LOTO Oscilloscope_Measure power supply open loop gain/power supply loop frequency response curve/PSM from scratch

LOTO2018

LOTO Oscilloscope_Measure power supply open loop gain/power supply loop frequency response curve/PSM from scratch [Copy link]

We have previously written an article that demonstrated how to measure the open-loop gain curve of a power supply loop from theory to practice, but it focused on theory and principles without showing many details. So this article starts from a different perspective, starting from scratch, and demonstrates step by step how to conduct practical tests.

The link to the previous article is: "LOTO oscilloscope measured open-loop gain frequency response curve/power supply loop response stability" https://www.bilibili.com/read/cv18591666/

We first get a power board. As shown in the figure below, we bought a very small and simple power module. Like most power systems, it is a negative feedback closed-loop system, which is convenient for demonstration and very representative.

We found the typical application schematic diagram of this power supply through the chip model. This circuit is basically made according to this schematic diagram, as shown below:

Next we need to disconnect the power loop at the location indicated by the purple arrow and insert a 50 ohm injection resistor in series. As shown in the figure below, we found this location on the power board and cut off the copper:

Solder an injection resistor and extend the pins at both ends of the resistor to facilitate signal injection and testing:

Now it becomes like this:

We started wiring with the LOTO oscilloscope OSCA02S, the oscilloscope with signal source module, and the injection transformer module Trans01.

Next, we can set the parameters of the sine wave sweep and adjust the signal amplitude, observe the signal waveform of the AB channel of the oscilloscope, and try to reduce the signal amplitude to ensure that the waveform is not distorted. Since we inject from point A, we adjust the amplitude of the signal source to prevent distortion in channel A. Channel B is the signal returned through the feedback channel, and it will always be distorted in a certain frequency band.

We can look at the AB waveform at different frequency points, and try to start the test from the frequency point where both AB channels are not too distorted.

As shown in the figure above, at 400HZ, the waveform of injection point B is distorted, but as shown in the figure below, after 1KHZ, the waveforms of both channels are not distorted:

Since the frequency range of our injection transformer is 100HZ~10MHZ, the low frequency must be at least higher than 100HZ, but lower than the crossover frequency of the power supply. We can first sweep the frequency to get a rough idea of the crossover frequency value and the sweep step size, and then make adjustments later.

We also have related articles and video demonstrations of the frequency sweeping steps. Let’s take a look at the results:

The green solid line in the figure is the open-loop gain curve, and the green dotted line is the phase angle. We can see that when the G=1 gain of this power board is 0DB, the crossover frequency is about 1.57K Hz, the phase angle is about 80 degrees, and the phase margin is more than 100 degrees. The crossover frequency of this power board is still very low, and the general power supply is around 10K to tens of kHz. The lower the crossover frequency, the more inconvenient it is to test, because it is closer to the low-frequency area, and the injection transformer has a low-frequency starting frequency limit.

By sweeping the frequency multiple times and storing the frequency response curves, we can see that basically changing some sweep parameters will have a small impact, as shown in the different colored curves in the figure below, but it is generally consistent.

We put the relevant video demonstration at: https://www.bilibili.com/video/BV1KC4y1G7mb/?spm_id_from=333.999.0.0

秦天qintian0303

LOTO oscilloscope is a module without screen and operation.