Ultrasonic flow velocity measuring method using phase difference measurements

Abstract

An ultrasonic flow velocity measuring method comprises a phase difference flow velocity measuring method irrelevant to the change of the sound velocity, if a flow velocity is measured based on a phase difference method without transmitting/receiving an ultrasonic pulse in a river, an open sluice channel or a pile having a large inner diameter, including amplitude-modulating at least one continuous ultrasonic sine waves of a carrier frequench ƒC into signals of an amplitude-modulated frequench ƒM, if at least one phase difference ΔφC transited and received in directions similar and contrary to the flow velocity is equal to nπ+aπ, obtaining nπ, using the signal of the amplitude-modulation ƒM; and measuring a part of aπ<π, obtaining the phase difference between the signals of the carrier frequench ƒC and obtaining the total phase differences ΔφC precisely.

Description

BACKGROUND OF THE INVENTION

The invention is related to providing a method of measuring a flow velocity using a phase difference of ultrasonic waves for calculating a flow rate of fluid in a large river or open sluice way channel and a flow rate of liquid or gas in a pipe having a large inner diameter.

PRIOR ARTS

The core portion of a recent well-known ultrasonic flow rate measuring system for a large open sluice way channel or a pipe having a large inner diameter is designed to measure a flow velocity of liquid or gas. The system is normally called “a flowmeter.”

Most of the flow rates measuring systems are supposed to measure the flow velocity based on a flow velocity measuring method utilizing an ultrasonic transit time difference.

As shown in FIG. 1 , a flow velocity measuring system using an ultrasonic transit time difference operates as follows: ultrasonic transducers 1 and 2 for transmitting/receiving an ultrasonic wave are mounted at an angle α to face each other. A switch circuit 3 functions to switch the ultrasonic transducers 1 and 2 in turns to the inputs of transmitting and receiving circuits. An example of a transmitting and receiving circuit is an ultrasonic pulse oscillator 4 and an ultrasonic receiving signal amplifier 5 . Next, a pulse shaping circuit 6 receives an amplified signal and shapes it into a pulse signal of a shorter period. A time interval measuring apparatus 7 measures transit times t 1 and t 2 at an interval distance L from the transmitting time till the receiving time. An arithmetic logic unit 8 computes a flow velocity based on expression (1).

That is to say, the transit time t 1 , which the ultrasonic pulse is transmitted from transducer 1 to transducer 2 (as shown in FIG. 1 ), is measured. On the contrary, the transit time t 2 , which the ultrasonic pulse is transmitted from the transducer 2 to the transducer 1 , is measured. These times measured are made as follows: $$ t 1 = L C + V ⁢   ⁢ cos ⁢   ⁢ α ;   ⁢ t 2 = L C - V ⁢   ⁢ cos ⁢   ⁢ α

Therefore, the transit time difference Δt that t 2 −t 1 can be presented as follows: $$ Δ ⁢   ⁢ t = 2 ⁢ L ⁢   ⁢ cos ⁢   ⁢ α ⁢   ⁢ V C 2 ( 1 )

Wherein, C is a sound velocity of liquid or gas, L is an interval between transducers 1 and 2 and V is an average flow velocity in the interval L.

The flow velocity V from the expression (1) is deduced as follows: $$ V = Δ ⁢   ⁢ tC 2 2 ⁢   ⁢ L ⁢   ⁢ cos ⁢   ⁢ α ( 2 )

It may be called “A Transit Time Difference Flow Velocity Measuring Method”, because the flow velocity V is proportional to the transit time difference Δt. It seems that the transit time difference flow velocity measuring method is related to the sound velocity, because there is an item C 2 , which is the square of the sound velocity, in the expression (2). The item C 2 of the sound velocity must be simultaneously measured at the time of the flow velocity measurement. The square of the sound velocity is represented as follows: $$ C 2 = L 2 t 1 · t 2

The sound velocity item C 2 is substituted into the expression (2) to make the final flow velocity measuring expression as follows: $$ V = L 2 2 ⁢ L ⁢   ⁢ cos ⁢   ⁢ α · t 2 - t 1 t 1 · t 2 = L 2 2 ⁢ d ⁢ t 2 - t 1 t 1 · t 2 ( 3 )

Then, the flow velocity is obtainable by measuring only the ultrasonic transit times t 2 and t 1 and computing the expression (3), because L 2 /2d=const.

Typical prior arts are disclosed in U.S. Pat. No. 5,531,124 granted on Jul. 2, 1996, Japanese Patent No. 2,676,321 granted on Jul. 25, 1998, Manual of Ultrasonic flow Measuring and Apparatus thereof and Ultrasonic Flowmeter related to Model UF-2000C manufactured by the Ultra flux Co.

The transit time difference flow velocity measuring method has a great advantage in that the flow velocity measuring is simply performed as illustrated in the expression (3), even though the sound velocity is seriously changed in fluid. That is, although the expression (3) seems like being related to the square of the sound velocity according to a deliberative method of the flow velocity measuring expression, it is not principally related to the flow velocity.

For example, the difference between the reciprocal numbers with respect to the transit times t 1 and t 2 is obtained as follows: $$ 1 t 1 - 1 t 2 = 2 ⁢ V ⁢   ⁢ cos ⁢   ⁢ α L ,

The items of the sound velocity C are offset to each other. Therefore, the flow velocity V is as follows: $$ V = L cos ⁢   ⁢ α ⁢ ( 1 t 1 - 1 t 2 ) = L 2 2 ⁢   ⁢ d ⁢ ( t 2 - t 1 t 1 · t 2 )

Wherein, d=L cos α.

As a result, the expression obtained is the same as the one (3).

It has a great advantage in that the transit time difference flow velocity measuring method has no relation with the change in the great range of the sound velocity C in fluid. But, the transit time difference flow velocity measuring method is limited to its use. For example, when the transit distance L is very small and/or the flow velocity V is very low, it is very difficult to measure the flow velocity, precisely. If L=0.05 m, V=0.1 m/s, α=45° and C≈1500 m/s, Δt≈3.14×10 −9 s.

If it is intended to measure a very little time difference within the error range of 1%, the time difference absolute measuring error should not exceed the range of 3×10 −11 s. Measuring the time difference based on such a method needs a relative complex time interval measuring apparatus. Also, an apparatus for catching moments of transmitting/receiving the ultrasonic pulses must be very stable and precise. As mentioned below, the transit time difference flow velocity measuring method causes many problems, when the gas flow velocity is measured in a pipe, or the horizontal flow velocity is measured in a channel or river.

In addition to the transit time difference flow velocity measuring method, an ultrasonic phase difference flow velocity measuring method is also well known. For example, there are Dutch Patent Laid-Open Publication No. DE19722140 on Nov. 12, 1997 and Japanese Patent Laid-Open Publication No. Hei 10-104039 published on Apr. 24, 1998, both of which are entitled as “A multi-channel flow rate measuring system.”

FIGS. 2A and 2B show a typical configuration of a phase difference flow velocity measuring system. Ultrasonic transducers 1 , 1 ′ and 2 , 2 ′ are positioned to face each other. A sine wave oscillator 9 generates a sine wave having a frequency ƒ. A phase shifter 10 adjusts the phase of received ultrasonic signals. An amplifier 11 amplifies the received signals from the phase shifter 10 and the transducer 1 ′. A phase difference discriminator 12 measures the phase difference between the received phase signals. When the sine wave oscillator 9 is operated, the transducers 2 and 2 ′ transmit ultrasonic waves at the same phase. At that time, the phase signals, which the receiving transducers 1 and 1 ′ receive are as follows:

φ 1 =2 πƒ·t 1 +φ 0 ; φ 2 =2 πƒt 2 +φ 0

Wherein, $$ t 1 = L C - V ⁢   ⁢ cos ⁢   ⁢ α :   ⁢ t 2 = L C + V ⁢   ⁢ cos ⁢   ⁢ α

φ 0 is an initial phase that the ultrasonic wave is firstly transmitted. Therefore, the phase difference Δφ between the received signals is as follows: $$ Δφ = φ 1 - φ 2 = 2 ⁢ π ⁢   ⁢ f ⁢   ⁢ Δ ⁢   ⁢ t = 2 ⁢ π ⁢   ⁢ f ⁢ 2 ⁢ LV ⁢   ⁢ cos ⁢   ⁢ α C 2 ( 4 )

Herein, the flow velocity is as follows: $$ V = Δφ ⁢   ⁢ C 2 4 ⁢ π ⁢   ⁢ fL ⁢   ⁢ cos ⁢   ⁢ α ( 5 )

The phase difference method has features in that the ultrasonic waves can be continuously transmitted and the phase difference Δφ is proportional to the frequency ƒ unlike the transit time difference method. Therefore, even if L and V are very small, when the ultrasonic frequency ƒ is selected at a higher one, the phase difference becomes larger, so that the phase difference measuring is conveniently and precisely done.

Also, if L is relatively larger, the damping factor is very small over the ultrasonic pulse, because the ultrasonic continuous waves are transmitted/received. Further, even though the amplitude of the received signal significantly pulsates, the received signal can be sufficiently amplified, because the receiving moment is not measured. And, an automatic gain control circuit can be used in the method. It means that there is not any problem in measuring the phase difference at all. Only, the phase difference method is preferably used under the condition that the sound velocity C is almost not changed or in the case that any other means measures the sound velocity C. For example, in order to measure the gas flow rate, the sound velocity of gas can be easily calculated under the condition that a pressure gauge and a thermometer are mounted in the pipe.

As mentioned above, the great advantage of the ultrasonic transit time difference method can be utilized even under the situation that the sound velocity in fluid is significantly changed. But, if the interval L between the transducers becomes larger, the following problems occur due to the transmitting/receiving of the ultrasonic pulse.

First, the ultrasonic pulse has a larger damping factor over the sine wave because of its sufficient harmonic wave components or overtones. If the ultrasonic transit distance L becomes larger, it is difficult to receive the transmitted ultrasonic wave and the received pulse becomes a bell form due to the serious damping problem.

For all that, it cannot help increasing the ultrasonic wave intensity that can be auxiliarily adjusted. If the intensity becomes higher, the cavity phenomenon occurs in a river, so that the ultrasonic wave is not transmitted. Especially, as the pulse frequency becomes lower in order to reduce the damping factor, the ultrasonic intensity also becomes lower, which causes the cavity phenomenon.

Second, the ultrasonic pulse is not damped only by the distance L in the procedure of being transmitted, but the amplitude of the ultrasonic wave seriously pulsates, by which the ultrasonic wave is diffused and reflected because of various sizes of eddy currents, the concentration change of floating particles, the temperature change of water, etc. in the open sluice way channel. It sometime happens that the ultrasonic wave is not received.

When the flow velocity of gas is measured, the damping factor of the ultrasonic pulse is larger than that in liquid. The serious damping and pulsation of the ultrasonic pulse cause many errors, when it is subjected to catch the moment that the ultrasonic pulse reaches. Thus, the flow velocity measuring error is increased.

Due to these reasons, the ultrasonic transit distance L is limited in that the ultrasonic pulse is transmitted/received and the flow velocity is measured based on a time difference method. Thus, it has great difficulty in measuring the flow velocity in the larger open sluiceway channel or river and a larger pipe.

If the phase difference method is used for measuring the flow velocity, its damping factor is decreased two or three times over that of the ultrasonic pulse, because the ultrasonic continuous waves (sine waves) are transmitted/received. Also, the phase difference method is not relevant to the amplitude pulsation of the received signals, because it is not related to catching the moment that the ultrasonic pulse reaches, but the phase difference between two sine waves is measured. Nevertheless, the phase difference method is limited to its use. If the phase difference Δφ between two sine waves is equal to mπ+β, a general phase difference measuring apparatus cannot detect m (1, 2, 3, . . . ). If the ultrasonic transit distance L or the flow velocity V is larger, Δφ becomes greater than π. For example, if it is intended to measure the flow rate of gas in the pipe having an inner diameter Φ of 300 mm, the cross-sectional average flow velocity V of gas is generally 10-30 m/s. Then, assuming that the sound velocity C is 400 m/s, the ultrasonic frequency ƒ is selected at 400 KHz in order to be beyond the frequency band of noises and an angle α is 45°, the changing width of the phase difference Δφ is as follows:

Δφ=9.42˜28.26rad≈(2π+0.998π)˜(8π+0.995π)

That is, Δφ>π.

If L=10 m, V=3 m/s, ƒ=200 KHz and C=1500 m/s in a relatively smaller open channel, the phase difference Δφ is as follows:

Δφ≈16.746rad=5π+0.33π>π.

Thus, the phase difference method cannot be used in measuring the flow velocity in the relatively smaller open channel. In other words, the transit time difference method has an advantage in being used under the situation that the sound velocity is changed to a larger range. But, it has disadvantages in that if the flow velocity-measuring interval L is larger, the ultrasonic pulse becomes unstable, because the ultrasonic pulse is greatly damped due to its own property during the transmitting/receiving.

The phase difference method has advantages in that the damping factor is relatively smaller and the received signal is easily processed, because the ultrasonic sine wave is transmitted/received. But, if the phase difference exceeds π radians by which the interval L and the flow velocity V is larger or the sound velocity is lower, it is not possible to measure the flow velocity based on the phase difference method. Also, the phase difference method has a disadvantage in that the sound velocity should be separately measured.

An object of the invention is to provide an ultrasonic flow velocity measuring method for measuring a flow velocity based on a phase difference method, smoothly, if a flow velocity measuring interval L is relatively larger, for example if a horizontal average flow velocity is measured in an open sluice way channel or river.

The other object of the invention is to provide an ultrasonic flow velocity measuring method for measuring the flow velocity based on a phase difference method, smoothly, if a flow velocity measuring interval L is relatively larger, for example if a gas flow velocity is measured in a pipe of a relatively larger inner diameter.

Another object of the invention is to provide an ultrasonic flow velocity measuring method for measuring the flow velocity based on a phase difference method, smoothly, if a gas or liquid flow velocity is measured in a pipe of a relatively larger inner diameter.

Still another object of the invention is to provide an ultrasonic flow velocity measuring method for measuring the flow velocity based on a phase difference method, smoothly, if a flow velocity is relatively larger and the sound velocity is relatively lower.

SUMMARY OF THE INVENTION

According to the invention, an ultrasonic flow velocity measuring method based on a phase difference method not depending upon a sound velocity, comprises: amplitude-modulating a continuous ultrasonic wave of a carrier frequench ƒ C into at least one signal of an amplitude-modulation frequench ƒ M , if a phase difference between the ultrasonic waves transited in directions similar and contrary to the flow velocity exceeds π radians beyond the measuring range of a general phase difference discriminator and becomes mπ+β, and transiting/receiving the amplitude-modulated signals in directions similar and contrary to the flow velocity; obtaining m using the signal of amplitude-modulation frequench ƒ M , measuring a part of aπ<π, obtaining the phase difference between the signals of the carrier frequench ƒ C and obtaining the total phase differences Δφ C , precisely, thereby enabling the very accurate measurement of the phase difference between the ultrasonic waves.

BRIEF DESCRIPTION OF THE DRAWINGS

The invention now will be described in detail with reference to the accompanying drawings, in which:

FIG. 1 is a schematic block diagram illustrating an ultrasonic transit time difference flow velocity measuring system according to a prior art;

FIGS. 2A and 2B are schematic block diagrams illustrating an ultrasonic phase difference flow velocity measuring system according to a prior art;

FIG. 3 is a timing chart illustrating the processing of an ultrasonic transit time difference flow velocity measuring method according to the invention;

FIG. 4 is a schematic block diagram illustrating an ultrasonic transit time difference flow velocity measuring system according to the invention;

FIG. 5 is a schematic block diagram illustrating an ultrasonic phase difference flow velocity measuring system according to the invention; and,

FIG. 6 is a schematic block diagram illustrating an ultrasonic phase difference flow velocity measuring system according to another embodiment of the invention.

DETAILED DESCRIPTION OF THE INVENTION

Firstly, an ultrasonic transit time difference flow velocity measuring method of the invention will be explained in detail referring to the accompanying drawings:

FIG. 3 is a timing chart or sequence illustrating a flow velocity measuring method. It is known that a carrier frequench ƒ C is generally selected by considering a noise frequency band caused in the fluid flowing, the security with respect to the directivity diagram of an ultrasonic transducer, and an ultrasonic damping factor in fluid, etc.

When a flow velocity is measured, an ultrasonic wave of a selected carrier frequench ƒ C ( FIG. 3 , VI) is amplitude-modulated into a signal of an amplitude-modulation frequench ƒ M ( FIG. 3 , I) lower than the carrier frequench ƒ C for a period of τ 2 ( FIG. 3 , V) and then transited in a direction similar or contrary to the flow velocity. And, if a time point to be amplitude-modulated is considered as a starting point, a time is measured from the starting point till a designated falling edge of a signal of the amplitude-modulation frequench ƒ M , while the amplitude-modulated ultrasonic wave is transited/received through a constant interval distance L and the received signal is demodulated. The time is defined into ultrasonic transit times t 1 and t 2 propagated in a direction similar or contrary to the flow velocity. In other words, the amplitude-modulated ultrasonic wave acts as a mark signal for measuring the transit time of the ultrasonic wave. And, because the ultrasonic wave is a kind of a sine wave that is continuously transited and amplitude-modulated for a constant time interval to measure the flow velocity, the ultrasonic frequency band is ƒ C ±ƒ M , which is significantly narrower than that of the shorter ultrasonic pulse, so its damping factor becomes smaller. And, even if the damping factor is changed too much, the processing of the receiving signal is easy and it doesn't make an effect on the measuring of the transit time.

But, when the signal of the ultrasonic carrier frequench ƒ C is amplitude-modulated into the signal of the amplitude-modulation frequench ƒ M , it should be amplitude-modulated at the same phase as that of the amplitude-modulation frequency ƒ M , for example a zero phase as shown in FIG. 3 , V. When the amplitude-modulated voltage is applied to the ultrasonic transducer, the ultrasonic wave of a type equal to the voltage applied is not transited, but a first half-period of a modulated ultrasonic wave is distorted in a shape. Furthermore, a signal obtained by receiving/demodulating the amplitude-modulated ultrasonic wave is not corresponding to the shape of the signal of the amplitude-modulation frequench ƒ M . Considering these points, the amplitude-modulated signal applied to the ultrasonic transducer is inputted into a demodulator to be demodulated, the signal of the amplitude-modulation frequench ƒ M is detected from the demodulated signal, and the moment that the first period of the modulation signal passes over the zero potential is caught using a zero-crossing discriminating circuit. Herein, the moment caught is considered as a starting point for measuring the ultrasonic transit time as shown in FIG. 3 , VII and VIII.

Similarly, the amplitude-modulation signal received is also demodulated by the demodulator as mentioned above, the signal of the amplitude-modulation frequench ƒ M is detected from the demodulated signal, and then the moment that the first period of the modulation signal passes over the zero-crossing point is caught to function as a stop signal of the time interval as shown in FIG. 3 , X and XI.

As described above, the ultrasonic transit time measuring accuracy can be significantly enhanced, by which only one demodulator demodulates the transiting/receiving signals and the moments that the first period of the demodulation signal pass over the zero-cross point are used as the time interval measuring start and stop signals.

As shown in FIG. 3 , VIII and XI, it is irrelevant to use the moments that one and a half period of the signal having the amplitude-modulation frequench ƒ M , not the first half period, passes over the zero-crossing point as the time interval measuring start and stop signals. Of course, the delay time is generated at the demodulator, the amplifier, the zero-crossing circuit etc., but it is not necessary to compensate for the delayed time, because the system generates the same delayed time, whenever the flow velocity is measured.

And, the signal of the amplitude-modulation frequench ƒ M should catch up with the following conditions:

The first condition is when the signal of the amplitude-modulation frequench ƒ M is significantly higher than a damping pulsation frequench ƒ p , for example ƒ M >>ƒ p . The ultrasonic wave has the damping factor changed due to many factors during the transiting in fluid. What the damping factor is changed is to make the ultrasonic wave amplitude-modulated. Thus, the amplitude-modulation frequench ƒ M should be higher than the damping pulsation frequench ƒ p in that the damping factor pulsates, which is not a noise frequency generated in fluid. The damping pulsation frequench ƒ p is not high and does not exceed 100 Hz, generally.

The second condition is when a carrier period should be contained more than 20th times in an amplitude-modulation period, for example ƒ M >>ƒ C /20. The condition concerns the amplitude-modulation of the carrier frequench ƒ C , in which the phase of the carrier frequench ƒ C at the start point of the amplitude-modulation is not always uniform, even if the signal of the carrier frequench ƒ C is amplitude-modulated at a zero-crossing point as shown in FIG. 3 , V. For it, the amplitude-modulated ultrasonic wave raises the transient phenomena and distorts the waveform in the interval of a first one-fourth period of the signal of the amplitude-modulation frequench ƒ M . In order to prevent a wave-distorted portion from exceeding one-fourth period, the carrier frequench ƒ C should include at least five periods in the first one-fourth period of the signal of the amplitude-modulation frequench ƒ M . Thus, the signal of the carrier frequench ƒ C should exist over 20 (=4×5) in one period of the amplitude-modulation signal ƒ M . In addition, it is preferable that the carrier frequench ƒ C is higher than the amplitude-modulation frequench ƒ M in order to filter the signal of the amplitude-modulation frequench ƒ M from the pulsating signal of the carrier frequency ƒ C .

The third condition is when the continuous time of amplitude-modulated signals desirably exceeds at least five periods of the signal of the amplitude-modulation frequench ƒ M (5/ƒ M ), if the amplitude-modulated signal is demodulated to detect the signal of the amplitude-modulation frequench ƒ M . If the amplitude-modulated signal having the amplitude-modulation period to be repeated two or three times is demodulated, the outputting signal of the demodulator is distorted.

The fourth condition is when the ultrasonic wave is transited/received in turns in a direction similar or contrary to the flow velocity, it is desirable that the continuous time of the amplitude-modulated ultrasonic wave does not exceed one-second of the ultrasonic transit time. The example is as follows: $$ 5 / f M ≤ L 2 ⁢ ( C + υ ) ⁢   ⁢ or ⁢   ⁢ f M ≥ 10 ⁢ ( C + υ ) L

As described above, the signal of the amplitude-modulation frequench ƒ M satisfied with four conditions is selected by the following expression: $$ f P ⪡ 10 ⁢   ⁢ ( C max + υ max L ) ≤ f M ≤ 0.05 ⁢ f C ( 6 )

Wherein, C max is a maximum sound velocity that can be expected in fluid, and υ max(= V max cos α) is a maximum flow velocity measuring value.

It is preferable in selecting the amplitude-modulation frequench ƒ M satisfying with the expression (6) that the relatively lower frequency is selected if possible, because the transient phenomena happens, when the voltage applied to the ultrasonic transducer is rapidly changed. It is desirable that the amplitude-modulation percentage m does not exceed 50%. According to the experiments, the amplitude-modulation percentage m of 25-30% is very reasonable. The ultrasonic damping factor pulsates at the lower frequench ƒ P , the changing ratio of which is generally about 50%. Thus, if m>50%, it is likely that the amplitude-modulated wave will be cut off. For example, assuming that L=10 m, α=45°, C max =1500 m/s, ƒ C =500 KHz, ƒ P <<1507<ƒ M 25×10 3 Hz. Thus, ƒ M can be selected in the range of 10 to 20 KHz. Considering the transient phenomena of the ultrasonic wave, it is not necessary to select the signal of the higher amplitude-modulation frequench ƒ M that is equal to 25 Hz.

FIG. 4 is a schematic block diagram illustrating the configuration of a system according to one embodiment of the invention for realizing a method of measuring a flow velocity as described above.

Ultrasonic transducers 1 and 2 are connected to a transducer switching circuit 3 to be switched into the transiting or receiving state. An outputting amplifier 18 excites the ultrasonic transducer 1 or 2 . A receiving amplifier 19 amplifies the signals from the ultrasonic transducer 1 or 2 , which is a narrow band amplifier that has the function of the automatic gain control (AGC) and amplifies only the frequency band of an amplitude-modulation signal.

An amplitude-modulator 17 amplitude-modulates a signal of an ultrasonic carrier frequench ƒ C . A carrier oscillator 13 generates the signal of an ultrasonic carrier frequench ƒ C . A modulating oscillator 14 generates a modulation signal of an amplitude-modulation frequench ƒ M lower than the carrier frequench ƒ C . Herein, both of the carrier oscillator 13 and the modulating oscillator 14 are sine wave oscillators. A demodulator 20 demodulates the amplitude-modulated signal to detect the amplitude modulation frequench ƒ M . A narrow band amplifier 21 amplifies the signal of the amplitude-modulation frequench ƒ M . A zero-crossing circuit 22 outputs a square pulse, when a first period of the outputting signal of the amplitude modulation frequench ƒ M from the narrow band amplifier 21 passes over the zero point. A time interval measuring apparatus 7 measures the time interval between two pulses. An arithmetic logic unit 8 computes a flow velocity based on an ultrasonic transit time difference flow velocity measuring expression. A switch circuit 23 permits the outputting signal of the amplitude-modulation frequench ƒ M from the modulating oscillator 14 to be passed therethrough in a given time interval. A zero-crossing circuit 15 generates a square pulse, when the first period of the signal of the amplitude-modulation frequency ƒ M passes over the zero-point. A monostable multivibrator 16 is operated by the zero-crossing circuit 15 to generate a pulse of a given length.

A switch circuit 24 is switched by the pulse of the monostable multivibrator 16 to allow the outputting signal of the modulation oscillator 14 to be applied to the amplitude-modulator 17 . A switch circuit 25 allows an ultrasonic modulated output to be applied to the demodulator 20 and then is switched to permit the outputting signal from the receiving amplifier 19 to be inputted into the amplitude-modulator 20 . A voltage attenuator 27 adjusts the outputting voltage of the outputting amplifier 18 . A switch circuit controller 26 controls the switch circuits 3 , 23 and 25 .

The operation of the ultrasonic flow velocity measuring system as shown in FIG. 4 will be explained in detail below with reference to FIG. 3 .

The carrier oscillator 13 and the modulation oscillator 14 are firstly oscillated to generate sine waves of the ultrasonic carrier frequench ƒ C and the amplitude modulation frequench ƒ M , respectively, as shown in FIG. 3 , VI and I. When a flow velocity measuring moment reaches, the switch circuit controller 26 applies a square pulse of a length τ 1 (referring to FIG. 3 , II) to the switch circuit 23 . The switch circuit 23 permits the signal of the amplitude modulation frequench ƒ M from the modulation oscillator 14 to be inputted to the zero-crossing circuit 15 . Then, as the operation potential level of the zero-crossing circuit 15 is set at a low level “−”(negative level), the zero-crossing circuit 15 generates a square pulse (referring to FIG. 3 , III), when the first one-half period of the outputting signal from the modulation oscillator 14 passes through the zero point (U=0). The square pulse is inputted into the monostable multivibrator 16 , and the monostable multivibrator 16 generates a square pulse of the length τ 2 ( FIG. 3 , IV). The square pulse of τ 2 switches the switch circuit 24 in order to permit the signal of the amplitude-modulation frequench ƒ M from the modulation oscillator 14 to be inputted to the amplitude-modulator 17 . Thus, the signal of the ultrasonic carrier frequench ƒ C is amplitude-modulated for the time of τ 2 as shown in FIG. 3 , VI. Like it, the ultrasonic carrier frequench ƒ C is always supposed to be amplitude-modulated into the same phase of the amplitude-modulation frequench ƒ M .

The amplitude-modulated signal from the amplitude-modulator 17 is amplified by the outputting amplifier 18 and then applied to the ultrasonic transducer 1 . The ultrasonic transducer 1 transits the amplitude-modulated ultrasonic wave through fluid to the transducer 2 .

At the same time, the outputting signal of the outputting amplifier 18 is inputted through the voltage attenuator 27 and the switch circuit 25 to the demodulator 20 to detect the modulation signal ƒ M ( FIG. 3 , VII). The narrow band amplifier 21 amplifies the amplitude-modulation signal demodulated by the demodulator 20 and applies the amplified signal to the zero-crossing circuit 22 . The zero-crossing circuit 22 generates a shorter square pulse ( FIG. 3 , VIII) at the moment that the first one and a half period “−”of the signal of the modulation frequench ƒ M passes through the zero-point. The shorter square pulse is inputted into the time interval measuring apparatus 7 to function as a time measuring start signal.

Thereafter, the switch circuit 25 cuts off the input to the attenuator 27 and forces the outputting signal from the receiving amplifier 19 to be applied to the demodulator 20 . In other words, the amplitude-modulated ultrasonic wave which is transited from the transducer 1 through an interval distance L is received by the transducer 2 and amplified by the receiving amplifier 19 . The outputting signal ( FIG. 3 , IX) from the receiving amplifier 19 is applied through the demodulator 20 and the amplifier 21 to the zero-crossing circuit 22 . The zero-crossing circuit 22 generates the shorter square pulse ( FIG. 3 , XI) and applies it to the time interval measuring apparatus 7 to function as a time measuring stop signal.

Therefore, the time interval measuring apparatus 7 measures the time interval t 1 between the first and second square pulses from the zero-crossing circuit 22 . After finishing the measurement of the time interval t 1 , the transducer switch circuit 3 is switched to connect the transducer 2 to the outputting amplifier 18 . Then, the switch circuit 25 is connected to the attenuator 27 , and the switch circuit 23 is switched, again. And, the next operations are repeated in the same manner as the measuring ones of the time interval t 1 . Therefore, a time t 2 is measured until the amplitude-modulated ultrasonic wave is transited from the transducer 2 and received by the transducer 1 .

The time intervals t 1 and t 2 are inputted into the flow velocity arithmetic logic unit 8 to compute the flow velocity based on the flow velocity measuring expression (3). The flow velocity arithmetic logic unit 8 outputs a signal corresponding to the flow velocity V. The outputting signal of the flow velocity V is provided to a flow rate measuring arithmetic logic unit (not shown), if the system is a flow rate measuring system.

Herein, important things are as follows: it has features in that in order to measure the time intervals t 1 and t 2 , the amplitude-modulated outputting signal inputted into the transducer 1 (or 2 ), the signal received by the transducer 2 (or 1 ) pass through one demodulator and the zero-crossing circuit, and the start and stop pulse signals inputted into the time interval measuring apparatus 7 are shaped into a square pulse.

The well-known phase difference flow velocity measuring expression (5) depends on the square of the sound velocity (C 2 ). In the expression (5), Δφ also is a phase difference between the ultrasonic waves transited in the directions similar and contrary to the flow velocity. Except for the flow velocity measuring method of the expression (5), a phase difference flow velocity measuring expression that does not depend on the sound velocity C can be derived.

The phase difference Δψ 1 between the ultrasonic transiting wave and the received wave next to be transited toward the flow velocity direction and the phase difference Δψ 2 between the ultrasonic transiting signal and the received signal next to be transited in a direction contrary to the flow velocity are as follows: $$ Δψ 1 = 2 ⁢ π ⁢   ⁢ f ⁢ L C + υ ⁢   ⁢ Δ ⁢   ⁢ ψ 2 = 2 ⁢ π ⁢   ⁢ f ⁢ L C - υ ( 7 )

Wherein, υ=V cos α, and L is an interval distance between ultrasonic transducers.

The difference between the reciprocal numbers of the phase differences Δψ 1 and Δψ 2 is as follows: $$ 1 Δψ 1 - 1 Δψ 2 = 2 ⁢ V ⁢   ⁢ cos ⁢   ⁢ α 2 ⁢ π ⁢   ⁢ fL ( 8 )

Wherein, V is as follows: $$ V = π ⁢   ⁢ fL cos ⁢   ⁢ α ⁢ ( 1 Δψ 1 - 1 Δψ 2 ) ( 9 )

The flow velocity measuring method is highly worth being used, because it is not necessary to measure the sound velocity, separately, even under the condition that the sound velocity is significantly changed. But, only if the measuring error of the phase differences Δψ 1 and Δψ 2 are very small enough to be ignored, the flow velocity can be measured based on the expression (9).

For example, Δψ 1 =2.0 radians, and Δψ 2 =2.2 radians. Assuming that the phase differences are measured in the range of the error of 0.5%, the measured phase differences are as follows: $$ Δψ 1 ′ = 2.0 ⁢ ( 1 + 1.005 ) = 2.01 $$ Δψ 2 ′ = 2.2 ⁢ ( 1 - 0.005 ) = 2.189

As a result, $$ 1 Δ ⁢   ⁢ ψ 1 ′ - 1 Δψ 2 ′ = 0.040682835

But, the actual value is as follows: $$ 1 2.0 - 1 2.2 = 0.0454545

Therefore, the error is as follows: $$ 0.0406828 - 0.04545 0.04545 = - 0.105 = 10.5 ⁢ %

That is to say, the phase difference was measured in range of the error of 0.5%, but the error between the differences of the reciprocal numbers with respect to the phase differences was increased more than 20 times. Thus, the flow velocity measuring error was more than 10%.

In order to realize the phase difference of the flow velocity measuring method not depending on the sound velocity C, the phase difference must be very precisely measured.

It appears that the following problem from the expression (7) happens. As the interval distance L is increased, the sound velocity C is lowered, and the ultrasonic frequency is increased, the phase difference Δψ 1,2 is too much increased more than π. Of course, if L, C and υ are given, the ultrasonic frequency ƒ, which enables the phase difference Δψ not to exceed the measuring range π of a general phase difference discriminator, can be selected, but it must be far too much higher than a noise frequency band generated in fluid.

For example, assuming that the inner diameter D of a natural gas pipe is equal to 0.3 m, C 420 m/s, V=30 m/s, α=45° and L=0.425 m, the ultrasonic frequency ƒ that does not exceed the phase difference π is as follows: $$ f ≤ C + V ⁢   ⁢ cos ⁢   ⁢ α 2 ⁢ π ⁢   ⁢ L = 420 + 30 · cos ⁢   ⁢ 45 ° 2 ⁢ π · 0.424 = 165.6 ⁢   ⁢ HZ

Such like a frequency band is included in a noise frequency one. Furthermore, it makes it impossible to manufacture a compact transducer that transits the sound wave of 165 Hz.

In order to be escaped from the noise band, if the ultrasonic carrier frequency ƒ C is selected to be 40 KHz, the phase difference in said examples is as follows: $$ Δ ⁢   ⁢ ψ 1 = 2 ⁢   ⁢ π · 4 · 10 4 · 0.424 420 + 30 ⁢   ⁢ cos ⁢   ⁢ α = 241.522 ⁢   ⁢ … ⁢   ⁢ rad > 76 ⁢   ⁢ π + β

Herein, the general phase difference discriminator cannot measure 76π.

In order to resolve these problems, the invention selects an ultrasonic carrier frequench ƒ C escaped far away from the noise band as a carrier, amplitude-modulates it into an amplitude-modulation frequench ƒ M lower than the ultrasonic carrier frequency ƒ C , transits it in the directions similar and contrary to the flow velocity and measures the phase differences between the transiting signal and the received signal as follows:

First, the amplitude-modulation frequench ƒ M is selected so that the phase differences Δψ M1 and Δψ M2 between the transiting wave of an amplitude-modulated signal and a signal that is received and demodulated next to be transited in the direction contrary to the flow velocity satisfy with the following conditions: $$ Δψ M1 = 2 ⁢ π ⁢   ⁢ f M ⁢ L C max + υ max = n ⁢   ⁢ π + b ⁢   ⁢ π ⁢ ⁢ Δψ M2 = 2 ⁢ π ⁢   ⁢ f M ⁢ L C min - υ max = n ⁢   ⁢ π + a ⁢   ⁢ π ( 10 )

Wherein, n=const (1, 2, 3, . . . ); a<1.0, b<1.0, C max and C min are maximum and minimum sound velocities in fluid and υ max =V max cos α, which is a maximum flow velocity measuring range.

In this case, as nπ is previously known, the phase differences Δψ M1 and Δψ M2 are supposed to be measured, only if aπ and bπ are measured and next nπ is added thereto. Herein, aπ is a maximum measuring limit and bπ is a lowest measuring limit. Because it is unstable, if a=1 and b=0, it is desirable that a is selected to be 0.95, and b is selected to be 0.2.

The n that satisfies with expression (10) is as follows:

The relative expression from the expression (10) is given as follows: $$ n + b n + a = C min - υ max C max + υ max

Wherein, n is as follows: $$ n = a ⁡ ( C min - υ max ) - b ⁡ ( C max + υ max ) C max - C min + 2 ⁢ υ max ( 11 )

The amplitude-modulation frequench ƒ M based on such like obtained n is as follows: $$ f M = n + a 2 ⁢ L ⁢ ( C min - υ max ) ⁢ ⁢ or , &IndentingNewLine; ⁢ f M = n + b 2 ⁢ L ⁢ ( C max + υ max ) ( 12 )

Therefore, the signals of the carrier frequench ƒ C are amplitude-modulated into the selected amplitude-modulation frequench ƒ M , and the amplitude-modulated signal is transited/received. At that time, if the phase differences Δψ M1 and Δψ M2 between the signals of amplitude-modulation frequench ƒ M transited and received are measured in the range of a constant error δ M , the calculation results of the phase differences Δψ M1 and Δψ M2 are as follows: $$ Δψ M1 ′ = n ⁢   ⁢ π + b ⁢   ⁢ π ⁡ ( 1 ± δ M ) ⁢ ⁢ Δψ M2 ′ = n ⁢   ⁢ π + a ⁢   ⁢ π ⁡ ( 1 ± δ M ) ( 13 )

Wherein, aπ=Δψ MM1 and bπ=Δψ MM2 , which are phase differences that the phase difference discriminator can measure. Multiplying the phase difference by ƒ C /πƒ M becomes a value that divides the phase differences Δψ C1 and Δψ C2 between the carrier frequencies into π. $$ Δ ⁢   ⁢ ψ M1 ′ × f C π ⁢   ⁢ f M = m 1 + β ⁢ ⁢ Δ ⁢   ⁢ ψ M2 ′ × f C π ⁢   ⁢ f M = m 2 + γ ( 14 )

Wherein, β<1.0, γ<1.0 and m 1 and m 2 are integers (1, 2, 3, 4, . . . ).

If the phase differences Δψ C1 and Δψ C2 are measured as described above, it is noted that m 1 π+βπ and m 2 π+γπ are obtainable.

The values that the discriminator measures the phase difference between the carrier frequencies are as follows: $$ Δψ CM1 ′ = β ⁢   ⁢ π ⁡ ( 1 ± δ c ) ⁢ ⁢ Δψ CM2 ′ = γπ ⁡ ( 1 ± δ c ) ( 15 )

If the m 1 π and m 2 π are added to the measured values, the difference between a phase upon the transiting of the carrier frequency wave and a phase of the received signal next to be transited in the direction contrary to the flow velocity are as follows: $$ Δψ C1 ′ = m 1 ⁢ π + βπ ⁡ ( 1 ± δ c ) ⁢ ⁢ Δψ C2 ′ = m 2 ⁢ π + γ ⁢   ⁢ π ⁡ ( 1 ± δ c ) ( 16 )

The phase differences Δψ C1 ′ and Δψ C2 ′ obtained like above are substituted into the flow velocity measuring expression to compute the flow velocity as follows: $$ V ′ = π ⁢   ⁢ f C ⁢ L cos ⁢   ⁢ α ⁢ ( 1 Δ ⁢   ⁢ ψ C1 ′ - 1 Δψ C2 ′ ) ( 17 )

If the phase difference of the carrier frequencies is measured according to the above method, the measuring error is reduced tens or hundreds times over the error δ C of the phase difference discriminator. $$ δ Δψ 1 = Δψ C1 ′ - Δψ C1 Δψ C1 = βπδ C m 1 ⁢ π + βπ = ± δ C 1 + m 1 β ⁢ ⁢ δ Δψ C2 = Δψ C2 ′ - Δψ C2 Δψ C2 = ± δ C 1 + m 2 γ ( 18 )

Wherein, m 1 and m 2 >>1, β and γ<1.0. Thus, δ ΔψC1 and δ ΔψC2 are too much smaller than δ C .

As described above, according to the invention, because the phase difference is accurately measured when the ultrasonic wave is transited and received, the flow velocity can be measured based on the phase difference flow velocity measuring expression not depending upon the sound velocity. Also, even if L and V are larger, C is lower and the phase difference between the ultrasonic waves exceeds far away from π radians, the flow velocity can be easily measured.

For example, when the flow velocity of natural gas flowing in a pipe having an inner diameter of 300 mm is measured, it is assumed that C min =420 m/s, C max =650 m/s, L=0.425 m, V max cos α=30 m/s and the ultrasonic carrier frequench ƒ C is selected at 40 KHz by considering the noise in the pipe. Assuming that as the measuring range of the phase difference discriminator is 0-π, bπ is selected as 0.2π, for example b=0.2, when the phase difference becomes minimum in the range, and aπ is selected as 0.95π, for example a=0.95, when the phase difference becomes maximum in the range, the amplitude-modulation frequench ƒ M is as follows: $$ n = 0.95 ⁢ ( 420 - 30 ) - 0.2 ⁢ ( 450 + 30 ) 450 - 420 + 2 · 30 = 30.5

Assuming that n is selected at 3 and stored at the memory of the system, $$ f M ≤ 3.05 + 0.95 2 · 0.424 ⁢ ( 420 - 30 ) = 1839.62 ⁢   ⁢ HZ

Assuming that ƒ M is selected at 1830 Hz, during the transiting of the ultrasonic wave amplitude-modulated into the amplitude-modulation frequench ƒ M of 1830 Hz in the directions similar and contrary to the flow velocity, the received signal is demodulated to detect the amplitude-modulation frequench ƒ M . Then, if the phase difference between the phase of the amplitude-modulation frequench ƒ M of the transiting signal and the receiving signal phase is measured, the results are as follows:

If the flow velocity V cos α is equal to 20 m/s and C is equal to 450 m/s, $$ Δψ M1 = 2 ⁢ π ⁢   ⁢ f M ⁢ L C + υ = 2 ⁢ π ⁢   ⁢ 1830 ⁢ 0.424 450 + 20 = 10.372877 ⁢   ⁢ ⋯ = 3 ⁢ π + 0.30178 ⁢ π ⁢   ⁢ ( n = 3 ) $$ Δψ M2 = 2 ⁢ π ⁢   ⁢ f M ⁢ L C - υ = 3 ⁢ π + 0.60893 ⁢ π ⁢   ⁢ ( n = 3 )

Herein, the phase differences that the discriminator can measure are 0.30178π and 0.60893π. If the measuring error of the phase differences is performed in the range of ±1%, the computed phase difference is as follows: $$ Δψ M1 ′ = 3 ⁢ π + 0.30178 ⁢ π ⁡ ( 1 + 0.01 ) = 10.382328 ⁢   ⁢ rad $$ Δψ M2 ′ = 3 ⁢ π + 0.60893 ⁢ π ⁡ ( 1 - 0.01 ) = 11.31865 ⁢   ⁢ rad

Next procedure is as follows: $$ Δψ M1 ′ · f C π ⁢   ⁢ f M = 10.3823 · 40 · 10 3 π1830 = 72.235819 ⁢   ⁢ ⋯

Herein, m 1 (=72) is stored at the memory of the system. $$ Δψ M2 ′ · f C π ⁢   ⁢ f M = 11.31865 ⁢ 40 · 10 3 π1830 = 78.75056 ⁢   ⁢ ⋯

Herein, m 2 (=78) is stored at the memory of the system.

The actual phase difference between the carriers is as follows: $$ Δψ C1 = 2 ⁢ π ⁢   ⁢ f C ⁢ L C + υ = 226.7294102 = 72.17021276 ⁢   ⁢ π

Wherein, it is noted that m 1 (=72) is coincident with the stored value and the phase difference Δψ CM1 between the carriers that can be directly measured is equal to 0.17021276. $$ Δψ C2 = 2 ⁢ π ⁢   ⁢ f C ⁢ L C - υ = 247.8205182 = 78.88372094 ⁢   ⁢ π

Wherein, m 2 (=78) is coincided with the stored value, and the phase difference Δψ CM2 between the carriers is equal to 0.88372094.

If the phase differences Δψ CM1 and Δψ CM2 are measured in the range of the error of ±1%,

Δψ CM1 ′=0.54rad, Δψ CM2 ′=2.748rad

The calculating results of the phase difference Δψ C1 and Δψ C2 are as follows: $$ Δ ⁢   ⁢ ψ C1 ′ = 72 ⁢   ⁢ π + 0.54 = 226.73467 ⁢   ⁢ rad $$ Δ ⁢   ⁢ ψ C2 ′ = 78 ⁢   ⁢ π + 2.748 = 247.7922 ⁢   ⁢ rad

These phase differences are substituted into the flow velocity measuring expression to compute the flow velocity as follows: $$ V ′ ⁢ cos ⁢   ⁢ α = π ⁢   ⁢ f C ⁢ L ⁡ ( 1 Δψ C1 ′ - 1 Δψ C2 ′ ) = π ⁢   ⁢ 40 · 10 3 · 0.424 ⁢ ( 10 - 3 0.0226 ⁢   ⁢ ⋯ - 10 3 0.24779 ⁢   ⁢ ⋯ ) = 19.97 ⁢   ⁢ m ⁢ / ⁢ s

The first flow velocity V cos α is equal to 20 m/s, but the actual measured flow velocity becomes 19.95 m/s. Thus, the measuring error becomes about ±0.15%. That is to say, the phase differences are measured two times in the range of ±1%. As a result, the flow velocity measuring error is reduced by 0.15%.

Such like error reduced reason is why the measuring errors of the phase difference Δψ C1 and Δψ C2 are significantly decreased. $$ δ ψ C1 = Δψ C1 ′ - Δψ C1 Δψ C1 = 226.73467 - 226.72941 226.72941 = 0.00002323 = 0.0023 ⁢ %

If the phase difference Δψ CM1 is measured at δ C (=1%), the measuring error Δψ C1 is reduced m 1 /β (=72/0.1702≈423) times (referring to the expression 18). It is assumed that the phase differences Δψ MM1 , Δψ MM2 and Δψ CM1 , Δψ CM2 are measured at the error of ±1% from the above example, but it is normal that the phase difference is measured at the error of ±0.5%.

As described above, according to the invention, the flow velocity of gas, in which the flow velocity is high and the sound velocity is low, can be accurately measured based on the phase difference method irrelevant to the sound velocity change in a pipe of a larger inner diameter.

In FIG. 5 , a schematic block diagram illustrating the configuration of a system for realizing a method of measuring a flow velocity based on the phase difference method is shown as one embodiment of the invention.

Ultrasonic transducers 1 and 1 ′ are an ultrasonic receiving transducer to receive an ultrasonic wave, and ultrasonic transducer 2 is an ultrasonic transiting transducer to transit ultrasonic waves at a wider directivity angle. A carrier oscillator 13 and a modulating wave oscillator 14 generate a signal of an ultrasonic carrier frequench ƒ C and a signal of an amplitude-modulation frequench ƒ M , respectively. An amplitude-modulator 17 amplitude-modulates an ultrasonic carrier frequench ƒ C . An outputting amplifier 18 excites the ultrasonic transducer 2 . Receiving amplifiers 19 and 19 ′ amplify the signals from the ultrasonic transducers 1 and 1 ′, respectively.

Demodulators 20 and 20 ′ demodulate the amplitude-modulated signal to detect the signal of the modulation frequench ƒ M . Narrow band amplifiers 21 and 21 ′ amplify the signals outputted from the demodulator 20 and 20 ′. Phase difference discriminators 28 and 28 ′ detect the phase differences Δψ MM1 and Δψ MM2 between the signals of the amplitude-modulation frequench ƒ M . Phase difference discriminators 31 , 31 ′ detect the phase differences Δψ CM1 and Δψ CM2 between the carriers of the carrier frequency ƒ C . Amplifier-limiters 30 and 30 ′ amplify the amplitude-modulated signals and limit them to a predetermined level. Phase shifters 29 and 29 ′ function to force the output of the phase difference discriminators 28 and 28 ′ to be adjusted to zero, if the flow velocity V is zero. An arithmetic logic unit 32 computes the phase differences Δψ C1 and Δψ C2 between the carriers of the carrier frequench ƒ C and then the flow velocity according to the invention.

The ultrasonic flow velocity measuring system is operated as follows:

The amplitude-modulator 17 amplitude-modulates the signal of the carrier frequench ƒ C generated by the carrier oscillator 13 into the modulation frequench ƒ M generated by the modulation oscillator 14 . The amplifier 18 amplifies the amplitude-modulated signal and supplies it to the transiting ultrasonic transducer 2 . If the transducer 2 transits the amplitude-modulated signal in the directions similar and contrary to the flow velocity, the receiving transducer 1 receives the signal transited in the directions similar and contrary to the flow velocity V and converts it into electrical signals. The outputting signal from the receiving transducer 1 is amplified by the receiving amplifier 19 for amplifying the frequency band of ƒ C ±ƒ M and inputted into the demodulator 20 . The demodulator 20 generates the signal of the amplitude-modulation frequench ƒ M at its output. The signal is inputted through the phase shifter 29 into the narrow band amplifier 21 . The narrow band amplifier 21 again filters the amplitude-modulated signal and applies it to the phase difference discriminator 28 of a lower amplitude-modulation frequench ƒ M . The discriminator 28 detects the signal corresponding to the phase difference of Δψ MM2 that is smaller than π and inputs its outputting signal into the arithmetic logic unit 32 thereby to compute the phase difference and the flow velocity.

The ultrasonic wave transited in the flow velocity direction is received by the receiving transducer 1 ′, and the phase difference Δψ MM1 is detected through a receiving amplifier 19 ′, a demodulator 20 ′, a narrow band amplifier 21 ′ and the discriminator 28 ′ as mentioned above. At the same time, the outputting signal from the receiving amplifier 19 ′ is amplified by the amplifier-limiter 30 ′ and inputted into the phase difference discriminator 31 ′. The phase difference discriminator 31 ′ generates the signals corresponding to the phase differences Δψ CM1 and Δψ CM2 and inputs them to the arithmetic logic unit 32 .

The arithmetic logic unit 32 is supposed to force the integers of n, ƒ M , ƒ C , L, and cos α to be inputted thereinto in advance and obtains m 1 and m 2 according to the expression (13), calculates the phase differences Δψ C1 and Δψ C2 of the carriers according to the expression (15) and computes the flow velocity V according to the expression (16). Such like obtained flow velocity may be used in computing the flow rate, if it is adapted to a flowmeter.

There is a case that the sound velocity C is measured in another way. For example, if a flowmeter for measuring a volume flow rate is installed to measure the mass flow rate of gas, the gas pressure and temperature are separately measured. In that case, the sound velocity can be computed using the measuring results of the gas pressure and temperature. If the liquid flow rate is measured, there is a case that the sound velocity C in liquid may be previously known without being changed. In that case, the ultrasonic wave transited in the directions similar and contrary to the flow velocity are received, and the phase difference Δφ C between the received signals is measured, so that the flow velocity V could be measured based on the expression (5). At that time, if Δφ C >>π, the phase difference Δφ C is measured as follows: in order to amplitude-modulate the signal of the ultrasonic carrier frequench ƒ C into the amplitude-modulation frequench ƒ M , the amplitude-modulation frequench ƒ M is selected as follows: $$ f M ≤ C min 2 4 ⁢ L ⁢   ⁢ V max ⁢ cos ⁢   ⁢ α ( 19 )

Wherein C min is a lowest sound velocity that can be expected in fluid.

The phase difference Δφ M between the receiving signals of such like selected amplitude-modulated frequency does not exceed π in the maximum flow velocity measuring value. The amplitude-modulated signal received is demodulated, so that the phase difference Δφ M between the amplitude-modulated frequency signals is measured, and then m is obtained by the following expression (20). $$ Δφ M × f C π ⁢   ⁢ f M = m + a = Δφ C π ( 20 )

Wherein a<1.0.

The aπ in the expression (19) is a part that enables the measurement of the phase difference between the carriers. At the same time, the phase difference aπ between the carriers is measured, and Δφ C is calculated by the following expression.

Δφ C =mπ+aπ   (21)

Next, the Δφ C is substituted into the expression (5) to compute the flow velocity V. Herein, the phase difference to be measured is aπ. When measuring absolute error Δaπ of the aπ is equal to δ aπ ×aπ (δ aπ is a relative error.), the measuring error of Δφ C is as follows: $$ δ φ C = δ a ⁢   ⁢ π ⁢ a ⁢   ⁢ π ( m + a ) ⁢ π = δ a ⁢   ⁢ π 1 + m / a

Therefore, δ φ C <<δ aπ and the accuracy of the flow velocity calculation is enhanced.

Another embodiment of a system for realizing a method for measuring a flow velocity such like this method is shown as a schematic diagram in FIG. 6 .

Referring to FIG. 6 , the reference numbers are referenced by the same numbers to the same parts as those of FIG. 5 . Only, a flow velocity arithmetic logic unit is supposed to force the integers of ƒ M , ƒ C , L and cos α to be inputted therein in advance and to compute the flow velocity based on the expressions (18), (19) and (5).

Accordingly, the invention can amplitude-modulate an ultrasonic wave and measure a flow velocity in a higher reliability based on an ultrasonic time difference method in a larger river, a larger sluiceway channel, and a pipe of a larger inner diameter. Also, the invention provides a phase difference flow velocity measuring method not depending upon a sound velocity, using a general phase difference discriminator having the phase difference measuring range of π, even if the phase difference exceeds π radian.

Claims

1. A method of transiting/receiving an ultrasonic wave at a constant angle α in a direction similar or contrary to a flow velocity and measuring a flow velocity using an ultrasonic phase difference changed proportional to the flow velocity comprising:amplitude-modulating an ultrasonic wave of a carrier frequench ƒC into an amplitude-modulation frequench ƒM which is lower than the carrier frequench ƒC; transiting the amplitude-modulated signals in directions similar and contrary to the flow velocity; demodulating the amplitude-modulated signals transited through an interval distance L and then detecting signals of the amplitude-modulation frequench ƒM; measuring a phase difference ΔφM1 between the signals of the amplitude-modulation frequench ƒM transited and received in the direction similar to the flow velocity and a phase difference ΔφM2 between the signals of the amplitude-modulation frequench ƒM transited and received in the direction contrary to the flow velocity; obtaining multiples of m1 and m2 by π excluding different phase components βπ and γπ measured by at least two phase difference discriminators from phase differences ΔφC1 and ΔφC2 between the signals of an ultrasonic carrier frequench ƒC transited and received in the direction similar to the flow velocity and between the signals of the carrier frequench ƒC transited and received in the direction contrary to the flow velocity by the following expressions; ΔφC1π=ΔφM1⁢fCπ⁢ ⁢fM=m1+βΔφC2π=ΔφM2⁢fCπ⁢ ⁢fM=m2+βwherein β is smaller than 1.0 and γ is smaller than 1.0, storing m1 and m2 at the memory of a system, measuring the phase different components βπ and γπ, adding m1π and m2π to the phase different components to calculate the phase differences ΔφC1 and ΔφC2 and computing the flow velocity V based on the following expression: V=π⁢ ⁢fC⁢Lcos⁢ ⁢α⁢(1ΔφC1-1ΔφC2)selecting the amplitude-modulation frequench ƒM as follows: fM=n+a2⁢L⁢(Cmin-vmax)n=a⁡(Cmin-vmax)-b⁡(Cmin+max)Cmax-Cmin+2⁢vmax)storing n at the memory of the system as follows; and ΔφM1=nπ+aπ ΔφM2=nπ+bπmeasuring the phase differences aπ and bπ measured by the phase discriminators in the above expression and adding nπ to the phase differences aπ and bπ thereby to obtain the phase differences ΔφM1 and ΔφM2, wherein a<1.0 is a factor for selecting a maximum measuring range (aπ)max of one of the phase difference discriminators, which is 0.95, and b<1.0 is a factor for selecting a maximum measuring range (bπ)max of the other phase difference discriminator, which is about 0.2, Cmax and Cmin are maximum and minimum sound velocities, and υmax=Vmax×cos α is a maximum flow velocity measuring range; the Vmax is a maximum flow velocity and the cosα is a directivity angle of the ultrasonic wave.

2. The phase difference flow velocity measuring method as claimed in claim 1, in which:the method of measuring the phase difference ΔφC between the signals of the ultrasonic waves transited and then received in the directions similar and contrary to the flow velocity, if a sound velocity C is separately measured or constant, and computing the flow velocity V by the following expression: V=ΔφC⁢C24⁢π⁢ ⁢fL⁢ ⁢cos⁢ ⁢αwherein C is a sound velocity, ƒ is an ultrasonic wave frequency, L is a transit distance of the ultrasonic wave and cos α is a directivity angle of the ultrasonic wave, furthermore comprises steps of: amplitude-modulating at least one ultrasonic wave of the carrier frequench ƒC into an amplitude-modulation frequench ƒM, transiting the amplitude-modulated signals in the directions similar and contrary to the flow velocity, demodulating at least two signals transited and then received and measuring the phase difference ΔφM<π between the demodulated signals and obtaining a multiple m that exceeds aπ of the phase difference ΔφC by the following expression: ΔφM×fcπ⁢ ⁢fM=m+a⁢ ⁢(a<1,0)and; measuring a phase difference aπ between the received signals of the ultrasonic carrier frequench ƒC by the phase difference discriminators in order to obtain aπ, adding mπ thereto to obtain ΔφC and then calculating the flow velocity according to the expression, wherein the amplitude-modulation frequench ƒM is selected by the following expression: fM≤Cmin24⁢LVmax⁢cos⁢ ⁢αwherein Cmin is a lowest sound velocity, L is a transit distance of the ultrasonic wave, Vmax is a maximum flow velocity and cos α is a directivity angle of the ultrasonic wave.