Magnetic flowmeter assembly with burr removal capability

Abstract

A system and related method for correcting for measurement glitches detected by a magnetic flowmeter assembly in determining fluid flow rate is provided. The magnetic flowmeter assembly includes a pair of electrodes that receive instantaneous first and second electrode potentials (voltage potentials) (Ue1 and Ue2) in order to determine an instantaneous induced voltage Ue across the fluid. A signal processor sends a digital signal of the induced voltage (voltage signal Ue) to a microprocessor that processes the signal data and calculates a corresponding instantaneous fluid velocity v. In addition, an instantaneous glitch detection variable Um is calculated based on the electrode potentials (Ue1 and Ue2) in order to detect the presence of measurement glitches in the electrode potentials, with the signal processor providing a digital signal of Um to the microprocessor. Thus, the flowmeter assembly is able to correct for measurement glitches by manipulating the instantaneous voltage signal Ue such that the calculated and displayed fluid flow rate is not skewed by the measurement glitches.

Description

TECHNICAL FIELD

[0001] The present invention relates generally to magnetic flow meters for performing fluid flow measurements, and more particularly, to flow meters capable of correcting fluid flow measurements in the presence of glitch readings. BACKGROUND

[0002] Magnetic flow meters measure the velocity of electrically conductive fluid passing through a conduit by generating a magnetic field and measuring a resulting voltage. These flow meters rely on Faraday's Law, wherein the flow of electrically conductive fluid through a magnetic field induces a voltage signal that is sensed by electrodes, and the sensed voltage is proportional to the velocity of the fluid.

[0003] Magnetic flow meter assemblies are typically either insertion magnetic meters or full bore magnetic meters. Insertion magnetic meters typically include a sensor body placed into the fluid flow and electrodes disposed at a distal end of the sensor body. The sensor can include an electrically conductive coil that generates a magnetic field that, in combination with the fluid flow, creates an electromotive force (voltage) that is then sensed by the electrodes. Full bore magnetic meters typically include a tubular body disposed axially along a fluid conduit, wherein electrodes are molded into the wall of the body and an electrically conductive coil is disposed on the tubular body that generates a magnetic field across the body.

[0004] While these flow meters are generally effective, there are deficiencies. For example, accurate fluid flow velocity measurements depend on several conditions, such as symmetrical fluid contact with the electrodes, or electromagnetic environments that have been accounted for in the accurate operation of the flow meter. However, deviations from such conditions can occur, and thereby result in a measurement "glitch," which is generally due to a sudden unwanted electronic signal that can result in an inaccurate flow measurement. Examples of deviations include asymmetrical fluid contact with the electrodes, thereby affecting the measured resulting voltage signal and corresponding fluid flow velocity calculation. Such inadequate fluid contact with the electrodes can be due to air pockets (bubbles) contained within the fluid flow, or due to the fluid flow splashing around within the conduit, particularly during start-up flow (initial flow through the conduit). Another example of a deviation includes electromagnetic interference (EMI) or from electromagnetic compatibility (EMC) events, wherein external energy sources including radio frequencies can affect the flow meter operation, for example, affecting the circuitry that generates the magnetic field, thereby affecting the calculated fluid velocity. Examples of such EMI and EMC events can be due to the operation of pumps, variable flow drives, and other heavy electrical equipment, and it can also be due to conducted or radiated noise, electronic PCBA, which results in transients in the signal that is proportional to the flow.

[0005] Existing methods for correcting for such measurement glitches include the use of post-measurement flow analysis, where such large outliers in the measurement data can be normalized to reduce the bias from the recorded fluid flow rate. However, even with such normalization or filtering, the resulting fluid flow rate profile can still include substantial errors, as such measurement glitches are still considered in the overall fluid flow rate calculation.

[0006] Accordingly, it should be appreciated that there remains a need for magnetic flowmeter assemblies that address these issues. The present invention fulfills these needs and others. SUMMARY

[0007] Briefly and generally, the present invention provides a system and related method for detecting measurement glitches and correcting for corresponding fluid flow rates measured by a magnetic flowmeter assembly. The magnetic flowmeter assembly includes a pair of electrodes in contact with a fluid and providing a corresponding electrode potential, which enables a voltage signal Ue (induced voltage) and a glitch detection variable Um to be determined. The flowmeter further includes a microprocessor configured to receive digitized signals of Ue and Um, and further configured to identify the presence of a measurement glitch based on a step change to Um. Accordingly, the microprocessor is able to correct the fluid velocity calculation by manipulating the voltage signal Ue over a prescribed duration of time, so as to minimize or eliminate the effects of the measurement glitch. In particular, a method for operating a magnetic flowmeter according to the present invention, the flowmeter configured to measure the velocity of an electrically conductive fluid in a flow path, the method comprising: driving at least one coil assembly with a drive current provided by a coil driver, the at least one coil assembly positioned proximate to the fluid flow path; measuring a voltage value in the fluid flow path via a pair of electrodes, such that the flowmeter is configured to: (a) determine a voltage signal Ue (induced voltage) from the measured voltage value; (b) determine a glitch detection variable Um from the measured voltage value; (c) identify the presence of a measurement glitch based on a step change to Um; and (d) correct the fluid velocity calculation by manipulating the voltage signal Ue over a prescribed duration of time, so as to minimize or eliminate the effects of the measurement glitch. 1 +Ue2) / 2, calculating a reference voltage value Um based on signal voltages Uel and Ue2 from the electrode pair, (b) detecting a glitch in the measured voltage value based on the reference voltage value Um exceeding a prescribed value range, and (c) calculating an induced voltage value Ue = Ue2 - Uel, detecting an outlier voltage value from the measured voltage value based on the glitch; correcting the measured voltage value by negating the outlier voltage value so as to fix the measured voltage value measured prior to the detection of the outlier voltage value, wherein the measured voltage value is fixed for a predetermined time period to enable the voltage value in the fluid flow path to stabilize; determining a corrected fluid velocity measurement from the calculated induced voltage value, the corrected fluid velocity, and the calculated induced voltage value corrected for the outlier voltage value, wherein the reference voltage value Um is monitored in real time while the induced voltage value Ue considered and published for fluid flow velocity calculation is at a prescribed delay so as to ensure that the published voltage and velocity are not skewed due to the detected bias. Preferably, the calculated induced voltage value is determined at a prescribed time after the voltage value is measured, such that the flow meter calculates the corrected fluid velocity measurement over a prescribed time delay. The outlier voltage value can be determined by an analog-to-digital converter.

[0008] More particularly, by way of example and not limitation, the microprocessor removes the voltage signal corresponding to the glitch detection variable from the voltage signals considered for fluid flow velocity calculation, and instead uses the voltage signal just prior to the occurrence of the measured glitch as gap fill. Further, the gap fill voltage signal can be used for a prescribed duration of time to minimize any residual effects on the actual voltage signal by the measured glitch.

[0009] In detailed aspects of the exemplary embodiment, the voltage signal and corresponding fluid flow velocity can be published over a prescribed delay, thereby enabling the microprocessor to detect the measured glitch and apply the correction measure.

[0010] In alternative embodiments, a fast analog-to-digital converter can be used to minimize the need for delayed publication of the voltage signal and calculated fluid flow velocity. The present invention also provides an apparatus for operating a flow meter configured to measure a velocity of a conductive fluid in a flow path, the apparatus comprising a coil assembly circuit configured to generate a magnetic field proximate to the fluid flow path, the coil assembly comprising: at least one coil assembly; a voltage source configured to provide a voltage output; and a coil driver configured to provide a drive current to the at least one coil assembly based on the voltage output; an electrical electrode pair configured to measure a voltage value proximate to the fluid flow path. The apparatus also comprises a computer processor configured to: receive a measured voltage value in the fluid flow path via the electrical electrode pair; correct the measured voltage value by Um = (Ue2 - Uel) / 2, calculating a reference voltage value Um based on signal voltages Uel and Ue2 from the electrode pair, (b) detecting a glitch in the measured voltage value based on the reference voltage value Um exceeding a prescribed value range, and (c) calculating an induced voltage value Ue = Ue2 - Uel, detecting an outlier voltage value from the measured voltage value based on the glitch; correcting the measured voltage value by negating the outlier voltage value so as to fix the measured voltage value measured prior to the detection of the outlier voltage value, wherein the measured voltage value is fixed for a predetermined time period to enable the voltage value in the fluid flow path to stabilize; determining a corrected fluid velocity measurement from the calculated induced voltage value, the corrected fluid velocity, and the calculated induced voltage value corrected for the outlier voltage value, wherein the reference voltage value Um is monitored in real time while the induced voltage value Ue considered and published for fluid flow velocity calculation is at a prescribed delay so as to ensure that the published voltage and velocity are not skewed due to the detected bias. Preferably, the calculated induced voltage value is determined at a prescribed time after the voltage value is measured, such that the flow meter calculates the corrected fluid velocity measurement over a prescribed time delay. The outlier voltage value can be determined by an analog-to-digital converter.1 + Ue2) / 2, calculating a reference voltage value Um based on the signal voltages Uel and Ue2 from the electrode pair; detecting a glitch in the measured voltage value based on the reference voltage value Um exceeding a prescribed value range; detecting an outlier voltage value from the measured voltage value based on the glitch; correcting the measured voltage value by negating the outlier voltage value so as to fix the measured voltage value measured prior to the detection of the outlier voltage value, wherein the measured voltage value is fixed for a predetermined time period to enable the voltage value in the fluid flow path to stabilize; determining a corrected fluid velocity measurement from the calculated induced voltage value, the corrected fluid velocity, and the calculated induced voltage value corrected for the outlier voltage value, wherein the reference voltage value Um is monitored in real time while the induced voltage value Ue considered and issued for fluid flow velocity calculation is placed in a prescribed delay so as to ensure that the issued voltage and velocity are not skewed due to the detected bias. Preferably, the computer processor drives the first and voltage sources to a non-zero potential such that the magnitude of the voltage induced in the first coil assembly is equal to the magnitude of the voltage induced in the second coil assembly. The computer processor can also be configured to monitor zero flow measurements to detect changes in performance of the magnetic coil assembly. The computer processor can also be configured to monitor zero flow measurements to detect changes in grounding of the fluid medium, or to monitor zero flow measurements to detect changes in electrical potential of the electrically conductive fluid.

[0011] For purposes of summarizing the application and the advantages achieved over the prior art, certain objects and advantages of the application have been described herein. It is to be understood that not necessarily all such objects or advantages can be achieved in accordance with any particular embodiment of the application. Thus, for example, those skilled in the art will recognize that the application can be embodied or carried out in a manner that achieves or optimizes one advantage or group of advantages as taught herein without necessarily achieving other advantages as can be taught or suggested herein.

[0012] All of these embodiments are intended to be within the scope of the application disclosed herein. These and other embodiments of the application will become readily apparent to those skilled in the art from the following detailed description of the preferred embodiments, wherein the application is not limited to any particular preferred embodiment disclosed. BRIEF DESCRIPTION OF DRAWINGS

[0013] Embodiments of the application will now be described, by way of example only, with reference to the accompanying drawings:

[0014] Figure 1 is a simplified perspective view of a magnetic flowmeter assembly according to the present application depicting a signal processor and microprocessor receiving input from an electrode pair.

[0015] Figure 2depicts a time varying voltage signal for a stagnant fluid, which depicts an ideal depiction of a voltage signal for a stagnant fluid, and a typical depiction of a voltage signal for a stagnant fluid.

[0016] Figure 3 depicts a time varying voltage signal for a fluid flowing within a magnetic flowmeter, which depicts the voltage potential measured by a pair of electrodes.

[0017] Figure 4 depicts a published voltage signal for a fluid flow affected by the occurrence of a measurement glitch.

[0018] Figure 5 depicts an existing filtering method for correcting a voltage signal affected by the occurrence of a measurement glitch.

[0019] Figure 6 depicts the occurrence of a measurement glitch for a fluid flow, which illustrates the effect on a glitch detection variable, and illustrates the actual voltage signal over a prescribed delay and the corrective action imposed by the flowmeter.

[0020] Figure 7 depicts a signal conditioning path for a flowmeter when using a fast analog to digital converter to detect and correct fluid velocity, thereby minimizing the delay in publishing a voltage signal.

[0021] Figure 8 is a simplified perspective view of a magnetic flowmeter assembly according to the present invention, which includes a bracket coupled to a pair of coils forming a magnetic circuit for a pipe.

[0022] Figure 9 is a simplified perspective view of a magnetic flowmeter assembly of Figure 8 , which further includes a shielded housing and electronics assembly.

[0023] incorporated by reference

[0024] In certain embodiments of the present application, the magnetic flowmeter assembly can be configured as described and claimed in the following co-pending patent applications of the Applicant: 1) U.S. Application No. 16 / 146,090, filed September 28, 2018, entitled “FULL BORE MAGNETIC FLOWMETER ASSEMBLY;” 2) U.S. Application No. 16 / 243,868, filed January 9, 2019, entitled “MAGNETIC FLOWMETER ASSEMBLY HAVING INDEPENDENT COIL DRIVE AND CONTROL SYSTEM;” 3) U.S. Application No. 16 / 243,980, filed January 9, 2019, entitled “MAGNETIC FLOWMETER WITH MEDIA CONDUCTIVITY MEASUREMENT;” 4) U.S. Application No. 16 / 244,060, filed January 9, 2019, entitled “MAGNETIC FLOWMETER ASSEMBLY WITH ZERO-FLOW MEASUREMENT CAPABILITY;” and 5) U.S. Application No. 16 / 271,718, filed February 8, 2019, entitled “FULL BORE MAGNETIC FLOWMETER ASSEMBLY WITH TEMPERATURE SENSING ELEMENT,” which are hereby incorporated by reference for all purposes. DETAILED DESCRIPTION

[0025] Reference is now made to the drawings, and in particular to Figure 1 , illustrating a system and method for correcting measurement glitches detected by a magnetic flowmeter assembly 10 in determining fluid flow velocity. The magnetic flowmeter assembly 10 includes a pair of measurement electrodes (26, 28) that receive instantaneous first and second electrode potentials (voltage potentials), Uel and Ue2, in order to determine an instantaneous induced voltage across the fluid, Ue. A signal processor 34 sends a digital signal of the induced voltage (voltage signal, Ue) to a microprocessor 36, which processes the signal data and calculates a corresponding instantaneous fluid velocity (v), (Ue = Uel - Ue2). In addition, an instantaneous glitch detection variable Um is calculated based on the electrode potentials (Uel and Ue2) (Um = (Uel + Ue2) / 2) in order to detect the presence of measurement glitches in said electrode potentials, where the signal processor 34 will provide a digital signal of Um to the microprocessor 36. Thus, the magnetic flowmeter assembly 10 is able to make corrections in terms of measurement glitches by manipulating the instantaneous voltage signal Ue, such that the calculated and displayed fluid flow velocity is not skewed by said measurement glitches.

[0026] With continued reference to Figure 1 , the magnetic flowmeter assembly 10 has a tubular body 12 (e.g., a pipe) having two opposite ends 14 and 16 aligned along a horizontal axis (A X ) and defining a fluid flow path 24 for the transport of an electrically conductive fluid. The magnetic flowmeter assembly 10 includes a pair of coil assemblies (18, 20) coupled to an intermediate region of the magnetic flowmeter assembly 10 and configured to deliver electrical current received from at least one coil driver. The coil assemblies (18, 20) generate a magnetic field 22 within the fluid flow path 24 of the tubular body 12 via the electrical current delivered in the tubular body 12. The aforementioned pair of measurement electrodes (26, 28) are attached to the tubular body 12 and configured to detect a voltage (Ue) induced by the electrically conductive fluid passing through the magnetic field 22.

[0027] The pair of coil assemblies (18, 20) can be externally coupled to the tubular body 12 aligned along a vertical axis (Az) orthogonal to the longitudinal axis (Ax). The pair of measurement electrodes (26, 28) can be aligned along an axis (Ay) orthogonal to the longitudinal axis (Ax) and the vertical axis (Az) and configured to detect a voltage induced by being in electrical communication with the fluid within the fluid flow path 24. The magnetic flowmeter assembly 10 further includes a plurality of auxiliary electrodes 19 (a, b, c) including a first auxiliary electrode 19 (a) and a second auxiliary electrode 19 (b) disposed upstream of the pair of measurement electrodes (26, 28). The first and second auxiliary electrodes are aligned with the axis (Az) on opposite sides of the pipe such that the axis (Ay) and the axis (Az) are coplanar. A third auxiliary electrode 19 (c) is disposed downstream of the pair of measurement electrodes (26, 28). The pair of measurement electrodes (26, 28) and the auxiliary electrodes (19a, b, c) are each mounted to a corresponding aperture formed in a wall of the tubular body 12.

[0028] The housing is externally coupled to the tubular body 12 and configured to hold at least one processor (signal processor 34) electrically coupled to the electrodes (as in Figure 9The signals from each of the two measuring electrode pairs (26, 28) in exemplary embodiments follow a signal conditioning path consisting of a high sensitivity operational amplifier and an adjustable gain instrumentation amplifier. The voltage signal Ue is the difference between the two measuring electrodes (i.e. Ue = Ue2 [voltage potential of the second electrode] - Uel [voltage potential of the first electrode]). The voltage signal Ue is further processed by a high resolution analog to digital converter (ADC) and a processor (signal processor 34). In exemplary embodiments, a 24-bit ADC can be used. The digitally converted voltage signal is conditioned and processed by a second processor (microprocessor 36) to accurately display and / or provide an output (i.e. digital output, 4-20 mA analog output) proportional to the fluid velocity. In addition, the microprocessor 36 will calculate and display the corresponding fluid velocity.

[0029] In exemplary embodiments, the output of the instrumentation amplifier is the input to the analog to digital converter (ADC). Uel, Ue2 and Ue are converted to digital entities by the ADC and transmitted via high speed SPI (serial peripheral interface) to the microprocessor which calculates the flow rate.

[0030] The magnetic flowmeter assembly 10 relies on Faraday's Law of electromagnetic induction to measure the velocity of a conductive fluid in a tubular body. Specifically, Faraday's Law states that a voltage is induced across any conductor moving at right angles through a magnetic field proportional to the velocity of the conductor.

[0031] Ue is proportional to v x B x L

[0032] Where:

[0033] Ue = induced voltage (i.e. signal voltage)

[0034] v = average velocity of the conductive fluid

[0035] B = magnetic field strength

[0036] L = length of the conductor (i.e. distance between electrodes)

[0037] Alternatively, the average fluid velocity v is proportional to Ue / (B x L)

[0038] As described above, the flow of a conductive liquid through a magnetic field B creates a voltage signal Ue that can be sensed by the measuring electrode pairs (26, 28) that in turn can be used to calculate the velocity v of the conductive fluid. The microprocessor can be configured to publish the induced voltage and fluid velocity at specific time intervals, such as every 200 milliseconds.

[0039] Referring now to Figures 2-3 , a graph depicting when the fluid is stagnant (i.e. v = 0) Figure 2) and when the fluid is flowing ( Figure 3 ). When the fluid is stagnant, ideally, the electrode potentials (Uel and Ue2) are very close to zero, and thus Ue is also close to zero ( Figure 2 ). However, in reality, even when the fluid is stagnant, the electrode potentials are not close to zero, and they move in tandem because they are both in contact with the fluid potential ( Figure 2 ). Once the fluid starts flowing through the fluid flow path, such that the velocity is no longer zero, the two electrode potentials Uel and Ue2 start moving away from each other ( Figure 3 ), thereby creating an induced voltage that is proportional to the flow at each instant on the time axis (tl, t2, t3,...).

[0040] Magnetic flowmeters are typically very accurate (e.g., < 1% measurement error). However, as Faraday's equation illustrates, incidental changes in the magnetic field strength B or induced voltage can lead to significant errors in the fluid velocity measurement. As noted above, conditions such as variations in the fluid flow profile or EMI / EMC interference can directly or indirectly affect the voltage potentials measured by the electrodes.

[0041] Referring now to Figure 4 , the impact on the voltage signal Ue due to the presence of a measurement glitch is shown, in this example, the measurement glitch is caused by fluid splashing around the measurement electrode pair (26, 28) out of sync. More particularly, Figure 4 the graph depicts the difference between Uel and Ue2 over time, as measured. In the exemplary embodiment, measurements are taken every 100 ms. As depicted, there is a significant deviation between the actual Ue measured and the occurrence of the measurement glitch, which will significantly impact the calculated fluid flow velocity. Moreover, as depicted, it is apparent that the measured Ue will oscillate for a period of time before returning to the actual measured value of Ue, thereby extending the impact of the measurement glitch on the fluid flow velocity.

[0042] Referring now to Figure 5 , the existing filtering method is shown, which is used to correct for such measurement glitches by normalizing the average calculated Ue, thereby reducing the impact of the measurement glitch. But as depicted, there is still a significant error margin, and thus the impact of the observed measurement glitch is not minimized. More particularly, Figure 5 the graph depicts the application of filtering methods (e.g., IIR and FIR) to Uel and Ue2 over time.

[0043] Referring now to Figure 6, depicts a graphical representation of the magnetic flowmeter assembly 10 correcting the voltage signal Ue for detected measurement glitches. Upon detecting a measurement glitch, the flowmeter calculates a glitch detection variable Um via another signal conditioning path, where Um is based on the first and second electrode potentials, specifically:

[0044] Um = (Ue1+ Ue2) / 2

[0045] The glitch detection variable Um is continuously calculated and processed by the second signal processor, including as a digital signal transmitted to the microprocessor 36 along with the corresponding determined voltage signal Ue. Using the glitch detection variable Um, the microprocessor is configured to detect the presence of a measurement glitch by monitoring any step change in Um that deviates significantly from previous trends. Upon detecting a step change, the microprocessor is configured to remove the corresponding Ue from being considered for fluid velocity calculations, and instead use the previous Ue (as determined prior to the occurrence of the measurement glitch) to act as a gap fill for the removed Ue. Moreover, given that a measurement glitch can cause the voltage signal to oscillate for multiple measurement intervals, the gap fill Ue can be substantially "frozen" and used for a prescribed number of measurement intervals when calculating fluid flow velocity. Thus, by maintaining the same gap fill Ue for multiple measurement intervals, the impact of measurement glitches on the calculated fluid velocity is minimized or eliminated.

[0046] The criteria for identifying a measurement glitch based on a step change in the glitch detection variable can be based on various data analysis methods. In an exemplary embodiment, the step change can be based on a percentage (%) change in Um from a previous value, where a minimum percentage change is identified as a threshold. Additionally or alternatively, the step change can be identified based on an absolute value change in Um that exceeds a prescribed threshold.

[0047] Referring to Figure 6 and in an exemplary embodiment, the microprocessor monitors the glitch detection variable Um in real-time while the voltage signal Ue being considered and issued for fluid flow velocity calculations is at a prescribed delay, e.g., a prescribed time delay, a prescribed number of measurement intervals, etc. It should be noted that the electrode potentials (Ue1 and Ue2) being considered for the glitch detection variable Um and the voltage signal Ue are the same, i.e., based on the same instant in time, however the two values are issued at different times. Thus, this prescribed delay enables the microprocessor to detect the presence of a measurement glitch and subsequently apply any correction measures as needed in order to ensure that the issued voltage and velocity are not skewed due to any detected deviations. In an exemplary embodiment, the prescribed delay in issuing the voltage signal can be 200 milliseconds.

[0048] Continuing to refer to Figure 6, curve AA represents the burr detection variable Um, where the spike is clearly visible at the 19 ms mark on the time axis. As depicted by curve BB, the voltage signal Ue, which is computed after Um on a 4 ms delay, is significantly off from the actual Ue value based on the previous trend without any correction measure being applied. Thus, curve CC represents the correction measure applied by the microprocessor, where the voltage signal Ue just prior to the measurement of the burr is used for several measurement intervals. Furthermore, as pointed out by curve BB, the voltage signal continues to oscillate slightly until the 35 ms point, but it is apparent that the impact is minimal when compared to the actual voltage signal before and after the measurement of the burr.

[0049] Reference is now made to Figure 7 , depicting an alternative arrangement for a flow meter assembly to correct computed fluid flow velocity, where the above-mentioned delay in publishing the voltage signal Ue can be minimized by using a fast analog-to-digital converter (ADC). Specifically, an operational amplifier pair is used to sum the electrode potentials (Uel and Ue2) such that a fast ADC is configured to compute the burr detection variable Um, and identify a step change based on a prescribed threshold of the detection.

[0050] The exemplary embodiment depicts a full bore magnetic flow meter, but the burr detection system can be incorporated in other types of magnetic flow meters (e.g., insertion mag meters).

[0051] Reference is now made to Figure 8 In the exemplary embodiment, the coil assemblies 18, 20 are coupled to the tubular body (pipe) 12 in the middle thereof. The coil assemblies are mounted on the outside of the pipe, aligned along the axis (Az). More particularly, each coil is held in place by a bracket 21 that circumscribes the tubular body 12. A pole 25 is disposed between the coil 18 and the pipe. The pole is formed of an electrically conductive material (e.g., the same metal as the magnetic bracket), a soft magnetic carbon steel with Fe% > 99.4, and is shaped to conform to the pipe circumference. Non-conductive (air gap) spacers 27 are disposed on opposite ends of the coil. For each coil, a first air gap spacer 27 is sandwiched between the coil and the corresponding pole 25, and a second air gap spacer 27 is sandwiched between the coil and the bracket 21. In each coil, there is a core made of a material with good magnetic properties. These cores transfer the flux lines from the coil into the pole shoe and the magnetic bracket.

[0052] The bracket 21 also serves as a magnetic circuit for the magnetic field generated by the coils (18, 20), which conducts the outwardly propagating magnetic field to add to the inwardly propagating magnetic field. The bracket has a generally octagonal shape, which facilitates assembly and operation of the magnetic flowmeter assembly 10. More particularly, the bracket 21 is formed of two generally c-shaped components 29 that are slidably mated to one another about the pipe to couple to one another. In this manner, the bracket 21 can be used on pipes having different diameters. Attachments (e.g., bolts) couple the coils to the bracket along the axis (Az).

[0053] The magnetic flowmeter assembly 10 is configured to generate a strong alternating magnetic field (flux) B that is uniformly distributed across the cross-section of the pipe. The use of an alternating magnetic field avoids electrode material migration. The configuration of the bracket 21 (e.g., including shape and material) facilitates the resulting magnetic field (flux) B within the tubular body 12. In an exemplary embodiment, the bracket 21 is formed of a "soft" magnetic material (such as soft iron material), which refers to relative permeability, meaning that it does not have a residual magnetization when turned off. Thus, for the magnetic field that propagates outwardly from the pipe, magnetic losses are minimized, and added to the magnetic field that propagates inwardly to the pipe.

[0054] Reference is now made to Figure 9 The magnetic flowmeter assembly 10 also includes a housing 38 that is configured to protect the magnetic field generator (which includes the coils 18, 20 and the bracket 21) from environmental exposure. The magnetic flowmeter assembly 10 also includes an electronics assembly 40 that is attached to the housing of the assembly. The electronics assembly is in electrical communication with the pair of measuring electrodes (26, 28) and the coil assembly (18, 20) to operate the magnetic flowmeter assembly 10. In an exemplary embodiment, the electronics assembly can house components such as a coil driver, operational amplifier, analog-to-digital converter (ADC), processor (e.g., signal processor, microprocessor), among others.

[0055] The application has been described above, according to the presently preferred embodiments, so that an understanding can be had of the application. However, it is not intended that the application be limited to the forms shown, which are to be considered illustrative rather than restrictive. It is contemplated that various other embodiments of the application will be apparent to those skilled in the art from the teachings herein, and it is intended that the application encompass these and other embodiments.

[0056] While the application has been disclosed in connection with only exemplary embodiments thereof, it will be understood that various other embodiments of the application can be provided, including any and all combinations of the features discussed herein, without departing from the scope of the application.

Claims

1. A method for operating a magnetic flowmeter, the flowmeter configured to measure a velocity of a conductive fluid in a flowpath, the method comprising: driving at least one coil assembly proximate to a fluid flowpath with a drive current provided by a coil driver; measuring a voltage value in the fluid flowpath via an electrode pair, such that the flowmeter is configured to: (a) calculating a reference voltage value Um based on signal voltages Uel and Ue2 from the electrode pair by Um = (Uel + Ue2) / 2, 1 + Ue2) / 2, (b) detect a glitch in the measured voltage value based on the reference voltage value Um exceeding a prescribed range of values, and (c) calculate an induced voltage value Ue = Ue2 - Uel, detect an outlier voltage value from the measured voltage value based on the glitch; correct the measured voltage value by negating the outlier voltage value to fix the measured voltage value measured prior to the detection of the outlier voltage value, wherein the measured voltage value is fixed for a predetermined period of time to enable the voltage value in the fluid flowpath to stabilize; determine a corrected fluid velocity measurement from the calculated induced voltage value, the corrected fluid velocity, and the calculated induced voltage value corrected for the outlier voltage value, wherein the reference voltage value Um is monitored in real time while the induced voltage value Ue considered and issued for fluid flow velocity calculations is at a prescribed delay to ensure that the issued voltage and velocity are not skewed due to the detected bias.

2. The method of claim 1, wherein the calculated induced voltage value is determined a prescribed time after measuring the voltage value, such that the flowmeter calculates the corrected fluid velocity measurement over a prescribed time delay.

3. The method of claim 2, wherein the outlier voltage value is determined by an analog-to-digital converter.

4. An apparatus for operating a flowmeter, the flowmeter configured to measure a velocity of a conductive fluid in a flowpath, the apparatus comprising: a coil assembly circuit configured to generate a magnetic field proximate to a fluid flowpath, the coil assembly comprising: at least one coil assembly; a voltage source configured to provide a voltage output; and a coil driver configured to provide a drive current to the at least one coil assembly based on the voltage output; an electrical electrode pair configured to measure a voltage value proximate to the fluid flowpath; and a computer processor configured to: receive a measured voltage value in the fluid flowpath via the electrical electrode pair; by Um = (Ue 1 + Ue2) / 2, a reference voltage value Um is calculated based on the signal voltages Ue1 and Ue2 from the electrode pair; detect a glitch in the measured voltage value based on the reference voltage value Um exceeding a prescribed range of values; detect an outlier voltage value from the measured voltage value based on the glitch; correct the measured voltage value by negating the outlier voltage value to fix the measured voltage value measured prior to the detection of the outlier voltage value, wherein the measured voltage value is fixed for a predetermined period of time to enable the voltage value in the fluid flowpath to stabilize; determining a corrected fluid velocity measurement from the calculated induced voltage value, the corrected fluid velocity, and the calculated induced voltage value corrected for the anomalous voltage value, wherein the reference voltage value Um is monitored in real time while the induced voltage value Ue considered and issued for fluid flow velocity calculations is placed at a defined delay so as to ensure that the issued voltage and velocity are not skewed due to detected deviations.

5. The apparatus of claim 4, wherein the computer processor drives the first and voltage sources to a non-zero potential such that the magnitude of the voltage induced in the first coil assembly is equal to the magnitude of the voltage induced in the second coil assembly.

6. The apparatus of claim 4, wherein the computer processor is further configured to monitor zero flow measurements to detect changes in performance of the magnetic coil assembly.

7. The apparatus of claim 4, wherein the computer processor is further configured to monitor zero flow measurements to detect changes in grounding of the fluid medium.

8. The apparatus of claim 4, wherein the computer processor is further configured to monitor zero flow measurements to detect changes in electrical potential of the electrically conductive fluid.

Images