MXPA99009800A - Electromagnetic flowmeter - Google Patents

Abstract

A method of operating an electromagnetic flowmeter comprising a flow measurement duct, means for generating a magnetic field in fluid flowing in the duct across the direction of flow, and means for measuring a voltage thereby induced in the fluid as indicative of the flow, the method comprising measuring an output of said measuring means in the absence of said magnetic field and determining therefrom the presence or absence of fluid in the duct.

Description

MAGNETIC ELECTRO FLOW METER The present invention relates to an electromagnetic flow meter, in particular to a battery-powered electromagnetic flow meter, mainly but not exclusively designed for domestic applications. An electromagnetic flow meter is used to measure the flow velocity of a conductive medium, such as water, through a flow tube. An electromagnetic flow meter, such as that described, for example, in U.S. Patent Serial No. 2 081 449, comprises a magnetic circuit for developing a magnetic field vector in a plane normal to the direction of the magnetic field. Fluid flow in the flow tube. As the fluid flows through the fl ow tube, a voltage is introduced into the fluid subject to magnetic flux and in a direction orthogonal to the direction of fluid flow and magnetic flux. The induced voltage of V is detected by a pair of electrodes placed in the fl ow tube, where Vj is related to the magnetic field strength ß and to the velocity v of the fluid flowing in the flow tube by the expression V. = B x l. v. k where / is the separation of the electrodes and k is a sensitivity factor that depends on the geometry of the flow tube. By measuring the magnitude and polarity of the induced voltage Vj, the magnitude and direction of the flow velocity of the fluid through the flow tube and hence the flow velocity can be calculated.

Conventionally, various combinations of permanent magnets and electromagnetic systems are used to create either a constant or alternating magnetic field of known magnitude. Said electromagnetic flow meters have high electrical energy requirements for reproduction of a required vector of high magnetic field through the flow tube, and consequently employ an excitation circuit driven by a main system. Thus, an external power supply is required, and the cost of operating such an electromagnetic flow meter is significant over time. The preferred embodiment of the present invention seeks to provide an electromagnetic flow meter with a very low energy consumption, which can, if desired, be driven by an internal battery for many years (for example 8 to 10). A meter of this type is particularly suitable for installation in a large number of domestic water supply networks, and only requires periodic reading of the water consumed either by an on-site visit or preferably by probing or data loading. remote A first aspect of the present invention provides a method of operating an electromagnetic flow meter comprising a flow measuring duct, means for generating a magnetic field in fluid flowing in the duct through the direction of flow, and means to measure such an induced voltage in the fluid as an indication of the flow, the method comprises measuring an output of said measurement means in the absence of said magnetic field and determining therefrom the presence or absence of fluid in the duct. This aspect of the present invention extends to an electromagnetic flow meter comprising a flow measuring duct, generating means for generating a magnetic field through the fluid flowing in the duct, measuring means for measuring a voltage as induced in the fluid and to derive a measurement of the flow thereof and characterized by means for detecting the presence or absence of fluid in the duct from an outlet of the measuring means when the generating means are inactive. Preferably, the measuring means is adapted to identify noise which means that the duct is empty of said fluid. The detection of flow voids (empty tube events), which is generally necessary to meet the requirements of public water services, it can be achieved then in an efficient way as regards energy. The measuring means can be arranged to measure said voltage for short periods of time separated by relatively long intervals, the frequency of operation of said periods is such that the flow measured by the measuring means during said periods is representative of the flow during the relatively long intervals. In the example of a domestic battery-operated meter, this can substantially reduce the meter's energy consumption. Additionally, the measurement means may be configured to measure said voltage during some of the aforementioned periods when the generating means are active, to identify said noise. The electromagnetic flow meter may further comprise means for varying the duration of such intervals in accordance with the measured flow rate, and / or in accordance with the variation in the measured flow rate. Thus, in another aspect of the present invention, an electromagnetic flow meter comprises an electromagnetic flow meter comprising a flow measuring duct, means for generating a magnetic field through the fluid flowing in the duct, means for deriving a flow measurement of such a voltage induced in the fluid, and characterized in that the generating means are configured to generate the magnetic field for short periods of time, separated by relatively long intervals, the frequency of occurrence of said periods is such that the flow measured by the measuring means during such periods is representative of the flow during relatively long intervals, the flow meter further comprises means for varying the duration of such intervals in accordance with the measured flow rate, and / or in accordance with the variation in the measured flow velocity. Preferably, the measurement means is configured to derive a series of voltage-spaced flow measurement signals thus induced in the fluid flow, each signal comprising a component representing the flow and a variable direct-current component unrelated to the flow. In addition, the electromagnetic fl ow meter additionally comprises means for determining the direct current component of said signal and for adjusting the direct current level of a subsequent signal in response thereto. This can ensure that the displacement in the direct current level of the flow measurement signal due to the electrochemical effects is corrected so that the dynamic input range of the processing circuits of the subsequent signal is not exceeded. Otherwise, the signal would be fixed, causing errors in the measurement. Preferably, the determination means predicts the direct current component of the subsequent signal of a plurality of values of the direct current component obtained from the previous signal. Thus, in a further aspect of the present invention, an electromagnetic flow meter comprises a flow measuring duct, generating means for generating a magnetic field through the fluid flowing in the duct, means for deriving a series of signals. for measuring voltages spaced in time from voltages thus induced in the fluid flow, the signal comprises a representative component of the flow and a variable direct current component unrelated to the flow, and means for determining the direct current component of said signal and to adjust the direct current level of a subsequent signal in response thereto, the determining means predicts the direct current component of the subsequent signal of a plurality of values of the direct current component obtained from the previous signal. Preferably, the first flow measurement signal comprises a plurality of pulses, the determination means predict the direct current component of the subsequent signal by applying an algorithm to values of the DC component obtained each from the respective pulse. Conventionally magnetic circuits of electromagnetic flow meters are formed of mild steels of high impurity, which contain a ferrite structure contaminated relatively with impurities. However, these materials possess extremely bad magnetic properties, such as incremental and low initial permeabilities at high and low flux density levels respectively, a low or undefined residual flux density, a high or undefined coercivity, a random crystalline texture or underdeveloped, a loss of energy and anisotropy of random or undefined permeability, and high energy losses at low excitation and induction frequencies. The energy loss is mainly composed of three components: (1) current loss that depends on the excitation frequency, the electrical resistivity, the peak flux density and the thickness of the magnetized body material; (2) losses due to hysteresis that depends on the composition, history of the processing, conditioning and metallurgical processing of the material, the composition, the previous and subsequent processing of the alloy, the heat treatment and the enameling of the material that collectively infl uence the levels of impurities and the mechanism of precipitation and contami ination of the metallic structure. The main factors associated with the losses by hysteresis are the size of the ferrite grain, the alignment and the condition of the ferrite grains, the precipitant and contaminants within the grains and grain boundaries, the alignment of the recrystallized grains , and the degree of the grains with respect to the thickness of the material; (3) parasitic current losses that can be attributed to many factors including the non-sinusoidal movement of the domi- nation wall in the material and the damping of the domain wall. In a preferred embodiment of the present invention, the generating means comprises first and second pole pieces for directing the magnetic field through the measuring duct, the pole pieces are of a material exhibiting a preferential crystalline anisotropy. The crystalline anisotropy can be substantially aligned along the long axis of said pole pieces. This can allow an efficient magnetic circuit with low losses to be provided and which is capable of being energized by a battery on board for a considerable period of time. Thus, in another aspect of the present invention, an electromagnetic flow meter comprises a flow measuring duct, generating means for generating a magnetic field through the fluid flowing in the duct, means for deriving a flow measurement. of a voltage thus induced in the fluid, and characterized in that the generating means comprise pole pieces to direct the magnetic field through the measuring duct, the pole pieces are of a material exhibiting a preferential crystalline anisotropy. This aspect of the present invention extends to a method of manufacturing pole pieces for use in a magnetic circuit of an electromagnetic flow meter comprising the steps of forming the pole pieces of magnetically soft material, and enamelling the pole pieces. While a magnetic field is applied simultaneously to the pole pieces to induce a preferential crystalline anisotropy in the pole pieces. The enameling is preferably carried out in a decarburizing atmosphere (preferably humid). The enameling can be carried out at a temperature of at least 780 ° C for at least two hours. The pole pieces are preferably of a magnetically soft material, preferably containing 0.03% by weight or less of carbon. The generating means may additionally comprise a magnetizing coil positioned on the first pole piece only, and a housing containing the pole pieces and the coil and through which the flow measuring duct passes, the pole pieces and the housing they form a magnetic circuit. Preferably, the generating means is configured to generate the magnetic field in alternate directions, additionally the flow meter comprises means for suppressing the noise in a signal representative of the electromagnetically induced flux by a change in the direction of the magnetic field. This can moderate the rate of change of the magnetic field when it changes direction, reducing the induced noise that might otherwise overwhelm the flow signal detection and processing circuits. Alternatively or additionally, it suppresses the effect of any undulation in the electric current produced by the magnetic field. Thus, in a further aspect of the present invention, an electromagnetic flow meter comprises a flow measuring duct, generating means for generating a magnetic field through fluid flowing in the duct, means for deriving a fl ow measurement of a voltage thus induced in the fluid, characterized in that the generating means are configured to generate the magnetic field in alternate directions, and comprise means for suppressing noise in a signal representative of electromagnetically induced flux by a change in direction of the magnetic field. Preferably, suppression means comprise a conductive medium in which parasitic currents are generated, said eddy currents moderating the rate of change of the magnetic field. The conductor member comprises a disk of diameter substantially equal to that of the coil, and the disk may be at one axial end of the coil facing the second pole piece. The conductor member can surround said pole piece. Thus, the suppression means may comprise a conductive ring surrounding a pole piece adjacent to the flow measuring duct. Preferably, the flow measurement duct comprises a smaller cross-sectional flow-rate measurement portion than another portion thereof, said portions being joined by a converging or diverging section of the duct, the walls of which are defined by inverse transition curves with a point of inflection between them. Preferably, the measurement section has at least one pair of opposed parallel flat walls. Thus, another aspect of the present invention provides an electromagnetic flow meter comprising a fluid flow duct with a cross-sectional flow measurement portion smaller than another portion thereof, said portions being joined by a cross section. convergent or divergent of the pipeline, whose walls are defined by reverse transition curves with a point of inflection between them. Alternatively, the flow measurement duct comprises a flow measurement portion of a smaller cross section than another portion thereof, said portions being joined by a converging or diverging section of the duct, the walls of which at any location axially of the ducts. same have an inclined angle of no more than nine degrees, the walls of the converging or diverging section are defined by one or more transition curves. Preferably the measurement section is a substantially rectangular cross section, and, if so, the other mentioned portion may be of a circular cross section. Thus, in a further aspect of the present invention there is provided an electromagnetic fl ow meter comprising a fluid flow duct with a flow measurement portion of a smaller cross-sectional area than that of another portion thereof, the mentioned portions are joined by a converging or diverging section of the duct, whose walls at any location axially thereof have an included angle of no more than nine degrees, the walls of the converging or diverging section being defined by one or more transition curves. This can provide an efficient passage of the fluid through the measuring portion of the meter. Preferably, the flow meter comprises means for receiving a battery to power the generating means, and, if so, it can then comprise a battery of this type. The preferred embodiment of the present invention will now be described, simply by way of example only, with reference to the accompanying drawings, in which: Figure 1 shows an exploded view of an electromagnetic flow meter according to an embodiment of the present invention; Figure 2 shows an exploded view of the dynamic flow components of the electromagnetic flow meter shown in Figure 1 in greater detail; Fig. 3A shows a horizontal cross-sectional view of the flow tube of the electromagnetic flow meter shown in Fig. 1; Figure 3B is a simplified vertical section of the flow tube, in line B-B of Figure 3A; Figures 3C, 3D and 3E are cuts through the flow tube in lines 3C, 3 D and 3E of Figure 3 B. Figure 4 shows a cross-sectional view of a perforated end shown in Figure 1; Figure 5 shows an exploded view of the magnetic circuit of the electromagnetic flow meter shown in Figure 1 in greater detail; Figure 6 shows an exploded view of the upper portion of the electromagnetic flow meter shown in Figure 1 in greater detail; Figure 7 shows an alternative form of the magnetic circuit and the energizing coil of the electromagnetic flow meter shown in Figure 1; Figure 8 shows a circuit diagram of the flow meter; Figure 9 shows a circuit diagram of an analog-to-digital converter of the electromagnetic flow meter circuit shown in Figure 8. Figure 10 shows waveforms used in the flow meter; and Figure 11, schematically shows the magnetic field generating circuit of the flow meter. Referring now to Figure 1, an electromagnetic flow meter 10, in accordance with a preferred embodiment of the present invention, comprises a meter body 12 and a meter cover 14 of plastic materials. As described below, the flow meter operates at low levels of magnetic field and induced voltage, and is then susceptible to external sources of interference. For the purpose of protecting the external interference electromagnetic flow meter, the meter body 12 and the meter cover 14 can be covered by plasma ionization / discharge, controlled vapor deposition, magnetron electronic deposition or spray coating. to provide an EMC / R FI database. The external surfaces of the body of the meter 12 and the cover of the meter 14 can be treated to make them impermeable to steam in order to prevent the entry of water vapor.

A flow tube 16 (Figure 2) through which, in use flowing fluid conductor is placed in the body of the meter 12. The flow tube 16 includes a pair of non-magnetic, non-permeable flanges 18 at the ends of the tube. inlet and outlet of the flow tube 16 to allow the flow tube to be connected to the body of the meter 12 by fixing plates 19. The seals 20, for example, O-rings, are placed between the flow tube 16 and the body of the meter 12 to prevent leakage of fluid from the joint formed between the flow tube 16 and the body of the meter 12. As shown in Figures 3A to 3E, the flow tube 16 comprises a non-magnetic, non-permeable tapered tube. , non-conductive, for example of preferably treated plastic material to make it impervious to infiltration of water under pressure. The tube has a cross section at its inlet and outlet ends 24, joined by contraction (converging) and diffusion (diverging) portions to an intermediate portion 26 of rectangular cross-section through which the velocity of the fluid flow is determined. through the meter. The portion 26 of the flow tube has a rectangular cross section which is referred to herein as the measuring duct. The flow tube 16 is shaped to achieve a uniform flow profile through the metering duct for a range of fluid flow velocities in the flow tube 16 with minimal variation in fluid pressure leaving the flow tube at outlet 24. The wide-aspect ratio (aspect) of duct 26 can be adjusted to achieve the dynamic flow characteristics desired for the electromagnetic measurement of the flow velocity, but will generally be in the range of 1.5. 23. In this preferred embodiment, the flow meter is designed to withstand a constant continuous working pressure of 16 bar, with a pressure drop across the meter of less than 0.5 bar at a flow rate of 3000 liters per hour. More specifically, the rectangular section of the measuring duct 26 has a cross section of about one third of the area of the inlet and outlet portions 24. This reduced area produces a high-speed relative flow which improves the sensitivity of the meter. The rectangular shape of the measuring duct also promotes a flow regime that is relatively uniform across its width (apart from boundary layer effects) for a wide range of flow rates that lead to a meter calibration characteristic linear With reference now to Fig. 3C, the upper and lower sides of the rectangular section 26 are wide strings of the flow tube of circular section (ie, close to its diameter) at the inlet and outlet ends 24. Consequently, almost all the reduction and the subsequent increase in The flow area is achieved by joining the top and bottom of the section as shown in Figures 3D and 3E. This union is carried out by inverse transition curves as shown in Figure 3B, that is, the degree of convergence or divergence is initially very small at station C, it increases to a maximum at a point of reflection at the center of the convergent or divergent section (approximately at station D) and then decreases from n ew to a small value at the end of the section, station E. The maximum degree of convergence or divergence at the inflection point is such that the included angle of the walls (or more precisely, of the tangents thereto) is less than or equal to g. The union of the vertical sides of the section is much less pronounced. For these sides, although transition curves are preferable, it may be sufficient to simply adopt a straight line junction. Of course, if a flow measurement section is selected that differs markedly from the diameter of the inlet and outlet sections in the vertical and horizontal dimensions, the transition curves must be adapted for the sides as well as for the top and bottom. from the section . A divergent flow tube as described can prevent excessive cavitation and turbulence in the diverging section. If the meter is not required to handle bidirectional flow, the converging section can be made shorter but in general it is preferable to make the flow tube symmetrical with respect to its center for ease of assembly as well as for hydrodynamic efficiency. The flow meter is adapted to be interposed between portions upstream and downstream of the flange of a water supply pipe or other fluid line. The flanged metal flanged ends 28 are attached to the meter body 12 by marked end fasteners 30. The perforation of each marked end 28 can be profiled as shown in Figure 4 to form an extension of the diffusion and contraction sections of the flow tube, thus allowing to achieve a lower ratio of section change. This can be particularly useful if the diameter of the tube is significantly larger than the width of the rectangular measurement section 26. Then the diameter can be gradually reduced to that of the flow tube at station C, Figure 3C, before the junction begins to the rectangular section. The marked ends 28 are adapted to be bolted to the flanges of the tube. The seals 32 are positioned between each marked end 28 and the body of the gauge 12 to prevent leakage of fluid from the seal formed between the respective marked end 28 and the body of the gauge 12. A pair of electrode housings 34 are provided in the flow tube 16. With reference to Figures 2 and 3, an electrode 36 is placed in each electrode housing 34 so that the electrodes 36 are positioned orthogonally through the direction of fluid flow in said tube. flow and also orthogonal to the magnetic field. The electrodes may be formed of any suitable non-contaminating conductive material, such as stainless steel, bronze or plated copper. The electrodes 36 can have a rectangular, the optical or circular cross section and have a height substantially equal to the height of the measuring conduit 26. The material from which the electrodes are formed can be subjected to cleaning and surface treatments such as stabilization, to allow the formation of a uniform layer of oxide on the contact surface of the electrodes exposed to the fluid flowing in the flow tube 16. Such surface treatments can reduce the susceptibility of the electrodes to electro-chemical effects such as polarization, and ensure a consistent electrode impedance from meter to meter. The electromagnetic flow meter is powered by a replaceable or rechargeable internal battery (Figure 6) in this case a lithium thionyl chloride cell size D with capacity of 15 amp-hour 84. The battery is housed in a battery compartment 86 on the cover of the meter 14, with its longitudinal axis parallel to the flow tube. The battery compartment is designed to allow access to the battery and easily replace it with the meter in place. Thus, an opening 88 in the battery compartment 86, closed by a flange and waterproof and fracture-proof access seals 90, allows the battery to be removed from the meter. Above the battery compartment is mounted a printed circuit board 92 comprising a coil driver circuit and signal processing circuits (Figures 8, 9 and 11) and a liquid crystal screen 140, visible through a window 146 on the cover of the meter 14. Compared to the electromagnetic flow meters energized by energy from a conventional main source, the available energy for this meter is much smaller, and the magnetic circuit, the excitation circuit of the coil and the The signal processing circuits of the meter are designed to compensate for this. With particular reference to Figure 5, the magnetic circuit is asymmetric with respect to the measuring conduit 26. It comprises upper and lower pole pieces 42, 44 within a cylindrical housing formed by an upper can 48 and a lower can 50. U na Excitation coil (not shown) is wound around a coil 46 positioned around the upper pole piece 42, which is significantly longer than the pole piece 44. The pole pieces 42 and 44 are formed of magnetically soft material. which has superior magnetic characteristics with respect to materials conventionally used in electromagnetic flow meters. The material from which the pole pieces are formed comprises a magnetically soft, magnetically permeable purified material of low low carbon impurity, such as silicon steel. The pole pieces may be machined from a cast or cast rod having a diameter typically between 10 and 20 mm, preferably 12.5 mm or alternatively it may be cast in the desired shape. Table 1 shows the typical impurity concentrations, expressed as weight percentages (% weight) of the pole pieces before any subsequent heat treatment. Table 1 The purified steel composition for the rod from which the pole pieces are formed reduces the detrimental effects of elements such as carbon and manganese and compounds such as carbides, oxides, nitrides and sulfides in the ferrite structure of the fully recrystallized material. The components are heat treated in a decarburizing atmosphere to promote the required recrystallization in the ferrite grain structure, releasing stored plastic deformation energy, which in turn is a precursor to develop the optimum magnetic properties of the material. The homogeneous recrystallization of the material and the uniform development of the recrystallized ferrite grains result in improved operation of the electromagnetic flow meter. After machining the pole pieces of the rod, or melting the pole pieces, the pole pieces are decarburized and enamelled at a temperature between 800 ° C and 825 ° C for a period between 2 and 4 hours in a humid atmosphere of hydrogen and nitrogen, which commonly contain 70-80% hydrogen, at a dew point of 25 ° C to 35 ° C, to reduce carbon and other impurities in the magnetic structure of the metal. The decarburization to the aforementioned dew point moderates the oxidation rate of the low carbon material, produced by the concentrations of silicon and aluminum in the material, thus controlling the thickness of the thin layer of oxide formed on the surface of the material during the enameling. After enamelling, the cooling rate is controlled so that it is no more than a maximum of 60 ° C per hour in the same atmosphere to avoid the formation of any discolored oxide layer on the surfaces of the material. To accelerate the decarburization process, the main material can be mixed with transition metals such as Zr (in the range of 0.01 to 0.23% weight) Nb (in the range of 0.012 to 0.29% weight) and Ti (in the range of 0.01 1 to 0.22% weight) to help the removal of excessive nitrides and carbides. The enameling is preferably magnetic enameling. Specifically, while the temperature is increased to the enameling temperature, a current is caused to flow through a coil surrounding the oven in which the pole pieces are placed, the rod being placed substantially with its axis parallel to that of the coil The uniform magnetic vector provided is maintained during the enameling and cooling of the material, and produces a crystal anisotropy in the structure of the ferrite grain, at the moment with the axis of the pole piece. Typical field strengths are 100 to 10000 A / m including all the material of the component being treated. Table 2 shows the typical concentrations of impurities, expressed as percentages of weight (% weight) of the rod after enameling. Table 2 The pole pieces may have circular, rectangular or other surfaces. It is also preferred that the pole pieces 42 and 44 are machined so as to be concave in a direction orthogonal to the longitudinal axis thereof. This acts to converge the magnetic flux created thereby increasing the magnetic potential difference and the profile gradient of the resulting bi-directional magnetic field created in the direction of fluid flow in the measuring conduit 26. After enameling, the upper pole pieces and below are subjected to various surface treatments to protect them from atmospheric contamination and corrosion and delay magnetic aging. These treatments include galvanic coating and galvano coating and spray lacquering. Each of the pole pieces 42, 44 has an enlarged face portion 43 (Figure 5) which is recorded in a flat settlement 45 in the flow tube. The periphery of each face 43 is formed by a copper ring in which eddy currents are generated to suppress the high frequency wave that may be present in the magnetic field., as will be described hereinafter with reference to Figure 10. The field is generated by the energization of the magnetic circuit by a single excitation coil 82 excited by the exciting circuit of the coil 80 (FIGS. 7, 8). ) positioned around the upper pole piece 42. In the embodiment shown in Fig. 1, the bobbin is carried by the bobbin 46. The bobbin is formed by injection molding a polymer, such as a PPS polymer comprising a polymer. % fiberglass. In this mode, the coil has a height of approximately 37.5 mm and the diameter of the coil flanges is 55.5 mm. The coil has a large current density with respect to its dimensions, to achieve an acceptable high density magnetomotive force. Commonly the coil comprises from 4000 to 4700 turns of copper wire, preferably approximately 4500 turns of wire, with a caliber of 0.335 to 0.355 mm. With the wire wound to the coil so that the outer diameter of the coil is substantially equal to the diameter of the flanges, there are commonly 60 to 63 layers of wire, with 72 to 75 turns per layer. The rolled wire is attached to the coil by an adhesive. As an alternative to winding the coil wire around a coil, the self-adhering wire can be wound directly around the upper pole piece 42 as shown in Figure 7. A further alternative is to pre-form the coil and adhering the coil directly into the upper pole piece 42. As another further alternative, the exciting coil may comprise three separate cascaded concentric coils connected in series. The upper can 48 comprises a cylindrical sheath 52 and an end receiving disc 54 having an edge (not shown). The disc 54 has a diameter that is slightly larger than the internal diameter of the sheath 54; for attaching the disc 54 to the sheath 52 the sheath is elastically deformed radially outward to increase the internal diameter in order to allow the sheath 52 to fit in the disc. Upon relaxing the fume 52, the sheath contracts so that the inner walls of the sheath strongly grip the edge of the disc 54. With the coil 46 disposed on the upper pole piece 42, the upper can 48 is placed on the coil and the pole piece 42. The upper pole piece 42 is attached to the disk 54 by a ferrous fastening screw or fastener 56. The screw or piping may be of the same material as the pole pieces. A rubber spacer washer or nut 58 can be provided to securely secure the spool 46 relative to the upper pole piece 42. A pair of openings 60 (only one shown in FIG. 3) allows the flow pipe 16 passing through the sheath 52. The lower can 50 comprises an end closing disc 62 having an edge 64. The lower can is placed on the lower pole part and the lower pole part is attached to the disc 62 by a by means of a ferrous fixing screw or screw (not shown). Then, the lower can is attached to the upper can, the inner surface of the sleeve 52 securely grasps the edge 64. The inductive cylindrical housing formed by the upper and lower cans thus mechanically locates the upper pole piece, the piece of lower pole and the excitation wind. Figure 7 shows the cylindrical housing located around the upper and lower pole pieces. The upper and lower cans are formed of magnetically soft laminated material, for example, non-oriented, low-silicon electric steel with a caliber of 0.50 to 0.65 mm. The sleeve 52 is formed by embossing the laminate and the discs 54 and 62 are formed by perforating the laminated material. The openings 60 in the outlet 52 are formed in the sheath 52 by drilling holes in the sheath. The material is subjected to heat treatments and finishing similar to those of polo pieces. Tables 3 and 4 show typical impurity concentrations, expressed as percentages by weight (% weight) of the laminated material before and after enameling.

Table 3 Table 4 The can is housed in a plastic box 70 (Fig. 1) which places the assembled magnetic circuit in the box 12, 14. The surface of the box 70 can be treated in the same way to the box 12, 14 to protect the circuit magnetic against external electromagnetic interference. As will be described later with reference to Figures 10 and 11, the coil excitation circuit 80 is adapted to supply direct currents with alternating polarities to the excitation coil 82, thus generating a bidirectional (alternating) magnetic field orthogonally with with respect to the flow of fluid in the flow tube 16. A bidirectional field is necessary to prevent ionization and electrochemical influences from acting on the electrodes 36; with a unidirectional magnetic field in the measuring duct 26, the voltage induced between the electrodes is also unidirectional. As a result, the particles in the fluid are polarized and have to adhere to the surface of the electrodes 36, which can eventually lead to a total or partial blockage of the duct 26, creating measurement errors and premature failure of the meter. Accordingly, the excitation current is switched, thus creating a bidirectional magnetic field. The steady state magnitude of the current is typically between 3 mA and 7 mA. In this mode, the excitation current is supplied for periods of 60 ms at intervals of a few seconds, each 60 ms period consisting of three half cycles of 20 ms. This simple duty cycle helps ensure a prolonged battery life, and the measurement interval is short enough so that the flow measured in successive 60 ms measurement windows is representative of the flow in the window interval. The duration of the measurement cycles (20 ms) is such that the ground interference of 50 Hz can be coupled within the measurement frequency spectrum and therefore can be rejected in the successive filtering and measuring steps. As the switching of the excitation current is carried out rapidly, the rate of change resulting from the magnetic field can cause a large transient induced voltage between the electrodes 36. This large voltage would otherwise saturate the signal processing circuits of the meter. , it is excluded by taking samples of the induced voltage signal only after the transient voltage has decreased. In order to dampen the rate of change of the magnetic field especially in the region of the upper pole piece 42, a conductive disk 49, preferably formed of aluminum or copper, is placed on the end of the coil that faces the the lower pole piece 44. The disk commonly has a uniform thickness of between 50 and 100 microns, in this mode 76 microns. A landing point is connected to the disk. The rapidly changing magnetic field induces eddy currents in the disk 49, which create an opposite field in the disk, dampening the net change rate of the magnetic field at the end of the upper pole piece 42, thus reducing the noise created by the change of direction of the field. In operation, the alternating magnetic flux passes around the magnetic circuit constituted by the upper pole piece 42, the fluid in the measuring duct 26, the lower pole piece 44, the disk 62, the sheath 52 and the disk 54. Depending on the excitation current and the geometry of the magnetic circuit, the magnetic field is in the range of 2.5mT to 5.0 mT, commonly about 4.5 mT. Figure 10 illustrates the waveform of the coil excitation signal supplied by the coil exciter circuit 80 shown in Figures 8 and (in greater detail) 1 1. The excitation circuit is controlled by a clock signal provided from a supervisor microprocessor 90 (Figure 8). At intervals of a few seconds, the coil is energized for three consecutive half cycles of 20 ms each. At the start of the first half cycle 150, a voltage pulse 152 of 12 volts and 2 ms of duration is applied from an accelerator circuit 1 53 (Figure 11). This is immediately followed by a voltage pulse of 0.6 volts 154 of a reducing circuit 155. As the coil is highly inductive, the accelerating voltage produces a peak of transient current 156 followed by a recovery 158 after which the current it is set at a stable value 160. The peak and rebound occupy 3 to 5 ms, the stable current 160 lasts for about 15 to 18 ms, and during this period samples are taken only of the voltage induced through the electrodes 36. as described herein. For a precise flow measurement, it is important that the magnetic field produced by the stable current 160 be substantially constant. The current 160 may have a ripple arising in the reducing circuit; the effect of this in the magnetic field is suppressed by opposite parasitic currents generated in the peripheral copper rings of the faces of the pole pieces 43. In the second half cycle the voltage applied to the coil is switched by a bridge in the form of H 168 (FIG. 11) to provide an acceleration voltage of 12 volts and a reduction voltage of 0.6 volts of opposite polarity to that of the first half cycle. The current waveform in the coil is also reversed. In the third half cycle the applied voltage is switched back to the new one, and the voltage and current are the same as in the first half cycle. In each half cycle, the purpose of the acceleration pulse is to establish a substantial magnetic flux in the correct direction as quickly as possible. In the absence of an acceleration pulse, the highly inductive nature of the circuit would cause the magnetic field to form only slowly at the constant value necessary for the fl ux measurement to be valid. Then, the duration of the cycle would have to be longer, or a higher (constant) excitation voltage would be necessary. This could also cause a greater consumption of battery power. With reference to Figure 11, the coil excitation circuit 80 consists of the accelerator circuit 153 and the reducing circuit 155, which receive a 3.6-volt input via line 163 of the battery and respectively accelerate and reduce that voltage to 12 volts and 0.6 volts, respectively. The circuits 153, 155 are static, except when triggered by a timing circuit 164 under the control of the supervisory microprocessor 90. Each contains voltage stabilizing functions so that the output voltages (and especially those of the uctor network 155) remain constant as the battery ages. The timing circuit supplies the reduction and acceleration voltages to a multiplexer 166 where they are combined sequentially, and applied to the switching bridge of the coil 168 (also controlled by the timing circuit 164) where the polarity is reversed of the second half cycle, and then the three are applied sequentially to the coil 82. A resistor 170 is provided in series with the coil 82 pair to limit the induced current that would otherwise be excessive during the acceleration phase of the excitation cycle. The voltage 132 developed through the resistor 170 (which has a low temperature coefficient and is therefore substantially stable in terms of temperature) is used as a reference signal for an analog-to-digital converter (FIGS. 9) which, as described herein, utilizes the signal to compensate for variations in the magnetic field in the measuring conduit 26. As the fluid flows through the flow tube 16, a voltage is induced in the fluid in the measuring conduit 26 subjected to magnetic flux and is detected by the electrodes 36. For low fluxes, the induced voltage can be as low as 0.3 micro volts. The induced voltage detected by the electrodes 36 is supplied to the signal processing circuits via coaxial cables in the form of a mesh with conductive sleeves filled with graphite or carbon with an elastomer base or intertwined and wrapped shielding layers of R FI / EMC, the which collectively network the susceptibility of the signal to external magnetic and electrical interference. Each cable can be housed alternatively in a respective shielding tube, or both cables can be housed in a single tube. The tube is made of a nickel-based alloy such as Mu metal, or of aluminum surrounded by a nickel-based sheet. The signal output from the electrodes 36 to the signal processing circuits is shown in Figures 8 and 9. These circuits are energized in a synchronous manner with the coil exciter circuit under the control of the microprocessor clock signal. The signal from the electrodes 36 is taken via an RC 100 coupling to a differential preamplifier having a common mode rejection ratio high in the bandwidth of the signal spectrum and a previously established and controllable gain. In order to accommodate the differential signal within the same rail of the power supply, a small common mode voltage Vc is injected in the mode of the center of the RC 100 coupling. After the fixed gain amplification of the differential signal made by the pre-amplifier 1 10, the signal is inserted as an input to a filter low pass and amplifier (LPF and gain) 120, which removes the high frequency noise its perpuesto of the signal and amplifies the signal additionally for compatibility with the limited dynamic range of the analog-to-digital converter 130.

The amplified analog signal is converted to a digital format by the analog-to-digital converter 130 with a resolution of 15 bits by a frequency-to-voltage conversion technique controlled by a microprocessor. An example of the analog-to-digital converter is shown in Figure 9. In the analog to digital converter 130, the amplified and filtered signal is integrated by the integrator 1 31 at pre-determined time intervals with respect to the reference voltage 132. developed in the current imminator resistor of the coil 170. The integration time is set by the control logic circuit 134 under the control of the microprocessor 90. The voltage 132 varies with the steady state (reducing) current that passes to through it. Thus, if the current varies for example due to a change in temperature or battery aging, this is reflected in the reference voltage 132. A change in the coil current causes a change in the magnetic field and by thus a parasitic change in the input of the flow rate signal (voltage) to the analog-to-digital converter 130. By utilizing the voltage 132 as the reference voltage in the integrator 131 a radiometric relationship is provided between a change in the excitation current and the signal conversion, ensuring that the integrity of the flow measurement is maintained. The output of the integrator is a ramp that is fed to a minimum level circuit 133 that discharges the capacitor 135 of the integrator and emits a pulse each time a minimum level voltage is achieved. The resultant pulse train is counted during a time interval established by the control logic circuit 134 to provide a digital conversion of the analog flow signal, which is converted by the microprocessor 90 to flow rate and / or to the total volume of fluid that has passed through the meter, for storage in a non-volatile memory and display as required by the liquid crystal screen 140. The accumulated data may be transmitted to a mobile or centralized database via a medium of interface which may include several radio frequency communication means at an appropriate bandwidth using various modulation methods including FSK, PSK and ASK. The transmission protocols can be selected as appropriate for the particular application. The analog-to-digital converter 130 can only accept a limited dynamic range in its input signal. However, the output of the amplifier / LPF 120 has a direct current level relative to a circuit landing due to the electromechanical effects between the fluid in the measuring conduit 26 and the electrodes 36. To maintain the level of direct current in the acceptable limits, a direct current voltage of the signal is added or subtracted in the LPF / amplifier 120, under the control of the microprocessor 90. The microprocessor applies a non-linear three-point regressive algorithm to the flow signals V0, V1 t V2 measurements during the three successive half cycles of 20 ms of Figure 10. As the direct current voltage level slowly varies its value for 60 ms, a measurement window can be used to predict its V3 value during the next window two seconds later . In this particle algorithm r (others can be used) v3 = V2 + (Vi - V2) 2 / (V? - Vo) The value V3 is supplied to an analog-to-digital converter 1 50 where it is converted to a voltage direct current that is applied to the LPF / amplifier 120 to adjust the direct current level of its output so that it is within the dynamic range of the analog-to-digital converter 130. The above description of the operation of the signal processing circuits It has been within the context of measuring a fluid that flows. It is also necessary to consider two other operating conditions: one when the flow tube is full of fluid but there is no flow and the other when the flow tube has no fluid. The voltage across the electrodes 36 will have one of the three characteristic states; although the transitions between these states is not abrupt, they are sufficiently marked so that a distinction can be made by defective logic algorithms in the microprocessor 90. In the first state, the flow of the fluid is passed through the measuring conduit 26. of the flow signal, the voltage monitored through the electrodes 36 also contains the noise arising from the electro-chemical effects of the fluid, and the noise of the flow. The above represents a feature that varies slowly, and is removed by the three-point non-linear regressive algorithm, and the latter is generally proportional to the flow velocity.

In the second state, the flow tube 16 is filled with stationary fluid, and the monitored voltage is dominated by the aforementioned electro-chemical effect.

The third state is when the measurement duct is empty. Then the electrodes 36 are not in contact with the conductive fluid and the impedance between them is very high. They are also exposed to parasitic electric and magnetic fields that now become signi fi cant. The voltage detected through the electrode 36 varies markedly from measurement to measurement but in an inconsistent and unpredictable manner, and is independent of whether the coil 82 is energized. Based on the nature of the meter readings, three criteria are adopted to determine the filling status of the flow tube as indicated below: (i) The consistency of the meter readings. If the consistency is bad, an empty tube event may already have occurred. (ii) The noise level in the flow tube (tube noise) when the excitation coil is not energized. If the noise is high, an empty tube event may already have occurred. (i ii) Correlation between the tube flow and the registered flow velocity. As the noise is generally proportional to the flow velocity, if the two are out of proportion (ie, the noise at the measured flow rate is significantly higher than it should be) an empty tube event may already have occurred. . It is evident that each or none of these three criteria responds to the transition between the full and empty tube state, ie situations in which the flow tube is only partially full or the fluid flowing contains empty spaces, and any These conditions cause measurement inaccuracies. In order to address these transition characteristics when implementing these criteria, a fuzzy logic technique is employed. In order to test the filling state of the tube, a number of consecutive meter readings are recorded before the test. The meter reading obtained in the absence of an excitation current is also recorded. This is achieved by the microprocessor during one of the 60 ms periods energizing the signal processing circuits but not the coil 82. Typically, this can be done once in every eight of thirty 60ms cycles. The instantaneous flow rate measured in the last test, the consistency of the flow rate, and the tube noise are each treated as independent and logical variables. Based on these meter readings, the logical values are assigned individually to these variables in accordance with the inherent relationship between them. Then, these variables are processed collectively to generate an updated logical function value, which is compared with a series of previously defined fuzzy minimum levels, in order to determine if the measuring tube is light or not. The microprocessor 90 is also used to realize an additional saving of the battery by adjusting the interval between the measurement periods of 60 ms. When the flow rate is high, or is changing significantly from measurement to measurement, the measurement periods are performed at relatively short intervals.

If the measured flow rate is zero or consistent at a relatively low speed, then the interval is extended, for example it is doubled. Each aspect disclosed in the description, and / or the claims and drawings, may be provided independently or in any appropriate combination. In particular, an aspect of a dependent claim can be incorporated into a claim on which it does not depend.

Claims (28)

1. CLAIMS 1. A method for operating an electromagnetic flow meter comprising a flow measuring duct, means for generating a magnetic field in fluid flowing in the duct through the direction of flow, and means for measuring such a voltage induced in the fluid As indicative of the flow, the method comprises measuring a series of pairs of outputs of said measurement means, each pair of outputs comprises an output in the presence of the magnetic field and an output in the absence of said magnetic field, and determining the series measurements of pairs of outputs the presence or absence of fluid in the duct.
2. 2. An electromagnetic flow meter comprising a flow measuring duct, means for generating a magnetic field through fluid flowing in the duct, measuring means for measuring a voltage thus induced in the fluid and for deriving a measurement of flow thereof and characterized by means for detecting the presence or absence of fluid in the duct of a series of output pairs of the measurement means, each pair of outputs comprises an output of the measurement means when the generation means they are active and an exit from the measurement means when the generation means are inactive.
3. 3. A flow meter according to claim 2, wherein the measuring means is adapted to identify noise indicating that the duct is empty of said fluid.
4. 4. A flow meter according to claim 3, wherein the measuring means are configured to measure said voltage for short periods of time separated by relatively long intervals, the frequency of occurrence of said periods is such that the The flow measured by the measurement means during said periods is representative of the flow during the relatively long intervals.
5. 5. A flow meter according to claim 4, wherein the measuring means is configured to measure said voltage during some of said periods when the generating means are inactive, to identify such noise.
6. 6. A flow meter according to claim 5, comprising means for varying the duration of said ranges in accordance with the measured flow rate, and / or in accordance with the variation in the measured flow rate.
7. 7. A flow meter according to any of claims 2 to 6 wherein the measuring means is configured to derive a series of voltage measurement signals spaced in time from voltages thus induced in the fluid flow, each signal comprises a representative component of the flow and a variable direct current component not related to the flow, the flow meter additionally comprises means for determining the direct current component of said signal and for adjusting the direct current level of a subsequent signal mentioned in response to it.
8. 8. A flow meter according to claim 7, wherein the determination means predicts the direct current component of the subsequent signal of a plurality of values of the direct current component obtained from the previous signal.
9. 9. A flow meter according to claim 8, wherein the first flow measurement signal comprises a plurality of pulses, the determination means predicting the direct current component of the subsequent signal when applying an algorithm to component values. of direct current obtained each of the respective respective pulse.
10. 10. A flow meter according to any of claims 2 to 9, wherein the generating means comprises first and second pole pieces for directing the magnetic field through the measuring duct, the pole pieces are of a material that exhibits preferential crystalline anisotropy. eleven .
11. A flow meter in accordance with the claim 10 wherein the crystalline anisotropy is aligned with the direction of the magnetic field in the pole pieces.
12. 12. A flow meter according to claim 10 or 1 wherein the pole pieces are of a magnetically soft material.
13. 13. A flow meter according to claim 12 wherein the pole pieces are of a material containing 0.03% by weight or less of carbon.
14. 14. A flow meter according to any of claims 10 to 13, wherein the generating means further comprises a magnetizing coil placed only on the first pole piece, and a housing containing the pole pieces and the coil and through which the flow measurement duct passes, the pole pieces and the housing form a magnetic circuit.
15. 15. A flow meter according to any of claims 2 to 14, wherein the generating means is configured to generate the magnetic field in alternating directions, the flow meter further comprising means for suppressing noise in a signal representative of the flow induced electromagnetically by a change in direction of the magnetic field.
16. 16. A flow meter according to claim 15, wherein the suppression means comprises a conductive member in which parasitic currents are generated, said eddy currents moderating the rate of change of the magnetic field.
17. 17. A flow meter according to claims 14, 15 and 16 wherein the conductor member comprises a disk of diameter substantially equal to that of the coil.
18. 18. A flow meter according to claim 17 wherein the disk is at an axial end of the coil facing the second pole piece.
19. 19. A flow meter according to claims 16 to 18 wherein the conductor member surrounds a single pole piece.
20. 20. A flow meter according to claims 10 and 16 wherein the suppression means comprises a conductive ring surrounding a pole piece adjacent to the flow measuring duct. twenty-one .
21. A flow meter according to any of claims 2 to 20, wherein said flow measurement duct comprises a cross-sectional flow measurement portion smaller than another portion thereof, said portions being joined by a convergent or divergent section of the duct, whose walls are defined by inverse transition curves with a point of inflection between them.
22. 22. A flow meter according to any of claims 2 to 20, wherein said flow measurement duct comprises a transverse flow measurement portion smaller than another portion thereof, said portions being joined by a Converging or diverging section of the duct, whose walls at any axial location thereof have an included angle of no more than nine degrees, the walls of the converging or diverging section are defined by one or more transition curves.
23. 23. A flow meter according to claim 21 wherein the measuring section has at least one pair of opposed parallel planar walls.
24. 24. A flow meter according to claim 22 wherein the measuring section is of a substantially rectangular cross section.
25. 25. A flow meter according to claim 22 wherein the measuring section is of a substantially circular cross section.
26. 26. A flow meter according to any of claims 2 to 25 comprising means for receiving a battery for feeding the generating means.
27. 27. A flow meter according to claim 26 comprising said battery.
28. 28. An electromagnetic flow meter or a method of operating an electromagnetic flow meter is described herein with reference to any of the accompanying drawings or examples. RESU MEN A method for operating an electromagnetic flow meter comprising a flow measuring duct, means for generating a magnetic field in fluid flowing in the duct through the direction of flow, and means for measuring a voltage so induced. in the fluid as indicative of the flow, the method comprises measuring a leak of said measurement means in the absence of said magnetic field, and determining therefrom the presence or absence of fluid in the duct.