GB1566790A - Vortex fluid flow meters - Google Patents

Description

(54) VORTEX FLUID FLOW METERS (71) We FLONIC, a French Society Anonyme of 12 place des Etats-Unis, 92120 Montrouge, France do hereby declare invention, for which we pray that a patent may be granted to us, and the method by which it is to be performed, to be particularly described in and by the following statement: The present invention relates to vortex fluid flow meters, in which fluid flow through a conduit is measured by sensing the frequency of vortices generated by an obstacle placed across the conduit.

The generation of a street of vortices by an obstacle placed in a fluid stream is well known, and many flowmetering devices based on this property have already been described.

Indeed it is known that the frequency of emission of the alternating vortices is linearly related to the local speed of flow of the fluid; the coefficient of proportionality depends practically only on the geometric dimensions of the emitting obstacle. The measurement of this frequency of emission is generally done by means of sensors responsive to the emission or the passage of the vortices, most often built into the obstacle and adapted to detect the alternating variations of local pressure which occur at the walls of the obstacle.

Nevertheless, these known devices fail to provide together a ruggedness which permits them to be used in severe environmental conditions, different from the conditions in a laboratory, and accuracy over a large measurement range.

In this connection it is known that the amplitude of the fluctations sensed by the vortex-responsive sensor is a quadratic function of the speed of flow of the fluid, and that consequently this amplitude decreases considerably for low speeds of flow. One of thus led to use a sensor having a high sensitivity in order to be able to detect the existence of the vortices. But this sensor is then also sensitive to the parasitic noises which are found in the environment of the conduit in industrial situations, such as the noise caused by compressors, pressure reducers, valves, turbulence etc. . ., or to the various industrial acoustic moises which propagate in the metal of the conduit or in the fluid itself.

The result is that the vortex-responsive sensors deliver spurious signals, even when the fluid flow is zero.

According to the invention a vortex fluid flow meter comprises conduit means, an obstacle disposed with said conduit means transversely thereof to generate vortices in fluid flowing through said conduit means, said obstacle having two opposed lateral zones past which said fluid flows and subject to variations of pressure due to said vortices; respective pressure pick-off ports in said lateral zones; respective passageways within said obstacle coupling said pressure pick-off ports to respective outlets; and a differential pressure sensor disposed outside said conduit means and coupled to said outlets by flexible tubes, and arranged to generate an electrical signal representative of the frequency of said pressure variations.

As the differential pressure sensor is not built into the obstacle, but is outside the conduit and connected to the passageways by flexible tubes, the sensor is subject only to the differential pressure variations existing at the chosen zones on the obstacle.

More over, it has also been discovered experimentally that the emission of vortices behind an obstacle placed diametrically across a conduit can be disturbed when the frequency of emission of these vortices is capable of generating and maintaining a local resonance. A particularly troublesome resonance, for example, is the establishment of a system of standing waves perpendicular to the direction of the fluid flow and having a half wave-length approximately equal to the diameter of the conduit in which the fluid is flowing. In this case, a whistling noise appears which is accompanied by an abnormal pressure drop at the terminals of the flowmeter; simultaneously, the emission frequency is itself disturbed: in general, it increases by several per cent in value and causes errors in the measurement of the fluid flow.A change in the fluid flow causes a decrease in the intensity of the whistling and the frequency of emission of the vortices becomes metrologically correct once again.

Accordingly the flow meter defined above may include at least one acoustic resonator mounted on said conduit means in the neighbourhood of a median axial plane of said conduit means, which plane is transverse to said obstacle. Preferably. where the conduit means is circular in cross-section, said resonator has a resonant frequency with a wavelength of approximately 7,d/N. where d is the diameter of said conduit means and N is a positive integer.

The effect of the resonator is to attenuate the phenomenon of parasitic resonance, which otherwise disturbs the measurement, and thus enable the flowmeter to maintain the same metrological accuracy over its entire measurement range, including the flow zones in which the disturbing phenomena are produced.

The obstacle may have two opposed lateral faces past which said fluid flows and which are inclined to the axis of the conduit means at an angle greater than 0 and less than 100.

The differential pressure sensor may comprise a fluid flow gauge having a movable member arranged to be actuated by currents offluid induced in at least one direction through said gauge in response to said pressure variations. and means for detecting actuations of said member and for generating said electrical signal in accordance therewith.

The flow meter may further include a circuit for processing said electrical signal generated by said differential pressure sensor, said circuit comprising: preamplifier means arranged to receive said signal; two signal paths connected in parallel to the output of said preamplifier means, a first of the paths being arranged selectively to transmit signals between a predetermined minimum frequency and a predetermined threshold frequency, and the second path being arranged selectively to transmit signals between said threshold frequency and a predetermined maximum frequency:: a first comparator having a first input connected to the output of said first path and a second input arranged to receive a signal respresentative of said minimum frequency; first switch means coupled to the output of said first path and controlled by said first comparator to close when the frequency of said signal is not less than said minimum frequency; a second comparator having a first input connected to the output of said second path and a second input arranged to receive a signal representative of said threshold frequency; and changeover switch means having one input terminal coupled to the output of said second path and the other input terminal coupled to said first switch means, and controlled by said second comparator to couple to its common terminal either said one input terminal or said other input terminal according to whether the frequency of said signal is greater or less than said threshold frequency.

Various vortex fluid flow meters in accordince with the invention will now be described, by way of example, with reference to the accompanying drawings, in which: Figure 1 is an axial section of part of a first flowmeter. the section passing through the median line of an obstacle of the flowmeter; Figure 2 is a transverse section of a second flowmeter similar to that shown in Figure 1, the section being taken along a line corresponding to the line 2-2 in Figure 1; Figures 3 and 4 are sectional and side views respectively of an obstacle of a third flowmeter; Figure 5 is a fragmentary view of a fourth flowmeter; Figure 6 is an explanatory diagram showing the excitation of a standing wave downstream of a vortex shedding obstacle; ; Figure 7 is an axial section of part of a fifth flowmeter, transverse to the obstacle thereof, showing an arrangement for suppressing standing waves; Figure 8 is an axial section of part of a sixth flowmeter similar to the other flowmeters described herein but in which the obstacle is angularly offset; Figure 9 is an axial section of part of a seventh flowmeter similar to the other flowmeters described herein but in which the obstacle is trapezoidal in cross-section; Figure 10 is a transverse section of an eighth flowmeter; Figure 11 is a schematic sectional view of a differential pressure sensor in the flowmeter of Figure 10; Figure 12 is a diagram of a circuit for processing signals produced by a differential pressure sensor in a flowmeter; and Figure 13 is a diagram showing the frequency response of the circuit of Figure 12.

Referring to Figures 1 and 2, a flowmeter is shown comprising a circular conduit element 10 intended to be connected in the usual manner by means of end flanges to a conduit through which the fluid whose flow Is to be measured passes (in the direction of the arrow). A vortex shedding obstacle 11 is mounted diametrically across the conduit element 10 perpendicular to the direction of flow of the fluid; this obstacle 11 is for example rectangular in cross section, or trapezoidal or circular.

The obstacle 11 has at its ends two cylindrical posts 12, 13 of different diameters which fit into two corresponding housings 14, 15 formed at diametrically opposed points on the conduit element 10. The obstacle 11 is maintained in place by means of a bolt 16 for example, which engages a flat 17 on the conduit element 10 to bring the cylindrical posts 12, 13 into contact with complementary shoulders 12a, 13a formed in the conduit element 10. Fluid tightness is achieved by means of seals 18, 19 in grooves formed in the cylindrical posts 12, 13.

The fluid flows in the conduit perpendicular to the plane of Figure 2, and, in known manner, vortices are generated and shed alternately from the two lateral faces 20, 21 of the obstacle 11.

The shedding of these vortices creates differential pressure fluctuations on the obstacle 11 in particular at the center of the fluid stream in the two cross hatched regions 22, 23. Openings or ports 24, 25 are formed in the obstacle 11 in the regions 22, 23 and they communicate by internal conduits or passageways 26, 27 with two outlets 28, 29. Two flexible tubes 30, 31 connect these outlets to a differential pressure sensor 32, for example of the type constructed by VALIDYNE, Northridge California, model DP 15 TL whose response extends from 0 to several kHz, and which converts the differential pressure variations on the faces 20, 21 of the obstacle 11 into corresponding alternating electric signals.The sensor 32 is connected by a cable 33 to a source of electric power, and is connected at its output to a shaping circuit 34, such as a trigger circuit, for example a Schmitt trigger, or a monostable flipflop which supplies uniform pulse signals.

The frequency of these pulses is representative of the instantaneous flow rate of the fluid in the conduit, and is indicated by a frequency meter 35.

Alternatively, or (as is shown in Figure 2) in addition, the pulses are merely counted by a totalizer 36 irrespective of their instantaneous frequency to measure the cumulative flow.

It has been observed that this device operates very well, with a very linear response, even for gaseous fluids and even at high pressure.

In the embodiment shown in Figure 2 there is a single opening 24 or 25 in each lateral face 20, 21 of the obstacle 11, these openings serving as pressure pick-off ports. However, each lateral face of the obstacle can have several openings for sensing the pressure, for example, three openings 25 as shown in dotted lines in Figure 1, these openings communicating with the respective internal conduits 26, 27.

It can be advantageous with certain dirty or aggressive fluids, to cover these openings with a flexible diaphragm, the pressure variations being then transmitted to the differential pressure sensor by the intermediary of an incompressible auxiliary fluid, such as oil.

Figures 3 and 4 show an obstacle incorporating this latter feature. Two chambers of elongated shape 41,42 are provided respectively in the lateral faces 20, 21 of the obstacle 11 and communicate with the internal conduits 26,27 via holes 43, 44. The chambers 41,42 are closed in a fluid tight manner by thin metallic diaphragms 120, 121, for example made of stainless steel, and fixed by welding or any other appropriate means on the faces 20, 21. The whole of the volumes formed by the chambers 41,42, the internal conduits 43, 44 and 26, 27, as well as the flexible tubes (30, 31 in Figure 2) leading to the differential pressure sensor, are filled with oil.In this case, the conduits 26, 27 are extended downwardly and are provided, near their lower ends, with respective filling orifices 45, 46 which can be closed by screws 47, 48.

The differential pressure sensor 32 can advantageously be resiliently mounted, that is mounted on the conduit by the intermediary of an elastic system playing the role of mechanical filter and thus decoupling the sensor 32 from parasitic vibrations which might be propagated in the conduit. Figure 5 shows an example of such mounting in which the flexible tubes 130, 131 connecting the differential pressure sensor 32 to the conduit element 11 have the form of a spiral spring. The sensor 32 is enclosed in a protective casing 49 bolted to the external head l lA of the obstacle 11, and provided with passages for the electrical supply conductors of the sensor and for the output electrical signals derived from the differential pressure variations.

The sensitive element of the sensor may be for example of the diaphragm type which is quasi-static due to its own stiffness, thereby avoiding parasitic vibrations of the sensor at high frequency with respect to the frequency of the signal. Such a sensor also avoids any fluid flow in the flexible tubes connecting the obstacle to the sensor. The pressure variations generated by the shedding of vortices can thus be transmitted to the sensor without intermediate pressure drops decreasing the amplitude of the signal.

On the other hand, if a differential pressure sensor is used which has a freely movable sensitive element, this element should not be damped too much, and should have a bandwidth sufficient to respond to the pressure variations at the rapid frequency of shedding of the vortices.

For high flow rates in conduits of large diameter. the dimensions of the obstacle should be chosen such that the pressure variations occur at low frequencies. for example of the order of 0.1 to 5 Hz. A differential pressure sensor having a small pass band is then suitable.

For flows in pipes of small diameter. the frequency of the pressure oscillations can reach and exceed 1000 Hz. and in this case. the differential pressure sensor chosen should have a response suttlciently rapid so as not to damp the higii frequencies of the signals.

As has been explained above. certain parasitic resonance phenomena can disturb the flow measurement and make the flow meter lose its accuracy at certain flow rates where these disturbing phenomena are produced. Figure 6 shows schematically the existence of a standing wave down stream of a vortex shedding obstacle. This figure shows schematically a cylindrical conduit element 51 in which a vortex shedding obstacle 52 is mounted, the obstacle comprising a body of rectangular cross section for example. the axis of this obstacle 52 coinciding with a diameter of the conduit 51. perpendicular to the plane of the figure.The fluid flows in the direction of the arrow and creates a vortex street of alternating vortices. shown schematically at 53 and 53', which detach themselves from the faces 52A and 5'B of the obstacle 52. A sensor (not shown) of the type described previously detects the frequency of the shedding of these vortices from which frequency the flow measurement can be deduced.

A resonance can occur locally. shown schelllatically by a standing wave 54 having pressure anti-nodes on the wall of the element 51 in the vicinity of the plane of the figure and a pressure node on the axis. The fluctuating pressure existing in the wake of the vortex shedding obstacle 5'. whose frequency can be different from the frequency of the standing wave. is responsible for sustaining the standing wave.

Figure 7 shows an arrangement for avoiding such standing waves. In the wall of the element 51. in a direction perpendicular to the axis of the obstacle 52. i.e. in the plane of the figure, there are formed openings 59. each opening receiving a short tube 55 whose outer end is provided with a thread 56. On this thread is screwed a hollow body forming an acoustic resonator. of the Helmhotz type 57 (spherical or hemispherical). or alternatively of the stationary wave type 58. of cylindrical shape and terminated by a plane of reflection. The specific frequency of the cavity thus formed is adjusted by screwing the body 57 or 58 along the tube 55. The dimensions of these bodies have been shown on a large scale to facilitate the understanding of the figure: they are in reality of dimensions much smaller than the diameter of the conduit 51.

These absorption resonantors can be disposed at any point on the wall in the plane of the figure where troublesome acoustic reflections are produced: for example opposite the obstacle 52. like the resonator 58. downstream like the resonator 57, or even upstream. The number of these resonators is not critical and is chosen as a function of the quality of the desired damping of the parasitic resonance.

A conduit element equipped with a vortex shedding obstacle can also be provided with non-adjustable cavity resonators. which are tuned once and for all. since for an obstacle and a conduit of well determined geometry there is a corresponding resonant cavity which also has well defined dimensions. These cavities can times be machined or welded permanently on to the flowmeter constituted by the conduit element equipped with its obstacle.

The acoustic impedances of these resonators can be tuned to a wave-length approximately double the diameter of the conduit element. or to a submultiple of this length: that is. to a wavelength of approximately 2d/N, where d is the diameter of the conduit and N is a positive integer.

In a practical example in which the fluid conduit was 80mm in diameter and the obstacle had a frontal height of 16mum. the appearance of a parasitic resonance was observed at a frequency of about 430 Hz. This parasitic frequency was effectively attenuated by means of a resonator tuned to about 1940 Hz.

Althougli in the preceding examples the obstacle is rectangular in section. this shape is not essential. The Strouhal number S. given by S = f.d/V where f is the vortex frequency d is the width of the obstacle transverse to the fluid flow V is the fluid flow rate remains constant for a large range of flowrates when a rectangular obstacle is used: in other words, the vortex frequency is closely proportional to the fluid speed. However. it has been found that for a Reynolds number less than 20.000. that is. for example. for low flows of gas at low pressure. the average vortex frequency would be higher than the theoretical value derived from the above equation. Thus the error of the flowmeter expressed as a function of flowrate is constant for gas pressures above about 4 bars. but rises by several per cent at low flowrates. causing an over-estimate when the pressure of the gas being metered is very low.

This type of error is particularly disadvantageous in relation to the calibration of gas meters, because these meters are generally calibrated at atmospheric pressure though they will subsequently be applied to the metering of gas under pressure.

Analysis of the flowmeter signal with a spectrum analyser has confirmed that the rise in error is not due to the appearance of parasitic oscillation, but rather is due to an inherent property of turbulent flow.

We have dound that the error characteristic of this type of flowmeter at low gas pressure can be substantially improved by modifying sliettly the geometry of the obstacle.

Accordingly the obstacle, which may be rectangular or trapezoidal in section, is aranged with its lateral faces turned towards the wall of the conduit through an angle, relative to the conduit axis, which is greater than Oe and less than 10". It appears that particularly good results are obtained with and angle of 4 .

If the obstacle is rectangular, it is turned in its entirety through the desired angle relative to the conduit axis; if it is trapezoidal, the lateral faces of the trapezium cross-section are inclined at the desired angle relative to the axis of symmetry (which is also the axis of the conduit).

It has been found that when a flowmeter is equipped with such an obstacle, over e;timation at low flowrates and low gas pressures is substantially reduced, and that vortex generation appears more stable and regular and provides improved measurement behaviour.

Referring to Figure 8, there is shown at 10 a cylindrical conduit traversed by a fluid whose flow is to be measured; a vortex-generating obstacle 11 is disposed transversely of the conduit along a diameter thereof, and is associated with one or more senors as described above (not shown) responsive to the generation or the passage of vortices. This obstacle 11 is rectangular in section, but instead of being oriented symmetrically with its lateral faces l lA, 11B parallel to the conduit axis, it is displaced about its axis so that the lateral faces 1 lA, l lB point towards the wall of the conduit 10, at an angle a of 4" relative to the axis AA of the conduit 10.

Good results have been obtained with a rectangular section having a ratio of its smaller dimension d' to its larger dimension d of about 0271 and a ratio of its larger dimension d to the diameter D of the conduit 10 of about 0.21.

Since the faces 11A, 1 lB both subtend the same angle with the axis AA, the obstacle shown in Figure 8 operates for fluid flow in either direction along the conduit 10.

The obstacle 124 shown in Figure 9 is oriented symmetrically about the axis AA of the conduit 10, but its cross-section is shaped as a trapezium, the larger end face 124C being upstream (the fluid flows in the direction of the arrow F), and each lateral face 124A,124B being at an angle a of 40 to the axis AA. The ratios defined above have vides of d'ld = 0.63 and d/D = 0.23, d' being in this case the altitude of the trapezium.

Although a value of 40 for the angle a has been found to provide particularly good results, any value greater than 0 and up to 10 , either in the case of a rectangular section (Figure 8) or in the case of a trapezoidal section (Figure 9), provides improvement in the metrological properties of the vortex emission frequency in conditions of low signal measurement described above.

It can be seen that the linear relationship between the vortex emission frequency f and the rate of fluid flow V is given by f = S.V/d, where d is the width of the obstacle transverse to the fluid flow and S is Strouhal's number, which generally has a value of about 0.21. In practice, we have found that the generation of vortices is particularly reliable metrologically, that is, the frequency fis closely proportional to the flowrate V, if the dimension d of the obstacle is of the order of 1/5 of the diameter of the conduit.

In the case of large-diameter conduits, for example of the order of 1 metre in diameter, this formula for the dimension d results in a frequency f of 1 Hz for a flowrate of 1 metre per second. This frequency is low enough to permit the use of vortex sensors of a different type to those used in the past.

It has been proposed, for example, to employ sensors of the thermal, piezo-electric, capacitive, inductive and ultra-sonic types.

These sensors, being analogue, require an appropriate shaping circuit to permit derivation from their output singals of the instantaneous flowrate and the cumulative flow by means of frequency meters and totalizers respectively.

In contrast, the new type of sensor described below is able to provide directly, without special electronic processing, discrete signals at the vortex frequency representing the flow to be measured. This sensor involves a fluid gauge having a movable member which is actuated by successive currents, in at least one direction, of fluid traversing the gauge in response to generation of a vortex, the movable member being associated with means to detect and count actuations of the member.

The fluid gauge (which is itself a fluid flow meter) can be of either of the kinds for measuring speed and volume respectively, in which a movable member causes rotation of an output shaft of the gauge, such as a turbine gauge, paddle-wheel gauge or oscillating arm gauge. The gauge is coupled via its inlet and outlet to the two pressure pick-off ports on the obstacle.

The gauge can be either of two types: reciprocatory, in which the movable member is actuated in one direction or the other in accordance with the direction of fluid flow in the gauge; or non-reciprocatory, in which the member can only be actuated in the direction corresponding to fluid flow from the inlet of the gauge to the outlet.

The use of a fluid gauge, such as a domestic water gauge, as a pressure sensor provides the following advantages: as a differential pressure sensor it is very sensitive: calculation shows that in a water gauge responsive to a minimum flowrate of 10 litres/hour, the corresponding pressure drop between inlet and outlet is between 1F and 10 millibar. Other differential pressure sensors do not combine similar simplicity and reliability with this sensitivity; it is of simple unitary construction, and is thus virtually insensitive to shocks and external vibration at high frequency.

Alternate fluid currents derived via the pressure pick-off ports from the generation of every other vortex on one face of the obstacle actuate the movable member, thereby turning the output shaft of the gauge in a given direction;-the generation of the intervening vortices on the other face of the obstacle either causes turning of the shaft in the other direction if the gauge is reciprocatory. or no movement if the gauge is non-reciprocatory. In the first case, (reciprocatory), the means for detecting actuation of the member is arranged to detect actuation in both directions. In the second case (non-reciprocatory) this means is arranged to detect actuations of the member which result from fluid currents through the gauge in the forward direction, that is for those vortices which cause a current to flow from inlet to outlet.

The frequency of actuation of the movable element is either the same as the frequency of vortex emission for a reciprocatory gauge, or half this frequency for a non-reciprocatory gauge. In either case, the actuation frequency is representative of the instantaneous flowrate of the fluid in the conduit, and the total number of actuations is representative of the cumulative flow.

The means for detecting actuation of the movable member can be any one of a variety of detectors: electrical, magnetic, optical, etc.

It is not necessary for the output shaft of the gauge to perform a large number of revolutions for each actuation, since it is the number of commencements of actuation of the movable member which is significant, rather than the number of revolutions as such performed by the output shaft. As an example, in one embodiment the detecting means comprises two end-stops which limit, to an angle less than 3600, the arcuate movement of an arm secured to the output shaft. Each end-stop has an associated proximity switch to provide a signal when the arm reaches the respective limit of its movement.These proximity switches could be, for example,: electrical (micro-switches); magnetic (a magnet carried by the same or another arm on the output shaft and arranged to operate two reed relays or Hall effect detectors); optical (a mirror carried by the same or another arm on the output shaft and arrange to reflect a light beam onto a photo sensor); or proximity sensors (inductive, capacitive, etc).

If it is only required to detect rotation of the output shaft in one direction (as in the case of a non-reciprocatory gauge), the detecting means need have only one proximity switch, associated with the end-stop which checks angular movement of the arm when the gauge is traversed by a fluid current in the forward direction. The arm is urged back to a rest position by a return spring in the absence of such currents.

Referring to Figure 10. there is shown a conduit 10 traversed by fluid to be measured, and an obstacle 11 mounted transversely across this conduit 10. The vortex-generating obstacle 11 has a suitable cross-section, such as a rectangular or trapezoidal section appropriate to the generation of distinct vortices. This obstacle 11 is secured in the conduit 10 by a bolt 212, the conduit 10 having housings provided with faces which engage complemtary shoulders on the obstacle 11, as previously described with reference to Figure 2.

Fluid flows along the conduit 10 at right angles to the plane of Figure 10, and vortices are generated alternately in the zones 213, 214 adjacent the lateral faces 215,216 of the obstacle 11. Openings 217,218 in the faces 215.216 constitute pressure pick-off ports, which are coupled to a differential pressure sensor 223 via internal passageways or conduits 219, 220 and external tubes 221, 222, in order to detect alternating pressure changes associated with the generation of vortices.

The differential pressure sensor 223 comprises a fluid gauge, for example a domestic water meter of the reciprocatory turbine type, in which a turbine is actuated to rotate alternately in one direction and then the other in accordance with the aformentioned alternating pressure changes. The alternating actuations in opposed senses of the turbine are detected by a detector 224 associated with the water meter, to provide signals representative of the frequency of generation of vortices (and thtis representative of the flow in the conduit 10) to a frequency meter 225 and a totalizer 226. Consequently the frequency meter 225 indicates the instantaneous flowrate, and the totalizer 226 indicates the cumulative flow.

Figure 11 is a view from above of the detector 224 mounted on the turbine chamber of the gauge. The output shaft 230 of the turbine of the gauge 223 carries an arm 231 whose angular travel is limited by two endstops 232,233. Associated with each of these end-stops 232, 233 is a proximity switch 234, 235 of one of the types discussed above: for example, a normally-open micro-switch. These switches 234, 235 are mounted adjacent the end-stops 232, 233 to operate when the arm 231 engages the respective one of the end stops 232, 233, and are electrically connected to the indicators 225, 226 of Figure 10 by leads 236.

In operation, with fluid flowing through the conduit 10, each vortex generated on one of the lateral faces of the obstacle 11, for example the face 215 (Figure 10), is accompanied by a pressure change thereon which, in turn, causes a current of fluid to traverse the gauge in the circuit comprising the passageway 219, the tube 221, the gauge 223, the tube 222 and the passageway 220.

The turbine in the gauge 223 is turned by this current until the arm 231 is arrested by an end-stop, for example the end-stop 232, whereupon the associated proximity switch 234 is closed to send a signal to the indicators 225 and 226.

When the next vortex is generated, a current is produced in the opposite direction in the circuit 219,221,223, 22',220, so the turbine of the gauge 223 is turned back again. The arm 231 is thus brought into contact with the other end-stop 233, operating the proximity switch 235 and sending another signal to the indicators 225, 226, and so on for each succeeding vortex.

If the gauge 223 were non-reciprocatory, the movable member of the gauge would only be actuated by fluid currents in one direction, for example those urging the arm 231 against the end-stop 232. The other proximity switch 235 would therefore be omitted, and a spring fitted so that, in the absence of a current through the gauge, the arm 231 would be urged against the end-stop 233. In this way, only movements corresponding to appropriate currents through the gauge would be registered, that is, those corresponding to one vortex in two.

As mentioned above, the proximity switches could be any of many different types, and, if sufficiently robust, could also act as the end stops.

In order to protect aganist the components of parasitic frequencies noise or vibration from various sources, which in certain circumstances, and notwithstanding the arrangements described above, can still affect a vortex sensor, it is advantageous to be able to extract the usable signal at the fundamental frequency representing the actual fluid flowrate from parasitic signals originating in the fluid flow and in external electrical sources.

To this end, the circuit described below operates selectively on the vortex detector signals in accordance with whether the signals are nearer the lower limit of the measured range or nearer the upper limit.

The circuit has two signal paths, respectively for low- and high-frequency signals, and output comparators which control switches so that: the signal from one of the two paths is coupled to the output of the circuit, in accordance with the relative values of the measured flowrate and a threshold flowrate; or the signals from both paths are blocked, in accordance with the relative values of the measured flowrate and a predetermined minimum flowrate.

The two paths are designed to provide frequency responses having a sharp cut-off in the vicinity of the frequency representing the threshold flowrate, so that their combination represents a bandpass filter having a bandwidth in two portions, one on each side of the threshold frequency, of which only one portion is operative at a time.

Referring to Figure 12, the circuit 310 includes an input 311 coupled to the output of a flowmeter sensor 312. The circuit 310 also includes, in general terms, a preamplifier 313 to provide adequate signal level for the remainder of the circuit; two signal paths, for low and high frequencies respectively, designated overall I and II, which are connected in parallel to the output of the preamplifier 313 and will be described in more detail hereinafter; and a switching circuit comprising two analogue comparators 314 and 315, respectively connected to the outputs of the paths I and II and which operate, respectively, a switch 316 connected to the output of the path I, and a changeover switch 317.This switches 317 has two inputs, one connected to the output of the switch 316 and the other connected to the output of the path 11, so that it can couple the output 318 of the circuit 310 either to the switch 316 or to the path II. The output 318 is connected to a frequency meter and a totalizer as already described to indicate respectively the instantaneous flowrate and the cumulative flow.

Considering the circuit in more detail, the path I includes a low-pass RC network 321, having a resistor 322 coupled to a capacitor 323 and a resistor 324 in parallel.

This network 321 feeds a low-frequency amplifier 325 comprising a differential amplifier and an RC negative feedback circuit 326. The output signal of the path I is supplied by a monostable circuit 327 coupled to the amplifier 325 via another low-pass Rc network 328/329. This monstable circuit 327 has a quasi-stable period T and generates output pulses of uniform amplitude and duration.

Similarly, the high-frequency path II has a high-pass CR network 331 with a capacitor 332 and a resistor 333; in addition, two rectifiers 334 are connected in opposite senses in parallel with one another and with the resistor 333. The network 331 feeds a highfrequency amplifier 335 comprising a differential amplifer and a resistive negative feedback circuit 336. The output signal of the path II is supplied by a monostable circuit 337 coupled to the amplifier 335 via another high-pass CR network 338/339. The monostable circuit 337 has a quasi-stable period t and generates output pulses of uniform amplitude and duration.

At the outputs of the paths I and II, the uniform pulses from the monostable circuits 327 and 337 are converted to respective d.c.

voltages proportional to their repetition frequencies. by two RC integrating networks 341, 342, the time constant of the network 341 being longer than that of the network 342.

These two d.c. voltages Vl3l and V}1 l. are supplied to respective differential amplifiers constituting the respective comparators 314 and 315.

The comparator 314 also receives a d.c.

voltage V,njn corresponding to the minimum frequency Frnin representing the minimum flowrate to be measured. The output of the comparator 314 is coupled to the control terminal of the switch 316, which, for example, is an MOS transfer gate.

The comparator 3 15 receives a d.c. voltage V0 corresponding to the threshold frequency F,,and its output is coupled to the control terminal of the changeover switch 3 17, which is similarly formed by semiconductor devices.

Operation of the circuit 310 will be described with reference to Figure 13 which shows, as a function of frequency, the responses T1 and Til of the paths I and ll, with sharp cutoffs adjacent the threshold frequency F,,and their combined response TR. The curve A indicates the variation of signal amplitude as a quadratic function of frequency.

The low-frequency path I eliminates parasitic components at frequencies above the threshold frequency Fo, by means of the cascaded low-pass filters formed by the circuits 321, 325, 328/329. and converts signals below this frequency Fo into pulses of period T by means of the monostable circuit 327. The period T is selected in accordance with the range of flowrates to be measured. and is typically of the order of several milliseconds in duration.

Similarly, the passband of the high-frequency path II results in signals at frequencies below F0 being strongly attenuated in this path to reduce the effect of spurious low-frequency signals which become significant when the flowrate increases and which result from phenomena in the flow itself and from mechanical vibration.

This attenuation is provided by the cascaded filters formed by the circuits 331,335.

338/339. In particular. the efficacy of the filter 331 is enhanced by the rectifiers 334 whose resistance decreases when the input signal amplitude increases, thereby reducing the time constant of the network 331. This results in an increase in the low-frequency cutoff of the filter when the signal amplitude A increases (as a result of an increase in flowrate).

Thus the separation of the required highfrequency signal from the spurious low-frequency signals becomes more pronounced as the input signal amplitude rises.

The monostable circuit 337 converts signals at frequencies above F0 into pulses t, chosen. in accordance with the maximum flowrate. to be of the order of a millisecond in duration.

The durations of the period T and t of the monostable circuits 327 and 337 are chosen to ensure that, for the respective range of frequencies, these circuits 327 and 337 cannot be spuriously triggered by brief parasitic pulses occurring between two required signals and having an amplitude which falls below and then rises back above the triggering level of the respective circuit 327 or 337. If there were only a single signal path, it would be necessary to choose a quasi-stable period t such that the period 2t would be less than the period of the signal corresponding to the maximum flowrate, in order to avoid loss of parts of the required maximum-flowrate signal; there would then be inadequate protection against spurious signals when the flowrate were low, unless the range of measurement of the meter were severely restricted.By providing two frequency-selective paths. it is possible to select quasi-stable periods t and T for the monostable circuits 327 and 337 which differ by a factor between 5 and 10, thereby providing protection against spurious signals over a wider frequency range embraced by the respective frequency bands of the paths I and 11.

The output pulses from the monostable circuits 327 and 337 are subsequently converted into d.c. voltages V13- and VlI l proportional to the respective pulse repetition frequencies by the filters 341 and 342. The time constant of each filter 341,342 is chosen such that the energy supplied by any occasional lone spurious pulses is insufficient to charge the filter capacitor to a voltage capable of triggering the respective comparator 314,315.

In order to prevent against errors caused by spurious signals in conditions of zero or very low flow. outside the measuring range of the flowmeter. the d.c. reference voltage Vmin is set at a level corresponding to the lower limit of the flowmeter's range, and is compared in the comparator 314 with the voltage VBI supplied by the filter 341. The resulting output signal of the comparator 314 opens the switch 316 if VBF < V min and closes it if V min < V Bi:.

Thus, if the flowrate falls below the lower limit, or if spurious signals occur (typically as infrequent groups of small numbers of pulses) when the flowrate is zero or otherwise below the lower limit, as may often happen in vortex flowmeters, output pulses are prevented from reaching the output 318.

The voltage VI is compared in the comparator 3 15 with the reference voltage V0 corresponds to the threshold frequency Fro . When VHF < VO, the comparator 315 operates the changeover switch 317 to couple the output of the switch 316 to the output 318, thereby disconnecting the path II. If, on the other hand, VHF < V0, the comparator 315 operates the changeover switch the other way so as to couple the output of the path II to the output 318, thereby disconnecting the path I.

Thus, depending on whether the flowrate is a) very low, b) between the lower limit of the flowmeter and the threhold flowrate, or c) greater than this threhold flowrate, there appears at the output 318 either a) no signal, b) the signals derived by the path 1, or c) the signals derived by the path 11.

By way of example, goods results have been obtained with a circuit for processing vortex flowrnOter signals and having the following parameters: range of measurement 10 -- 300 pulses/s threshold frequency Fo : 55 Hz quasi-stable periods : t -- 1 ms T -- 6.5 ms: WHAT'WE CLAIM IS: 1.A vortex fluid flow meter comprising: conduit means; an obstacle disposed within said conduit means transversely thereof to generate vortices in fluid flowing through said conduit means, said obstacle having two opposed lateral zones past which said fluid flows and subject to variations of pressure due to said vortices; respective pressure pick-off ports in said lateral zones; respective passageways within said obstacle coupling said pressure pick-off ports to respective outlets; and a differential pressure sensor disposed outside said conduit means and coupled to said outlets by flexible tubes, and arranged to generate an electrical signal repre sentative of the frequency of said pressure variations.

2. A meter as claimed in claim 1, wherein each pressure pick-off port is closed by a flexible diaphragm, and said ports, said passage ways and said flexible tubes are filled with an incompressible fluid.

3. A meter as claimed in claim 1 or claim 2, wherein said differential pressure sensor is mounted on said conduit means by elastic mounting means to decouple said sensor from vibration of said conduit means.

4. A meter as claimed in claim 3, wherein said flexible tubes are of elastic material and constitute said mounting means.

5. A meter as claimed in any one of the pre ceding claims, including at least one acoustic resonator mounted on said conduit means in the neighbourhood of a median axial plane of said conduit means, which plane is transverse to said obstacle.

6. A meter as claimed in claim 5, wherein said conduit means is circular in cross-section, and said resonator has a resonant frequency with a wavelength of approximately 2d/N, where d is the diameter of said conduit means and N is a positive integer.

7. A meter as claimed in claim 5 or claim 6, wherein said resonator is mounted opposite said obstacle.

8. A meter as claimed in claim 5 or claim 6, wherein said resonator is mounted offset along said conduit means with respect to said obstacle.

9. A meter as claimed in any one of claims 5 to 8, wherein the resonant frequency of said resonator is adjustable.

10. A meter as claimed in any one of the preceding claims, wherein said outlets are disposed outside said conduit means.

11. A meter as claimed in any one of the preceding claims, wherein the obstacle has two opposed lateral faces past which said fluid flows and which are inclined to the axis of the conduit means at an angle greater than 0 and less than 10 .

12. A meter as claimed in claim 11, wherein said faces are inclined to said axis at an angle of 13. A meter as claimed in claim 11 or claim 12, wherein said obstacle is trapezoidal in crosssection and is symmetrically disposed about the axis of the conduit means with its larger end face facing upstream.

14. A meter as claimed in claim 13, wherein the ratio of the altitude of the trapezoidal section to the width of said larger end face is substantially equal to 0.63.

15. A meter as claimed in claim 13 or claim 14, wherein the ratio of the width of said larger end face to the diameter of said conduit means is substantially equal to 0.23.

1 6. A meter as claimed in any one of claims 1 to 12, wherein said obstacle is rectangular in cross-section.

17. A meter as claimed in claim 16, wherein the ratio of the lateral faces of the rectangular section to the end faces is substantially equal to 0.71.

18. A meter as claimed in claim 16 or claim 17, wherein the ratio of the end faces of the rectangular section to the diameter of said conduit means is substantially equal to 0.21.

19. A meter as claimed in any one of the preceding claims, wherein said differential pressure sensor comprises a fluid flow gauge having a movable member arranged to be actuated by currents of fluid induced in at least one direction through said sauge in response to said pressure variations, and means for detectong actuations of said member and for generating said electrical signal in accordance therewith.

20. A meter as claimed in claim 19, wherein said detecting means comprises a shaft arranged to be rotated in response to actuation of said member, an arm carried by said shaft. at least one end-stop arranged to limit arcuate move ment of said arm, and at least one proximity sensing means adjacent said end-stop to provide a signal when said arm engages said end-stop.

21. A meter as claimed in claim 20, wherein said gauge is reciprocatory and said detecting means includes a second end-stop arranged with said one end-stop to limit,arcuate movement of

**WARNING** end of DESC field may overlap start of CLMS **.

Claims (39)

1. **WARNING** start of CLMS field may overlap end of DESC **.

   way so as to couple the output of the path II to the output 318, thereby disconnecting the path I.

   Thus, depending on whether the flowrate is a) very low, b) between the lower limit of the flowmeter and the threhold flowrate, or c) greater than this threhold flowrate, there appears at the output 318 either a) no signal, b) the signals derived by the path 1, or c) the signals derived by the path 11.

   By way of example, goods results have been obtained with a circuit for processing vortex flowrnOter signals and having the following parameters: range of measurement 10 -- 300 pulses/s threshold frequency Fo : 55 Hz quasi-stable periods : t -- 1 ms T -- 6.5 ms: WHAT'WE CLAIM IS: 1.A vortex fluid flow meter comprising: conduit means; an obstacle disposed within said conduit means transversely thereof to generate vortices in fluid flowing through said conduit means, said obstacle having two opposed lateral zones past which said fluid flows and subject to variations of pressure due to said vortices; respective pressure pick-off ports in said lateral zones; respective passageways within said obstacle coupling said pressure pick-off ports to respective outlets; and a differential pressure sensor disposed outside said conduit means and coupled to said outlets by flexible tubes, and arranged to generate an electrical signal repre sentative of the frequency of said pressure variations.
2. 2. A meter as claimed in claim 1, wherein each pressure pick-off port is closed by a flexible diaphragm, and said ports, said passage ways and said flexible tubes are filled with an incompressible fluid.
3. 3. A meter as claimed in claim 1 or claim 2, wherein said differential pressure sensor is mounted on said conduit means by elastic mounting means to decouple said sensor from vibration of said conduit means.
4. 4. A meter as claimed in claim 3, wherein said flexible tubes are of elastic material and constitute said mounting means.
5. 5. A meter as claimed in any one of the pre ceding claims, including at least one acoustic resonator mounted on said conduit means in the neighbourhood of a median axial plane of said conduit means, which plane is transverse to said obstacle.
6. 6. A meter as claimed in claim 5, wherein said conduit means is circular in cross-section, and said resonator has a resonant frequency with a wavelength of approximately 2d/N, where d is the diameter of said conduit means and N is a positive integer.
7. 7. A meter as claimed in claim 5 or claim 6, wherein said resonator is mounted opposite said obstacle.
8. 8. A meter as claimed in claim 5 or claim 6, wherein said resonator is mounted offset along said conduit means with respect to said obstacle.
9. 9. A meter as claimed in any one of claims 5 to 8, wherein the resonant frequency of said resonator is adjustable.
10. 10. A meter as claimed in any one of the preceding claims, wherein said outlets are disposed outside said conduit means.
11. 11. A meter as claimed in any one of the preceding claims, wherein the obstacle has two opposed lateral faces past which said fluid flows and which are inclined to the axis of the conduit means at an angle greater than 0 and less than 10 .
12. 12. A meter as claimed in claim 11, wherein said faces are inclined to said axis at an angle of
13. 13. A meter as claimed in claim 11 or claim 12, wherein said obstacle is trapezoidal in crosssection and is symmetrically disposed about the axis of the conduit means with its larger end face facing upstream.
14. 14. A meter as claimed in claim 13, wherein the ratio of the altitude of the trapezoidal section to the width of said larger end face is substantially equal to 0.63.
15. 15. A meter as claimed in claim 13 or claim 14, wherein the ratio of the width of said larger end face to the diameter of said conduit means is substantially equal to 0.23.
16. 1 6. A meter as claimed in any one of claims 1 to 12, wherein said obstacle is rectangular in cross-section.
17. 17. A meter as claimed in claim 16, wherein the ratio of the lateral faces of the rectangular section to the end faces is substantially equal to 0.71.
18. 18. A meter as claimed in claim 16 or claim 17, wherein the ratio of the end faces of the rectangular section to the diameter of said conduit means is substantially equal to 0.21.
19. 19. A meter as claimed in any one of the preceding claims, wherein said differential pressure sensor comprises a fluid flow gauge having a movable member arranged to be actuated by currents of fluid induced in at least one direction through said sauge in response to said pressure variations, and means for detectong actuations of said member and for generating said electrical signal in accordance therewith.
20. 20. A meter as claimed in claim 19, wherein said detecting means comprises a shaft arranged to be rotated in response to actuation of said member, an arm carried by said shaft. at least one end-stop arranged to limit arcuate move ment of said arm, and at least one proximity sensing means adjacent said end-stop to provide a signal when said arm engages said end-stop.
21. 21. A meter as claimed in claim 20, wherein said gauge is reciprocatory and said detecting means includes a second end-stop arranged with said one end-stop to limit,arcuate movement of

    said arm to less than 3600 and a second proxim ity sensing means associated therewith.
22. 22. A meter as claimed in claim 20, wherein said gauge is non-reciprocatory and said detecting means includes a return spring operative, in the absence of fluid currents in the gauge, to urge tile arm against a second end-stop.
23. 23. A meter as claimed in any one of the preceding claims, further including a circuit for processing said electrical signal generated by said dlfterentia! pressure sensor. said circuit f smiiriss reamplifier means arranged to receive said signal: two signal paths connected in parallel tu the output nf sai l preamplifier means, a first of the paths bering arranged seiectiveiy to transmit signals between a predetermined minimum frequency vnd a pfedetermined threshold frequency, and the oeco1ld.path being arranged selectively to transmit signais between said threshold fre qu ic:; and a predetermined maximum f. equellcy: a first comparator having a first in ,.ut connected to the output of said first path and a second input arranged to receive a signal representative of said minimum frequency:: first switch means coupled to the output of said first path and controlled by said first comparator to close when the frequency of aid signal is not less than said minimum fre quench.

    a second comparator having a first input connected to the output of said second rath and a second input arranged to receive a signal representative of said threshold fre quencv; and cnnulgeovel switch means having one input tenninal coupled to the output of said second path and the other input terminal coupled to said first switch means, and controlled by said second comparator to couple to i common terminal either said one input terminal or said other input terminal according to whether the frequency of said signal is greater or less than said threshold frequency.
24. 24. A meter as claimed in claim 23, wherein each signal path comprises the series combination of a filter network, a band-limited amplifier and a monostable circuit.
25. 25. A meter as claimed in claim 24, wherein the fitter network of said second path comprises a capacitance and resistance in series, twe rectifying means being connected each in parallel with the resistance and in opposite sense to one another.
26. 26. A meter as claimed in claim 24 or claim 25, wherein the respective monostable circuits of the two paths have different quasi-stable state periods.
27. 27. A meter as claimed in claim 26, wherein the quasi-stable state period of the monostable circuit of said first path is between five and ten times that of the monostable circuit of said second path.
28. 28. A meter as claimed in any one of claims 24 to 27, wherein the first input of each comparator is connected to the output of the respective path via a respective integrating cir cuit to provide a voltage proportional to the repetition frequency of pulses supplied thereto by the respective monostable circuit.
29. 29. A meter as claimed in claim 28, wherein said integrating circuit associated with said first path has a time constant longer than that of the integrating circuit associated with said second path.
30. 30. A vortex fluid flow meter substantially as hereinbefore described with reference to Figure 1 of the accompanying drawings.
31. 31. A vortex fluid flow meter substantially as hereinbefore described with reference to Figure 2 of the accompanying drawings.
32. 32. A vdrtex fluid flow meter substantially as hereinbefore described with reference to igure 3 and 4 of the accompanying drawings.
33. 33. A vortex fluid flow meter substantially as hereinbefore described with reference to Figure 5 of the accompanying drawings.
34. 34. A vortex fluid flow meter substantially as hereinbefore described with reference to Figure 7 of the accompanying drawings.
35. 35. A vortex fluid flow meter substantially as hereinbefore described with reference to Figure 8 of the accompanying drawings.
36. 36. A vortex fluid flow meter substantially as hereinbefore described with reference to Figure 9 of the accompanying drawings.

    ts
37. 37. A vortex fluid flow meter substantially as hereinbefore described with reference to Figure 10 of the accompanying drawings.
38. 38. A vortex fluid flow meter substantially as hereinbefore described with reference to Figures 10 and 11 of the accompanying drawings.
39. 39. A vortex fluid flow meter as claimed in any one of claims 30 to 38, including a signal processing circuit substantially as hereinbefore described with reference to Figures 12 and 13 of the accompanying drawings.