TI Power Switch Design Tips 2 - Driving Noise Power

MOS power IC full range

Probe domestic switch needle KG-300K needle head diameter is 3.0mm normally open switch needle

Passive 3225G 8MHZ 10MHZ

Noise-free power supplies are not accidentally designed. A good power layout is designed to minimize experiment time. Taking minutes or even hours to look at the power layout carefully can save you days of troubleshooting time.

Figure 1 shows the block diagram of some of the main noise-sensitive circuits inside the power supply. The output voltage is compared to a reference voltage to generate an error signal, which is then compared to a ramp to generate a PWM (Pulse Width Modulation) signal for driving the power stage.

Power supply noise comes mainly from three places: error amplifier input and output, reference voltage, and ramp. Careful electrical and physical design of these nodes helps minimize troubleshooting time. In general, noise is capacitively coupled to these low level circuits. An excellent design ensures tight layout of these low level circuits and away from all switching waveforms. The ground plane also has a shielding effect.

Figure 1. Many noise formation opportunities for low-level control circuits

The error amplifier input may be the most sensitive node in the power supply because it typically has the most connected components. If you combine it with the extremely high gain and high impedance of this stage, you will have endless troubles. During the layout process, you must minimize the length of the node and place the feedback and input components close to the error amplifier as close as possible. If there is a high frequency integrating capacitor in the feedback network, you must place it close to the amplifier and the other feedback components follow. Also, a series resistor-capacitor may also form a compensation network. The most ideal result is to place the resistor close to the input of the error amplifier, so that if a high-frequency signal is injected into the resistor-capacitor node, then the high-frequency signal has to withstand a higher resistance impedance - and the capacitor has a high-frequency signal The impedance is small.

Slopes are another potential source of noise problems. The ramp is typically generated by capacitor charging (voltage mode) or by sampling (current mode) from the power switch current. In general, the voltage mode ramp is not a problem because the impedance of the capacitor to the high frequency injected signal is small. Current ramps are tricky because of rising edge peaks, relatively small ramp amplitudes, and power stage parasitics.

Figure 2 shows some of the problems with current ramps. The first image shows the rising edge peak and the resulting current ramp. The comparators (depending on their speed) have two potential trip points, and the result is an out-of-order control that sounds more like a fried bacon.

This problem can be well solved by using rising edge blanking in the control IC, which ignores the initial portion of the current waveform. The high frequency filtering of the waveform also helps to solve this problem. Also place the capacitor as close as possible to the control IC. As these two waveforms show, another common problem is subharmonic oscillation.

This wide-narrow drive waveform appears to be insufficient slope compensation. Adding more voltage ramps to the current ramp will solve the problem.

Figure 2 Two common current mode noise problems

Although you have designed the power layout quite carefully, there is still noise in your prototype power supply. What should I do? First, you have to make sure there is no problem with the loop response that eliminates the instability. Interestingly, the noise problem may look like instability at the crossover frequency of the power supply. But the real situation is that the loop is correcting the injection error with its fastest response speed. Again, the best approach is to identify that noise is being injected into one of three places: the error amplifier, the reference voltage, or the ramp. You only need to solve it step by step!

The first step is to check the nodes to see if there is significant nonlinearity in the ramp or if there is a high frequency change in the error amplifier output. If no problems are found after the inspection, the error amplifier is removed from the circuit and replaced with a clean voltage source. This way you should be able to change the output of this voltage source to smoothly change the power output. If this works, then you have narrowed the problem down to the reference voltage and error amplifier.

Sometimes the reference voltage in the control IC is susceptible to the switching waveform. This situation may be improved by adding more (or appropriate) bypasses. In addition, using a gate drive resistor to slow down the switching waveform may also help to solve this problem. If the problem is with the error amplifier, it can be helpful to reduce the compensation component impedance as this reduces the amplitude of the injected signal. If all of these methods do not work, then the error amplifier node is removed from the printed circuit board. Air wiring the compensation components helps us identify where there is a problem.