I have a board where I use the Jetson Nano module as the CPU.

During the power-on sequence and data processing the Jetson introduces about 400mv peak to peak noise/ringing on the 5V (SMPS) rail causing my Hall effect sensors to misbehave.

I have tried additional decoupling capacitance, a ferrite bead + additional capacitor near the Jetson on the 5V line and a uni and bidir TVS on the 5V rail, so far nothing worked.

What is an effective way to combat this besides giving the Jetson its own power supply?

The 5V supply

  • \$\begingroup\$ Schematic please. \$\endgroup\$ – Andy aka Jul 31 '20 at 9:06
  • \$\begingroup\$ Personally, I would play with the compensation network as it determines the speed of the system's response to sudden load changes. By the way, what is FS2? \$\endgroup\$ – Rohat Kılıç Jul 31 '20 at 9:24
  • \$\begingroup\$ @RohatKılıç what exactly would you play around with then? And that's a polyfuse that has no business being there. \$\endgroup\$ – Alex Er Jul 31 '20 at 14:34
  • 1
    \$\begingroup\$ I mean R11-C13-C12. For example, decreasing the RC constant (e.g. decreasing R11) may lead the feedback network to respond faster. You can start with R11=10k, for example. I asked about FS2 because I first thought that it may cause a voltage drop across itself. \$\endgroup\$ – Rohat Kılıç Jul 31 '20 at 14:59
  • \$\begingroup\$ @RohatKılıç good call, will try that now. Thanks! \$\endgroup\$ – Alex Er Aug 3 '20 at 9:33

Tweaking the compensation network of the power supply solved the issue, I ended up increasing R11 to 47k (calculated 46k8). Carefully (re)read the datasheet if your SMPS is not working properly, there's a lot you might have overlooked at first. Or try WeBench. I did both to be sure.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.