Like many others I got the undervoltage warning on a pi3b+. I see most answers blame a poor power supply. I plugged mine to a bench power supply to test this, I set it to 4 amp and kept getting undervoltage warnings until I supplied 5.5V. Is it possible that the rasbpberry pi is actually defective and needs more than 5V (my usb power supply is actually providing 5.10) ? Is there anything I can do to investigate this further?

  • Perhaps between the power supply and the Pi some volts are being dropped? – joan Mar 26 '19 at 11:41
  • That is possible but I tried many cables with the usb power supply and I always got undervoltage. (Although I tried only 1 with the bench power supply because I had to cut it ) – Enrico Mar 26 '19 at 11:49
  • So what was the voltage at the end of the microUSB cable? – joan Mar 26 '19 at 11:50
  • I cannot measure the microusb end, I get 5.6 in the middle of the cable and 5.3/5.4 on the GPIO – Enrico Mar 26 '19 at 12:06
  • By "on the GPIO" you mean the 5V rail? (Not technically a GPIO but situated by the on the header.) If so, that should be essentially the same as the end of the USB cable since they are connected internally. If not the same, that would indicate some issue with the board itself. – Brick Mar 26 '19 at 15:43

There is nothing wrong with the Pi3B+ power supply - it is an excellent design (which you can find documented on the Foundation site). Some few boards have had failures of the 3.3V rail.

The many power problems Pi users have are a combination of poor power supplies and inadequate cables. If you are getting 0.2 drop on half of your cable it is rubbish.

Increasing the voltage to compensate for inadequate cables is poor practice and potentially dangerous.

See Raspberry Pi Power Limitations

  • I just tested many cables and many power supplies (including a laptop). I get around 0.01 V drop from the 5V rail when the pi is off (so it's not the cable). But during boot or any operation that uses IO or CPU i get up to 0.3V drops. This with a 4 amp bench power supply, an apple 5W adapter, an anker multiport usb charger rated for 5V 8A, and an old macbook air. So the only way so far not to get undervoltage warning is to supply at least 5.3V. It is possible that all of these power supplies can't keep 5V steady but unlikely IMO. What else can I check? Can the SD card cause this behaviour? – Enrico Mar 26 '19 at 22:47
  • Well.... I knew most usb cables were bad... but this is ridiculous. I kept trying more cables and eventually found one that doesn't cause undervoltage. My above conclusions were wrong, the voltage drop is lower when there's less current going through. Now I have 1 cable that I will hold like a sacred relic. – Enrico Mar 26 '19 at 23:08
  • 1
    "I get around 0.01 V drop from the 5V rail when the pi is off (so it's not the cable)". Of course, the voltage drop depends on the load. The cable and the Pi act as a voltage divider, if you will. Plus there are inductive effects you may not see unless you use an oscilloscope (rapid, large spikes up and down). USB-IF certified charging cables (i.e. official USB Cables using the real logo) should never cause you problems at 2A, that means your local supply of cable is polluted with out of spec cables. – crasic Mar 26 '19 at 23:23

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.