MP2338 and output voltage transients

sagonzal · May 18, 2026, 3:58pm

Hi, I am using 2 x MP2338 in a board to produce 5V0 and 3V3 rails out of 13Vdc input. Output voltage is working (references being followed and accurate voltage output is being tracked) but a periodic perturbation makes a glitch at every output.

The 5V0 rail is the less problematic and the 3V3 rail is the more affected. Both are following reference designs (per the data sheet) and there is no load o very light load at the output.

The scope at the output of the inductors indicate the device is actually not controlling properly the output voltage. Please see the output of every rail after at the capacitor bank on each rail.

Any suggestion ?

Do you have a compensation guideline for this specific IC ?

Regards,

Sergio

sagonzal · May 20, 2026, 2:29pm

A clarification: my previous scope measurements we see the voltage at the INPUT of the inductor but that is clearly the root of the problem. Can’t get to see the classical saw-tooth voltage. This tells me the device is entering some discontinuous conduction mode (DCM).

To add more measurements in order to understand the problem, I’m attaching output voltage scope readings showing this perturbation whcih is currently around 3.2kHz.

Which is odd that in both rails we have the same perturbations. this make me questions the 10kOhm resistor entering the feedback; this increases the noise at the input of the error amplifier.

Has MPS seen anything like this ?

Any ideas ?