U.S. patent application number 13/646665 was filed with the patent office on 2014-04-10 for device and method for measurement of ultrasonic transit times.
This patent application is currently assigned to Spire Metering Technology LLC. The applicant listed for this patent is SPIRE METERING TECHNOLOGY LLC. Invention is credited to Chang Shen.
Application Number | 20140096586 13/646665 |
Document ID | / |
Family ID | 50431672 |
Filed Date | 2014-04-10 |
United States Patent
Application |
20140096586 |
Kind Code |
A1 |
Shen; Chang |
April 10, 2014 |
DEVICE AND METHOD FOR MEASUREMENT OF ULTRASONIC TRANSIT TIMES
Abstract
A device for measurement of ultrasonic wave transit times of an
ultrasonic flow sensor consists of: 1) a synchronization signal
generator, 2) a reference pulse generator, 3) a sine wave
generator, 4) an analog signal amplifier, 5) a comparator, 6) a
plurality of latch circuits, 7) a digital adder, 8) an integrator,
9) an A/D converter, 10) a master counter, 11) a plurality of edge
counters, and 12) an arithmetic circuit. The device measures the
ultrasonic wave transit times using a method of averaging the
ultrasonic wave arriving times at different measuring points
(triggering point). This method has less dependency on triggering
threshold level and the ultrasonic signal amplitude, and thus has
less dependency on threshold drift, threshold stability, system
gain fluctuation, electronic noise and signal amplitude variations.
As a result, the method can greatly improve the velocity
measurement accuracy and system robustness of an ultrasonic flow
sensor.
Inventors: |
Shen; Chang; (Acton,
MA) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
SPIRE METERING TECHNOLOGY LLC |
Acton |
MA |
US |
|
|
Assignee: |
Spire Metering Technology
LLC
Acton
MA
|
Family ID: |
50431672 |
Appl. No.: |
13/646665 |
Filed: |
October 6, 2012 |
Current U.S.
Class: |
73/1.24 |
Current CPC
Class: |
G01F 25/00 20130101 |
Class at
Publication: |
73/1.24 |
International
Class: |
G01F 25/00 20060101
G01F025/00 |
Claims
1. A device for measurement of ultrasonic wave transit times of an
ultrasonic flow sensor consists of: a synchronization signal
generator, a reference pulse generator, a sine wave generator, an
analog signal amplifier, a comparator, a plurality of latch
circuits, a digital adder, an integrator, an A/D converter, a
master counter, a plurality of edge counters, and an arithmetic
circuit.
2. The device for measurement of ultrasonic wave transit times of
claim 1, wherein the synchronization signal generator performs the
following functions: 1) initiating the measurement cycle, 2)
triggering the sine wave generator to start sending sine wave
signal to the transmitter of the ultrasonic flow sensor, 3)
triggering the reference pulse generator to start generating high
frequency clock signal, and 4) commanding the master counter to
start counting the reference clock.
3. The device for measurement of ultrasonic wave transit times of
claim 1, wherein the reference pulse generator sends high frequency
clock signal to: 1) the master counter, 2) the edge counters, and
3) the latch circuits. These clocks are used to measure the
arriving time of the ultrasonic wave at different measuring
points.
4. The device for measurement of ultrasonic wave transit times of
claim 1, wherein the sine wave generator sends sine waves to the
transmitter of the ultrasonic flow sensor. After certain period of
time delay, the sine wave signal arrives at the receiver of the
ultrasonic flow sensor with modulated amplitudes.
5. The device for measurement of ultrasonic wave transit times of
claim 1, wherein the analog signal amplifier is AC-coupled. It
amplifies the output signal from the receiver of the ultrasonic
flow sensor.
6. The device for measurement of ultrasonic wave transit times of
claim 1, wherein the comparator compares the signal received from
the analog signal amplifier with the predefined threshold value.
When the received signal becomes higher than the threshold value,
it outputs a positive pulse. On the other hand, when the received
signal becomes lower than the threshold value, it outputs a
negative pulse.
7. The device for measurement of ultrasonic wave transit times of
claim 1, wherein the latch circuits are employed to measure the
time intervals which are less than one reference clock cycle and
cannot be counted by the master counter and the edge counters.
8. The device for measurement of ultrasonic wave transit times of
claim 1, wherein the digital adder is used to add the outputs of
all the latch circuits together and then output the summed signal
to the integrator.
9. The device for measurement of ultrasonic wave transit times of
claim 1, wherein the integrator converts the short pulse from the
adder to an analog exponential signal.
10. The device for measurement of ultrasonic wave transit times of
claim 1, wherein the A/D converter converts the analog output of
the integrator to a digital value.
11. The device for measurement of ultrasonic wave transit times of
claim 1, wherein the master counter is used to measure the arriving
time of ultrasonic wave at the first measuring point.
12. The device for measurement of ultrasonic wave transit times of
claim 1, wherein the edge counters are used to measure the arriving
time of ultrasonic wave at the subsequent measuring points after
the first measuring point.
13. The device for measurement of ultrasonic wave transit times of
claim 1, wherein the arithmetic circuit calculates the ultrasonic
wave transit times from the inputs of the master counter, the edge
counters and the A/D converter, by averaging the arriving times at
different measuring points. This time measurement method has less
dependency on triggering threshold level and ultrasonic signal
amplitudes, and thus has less dependency on threshold drift,
threshold stability, system gain fluctuation and signal amplitude
variations. With multi-point triggering and multiple transit-time
averaging as explained in previous section, this time measurement
method is also more robust again noise interference than prior art.
As a result, the method can greatly improve the velocity
measurement accuracy of an ultrasonic flow sensor.
14. The device for measurement of ultrasonic wave transit times of
claim 1, wherein an even number of latch circuits are employed. And
the number of edge counters is always one less than the total
number of the latch circuits. For example, if the total number of
latch circuits is eight, the total number of the edge counters is
seven.
15. The device for measurement of ultrasonic wave transit times of
claim 14, wherein two latch circuits and one edge counter are
employed. The ultrasonic wave transit time is calculated based on
the average of the arriving times at the first and second measuring
points.
16. The device for measurement of ultrasonic wave transit times of
claim 14, wherein four latch circuits and three edge counters are
employed. The ultrasonic wave transit time is calculated based on
the average of arriving times at the 1-4 measuring points.
17. The device for measurement of ultrasonic wave transit times of
claim 16, wherein the ultrasonic wave transit time is calculated
based on the average of arriving times at the third and fourth
measuring points.
18. The device for measurement of ultrasonic wave transit times of
claim 14, wherein eight latch circuits and seven edge counters are
employed. The ultrasonic wave transit time is calculated based on
the average of arriving times at the 1-8 measuring points.
19. The device for measurement of ultrasonic wave transit times of
claim 18, the ultrasonic wave transit time is calculated based on
the average of arriving times at the fifth and sixth measuring
points.
20. The device for measurement of ultrasonic wave transit times of
claim 18, the ultrasonic wave transit time is calculated based on
the average of arriving times at the seventh and eighth measuring
points.
Description
CROSS-REFERENCE TO RELATED APPLICATIONS
[0001] Not Applicable.
STATEMENT REGARDING FEDERALLY SPONSORED RESEARCH OR DEVELOPMENT
[0002] Not Applicable.
REFERENCE TO SEQUENCE LISTING, A TABLE, OR A COMPUTER PROGRAM
LISTING COMPACT DISC APPENDIX
[0003] Not Applicable.
FIELD OF THE INVENTION
[0004] The present disclosure relates to a device for measuring
ultrasonic wave transit times from the transmitter to the receiver
of an ultrasonic flow sensor.
BACKGROUND OF THE INVENTION
[0005] The statements in this section merely provide background
information related to the present disclosure and may not
constitute prior art.
[0006] An ultrasonic flow sensor measures the average velocity of
liquid or gaseous media by means of ultrasonic transducers based on
the principle that the transit time of an ultrasonic wave from the
transmitter of a transducer to the corresponding receiver is
determined by the fluid velocity and the ultrasonic wave
propagating direction. Normally, a pair of transducers is used, one
is installed in upstream and the other is installed in downstream.
Each transducer can be used as a transmitter or a receiver. One
ultrasonic wave is transmitted from the upstream transducer to the
downstream transducer. The second ultrasonic pulse is transmitted
from the downstream transducer to the upstream transducer. The
transit time in each direction is measured by an electronic device.
The difference of the two transit-time data is proportional to flow
velocity. It is then used to calculate the average flow velocity of
the fluid.
[0007] In conventional electronic devices of an ultrasonic flow
sensor, the ultrasonic wave transit time is measured by a time
counter to count a reference clock using the following method. 1)
Sending ultrasonic pulse wave to the transmitter, starting the
timer counter. 2) Monitoring the ultrasonic signal received by the
receiver, when the received signal becomes higher than the
predefined threshold value, immediately stopping the time counter,
and recording the arriving time. This arriving time is treated as
the transmit time.
[0008] In the above approach, an analog integrator may be used to
measure the residual time from the counter stopping moment to the
rising edge of the next cycle of the reference clock. This residual
time is then combined with the previous transit-time to obtain a
transit-time with higher accuracy.
[0009] However, this measurement method is susceptible to
electronic noise and condition variation. Both the strength of the
received signal and the predefined threshold value are subject to
electronic noise. In addition, the strength of the received signal
varies with the fluid properties such as temperature, velocity,
turbulence, solids concentration, etc. As a result, the measured
transit time changes not only with flow velocity, but with the
fluid properties and electronic noise level. This significantly
reduces the velocity measurement accuracy and stability of an
ultrasonic flow sensor.
BRIEF SUMMARY OF THE INVENTION
[0010] Further areas of applicability will become apparent from the
description provided herein. It should be understood that the
description and specific examples are intended for purposes of
illustration only and are not intended to limit the scope of the
present disclosure.
[0011] The object of the present invention is to provide an
electronic device which can accurately and reliably measure the
transit times from the transmitter to the receiver and hence
improving the velocity measurement accuracy of an ultrasonic flow
sensor.
[0012] In more detail the present invention provides an electronic
device for measurement of ultrasonic wave transit times of an
ultrasonic flow sensor consists of: 1) a synchronization signal
generator, 2) a reference pulse generator, 3) a sine wave
generator, 4) an analog signal amplifier, 5) a comparator, 6) a
plurality of latch circuits, 7) a digital adder, 8) an integrator,
9) an A/D converter, 10) a master counter, 11) a plurality of edge
counters, and 12) an arithmetic circuit (microprocessor). The
device measures the ultrasonic wave transit times using a threshold
level to trigger both the rising edge and falling edge of the
received ultrasonic signal, and using a method of averaging the
ultrasonic wave arriving times at different measuring points. This
method has less dependency on the threshold level and the
ultrasonic signal amplitude, thus, has less dependency on threshold
drift, threshold stability, system gain fluctuation, electronic
noise and signal amplitude variations. As a result, this method can
greatly improve the velocity measurement accuracy and system
robustness of an ultrasonic flow sensor.
BRIEF DESCRIPTION OF THE SEVERAL VIEWS OF THE DRAWINGS
[0013] The drawings described herein are for illustration purposes
only and are not intended to limit the scope of the present
disclosure in any way.
[0014] FIG. 1 shows a block diagram of the first embodiment of a
device for measurement of the ultrasonic wave transit times of the
present disclosure.
[0015] FIG. 2 shows the operational waveform diagram of the device
shown in FIG. 1.
[0016] FIG. 3(a) illustrates the transit time measurement error
caused by the threshold fluctuation of prior art.
[0017] FIG. 3(b) illustrates the transit time measurement error
caused by the signal amplitude fluctuation of prior art.
[0018] FIG. 4(a) illustrates the transit time measurement error
reduction of the first embodiment of the present disclosure against
threshold fluctuation interference.
[0019] FIG. 4(b) illustrates the transit time measurement error
reduction of the first embodiment of the present disclosure against
signal amplitude fluctuation interference.
[0020] FIG. 5 shows a block diagram of the second embodiment of a
device for measurement of the ultrasonic wave transit times of the
present disclosure.
[0021] FIG. 6 shows the operational waveform diagram of the device
shown in FIG. 5.
[0022] FIG. 7(a) illustrates the transit time measurement error
reduction of the second embodiment of the present disclosure
against threshold level fluctuation interference.
[0023] FIG. 7(b) illustrates the transit time measurement error
reduction of the second embodiment of the present disclosure
against signal amplitude fluctuation interference.
[0024] FIG. 8 shows a block diagram of the third embodiment of a
device for measurement of the ultrasonic wave transit times of the
present disclosure.
[0025] FIG. 9 shows the operational waveform diagram of the device
shown in FIG. 8.
[0026] FIG. 10 illustrates the transit time measurement error
reduction of the third embodiment of the present disclosure.
DETAILED DESCRIPTION OF THE INVENTION
[0027] The following description is merely exemplary in nature and
is not intended to limit the present disclosure, application, or
uses. It should be understood that throughout the drawings,
corresponding reference numerals indicate like or corresponding
parts and features.
[0028] FIG. 1 illustrates the first embodiment of an electronic
device for measuring the ultrasonic wave transit times of the
present disclosure. In this first embodiment, the device 10
consists of a synchronization signal generator 20, a reference
pulse generator 30, a sine wave generator 40, an analog signal
amplifier 60, a comparator 70, two latch circuits 80 and 81, a
digital adder 90, an integrator 100, an A/D converter 110, a master
pulse counter 120, an edge counter 121, and an arithmetic circuit
130.
[0029] Referring to FIGS. 1 and 2, the synchronization signal
generator 20 outputs a pulse shown in FIG. 2a. This pulse is used
to perform the following functions: 1) initiating the measurement
cycle, 2) triggering the sine wave generator 40 to start sending
sine wave signal to the transmitter of the ultrasonic flow sensor
50, 3) triggering the reference pulse generator 30 to start
generating high frequency clock signal, and 4) commanding the
master counter 120 to start counting the reference pulses.
[0030] Referring to FIGS. 1 and 2, after receiving the
synchronization pulse, the reference pulse generator 30 starts
sending high frequency clock signal to: 1) the master counter 120,
2) the edge counter 121, and 3) the two latch circuits 80-81, as
shown in FIG. 2e.
[0031] Referring to FIGS. 1 and 2, after receiving the
synchronization pulse, the sine wave generator 40 starts sending
sine wave signals (FIG. 2b) to the transmitter of the ultrasonic
flow sensor 50. After certain period of time delay, the sine wave
signal arrives at the receiver of the ultrasonic flow sensor 50
with modulated amplitude, as shown in FIG. 2c.
[0032] Referring to FIGS. 1 and 2, the AC-coupled analog signal
amplifier 60 amplifies the output signal from the receiver of the
ultrasonic flow sensor 50.
[0033] Referring to FIGS. 1 and 2, the comparator 70 compares the
signal received from the analog signal amplifier 60 with the
predefined threshold value. When the received signal becomes higher
than the threshold value, it outputs a positive pulse. On the other
hand, when the received signal becomes lower than the threshold
value, it outputs a negative pulse, as shown in FIG. 2d.
[0034] Referring to FIGS. 1 and 2, after receiving the positive
pulse from the comparator 70, the master counter 120 stops counting
the reference clock, as shown in FIG. 2e. The time interval, C0,
measured by the master counter 120 can be described by the
equation:
C0=NTr.
Where N is the output of the mater counter 120, Tr is the period of
the reference clock. The master counter 120 can only count complete
clock cycles, its output N is a positive integer number, any time
less than one clock cycle will not be counted.
[0035] Referring to FIGS. 1 and 2, after receiving the positive
pulse from the comparator 70, the edge counter 121 starts counting
the reference pulses. After receiving the negative pulse from the
comparator 70, the edge counter 121 stops counting the pulses, as
shown in FIG. 2e. The time interval, C1, measured by the edge
counter 121 can be described by the equation:
C1=N1Tr.
Where N1 is the output of the first counter 121. Similar to the
master counter, the edge counter 121 can only count complete clock
cycles, its output N1 is a positive integer number, any time less
than one clock cycle will not be counted.
[0036] Referring to FIGS. 1 and 2, the latch circuit 80 is used to
measure the time interval t1, between the positive pulse from the
comparator 70 and the first upward edge of the reference clock fed
into the edge counter 121, as shown in FIG. 2f.
[0037] Referring to FIGS. 1 and 2, the latch circuit 81 is used to
measure the time interval t2, between the negative pulse from the
comparator 70 and the next upward edge of the reference clock after
the edge counter 121 is stopped, as shown in FIG. 2f.
[0038] Referring to FIG. 1, the outputs of the latch circuits 80
and 81 are fed into the adder circuit 90. They are added together,
and then output to the integrator circuit 100.
[0039] Referring to FIGS. 1 and 2, since the t1 and t2 time
intervals from the latch circuits 80 and 81 are very short, the
integrator circuit 100 is used to convert these short pulses to
analog exponential waves, as shown in FIG. 2g.
[0040] Referring to FIG. 1, the analog signal from the integrator
100 is then converted to a digital value by the A/D converter 110,
and fed into the arithmetic circuit 130.
[0041] Referring to FIG. 2, the arriving time of the ultrasonic
wave, T1, measured at the first measurement point P1, can be
described by the equation:
T1=C0-t1=NTr-t1
[0042] Referring to FIG. 2, the arriving time of the ultrasonic
wave, T2, measured at the second measurement point P2, can be
described by the equation:
T2=C0+C1-t2=(N+N1)Tr-t2.
[0043] Referring to FIG. 2, the arithmetic circuit 130 calculates
the ultrasonic wave transit time based on the following
formula:
Tm=(T1+T2)/2=NTr+N1Tr/2-(t1+t2)/2.
[0044] Referring to FIGS. 3(a) and (b), the ultrasonic wave transit
time with prior art is based on the following formula:
Tm=NTr-t1. [0045] Obviously, the transit-time Tm obtained by prior
art differs from the one obtained by the first embodiment of the
present disclosure by N1Tr/2. This difference does not have any
impact on the flow measurement, because the flow rate is calculated
based on transit-time difference between upstream Tm and downstream
Tm. In addition, the difference can be calibrated so to have
accurate transit-time measurement.
[0046] FIG. 3 (a) illustrates the transit time measurement error
caused by the threshold level fluctuation of prior art. In this
case the time measurement error is:
.DELTA.Tm=|.DELTA.t1|.
[0047] FIG. 3 (b) illustrates the transit time measurement error
caused by signal amplitude fluctuation of prior art. In this case
the time measurement error is:
.DELTA.Tm=|.DELTA.t1|.
[0048] FIG. 4(a) illustrates the transit time measurement error
caused by threshold fluctuation of the first embodiment of the
present disclosure. In this case the time measurement error is:
.DELTA.Tm=|.DELTA.t1+.DELTA.t2|/2.
[0049] FIG. 4(b) illustrates the transit time measurement error
caused by signal amplitude fluctuation of the first embodiment of
the present disclosure. In this case the time measurement error
is:
.DELTA.Tm=|.DELTA.t1+.DELTA.t2|/2.
It is noted from the FIGS. 4(a) and (b) that .DELTA.t1 and
.DELTA.t2 change in opposite direction in similar magnitude, when
.DELTA.t1 increases, .DELTA.t2 decreases, and vice versa. i.e.:
.DELTA.t1.apprxeq.-.DELTA.t2
As a result, their average, .DELTA.Tm, is always smaller than
|.DELTA.t1|. In effect,
.DELTA.Tm.apprxeq.0.
This indicates that the transit-time obtained by the present
invention does not change with threshold drifting or signal
amplitude variation. By contract, the transit-time obtained by
prior art is sensitive to threshold drifting and signal amplitude
variation. As a result of this, the transit time measurement
accuracy and reliability are greatly improved by using the method
of the present disclosure compared to the method of prior art.
[0050] FIG. 5 illustrates the second embodiment of an electronic
device for measuring the ultrasonic wave transit times of the
present disclosure. In this second embodiment, the device 10
consists of a synchronization signal generator 20, a reference
pulse generator 30, a sine wave generator 40, an analog signal
amplifier 60, a comparator 70, four latch circuits 80-83, a digital
adder 90, an integrator 100, an A/D converter 110, a master counter
120, three edge counters 121-123, and an arithmetic circuit
130.
[0051] Referring to FIGS. 5 and 6, the synchronization signal
generator 20 outputs a pulse shown in FIG. 6a. This pulse is used
to perform the following functions: 1) initiating the measurement
cycle, 2) triggering the sine wave generator 40 to start sending
sine wave signal to the transmitter of the ultrasonic flow sensor
50, 3) triggering the reference pulse generator 30 to start
generating high frequency clock signal, and 4) commanding the
master counter 120 to start counting the reference clock
cycles.
[0052] Referring to FIGS. 5 and 6, after receiving the
synchronization pulse, the reference pulse generator 30 starts
sending high frequency clock signal to: 1) the master counter 120,
2) the edge counters 121-123, and 3) the latch circuits 80-83, as
shown in FIG. 6e.
[0053] Referring to FIGS. 5 and 6, after receiving the
synchronization pulse, the sine wave generator 40 starts sending
sine waves (FIG. 6b) to the transmitter of the ultrasonic flow
sensor 50. After certain period of time delay, the sine wave signal
arrives at the receiver of the flow meter with modulated amplitude,
as shown in FIG. 6c.
[0054] Referring to FIGS. 5 and 6, the AC-coupled analog signal
amplifier 60 amplifies the output signal from the receiver of the
ultrasonic flow sensor 50.
[0055] Referring to FIGS. 5 and 6, the comparator 70 compares the
signal received from the analog signal amplifier 60 with the
predefined threshold value. When the received signal becomes higher
than the threshold value, it outputs a positive pulse. On the other
hand, when the received signal becomes lower than the threshold
value, it outputs a negative pulse, as shown in FIG. 6d.
[0056] Referring to FIGS. 5 and 6, after receiving the positive
pulse from the comparator 70, the master counter 120 stops counting
the reference clock, as shown in FIG. 6e. The time interval, C0,
measured by the master counter 120 can be described by the
equation:
C0=NTr,
where N is the output of the mater counter 120, Tr is the period of
the reference clock. The master counter 120 can only count complete
clock cycles, its output N is a positive integer number, any time
less than one clock cycle will not be counted.
[0057] Referring to FIGS. 5 and 6, after receiving the first
positive pulse from the comparator 70, the edge counter 121 starts
counting the reference clock. After receiving the first negative
pulse from the comparator 70, the edge counter 121 stops counting
the clock, as shown in FIG. 6e. The time interval, C1, measured by
the edge counter 121 can be described by the equation:
C1=N1Tr,
where N1 is the output of the counter 121. Similar to the master
counter 120, the edge counter 121 can only count complete clock
cycles, its output N1 is a positive integer number, any time less
than one clock cycle will not be counted.
[0058] Referring to FIGS. 5 and 6, after receiving the first
negative pulse from the comparator 70, the edge counter 122 starts
counting the reference clock. After receiving the second positive
pulse from the comparator 70, the edge counter 122 stops counting
the clock, as shown in FIG. 6e. The time interval, C2, measured by
the edge counter 122 can be described by the equation:
C2=N2Tr,
where N2 is the output of the counter 122. Similar to the master
counter 120, the edge counter 122 can only count complete clock
cycles, its output N2 is a positive integer number, any time less
than one cycle pulse will not be counted.
[0059] Referring to FIGS. 5 and 6, after receiving the second
positive pulse from the comparator 70, the edge counter 123 starts
counting the reference clock. After receiving the second negative
pulse from the comparator 70, the edge counter 123 stops counting
the reference clock, as shown in FIG. 6e. The time interval, C3,
measured by the edge counter 123 can be described by the
equation:
C3=N3Tr,
where N3 is the output of the counter 123. Similar to the master
counter 120, the edge counter 123 can only count complete clock
cycles, its output N3 is a positive integer number, any time less
than one clock cycle will not be counted.
[0060] Referring to FIGS. 5 and 6, the latch circuit 80 is used to
measure the time interval t1 between the first positive pulse from
the comparator 70 and the first upward edge of the reference clock
fed into the edge counter 121, as shown in FIG. 6f. Since time
interval t1 is less than one complete reference clock, it cannot be
measured by the master counter 120.
[0061] Referring to FIGS. 5 and 6, the latch circuit 81 is used to
measure the time interval t2 between the first negative pulse from
the comparator 70 and the first upward edge of the reference clock
fed into the edge counter 122, as shown in FIG. 6f. Since time
interval t2 is less than one complete reference clock, it cannot be
measured by the edge counter 121.
[0062] Referring to FIGS. 5 and 6, the latch circuit 82 is used to
measure the time interval t3 between the second positive pulse from
the comparator 70 and the first upward edge of the reference clock
fed into the edge counter 123, as shown in FIG. 6f. Since time
interval t3 is less than one complete reference clock, it cannot be
measured by the edge counter 122.
[0063] Referring to FIGS. 5 and 6, the latch circuit 83 is used to
measure the time interval t4 between the second negative pulse from
the comparator 70 and the next upward edge of the reference clock
after the edge counter 123 is stopped, as shown in FIG. 6f. Since
time interval t4 is less than one complete reference clock, it
cannot be measured by the edge counter 123.
[0064] Referring to FIG. 5, the outputs of latch circuits 80-83 are
fed into the adder circuit 90. They are added together and then
output to the integrator circuit 100.
[0065] Referring to FIGS. 5 and 6, since the time intervals t1-t4
from the latch circuits 80-83 are very short, the integrator
circuit 100 is used to convert these short pulses to triangular
waves, as shown in FIG. 2g.
[0066] Referring to FIG. 5, the analog signal from the integrator
100 is then converted to a digital value by the A/D converter 110,
and fed into the arithmetic circuit 130.
[0067] Referring to FIG. 6, the arriving time of the ultrasonic
wave, T1, measured at the first measurement point P1, can be
described by the equation:
T1=C0-t1=NTr-t1.
[0068] Referring to FIG. 6, the arriving time of the ultrasonic
wave, T2, measured at the second measurement point P2, can be
described by the equation:
T2=C0+C1-t2=(N+N1)Tr-t2.
[0069] Referring to FIG. 6, the arriving time of the ultrasonic
wave, T3, measured at the third measurement point P3, can be
described by the equation:
T 3 = C 0 + C 1 + C 2 - t 3 = ( N + N 1 + N 2 ) Tr - t 3.
##EQU00001##
[0070] Referring to FIG. 6, the arriving time of the ultrasonic
wave, T4, measured at the fourth measurement point P4, can be
described by the equation:
T 4 = C 0 + C 1 + C 2 + C 4 - t 4 = T 3 + C 3 + t 3 - t 4 = ( N + N
1 + N 2 + N 3 ) Tr - t 4. ##EQU00002##
[0071] Referring to FIG. 5, the arithmetic circuit 130 calculates
the ultrasonic wave transit time based on the following
formula:
Tm = [ ( T 1 + T 2 ) / 2 + ( ( T 3 + T 4 ) / 2 - Tx ) ] / 2 = NTr +
T_ 123 - ( t 1 + t 2 + t 3 + t 4 ) / 4. ##EQU00003## [0072] Here Tx
is the period of the received ultrasonic signal. The center of T3
and T4 is always one period away from the center of T1 and T2.
T.sub.--123 is expressed as following,
[0072] T.sub.--123=(3N1+2N2+N3)Tr/4-Tx/2.
[0073] FIG. 7 illustrates the transit time measurement error of the
second embodiment of the present disclosure. In this case the time
measurement error is:
.DELTA.Tm=|.DELTA.t1+.DELTA.t2+.DELTA.t3+.DELTA.t4|/4.
Since (.DELTA.t1, .DELTA.t2) and (.DELTA.t3, .DELTA.t4) change in
opposite directions, their average is always smaller than |t1|.
This indicates that the ultrasonic wave transit time measurement
accuracy is greatly improved using the method of the present
disclosure compared to the method of prior art.
[0074] FIG. 8 illustrates the third embodiment of an electronic
device for measuring the ultrasonic wave transit times of the
present disclosure. In this embodiment, the device 10 consists of a
synchronization signal generator 20, a reference pulse generator
30, a sine wave generator 40, an analog signal amplifier 60, a
comparator 70, eight latch circuits 80-87, a digital adder 90, an
integrator 100, an A/D converter 110, a master counter 120, seven
edge counters 121-127, and an arithmetic circuit 130.
[0075] Referring to FIGS. 8 and 9, the synchronization signal
generator 20 outputs a pulse shown in FIG. 9a. This pulse is used
to perform the following functions: 1) initiating the measurement
cycle, 2) triggering the sine wave generator 40 to start sending
sine wave signal to the transmitter of the ultrasonic flow sensor
50, 3) triggering the reference pulse generator 30 to start
generating high frequency clock signal, and 4) commanding the
master counter 120 to start counting the reference clock
cycles.
[0076] Referring to FIGS. 8 and 9, after receiving the
synchronization pulse, the reference pulse generator 30 starts
sending high frequency clock signal to: 1) the master counter 120,
2) the edge counters 121-127, and 3) the latch circuits 80-87, as
shown in FIG. 9e.
[0077] Referring to FIGS. 8 and 9, after receiving the
synchronization pulse, the sine wave generator 40 starts sending
sine wave signals (FIG. 9b) to the transmitter of the ultrasonic
flow sensor 50. After certain period of time delay, the sine wave
signal arrives at the receiver of the ultrasonic flow sensor 50
with modulated amplitude, as shown in FIG. 9c.
[0078] Referring to FIGS. 8 and 9, the AC-coupled analog signal
amplifier 60 amplifies the output signal from the receiver of the
ultrasonic flow sensor 50.
[0079] Referring to FIGS. 8 and 9, the comparator 70 compares the
signal received from the analog signal amplifier 60 with the
predefined threshold value. When the received signal becomes higher
than the threshold value, it outputs a positive pulse. On the other
hand, when the received signal becomes lower than the threshold
value, it outputs a negative pulse, as shown in FIG. 9d.
[0080] Referring to FIGS. 8 and 9, after receiving the positive
pulse from the comparator 70, the master counter 120 stops counting
the reference clock, as shown in FIG. 9e. The time interval, C0,
measured by the master counter 120 can be described by the
equation:
C0=NTr,
where N is the output of the mater counter 120, Tr is the period of
the reference clock. The master counter 120 can only count complete
clock cycles, its output N is a positive integer number, any time
less than one clock cycle will not be counted.
[0081] Referring to FIGS. 8 and 9, after receiving the first
positive pulse from the comparator 70, the edge counter 121 starts
counting the reference clock. After receiving the first negative
pulse from the comparator 70, the edge counter 121 stops counting
the reference clock, as shown in FIG. 9e. The time interval, C1,
measured by the edge counter 121 can be described by the
equation:
C1=N1Tr,
where N1 is the output of the counter 121. Similar to the master
counter 120, the edge counter 121 can only count complete clock
cycles, its output N1 is a positive integer number, any time less
than one clock cycle will not be counted.
[0082] Referring to FIGS. 8 and 9, after receiving the first
negative pulse from the comparator 70, the edge counter 122 starts
counting the reference clock. After receiving the second positive
pulse from the comparator 70, the edge counter 122 stops counting
the reference clock, as shown in FIG. 9e. The time interval, C2,
measured by the edge counter 122 can be described by the
equation:
C2=N2Tr,
where N2 is the output of the counter 122. Similar to the master
counter 120, the edge counter 122 can only count complete clock
cycles, its output N2 is a positive integer number, any time less
than one clock cycle will not be counted.
[0083] Referring to FIGS. 8 and 9, after receiving the second
positive pulse from the comparator 70, the edge counter 123 starts
counting the reference clock. After receiving the second negative
pulse from the comparator 70, the edge counter 123 stops counting
the reference clock, as shown in FIG. 9e. The time interval, C3,
measured by the edge counter 123 can be described by the
equation:
C3=N3Tr,
where N3 is the output of the counter 123. Similar to the master
counter 120, the edge counter 123 can only count complete clock
cycles, its output N3 is a positive integer number, any time less
than one clock cycle will not be counted.
[0084] Referring to FIGS. 8 and 9, after receiving the second
negative pulse from the comparator 70, the edge counter 124 starts
counting the reference clock. After receiving the third positive
pulse from the comparator 70, the edge counter 124 stops counting
the reference clock, as shown in FIG. 9e. The time interval, C4,
measured by the edge counter 124 can be described by the
equation:
C4=N4Tr,
where N4 is the output of the counter 124. Similar to the master
counter 120, the edge counter 124 can only count complete clock
cycles, its output N4 is a positive integer number, any time less
than one clock cycle will not be counted.
[0085] Referring to FIGS. 8 and 9, after receiving the third
positive pulse from the comparator 70, the edge counter 125 starts
counting the reference clock. After receiving the third negative
pulse from the comparator 70, the edge counter 125 stops counting
the reference clock, as shown in FIG. 9e. The time interval, C5,
measured by the edge counter 125 can be described by the
equation:
C5=N5Tr,
where N5 is the output of the counter 125. Similar to the master
counter 120, the edge counter 125 can only count complete clock
cycles, its output N5 is a positive integer number, any time less
than one clock cycle will not be counted.
[0086] Referring to FIGS. 8 and 9, after receiving the third
negative pulse from the comparator 70, the edge counter 126 starts
counting the reference clock. After receiving the fourth positive
pulse from the comparator 70, the edge counter 126 stops counting
the reference clock, as shown in FIG. 9e. The time interval, C6,
measured by the edge counter 126 can be described by the
equation:
C6=N6Tr,
where N6 is the output of the counter 126. Similar to the master
counter 120, the edge counter 126 can only count complete clock
cycles, its output N6 is a positive integer number, any time less
than one clock cycle will not be counted.
[0087] Referring to FIGS. 8 and 9, after receiving the fourth
positive pulse from the comparator 70, the edge counter 127 starts
counting the reference clock. After receiving the fourth negative
pulse from the comparator 70, the edge counter 127 stops counting
the reference clock, as shown in FIG. 9e. The time interval, C7,
measured by the edge counter 127 can be described by the
equation:
C7=N7Tr,
where N7 is the output of the counter 127. Similar to the master
counter 120, the edge counter 127 can only count complete clock
cycles, its output N7 is a positive integer number, any time less
than one clock cycle will not be counted.
[0088] Referring to FIGS. 8 and 9, the latch circuit 80 is used to
measure the time interval t1 between the first positive pulse from
the comparator 70 and the first upward edge of the reference clock
fed into the edge counter 121, as shown in FIG. 9f. Since time
interval t1 is less than one complete reference clock, it cannot be
measured by the master counter 120.
[0089] Referring to FIGS. 8 and 9, the latch circuit 81 is used to
measure the time interval t2 between the first negative pulse from
the comparator 70 and the first upward edge of the reference clock
fed into the edge counter 122, as shown in FIG. 9f. Since time
interval t2 is less than one complete reference clock, it cannot be
measured by the edge counter 121.
[0090] Referring to FIGS. 8 and 9, the latch circuit 82 is used to
measure the time interval t3 between the second positive pulse from
the comparator 70 and the first upward edge of the reference clock
fed into the edge counter 123, as shown in FIG. 9f. Since time
interval t3 is less than one complete reference clock, it cannot be
measured by the edge counter 122.
[0091] Referring to FIGS. 8 and 9, the latch circuit 83 is used to
measure the time interval t4 between the second negative pulse from
the comparator 70 and the first upward edge of the reference clock
fed into the edge counter 124, as shown in FIG. 9f. Since time
interval t4 is less than one complete reference clock, it cannot be
measured by the edge counter 123.
[0092] Referring to FIGS. 8 and 9, the latch circuit 84 is used to
measure the time interval t5 between the third positive pulse from
the comparator 70 and the first upward edge of the reference clock
fed into the edge counter 125, as shown in FIG. 9f. Since time
interval t5 is less than one complete reference clock, it cannot be
measured by the edge counter 124.
[0093] Referring to FIGS. 8 and 9, the latch circuit 85 is used to
measure the time interval t6 between the third negative pulse from
the comparator 70 and the first upward edge of the reference clock
fed into the edge counter 126, as shown in FIG. 9f. Since time
interval t6 is less than one complete reference clock, it cannot be
measured by the edge counter 125.
[0094] Referring to FIGS. 8 and 9, the latch circuit 86 is used to
measure the time interval t7 between the fourth positive pulse from
the comparator 70 and the first upward edge of the reference clock
fed into the edge counter 127, as shown in FIG. 9f. Since time
interval t7 is less than one complete reference clock, it cannot be
measured by the edge counter 126.
[0095] Referring to FIGS. 8 and 9, the latch circuit 87 is used to
measure the time interval t8 between the fourth negative pulse from
the comparator 70 and the upward edge of the next reference clock
after the edge counter 127 is stopped, as shown in FIG. 9f. Since
time interval t8 is less than one complete reference clock, it
cannot be measured by the edge counter 127.
[0096] Referring to FIG. 8, the outputs of the latch circuits 80-87
are fed into the adder circuit 90. They are added together and then
output to the integrator circuit 100.
[0097] Referring to FIGS. 8 and 9, since the time intervals t1-t8
from the latch circuits 80-87 are short pulses, the integrator
circuit 100 is used to convert these short pulses to triangular
waves, as shown in FIG. 9g.
[0098] Referring to FIG. 8, the analog signal from the integrator
100 is then converted to a digital value by the A/D converter 110,
and fed into the arithmetic circuit 130.
[0099] Referring to FIG. 9, the arriving time of the ultrasonic
wave, T1, measured at the first measurement point P1, can be
described by the equation:
T1=C0-t1=NTr-t1.
[0100] Referring to FIG. 9, The arriving time of the ultrasonic
wave, T2, measured at the second measurement point P2, can be
described by the equation:
T2=C0+C1-t2=(N+N1)Tr-t2.
[0101] Referring to FIG. 9, The arriving time of the ultrasonic
wave, T3, measured at the third measurement point P3, can be
described by the equation:
T3=C0+C1+C2-t3=(N+N1+N2)Tr-t3.
[0102] Referring to FIG. 9, The arriving time of the ultrasonic
wave, T4, measured at the fourth measurement point P4, can be
described by the equation:
T4=C0+C1+C2+C3-t4=(N+N1+N2+N3)Tr-t4.
[0103] Referring to FIG. 9, The arriving time of the ultrasonic
wave, T5, measured at the fourth measurement point P5, can be
described by the equation:
T 5 = C 0 + C 1 + C 2 + C 3 + C 4 - t 5 = ( N + N 1 + N 2 + N 3 + N
4 ) Tr - t 5. ##EQU00004##
[0104] Referring to FIG. 9, The arriving time of the ultrasonic
wave, T6, measured at the fourth measurement point P6, can be
described by the equation:
T 6 = C 0 + C 1 + C 2 + C 3 + C 4 + C 5 - t 6 = ( N + N 1 + N 2 + N
3 + N 4 + N 5 ) Tr - t 6. ##EQU00005##
[0105] Referring to FIG. 9, The arriving time of the ultrasonic
wave, T7, measured at the fourth measurement point P7, can be
described by the equation:
T 7 = C 0 + C 1 + C 2 + C 3 + C 4 + C 5 + C 6 - t 7 = ( N + N 1 + N
2 + N 3 + N 4 + N 5 + N 6 ) Tr - t 7. ##EQU00006##
[0106] Referring to FIG. 9, The arriving time of the ultrasonic
wave, T8, measured at the fourth measurement point P8, can be
described by the equation:
T 8 = C 0 + C 1 + C 2 + C 3 + C 4 + C 5 + C 6 + C 7 - t 8 = ( N + N
1 + N 2 + N 3 + N 4 + N 5 + N 6 + N 7 ) Tr - t 8. ##EQU00007##
[0107] Referring to FIG. 8, the arithmetic circuit 130 calculates
the ultrasonic wave transit time based on the following
formula:
Tm = [ ( T 1 + T 2 ) / 2 + ( T 3 + T 4 ) / 2 - Tx + ( T 5 + T 6 ) /
2 - 2 Tx + ( T 7 + T 8 ) / 2 - 3 Tx ] 4 = ( N + N 1 + N 2 + N 3 + N
4 + N 5 + N 6 + N 7 ) / 8 - 1.5 Tx = NTr + T_ 1 _ 8 - ( t 1 + t 2 +
t 3 + t 4 + t 5 + t 6 + t 7 + t 8 ) / 8. ##EQU00008##
[0108] Here Tx is the period of the received ultrasonic signal. The
center of T3 and T4 is always one period away from the center of T1
and T2. Similarly, the center of T5 and T6 is always one period
away from the center of T3 and T4, and etc. The term
T.sub.--1.sub.--8 can be expressed as follows,
T.sub.--1.sub.--8=(7N1+6N2+5N3+4N4+3N5+2N6+N7)Tr/8-1.5Tx.
[0109] FIG. 10 illustrates the transit time measurement error of
the third embodiment of the present disclosure. In this case the
time measurement error is:
.DELTA.Tm=|.DELTA.t1+.DELTA.t2+.DELTA.t3+.DELTA.t4+.DELTA.t5+.DELTA.t6+.-
DELTA.t7+.DELTA.t8|/8.
[0110] (.DELTA.t1, .DELTA.t2), (.DELTA.t3, .DELTA.t4), (.DELTA.t5,
.DELTA.t6), and (.DELTA.t7, .DELTA.t8) change in opposite
directions. As a result, their average is always smaller than
|.DELTA.t1|. In effect, .DELTA.Tm.apprxeq.0. This indicates that
the ultrasonic wave transit time measurement accuracy is greatly
improved using the method of the present disclosure compared to the
method of prior art.
[0111] Noise in the received signal could cause threshold
triggering error, thus, cause transit time measurement error.
However, with multiple triggering mechanisms as illustrated in
FIGS. 7 and 10, the transit times obtained at each triggering point
are averaged to provide a result with reduced error. This not only
makes the whole system more robust, but also improves the
transit-time measurement accuracy, thus, the flow velocity
accuracy, significantly.
* * * * *