Phase compensation of op amp In order to make the op amp work properly, a phase compensation capacitor is often added between the input and output in the circuit. 1. There are theoretical calculations for compensation capacitors, but when it comes to mature design, it seems that most people rely on their previous debugging experience. Generally, the value of the capacitor should take into account the frequency response of the system (simply put, the larger the capacitance, the narrower the bandwidth), and then there is the oscillation problem; if you have to calculate, you can look at the distributed capacitance at the input of the op amp. For example, the negative feedback amplifier circuit is to ensure that the product of the resistance of the input resistor and the distributed capacitance = the product of the resistance of the feedback resistor and the capacitance you want to add... 2. Two functions 1. Change the phase shift of the feedback network to compensate for the phase lag of the op amp 2. Compensate for the effect of the capacitor at the input of the op amp (in fact, it is ultimately phase compensation...) Because the op amps we use are not ideal. Generally, the operational amplifiers used in practice have a corresponding phase shift effect on signals of a certain frequency. Such signals fed back to the input will make the amplifier circuit unstable or even oscillate. For this reason, corresponding capacitors must be added to provide a certain phase compensation. Compensation capacitors are usually built into the op amp. Of course, they can be added to the circuit if necessary. Their value depends on the signal frequency and circuit characteristics. Op amp input compensation capacitor The input parasitic capacitance Cs of a general linear amplifier (that is, an amplifier circuit that introduces negative feedback) will affect the stability of the circuit. The compensation measures are shown in the figure. There is usually a parasitic capacitance Cs of about a few picofarads at the input end of the amplifier. This capacitance includes the input capacitance of the op amp and the distributed capacitance of the wiring. It forms a lag network with the feedback resistor Rf, causing the output voltage phase lag. When the frequency of the input signal is very high, the bypass effect of Cs makes the high-frequency response of the amplifier worse, and the upper limit frequency of its frequency band is approximately: ωh=1/(2πRfCs) If the resistance value of Rf is large, the upper limit frequency of the amplifier will be seriously reduced. At the same time, the additional lag phase introduced by Cs and Rf may cause parasitic oscillation, which will cause serious stability problems. A simple solution to this problem is to reduce the resistance of Rf so that ωh is higher than the frequency range of actual application, but this method will reduce the voltage gain of the operational amplifier (because Av=-Rf/Rin). In order to keep the voltage gain of the amplifier circuit high, a more common method is to connect a compensation capacitor Cf in parallel to Rf so that the RinCf network and the RfCs network form a phase compensation. RinCf will cause the output voltage phase to advance. Since the value of Cs cannot be accurately known, the phase advance and lag cannot be fully compensated. Generally, a variable capacitor Cf is used to minimize the additional phase shift by experiment and adjustment of Cf. If Rf=10kΩ, the typical value of Cf is 3~10pF. For voltage followers, the Cf value can be slightly larger. Compensation of op amp output capacitance For many integrated operational amplifier circuits, if the value of the output load capacitance CL is much larger than 100pF, the output capacitance (including parasitic capacitance) and the output resistance will cause additional phase shift. The accumulation of this additional phase shift may produce parasitic oscillation, making the amplifier operation seriously unstable. The solution to this problem is to connect a resistor Ro in series at the output of the op amp to isolate the load capacitor CL from the amplifier circuit. As shown in the figure, a feedback resistor Rf is connected behind Ro to compensate for DC attenuation. Adding feedback capacitor Cf will reduce the high-frequency closed-loop voltage amplification factor. The selection method of Cf is: make the capacitive reactance Xcf ≤ Rf/10 of the amplifier circuit at the unit gain frequency fT, and Xf=l/(2πfTCf). Under normal circumstances, Ro=50~200Ω, and Cf is about 3~10pF. In addition to the above-mentioned unstable factors, there are other unstable factors, some of which come from the integrated chip itself. Some are derived from the system circuit (such as the coupling problem of the internal impedance of the power supply). Sometimes it is difficult to solve the instability problem using many methods, but the problem can be solved by using appropriate compensation methods. For example. When the amplifier does not need a wide frequency band and optimal conversion rate, the over-compensation method for the integrated operational amplifier will achieve good results, such as increasing the compensation capacitor by 9 times or the multiple required to achieve stability. For μA301 operational amplifiers, the effect is generally good.
|