3D之家

service and support

service and support

SERVICE AND SUPPORT

3D之家 > Service and Support > Instrument Knowledge

Research on Ultrasonic Flowmeter Based on Voltage Modulation

2016-05-23
Diyuan Instrument Company News, Academic Papers, Blog Manuscript Call Form
details 0 Introduction Ultrasonic waves are generated by mechanical vibration, and their characteristics include short vibration frequency and wavelength, directional propagation, and strong penetrability. Ultrasonic flowmeter is an instrument that measures the volume of flow by detecting the effect of the ultrasonic beam on the flow of fluid. The time-of-flight method is one of them. It is obtained by measuring the difference between the forward and backward flow time of ultrasonic signals in the gas. Taking flow rate is a very ideal energy-saving flow meter. The state of fluid flow is not necessarily ideal, so if you want to improve the measurement accuracy of the flow meter, you need to improve the accuracy of data detection to eliminate the effects of non-steady state. With the introduction of high-speed integrated chips and analog switches, the system can accurately detect the zero-crossing mark. The accuracy of the entire detection process is determined by the timing chip and sampling chip. With the application of DSP and A / D sampling in ultrasonic testing and the use of zero-crossing curve fitting methods, the accuracy of data processing can be improved as much as possible. But usually to obtain more accurate waveform data, the higher the number of ADC bits is required, and this is inconsistent with the index of the sampling rate. If you rely entirely on the digitized waveform to determine the transit time, it is difficult to take into account the sampling accuracy and rate. This study designed an ultrasonic gas flowmeter amplitude-modulation excitation circuit and echo circuit. After digitally determining the waveform characteristic points, combined with the analog circuit to determine the position of the zero-crossing point, it is possible to maintain a lower sampling rate while ensuring the accuracy of time measurement. . 1 Working principle of ultrasonic flowmeter The time difference method of ultrasonic flowmeter is approximately a straight line, and the cross section of the working area is shown in Figure 1. Figure 1 Sectional view of the meter body. On the pipeline through which the fluid passes, two AB transducers with an angle of θ (the value of θ in the figure is 45 °) are installed. The diameter of the pipe is D, the linear distance between the two transducers is L, and the velocity of the ultrasonic wave in a stationary fluid is C. A beam of ultrasonic pulses travels through the fluid:

V = 0, t1 = L / C (1) A to B, that is, the ultrasonic wave propagation time in the downstream direction is:

t2 = L / (C + V cosθ) (2) B to A, that is, the propagation time of the ultrasonic wave in the reverse flow is:

t3 = L / (C -V cosθ) (3) The data of the forward and backward flow of the fluid are measured experimentally. According to formulas (2, 3), the average linear velocity V on the cross section of the channel under different conditions can be obtained. In the formula, L, C, and θ are all known. Only accurate t2 and t3 are required to obtain accurate data. So the key to ensuring the accuracy of the flow meter is to accurately sample and process the transit time. 2 Excitation Circuit and Signal Receiving Circuit 2.1 Design of Ultrasonic Excitation Circuit The traditional ultrasonic excitation method generally uses a square wave with a certain duty cycle to excite the transducer [7-8]. This study uses an amplitude-modulated excitation pulse method. The CMOS analog switch CD4052 is used to select the signals of several external voltage signal sources. To avoid excessive current during channel switching, the input terminals are connected to current limiting resistors R1 ~ R3. The excitation intensity can be adjusted by adjusting the external voltage. The circuit is shown in Figure 2. Figure 2 Transducer excitation circuit The transducer is selected as LHQ200-3, and its working center frequency is 200kHz ± 4%. The center frequency of the excitation pulse waveform should be consistent with the working frequency of the transducer. By controlling the pin AB, you can Implement voltage channel switching to generate the required waveform. The results of the excitation waveform observation in the oscilloscope are shown in Figure 3. The first 10 cycles are 5V small voltage, then 2 cycles of 24V high voltage, and finally the 2 cycles of 24V voltage suppression. The time interval between two adjacent excitation waveforms is about 2ms, which prevents the interval between transmitting waveforms from being too short to generate adjacent signals to each other. Excited. CD4052 is a CMOS analog switch. The chip's power supply voltage VCC is set to 24V, and the control signal sent by the DSP control chip is too small to meet the control requirements. The CD4052 can be effectively controlled by zooming in to a high level above 0.7VCC. Figure 3 Excitation pulse waveform Figure 2.2 Design of the ultrasonic receiving circuit The role of the ultrasonic receiving circuit is to sufficiently amplify and filter the small signal received by the receiving transducer into a sufficiently large signal. After propagation in the pipeline, the waveform received by the receiving transducer becomes very weak, only a few millivolts to tens of millivolts, so the received signal must be amplified and filtered. The ultrasonic frequency is high, so the amplifier chip is required to have a sufficient bandwidth, and the circuit is designed for two-stage NE5534 amplification. It is a single low-noise high-speed dual-supply power amplifier as shown in Figure 4. The magnification of the circuit in Figure 4 is determined by the input and output resistance. The magnification is about 400 times, which can enlarge the maximum voltage to a suitable range. Figure 4 Receive amplifier circuit Before sampling, this research designed a π-type low-pass filter to do anti-aliasing filter processing. By observing the oscilloscope, you can see that it has a certain suppression effect on waveform glitches and other interference. 3 Experimental data analysis The processed waveforms of the signals received by the transducer under different wind volumes are shown in Figure 5. The receiving waveform when there is no wind speed is shown in Figure 5 (a), and the downstream receiving waveform when the air volume is 207m3 / h is shown in Figure 5 (b). The CH2 trigger signal is used as the reference. Each received waveform has a point where the peak value increases before and after. This point is basically stable at the 12th peak of the waveform. As shown in Figure 5, as the wind volume increases, the position of the peak (with the falling edge of the trigger waveform) The time difference t) will change, and the amount of change is small and closely related to the velocity of the fluid. By comparing with the emission waveform, it is concluded that this is the second 24V voltage excitation result in the excitation voltage, and the time difference between the forward and reverse currents at different wind speeds can be obtained through the position detection result of the first zero crossing point after the peak. In A / D conversion, this point is the timing end time. After the zero-crossing curve is fitted during the processing, more accurate zero-crossing information can be obtained. The median filtering can effectively reduce the deviation caused by unstable flow velocity. V calculated according to formula (2, 3) is the average velocity on the propagation channel line, which needs to be modified to obtain the average velocity of the pipe section VM = V / K, the flow correction coefficient K = (2n + 1) / 2n, When the Reynolds coefficient is 104, n is 6.5, and the processing result is shown in Table 1. The reference flow rate is from a 0.5-stage turbine flowmeter. (A) Received waveform without wind speed

(B) The measurement error of the downstream receiving waveform when the air volume is 207m3 / h includes the indication error and the reference error. The measurement accuracy of the instrument is indicated by the reference error. The indication error is the difference between the measured value and the true value, and the reference error is the measurement. The ratio of the absolute value of the value to the full scale value of the meter. When full scale measurement value is taken as the maximum air volume VM, the error analysis table is shown in Table 2.

Table 1 Data processing table
Benchmark flow
/ (M 3 · h -1
Downstream zero crossing position Countercurrent zero-crossing position Countercurrent and zero-crossing median difference / (10 -7 s) Average speed V M / (m · s -1 )
50 883.000 0 915.000 0 31.0000 1.6671
884.000 0 964.000 0
885.000 0 914.379 9
886.000 0 914.000 0
882.670 2 919.999 9
101 863.281 0 929.326 6 65.0000 3.4956
866.659 8 929.666 7
864.690 9 931.000 0
869.542 6 930.000 0
865.000 0 932.709 8
122 853.343 2 938.549 3 79.4999 4.2754
857.297 2 929.999 9
854.000 1 936.676 0
852.616 1 929.679 2
860.000 0 933.500 0
185 827.290 4 945.000 2 120.1976 6.4640
824.000 0 944.524 2
824.326 6 942.000 2
828.460 7 991.372 9
821.000 0 933.361 8
207 817.999 9 956.463 7 135.0001 7.2460
862.480 9 955.000 1
821.409 9 957.501 9
820.000 0 952.511 8
819.000 0 943.351 2
Table 2 Error analysis table
Benchmark flow
/ (M 3 · h -1
Reference average speed / (m · s -1 ) Velocity display error / (m · s -1 ) Flow rate reference error / (%)
50 1.768 4 -0.101 3 1.398
101 3.572 1 -0.076 5 1.056
122 4.314 9 -0.039 5 0.545
185 6.543 0 -0.079 0 1.09
207 7.321 1 -0.075 1 1.036
According to the results in Table 2, it can be known that the measured average flow velocity does not deviate much from the actual, and the measurement citation error is kept within 1.5%, which can meet the measurement requirements. One acquisition process is 2ms, the acquisition time is 800μs, and the remaining time is allocated for data processing and data display. For subsequent data, only the sampling of the zero-crossing position can be used to meet the measurement effect. This can effectively solve the problem that the sampling accuracy and efficiency cannot be taken into account in A / D sampling. 4 Conclusion This experiment designed a method based on voltage amplitude modulation, which has the advantages of easier implementation and simple subsequent data sampling and processing. Calibration experiment results show that the method has good measurement accuracy in general flow fields, but also has the disadvantage of poor anti-interference ability common to the amplitude modulation method. In the industrial environment, the meter will be subject to a lot of interference from the surroundings, which will bring great errors to the measurement results. Therefore, appropriate measures need to be taken in terms of anti-interference, and the effect needs to be verified in subsequent studies. Article excerpt: Dong Denghui, Bao Min, Zheng Xibin. Research on Ultrasonic Flowmeter Based on Voltage Modulation [J] .Mechanical & Electrical Engineering. 2014 (02)

25 years quality guarantee

trustworthy

加盟咨询 客户服务 留言板 Customer service message board for national high-tech enterprises joining consulting