“The measurement of nonlinear distortion generally adopts the fundamental wave suppression method (single tone method), which can be realized by the fundamental wave suppression network. The fundamental wave suppression network is the notch filter, which can filter out the fundamental wave voltage component. The common ones are the RC trap circuit composed of the Wien bridge and the trap circuit composed of the double T-shaped bridge.
“
1 Introduction
The measurement of nonlinear distortion generally adopts the fundamental wave suppression method (single tone method), which can be realized by the fundamental wave suppression network. The fundamental wave suppression network is the notch filter, which can filter out the fundamental wave voltage component. The common ones are the RC trap circuit composed of the Wien bridge and the trap circuit composed of the double T-shaped bridge.
A high-performance distortion meter must use a high-performance trap, which should be able to completely filter out the fundamental without attenuating other harmonics. The new distortion meter produces fundamental attenuation or notch depth of 100 dB or more, while attenuating harmonics by only 1 dB or less. Achieving such high performance requires filters with very high Q values, and the tuning must be very accurate, which is nearly impossible with the usual manual tuning. High-performance distortion meters can automatically tune to the fundamental frequency with a deviation of only a few percent. The measurement of distortion is mainly to design and select a high-performance notch filter circuit.
The Venturi bridge notch filter is a commonly used device in the design of distortion meters, and its fundamental attenuation depth can generally reach more than 80 dB. However, manual tuning is often used in old-fashioned distortion meters. On the basis of the original manual tuning Venturi bridge trap network, an intelligent numerical control tuning Venturi bridge trap is designed.
2. The principle of the Venturi bridge trap
The fundamental wave suppression circuit (wave trap) composed of the Wien bridge is shown in Figure l.The relationship between the components of the bridge is
Because Rl=2R2, for any frequency signal, UAD=Ui/3. It can be known from the calculation: when the input signal frequency f=fo, UBD=Ui/3, then UAB=0. At this point, the bridge is in balance and the output is zero. When the input signal frequency f deviates from fo, the bridge is out of balance, and there is a voltage output.
The selection characteristics of the Wien bridge passive filter circuit are poor. In practical work, a notch filter with very narrow stopband and strong selectivity is required. For this purpose, an active notch circuit composed of a Wien bridge is used, as shown in Figure 2. At this time, the frequency of the notch is l kHz.
Al and A2 are voltage follower configurations with buffer isolation, high input impedance and low output impedance, and have no effect on the resonant frequency of the frequency selection circuit. Part of the voltage output by A1 is fed back to the non-inverting terminal of A2, and is A2 is output to the bridge arm. Adjusting Rp can adjust the feedback amount, thereby changing the Q value, so as to achieve the function of sharp passband frequency selection. If no positive feedback is added, the characteristic curve of the second harmonic will drop near 1 kHz, and accurate measurement cannot be performed. If the feedback amount is related to the frequency characteristic, adjust it with a variable resistor Rp; if the attenuation characteristic has been adjusted and the Q value has been selected, then Rp can be replaced with a fixed resistor. Adding resistor R8 to the Al feedback loop is to offset the input bias current to reduce DC drift. The role of C3 is to suppress spikes.
When f=fo, the bridge is balanced, and the output of Al is 0; when f deviates from fo, the bridge is unbalanced and there is an output voltage. Therefore, the circuit can suppress the fundamental wave and let the harmonics pass.
If take fo=l kHz, C=0.01μF, calculate R from R=l/2πfoC, and obtain R=15 kΩ. A1 and A2 are integrated operational amplifiers, NE5532A is optional.
A high-Q notch filter has good selectivity. However, the center frequency fo is easy to shift, which will cause a large measurement error. Therefore, when measuring the distortion degree, a two-stage or even three-stage series tuning design can be used, so that it has an attenuation bandwidth of ±1% of the center frequency.
3. System modules
The intelligent numerical control tuning Venturi bridge trap includes the notch frequency tuning Venturi bridge, RMS detector, A/D sampling circuit and single-chip control circuit, as shown in Figure 3.
In the system, after a signal of unknown frequency is input to the Wien bridge, a notch is performed at a certain frequency point, and the residual signal output by the Wien bridge is detected by the RMS detection circuit; The DC voltage generated after the detection is sampled, converted into a digital signal, and the data is transmitted to the single-chip microcomputer; the single-chip computer judges this data, when the collected DC level is the value, the resonant center frequency of the Wien bridge is exactly the required value The notch frequency (that is, close to the fundamental frequency); if the collected DC level is not a value, then the microcontroller will control to change the resistance and capacitance of the Wien bridge to make its center frequency close to the fundamental frequency. Through the above process, the intelligent numerical control tuning of the Venturi bridge trap is realized.
The Wien bridge in Figure 3 is improved on the basis of Figure 2. R and C in Figure 2 are no longer composed of single-value resistors and capacitors, C is composed of a parallel capacitor network, the resistor R is replaced by a numerically controlled potentiometer, and R and C can be controlled by a single-chip microcomputer.
The function of detection is to convert the residual signal output by the Wien bridge into a detectable value, and provide it to the analog-to-digital converter for sampling and conversion into a digital signal.
The function of the A/D sampling circuit is to sample the analog signal output by the effective value detection, convert it into a digital signal, and then process it by the single-chip microcomputer.
The single-chip control circuit mainly realizes the processing of the data after sampling, the selection of the capacitance file (controlling the on-off of the relay) and the control of the numerical control potentiometer.
4. System design and implementation
4.l Wien Bridge
The key part of the system hardware circuit is the Wien bridge. The overall circuit structure of the system is shown in Figure 4. The purpose of this system is to achieve automatic tuning of the notch. Therefore, R and C, which play a decisive role in the resonant center frequency of the Wien bridge, must be improved from fixed values to changes that can be automatically changed within a certain range. Considering that the double-connected variable capacitor is difficult to buy, and the double-connected variable resistor itself is not very convenient, it is not very convenient to use, so the step-by-step capacitor is used to realize the coarse adjustment of the resonance center frequency, and the numerical control potentiometer realizes the fine adjustment scheme. Use ordinary monolithic capacitors, CBB capacitors with large capacitance are available. The 100-tap X9C103 numerically controlled potentiometer, which is relatively easy to buy, is used. The interface between X9C103 and the single-chip microcomputer is a 3-bus mode, and the three control ports are U/D, INC, and CS. In the actual design, the three ports are respectively connected with the P2 of the single-chip microcomputer. O, P2.1 and P2.2 are connected. The sliding head of the X9C103 has a fixed resistance value of 40 Ω, so the actual resistance value change range is 40 Ω~10040 Ω, and the step size is 100 Ω.
The binning capacitor is connected to the relay. Using double pole single throw relays, each relay controls 2 capacitors of the same value. As the switch in Figure 4, the relay is usually in a normally-off state. The on-off of the relay is controlled by the single-chip microcomputer to switch on the required capacitance gear. The 7-speed capacitor corresponds to 7 relays, which are respectively connected to the P1.0-P1.6 ports of the single-chip microcomputer. . The selection of the capacitance value of each grade is very important in this circuit. First of all, it is necessary to consider whether the frequency tuning range can cover the entire frequency band required by the system. Each grade of capacitance corresponds to a certain range of frequencies. The step size corresponding to the frequency should be small to reduce the error between the notch center frequency and the fundamental frequency. When the Wien bridge is working, the latter capacitor is gated, which can reduce the interference of the relay to the RC resonant network. Considering the above 3 points, after calculation and practice, we selected 7 suitable capacitors within the frequency range of 10Hz to 1 MHz, as shown in Table 1.
4.2 Detection and A/D Conversion
Considering that the input signal itself is an irregular distortion signal in the distortion measurement, and the RMS detection circuit composed of discrete components calculates the RMS according to a certain relationship after detecting the peak value of the signal, which can generally only be used for Detection of regular signals (such as sine wave and other signals), the output error is relatively large, not suitable for distortion meter, so the system AC detection signal – DC RMS conversion using AD536 conversion circuit. AD536 is a monolithic integrated circuit specially used for true RMS-DC conversion introduced by ADI Company in the United States. Its performance is comparable to or better than hybrid or analog-digital devices, and the price is much lower. The AD536A directly calculates the true rms value of any complex input waveform that contains an AC component of DC and converts it to a DC output signal. The AD536A can be widely used to measure various noise and mechanical sensing signals of standard sine wave or aperiodic, non-sinusoidal and superimposed DC level. In order to reduce the ripple component in the output, a backward filter is used, such as R9, C18 and C19 in Figure 4, which play a filtering role.
In this system, the output signal after each resonant frequency adjustment of the Wien bridge must be measured and compared with the previous and previous measurements. With the above true RMS detection, the RMS value of the signal after the notch can be directly detected. . The author uses ADC0809 universal 8-bit parallel analog-to-digital converter to convert the detected DC signal into binary data for processing by a single-chip microcomputer. ADC0809 has 8 analog input channels, this system only needs to use one. The 8-bit data output end of ADC0809 is connected to the P0 port of the microcontroller, the CLK signal is connected to the ALE port of the microcontroller, CE and START are connected to the P2.6 and P2.7 ports of the microcontroller respectively, and the EOC is connected to the INT0 port of the microcontroller. Because an ALC (automatic level control) circuit has been designed before this system to reasonably control the voltage amplitude of the input Wien bridge, the reference voltage of ADC0809 can be 5 V, using LM336-5 integrated voltage regulator power supply.
4.3 Microcontroller Control
This system selects AT89C51 single-chip microcomputer. Regardless of cost, processing speed or storage capacity, the choice of AT89C51 is very reasonable. After calculation, it can be seen that when the capacitance is relatively small, the corresponding frequency change is relatively large when the numerical control potentiometer changes by 100 Ω. In order to shorten the tuning time, the program design starts from the capacitor with the capacitance value to scan in sequence to search for the appropriate notch center frequency.
After the system is started, the microcontroller program is initialized first (that is, the capacitance value is selected as 0.22 nF, and the value of the numerical control potentiometer is 40 Ω), and then the microcontroller controls the ADC0809 to sample, and reads the P0 port data for processing. First perform a rough scan on the entire system, that is to say, do not change the resistance value of the numerical control potentiometer first, only perform capacitance shifting, when the program scans the entire 7-speed capacitor, compare the data of ADC0809, and select the corresponding capacitance file. as the file required by the system. Then use the digital potentiometer to scan, X9C103 has 100 taps, but it is impossible to scan 100 times.In the program written by the author, it is set to scan 5 values after scanning to 1 value. If these 5 values
are larger than the previous value, then that value is the value, and the corresponding numerical control potentiometer value is required, and the system will be stable. The X9C103 interface is a 3-bus type, the communication protocol is relatively simple, and the programming is more convenient. Figure 5 shows the software flow of the system.
5. Conclusion
This paper describes the design of the fundamental wave suppression network in the distortion measuring instrument. In order to complete its function well, an intelligent numerically controlled tuned Wien bridge notch filter is designed and produced. Although the notch depth of the Wien bridge notch filter is still Yes, it can reach more than 60 dB, but the notch frequency band is not wide enough, and the multi-stage series connection can realize the broadening of the notch frequency band. Generally, the three-stage notch can have very good performance. The notch filter is also suitable for other similar occasions.
The Links: PM100CVA120 KCS057QV1AJ-G23