CN1058565C - Improved stability coriolis mass flow meter - Google Patents

Abstract

An optimal Coriolis mass flowmeter has improved stability to excitation caused by external disturbances. Its main improvement includes using the modal polarization of the flow pipe to determine the installation position of the sensor device, so that the first bending waveform with the same phase, the first bending waveform with different phases, the first twisted waveform with different phases, the second twisted waveform with different phases, the second twisted waveform The interference of external excitation of one or several waveforms among the two different phase warping waveforms and the third different phase warping waveform is minimized.

Description

Coriolis mass flow meter

This application is a continuation-in-part of our US Patent Application (Serial No. 364,032) filed June 9, 1989.

In the prior art concerning the measurement of the mass flow rate of a flowing substance it is known that flowing a fluid through an oscillating flow conduit induces Coriolis forces which oscillate substantially transverse to and about the direction of fluid flow It is also known that the magnitude of this Coriolis force is related to the mass flow rate of the fluid flowing through the pipe and the angular velocity of the pipe when it oscillates.

A major technical problem associated with past efforts to design and manufacture Coriolis mass flow meters has been the need to accurately measure or precisely control the angular velocity of the oscillating flow conduit so that the measured Coriolis force can be used to calculate the flow rate through the flow conduit. The mass flow rate of the fluid. Even if the angular velocity of the flow conduit can be precisely determined or controlled, another serious problem arises from the precise measurement of the magnitude of the effect caused by the Coriolis force. Part of the reason for this problem is that the magnitude of the resulting Coriolis force is small compared to other forces such as inertia and damping, so the resulting effect caused by the Coriolis force is small. In addition, since the Coriolis force is small, effects caused by external forces such as oscillations caused by, for example, pressure surges in adjacent machinery or fluid piping can lead to erroneous determinations of mass flow rates. These sources of error, such as discontinuities in flow piping, unstable installation of piping, use of piping that lacks mechanically reproducible bending characteristics, etc., often completely mask the effects caused by the resulting Coriolis forces, greatly weakening mass flowmeters. actual use.

In several U.S. patents (named "method and structure of flow measurement", patent No. Re 31450 and publication date is November 29, 1983; title is "mass flow measurement method and device", patent No. 4422338 and publication dated December 27, 1983; and, titled "Parallel Channel Coriolis Mass Flowmeter", patent No. 4491025 and publication date January 1, 1985), the mechanical structure and measurement technology have been disclosed, and Other advantages are disclosed, namely (a) there is no need to measure or control the magnitude of the angular velocity of the oscillating flow conduit of the Coriolis mass flow meter; (b) simultaneously providing the sensitivity and precision necessary for measuring the effects of the Coriolis force; and (c) reduce susceptibility to many errors present in early experimental mass flow meters. The mechanisms disclosed in these patents contain curved continuous flow conduits which are pressure insensitive joints or pipe sections such as bellows, rubber joints or other pressure deformable sections. These flow ducts are rigidly mounted at their inlet and outlet ends, with their curved portions cantilevering out of the brackets. For example, in flowmeters according to any of the aforementioned patents, the communication tubes are welded or brazed to the brackets so that they oscillate in a spring-like manner about some axis which is substantially located in the solid position of the communication tube. Where the mounting points adjoin, or as disclosed in U.S. Patent 4,491,025, these axes are substantially positioned to securely connect the strut means for connecting two or more The pipe is clamped securely.

Constructing the flow channel in this way results in a mechanical state in which the force opposing the Coriolis force generated in the oscillating flow channel is essentially a linear spring force in the flow situation. The Coriolis force, substantially opposed to the linear spring force, that causes an oscillating flow conduit containing a flowing fluid to deflect or twist about axes that are located between and equidistant from those of the flow conduit that can express Those parts of the Coriolis force. The amount of deflection is a function of the resulting Coriolis force and the magnitude of the linear spring force relative to the Coriolis force. In addition, these rigidly mounted, continuous flow conduits are designed so that their resonance frequencies about the axes of oscillation (where the frames or struts are substantially located) are different from, and preferably lower than, the resonance frequencies about other axes frequency, with respect to these axes the Coriolis force acts.

A number of specifically shaped, securely fitted curved flow conduits are disclosed in the prior art. These typically include U-shaped ducts "whose branches are converging, diverging or substantially skewed" (US Patent Re 31450, column 5, lines 10-11). Straight, firmly installed flow ducts are also known in the prior art, and their general operating principle is the same as for curved ducts.

As mentioned above, the Coriolis force is generated when fluid flows through the flow conduits, and these conduits are driven to oscillate. Thus, under flow conditions, the portion of each flow channel that is acted upon by the Coriolis force will deflect (ie, twist) and cause it to move along the flow channel ahead of the flow channel on which the Coriolis force is acting. another part of the The moment at which a first portion of an oscillating flow conduit deflected by the Coriolis force passes a predetermined point in the path of oscillation of the flow conduit and the instant at which a second portion of the conduit passes a predetermined point coincident in that path The time or phase relationship between them is a function of the mass flow rate of the fluid flowing through the flow conduit. This time difference can be used with many types of sensors, including optical sensors (described in detail in U.S. Patent Re 31450), electromagnetic speed sensors (described in detail in U.S. Patents 4,422,338 and 4,491,025), or position or acceleration sensors (also disclosed in U.S. Patent 4422338) to measure. In US Patent 4491025 an embodiment of a parallel route dual flow duct with sensors for measuring the optimum time difference is described. This embodiment provides a Coriolis mass flowmeter configuration which operates in a tuning fork-like manner as described earlier in U.S. Patent Re 31450. Detailed discussions of methods and apparatus for determining mass flow by synthesizing motion sensor signals have occurred in U.S. Patents Re 31450 and 4422338, and in patent application PCT/US88/02360 published on WO89/00679.

In the flowmeter designs described above, the sensors are usually placed at symmetrical locations along the inlet and outlet locations of the flow conduit, which provides the sensitivity required to enable the measurements to be made by the selected sensors and to allow the measured mass flow rate to be measured. Accuracy is within ±0.2%.

About 100,000 Coriolis mass flowmeters have been made using one or more of the inventions in US Patents Re 31450, 4422338 and 4491025, and these flowmeters are widely available in the market. Over a decade of experience in the commercial use of these flowmeters in measuring mass flow with a variety of different fluid products has shown that, in general, end users are satisfied with the sensitivity and accuracy of Improved overall stability, including zero stability, reduces factory maintenance of these flowmeters, including flowmeter recalibration. In general, the instability of a flow meter results from its susceptibility to the transfer of undesired mechanical energy to the flow meter from an external source. These external forces can also affect the stability of the zero position of the flowmeter (that is, the measured value when there is no flow).

Although the above commercial experience basically shows that there is no problem of fatigue failure in the flow pipe in actual use, it will be a further step to gradually improve the life of the pipe by reducing the source of fatigue failure. Similarly, providing a hermetically sealed pressure-tight housing increases the suitability of the flowmeter for high pressure applications with hazardous materials and pressures up to 1000 psig and beyond. Even when the attained rated pressure is balanced against the expense of manufacturing the housing, the use of the housing described in this application enables flowmeters to be rated at a pressure of at least 300 psi (approx. 2 to 1/2 inch) and 150 psi (for large flow line).

The present invention provides an improved mass flow meter, which can significantly improve the overall stability, including reducing the sensitivity to external forces and improving zero stability, reducing pressure drop characteristics and better tolerance to fluid pressure . Numerous design improvements have been made to Coriolis mass flowmeters manufactured in accordance with one or more of the aforementioned patents as a result of optimizing their already successful performance and operating characteristics.

The present invention relates to a Coriolis mass flow meter comprising one or more flow conduits which are excited to oscillate at the resonant frequency of the flow conduit containing fluid passing therethrough. The excitation frequency is maintained at this resonance by a previously described feedback system which monitors changes in the resonance characteristics of the fluid-filled conduit as a result of changes in fluid mass due to changes in fluid density. The flow conduit of these Coriolis mass flow meters is mounted so that it oscillates about an axis of oscillation substantially at the point of mounting or strut. The resonant frequency of the oscillation is related to the axis of oscillation. The flow conduit also deforms (twists) about a second axis about which the flow conduit deflects or twists in response to the Coriolis force caused by fluid flowing through the oscillating flow conduit. This latter axis is related to the deflection caused by the Coriolis force and is substantially transverse to the axis of oscillation. The improved flowmeter provided by the present invention has improved stability and reduced sensitivity to external force disturbances, mainly due to the optimization of sensor settings, which will be described in detail later. Other improvements resulting in improved overall stability of the improved flow meter include at least a four-fold reduction in the mass of the sensor and the driver.

In a preferred embodiment, a modified U-shaped flow conduit design is provided, which has two substantially straight inlet and outlet branches that converge on each other at the Process line manifold, and, at Bends are made at two symmetrical locations along the length of the pipe, the two bends being separated by a substantially straight intermediate portion. It is also contemplated that some modified U-shaped flow ducts will have converging inlet and outlet branches separated by a continuously curved rather than straight middle section, and that other U-shaped ducts will be implemented in accordance with popular commercial Examples will have substantially parallel inlet and outlet branches. Connected to each flow conduit at symmetrical positions are two motion sensors, positioned so that the sensitivity to external forces of the signals they monitor and transmit to the electronics of the flowmeter is greater than previously known commercial mass The sensitivity of the flowmeter is significantly reduced. In a preferred embodiment, this is accomplished by placing the sensor between, but as close as possible to, the pipeline in a second out of phase twist mode and the flow in a third out of phase twist mode nodes on each side of the tube, and the actuators are placed equidistant from these sensors. The mass of the motion sensors plus their housings, and the actuators plus their housings, is substantially less than the mass of the corresponding parts of previously commercial mass flow meters. The susceptibility of the flow conduits to fatigue failure is optionally reduced by the provision of novel struts having a novel nipple-shaped sleeve for defining the axis about which each flow conduit oscillates, but , can alternatively use traditional struts, and in some embodiments, the struts are omitted. In one embodiment, a converging U-shape is utilized to provide a flangeless wafer configuration manifold structure for connection to a monitored process line. A specially sealed pressure-tight housing is provided which houses the flow conduits, motion sensors, actuators and accompanying electrical connections. The present invention discloses several embodiments of this housing, among which the embodiments shown in Figures 8 and 9 are the most desirable.

Figure 1 shows a preferred Coriolis mass flowmeter according to the present invention;

Fig. 1A represents according to the positioning situation of all motion sensors after doing modal analysis to Fig. 1 embodiment;

Fig. 2 represents a kind of random novel strut structure;

Figure 3 shows a preferred Coriolis mass flowmeter according to the present invention, which has a sheet-like manifold structure partly located in a housing as shown in Figure 4;

Figure 4 shows a random high pressure shell structure;

Figures 5A-7I represent the stability test results of a shaking table (shaker table);

Figure 8 shows another optional high pressure housing construction which is the best in terms of expense and ease of fabrication;

Fig. 9 gives further details about the housing structure of Fig. 8; and

10A to 11L show further test results to be described later.

The main feature of the invention, ie, the reduction of the influence of external forces on the stability of the flowmeter, is achieved by an optimal arrangement of the sensors and struts within a limited range. We can find that in the frequency range of 0 to 2000 Hz, there are basically six vibration waveforms whose excitation may cause loss of flowmeter stability. These six waveforms are: (1) the first in-phase curved waveform (of a low frequency lower than the excitation frequency); (2) the first out-of-phase curved waveform, which corresponds to the fundamental excitation frequency; except that the excitation frequency is filled with Outside the natural frequency of the straight pipe of the fluid (while the mode analysis (mode analysis) is carried out with the empty pipe); (3) the first different phase torsion (also known as deflection or deflection) waveform; (4) the second different phase twisted waveforms; (5) a second different phase bent waveform, and (6) a third different phase bent waveform. The optimal placement of the sensors was obtained by performing a modal analysis of the flow tube, placing two nodes on that tube for each of the six waveforms, and depending on the flow tube geometry, size and material to determine those nodes that place the sensors in closest proximity. For example, for a flow tube with the geometry shown in Figure 1, modal analysis shows that the sensors are optimally placed at the second out-of-phase torsional node and the third out-of-phase bending node on either side of the flow tube between and as close as possible to each of these nodes. Those skilled in the art will appreciate that the nodes for the six enumerated waveforms may be positioned differently from each other depending on the geometry and other characteristics of the flow tube. In some flow tube shapes, two or more nodes of different waveforms will actually coincide, enabling sensors to be placed in coincident locations, thereby increasing flow tube sensitivity. As a rule of thumb, the present invention includes the discovery that, to increase flowmeter stability, the sensors can be placed as close as possible to at least two nodes on each side of the pipe, each node being one of the aforementioned nodes. A node of a different waveform among the waveforms, especially a node of a different waveform among the waveforms represented by (3) and (6) above. The present invention also includes the discovery that the effect of waveform (1), the first in-phase flexural waveform, can be reduced or eliminated by a strut arrangement that separates harmonics of waveform (1) and The harmonics of waveform (2) are separated.

Figure 1 shows a preferred embodiment of a Coriolis mass flowmeter with a preferred modified U-shape and motion sensor positioning and mounting. As with today's commercial flow meters, the flow conduit 112 is rigidly mounted at a point 131 on a manifold 130 . The strut 122 is fixedly mounted to the flow conduit 112 so as to define an axis of oscillation B' when the flow conduit 112 is excited by the exciter 114 in the manner of a tuning fork. As the flowing material flows through the conduit 112, the Coriolis force deflects the conduit about the deflection axis A'. Electrical connectors 125, 126, and 127 from actuator 114, motion sensors 118, and 116, respectively, are connected and supported on bracket 128 as shown, and minimize and stabilize stress (when connected); or , using another approach, instead of using a bracket (as discussed later) and using a printed circuit board fixed to the housing. The connectors shown in Figure 1 are single-wire or ribbon-shaped flexible connectors with buried wires, which are mounted in a stable, low-stress semi-ring. The use of flexible joints is contemplated in U.S. Patent Application Serial No. 865,715, filed May 22, 1986 (now abandoned), and its continuation (serial number 272,209, filed November 17, 1988, already waived), and its continuation application (Serial No. 337324, filed July 10, 1989) and others. The flexible connector also has a stable, stress-reducing half-ring shape. In Figure 1, test motion sensors 116 are located at the end of an inlet or outlet leg 117, just before the bend. The motion sensors 118 are located at positions determined by modal analysis according to the present invention, which effectively reduces the influence of external forces. In a commercial flow meter embodiment, only motion sensor 118 is present, and motion sensors 116, which were tested for comparison purposes, are eliminated. The flowmeter of Figure 1 includes a housing 140 containing the flow conduit and its associated accessories, which in turn secures the manifold 130, however, preferably a flowmeter housing as shown in Figure 4 or Figure 8 and described below may be used body.

The flowmeter of the embodiment of Fig. 1 is capable of reducing the effect of external forces. Another feature of its construction includes the balancing of the flow conduits with their attachments, and the use of reduced mass sensor and actuator assemblies.

In a Coriolis mass flowmeter, the flow pipes act as springs whose elastic force acts mainly on the inlet and outlet branches. The flow conduits of the embodiment of Figure 1 are a modified U-shape with no permanent deformation due to bending in the inlet and outlet branches. There is no bending during processing, so that there is no permanent deformation in the four areas where the elastic force acts in the dual-pipe flowmeter. As a result, if a similar material of substantially the same size is used, all four regions show substantially the same response to the elastic force. This improves the flow meter manufacturer's ability to balance flow piping. When using lighter weight magnets and coils and reducing the size of their supports, the mass of the motion sensors and actuators can be reduced, thus further improving the balance of the flow conduit.

As with current commercial flow meters, motion sensors and actuators include a magnet and a coil. The sensors are of the type disclosed in US Pat. No. 4,422,338 which linearly track the entire movement of the pipeline in its oscillating path. In the FIG. 1 embodiment of the present invention and other embodiments, the total weight of these sensors and actuators is reduced by at least about four times, preferably five to five times less than their total weight in current commercial flow meter embodiments. six times or more. When using a lightweight bobbin with molded pin connectors, and having 50 gauge wire wound around the coil, while continuing to use the same mass of magnets, the weight savings can be achieved. Lightweight sensors and actuators are mounted directly on the flow conduits, which eliminates the need for mounting brackets. For example, there is an embodiment of the same size and shape as the embodiment of FIG. 1 , with a flowtube having an outer diameter of about 0.25 inches and a lightweight coil having a mass of about 300 milligrams. Some prior coils, used in comparable sized mass flowmeters manufactured in accordance with U.S. Patents Re 31450, 4422338 and 4491025 and having substantially the same flow tube outer diameter dimensions, had a mass of about 963 mg. In the flowmeter embodiment of the same size as in Figure 1, the mass flowmeter adds a new assembly (1 actuator coil and 2 sensor coils, each weighing 300 mg, and 3 magnets, each weighing 1 gram). In contrast, the mass flow meter adds 22.2 grams of corresponding components (1 actuator coil, 2 sensor coils, 1 coil bracket, 1 magnet bracket and 3 magnets) in the compared previous commercial embodiment.

A representative modal analysis was performed for the embodiment of FIG. 1 . Based on the results of the analysis, the nodes of the second out-of-phase torsional waveform and the nodes of the third out-of-phase bending waveform can be located on a flow conduit having the dimensions and properties shown in Table 1 and FIG. 1A below.

Table 1

Pipe material: 316L stainless steel

Pipe length: 16 inches

Pipe outer diameter: 0.25 inches

Pipe wall thickness: 0.010 inches

Inlet Branch: 3 inches

Outlet branch pipe: 3 inches

Intermediate pipe section: 5 inches

Bend Radius: 1.25 inches

The final sensor location in the middle of the aforementioned wave nodes is located 22.5° from the horizontal axis extending from the center of the radius of the bend.

This final sensor location is not only between the two nodes, but as close as possible to each of the two nodes on each side of the pipe. As will be understood by those skilled in the art, using modal analysis of flow conduits of other precise shapes, sizes and materials, it is possible to determine the position of each node of all of the above waveforms and make the final sensor position The determination is quickly optimized.

In some embodiments of the improved flow meter of the present invention, the fundamental excitation frequency (first out-of-phase warping waveform) is increased relative to current commercially available flow meters manufactured by the applicant assignee, Thereby increasing its harmonic value. This results in better separation of individual harmonics of the excitation waveform from those of other waveforms. For example, in the Figure 1 embodiment having the above dimensions, for all frequency cases below 2000 Hz, the tuning functions for the other five waveforms of interest are each separated from the tuning of the excitation frequency by at least 20 Hz.

In the embodiment of Fig. 1, the strut arrangement acts to separate the first in-phase bending frequency from the fundamental excitation frequency, thereby avoiding possible excitation effects with the first in-phase bending frequency. The effects of external forces operating at frequencies corresponding to the remaining four waveforms of interest are partially reduced by having a balanced flow-through conduit structure. In addition, in this embodiment, the motion sensors are placed between but very close to the nodes of the two waveforms, which happen to be close to each other, so that the second out-of-phase twist waveform can be reduced. and the effect of a third different phase-bending waveform. It is conceivable to consider its size, shape and material through modal analysis so that other orientations can be chosen to reduce the effect of those waveforms which most affect the stability of any particular pipe.

Under varying fluid pressures from less than 10 psig to about 1000 psig, for current commercial D-type flowmeters made by the assignee of the application and for current commercial Coriolis mass flowmeters made by others The embodiment in Figure 1 is tested, and the results show that: when the pressure tends to 100 psi, the excitation frequency and torsional frequency change, which has a bad influence on the accuracy of mass flow measurement. So far, it can be concluded that the effect of high fluid pressure (which is the case in will be explained below).

For example, the tube wall thickness in the embodiment of the flowmeter shown in FIG. 1 having a flow tube with an outside diameter of 0.230 inches is increased from about 0.010 inches to about 0.012 inches to reduce the undesired flow caused by high fluid pressure. stability.

Figure 2 shows an optional strut structure according to the invention. A sheet of metal (such as 316L or 304L stainless steel) or other suitable material is stamped to form each strut 122 so that it has two nipples with a hole (bore 124) having the outer diameter of the flow conduit to the large hole 120. Sleeve for transition section 121 . These struts may be brazed or welded to the flow conduits to reduce stress concentrations at the connections 123 and at the main points around which the conduits oscillate. However, the use of conventional struts previously disclosed in the prior art does not depart from the scope of the present invention.

Figure 3 shows an exploded view of a preferred Coriolis flowmeter as shown in Figure 1, with production line assembly. Instead of the typical flanged manifold in the prior art, a novel plate-shaped flangeless structural member 230 is provided, the end 232 of which can be passed through the flange hole in the manufacturing process line or other commercial process lines. 238 and threaded connecting rod 234 held in place by nut 235 are bolted between two ready made flanges.

Figure 4 shows a form of housing which reduces pressure effects. This form of construction can be used to contain the entire flow line and sensor connection parts. It includes a tube 350 of sufficient diameter to contain the flow conduits 312, actuators, motion sensors and their associated wiring connections (not shown). The tube is bent into the shape of a flow conduit. Then, this pipe is cut into two substantially equal halves along its longitudinal direction, and then the flow conduit 312 is packed into one half of the pipe along with the relevant actuator, motion sensor and wiring, and the other half of the pipe cover is installed on the above-mentioned combination. and weld them together along the two longitudinal seams, and weld them to the manifold at the joint. In this way, a pressure tight enclosure is formed which is suitable for use in applications containing hazardous fluids and capable of withstanding high pressures on the order of at least 300 psig up to 500 psig or even higher. For some embodiments, a printed circuit board may be used to connect to the inside of the housing, with flexible connections from the actuator and motion sensors to the circuit board. A junction box is then connected to the housing, and connected to the circuit board by wires that can pass through a sealed joint on the top of the housing. The junction box is then connected to means for electronically processing the signal from the sensor to give a mass flow reading and, if desired, a fluid density reading.

Another housing that facilitates housing processing is shown in FIG. 8 , and a section of it is shown in FIG. 9 . The shell of this shape is made of molded steel plates with semicircular cross-sections, as shown in Figure 9 (only plate 2 or 4 in Figure 8 is shown), and these steel plates are welded to each other to form the shell . When specifically used in the embodiment of Fig. 1, the housing is made up of ten parts marked with 1-10 as shown in Fig. 8, and these ten parts are combined in the following manner:

Five parts (numbered 1, 2, 3, 4 and 5 in Fig. 8) form half of the housing, that is, cover half of the outer peripheral surface of the flow tube when combined, and these five parts are welded to the inlet and the flow tube including the flow tube. On the bracket of the outlet manifold. A printed circuit board (not shown in any of the figures) is secured to the housing and is positioned as close as possible to the pick-off coil portion of the sensors on the flow tube, and the pick-off coil terminals rely on the pick-off coils provided thereon. To these types of wires with bends or separate semi-circular wires are connected to these printed circuit boards. The wires are then routed along the housing to the central straight section of the housing (ie section 3, which contains the straight flow conduit section 12 of Figure 1), where the wiring feeds for the flowmeter electronics are located. The feedthrough (not shown in Figures 8, 9) includes posts directly connected to the wires, or may include a third printed circuit board connected to the wires, which is connected to the feedthrough post and then connected to a A junction box (not shown) on section 3 of the middle section of the body. After the wiring is completed, the remaining 5 other parts of the housing (not shown in Figure 8, which is a top view of the housing) are welded to the bracket and the aforementioned combined and welded parts one by one, welding It is best to use automatic welding. These last five parts make up the other half of the housing. The welds between these parts are shown as thick solid lines in FIG. 8 . In addition, all welds are welded along the seams (not shown) of the inner and outer peripheries of the shell, and these two welds are respectively inside and outside the outer cover made up of the flow tube and the bracket, and the shell's The top and bottom halves are connected.

The embodiment of the casing only shows two kinds of Fig. 4 and 8, and those skilled in the art understand that: it is easy to make a similar casing to be suitable for any size and curved or straight tube Coriolis mass flowmeter, and, according to The exact shape involved, the manufacture of the embodiment of Figures 8 and 9 may advantageously be selected from other numbers of molded (profiled) steel half-circumference parts. As has been clearly shown, those skilled in the art can easily design other wiring layouts as long as they do not depart from the basic principles of the present invention.

The shaking table test is used to test the influence of external vibration force and operation line noise when vibrating the circulation pipeline and its related accessories. During the operation of factory equipment, this kind of external force often appears. 5A to 5F show the test results of a currently commercial Mi-croMotion Co., Ltd. D25 Coriolis mass flow meter using a shaking table. Figures 6A to 6F are the test results of a similar experiment on a currently commercial Micro Motion Inc. Model D40 flowmeter. The inside diameter of the flow tubing is 0.172 inches for the D25 and 0.230 inches for the D40. 7A to 7I are experimental test results of similar experiments performed on a Coriolis mass flow meter of the type similar to the flow meter of FIG. 1 .

In each of Figures 5A to 7I, the X-axis is the axis passing through the flowmeter flange (i.e., parallel to the oscillation axis B'-B'), and the Y-axis is parallel to the plane where the flow channels are located (i.e., parallel to Deformation axis A'-A'). The Z axis is perpendicular to the plane in which the flow ducts are located. All test relevant parameters are summarized in Table 2:

            Table 2 Figure No. Full Scale Vertical Scale Electronic Part Vertical Scale Full Movement

Flow Rate Frequency Sweep Output Percentage of Scale Axis

(1bm/minute) (Hezhi) (Jiho) ( %) ( %) 5A 1.04 4.20 10 Z5B 10.4 4.20 100 Z5C-15-2000-Z5D 1.04 4.20 10 x5e 10.4 4.20 X5F-15-2000-X6A 1.04 4.200 10 10 10 10 Z6B 10.4 4.20 100 Z6C-15-2000-Z6D 1.04 4.20 10 x6e 10.4 4.20 100 X6F-15-2000-X7A 1.04 4.20 10 X7B 10.4 4.20 100 X7C-15-2000-X7D 1.04 4.20 10 Z7E 10.4 4.20 100 Z7F -15-2000-Z7G 1.04 4.20 10 Y7H 10.4 4.20 100 Y7i-15-2000-Y

These shaking table tests are performed on the complete mass flow meter without the housing. The output, shown on the bar chart charts of Figures 5A to 7I, is the reading of the motion sensor in response to the corresponding external vibration. The sequence represented in each set of graphs is the response of the flowmeter to a linear frequency ramp (ramp) ranging from 15 Hz to 2 kHz, and thereafter to input vibrations at random frequencies (see Figures 5C, 5F, 6C, 7C, 7F, 7I). Frequency ramps are generated over a 10 minute period, and random vibrations are generated over about a 5 minute period. Figures 5A, 5B, 5D, 5E, 6A and 6B each show the effect of external vibration at various frequencies in exciting harmonics of the six waveforms discussed above for the movement of the D25 and D40 flowmeters.

7A to 7F show the sensitivity to excitation caused by external vibrations about the X and Z axes (the same axes as in FIGS. 5A and 5F ) of the flowmeter shown in FIG. 1 embodiment of the present invention, and each The scale of the graph is the same. Figures 7G to 7I are drawn along the Y-axis. (Random vibrations are first produced in Figures 7D-7F). What to know: Optimal flow meters have been shown to significantly reduce the effects of external vibrations when exciting the harmonics of the six waveforms of motion. In this way, the optimal structural design means that the flowmeter is effectively isolated from external forces.

Additionally, tests were performed to measure the effect of external vibration on zero stability. This test provides the stability of the time difference (Δt) measurement at no flow, a result known as a jitter test. A frequency counter is used to directly measure the pulse width of an up-down counter before averaging or filtering the electronics. Tests were performed on a shaker table with random frequency inputs over a range of accelerations along the X, Y and Z directions. The results show that: for the D25 and D40 flowmeters, as the acceleration increases, the influence of external vibration causes the pulse width to diverge (expand) exponentially or several times from the average bit. For each axis, such divergence has been significantly reduced in the best flowmeters. Thus, as in the case of the vibration test, the jump test shows that the optimal flowmeter structure design effectively isolates the flowmeter from the influence of external forces.

Included in FIGS. 10A-10G are graphs depicted with further data obtained with a flowmeter embodiment of a structural design of FIG. 1 . As disclosed by me, the flowmeter has conventional struts and a flow tube that is 20% thicker than that of current commercial flowmeters of comparable size. According to the requirements of the present invention, the sensors used in the test are placed between and As close as possible to the position between the nodes of waveforms (4) and (5) above. When testing, the flow tube of this flowmeter embodiment is covered by a housing of the type shown in Figures 8 and 9 (half of the circumference of the tube), and a junction box (not shown in the figure) is attached to the outside of the housing. shown), the wires are routed through the housing to the junction box. The junction box is conventionally connected to another box (called a "Remote Control Flow Transmitter" or "RFT") that contains the flowmeter electronics and has a readout board for mass flow and density values, from which charts can be collected. Data in 10A-10D. The inside diameter of the flow tube wall of this flow meter embodiment is about 0.206 inches.

Figures 10A and 10B each represent a calibration curve between accuracy and flow rate tested using water with a mass flow rate of 0-45 lbs/min. The difference between Fig. 10A and Fig. 10B is: what Fig. 10A represents is a "22-point" calibration curve drawn with the points starting to measure under the situation of about (3-4) pounds per minute of mass flow; Fig. 10B is a A "45-point" calibration curve where more data points were collected, especially for mass flow rates below 5 lbs/min (starting at about 0.5 lbs/min flow).

Figure 10B also presents fluid line pressure drop data for mass flow rates ranging from 0 to about 45 lbs/min. The combination of Figures 10A and 10B shows that the embodiment of the flowmeter of the present invention is in good condition within ±0.2% of the published accuracy value, which is the assignee of the applicant, i.e. Micro Motion Co., Ltd. Characteristics of current commercial flowmeters. Figure 10B also demonstrates that this flow meter embodiment has an approved fluid line pressure drop performance.

10C and 10D are graphs of simulated offset values for mass flow and density, respectively, for fluid pressures from 0 to about 2000 psig hydraulic. In each case, there is a comparison curve on the mean empirical standard deviation value curve measured for a D-Series commercial mass flow meter sold by Micro Motion Incorporated. In FIG. 10C, each data point for the flow simulation offset and flow standard deviation is expressed in "seconds". In Figure 10D, the density simulation offset and density standard deviation are expressed in "g/cm3" units. In both cases, the data show that the flowmeter of the present invention performs well within the measured standard deviation data.

Figures 10E, 10F and 10G are graphs of shaker test data obtained in a manner similar to the data in Figures 5A to 7I, but expressed as shaker frequency (Hertz) versus analog output (i.e. mass flow) perturbation ( seconds), while Figures 5A to 7I are reproductions of bar graphs of shaker frequency (Hertz) versus motion sensor readings (seconds). Additionally, the data in Figures 10E, 10F, and 10G were collected on the same flowmeter assembly with the half-shell attachment that was used for the data of Figures 10A-10D. For comparison, similar curves for two different D25 type runner conduit arrangements, each without a housing attachment, are also shown in FIGS. 11A-11L. In each case, the X, Y and Z axes are as defined above, the vibrating table vertically scales the frequency sweep from 15 to 200 Hz, the remote frequency transmitter spans 5 g/s, and the correction factor is 1. An analog output read is available from this transmitter. Other relevant parameters are summarized in Table 3:

table 3

Drawing No. Motion Device of Vibrating Table

Scan Axis

10E Logarithmic sign X X Embodiment of the present invention

10F Logarithmic Sign Y Embodiment of the present invention

10G Logarithmic Sign Z Z Embodiment of the present invention

11A Linear X D25 type device 1

11B Logarithmic sign X D25 type device 1

11C Linear Y Y D25 type device 1

11D Logarithmic sign Y D25 device 1

11E Linear Z Z Type D25 device 1

11F Logarithmic sign Z D25 type device 1

11G Linear X D25 type device 2

11H Logarithmic sign X D25 type device 2

11I Linear Y D25 type device 2

11J Logarithmic sign Y D25 type device 2

11K Linear Y D25 device 2

11L Logarithmic symbol Y D25 type device 2

A linear sweep of the vibrating table takes the same amount of time to move through each 100 Hz vibration interval, ie, for example, 15 to 115 Hz for a scan as for example, 1000 to 1100 Hz. In a logarithmic symbol sweep, the elapsed time interval is lengthened at low frequencies such as 15-400 Hz, and shortened (or sped up) at high frequencies. It can be seen that the logarithmic sweep clearly points out the frequencies at which, in current commercial D-type flowmeters, external excitation causes harmonic disturbances. Figures 10E, 10F and 10G show that the flowmeter of the present invention is significantly less sensitive to this effect than the D25 flow line.

Although the preferred embodiment of a multiple flow conduit mass flowmeter is shown, it is contemplated that the invention described herein can also be implemented with a Coriolis mass flowmeter having only one flow conduit which is either connected to a Components such as leaf springs or a dummy pipe are used to form a tuning fork with the flow pipe, or to install it on a base with a larger mass when the mass of a single flow pipe is small.

While the above discussion has focused on the size and shape of the flow conduits for purposes of illustration, it is well within the knowledge of those skilled in the art that many changes and modifications may be made in the actual practice of the invention described herein, Such changes and modifications are intended without departing from the scope of the present invention as defined by the following claims.

Claims (11)

CLAIMS 1. A mass flow meter in which the mass flow rate of a flowing substance is determined from the effect of a measured Coriolis force. The flowmeter comprises: at least one continuous flow conduit without pressure sensitive joints or pipe sections, said conduit being securely mounted to said support means with said conduit's inlet and outlet ends; a plane parallel to the flow pipe is firmly mounted on the support means, so that the member and the pipe oscillate in a tuning fork manner; an excitation device is used to make the pipe and the member rotate around each of the a rigid support means oscillates adjacent to the bending axis; a pair of sensor means mounted on said pipe at opposite sides of said excitation means to monitor oscillatory motion of said pipe including flow conditions of said pipe The movement caused by the Coriolis force; and the signal processing device, which detects the signal from the sensor and converts this signal into a mass flow value, characterized in that each sensor device installation point is as close as possible to a pair of the pipeline The two closest nodes of the vibration waveform, the pair of vibration waveforms is obtained after (i) doing a wave state analysis to determine the nodes of each waveform, and (ii) selecting the two nodes with the closest On the basis of the waveform, then on the basis of the first in-phase bending waveform, the first different-phase bending waveform, the first different-phase twisting waveform, the second different-phase twisting waveform, the second different-phase bending waveform, or the third different-phase bending waveform selected from. 2. The mass flow meter of claim 1, wherein said second member is a second continuous flow conduit that is substantially identical to said first conduit. 3. The mass flow meter of claim 1 or 2, wherein each of said continuous flow conduits has a substantially straight inlet branch and a substantially straight outlet branch converging at said support means relative to each other and interconnected with respect to said support means by the remainder of said continuous conduits. 4. The mass flow meter of claim 3, wherein the inlet branch and the outlet branch are interconnected relative to the bracket by a substantially straight portion of the conduit, the substantially straight portion being bent at either end , to join each substantially straight branch section. 5. The mass flow meter according to claim 1, wherein the setting point of the sensor device is as close as possible to the second different phase torsional waveform and the third different phase on the inlet side and the outlet side of the flow pipe. Corresponding nodes of a phase-bending waveform. 6. The mass flow meter of claim 2, wherein said sensor means are set points, on respective inlet and outlet sides of each of the flow conduits, despite proximity to the second out-of-phase torsional waveform and the first Corresponding nodes of three different phase warped waveforms. 7. The mass flow meter of claim 2 having said flow conduits clamped together at points along the inlet and outlet branches by struts, said flow conduits being determined according to modal analysis The positions of the points which enable the separation between the frequencies of the first in-phase curved waveforms and the first out-of-phase curved waveforms of the flow conduits enable the separation of the first in-phase curved waveforms to The influence of flowmeter sensitivity is reduced to a minimum. 8. The mass flow meter of claim 7, wherein said continuous flow conduits have a substantially straight inlet branch and a substantially straight outlet branch which converge oppositely at said support means and interconnected with respect to the support means by the remainder of said continuous conduits. 9. The mass flow meter of claim 7, wherein said struts comprise nipple retaining sleeve means. 10. The mass flow meter of claim 2, wherein the support means includes a flangeless inlet and outlet plenum, the plenum comprising two separate flow chambers, one of which is used at the inlet One side separates the fluids and the other serves to recombine the fluids on the outlet side. 11. The mass flowmeter as claimed in claim 1 or 2, characterized in that it also comprises a pressure-tight housing having substantially the same geometrical shape as the flow conduit, which contains all of the conduits, the sensor device and the exciter, and welded on the supporting device.

Images