MXPA00005320A - Driver for oscillating a vibrating conduit - Google Patents

Abstract

A process parameter measurement device, and in particular a Coriolis mass flowmeter or vibrating tube densimeter, having a driver or drivers for inducing oscillation of at least one conduit. The drivers are attached at locations along the vibrating conduit selected to influence certain modes of interest. A driver is located near an area of maximum amplitude of a desired vibration mode and near an area of minimum amplitude of an undesired vibration mode. Various known experimental modal analysis or modeling techniques are used to determine the appropriate locations for one or more drivers. Multiple drivers located according to the present invention are used to influence, i.e. excite or suppress, multiple modes. In addition, multiple, separate drive circuits produce multiple, electronically isolated drive signals for delivering greater total power to the vibrating conduit.

Description

ACTUATOR TO MAKE A VIBRANT DUCT FIELD OF THE INVENTION The present invention relates to an apparatus and methods for using and controlling an actuator (s) for oscillating a vibrating conduit. More particularly, the present invention relates to optimally positioning an actuator (s) so that the multiple modes of vibration are appropriately influenced by the operation of the actuator (s).

DECLARATION OF THE PROBLEM It is known to use Coriolis mass flow meters to measure the rastic flow and other information for materials that flow through a conduit. Coriolis effect flowmeters, exemplary, are described in U.S. Patent Nos. 4,109,524 of August 29, 1978, 4,491,025 of January 1, 1985, and Re. 31,450 of February 11, 1982, all issued to J. E. Smith et al. These flow meters have one or more conduits of straight or curved configuration. Each configuration of the duct in a Coriolis etroastral flow has a set of natural vibration modes, which can be of a simple, torsional, radial or REF: 120214 coupled. Each conduit is driven to oscillate in resonance in one of these natural modes. The material flows into the flowmeter from a conduit connected to the input side of the flowmeter, is directed through the conduit or conduits, and exits the flowmeter through the outlet side. The natural vibration modes of the material-filled, vibrating system are defined in part by the combined mass and stiffness characteristics of the conduits and the material flowing within the conduits. When there is no flow through the flowmeter, all points along the pipe oscillate, due to an applied driving force, with an identical phase or a zero flow phase depending on the mode of vibration driven. When the material begins to flow, the Coriolis forces cause a change in the phase difference between any two points along the conduit. The phase on the input side of the duct stops the actuator, while the phase on the output side induces the actuator. The selective sensors are placed in the conduit to produce sinusoidal signals representative of the movement of the conduit. The signal output of the selective sensors is processed to determine the change in the phase difference between the selective sensors. The change in the phase difference between two signals of the selective sensor is proportional to the mass flow rate of the material through the conduit. A typical component of each Coriolis effect flow meter, and each densitometer of the vibrating tube, is the drive or excitation system. The drive system operates to apply a physical, periodic force to the conduit which causes the conduit to oscillate. The drive system includes an actuator mounted to the conduit (s) of the flow meter. The drive mechanism typically contains one of many well-known arrangements, such as, but limited to, a moving coil where a magnet is mounted to a conduit and a coil of wire is mounted to the other conduit in a relationship opposite the magnet. An actuator conduit continuously applies a drive signal, typically sinusoidal or square, periodic to the drive coil. Through the interaction of the alternating, continuous magnetic field, produced by the coil in response to the driving signal, periodic and the magnetic field, constant produced by the magnet, both flow conduits are initially forced to vibrate in a sinusoidal pattern, opposite which remains after. Those skilled in the art recognize that any device capable of converting an electrical signal to a mechanical force is suitable for the application as an actuator. (See the North American patent No. 4,777,833 issued to Carpenter and assigned in its cover to Micro Motion, Inc.) Also, one does not need to use a sinusoidal signal, but preferably some periodic signal such as the drive signal (see U.S. Patent No. 5,009,109 issued to Kalotay et al. And assigned on its cover to Micro Motion, Inc. .). A typical mode, although not the only mode, in which a double-tube Coriolis effect flowmeter is operated is the first out-of-phase bending mode. It is known to maximize the first bending mode out of phase through the careful placement of the actuator. See U.S. Patent No. 4,823,614 issued to Dahlin. However, the first out-of-phase bending mode is not the only mode of vibration present in the vibrating structure of a Coriolis effect flow meter operated in the first out-of-phase bending mode. Naturally, there are higher modes of vibration which can be excited. There are also, as a result of the fluid through the vibrating duct and the consequent Coriolis forces, response modes of the Coriolis effect such as the first torsional mode out of phase. There are also in-phase modes and lateral modes of vibration. Another reason why additional and undesirable modes are sometimes excited in a Coriolis effect flowmeter is when the manufacturing tolerances are such that the actuator elements are not placed symmetrically in the conduits or the actuator does not generate a pure uniaxial force in the direction proposal perpendicular to the plane of the tube. This results in the actuator placing eccentric forces in the conduits, thereby exciting the multiple modes of vibration. In addition to the multiple modes that are excited by the energized excitation of the conduits, the modes can be excited due to vibrations external to the flow meter. For example, a pump placed elsewhere in a process line could generate a vibration along a pipe that excites a vibration mode in a Coriolis effect flowmeter. Finally, there are belts of vibration modes present in the Coriolis effect flowmeter which is proposed to oscillate in only an individual mode such as the first bending mode out of phase. Even within a relatively small range of frequencies near the driven mode there are typically at least several additional modes of vibration. In this way, a Coriolis effect flowmeter to oscillate or resonate in a mode actually has one (one) duct (s) that oscillate in many other modes besides the proposed mode. Selective sensors in a vibrating duct (s) produce feedback signals representative of the vibration of the duct (s). In this way, if the conduit (s) are vibrated in multiple modes, then any feedback signal from a selective sensor in the vibrating conduit (s) will have a modal content representative of the modes multiple vibration. This can lead to problems in the feedback circuit of the drive signals since unwanted vibration modes can be reinforced by the drive signal itself. For example, a pump could cause a vibration in a pipe to which a Coriolis effect flowmeter is attached. The Coriolis effect flow meter is caused to vibrate in a certain way due to the vibration of the pump. This vibration mode is represented by a certain modal content in the drive feedback signal (from one of the transducers). The drive feedback signal is processed to produce a drive signal. The drive signal, which still has a modal content in the vibration mode induced by the vibration of the pump, is used to drive the Coriolis effect flowmeter to vibrate. In this way, the flowmeter is activated to vibrate in an undesired way. Another exemplary problem concerns the intrinsic security requirements. In order to comply with the intrinsic safety requirements established by the various approval agencies, the total power available in the actuator of a Coriolis effect flow meter is limited. This power limitation may be a problem for Coriolis effect flowmeters particularly with respect to larger flow meters and more particularly with respect to larger flow meters that measure fluids with retained gas. Therefore, it is important to introduce energy to a flow meter in a way that only the desired vibration mode (s) are energized, therefore the energy enters the desired modes and energy is not 'consumed' in the modes unwanted A further problem is that, in the example of a historical Coriolis effect meter operated in the first out-of-phase bending mode, the positioning of the actuator is also a position of maximum amplitude for the second out-of-phase bending mode. Therefore, the second out-of-phase bending mode is permanently excited in a powered Coriolis effect meter to oscillate in the first out-of-phase bending mode. The drive feedback signal, and subsequently the drive signal, therefore contains a strong response in the second out-of-phase bending mode. U.S. Patent No. 5,301,557 issued April 12, 1994 to Cage et al. And assigned on its cover to Micro Motion, Inc. ("the 557 patent") discloses a method of positioning selective sensors in the conduit (s) ( s) of a Coriolis effect flow meter. The '557 patent discloses a method for mounting transducers at locations in u (a) conduit (s) close to a node (s) of an undesired vibration mode (s). Therefore, transducers are less likely to produce signals that have a strong component of the desired mode (s). The '557 patent does not teach anything about the placement of actuators or the use of the drive signal where undesired modes are suppressed. There is a need to optimally position the actuator elements in the vibrating tube (s) so that undesired modes are minimized. There is a further need to increase the force of the actuator available in a Coriolis effect flow meter while still meeting the intrinsic safety requirements. There is a need to influence the multiple modes in the vibrating tube (s) of a flowmeter such as to excite two modes or to excite one mode and suppress another.

DECLARATION OF THE SOLUTION The problems identified above, and others, are solved and a technical advance is achieved in the field by the drive system of the present invention. The present invention provides a method and apparatus for using modal analysis techniques to optimally position an actuator (s) in the vibrating tube (s) of a flow meter so that the vibration modes they are appropriately influenced. One or more actuators are placed in the location (s) in the conduit (s) such that the energy is introduced to the vibrating structure that excites a desired mode (s) of vibration (s) and does not excite one (a) mode (s) of unwanted vibration (s). The excitation of the undesired modes is minimized, the force of the actuator in the desired vibration mode is maximized, and the conduit (s) is (are) so driven most effectively in the mode (s) desired (s). A method is provided for optimally placing the actuators in the vibrating ducts so that the influence of the multiple vibration modes is controlled. Influencing a mode is to excite or suppress the mode. A Finite Element (FE) model of the vibrating duct (s) is constructed. The coefficients of the eigenvector for the modes of interest are extracted from the FE model. Alternatively, modal analysis techniques are used to determine the coefficients of the eigenvector for the modes of interest. The coefficients of the eigenvector of the modes of interest are represented graphically to identify those regions along the vibrating tube (s), where, for example, a desired mode is near a point of maximum amplitude and an undesired mode. it is near a point of minimum amplitude. An actuator is placed within this region. An actuator placed in this region is less likely to excite the undesired mode while appropriately exciting the desired mode. The same technique is used to place one (one) actuator (s) so that the multiple desired modes are excited more effectively or to ensure that multiple unwanted modes are not excited. The coefficients of the eigenvector for the modes of interest are used alternatively to produce a Frequency Response Function (FRF) for the vibrating duct (s). An FRF characterizes the dynamics between a force applied to a structure in one location and the resulting movement of the structure in another location. The FRF is used to quantitatively determine the optimal location (s) of the actuator (s) in the vibrating tube (s) as an alternative to the observed graphical representation method. previously. A vibrating U-shaped conduit provides an example of the present invention for optimally positioning the actuators. Historically an individual actuator is placed in the center of the sinus end of a U-shaped conduit. A sinusoidal signal is supplied at the frequency of the first bending mode out of phase to the individual actuator to cause the conduit (s) ) vibrate (n). The center of the sinus end of the duct (s) is a point of maximum amplitude of both the first and second out-of-phase bending modes of the U-shaped duct (s). In this way, this location of the actuator tends to excite the second, unwanted, out-of-phase bending mode as well as the first desirable out-of-phase bending mode The present invention provides a flowmeter with the actuator (s) placed in the position (s) such that the first bending mode out of phase is excited (desirable) but the excitation of the second out-of-phase bending mode is initiated This is done according to the present invention when placing at least one actuator in a location that is close to a position of maximum amplitude of the first bending mode out of phase, desired and close to a position of minimum amplitude of the second mode of bending out of phase, undesired. A further example of the present invention is where one wishes to excite both the first bending mode out of phase and the first torsional mode out of phase but not the second bending mode out of phase. Historically, this would be done by placing, in the example of a U-shaped conduit (s), an individual actuator in the center of the end of the duct sine to excite the first bending mode out of phase and a couple of actuators on the opposite legs of the duct (s) to drive the first torsional mode out of phase. A completely different approach and structure result in accordance with the present invention. A FE model of the vibrating structure is constructed. The coefficients of the eigenvector for the first bending mode out of phase, the first torsional mode out of phase and the second bending mode out of phase are extracted from the FE model. The coefficients of the eigenvector are plotted with respect to the distance along the conduit and the locations of the actuators are selected. The selected locations of the actuators are positions along the conduit where the first bending mode out of phase and the first torsional mode out of phase are close to the points of maximum amplitude and the second bending mode out of phase is close to a point of minimum amplitude. In this way, the input of energy appropriately phased to the acclimates in these locations tends to excite the first mode of bending out of phase and the first torsional mode out of phase and likewise tends not to excite the second bending mode out of phase. phase. The optimal location of the actuators also allows for an increase in power input to the vibrating tube (s) in the desired mode (s) since more than the available power is supplied in the desired mode (s). This is advantageous where a high actuator power is required such as, but not limited to, where intrinsic safety requirements are a problem. Additional gains in drive power are achieved in accordance with the present invention by using multiple actuators, which are properly positioned and each of which is controlled by a separate drive circuit.

BRIEF DESCRIPTION OF THE DRAWINGS FIGURE 1 represents a Coriolis mass flow meter system; FIGURE 2 describes a representation of the finite element model of the vibrating ducts of the Coriolis mass flowmeter system of FIGURE 1; FIGURE 3 is a graph of the shape coefficients of the mode with respect to the position along the conduits of FIGURE 2; FIGURE 4 discloses a representation of a finite element model of a Coriolis mass flow meter system including an actuator and feedback elements according to the present invention; FIGURE 5 depicts a representation of a finite element model of a Coriolis mass flowmeter system that includes an actuator and feedback elements according to another embodiment of the present invention; FIGURES 6 and 6B are graphs of phase and magnitude of the speed of the conduit for the two different schemes of the actuators; FIGURES 7A and 7B are graphs of phase and magnitude of the duct velocity for a drive scheme using double actuators that supply force with equal amplitude but opposite phase; FIGURE 8 depicts a dual drive system having electrically isolated drive circuits for each drive element; FIGURE 9 depicts a representation of a finite element model of a straight tube Coriolis mass flowmeter according to the present invention; FIGURE 10 is a graph of the mode shape coefficients with respect to the position along the conduits of FIGURE 9; FIGURE 11 is a block diagram of a drive circuit for producing a multi-mode drive signal.

DETAILED DESCRIPTION General Coriolis Effect Flowmeter System - FIGURE 1 Figure 1 shows a Coriolis Effect Flowmeter 5 comprising a Coriolis effect meter assembly 10 and an electronic measuring device 20. The electronic measurement device 20 is connected to the assembly meter 10 by means of cables 100 to provide density, mass flow rate, volume flow rate and mass flow information, total on path 26. FIGURES 1-8 depict the structure and operation of a flow meter Coriolis effect model CMF300 manufactured by Micro Motion, Inc. of Boulder, Colorado. A certain structure of the Coriolis effect flowmeter is disclosed although it is apparent to those skilled in the art that the present invention could be practiced in conjunction with a vibrating tube densitometer without the additional measurement capability, provided by a Coriolis mass flowmeter. Also, although certain configurations of Coriolis effect flowmeters are shown and described herein, those skilled in the art of vibrating tube sensors recognize that the present invention is equally applicable to any vibrating tube flowmeter or densitometer regardless of the number and the shape of the vibrating ducts. In essence, the present invention is equally applicable to any vibrating tube flow meter or densitometer without considering the number and shape of the vibrating ducts. In essence, the present invention is applicable to any process parameter measuring device that employs a vibrating duct. The meter assembly 10 includes a pair of fins 101 and 101 ', distributors 102, 102'; a separator 107 and conduits 103A and 103B. Connected to conduits 103A and 103B are actuator 104 and selective sensors 105 and 105 '. The tension bars 106 and 106 'serve to define the axes and W around which the conduit oscillates. When the flowmeter 10 is inserted into a pipe system (not shown) which carries the material of the process being measured, the material enters the meter assembly 10 through the fin 101, passes through a distributor 102 where the material is directed to enter conduits 103A and 103B, flows through conduits 103A and 103B and back into manifold 102 where meter assembly 10 exits through fin 101 '.

The conduits 103A and 103B are appropriately selected and mounted to the manifold 102 to have substantially the same basic distribution, moments of inertia and elastic moduli around the bending axes -W and ¥! ' - ¥! ' , respectively. The conduits extend outwardly from the distributor in an essentially parallel fashion. The conduits 103A-103B are stimulated by the actuator 104 in opposite directions around their respective bending axes W and W and in what is referred to as the first out-of-phase bending mode of the flow meter. The actuator 104 may comprise any of many well-known arrangements, such as a magnet mounted to the conduit 103A and an opposite coil mounted to the conduit 103B and through which an alternating current is passed to vibrate both conduits. A suitable actuation signal is applied by the electronic measuring device 20, by means of the cable 110, to the actuator 104. The electronic measuring device 20 receives the left and right speed signals appearing on the cables 111 'and 111', respectively. The electronic measuring device 20 produces the drive signal that appears on the cable 110 and causes the actuator 104 to vibrate the tubes 103A and 103B.

The electronic measuring device 20 processes the left and right speed signals to calculate the speed of the mass flow and the density of the material passing through the meter assembly 10. This information is applied by the electronic measurement device 20 on the path 26 to a means of use (not shown). It is known to those skilled in the art that the Coriolis effect flowmeter 5 is very similar in structure to a vibrating tube densitometer. Vibrating tube densitometers also use a vibrating tube through which the fluid flows or, in the case of a densitometer of the type of sample, within which the fluid is maintained. Vibrating tube densitometers also employ a drive system to excite the duct to vibrate. Vibrating tube densitometers typically use only a single feedback signal since a density measurement only requires frequency measurement and a phase measurement is not necessary. The descriptions of the present invention herein apply equally to vibrating tube densitometers.

Mode Form Coefficients - FIGURES 2-3 FIGURE 2 depicts a finite element model of conduits 103A-103B of the flow meter 10 shown in FIGURE 1. For purposes of describing the present invention, only the vibrating portion of the flowmeter needs to be discussed. flowmeter, in this way FIGURE 2 represents only conduits 103A-103B. The model fixes to the ground the ends of the flow tubes that, in a physical flowmeter, are connected to the flowmeter distributor. The techniques of the finite element model are well known to those skilled in the art and do not form part of the present invention. The exemplary finite element model was constructed using SDRC-Ideas and analyzed by MSC / NASTRAN, a finite element code available from MacNeal-Schwendler. Those skilled in the art of the finite element model recognize that any finite element code could alternatively be used. The locations of the transducers were modeled to produce the representative output of the relative movement between the locations in the flow tube of the magnet and the coil corresponding to the right transducer, the actuator and the left transducer. These "scalar points" are a normal technique in dynamic, advanced analysis. See "A Finite Element for the Vibration Analysis of Fluid-Conveying Ti Eshenko Bea", (AIAA Paper 93-1552), for more information on the finite element model of Coriolis effect flowmeters. Each scalar point is marked with a number of nodes N101-N117 in FIGURE 2. Nodes N101-N-117 facilitate further discussion of mode shapes and their interactions along the length of conduits 103A 103. The actuator 104 and the transducers 105-105 'are shown in the same positions as shown in FIGURE 1. The actuator 104 and the transducers 105-105' are shown in FIGURE 2 and the FIGURES below, as an element in each conduit. This is because the actuators and transducers are typically comprised of a coil attached to a conduit and a magnet attached to a second conduit or a cover of the flow meter. The position of the actuator 104 in the node N109 is a known and typical actuator position for a curved tube Coriolis effect meter operated in a bending mode. FIGURE 3 is a graph of the coefficients of the eigenvector, normalized for certain modes of vibration as a function of a position along the conduits 103A-B. The vertical axis of the graph 300 is the coefficient of the eigenvector, normalized. The horizontal axis of the graph 300 is the position along the conduits 103A-B as indicated by the nodes N101-N117. The graph 300 includes the curve 302 which is comprised of the coefficients of the eigenvector for the first bending mode out of phase with respect to the positions of the nodes N101-N117. The graph 300 also includes the curve 304 which is comprised of the coefficients of the eigenvector for the second bending mode out of phase with respect to the positions of the nodes N101-N117. The third data set comprising the graph 3Q0 is the curve 306 which is comprised of the coefficients of the eigenvector for the first torsional mode out of phase with respect to the positions of the nodes N101-N117. FIGURE 3 represents a quantitative approach to characterize the shapes present in a vibrating duct. The coefficients of the eigenvector used to generate the curves 302-306 are generated in one of at least two ways. One approach is to construct a finite element model of the vibrant structure of interest from which the coefficient of the eigenvector is extracted for the modes of interest. Another approach is to use experimental, modal analysis techniques to determine the coefficients of the eigenvector of a physical model of a vibrating structure. The finite element model and experimental, modal analysis techniques are well known to those skilled in the art of complex mechanics. The nodes N101 and N117 are close to the tension bars 106 and 106 ', respectively. The tension rods 106, 106 'connect the conduits 103A-B together and are therefore a ligating location along the length of the conduits 103A-B where very little relative movement occurs between the conduits. In this way, all three curves 302-306 improve the zero amplitude at nodes N101 and N117. The conduits 103A-B are free to oscillate between the nodes N101 and N117. The maximum oscillation amplitude at each node N101-N117 in each mode is indicated by curves 302-306. FIGURE 2 represents an actuator 104 placed at node N109. The node N109 is the center of the conduits 103A-B which means that it is a position that is equidistant from each tension bar 106 and 106 '. This represents the typical position of the actuator, historically used to drive a Coriolis curved-tube flowmeter in the first out-of-phase bending mode. Note that the first bending mode out of phase, as indicated by curve 302, reaches a maximum at node N109, the center point of conduits 103A-B. In this way the node N109 is an efficient position in which the first bending mode out of phase is excited. In this context, "efficient" means that a movement of a relatively large conduit results from a relatively small input force. The node N109 is the most efficient position along the conduits 103A-B in which the first bending mode out of phase is excited. However, observe in FIGURE 3 that the node N109 is also a position of maximum amplitude for the second mode of bending out of phase, as indicated by the curve 304. In this way, the energy input to the conduits 103A- B at node N109 tends to excite both the first and second bending modes out of phase. This is an undesired condition since one typically does not want to excite the second mode of bending out of phase. As is well known to those skilled in the art of vibrating tube sensors, any mode that is excited by an actuator is also detected by the transducers and certain modes can detrimentally influence a mass flow or density measurement or the generation of an efficient drive signal.

Placement of the Actuator _, in Accordance With the Present Invention - FIGURES 4-6B FIGURE 4 depicts the conduits 103A-B with an actuator 401 positioned at a position along the conduits 103A-B in accordance with the teaching herein invention. With reference to the graph of the shape-mode coefficient 300 of FIGURE 3, note that the second bending mode out of phase (curve 304) is close to a position of minimum amplitude near the node N107 while the first mode of bending out of phase is still near a maximum amplitude at node N107. Therefore, the energy input to the conduits 103A-B at or near the node N107 tends to excite the first bending mode out of phase but not the second bending mode out of phase. FIGURE 4"shows the actuator 401 positioned at the node N107. The actuator 401, when operating to oscillate the conduits 103A-B, excites the first bending mode out of phase but does not excite, or excites minimally, the second mode of bending out of phase A flow meter according to the present invention and having an actuator, therefore, takes advantage of an awareness of the various vibration modes present in the vibrating structure to optimally position the actuator so that only the mode or desired drive modes are excited.The eccentric positioning of the actuator 401 gives rise to certain problems.A problem is that the energy enters the vibrating structure in a non-symmetrical manner due to the eccentric location of the actuator. close to a position of maximum amplitude of the first bending mode out of phase, in this way, an individual actuator in node N107 (or the corresponding node Nlll) tends to excite eccentrically the first torsional mode out of phase which can manifest as the phase shift induced within two points along the conduits 103A-B. Since the phase shift between the points along the conduit is the basis for measuring the mass flow rate made by a Coriolis effect flow meter, this can be a problem. FIGURE 5 depicts two actuators 501-502 placed at nodes N107 and Nlll, respectively. The locations of the actuators 501-502, specifically the nodes N107 and Nlll, are selected for the same reasons discussed above with respect to FIGURE 4 for the case of an eccentrically placed individual actuator. In the case of a flow meter operated to oscillate in the first bending mode out of phase, the force generated by the actuators 501-502 is of equal amplitude and phase. With reference to the graph of the shape coefficient of mode 300 of FIGURE 3, the nodes N107 and Nlll are close to the positions of maximum amplitude for the first mode of bending out of phase and the positions of minimum amplitude for the second mode of bending out of phase. In this way, the first out-of-phase bending mode is excited and the second out-of-phase bending mode is only minimally excited. In addition, the first out-of-phase torsional mode is not excited because the actuators 501-502 drive the conduit 103A with equal amplitude and equal phase. Therefore, the measurement of the mass flow rate of the Coriolis effect flow meter is not affected as is possible in the mode represented in FIGURE 4. Preferably that the graphical representation of the coefficients of the eigenvector as shown in FIGURE 3 , one can instead use the coefficients of the eigenvector in combination with the natural frequencies and the damping of the vibrating structure to generate an FRF for the vibrating structure. The FRF is used to determine a physical response in inches per second at a location in the vibrating structure to a force applied in pounds at another placement in the vibrating structure. This provides a quantitative approach to identify the optimal location of the actuator. The calculation and manipulation of frequency response functions is well known to those experts in the study of vibrating structures. The FRF matrix, which allows the calculation of the response at a point in the structure to an input at another point in the structure, is given in Equation 1: EQUATION 1 where H (?) is the matrix of the FRF as a function of the frequency in response units normalized by a drive excitation. Typical units are inches / second per pound. The indices of the FRF matrix correspond to the physical locations for the response and excitation, ie H?: I is the response at location i to a drive excitation at location j. The sum index r corresponds to the number of desired modes, which is defined by the number of columns in the matrix of the eigenvector F. Each row in F corresponds to the coefficient of the eigenvector of a physical location in the structure for which you want an answer or in which force should be applied. The matrix of the eigenvector F can be derived conveniently from a finite element analysis or can be measured experimentally. The term i?, Where i = V-l indicates that the answer is in terms of velocity. The term ? in the numerator and the denominator is the excitation frequency in radians / second. F | ri is the eigenvector r'th (column of the eigenvector matrix) normalized to unit modal mass,? is the modal damping for the r'th mode as a critical damping fraction, and? n is the undamped natural frequency of the r'th mode in radians / second. A physical response, X, to a given force, F, is calculated from Equations 1 and 2. Note that the response in this linear system for the multiple forces can be superimposed by adding the individual responses to an individual force.

X (w) = H (?) XF EQUATION 2 Equations 1 and 2 are used to calculate the physical velocity of a first point, for example the transducer 105 'at the node N113 in FIGURES 2 and 5, in response to a force applied to a second point, i.e. the location of the actuator (s). For the case of the conventional drive illustrated in FIGURE 2, the force is applied to the center point of the conduit 103A, the node N109. For the case of the double drives, illustrated in FIGURE 5, the force is applied symmetrically on both nodes N107 and Nlll. A "symmetrically" applied force is one applied to both nodes with the same magnitude and in one phase with each other. FIGURES 6A and 6B illustrate a comparison of the magnitude and phase of the physical speed of the transducer 105 'for the cases of single and double operation. The data of FIGS. 6A and 6B are generated from the RFR for the model CM300 flowmeter for both individual and dual drive cases. The graph 601 of Figure 6A illustrates the phase of the duct velocity with respect to the frequency in the transducer 105 '(node N113). The graph 602 of FIGURE 6B illustrates the magnitude of the velocity of the conduit with respect to the frequency in the transducer 105 '(node N113). The curve 603 of the graph 601 is the phase of the speed of the conduit at the node N113 in the case of an individual drive located at the node N109. The curve 604 of the graph 601 is the phase of the velocity of the conduit at the node N113 in the case of double drives at the nodes N107 and Nlll, respectively. Observe that there is no difference in the phase of the duct velocity between the two cases of individual, central and double drives. The curve 605 of the graph 602 is the magnitude of the velocity of the conduit at the node N113 in the case of an individual drive placed at the node N109. The curve 606 of the graph 602"is the magnitude of the velocity of the conduit at the node N113 in the case of double drives at the nodes N107 and Nlll, respectively Note that the response in both cases is the same in the first mode bending out of phase at 73 Hz. Also note that the response of the second bending mode out of phase, at 466 Hz is a factor of about 5 less than in the case of the double drive when compared to the case of the individual, conventional drive. This is due to the response of the nodes N107 and Nlll in the second bending mode is smaller than the response of N109, but it is not completely zero, as shown in FIGURE 3. The response of the second bending mode could be further reduced by moving the location of nodes N107 and Nlll to a point where the coefficient of the proper vector for this mode is closest to zero. In this case, this would mean moving the actuator at node N107 towards node N106 and the actuator at node Nlll towards node N112, for example with reference to FIGURE 3. Also observe that the total response, the area under the curve , for the double-drive case, the curve 606 is approximately half that of the individual drive case, curve 605. The total, lower response in the case of the double drive indicates that the proposed mode, ie the first mode of bending out of phase, it is more efficiently excited in the case of the double drive than in the case of the individual drive. The frequency response function and the resulting graphs of FIGURE 6 illustrate a quantitative approach for understanding the advantages of the drive system of the present invention.

Additional Drive Power - FIGURE 8 An additional advantage of the double drives as described above, in particular with respect to the drive scheme of FIGURE 5, is the ability to supply more drive power to the duct 103A. In practice, an actuator in a Coriolis mass flow meter is limited to approximately .5 watts of power for reasons of intrinsic safety. Those skilled in the art of industrial process control are familiar with intrinsic safety requirements. Essentially, these requirements are proposed to ensure that a process control device, such as a Coriolis effect flowmeter, does not expose sufficient energy, either stored or instantaneously, to an explosive environment such that the environment could catch fire. The designers of the Coriolis effect flowmeters are accustomed to relate (in the case of electromagnetic actuators) the drive current, magnetic field resistance and a number of turns of the coil wire to achieve adequate force of the actuator. However, it is sometimes difficult to supply force to an actuator such that the actuator causes the conduit to oscillate sufficiently for proper operation of the flowmeter. This is particularly true for ducts of larger dimensions and the conduits through which a fluid flows within which gas is retained. The system of the present invention is used under these conditions to provide additional drive power to vibrate a conduit. If a dual drive system is designed such that the two actuators are each part of the electrically isolated drive circuits then each actuator can supply approximately .5 watts of drive power to the conduit and still be able to satisfy the intrinsic safety requirements, necessary . FIGURE 8 represents a double drive system as shown in FIGURE 5 as well as the electronic measuring device 20. The electronic measuring device 20 includes drive circuit A 802 and drive circuit B 804. drive 802-804 are electrically isolated from each other and can therefore be treated as separate circuits for intrinsic safety calculation purposes. The drive circuit A 802 is connected to the actuator 501 on the path 806. The drive circuit B 804 is connected to the driver 502 on the path 808. Each drive circuit 802-804 provides up to the maximum allowable power under the safety requirements intrinsic, relevant to its respective actuator 501-502. In this way, each actuator 501-502 can provide, for example, .5 Watts of drive power to conduit 103A.

Excitation of Alternating Modes - FIGURES 7A and 7B Another application of the dual drive system of the present invention is for the excitation of alternate modes. As noted above, the first out-of-phase bending mode is the most common operating mode for existing Coriolis mass flow meters. However, the ingenuity of the present invention is applicable to any geometry of the conduit and to the use of any mode or driven modes. The first out-of-phase torsional mode, for example, can be efficiently excited by the drive system of the present invention. The drive schemes used to generate the data illustrated in FIGURE 6 are incapable of exciting the first torsional mode out of phase. The model CMF300 flowmeter has a first torsional mode out of phase at 184 H and, as seen in FIGURE 6, neither the individual drive system nor the double drive system generates a significant amplitude at this frequency. The central drive scheme, individual can not excite the torsional mode. However, the double drive scheme offers an alternative. The forces in each drive can be done in the same amplitude but in the opposite phase. In other words, the double actuators can be 180 ° out of phase with each other. When the double actuators are out of phase, the first bending mode out of phase is not excited but the first torsional mode out of phase is excited. FIGURES 7A and 7B illustrate the magnitude and phase of the physical speed of the transducer 105 'for the case of double actuation where the drives are 180 ° out of phase with each other. The graph 701 of FIGURE 7A illustrates the phase of the duct velocity with respect to the frequency in the transducer 105 ' (node N113). The graph 702 of FIGURE 7B illustrates the magnitude of the velocity of the conduit with respect to the frequency in the transducer 105 '(node N113). The curve 703 of graph 701 is the phase of the duct velocity at node N113 in the case of the double drives at nodes N107 and Nlll, respectively, where the drives have the same amplitude but opposite phase. The curve 704 of the graph 702 is the magnitude of the duct velocity at the node N113 in the case of the double drives at the nodes N107 and Nlll, respectively, where the drives have the same amplitude but opposite phase. Observe the strong response at 184 Hz, the first torsional mode out of phase, and the lack of a response in the first and second bending modes out of phase.

In this manner, the present invention provides a Coriolis effect flowmeter for driving in the first torsional mode out of phase with actuators placed for the torsion mode drive but not the bending modes.

Alternative Duct Geometry, Exemplary - FIGURES 9-10 The teaching of the present invention is not limited to a double curved duct vibrating sensor. Any number of one or more conduits in any geometrical configuration can benefit from the actuator (s) of the present invention. FIGS. 9-10 provide a further example of the teaching of the present invention. FIGURE 9"represents a finite element model of a Coriolis straight-line flowmeter, double 900. The nodes S101-S117 are indicated along the length of the conduits 902A-B, the conduits 902A-B are linked at least at each end by the tension bar 904 and the tension bar 904. FIGURE 10 illustrates the graph of the shape coefficient of mode 1000 for the 900 flowmeter. For the case where one wishes to operate the 900 flowmeter in the first mode of symmetrical flexion, out of phase, the graph 1000 is examined to place the positions where the first symmetrical bending mode, out of phase is near a maximum amplitude and the second symmetric bending mode, out of phase is near an amplitude The curve 1002 represents the coefficients of the eigenvector for the first symmetrical bending mode, outside the phase of the 900 flowmeter. The curve 1004 represents the coefficients of the eigenvector for the second bending mode. The curve 1006 represents the coefficients of the eigenvector for the first mode of asymmetric bending, out of phase of the 900 flowmeter. Examination of the coefficient chart of the 1000 mode form reveals that approximately half way between the node N106 and the node N107 the first symmetrical bending mode, out of phase is near a position of maximum amplitude and the second symmetrical bending mode, out of phase is near a position of minimum amplitude. The same is true between nodes Nlll and N112. In this way, the actuator 908 is placed between the nodes N106 and N107 and the actuator 908 'is positioned between the nodes Nlll and N112, as shown in FIGURE 10. When the actuators 908 and 908' are excited with equal amplitude and then phase of the first symmetrical bending mode, out of phase is excited and the second symmetrical bending mode, out of phase is not excited or minimally excited. FIGURES 9-10 illustrate the extent of the teaching of the present invention. Although the physical structures depicted in FIGURES 5 and 9 are very different, the corresponding mode shape coefficient graphs of FIGURES 3 and 10 indicate that the teaching of the present invention is applicable to both. The structures represented herein are only exemplary of the teaching of the present invention. The present invention is applicable in any vibrating tube flowmeter or densitometer.

Excitation of Multiple Modes - FIGURE 11 - Sometimes it is desirable to intentionally excite more than one mode. See for example, copending application serial number 08 / 689,839 filed on August 14, 1996, to the assignee Micro Motion, Inc., where two modes are excited and the changes in the ratio of the resonance frequencies of the two modes excited refer to the pressure of the fluid inside the vibrating duct. FIGURE 11 depicts a block diagram of a drive circuit 1100. The drive circuit 1100 resides, with reference to FIGURE 1 or FIGURE 8, within an electronic measuring device 20. The drive circuit 1100 is comprised of a Mode A driving circuit 1102, Mode B driving circuit 1104 and summation stage 1106. Mode A driving circuit 1102 receives a driving feedback signal on path 1108 and produces a driving signal in the frequency of a first mode (Mode A) on the path 1110. The mode B drive circuit 1104 receives a drive feedback signal on the path 1108 and produces a drive signal on the frequency of a second mode (Mode B) on path 1112. The drive signal on path 1110 of the mode A drive circuit and the drive signal on the path 1112 of the mode B drive circuit is the input to the summation stage 1106. The summation stage 1106 operates to linearly combine the two input drive signals to produce the driving signal applied on the path 1114. The signal The drive applied to the path 1114 is applied to the actuator (s) in the vibrating duct.

Referring now to FIGS. 9-11, assume that one wants to excite the flowmeter of FIGURE 9 as long as the first bending mode is out of phase (curve 1002) and the first torsional mode is out of phase (1006). The Mode A drive circuit 1102 is configured to produce a first drive signal at the frequency of the first bending mode out of phase. The Mode B drive circuit 1104 is configured to produce a second drive signal at the frequency of the first torsional mode out of phase. The first and second drive signals are summarized in the summation stage 1106 to produce the drive signal applied on the path 1114. The applied drive signal is fed to the actuators 908 and 908 'in the 900 flowmeter. FIGURE 10 indicates that the actuators 908 and 908 'are properly positioned to excite the first bending mode out of phase and the first torsional mode out of phase. The actuators 908 and 908 'are placed between the nodes S106 and S107 and Slll and S112, respectively. This is a region of, as seen in FIGURE 10, maximum amplitude for both the first torsional mode out of phase and the first bending mode out of phase and a region of minimum amplitude of the second bending mode out of phase. In this manner, the multi-mode drive circuit of FIGURE 11 excites the first bending mode out of phase and the first torsional mode out of phase but not the second bending mode out of phase. Those . Those skilled in the art of Coriolis effect mass flowmeters are familiar with many different ways to generate the drive signals by drive circuits 1102 and 1104. See for example U.S. Patent No. 5,009,109 filed on April 23, 1991 and assigned on its cover to Micro Motion, Inc. and co-pending application serial number 08 / 890,785 filed July 11, 1997, issued to the applicant Timothy J. Cunningham, which is incorporated by this act by reference to the same degree as a idea completely described in the present. Although the specific embodiments are described herein, it is expected that those skilled in the art may design alternative Coriolis effect flow meter drive systems employing the locations of the actuators and the multiple actuators that are within the scope of the following claims either literarily or under the Doctrine of Equivalents.

It is noted that in relation to this date, the best method known to the applicant to carry out the aforementioned invention, is that which is clear from the present description of the invention.

Having described the invention as above, the content of the following claims is claimed as property.

Claims (17)

1. An apparatus for measuring the characteristics of a material flowing through the apparatus, the apparatus having at least one conduit through which the material flows, an actuating means for vibrating at least one conduit as the material flows through the conduit. minus a conduit, sensors fixed to at least one conduit for generating the output signals representing the oscillations of at least one conduit caused by the Coriolis forces when at least one conduit and the material are vibrated by the actuating means and for transmitting the output signals to a signal processor, and the signal processor to generate a measurement of a characteristic of the material that responds to receive the output signals from the sensors, the apparatus is characterized by: the actuator means which is fixed at minus a conduit in a position that is determined to cause the actuator means to substantially maximize the amplitude in at least one desired mode of operation. The coefficients of the eigenvector for at least one desired mode in a plurality of nodes along the conduit and substantially minimize the amplitude in at least one undesired mode as determined from the coefficients of the eigenvector in the plurality of nodes at length of the conduit.

2. The apparatus according to claim 1, further characterized by: a control circuit of the drive for applying a current to the drive means; a first mode circuit in the drive control circuit for generating a first current causing the actuator means to oscillate at least one conduit at a frequency of a first desired mode; and a second mode circuit in the drive control circuit for generating a second current causing the actuator means to oscillate at least one conduit at a frequency of a second desired mode.

3. The apparatus according to claim 2, further characterized by: a summing circuit in the drive control circuit for summing the first current of the first mode circuit and the second second mode mode current in a driving current that is Applies to the actuator means.

4. The apparatus according to claim 1, wherein the actuator means is characterized by: a plurality of actuators each fixed to one of a plurality of positions that have been determined to maximize an amplitude of oscillations of at least one conduit and the material in at least one desired mode of coefficients of the eigenvector in a plurality of nodes along the conduit.

5. The compliance apparatus with claim 4, characterized in that at least one desired mode includes a first mode of bending out of phase.

6. The apparatus according to claim 5, further characterized in that the predetermined locations are also selected to minimize the amplitude for at least one undesired mode.

7. The apparatus according to claim 5, characterized in that at least one of the undesired modes includes a second bending mode out of phase.

8. The apparatus according to claim 5, further characterized by: a plurality of drive circuits each providing a drive current to one of the plurality of actuators.

9. The apparatus according to claim 8, characterized in that at least one of the desired modes includes a first torsional mode out of phase.

10. The apparatus according to claim 9, characterized in that at least one of the undesired modes includes a first bending mode out of phase.

11. The apparatus according to claim 8, characterized in that each of the plurality of drive circuits is electrically isolated from each other.

12. The apparatus according to claim 8, characterized in that the plurality of drive circuits apply the drive currents having amplitudes and phases substantially equal to the plurality of actuators.t.

13. The apparatus according to claim 8, characterized in that at least a portion of the plurality of drive circuits apply drive currents having substantially different amplitudes and phases to the plurality of actuators.

14. A method for attaching an actuator means to at least one conduit of an apparatus for measuring the properties of a material flowing through at least one conduit, the method is characterized by the steps of: extracting the coefficients of the eigenvector of the nodes along at least one conduit of a finite element model of the apparatus for at least one desired mode and at least one undesired mode; selecting a position that maximizes the amplitude of the oscillations of at least one conduit in at least one desired mode and minimizes oscillations in at least one undesired mode of the coefficients of the eigenvector of the nodes; and place the actuator system in position.

15. The method according to claim 14, further characterized by the steps of: selecting the multiple positions that maximize the amplitude in at least one desired mode and minimizing the amplitude in at least one undesired mode.

16. The method according to claim 14, characterized in that the selection step is characterized by the step of: applying the coefficients of the eigenvector to a graph; determine the maximum values in the graph for at least one desired mode and the minimum values for at least one undesired mode.

17. The method according to claim 14, wherein the selection step is characterized by the steps of: generating an FRF; and determine the position based on the FRF.