WO1983000220A1

WO1983000220A1 - Flow controller and flow sensor

Info

Publication number: WO1983000220A1
Application number: PCT/US1982/000875
Authority: WO
Inventors: Inc. American Flow Systems; Imad Mahawili; Timothy J. Boyle; Norvell A. Nelson
Original assignee: American Flow Systems Inc
Priority date: 1981-07-09
Filing date: 1982-06-28
Publication date: 1983-01-20
Also published as: EP0083633A1

Abstract

Structure for measuring the flow of fluid comprises a flow passage (16) through which fluid whose flow is being measured is passed and a tapered pin (2) slidably mounted in the flow passage, for moving in response to the flow of fluid. The amount of motion of the tapered pin is a function of the flowrate of the fluid. A restoring force is placed on the pin by means of a spring (6) thereby to restore the pin to its nominal position so as to block the flow of fluid in the absence of fluid flow. The pin is mounted so as to move a magnet (4), the motion of which is sensed by a sensing element, typically a Hall effect sensor (10). The output signal from the Hall effect sensor is then processed to provide a measure of the flowrate. In one embodiment, the processing of this signal from the Hall sensor is done by first digitizing the signal, then comparing the value represented by this digitized signal to values of flowrate versus signal value stored in a matrix in a memory as a function of pressure and then interpolating to find the actual flowrate. The flow sensor (3) operates on an impact principle and has a concave, cup-like shape (3A) on the surface impacted by the fluid to ensure the smooth flow of this fluid over the surface.

Description

FLO CONTROLLER AND FLOW SENSOR

FIELD OF THE INVENTION

This invention relates to an electronic flow control system using a flow impact sensor in conjunction with electronic signal processing data and electronic control circuitry.

Prior Art

Flow meters are well known. A summary of positive displacement flow meter prior art is given in the intro¬ duction to the copending patent application Serial No. 127,918 filed March 6, 1980 entitled "Electronically Controlled 'Flow Meter and Flow Control System", assigned to American Flow Systems, Inc., the assignee of this application. The '127' application discloses a positive displacement flow sensor using a flexible diaphragm in combination with ancillary electronic circuitry for con¬ tinuously measuring the flow rate of a fluid and for generating from the measured flow rate a control signal for use in controlling a flow control valve. While the above described flow meter uses the displacement of a diaphragm as a function of time to measure the flow rate, other flow meters make use of the momentum of the fluid whose flow is being measured to determine flow rate. Thus, U.S. Patent 3,805,611 issued April 23, 1974 entitled "Fluid Flow Meter" on an application of Hedland discloses a fluid flow meter with a piston slidably disposed within a chamber having an opening therein for the flow of fluid therethrough. A spring urges the piston^'in one direction and fluid pressure urges the piston in the opposite direction. A conically shaped piece is disposed axially of the ^"piston and in the opening therein so that the axial position of the piston varies in accordance with the rate of fluid flow through the opening. A magnet on the piston actuates a follower on the exterior of the chamber so that as the piston is moved in response to fluid flow, the magnet on the piston moves the follower on the exterior of the chamber to indicate the fluid flow.

Another flow meter of a type similar to that disclosed in the '611 patent is disclosed in U.S. Patent 4,254,664 issued March 10, 1981 entitled "Flow Meters" on an appli¬ cation of Graham. In the '664 patent, a contoured plug is axially mounted in an orifice in a flow channel of generally circular cross-sectional flow with the plug being normally pressed toward an input section so as to close the flow passage. The plug is tapered curvilinearly in cross-section with the curvature being s.uch as to produce a linear relationship between differential pressure across the annulus formed by the plug and the orifice and the flow through the annulus.

A similar structure is disclosed in U.S. Patent 4,181,835 issued January 1, 1980 on an application of Stadler, et. al., entitled "Gas Flow Indicator Having a Magnetic Field Sensitive Switch That Is Responsive to the Position of a Magnet Secured to a Piston". In the '835 patent, the piston used for measuring the flow is generally cylindrical in shape.

U.S. Patent 4,195,581 issued April 1, 1980 application entitled "Armored Rotameter" on an application of Fees discloses another type of structure described as "capable of accurately measuring low flow rates." This structure includes a metal body provided with a bore defining a vertical flow tube having a lower inlet into which a fluid to be measured is admitted and an upper outlet from which

OMPI SΛ. _ BO the fluid is discharged. A pin rides up and down a slot as a float moves up and down the tube as a function of flow rate. The pin is coupled to one end of a bar magnet link external to the meter body, the link being caused to swing to an extent and in a direction in accordance with float movement whereby an external bar magnet pointer follows the bar magnet link to indicate flow rate on the external scale. A portion of the flow tube in which the pin moves is tapered.

SUMMARY OF THE INVENTION

In accordance with this invention, a flow controller is provided incorporating a flow sensor which operates over both a "laminar" and what is called an "impact and discharge" flow regime-

In accordance with this invention, a flow sensor is provided which comprises a flow tube in one end of which is partially inserted a slidably moveable tapered pin. The tapered pin is pressed into the flow tube by means of a spring loaded base plate with a curved cup-like re¬ ceptacle for that end of the tapered pin extending out of the flow tube. The curved, cup-like receptacle is designed to completely reverse the direction of fluid flowing past the pin so as to prevent pressure variations sometimes existing in prior art impact sensors due to fluid impacting upon the plate and then moving perpendicularly to the direction of inlet fluid flow across the plate thereby creating a variable pressure^' region on the plate and iri erferring with the flow measurement. Upon the end of the base plate opposite to that where the carved cup extends a magnet is mounted. The base plate travels linearly in space as a result of the displacement of the tapered pin. A Hall effect sensor is mounted on the outside of the flow chamber to detect motion of the magnet. In accordance with one embodiment of this invention, the impact sensing structure comprises a cantilevered beam upon the free end of which is mounted a magnet. The magnet travels in a curvilinear motion as a function of the impact of the fluid on the beam. A Hall effect sensor is mounted on the outside of the flow chamber to detect motion of the magnet. The magnet moves in such a manner as to produce an output signal which partially corrects for the non-linearities inherent in the system.

In accordance with another embodiment of this in¬ vention, the sensors provided in accordance with this invention are used in accordance with a signal condi¬ tioning circuit and either a digital or an analog proces¬ sing circuit to produce either a digital or an analog measure of the flowrate of the fluid whose flow is being measured at a selected temperature and pressure.

In accordance with another embodiment of this in¬ vention, means are provided for compensating for pressure and temperature variations in the fluid whose flow is being measured, thereby to provide a measure of actual mass flow.

In another embodiment of this invention, a micro¬ processor is employed to process the signals generated by the flow sensor. This embodiment is particularly useful in the measurement of the mass flowrate of compressible fluids. To determine actual mass flow rate of a compres¬ sible fluid under conditions of variable pressure and temperature, temperature and pressure sensors are also required. The microprocessor therefore receives input signals representing flow data from the flow sensor, pressure from the pressure sensor and temperature from the temperature sensor. It processes these input signals to produce a measure of the actual mass flow. When this embodiment is used in a flow controller, this measure is

OMH SNATlb compared to the desired flow which is input to the system by the-operator. The control valve is then adjusted so as to eliminate the difference between actual flow and desired flow. If desired, the actual flow rate is displayed to the operator.

In accordance with this invention a friction elimi¬ nation mechanism is provided and is selectively activated to ensure accurate readings from the flow sensor. This entire process is repeated continuously. If flow control is not required, this embodiment can be reduced to a pressure and temperature compensated mass flow meter by eliminating the control valve circuitry.

Description of the Drawings

Figures la, lb and lc illustrate three different embodiments of the flow transducer mechanical portion of the flowmeter of this invention;

Figure 2 is a curve illustrating the output signal from a typical sensing device appropriate for use with this invention as a function of flowrate for one parti¬ cular fluid;

Figure 3 illustrates the output signal from a Hall effect sensor suitable for use with this invention versus flowrate for helium and nitrogen gases for a pin 2 having a taper of 0.062 "to 0.050";

Figure 4 illustrates the output signal from a Hall effect sensor suitable for use with this invention versus flowrate for watery-

Figure 5 illustrates the output signal from a Hall effect sensor as a function of pressure ^'for a number of constant flowrates of nitrogen; Figure 6 illustrates the temperature correction factor associated with the output signals from the flow- meter of this invention as a function of different flow- rates;

Figure 7 illustrates the electronic circuitry associ¬ ated with an embodiment of this invention used to convert the signals from the flow sensor, temperature sensor and pressure sensor to digital form and to generate a control signal in response thereto;

Figure 8 illustrates the signal conditioning circuit for use with the flow transducers disclosed in Figure 1 and the pressure sensor 101a disclosed in Figure 7;

Figure 9 illustrates the circuit used to produce a linear analog output signal from the output signal of the signal conditioning circuit disclosed in Figure 8;

Figure 10 illustrates the circuitry used to produce and display a digital output signal from the output signal of the signal conditioning circuit disclosed in Figure 8;

Figure 11 illustrates the relationship between absolut pressure and the pressure transducer output signal; and

Figure 12 is of use in illustrating the manner in which the flow sensor output signal and the pressure sensor output signal are used to generate a flow rate through three interpolations.

A Detailed Description

This invention will be more fully understood in con¬ junction with the following detailed description taken together with the above described drawings:

OMPI Figure la illustrates the structure of the mechanical portion of one embodiment of this invention. In Figure la, fluid whose flow is to be measured is passed into the structure through an input tube or nozzle 1 from a fluid source 16. In the other end of input nozzle 1 is partially inserted a tapered pin 2 carefully designed to provide a selected range of variable flow areas between the inner diameter of inlet nozzle 1 and the outer diameter of taper 2. Taper 2 is freely mounted and is therefore capable of moving relative to inlet 1. As taper 2 moves out of inlet 1, the flow area between the outer diameter of taper 2 and the inner diameter of inlet nozzle 1 increases thereby allowing a larger flow rate. The end 2a of taper 2 rests on the top surface 3A of deflector tube and impact plate 3.

As fluid enters input nozzle 1, the fluid forces taper 2 out of nozzle 1 against surface 3A of deflector tube and impact plate 3. The inner surface 3B of impact plate 3 (end face 3A of plate 3 is a portion of inner surface 3B) is preferrably carefully curved and contoured to reflect and turn 180° the fluid which impacts against " plate 3 after passing taper 2 so that this fluid then flows back up past taper 2 and the outside of tube to the exit connector 15. Connector 15 allows the fluid to flow from the chamber formed in housing 8 and in which is mounted impact plate 3, and a portion of nozzle 1. However, this inner surface 3B can also assume other shapes if desired provided the degradation in accuracy of flow measurement associated with a different shape is acceptable. The shape as shown of surface 3B actually eliminates vibration of plate 3 which commonly would be present should the fluid not be deflected as shown. Such vibrations arise from the fluid flow regimes that are capable of being obtained when impact plate 3 assumes other con¬ figurations in combination with the other elements of the structure shown. I pact plate 3 is mounted by means of spring 6 onto nozzle flange 17 which is mounted firmly in housing 8 by means of retainer nut 7. "0" rings 9 seal cavity la to insure that there is no leakage into or out of this cavity. Spring 6 insures that impact plate 3 is drawn closely adjacent to input nozzle 1 when no fluid is flowing thereby to bring tapered pin 2 firmly into nozzle 1. In some embodiments taper 2 can penetrate sufficiently far into input nozzle 1 to block the flow of fluid. However this is not essential and taper 2 can, in some instances, be inserted only partly into nozzle 1 thereby allowing an annular area between the outside of pin 2 and the inside of nozzle 1 through which fluid can flow.

Mounted on the bottom portion of impact plate 3 is a magnet 4. Magnet 4 is covered by either an elastomer such as Viton or a welded piece of steel to protect magnet 4 from corrosion by the fluid. Placed on the bottom of housing 8 is a printed circuit board 11 on the inner face of which is mounted a Hall effect sensor which can, in one embodiment, be of the type known as a linear output Hall effect transducer or "LOΞET". Any other Hall effect - sensors can also be used as appropriate.. The deflection of the plate 3 may be detected by the use of optical methods, capacitance techniques, or reed switches for on/off flow sensing devices, or by direct coupling of plate 3 to a variable potentiometer for direct sensing of flowrate by measuring change in potentiometer resistance. Printed circuit board 11 is mounted on the bottom of housing 8 by means of screws 12 and is spaced from housing 8 by spacers 13.

Impact plate 3 is held in cavity 1A by spring 6. Spring 6 is sized so that in its nominal position of rest a sufficient space 14 is left between the bottom of impact plate 3 and the bottom surface 8A of housing 8 to allow impact plate 3 to move freely within its expected range of

J\TREA^ OMH

^Z°ι -9- motion. Typically, this distance is about 0.100 inch. The spring.rate and the taper geometry of pin 2 are selected to control the minimum and maximum flowrates through the system. In one embodiment suitable for measuring the flow of nitrogen, the spring rate was selected to be 0.9 lb/in and a taper was selected to yield a minimum measured flowrate of about 1 cc/min and a maximum measured flowrate of about 300 cc/min.

Figure IB illustrates another structure of this invention wherein the input nozzle 1 and the taper 2 are substantially as shown in the embodiment of Figure 1A. However, taper 2 is allowed to rest freely on a canti¬ levered arm 3 at the free moving end of which is mounted a magnet 26. The magnet again is preferably coated with a protective material, such as Viton, to protect the magnet from the fluid (possibly corrosive) whose flow is being measured. Cantilever arm 23 is mounted on mounting block 25 by screw 24. Magnet 26 is mounted upon the free moving end of arm 23 so as to be capable of motion relative to a sensing device 10 (preferably a Hall effect sensor) mounted on the inner surface of printed circuit board 11. Printed circuit board 11 is mounted on the bottom of housing 8 in the same manner as the comparable board is mounted on the structure shown in Figure 1A. Spacer 25 is sized to hold magnet 26 at a selected distance above the bottom surface of housing 8.

The structure shown in Figure 2B operates in a manner similar to that shown in 1A except that when fluid enters the input nozzle 1 and thereby displaces taper 2 to allow fluid to enter cavity 1A in housing 8, and to exit from output connecter 18, cantilevered arm 23 is depressed. The movement of cantilevered arm 23 changes the relative locations of magnet 26 and Hall effect sensor 10 thereby changing the signal produced by sensor 10. The output signal from sensor 10 can be calibrated to the flowrate for each type of fluid, the flow of which is being measured. A substantial benefit is obtained from the structure shown in Figure IB from the fact that the motion of magnet 26 is nonlinear in relation to flow in such a manner as to tend to correct the output signal from Hall sensor 10 in the direction of being more linearly related to flowrate.

In a typical embodiment used to measure the flow of nitrogen, magnet 26 sweeps over an arc of approximately 0.100 inches in response to a flowrate which can vary from about 1 to 300 cc/min for example.

Cantilever arm 23 can be designed to have a number of different sizes to respond to a number of different flow- rates just as the spring 6 in the embodiment of Figure 1A can be similarly designed to accomodate a number of differen flowrates.

Figure 3C illustrates a third embodiment of the mechanical portion of this invention. 'In Figure 3C the structure is to some extent similar to that shown in Figure 1A and IB but differs where numbered differently. In Figure 3C fluid enters input nozzle 1 and displaces a tapered pin 2. Taper 2 is mounted on a support 34 which is held by spring 37 to the bottom portion 38 of housing 8. Spring-retaining bottom plate 38 is mounted above the bottom portion of housing 8. Formed on the bottom portion of housing 8 outside the housing is printed circuit board 11 mounted as is the printed circuit board shown in Figure 1A. Hall effect sensor 10 is mounted on the inner surface of printed circuit board 11 so as to detect the motion of impact plate 34. Impact plate 34 contains as a portion thereof a magnet 36. Again magnet 36 can be coated with a material such as Viton or a welded steel plate to protect corrosive liquids from attacking it. Spring 37 is selected to have a desired spring rate and together with the taper selected for taper 2 to control the range of flowrates capable of being measured by the structure shown in Figure 3C. Fluid which enters cavity 1A exits through opening 18 as shown.

Typical curves of fluid flowrate versus output signal from the sensor 10 for use with the mechanical structure shown in Figures 1A, IB and 1C are shown in Figures 2, 3, 4, 5 and 6. It should be understood that these curves represent, however, different regimes of fluid flow through the device. One feature of this invention is that it is capable of measuring flowrates for gases as low as one (1) cubic centimeter per minute, a flowrate which heretofore has been difficult to achieve. Yet the invention is capable not only of measuring flowrates at this level but also flowrates several orders of magnitude higher. Thus the structure of this invention operates in what is characterized as the laminar regime (corresponding to a flowrate of approximately 1 to 200 cubic centimeters per minute of nitrogen), and the impact and discharge regime corresponding to flowrates from about 200 cubic centimeters per minute up through the maximum flow capability of the device. The surprising and nonobvious results of this invention are achieved in all regions but with particular emphasis upon the laminar region wherein reproducible and accurate flow measurements are obtained for flows as little as 1 cubic centimeter/min for gases. To enhance reproducibility, a low frequency (two hundred to three hundred Hertz) vibration device is used intermitently to reduce internal friction in the structure and thereby obtain a more rapid and accurate response to changes in flows. In the laminar region, the force on taper 2 is predominantly the friction force of the fluid flowing past the taper. The flowrate is necessarily extremely low within the laminar flow regime and thus the drag forces created on the taper are the only substantial forces tending to move the taper out of input nozzle 1 thereby to allow the passage of the fluid. During the motion of the

gFREX; O PI_

^ . \J^ fluid the spring, (such as spring 6 shown in Figure 1A, cantilevered beam 23 shown in Figure IB, or spring 37 shown in Figure 1C) is sufficiently strong to restrain taper 2 from being forced out of nozzle 1. Thus there can be a slight oscillatory motion set up between the fluid attempting to flow past taper 2 and the spring attempting to restore taper 2 to its normal position. The equilibrium point is that which allows the flow of fluid in the steady state condition and this point is approached quite rapidly. No visible vibration appears in the flow using this system. Even when dither or vibration is used in an attempt to insure frictionless performance of the taper, the vibration is introduced at a frequency substantially different from the frequency of oscillation of the components of the flowmeter and thus does not appear in the fluid.

During the impact and discharge regime, the fluid flowrate is large enough that the momentum of the fluid is sufficiently high to be of greater significance than the friction forces. During the transition from the laminar regime to the impact and discharge regime, of course the two forces are substantially of the same magnitude and thus taper 2 responds to two different kinds of forces the friction drag forces along the skin of the taper and the impact forces of the fluid striking the taper and having its momentum changed as a result of the impact. During the higher flowrates, the impact flow regime predominates with the result that the taper behaves in a substantially quadratic manner according to theory.

To achieve the performance of this device in the laminar regime, the taper is carefully designed to insure proper response of the system over the range of flow in this regime. Thus in one embodiment particularly adapted for measurement of the flow of nitrogen over a flow range from one cubic centimeter per minute to 300 cubic centi¬ meters per minute, the internal diameter of the nozzle was .062 inches. A 0.5 inch taper 2 was used having a base outer diameter of 0.062 inches and a top outer diameter of 0.055 inches for a total taper of about 0.5°. A flowrate greater than 300 cc/min results in less accuracy and reproducibility because of slight deformation of the spring as it approaches the deformation limits defined by Hooke's law. Figure 4 shows two curves of the flowrate of water versus Hall sensor output signal for two different tapered pins 2 having the dimensions shown in Figure 4.

Taper 2 was fabricated of materials such as Teflon, Delrin, Kel-F or other plastics. Dissimilar materials are used for the taper 2 and nozzle 1 to prevent "lockup" of the taper in the nozzle, otherwise known as "galling." However, taper 2 and nozzle 1 can be made of the same material, if desired. In one embodiment using the same materials for taper 2 and nozzle 1, a 100° recess (made using a conical drill having a cone angle of 100°) in the nozzle tip eliminated the galling. The polishing of this recess after its formation removes sharp edges and results in taper 2 moving smoothly into and out of nozzle 1. Because both taper 2 and nozzle 1 are of the same material, adverse effects (such as taper 2 locking in nozzle 1) due to differential thermal expansion of the components when measuring the flow of fluid at other than the design temperature are eliminated.

The sensors described above in conjunction with Figures la, lb and lc can be used in a digital flow meter, an analog flow meter, and a flow controller (either digital or analog preferably digital for ease of implementation). The structure shown in Figures 8 and 10 depicts a digital flow meter at standard temperature and pressure. The structure shown in Figures 8 and 9 depicts an analog flow meter as standard temperature and pressure. When the sensing structure shown in Figures la, lb and lc are used to sense the fluid flow rate, the output signals from the structures of Figures 8 and 10 and Figures 8 and 9 repre¬ sents a mass flow rate at standard temperature and pressure or at some other single temperature and pressure for which the system is calibrated. Should it be desired to measure actual mass flowrate, the temperature and pressure of the fluid (in the case where the fluid is compressible) must be measured and the deviations in these parameters from the values of these parameters for which the system is calibrated must be taken into consideration. The structure shown in Figure 7 provides one means for carrying out this correction. While the structure in Figure 7 depicts in total a flow controller, by eliminating digital to analog converter 108, set point input 109, control valve amplifier 112 and control valve 114, the remaining circuitry function to correct the flowrate sensed by flow sensor 101c in accordance with the fluid temperature and pressure sensed by sensors 101b and 101a, respectively, and, in a manner to be described, to correct the flowrate for the actual temperatures and pressures of the fluid. The structure in Figure 7 does this through a digital processing technique. While this is the preferred embodiment of this invention, these corrections could also be made using analog circuitry.

Figure 8 illustrates the electronic circuit used in conjunction with the transducers disclosed in Figures 1A, IB or 1C to condition the signal from Hall sensor 10 prior to transmitting the signal to either an analog-to-digital converter or to an analog system. As shown in Figure 8 the structure comprises a precision voltage reference 20, such as the H00701 manufactured by National Semiconductor Corporation, to provide the required supply voltage to Hall sensor 21 (product 91SS12-2 manufactured by Micro Switch Division of Honeywell). Resistor Rl and potentio¬ meter R2 serve as a "zero adjust" circuit. By adjusting the setting of potentiometer R2 (typically 5K) the zero point on the output curve from Hall sensor 21 (see Figure 5) can be adjusted to a desired value. The output signal from Hall sensor 21 is passed through resistor R4 (10K) to the inverting input lead of operational amplifier 22 (LM747). The non-inverting input lead to operational amplifier 22 is connected to the adjustable potentiometer R2. The output signal from ampli¬ fier 22 is transmitted through resistor R6 (10K) to the inverting input lead of op amp 23. The non-inverting input lead of op amp 23 is connected through resistor R9 (10K) to ground. The output signal from op amp 23 is the output signal from the signal conditioning circuit and is designated as "A" in Figure 8. In the analog flow meter embodiments of this invention, output signal A is trans¬ mitted to the analog linearization circuit shown in Figure 9. In the digital flow meter embodiment of this invention, output signal A is transmitted to the digital conversion and display circuit shown in Figure 10. In the pressure and temperature compensated digital flow controller embodi¬ ment of this invention, output signal A is transmitted to analog multiplexer 103 shown in Figure 7. The op amp 22 contains a feedback circuit comprising parallel-connected capacitor C4 (0.1 microfarad) and resistor R5 (10K ohms). Op amp 23 includes a feedback circuit comprising series -connected resistors R7 (adjustable 50K ohms) and R8 (68K ohms) in parallel with capacitor C5 (0.47 microfarads). Adjustable resistor R7 is used to adjust the gain of the signal conditioning circuit shown in Figure 8. Since the transducers shown in Figure 1 are, in general, nonlinear, the output signal A of the signal conditioning circuit is also, in general, nonlinear as shown in Figure 11 (which shows the output signal from the Hall sensor associated with pressure sensor 101A).

Figure 9 illustrates the analog linearization and output circuit used with this invention when the structure shown in Figures 1A, IB or 1C is used as an analog flow meter. The analog circuit in Figure 9 produces a zero to 10 volt linear output signal from the generally non-linear output signal from the signal conditioning circuits 101 shown in Figures 7 and 8. Analog multiplier 31 (AD 534 commercially available from Analog Devices Inc.) is con¬ nected as a squarer to square the output signal A from amplifier 23 shown in Figure 9 and produce an output signal C₂A 2 on output pin 12. The output signal C₂A2 from multiplier AD534 is applied across resistors R12 (10K) and R13 (2k) to produce an output signal C₃C_A 2. Signal

C₃C₂A is transmitted to the inverting input lead of operational amplifier 32 (the well known M747) which is connected as a summing inverter. Resistor (potentiometer)

R13 is adjustable to allow the magnitude of the constant

C3 to be varied and thereby to allow the linearity of the final output signal from amplifier 33 to be adjusted.

2 Signal C₃C₂A from square circuit 31 is added by op amp 32 to another signal C,A derived by passing the signal A from amplifier 23 across resistors RIO and Rll (17K ohms) to provide a linear output signal from amplifier 33 given by C,A + C₃C₂A 2. The gain of op amp 32 is set to unity by two parallel-connected resistors R16 and R17 each of 100K ohms. The output signal from op amp 32 is transmitted through resistor R18 (100 ohms) to the inverting input lead of inverting op amp 33 ( M747N). The gain of op amp 33 is unity and is controlled by feedback resistor R19 (100K ohms). Op amp 33 merely inverts the output signal from op amp 32 to the proper polarity. The output signal from op amp 33 is approximately linear in the range of 0 to 10 volts and provides a direct linear measure of the flowrate sensed by the flow transducer.

Mathematically, the output signal from op amp 23 (Fig. 8) is denoted as A which equals f(p) (i.e. which is a function of the sensed flow). The output signal from amplifier 33 has its shape controlled by the constants C, , C₂ and C₃- Constant C, is determined by the values of the two series connected resistors R10 and Rll. In the embodi¬ ment described above, R10 is approximately 83K ohms and Rll is approximately 17K ohms thereby making C-^ equal approximately 0.17. C₂ is determined by the internal structure of multiplier 31 and typically has a value of 0.1. C₃ is determined by the setting of the wiper arm or adjustable terminal on potentiometer R13 having a total resistance of 2K ohms. The wiper arm on potentiometer R13 is set to give a value for C- of from 0 to 0.16. To actually set the value of C~ , several different known flowrates are sequentially passed by the flow transducer and the linearity adjust (the wiper arm on potentiometer R13) and the gain adjust (the wiper arm of the potentio¬ meter R7 connected as a variable resistor in the feedback circuit of amplifier 23) are adjusted to yield the proper linearity. The procedure can be iterative requiring several flowrate measurements and adjustments before proper linearity is achieved.

Figure 10 illustrates the digital circuit used with the digital flow meter embodiment of this invention. In Figure 10, the output signal A from the signal conditioning circuit of Figure 8 derived from the output lead of op amp 23 is applied to the input lead 13 of analog to digital converter 41 (comprising the well known AD574 circuit commercially available from Analog Devices Inc.). Converter 41 is clocked by the output of the LM555 timer 42 (National Semiconductor Corporation) to provide a periodic conversion to digital form of the analog output signal from the signal conditioning circuit shown in Figure 8. The output signal from converter 41 is a digital signal of twelve bits representing any one of 2 12 possible different signal levels. The eleven most significant outputs of converter 41 are used to address the 2K byte electrically program¬ mable read only memories (EPROMs) 43 and 44 (for example, the 2516 commercially available from Texas Instruments Inc.) to provide an output signal corresponding to the information stored in the appropriate EPROM at the address identified by the digital eleven bit output signal from converter 41. The information stored in EPROMS 43 and 44 corresponds to the digitized version of the voltage versus curve shown in Figure 11 when the output signal from op amp 23 (Figure 8) is derived from signal conditioning circuit 102A (Figure 7) and represents pressure when derived from op amp 23 in signal conditioning circuit 102C (Figure 7). In curve of Figure 11, the output voltage from the signal conditioning circuit shown in Figure 8 is plotted versus the flowrate. The output signal from the appropriate address in EPROM 43 or 44 (Figure 10) cor¬ responds to the digital representation of the flowrate sensed by the structure of Figures 1A, IB or 1C and is transmitted in binary coded decimal to displays 45, 46 and 47 to provide a display of three digits corresponding to the flowrate. Thus, the address of the location in memory in which this digital representation of the flowrate is stored also corresponds uniquely to this flowrate. Displays 45, 46 and 47 comprise the well known 5082-7302s manu¬ factured by Hewlett-Packard Co.

Another embodiment of this invention is a flow con¬ troller with pressure and temperature compensation. Of course, this embodiment can be reduced to a flow meter with pressure and temperature compensation by eliminating the control circuitry such as circuit elements 108, 109, 112 and 114, if it is so desired.

Figure 7 illustrates in block diagram form, the electronic circuitry of one embodiment of this invention used as a flow controller. As shown in Figure 7, micro¬ processor 105 receives as input signals certain processed data from pressure sensor 101A, temperature sensor 101B and flow sensor lOlC. Pressure sensor 101A and temperature sensor 101B are each of a type commonly known in the art. Flow sensor 101C is of the type previously described in this specification in conjunction with Figures 1A, IB and IC. Pressure sensor 101a can be any pressure sensor of a type commonly employed in the measurement of fluid pressures and preferably will be a pressure sensor of the type shown in copending application serial no.Q6/?Ri .620 entitled "Pressure Transducer" on an application of Boyle, Nelson and Mahawili filed the same day as this application and assigned to American Flow Systems, Inc. the assignee of this application.

The signals produced by sensors 101A, 101B and 101C are transmitted to signal conditioning circuits 102A, 102B and 102C respectively. Signal conditioning circuits 102A and 102C comprise the structure as shown in Figure 8 (described infra). The signal conditioning circuit shown in Figure 102B comprises a potentiometer connected so that the output current from the temperature sensor is passed through the potentiometer. The magnitude of the input signal generated across the potentiometer is merely a function of the value of the resistance presented by the potentiometer to this current. This current is proportional to temperature.

Analog multiplexer 103 routes a desired output signal from one of signal conditioning circuits 102A, 102B and 102C to analog-to-digital converter 104. A-to-D converter 104 converts this output signal to a digital signal. Multiplexer 103 and A-to-D converter 104 are driven by output signals from microprocessor 105 generated in response to the internal program of microprocessor 105.

Microprocessor 105 is operated under control of in¬ structions in program memory 106. Memory 106 also contains data to be used by the system in producing signals indi¬ cative of the mass flow rate of the fluid whose flow is being measured from the measurements of the pressure and temperature of the fluid and the flow rate passing the flow sensor 101C. Program memory 106 comprises a read only memory ("ROM") of well known design or an electri- cally programmable read only memory.

Address decoder 107 receives signals from program memory 106 and from microprocessor 105 for accessing or providing instructions to multiplexor 103, analog converter 104, memory 106 and certain components associated with the control portion of the circuitry and the user input and output portions of the circuitry comprising digital to analog converter 108, set point input 109 and output display 110.

Digital to analog converter 108 takes the output signal from microprocessor 105 representing in digital form the desired control voltage to be applied to the control valve amplifier 112. Amplifier 112 controls control valve 114 to control the flow rate being sensed by flow sensor 101c. This signal input to D-to-A converter 108 is produced by microprocessor 105 from the difference between the mass flow rate computed from the pressure, temperature, and flow sensor inputs and the set point input represented by the settings of input switches 109 of well known design (shown schematically only). The analog signal from converter 108 applied to power amplifier 112 results in amplifier 112 producing an output signal adequate to drive control valve 114.

Output display 110 displays the measured flowrate.

As a feature of this invention, a friction eliminating component is provided for use with the mechanical structure depicted above in Figures 1A, IB and 1C. Friction eliminato amplifier 111 provides power to a friction eliminator 113 which comprises a vibrating element. In one embodiment, friction eliminator 113 comprised a motor which vibrates owing to an unbalance in the rotating components of the motor. The data bus 116 is the structure over which all data transactions between the microprocessor 105 and the other components of the circuit take place. The control bus 117 is a group of signal lines, controlled by the microprocessor 105, which selects the circuit element to be activated for a particular data transaction with the microprocessor 105. The control bus 117 also includes signal lines for starting an analog-to-digital conversion and selecting which of the three sensors 101A, 101B or 101C is to be routed through the analog multiplexer 103 to the analog to digital con¬ verter 108. Another control line is responsible for turning on the friction eliminator amplifier 111 which drives friction eliminator 113.

Instructions and calibration data for the micro¬ processor 105 are contained in the program memory 106 and addressed through the address bus 115 which is also used by the address decoder 107 to derive some of the signals on control bus 117.

Each of the three sensors, flow 101C, pressure 101A and temperature 101B, is connected to a signal condi¬ tioning circuit 102C, 102A and 102C, respectively, which provides the offset and amplification to produce a $ to Iβ volt signal from each sensor. These signals are connected to the analog multiplexer 103 which, in response to control signals, determines which output signal from circuits 102A, 102B and 102C will be connected to the analog to digital converter 104. The analog to digital converter 104 periodically and repetitively converts the appropriate output signal to a digital form that can be processed by the microprocessor 105.

The output display 110 consists of a standard multi- digital LED display and data latches activated by signals on control bus 117.

OMPI -r. ^° ,^*^ The setpoint input 109 consists of a standard multi- digital thumbwheel switch with binary coded decimal outputs.

The control valve 114 requires a variable analog - voltage signal to be generated by the microprocessor 105. This is accomplished by connecting digital to analog converter 108 to the data bus and connecting a power amplifier 112 to the output of converter 108. Micro¬ processor 105 sends a digital signal to converter 108 which converts this signal to an analog voltage. Amplifier 112 applies the voltage to control value 114 at an increase power level.

The friction elimination circuit 111 is turned on and off by a signal on control bus 117 under microprocessor control. Circuit 111 is ^"simply a power transistor operated as a switch to turn the power to the friction eliminator mechanism 113 on or off.

Microprocessor 105 computes the mass flow rate from the magnitude of the input signals from the flow'sensor, pressure sensor, and temperature sensor. In the case when - the fluid being measured is a liquid, the effects of pressure and temperature are minor and the flow sensor output can be used directly to compute the mass flow. This is done by storing in the memory 106 the mass flow rate corresponding to selected output signals from flow sensor 101C. Subsequently, when the output signal from flow sensor 101C is measured, the mass flow rate is compute from the nearest data in memory corresponding to an output signal from flow sensor 101C greater than the measured one, and the nearest data in memory corresponding to an output signal from flow sensor 101C less than or equal to the measured one. The mass flow rate for the measured output signal from flow sensor 101C is computed by a linear interpolation between these two points. In the case when the fluid being measured is a gas (a compressible fluid), the flow sensor output signal and the pressure and temperature of the gas are all required to compute the mass flow rate. Ignoring, for the moment, the effects of temperature, the mass flow rate of a gas is a complex function of the flow sensor output signal and the actual pressure of the gas. That is, for a given flow sensor output signal, the mass flow rate is a function of the gas pressure, and for a given pressure the mass flow rate is a function of the flow sensor output signal.

In order to be able to evaluate the mass flowrate from the flow sensor output signal and the measured gas pressure, a matrix of mass flow rates corresponding to selected values of flow sensor output signals and measured pressures is stored program memory 106 (Figure 7). This matrix can be visualized as a standard two dimensional matrix in which the rows correspond to selected values of the output signal from flow sensor 101C, and the columns correspond to selected values of gas pressure. The value of each point or location within the matrix equals the mass flowrate for a flow sensor output signal corresponding "^* to the row of the matrix location and a pressure corres¬ ponding to the column of the matrix location.

The mass flow rate for any value of output signal from flow sensor 101C and pressure can be computed from the values stored in the matrix by interpolating among the four matrix values nearest to the measured point-

For example, Figure 12 shows the relationship of certain data points such as A,B,C, and D, representing known mass flow rates corresponding to selected values of the output signal from flow sensor 101C and the pressure. In terms of the matrix of mass flow data in program memory 106, selected flow sensor 101C output signals such as SI and S2 correspond to rows of the matrix and selected pressures such as PI and P2 correspond to columns within the matrix. Therefore, memory 106 contains mass flow rate data for the points (PI, 52) (PI, 51) (P2, 52). For example, point A refers to memory location which contains the true mass flow rate (excluding the effects of temper¬ ature) for a output signal of Si from flow sensor 101C and a pressure of PI. In general, an actual measurement of either the flow sensor 101C output signal or the pressure will fall between two of the selected points for which data has been stored. This is represented in Figure 12 by points S and P. S falls between SI (the next lower sensor output for which data has been stored) and 52 (the next higher sensor 101C output signal for which data has been stored) . P falls between PI (the next lower pressure for which data has been stored) and P2 (the next higher pressure for which data has been stored). The intersection of S and P is denoted in Figure 12 by X. This is the point which would contain the true mass flow due to flow sensor 101C output signal S and a pressure P if the memory had data for this point. In general,, the memory does not have data for this point but has data for points such as A,B,C, and D. A way of computing the mass flow rate corresponding - to point X from the available data, is to perform three interpolations among the four points shown in Figure 12 as A,B,C, and D. These points represent the mass flow rates due to combined flow sensor 101C output signals and pressure SI and Pi, S2 and PI, SI and P2 and S2 and P2 respectively. The first interpolation computes point E from the points A and B. E is computed by linear interpolation between A and B according to the position of S in the interval between SI and S2. Thus:

E=((S-S1)/(S2-S1))*B+((S2-S)/(S2-S1))*A

Similarily, F is computed from the points C and D. Thus: F=( (S-S1 )/( S2-S1 ) ) *D+( ( S2-S )/( S2-S1 ) ) *C

Finally, X, the true mass flow rate is computed by linear interpolation between the calculated values of E and F according to the position of P in the interval between PI and P2, Thus:

X=((P-P1)/(P2-P1))*F+((P2-P)/(P2-P1))*E

The accuracy of this calculation of X depends on the linearity of the mass flow as a function of the output signal of flow sensor 101C at constant temperature and pressure over the intervals A to B and C to D, and as a function of pressure at constant temperature and flow sensor output over the interval from E to F. Greater accuracy, if desired, can be achieved in. a well known manner by reducing the size of the intervals and/or using a nonlinear interpolation function.

Once the microprocessor has computed a mass flow rate (such as the one represented by X in Figure 12) which does not include the effects of temperature, it is necessary to adjust the result to compensate for this effect. Since the absolute temperature varies much less than the flow rate or the pressure in the operating range of an in¬ strument of the type disclosed herein, it is possible to use a simple function of the difference between the measured temperature and the calibration temperature to perform this adj stment. The calibration temperature is the temperature at which the matrix of data relating to flow sensor 101C output signals and pressures to mass flow rates (such as used to calculate X) is measured. A linear function of the actual temperature minus the calibration temperature has been found to provide adequate temperature compensation for this data over the normal operating range of 0 degrees Celsius (273°K) to 40 degrees Celsius (313°K). Other embodiments of this invention will be obvious to those skilled in the flow measuring and flow controlling arts and in the electronic circuit design arts in view of the above disclosure and this disclosure is meant to be illustrative only, and not to limit the scope of our claims.

Claims

We claim:

1. Structure for measuring the flow of fluid comprising:

a flow inlet passage through which fluid whose flow is being measured is passed;

means, slidably mounted in said flow inlet passage, for moving in response to the flow of fluid, the amount of motion being related to the flowrate of said fluid, said means for moving being mounted in a nominal position so as to block said flow passage in the absence of the flow of fluid, said means for moving being capable of measuring fluid flow in both the laminar and the impact and discharge flow regimes;

means for generating a restoring force on said means for moving to restore said means for moving to its nominal position in the absence of flow of said . fluid;

means, responsive to the position of said means for moving, for producing a first signal repre¬ sentative of the position of said means for moving, said first signal being representative of the flow¬ rate of said fluid; and

means for processing said first signal to produce an indication of the flowrate of said fluid.

2. Structure as in claim 1 wherein said flow passage comprises a tube of circular cross-section and wherein said means for moving is slidably mounted and partially inserted into said tube.

3. Structure as in claim 2 wherein said means for moving comprises a tapered pin and wherein said tapered pin is

OMPI mounted in said tube such that upon substantially complete insertion into said tube the cross-sectional flow area between said pin and said tube is reduced to zero and wherein said tapered pin is forced out of said tube by the flow of.fluid a distance representative of the flowrate of said fluid thereby to open between said pin and said inlet tube an annularly shaped cross-sectional flow area so as to allow the passage of fluid.

4. Structure as in claim 3 wherein said structure include

means for sensing the position of said tapered pin and for producing an output signal linearly related to said position.

5. Structure as in claim 4 wherein said means for sensing the position of said tapered pin comprises:

a magnet mounted on a base plate against which said tapered pin presses in response to the flow of fluid thereby to displace said base plate from its nominal position;

means for restoring said magnet and said tapered pin to their nominal rest positions so as to block the flow of fluid through said tube; and

means for sensing the position of said magnet and for producing said first signal.

6. Structure as in claims 1, 2, 3, 4 or 5 wherein said base plate includes a curved, cup-like concave surface facing said tapered pin such that said tapered pin rests in the bottom of said cup-like surface so that the fluid flowing past said tapered pin is reversed in flow direction by said curved, cup-like concave surface. 7. Structure as in claim 6 wherein said means for re¬ storing said magnet and said tapered pin to their nominal rest positions comprises a spring.

8. Structure as in claim 7 wherein said means for re¬ storing said magnet and said tapered pin to their nominal rest positions comprises a cantilevered beam, said magnet being mounted on the free end of said cantilevered beam and said tapered pin striking said cantilever beam at a distance removed from its pivot point, the normal spring forces of said cantilever beam tending to restore said beam to its rest position.

9. Structure as in claim 8 wherein said magnet travels a curved path in response to deflection of said cantilevered beam in response to motion of said tapered pin resulting from the flow of fluid, wherein said motion along said curved path reduces the degree of nonlinearity in the output signal from said means for sensing.

10. Structure as in claims .1, 2, 3 or 4 including means for sensing the pressure and temperature of said fluid thereby to provide correction signals for use in con¬ verting said fluid flowrate to a mass flowrate.

11. Structure as in claim 1 wherein said means for proces¬ sing comprises means for converting the output signal from said means for producing to a digital signal and means for using said digital signal to obtain a measure of the flowrate of said fluid.

l-d . Structure as in claim 1 wherein said means for proces¬ sing comprises an electrical circuit including means for > amplifying and linearizing said first signal from said means for producing to produce an analog signal linearly representing of the flowrate sensed by said means for sensing. 13. Structure as in claim 12 wherein said means for processing said first signal from said means for producing comprises

means for producing a first intermediate signal;

means for squaring said first intermediate signal to produce a second intermediate signal; and

means for adding said second intermediate signal to said first intermediate signal to produce a composi signal that varies substantially linearly with respect to flowrate changes detected by said means for moving.

14. Structure as in claim 13 including means for amplifyin said composite signal from said means for processing.

15. Structure as in claim 11 wherein said means for processing comprises

a stored table comprising a selected number of values of flowrate versus the first signal from said means for producing stored at a number of locations corresponding to the same number of values of said first signal; and

means for interpolating the actual flowrate from the values of flowrate stored in said table, thereby to produce an output signal representing the flowrate sensed by said flow sensor.

16. Structure as in claim 15 wherein said means for interpolating comprises

means for converting said first signal from said means for producing to a digital signal wherein the value of said digital signal is generally between two values of flow rates stored in a memory;

means for using said digital signal to select the addresses of the locations in said memory con¬ taining the flowrates most closely corresponding to said value of said first signal and for interpolating from these flowrates, the actual flowrate of said fluid.

17. Structure as in claim 16 including means for dis¬ playing the actual flowrate.

18. Structure as in claim 6 wherein said concave, cup¬ like surface is shaped so as to substantially reverse the flow of fluid impacting on said surface.

19. Structure as in claim 18 wherein said concave, cup¬ like surface ensures the smooth flow of fluid over said surface thereby to enhance the quality of the measured flow rate.

20. Structure as in claim 11 wherein said means for processing comprises: a stored table comprising a selected number of values of flow rate versus said first signal from said means for producing stored at a number of locations corresponding to the same number of values of said first signal; and

means for selecting the value of flow rate from the values of flow rate in said stored table repre¬ sented by said first signal thereby to produce from said table an output signal representing the flowrate sensed by said flow sensor. 21. Structure as in claim 20 wherein said means for selecting comprises:

means for converting said first signal from said means or producing to a digital signal wherein the value of said digital signal corresponds to the address of a location in a memory storing the values of flow rates sensed by said sensor; and

means for using said digital signal to select the address of the location in said memory containing the flow rate corresponding to said value of said first signal and for reading out a digital signal representing said flowrate of said fluid.

22. Structure for measuring the flow of fluid comprising:

means for generating a restoring force on said means for moving to restore said means for moving to its nominal position in the absence of flow of said fluid; and

means, responsive to the position of said means for moving, for producing 'a first signal representative of the position of said means for moving, said first signal being representative of the flowrate of said fluid.

;23. structure as in claim 22 wherein said flow passage comprises a tube of circular cross-section and wherein said means for moving is slidably mounted and partially inserted into said tube.

24. Structure as in claim 23wherein said means for moving comprises a tapered pin and wherein said tapered pin is mounted in said tube such that upon substantially complete insertion into said tube the cross-sectional flow area between said pin and said tube is reduced to zero and wherei said tapered pin is forced out of said tube by the flow of fluid a distance representative of the flowrate of said fluid thereby to open between said pin and said inlet tube an annularly shaped cross-sectional flow area so as to allow the passage of fluid.

25. Structure as in clai 24 wherein said structure includes

means for se___sing the position of said tapered pin and for producing an output signal related to said position.

26V Structure as in claim24 wherein said means for sensing the position of said tapered pin comprises:

means for restoring said magnet and said tapered pin to their nomi_22l rest positions so as to block the flow of fluid through said tube; and

27. Structure as in claim 2?, 23, 24,.25 or 26 wherein said base plate includes a curved, cup-like concave surface facing said tapered pin such ^'that said tapered pin rests in the bottom of said cup-like surface so that the fluid flowing past said tapered pin is reversed in flow direction by said curved, cup-like concave surface.

28. Structure as in claim 27 wherein said means for restori said magnet and said tapered pin to their nominal rest positions comprises a spring.

OM 29. Structure as in claim 28 wherein said means for restori said magnet and said tapered pin to their nominal rest positions comprises a cantilevered beam, said magnet being mounted on the free enc of said cantilevered beam and said tapered pin striking said cantilever beam at a distance removed from its pivot point, the normal spring forces of _=.aid cantilever beam tending to restore said beam to its rest position.

³⁰• Structure as in claim 29 wherein said magnet travels a curved path in response to deflection of said cantilevered beam in response to motion of said tapered pin resulting from the flow of fluid, wherein said motion along said curved path reduces the degree of nonlinearity in the output signal from said means for sensing.

Structure as in claim 27wherein said concave, cup-like surface is shaped s as to substantially reverse the flow of fluid impacting on said surface.

31. Structure as in claim 27 wherein said concave, cup-lik surface ensures the smooth low of fluid over said surface thereby to reduce the possibility of vibrations.

^^Q £Λ

OMPI ^™°