Note : Les descriptions sont présentées dans la langue officielle dans laquelle elles ont été soumises.
s
BACKGROUND OF THE INVENTION
The invention relates broadly to the field of flowmeters
and particularly to acoustical flowmeters for measuring the rate ~ ;~
of flow of a fluid along a confined path.
A typical acoustical wave flowmeter is disclosed in U.S.
Patent No. 3,109,112 issued October 29, 1963 to Robert A. Lester.
This flowmeter has a pair of transducers for generating and
receiving compressional waves in ei-ther the audible or ultrasonic -~
frequency range. These transducers are located within an enclosure ,
through which the fluid flows. The transducers are alternately
configured in transmit and receive mode so that compressional
waves are produced in the gas by the transmitting transducer and
received at the other transducer. By measuring the phase
difference between the transmitted and the received waves in ;
both directions, the velocity of the gas passing through the
transducer is determined.
A modification to the approach described in the above men-
tioned U.S. Patent relies on transducers designed with a transmit
and a receive section configured to substantially eliminate any
problems due to gas composition or temperature changes in the gas.
The above known confiyuration requires that the transducer
be located in the flow path of the gas whose velocity is to be ~ ,~
measured or located in a cavity in the conduit wall. In either
arrangement, the normal flow of the fluid is substantially altered
when passing the transducer thereby causing the accuracy of the
meter to suffer. Additionally, sedimentery particles in the
fluid or the like can collect around the transducers thereby
impairing the transducer's transmitting and receiving character-
istics.
A flowmeter has been developed in an effort to overcome
--1--
.. . .
these and other difficulties of known acoustical flo~meters.
The improved flowmeter is designed to have no obstructions along
the flow path and is described in U.S. Patent No. 4,003,252 issued
January 18, 1977 to Edward James DeWath and entitled "Acoustical
Wave Flowmeter". In that flowmeter, the transducers are generally
cylindrical in shape and are disposed within the walls of the
fluid conduit thereby eliminating all obstructions in the flow
path as well as eliminating cavities in the conduit wall in which
debris might collect.
While the invention of Edward James DeWath does eliminate
obstructions in the path of fluid flow as well as cavities in the
conduit wall, his flowmeter may become inaccurate or fall to
function entirely whenever the composition of the gas flowing
therethrough is different Erom that for which the meter is cali-
brated. Indeed, this is a common problem with most acoustical
flowmeters as the velocity of acoustical compressions within a
gas is a function of its chemical composition so that typical
acoustical flowmeter accuracy depends on calibrating it for the
specific gas whose flow is to be measured. Requiring recalibration
each time the flowmeter is used with a different gas is, at best,
an inconvenience. In applications where the gas composition changes
as the flowmeter is used as, for example, in pulmonary function
analyzers, known acoustic wave flowmeters are inaccurate unless -
correctlve feedback is provided from a gas analyzer to compensate
for gas composition change.
~ ccordingly, it is an object of the present invention to
provide a flowmeter which is accurate regardless of changes in
gas composition or temperature during use.
It is yet another objectof the invention to provide a
flowmeter including means to automatically adjust the flowmeter
-2-
operation as a function of dynamic gas compositional and temp-
erature changes and to provide a measure of the velocity of sound,
which is itself related to the average molecular weight of the gas.
SUMMARY OF THE INVENTION
The invention comprises an acoustical wave flowmeter
having, in its preferred form, two transducers mounted in the wall -
of a conduit so that there is no obstruction to gas flow and no
cavities in the wall to provide a locus for collecting particulate
matter. The transducers are connected to an electronic circuit
10 which alternately switches one of them into transmit mode and the ;
other into receive mode. Automatic circuit adjustment means is ~-~
provided to adjust the transmitting frequency so that the energy
in the acoustic compressions generated by a transmitting transducer
is maximized at a receivingtransducer thereby compensating for the
change in velocity of acoustic compressions in the gas caused by
changes in gas composition or temperature. ;~
The electronic circuit includes circuitry to measure and
store the phase difference between the signal producing acoustic ~
compressions at a transmitting transducer and the signal produced ;
20 by the receiving transducer in response to the acoustic compres- `
sions during each of two successive transmit-receive cycles.
Circuit means is provided to determine the difference between two
successive phase differences, at least one of which was pre-
viously stored, wherein the sign of the difference corresponds to
the direction of gas flow and the magnitude of the difference ;
corresponds to the rate of gas flow through the flowmeter. Cir- ;~
cuitry is also provided to form the sum of two successive phase
differences which is proportional to the velocity of sound in
the gas moving through the flowmeter.
~ 3- `
, ~
BRIEF DESCRIPTION OF THE DRAWINGS
The invention is described below in greater detail in
conjunction with the drawings wherein:
Fig. 1 is a schematic diagram of the flowmeter including
the transducer assembly and a block diagram of the electronic
circuitry associated therewith;
Fig. 2 is a detailed circuit diagram of a tlming pulse
generator for controlling the operation of the circuit of Figs.
4 and 5;
Fig. 3 is a timing pulse chart for the circuit of Fig. ~ :
2; and
Figs. ~ - 6 comprise a detailed circuit diagram for one
embodiment of the flowmeter circuitry according to the invention
which utilizes the timiny pulses generated by the circuitry
of Fig. 2.
DETAILED DESCRIPTION
Referring first to Fig. 1, the flowmeter of the present
invention includes a transducer assembly 10, shown in longitudinal
section, which comprises a substantially cylindrical body having
a central opening extending through the assembly 10 through which
gas flows in a direction as indicated by the arrows 12. These '
arrows 12 indicate gas flow through the transducer assembly 10 in
a direction from left to right as viewed in Fig. 1. Gas can flow
through the assembly 10, however, in the opposite direction if so
desired.
The transducer assembly 10 is generally made in accordance `~
with the description found in U.S. Patent No. 4,003,252 by Edward
James DeWath and entiteld "Acoustical Wave Flowmeter". Accordingly, ~ .
the transducer assembly 10 of the present invention may additionally
include a cylindrical casing made of metal or other suitable
material (not shown in Fig. 1) which surrounds the support member
6~3S
14. The support member 14 itself is preferably fabricated from a
polyurethane foam, foam rubber or other material having good
acoustical damping properties. Two annular no-tches 16 and 18
are formed into the inner wall of the support member 14 and are
disposed in spaced relation to each other along the path of fluid
flow as indicated by the arrows 12. Two addi-tional annular notches
20 and 22 are disposed at opposite ends of the support member 14. ;~
Disposed respectively within and in contact with the
surfaces defining each annular notch 16 and 18 is a cylindrical
transducer 24 and 26 each having an inner cylindrical bore passing
therethrough which is substantially coextensive with the inner
cylindrical bores 28 which extend between the annular notches of
the support member 14.
Transducers with other geometry such as arcuate transducers
may also be used with appropriate changes to the remaining parts
of the transducer assembly 10 so as to maintain a centrally located
bore therethrough with no obstructions and no cavities in the bore
wall to provide a locus for particulate matter to collect. ~;
Disposed respectively within each annular notch 20 and
22 is a terminal ring 30 and 32 which is preferably made of
sound ahsorbing material such as a polyurethane damping foam like
Y-370 Vibration Damping Tape manufactured by the 3M Corp.
Materials having similar sound absorbing properties may also be
used for the terminal rings 30 and 32. The terminal rings 30 and
32 each have an inner cylindrical bore which is substantially
coe~tensive with the adjacent cylindrical bores 2~ of the support
member 14. Accordingly, the fluid flow path through the trans-
ducer assembly lO, as indicated by the arrows 12, has a substan-
tially continuous wall so that the fluid flowing therethrough
encounters neither protuberances nor cavities.
The transducers 24 and 26 may comprise one of a number of
conventional devices capable of operating in radial or hoop mode
- . ~
for generating acoustical compressions with in the gas passing
through the assembly 10. Examples of suitable materials for the
transducers 24 and 26 include cylindrical bodies made for poly-
vinylfluoridene or other high polymer organic piezoelectric :~
materials, ceramic transducers fabricated from barium titanate,
lead zirconate titanate, or other polarized polycrystalline
ferroelectric ceramic materials,~uartz, tourmaline, or equivalent
electromechanical devices known to those skilled in the art.
The inner and outer surfaces of the transducer 2~ and
1~ 26 have conductive coatings thereon which oonstitute the electricaldrive electrodes. The conductive coatings on the inner surfaces
of the transducers 2~ and 26 are respectively connected via wires
34 and 36 to externally located crystal driver/receiver c.ircuits
38 and 40. The conductive coating on the outer surface of the
transducers 24 and 26 are respectively connected via wires 42
and 44 to the crystal driver/receiver circuits 38 and 40. The
electrical connection wire pairs 34, 43 and 36, 44 for the trans-
ducers 24 and 26 respectively pass through openings 46 and 48
which respectively communicate from outside the assembly 10 to `
the transducers within annùlar notches 16 and 18.
The operation of the transducer assembly 10 in cooperation
with the electronic circuitry is fundamentally the same as des-
cribed in the above-referenced Patent of Edward James DeWath.
Basically, however, the transducer assembly is connected by flex- ~`
ible hoses, tubing, or the like to an external source of fluid
whose flow rate is to be determined. The fluid itself may com- ~
prise either a gas or a li.quid although the illustrated embodi- .
ment of the invention is designed particularly for measuring gas
flow as it flows through the transducer assembly 10 in an axial
direction indicated by the arrows 12 or in the opposite direction.
A control circuit 50 is coupled respectively by lines 52 and 54
--6--
t;~`
,~; .
to the transducer driver/receiver circuits 38 and 40. The
control circuit 50 is operative to cause one of the transducer
driver/receiver circuits 38 or 40 to transmit an electrical
signal to the respectively coupled cylindrical transducer 24
or 26 while the other driver/receivercircuit 40 or 38 is con-
ditioned to receive a signal from the coupled transducer 26 or
24. As indicated earlier, the transducer 24 or 26 which receives
electrical signals from one driver/receiver circuit responds
thereto by producing an acoustic comprssion in -the fluid. At
the same time, the other transducer coupled to -the other driver/
receiver circuit 38 or 40 responds to the acoustic compressions
in the fluid by producing an electrical received signal which is
sensed by the coupled driver/receiver circuit. ; .
The acoustic comprssions produced at one of the transducers
24 or 26 take a finite time to propagate from the transmitting
transducer to the receiving transducer with the propagation time
being dependent on the direction and the velocity of fluid flow
through the transducer 10 and the velocity of sound in the ~luid
flowing within the transducer 10. These relationships are
described in greater detail in the above mentioned Patent.
As an alternative to the above described transducers, each
transducer 24 and 26 can be replaced by a pair of transducers
mounted in the assembly 10 in a similar manner. Of the pair of
transducers, one is specifically for transmitting, i.e., producing
acoustic compressions, and the other is a receiving transducer,
i.e., to produce a received signal in response to acoustic compres-
sions in the gas.
The transducer assembly 10 and the coupled circuitry,
operate in the following manner. The control circuit 50 first
causes one transducer, either transducer 24 or 26, to produce
acoustic compressions in the fluid and at the same time the other
~ 5~
transducer, either 26 or 24, is conditioned to receive the acoustic compression
from the first transducer. m is mode of operation comprises a first transmit-
receive cycle. A transmit signal is simultaneously sent from the control
circuit 50 over either line 52 or 54 to one driver/receiver circuit 38 or 40
and over the line 56 to a phase detector 58. A received signal from the
receiving transducer 24 or 26 is coupled by either transducer driver/receiver
circuit 38 or 40 over lines 60 or 62 respectively to the phase detector 58.
m e phase detector 58 responds to the transmit signal and to the received
signal by producing a signal on line 64 w~ich indicates the phase difference
between the transmitted and the received signal. This information is coupled ~ ;
over line 64 to a phase adder and subtractor 66 which includes storage means
for storing te~porarily the phase difference received from the phase detector 58.
The control circuit 50 then causes the transducers 24 and 26 to change
roles. That is, -the other transducer 24 or 26 is placed in transmlt mode to
produce acoustic compressions in the fluid and the first transducer 24 or 26 is
placed in receive mode. m is mode of operation comprises a second transmit-
receive cycle. The phase detector 58 is operative during the second transmit~
receive cycle to determine the phase difference between the transmitted signal
and the received signal and this second phase difference is transmitted over
the line 64 to the phase added and subtractor 66 which also tem~orarily stores
this second phase difference. ~ ~ -
The phase adder and subtractor 66, once the two phase differences have
been stored therein, calculates both a sum and a difference between the two ;
stored phase differences. The difference between the two stored phase
differences with one further correction described in connection with Fig. 6 is
presented on the output line 68 which, when the flowmeter of the invention is
calibrated, is indicative of the flow rate through the transducer assembly 10.
In addition, the algebraic sign of the difference between the two phase
differences calculated by the phase detector 58 is indicative of the direction
of flow through the transducer assembly 10 with a positive algebraic sign
representing one arbitrary direction of fluid flow through the assembly 10 and a
~ ~
negative algebraic sign representing the opposite direction of
fluid flow.
The sum of the two phase differences as calculated by ~ .
the phase adder and subtractor 66 is presented on the output line ~ :
70. The magnitude of the signal on the line 70 is proportional
to the velocity of sound in the gas passing through the assembly
10 compared to the velocity of sound in the reference gas used
for calibration. The sum is also transmitted over the line 72
to the control circuit 50 which is utilized thereby in a manner ~ ~
described in greater detail below. In addition, it is used to ;~ ; -;~.
correct the flow value on line 68. The value presented on the ~
OlltpUt line 70 is a relative one and the flowmeter, according to :-
the invention, must be calibrated so that the magnitude of the
va~we on line 70 can he interpreted.
As indicated generally above, a failing of the flowmeter
described in the above mentioned Patent by DeWath is that the ~
meter there described is not operational in environments where -
the velocity of sound i.n the gas passing through the transducer
changes dynamically during its use. We have found that the
cylindrical transducer system functions by building up a strong
resonant echo across the diameter of the tube at one or more
Eigen (characteristic) frequencies. For example, the relationship
may be approximately 2.3~ = D, where ~ is the wavelength of sound
in the fluid (cm) and D is the inside diameter of the crystal
and tube (cm). The frequency at which resonant echo occurs
across the diameter D is called the natural cavity resonant
frequency of the 02 mode. Other useful frequencies at which
resonant echo occurs are those at which D is substantially equal
to 0.73~, 1.4~, 3.2~ or 3.9~. The reason for the above described
f].owmeter failure then is due to the fact that the velocity of the
acoustic compressions within the gas passing through the flow
detecting transducers varies as a function of the gas composition.
This causes the amplitude of the acoustic compressions, detected
by the transducer in the assembly which is in receive mode to vary
because the inner diameter of the transducers is no longer equal
to the Eigen wave length thereby causing the magnitude of the
signal produced by the receiving transducer to fall off dramatically.
Accordingly, the received signal becomes more difficult to detect
thereby introducing the possibility of detection error or complete
failure to detect the acoustic compression. This failure can be
overcome by changing the inner diameter dimension of the transducers
within the assembly or by changing the frequency of the acoustic
comprssions produced by the transmitting transducer. As the
transducer assembly itself is not suitable for inner diameter
adjustment of the transducers, clearly, the better approach to
solving this problem is to adjust the frequency o~ the acoustic
compressions produced by the transmitting transducer. Accordingly,
the sum of the phase differences as calculated by the phase adder
and subtractor 66 is transmitted over the line 72 to the control
circuit 50. As indicated above, the signal on the line 72 is
related to the velocity of sound in the gas passing through the
assembly 10. This signal is utilized by the control circuit
50 to change the frequency of the acoustic compressions produced
by the transmitting transducer during each transmit-receive cycle
so as to maintain the wavelength constant and thereby to maximize
the signal received by the receiving transducer.
We have also found, after correcting the frequency in order
to hold t:he wavelength constant, that we must multiply the
apparent flow rate by -the velocity of sound in order to produce
an accurate flow rate measurement independent of change of gas
composition.
Experience has shown that flowmeters of the type cLescribed
in the above identified Patent are sometimes subject to error due
--10 -~
to reflection of acoustic compressions from other parts of the
system into which the transducer is connectecl. These reflections
frequently result from the acoustic compressions bouncing off
other fluid transmitting fittings which are coupled to the trans-
ducer assembly. Problems have also resulted when the transducer
itself is used in breath analyzers where it is physically located
in close proximity to the mouth of the individual whose breath is ~
being analyzed. -;
The present invention overcomes the reflection problem
by providing the terminal rings 30 and 32 which are disposed at
opposite ends of the assembly 10. These rings 30 and 32 are
made of an acoustic damping material such as that mentioned earlier
which is operative to substantially reduce the amplitude of
acoustic compressions passing therethrough. Indeed, the magnitude
of the damping provided thereby, if the proper material is
selected for the rings 30 and 32, is sufficient so that one end
of the assembly 10 may be inserted in close proximity to an
individuals mouth whose breath is being analysed and the problems
experienced with other sonic flowmeters due to acoustic reflections
is substantially eliminated. The damping provided by the rings
30 and 32 also substantially eliminates problems with reflections
when the assembly 10 is coupled into the fluid carrying systems
as well.
The foregoing discussion in connection with Fig. 1 has
generally described the circuitry and operation of the invention.
The circuitry of Figs. 2, 4-6 comprise one actual irnplementation
of the invention for use in pulmonary testing equipment, however,
those of skill in the art will recognize that alternate circuits
may be used, and the components may have to be changed to optimize
the circuit for use in other applications for the invention.
The circuitry of Fig. 2 comprises a pulse generator
for producing control pulses to operate the circuits shown in
Figs. 4-6. Fig. 3, on the other hand, is a pulse diagram showing ;
the pulse train for different outputs from the circuit of Fig. 2.
Briefly, the circuit of Fig~ 2 comprises an oscillator circuit,
integrated circuit type CD4047, to which is coupled a 121K
resistor and a 4700 PF capacitor for controlling the output
frequency which, for the resistor and capacitor mentioned, is
400Hz. ~wo series coupled J-K flip-flops, each comprising one
half of a CD4027 integrated circuit, are utilized to produce
timing pulses at a rate slower than that produced by the oscillator
integrated circuit CD4047. The signals produced by the oscillator
and the flip-flops are combined by the AND and I~ND gates of
Fig. 2 so as to produce the respective pulse strings as shown in
Fig. 3. A pulse diagram for the output QO is not shown in Fig. 3,
however, it comprises a square wave having a frequency twice that
for Ql and one leading positive going edge of a pulse on the line
labelled QO occurs at the same time as the leading positive going
edge of a pulse on the output of CD4047labelled Ql
Referring now to ~ig. 5, integrated circuit CD4046 has
a voltage controlled oscillator (VCO) which produces a square
wave signal at the output pin 4 whose frequency is controlled by
the resistors connected in series between pin 11 and the ground as
well as by the voltage applied at pin 9. For the particular
resistors shown in Fig. 5 with a voltage at pin 9 of about -~7.5
volts, the frequency of the square wave output from the voltage
controlled oscillator at pin 4 is nominally 51KHz. Changing the
voltage appearing at input pin 9 is operative to shift the
frequency of the voltage controlled oscillator. The circuit of
Fig. 5 will cause the frequency of the VCO to change dynamically
to respond to changes in the velocity of sound in the gas in a
manner described later in greater detail.
-12-
~ 5~
The square wave signal from the voltage controlled oscil-
lator at pin 4 connects via a line 100 to pin 3 of circuit CD4046
which internally connects to a phase detector (0DET). The square .:
wave signal also connects via a line 102 to a NAND gate whose
output connects to an operational amplifier LM318 and operational
multiplier XR2208. The function of these circuits is to produce
a sine wave signal at point 104 having the same frequency as the
square wave signal appearing at line 102. Those of skill in the ~
art will recognize, however, that the circuitry between line 102 ;
and point 104 comprises only one circuit of many known circuits ~`
for converting a square wave into a sine wave signal and that
other equivalent circuits may be used.
Disposed between point 104 and the output terminal AC
are three operational amplifiers comprising circuit types LM31
and 8043C. The LM318 operational amplifier is for amplifying of
the sine wave at point 104. The two 8043C operational amplifiers
are for adjusting the phase of the signal appearing at the output
terminal AC with adjustment resistors 106 and 108 being operative
to adjust the phase of the signal at the output terminal AC by
about 360. These resistors are adjusted during calibration of
the flowmeter and are preferably adjusted so that the signal at ;.
pin 9 of circuit CD4046 is +7.5 volts with no air flow through
the transducer assembly 10 (Fig. 1). The description that follows
will describe the mechanism by which phase adjustment of the signal
at the output terminal AC causes the voltage at pin 9 to change.
The circuitry of Fig. 4 comprises the transducer driver/
receiver circuits and shows how they are connected to the trans-
ducers 24 and 26. The circuitry of Fig. 4 is divided into a
driver section appearing to the left of the dotted line 110 and a :~
receiver section, exclusing the transducers 24 and 26, appearing
to the right of the dotted line 110. ;'
-13-
.`'~
The sine wave signal from the circuitry of Fig. 5 ls
connected to the input terminal AC in Fig. 4 and thereafter is
coupled to two transducer drivercircuits, the first driver
circuit including Q5 and the second driver clrcuit including Q6.
These transistors Q5 and Q6 gate the sine wave signal from the
input terminal AC to the respectively coupled transducer 24 or
26 thereby placing the transducer into transmit mode. The
gating signals are developed by the transistor pairs Ql, Q3 and
Q2, Q4 and the respectively coupled circuitry which includes that
10circuitry of Fig. 2 for developing the gate signals which appear
at terminals X and Y. As the gate signals at terminals X and Y
occur at different times and alternate as shown in Fig. 3, the
transducers 24 and 26 are alternately placed in transmit mode.
The receiver circuits 116 and 118 are also directly
coupled respectively to the transducers 24 and 26, however, the
receiver circuits 116 and 118 are either operative or inoperative
to respond to signals developed by the coupled transducers 24 or
26 depending on whether the respectively coupled input shorting "
transistor Q7 or Q8 is either off or on. Transistor Q7, for
20example, is controlled by a gating pulse signal Q3 and shorts
line 112 to ground whenever the signal Q3 is positive. In a
similar manner, the transistor Q8 shorts line 114 to ground
whenever the gatesignal Q3 is positive. From the pulse chart of
Fig. 3, it can be seen that Q3 is positive whenever X is positive
and hence receiver 116 is inoperative whenever the transducer 24
is in transmit mode and receiving signals from the input terminal
AC. Similarly, recei,ver 118 is inoperative due to Q3 being
posi,tive whenever the transducer 26 is in transmit mode. As such, '~
whenever a given transducer 24 or 26 is in transmit mode, the
respectively coupled receiver circuit 116 or 118 is inoperative.
However, whenever one transducer is in transmit mode, the receiver
circuit 116 or 118 coupled to the other transducer 24 or 26 is
-14-
enabled because the corresponding gating signal Q3 or Q3 is at
ground potential. For example, during a first transmit-receive
cycle, the transducer 24 is in transmit mode and receives signals
from the input terminal AC, the transistor Q7 is turned on by
the signal Q3 to turn receiver 116 off by grounding line 112 and
the receiver circuit 118 is operatively coupled to transducer 26
because transistor Q8 is turned off due to gate signal Q3 being ~.
at ground potential. At the same time, the gate signal Y is at - ;
ground potential thereby preventing the signal appearing at -the
input terminal AC from bei.ng coupled via transistor Q6 to the
transducer 26. The second transmit-receive cycle occurs when
transducer 26 reaeives signals from the input terminal ~C,
transistor Q7 is turned off b~ Q3, transistor Q8 is turned on by ~ 7
Q3, gate signal Y is positive and the gate signal X ls at ground
potential. As such, transducer 26 is in transmit mode, receiver
116 is operative and receiver 11~ is inoperative.
The receiver circuit 116 coupled to the transducer 24
includes two operational amplifiers 120 and 122 connected in
series and operative whenever transis-tor Q7 is not turned on to
amplify any signal appearing at the input line 112. The amplified
signal appears at pin 6 of operational amplifier 122 which is
connected to the inverting input of a comparator circuit 124~ The
comparator circuit 124 is enabled whenever the gate signal A is
at logic zero which allows the output signal to appear at pin 7.
The ou-tput at pin 7 is at a positive voltage (approximately
~15 volts) when the sine wave input to pin 3 of the comparator
124 is of negative potential and at zero volts whenever the input
at pin 3 is of positive poten-tial. Comparator 124 thus converts
the sine wave input into a square wave output. Comparator
circuit 126 which couples thereto is inoperative during the period
when comparator circuit 124 is operative and vice versa.
:~ -15-
s6~5
The receiver 118 includes two series connected operational
amplifiers 130 and 132 which amplify the signal appearing on
the input line 114 and present that amplified signal to the output
pin 6 of the operational amplifier 132. This output signal at
pin 6 of amplifier 132 is applied to the inverting input of a
comparator circui~ 126 which is enabled whenever the gate signal
B is at logic zero. The output at pin 7 of the comparator
circuit 126 is a square wave having a voltage of approximately
plus 15 volts whenever the input at pin 3 is of negative potential
and at approximately zero volts whenever the input at pin 3 is of
positive potential. Since gate signal A and B occur at different
times, the operation of the comparator circuit 124 or 126 does
not affect the operation of the other comparator circuit 126 or 124.
~ ccordingly, the comparator circuits 124 and 126 are
independently operative to produce square wave signals at their
respective outputs from the sinusoidal singals which are developed '
across the coupled transducer 24 or 26 whenever that transducer is
in receive mode and the required gating signals are present to
actuate the receiver circuit 116 or 118. ~s the output pins of
the comparator circuits are coupled together at terminal SIG,
the signal appearing at terminal SIG represents the logcial AND
of the signal developed at the output of comparators 124 and 126.
The output from the receiver circuits 116 and 118 is
transmitted via the terminal labeled SIG in Fig. 4 to the corres-
ponding input terminal labelled SIG in Fig. 5 and then to pin 14 of
the integrated circuit CD4046 which is internally connected to the
phase detector circuit contained therein. The phase detector
itself functionally operates the same as an EXCLUSIVE OR circuit
whose output is internally connected to pin 2 of integrated
circuit CD4046 and has a logic one output level whenever only one
input to the phase detector is also at a logic one level.
-16-
`
6~
Accordingly, the signal appearing at pin 2 of integrated circuit
CD4046 comprises a square wave signal which is pulse width
modulated where the width of each pulse is related to the phase -~
difference between the signal transmitted, i.e. the signal on
line 100 and the received signal, i.e., the signal appearing
at pin 14 of integrated circuit CD4046.
In normal operation of the flowmeter according to the
invention, if the composition of the gas passing through the
transducer assem~ly 10 (Fig. 1) changes from that when the meter
was calibrated, the phase of the signal appearing at the input
terminal SIG in Fig. 5 is different relative to the signal phase
appearing at that input terminal when circuitry was calibrated.
Accordinyly, the phase detector output will be different. thereby
causing the voltage across the one microfarad capacitor o.E the
low pass filter comprising a 1 meg ohm resistor and a 1 microfarad
capacitor coupled to the output of operational amplifier 156 to
change relative to that when the meter was calibrated in a
manner which is described later in greater detail. This change
in voltage across the 1 microfarad capacitor causes the voltage
at pin 9 of integrated circuit CD4046 to change thereby causing
the frequency of the signal produced by the voltage controlled
oscillator to change as well. The system will continue to adjust
the voltage controlled oscillator frequency until the phase
difference between the transmitted and the received signal no
longer causes a change in voltage across the one microfarad
capacitor in the low-pass filter.
The ability to change operating frequency in response to
gas composition change is critical to the ability of the present
invention to operate where prior art systems cannot. It can be
demonstrated that the transducers of the invention are able to
produce maximum energy transfer from the transmitting to the
~17~
S~
-
receiving transducer when operated at an Eigen (characteristic)
frequency which depends on the inner diameter of the transducer.
Accordingly, when air is the gas in the transducer assembl~ when
the meter is calibrated, the frequency of the voltage controlled
oscillator corresponds to that which produces acoustic compressions
having a fractional number of half wave lengths across the
transducer diameter in air which corresponds to an Eigen frequency.
If the gas density thereafter changes, the velocity of acoustic
compressions therein also changes and hence a different phase
difference is de-tected by the phase detector. This causes the
frequency of acoustic compressions produced by the voltage
eontrolled oscillator to change in a manner described in greater
detail later and it can be demonstrated that the new fre~uency -~
eorresponds to one where the diameter of eaeh transducer is again
the Eigen value number of half wavelengths at the new frequeney -
in the gas then in the transducer assembly. As such, maximum -
energy transfer between a transmitting and a receiving transducer
is maintained. ~ `;
The pulsewidth modulated signal appearing at the output
of the phase detector comprises one of two inputs to an AND
gate 140 whose output is coupled to an integrator eircuit indicated
generally at 142. The second input to the AND gate 140 is an
integrate enable signal IE which is a logic one whenever either
gate signal A or the gate signal B is logic zero, a condition
indicating that the phase detector output corresponds to the phase
difference between a transmitted and a received signal. As such,
the pulse width modulated signal is applied via the AND gate 140
when it is enabled by the integrate enable signal IE. The output
of the integrator circuit 142 appears at pin 6 of the operational
amplifier LM318 and comprises an integrated level during the
-18-
:-
~::
integrate enable period whose final level is related to the ;~
phase difference between the transmitted signal at one transducer
and the signal received at the other transducer and is coupled
via a line 144 to input pin 5 of two different sample and hold
circuits 146 and 148. Each sample and hold circuit 146 or 148
sampled the voltage appearing at pin 5 when a gate signal is ~
applied to each respective control input on gate pin 6. The ~ -
samples voltage appears at pin 11 and has the same level as
appeared at pin 5 when the gate signal was present. The voltage
at pin 11 of each sample and hold circuit 146 or 148 remains ~ ~
unchanged between gate pulses at pin 6. The sample and hold :~ .
circuit 146 is gated to the sample mode whenever the signal
appearing at input terminal. U is logic one. In a similar manne.r,
the sample and hold circuit 148 is gated to the sample mode
whenever the input terminal D is logic one.
Between the sampling operations at either of the .
sample and hold circuits 146 and 148, an integrator reset signal
appears at the input terminal IR and is operative to turn
transistor Q12 on to short circuit a capacitor appearing between
pin 6 and pin 2 of the operational amplifier within the integrator
circuit 142. This resets the integrator so that its output
voltage is zero.
In operation, the sample and hold circuits 146 and
148 are operative to store DC voltages which are representative
of the phase difference between the transmitted signal at one
transducer and the received signal at the other transducer. In
the case of the sample and hold circuit 146 which is gated by the
gate signal at input terminal U, a voltage is stored which,
according to an arbitrary definition, corresponds to the phase
difference between -the signal transmitted by the downstream
--19--
S6~S
transducer 26 and the signal generated in response thereto at
the upstream transducer 24, i.e., a so-called upstream phase
difference. On the other hand, the sample and hold circuit 148
is operative in response to the gate signal D to s-tore a voltage
representing the phase difference between the signal transmitted
by the upstream transducer 24 and the signal generated in
response thereto by the downstream transducer 26, i.e., a so-
called downstream phase difference.
The output from the sample and hold circuits 146 and
148 are respectively coupled to the inverting and the non-
inverting input of an operational amplifier 150 which is
configured to produce a voltage at its output pin 14 equal to
the difference between the voltage applied at the non-inverting
input terminal and the voltage appearing at the inverting input
terminal (0D ~ ~U) . As indicated earlier, this difference is
representative of the uncorrected flow rate of a gas through
the transducer assembly 10 (Fig. 1). In order to properly
calibrate the flowmeter, the non-inverting input terminal of the
operational amplifier has an offset adjustment network, indicated
generally at 152, connected thereto which is operative to adjust ~-
the voltage at the non-inverting input terminal so that the
voltage appearing at the output pin 14 is zero whenever the flow -~
rate through the transducer assembly 10 is also zero. This
offset adjust circuit 152 compensates for the various circuit
offsets especially those for the sample and hold circuits 146
and 148.
The output of the sample and hold circuits 1~6 and :148
are each coupled via a 20K resistor to input pin 5 of a further
sample and hold circuit 154. Since the outputs of sample and
hold circuits 146 and 148 are coupled in the manner shown, the
voltage at pin 5 of the sample and hold circuit 154 comprises
one half of the sum of the two phase differences which are stored
-20-
S6
' `~;
in the two sample and hold circuits 146 and 148. This sum is -
stored within the sample and hold circuit 154 in response to a
summing signal received at the input terminal labelled 5~ The
output of the sample and hold circuit 154 is coupled by a further ;~
operational amplifier 156 which produces a DC voltage at output
pin 8 and is related to the sum of the phase differences stored
in the sample and hold circuits 146 and 148 at the time when the
gate signal occurred. As indicated earlier, this voltage
(~D + 0U) at the output pin 8 of the amplifier 156 comprises
a relative indication of the velocity of sound in the gas in the
transducer assembly 10.
D ~U) appearing at pin 8 of amplifi !
156 is fed back to a low pass filter comprising a 1 Meg ohm
resistor and a one microfarad capacitor. The voltage across this
one microfarad capacitor is coupled to pin 9 of circuit CD 4046
which connects internally to the VCO and thereby adjusts its
frequency of operation. Accordingly, the voltage controlled
oscillator frequency is changed as the velocity of sound in the
gas in transducer 10 changes.
The circuitry of Fig. 4 and Fig. 5 in cooperation with
the circuits oP Fig. 2 is operative to produce a signal at output
pin 14 oP amplifier 150 which relates to the uncorrected flow
through the transducer assembly 10 and a further output at pin 8
of amplifier 156 which is related to the velocity of sound in the
gas in the transducer assembly 10. In accordance with the
conventions established in defining transducer 24 as the upstream
transducer and transducer 26 as the downstream transducer, when-
ever the voltage appearing at pin 14 of amplifier 150 is negative,
this negative voltage indicates that fluid indeed is flowing
through the transducer assembly in a direction from the upstream
-21-
:: ` :`: : ` :
transducer 24 toward the downstream -transducer 26. Additionally,
the magnitude of the voltage appearing at pin 14 of amplifier 150
is related to and can be corrected by the circuitry of Fig. 6
to indicate the flow rate for the fluid passing through the
assembly 10. On the other hand, if the voltage at pin 14 of
amplifier 150 is positive, this indicates that fluid is flowing ~ -'
through the assembly 10 in a direction from the downstream trans-
ducer 26 toward the upstream transducer 24. Again, the magnitude
of the voltage appearing at pin 14 of amplifier 150 corresponds ~i~
to the uncorrected flow rate through the transducer assembly 10.
For the circuit shown in Fig. 5, however, the output
of amplifier 156 is merely an indication related to the velocity
of sound in the gas. In order to determine whether the velocity ~ `
ls greater or less than than for which the instrument was
calibrated, one must record the amplitude at the time it is cal-
ibrated and then compare the cùrrent reading with that previously
recorded value. The circuit, however, can be readily modified
so that the output voltage is equal to zero whenever the
calibrating fluid is present in the transducer assembly. Then,
if the velocity of sound in the fluid subsequently changes, the
output voltage either goes positive or negative and the sign of
the voltage corresponds to whether the velocity has increased or `
decreased compared to the fluid for which the instrument was
calibrated. The magnitude of the output voltage then corresponds
to the relative difference between the sound velocity in the
gas currently passing through the transducers and the sound
velocity in the gas in the transducer assembly at the time it was
calibrated. In order to use the output voltage to provide an
indication of relative sound velocity, more circuits are necessary
since the output of amplifier 156 is dedicated for use as an error
voltage to adjust the frequency of the voltage controlled -
oscilla-tor VCO and as an input to the circuitry of Fig. 6
-22-
:. .. '~
A further alternative configuration permits the
output voltage to be adjusted to equal one when, for example,
air is present in the transducer assembly. Then, whenever the
sound velocity of gas flowing through the transducer assembly
changes, the magnitude of the output voltage corresponds to
the velocity in the gas relative to the velocity in air.
The circuitry of Fig. 6 includes a circuitry for
accepting the uncorrected flow rate from the output of amplifier
150 in Fig. 5 and producing a corrected flow rate output indica~
tion. It was found that the output of ampli-Eier 150 had an
error which is proportional to fl/f2 where fl is the initial
calibration frequency of the VCO in Fi~. 5 and f2 is the frequency
of the VCO when a gas having a different speed of sound is in the
transducer. The circuitry of Fig. 6 multiplies the uncorrected
flow rate output amplifier 150 by f2/fl to produce a corrected
flow rate signal at pin 7 of amplifier LM324B in Fig. 6.
The VCO error voltage which appears at the output of
amplifier 156 in Fig. 5 is proportional to frequency and is used
to establish the correction factor for correcting the flow rate.
The operational amplifier LM308 of Fig. 6 acts as a signal
conditioner and receives the VCO error voltage (0D + 0V) BY
adjustment of the zero adjust resistor coupled to the amplifier
LM308, a zero volt output at pin 6 is established when the error
voltage is at its nominal level (+7.5 volt). The signal
conditioner provides an output voltage of -1.0 volt per +5KHz
deviation.
The output at pin 6 of LM308 then modulates a duty
cycle modulator consisting of amplifiers LM32~A and LM311 and
circuitry coupled thereto. When the VCO operates at the calibra-
tion frequency, the output of LM308 is zero, the duty cycleadjust resistor coupled to LM324A is adjusted to 50% whereas
23
, :,,
~G5~5
FET switches Q14 and Q15 are off 50% of the time and are on
50% of the time. As a result, the gain of amplifier LM324B is
one and no correction is introduced to the uncorrected flow from
amplifier 150 in Fig. 5 and the output at pin 7 of LM324B represents
the flow rate of gas through the transducer.
When the VCO operating frequency changes, a voltage
appears at pin 6 of LM308 in Fig. 6 which is positive for a decrease
in frequency and negative for an increase in frequency. This ~-
voltage modulates the duty cycle generator causing the duty cycle
to change. The duty cycle change causes the gain of amplifier
LM324B to change in accordance with the change of VCO frequency
so that the uncorrected Elow rate signal from Fig. 5 is changed
by LM324B so that the output is proportional to flow rate through
the transducer.
The foregoing description of an acoustical wave
flowmeter has been made with particular emphasis on a preferred ;
electronic circuit which cooperates with a transducer assembly
to provide not only gas flow and direction but also a measurement
of relative sound velocity in the flowing gas. The description
has made some emphasis on the fact that the invention is suitable
as a flowmeter for measuring direction of flow, flow and sound
velocity in a gas, however, the apparatus is equally operable
for measuring direction of flow, flow and sound velocity in li~uids
as well although some circuit elements may require change in
value to optimize performance of the flowmeter for applications
other than gas flow direction, flow and sound velocity measurements.
Furthermore, those of skill in the art will recognize other
modifications to the flowmeter which might be made without departing
from the spirit and scope of the invention as defined in the claims.
Eor example, in place of the phase detector and the phase sum and
difference calculators, the invention may utilize other equivalent
-24-
means to calculate a representation proportional to or equal to -
the speed of acoustic compressions in the transducer traveling
from the transmitting to the receiving transducer. Each
calculated speed has two components, one being the speed of fluid ~ .
flow and the other being the speed of acoustic compressions in
the fluid without fluid flow. One such equivalent speed calcu-
lator may include means to determine the time difference between
the start of acoustic compressions at a transmitting transducer
and the time when the receiving transducer produces a received
signal in response to the acoustic compression. Each time
difference calculated is also proportional to the speed of
acoustic compressions traveling from the transmitting to the
receiving transducer.
Further benefits of the described flowmeter may be
obtained by combinlng or rearranging the signals representing
various parameters. For example, changes in gas composition
which cause a change in molecular weight of the gas mixture also
cause a change in the velocity of sound. Thus a transition from
gas mixture A having an average molecular weight, MA, to gas
mixture B having an average molecular weight, MB, and the fraction
of gas mixture A mixed with gas mixture B may be measured. Such
a techni~ue may be used with a gas flowmeter, for example to
provide a proportional measurement of the fraction of carbon
dioxide in the exhaled breath compared to that in the inhaled
mixture.
Since the velocity of sound is given by:
c ~M meters/second
where ~ = ratio of specific heat at constant pressure to that at
constant volume
k = Boltzmans constant, 1.38 x 10 joules/C
-25-
~ ~.
T = absolute temperature r Celsius
M = mass of the molecules in the gas, kiloyrams,
it can be seen that molecular weight is proportional to c 2 and
for those skilled in the art, an electrical output can be produced
which is proportional to changes in molecular weight; other
variables remaining constant. It is apparent that this simple
case can be extended to combine changes in specific heats and
temperature with changes in molecular weight such that gas mixture
A is typified by one set of conditions and gas mixture B by another
set of conditions.
Additionally, where pressure or density is changing
radically, the volumetric flow may be corrected to approximately
standard conditions or true mass flow by comhining a pressure
measurement with parameters available from the described flowmeter.
Mass flow is given by,
O_O ;
~ ~ mp/kT kilograms/second
M = V
where p = pressure, newtons/meter2.
Thus from the equation for~the velocity of sound, c, above
M = Vpy/c
To those skilled in the art, it is apparent that a
pressure transducer can be connected to measure the absolute
pressure in the flowmeter. Further, the output signal of the
pressure transducer multiplied by the volumetric flow, V, from the
flowmeter and divided by the square of the velocity of sound, c,
also a signal from the flowmeter, along with appropriate constants, ~ !
can provide an approximate value for true mass flow.
While a constant value for y will cause an error in
value for some changes in gas mixture, there are many where the
~ j, ,.~
s
changes in y will be insignificant. The use of the above described
parameters to produce a signal for volumetric flow reduced -to
standard temperature and pressure will also be apparent to those
skilled in the art.
Having described a flowmeter system capable of
accurately measuring flow rate independent of composition it is
also apparent that the techniques may be applied with benefit to
other sonic flowmeters wherein the transducers are not cylindrical
or arcuate and thus may be in the flow path or cause some
10 obstruction or recess along the flow path. That is, any sonic
flowmeter may benefit from these techniques when errors are caused
by changes in velocity of sound in the fLuid.
Those of skill in the art will recognize that the
acoustical flowmeter described above has achieved its principal 3
objective, i.e., the flowmeter accuracy is relatively independent
of changes in gas composition. Those of skill in the art will
also recognize that the above and other changes can be made to
the circuit described above to produce equivalent operation
without depart-i~ng from the spirit and scope of the invention as
20 defined in the claims.
