the event of a fault. Weighting of the individual voltage values is therefore unnecessary, which significantly simplifies the evaluation unit for determining the applied voltage and the resulting flow velocity. It is now sufficient to convert the measured voltage value into a flow velocity or volume flow rate using calibration.

[0017] The electrodes are short-circuited, particularly by cables, and preferably by a conductive sheet metal part. This allows for simple and stable installation and also provides a cost-effective alternative to conventional solutions.

[0018] The electrodes are connected to a control and evaluation unit, which uses the voltage induced in the electrodes to provide information about the flow velocity and volume flow in the measuring tube.

[0019] Adjusting the angles α and β can be carried out with a simulation program.

[0020] According to a further embodiment, the magnetic field generating device for generating a magnetic field in the medium that is perpendicular to the longitudinal direction of the measuring tube comprises at least one saddle coil or at least one pole piece with a coil attached thereto.

[0021] Typically, the magnetic field-generating device comprises two diametrically arranged coil systems. In a commercially available magnetic-inductive flowmeter, saddle coils or pole pieces with attached coils are usually installed. These surround the measuring tube and create a region in which the magnetic field in the medium is approximately homogeneous or perpendicular to the abscissa axis.

[0022] According to a further embodiment, the electrodes are arranged axially symmetrically to the vertical longitudinal plane of the measuring tube.

[0023] According to a further embodiment, two adjacent electrodes located on one side of the measuring tube form an angle δ = α / ( N - 1) in the cross-section, where the natural number N corresponds to the number of electrode pairs.

[0024] According to one embodiment, the electrodes are short-circuited with a stamped and bent part.

[0025] The invention is explained in more detail below using exemplary embodiments with reference to the accompanying figures of the drawing. It shows: Fig.1a : a schematic representation of an embodiment of the magnetic-inductive flowmeter to be designed according to the invention, Fig.1b : a schematic representation of a longitudinal section of the measuring tube (1), Fig.2a : two schematic representations of an aperture (B) each with a first aperture orientation (B1) and a second aperture orientation (B2), Fig.2b : two schematic representations of a 90° pipe bend (90°R) with a first pipe bend orientation (R1) and a second pipe bend orientation (R2), Fig.3 : a schematic representation of a design of the measurement configuration in which all electrodes on one side are short-circuited, Fig.4 : an overview of the measurement error depending on the angles α and βfor an 80-DN measuring tube and Fig.5 : the maximum deviation of the measured flow velocity due to a disturbed flow profile as a function of the nominal diameter of the measuring tube for three different electrode and device configurations using three pairs of electrodes.

[0026] The structure and measuring principle of a magnetic-inductive flowmeter is basically known. Fig.1a shows a schematic cross-section (9) of a known magnetic-inductive flowmeter. A medium with electrical conductivity is passed through a measuring tube (1). A magnetic field generating device (5) is mounted in such a way that the magnetic field lines are oriented perpendicular to a longitudinal direction (4) defined by the measuring tube axis. A saddle coil or a pole piece with a coil attached is preferably used as the magnetic field generating device (5). When a magnetic field is applied, a potential distribution is created in the measuring tube (1), which can be tapped using two electrodes attached to the inner wall of the measuring tube (1). These are generally arranged diametrically and form an electrode axis that runs perpendicular to the magnetic field lines and the longitudinal direction (4).Based on the measured voltage, the flow velocity of the medium u can be determined, taking into account the magnetic flux density, and the volume flow V, taking into account the pipe cross-sectional area. To prevent the voltage applied to the electrode system (6) from being dissipated through the pipe, the inner wall is lined with an insulating material. The magnetic field, generated by an electromagnet, for example, is generated by a pulsed direct current of alternating polarity. This ensures a stable zero point and makes the measurement insensitive to influences from multiphase substances, inhomogeneities in the liquid, or low conductivity.

[0027] According to the invention, at least two pairs of electrodes are used to determine the volume flow V̇ used. In a schematic representation, Fig.1a , an example of a magnetic-inductive flowmeter with three pairs of electrodes is shown.

[0028] In addition to the electrode system (6), which serves to measure a potential difference, additional electrodes in the form of medium monitoring or grounding electrodes are often installed in the measuring tube (1). These electrodes are used to measure an electrical reference potential or to detect partially filled measuring tubes (1) or to record the temperature of the medium using a built-in temperature sensor. These are shown in the schematic representation of the Fig.1 not taken into account.

[0029] In each case, a first electrode (6.1) of the electrode pair is located on the first side (I) of the measuring tube (1) and a second electrode (6.2) of the electrode pair is located on the second side (II). The outer electrodes of one side span an angle αin cross-section (9). The other electrodes are distributed within the opened circular segment, preferably on the inner wall of the measuring tube (1).

[0030] In the Fig.1a In the flowmeter shown, the electrode system (6) is in direct contact with the medium; however, as mentioned above, the coupling can also be capacitive.

[0031] A magnetic-inductive flowmeter comprises an inlet-side face (2) and an outlet-side face (3) (see Fig.1b ). In Fig.1b An arrow indicates the flow direction of the medium. A 90° pipe bend (90°R) or an orifice plate (B) mounted on the inlet-side face (2) affects the flow profile of the medium, resulting in a rotationally asymmetric flow profile in the measuring tube (1).

[0032] The magnetic field generating device (5) is typically designed to distribute the magnetic field lines as homogeneously as possible across the cross-section (9) of the measuring tube. This allows measurement errors of less than 0.2% to be achieved, particularly for fully developed flow profiles. In the case of a rotationally asymmetric flow profile, a homogeneous magnetic field can have a detrimental effect on measurement accuracy. This problem can be solved inventively by adapting the magnetic field generating device (5), in particular by adapting the circular arc angle. β .

[0033] By varying the angle β, which describes the extent to which a segment of the magnetic field generating device (5) attached to the measuring tube (1) encompasses the measuring tube (1), a further degree of freedom is obtained for reducing the measurement error. A segment coupling the magnetic field into the medium can comprise a pole piece having two limbs adjacent to a flat surface or two circular arcs attached to its flat surface. Alternatively, a pole piece can also take the shape of a circular arc. In general, a segment coupling the magnetic field into the medium can take on any contour consisting of at least one further sub-segment. For the determination of the maximum circular arc angle β the sub-segments are taken into account which are essentially responsible for coupling the magnetic field into the medium.

[0034] The measurement errors of the flow velocity u or the volume flow V̇ are Δ u = u va − u S u va or Δ V ˙ = V ˙ va − V ˙ S V ˙ va , where the flow rate u va and the volume flow V̇ va in the case of a flow with a fully developed flow profile, and the flow velocity u S and the volume flow V̇ S in the case of a rotationally asymmetric flow profile. The real volume flow V̇ real identical in both cases and in the case of the fully developed flow profile optimally equal to the measured volume flow V̇ va .

[0035] In the simulations, a magnetic-inductive flowmeter with three pairs of electrodes forms the basis for calculating the optimal parameters. The area of the electrodes is larger than point-shaped, but finite in size. The optimization of the angles α and β proceeds in the following steps: In the first step, the angles α and βadjusted to minimize the flow velocity measurement error in test measurements with a single disturbance. The disturbance is generated by an orifice plate (B) or a 90° pipe bend (90°R).

[0036] The aperture (B) covers 10% of the pipe cross-section (9) and has a chord that limits the aperture to the pipe. It assumes a first aperture orientation (B1) or a second aperture orientation (B2), which are rotated by 90° to each other. The chord is oriented perpendicular to the magnetic field in the first aperture orientation (B1) and parallel to the magnetic field in the second aperture orientation (B2). The first aperture orientation (B1) and the second aperture orientation (B2) of an aperture (B) are shown schematically in Fig.2a The black filled circle segment represents the area that blocks part of the cross-sectional area of the measuring tube. In the test measurement, the orifice plate (B) is mounted at a distance of 0-DN from the inlet-side end face (2). Alternatively, a 90° pipe bend (90°R) is mounted at the inlet to the inlet-side end face (2) at a distance of 0-DN, whereby the 90° pipe bend (90°R) assumes a first pipe bend orientation (R1) or a second pipe bend orientation (R2), which are rotated by 90° to each other. The first pipe bend orientation (R1) and the second pipe bend orientation (R2) of a 90° pipe bend (90°R) are shown schematically in Fig.2b The adjustment of the angles α and β is preferably carried out for the two faults with both orientations.

[0037] In the second step, according to the invention, the angle pair is determined whose maximum measurement error is minimal for all test measurements carried out.

[0038] In Fig.4 is an example of the simulated measurement error (Z-axis) for a 500-DN measuring tube (1) depending on the angle α (Y-axis) and angle β (X-axis). Based on the curve, the minimum of the measurement error for a specific disturbance, in this case a pipe bend (90°R) optimized with respect to a first and second pipe bend orientation (R1, R2), is determined. The first pipe bend orientation (R1) is characterized by a pipe axis (11) running perpendicular to the magnetic field and to the longitudinal direction (4), and the second pipe bend orientation (R2) is characterized by a pipe axis (11) running parallel to the magnetic field and perpendicular to the longitudinal direction (4) (see Fig.2b ). This procedure is repeated for all of the above-mentioned disturbances, whereby in the last step the angle pair is determined that has the smallest measurement error with respect to all test measurements. The values for the angles α and β are adjusted until the resulting measurement error for all test measurements is less than 0.5%, preferably less than 0.2%.

[0039] Fig.3 shows a preferred measurement configuration in which the individual electrodes located on one side are short-circuited. The measured voltage corresponds to an average voltage across all electrodes. For example, the electrodes are short-circuited on one side by individual cables or with a conformal sheet metal part and connected to a control and evaluation unit (12).

[0040] Fig.5 shows the maximum measurement error of the measured flow velocity due to a disturbed flow profile as a function of the nominal diameter of the measuring tube (1) for three different electrode and device configurations using three electrode pairs. The maximum measurement error results from the highest measurement error for the angle pair ( α - β ) with respect to all test measurements performed. The triangles show the maximum measurement error for a magnetic-inductive flowmeter with an angle β , which is used, for example, in a Promag W series magnetic-inductive flowmeter from Endress+Hauser. For the angle αAccording to EP 0 878 694 A1, an angle of 45° is assumed. By optimizing the weighting of the determined stress values, a maximum measurement error of 1% is obtained for a measuring tube (1) with 50 DN and a maximum measurement error of 2.25% for a measuring tube (1) with 900 DN. The circles show results of simulations in which the angle β is varied for optimization and the angle α is kept constant at 45°. By using β As an additional optimization parameter, the measurement error can be significantly reduced in an asymmetric flow profile. The squares show results of simulations where, in addition to the variable angle β also the angle α, which impacts the electrode area, is assumed to be variable for optimization. In this case, weighting of the individual voltages applied to the electrode pairs is omitted, which corresponds to short-circuited electrodes. This allows the measurement error caused by a rotationally asymmetric flow profile to be halved again. To determine the optimal angles α and β disturbances comprising an orifice plate (B) with a first orifice plate orientation (B1), an orifice plate (B) with a second orifice plate orientation (B2), a 90° pipe bend (90°R) with a pipe bend orientation (R1) and a 90° pipe bend (90°R) with a pipe bend orientation (R2) were used.

[0041] Based on the optimization procedure described above, a magnetic-inductive flowmeter with three electrode pairs, a 150 DN measuring tube (1) and a medium with a flow velocity of 1 m / s has a measurement error of 0.15% when an orifice plate (B) with orifice orientation (B1) is used and a measurement error of 0.01% when an orifice plate (B) with orifice orientation (B2) is used.

[0042] Based on the optimization procedure described above, a magnetic-inductive flowmeter with three electrode pairs, a 150 DN measuring tube (1) and a medium with a flow velocity of 1 m / s has a measurement error of 0.05% when using a 90° pipe bend (90°R) with pipe bend orientation (R1) and a measurement error of 0.02% when using a 90° pipe bend (90°R) with pipe bend orientation (R2).

[0043] Based on the optimization procedure described above, a magnetic-inductive flowmeter with three electrode pairs, a 300 DN measuring tube (1) and a medium with a flow velocity of 1 m / s has a measurement error of 0.04% when an orifice plate (B) with orifice orientation (B1) is used and a measurement error of 0.21% when an orifice plate (B) with orifice orientation (B2) is used.

[0044] Based on the optimization procedure described above, a magnetic-inductive flowmeter with three electrode pairs, a 300 DN measuring tube (1) and a medium with a flow velocity of 1 m / s has a measurement error of 0.04% when using a 90° pipe bend (90°R) with pipe bend orientation (R1) and a measurement error of 0.15% when using a 90° pipe bend (90°R) with pipe bend orientation (R2).

[0045] Based on the optimization procedure described above, a magnetic-inductive flowmeter with three electrode pairs, a 500 DN measuring tube (1) and a medium with a flow velocity of 1 m / s has a measurement error of 0.07% when an orifice plate (B) with orifice orientation (B1) is used and a measurement error of 0.04% when an orifice plate (B) with orifice orientation (B2) is used.

[0046] Based on the optimization procedure described above, a magnetic-inductive flowmeter with three electrode pairs, a 500 DN measuring tube (1) and a medium with a flow velocity of 1 m / s has a measurement error of 0.18% when using a 90° pipe bend (90°R) with pipe bend orientation (R1) and a measurement error of 0.09% when using a 90° pipe bend (90°R) with pipe bend orientation (R2). Bezugszeichenliste

[0047] 1Measuring tube 2Inlet-side end face 3Outlet-side end face 4Longitudinal direction 5Magnetic field generating device 6Electrode system 6.1First electrode of an electrode pair 6.2Second electrode of an electrode pair 7Vertical measuring tube longitudinal plane 8Radius 9Cross-section 10Abscissa axis 11Pipe axis 12Control and evaluation unit BBaperture B1First aperture orientation B2Second aperture orientation 90°R90°Pipe bend / elbow R1First aperture orientation R2Second aperture orientation IFirst side IISecond side

Claims

1. A method for designing a magneto-inductive flowmeter for measuring the flow velocity u or the volume flow V of a medium, wherein the magneto-inductive flowmeter comprises: - A measuring tube (1), for conducting the medium in a longitudinal direction (4) defined by a measuring tube axis, wherein the measuring tube has an inlet-side end face (2) and an outlet-side end face (3) which delimit the measuring tube (1) in the longitudinal direction (4); - at least one magnetic-field-generating device (5) positioned in the direction of a cross-section (9) of the measuring tube (1) for generating a magnetic field in the medium essentially perpendicular to the longitudinal direction (4); ∘ wherein the magnetic-field-generating device (5) is characterized by a segment which rests on the measuring tube or is spaced at a specified minimum distance from it and couples the magnetic field into the medium, ∘ wherein the segment in the cross-section (9) of the measuring tube (1) encompasses the measuring tube (1) at a maximum arc angle β, - an electrode system (6) with at least two pairs of electrodes which are configured to detect a voltage induced in the medium perpendicular to the magnetic field and to the longitudinal direction (4); ∘ wherein a vertical measuring tube longitudinal plane (7) divides the measuring tube (1) into a first side (I) and a second side (II), ∘ wherein a first electrode (6.1) of each pair of electrodes is located on the first side (I) of the measuring tube, ∘ wherein a second electrode (6.2) of each pair of electrodes is located on the second side (II), ∘ wherein an angle α in the cross-section (9) of the measuring tube creates a minimum circular sector, in which the electrodes 6.1, 6.2) located on each side of the measuring tube (1) are distributed, ∘ wherein the following applies to angle α: 30° ≤ α ≤ 60°, and the following applies to angle 6: 70° ≤ β ≤ 80°, wherein the method comprises the following process steps: - Adjusting angles α and β in such a way that a measurement error of the flow velocity becomes minimal in a test measurement with a single fault, ∘ wherein the fault is generated by an orifice plate (B) or a 90° pipe elbow (90°R); and - working out a pair of angles from angles α and β, the maximum measurement error of which is minimal for all test measurements that are performed.

2. The method as claimed in claim 1, wherein an insensitivity toward a rotationally asymmetrical flow profile with a Reynolds number of the medium in the measuring tube (1) is greater than or equal to 10,000, in particular greater than or equal to 50,000, and preferably greater than or equal to 100,000.

3. The method as claimed in claim 1 or 2, wherein the flowmeter has three pairs of electrodes.

4. The method as claimed in one of claims 1 to 3, wherein at least two electrodes, in particular all electrodes located on each side of the measuring tube relative to the vertical measuring tube longitudinal plane (7), are shortcircuited, in particular with a stamped / bent part.

5. The method as claimed in one of claims 1 to 4, wherein the magnetic-field-generating device (5) for generating a magnetic field in the medium perpendicular to the longitudinal direction (4) of the measuring tube comprises at least one saddle-shaped coil or at least one pole shoe with surface-mounted coil.

6. The method as claimed in one of claims 1 to 5, wherein the electrodes are arranged such that they are axially symmetrical to the vertical measuring tube longitudinal plane (7).

7. The method as claimed in one of claims 1 to 6, wherein two adjacent electrodes located on each side of the measuring tube (1) create an angle δ = α / (N - 1) in the cross-section (9) of the measuring tube, wherein a natural number N is equal to the number of pairs of electrodes.

8. The method as claimed in one of claims 1 to 7, wherein angles α and β are coordinated by means of a simulation process.