Rover Stuck in a Boot Loop

Hey y’all, our BlueROV2 (Raspberry Pi 4 + Navigator Rev 5, BlueOS) is in a boot loop and never reaches the BlueOS web interface. After extensive troubleshooting (multiple SD cards, multiple BlueOS images, a new Pi, swapped power sources, and isolation testing), our strongest finding is that the 5V converter powering the Pi runs absurdly hot during operation. We believe the converter is the root cause, but we want professional help

Symptom

On power-up, the system enters a repeating initialization cycle (the looping init beeps audible from the ESCs). The Pi powers on, LEDs light, the green ACT LED shows boot activity for a window, then activity stalls or resets. Across all attempts, we have never reached the BlueOS web interface at 192.168.2.2 or blueos.local.

Best behavior observed: on a proper 5V/3A USB-C wall supply (bench, off the rover), the Pi runs ~3 minutes, the green LED flickers, holds well until about 7 minutes, then resets / flickers / goes blank and comes back. In the rover on battery power, the loop is faster, and the initial noise repeats continuously.

Issues

The converter runs absurdly hot. After running through power, card, image, peripheral, and tether checks, our biggest finding is that the 5V converter feeding the Pi runs hot enough to be a clear outlier. We are confident this is abnormal. We suspect this is the root cause of the boot loop.

What we’ve already done

SD card / BlueOS image

  • Re-downloaded the latest stable BlueOS image from the official source

  • Flashed the .zip directly (NOT extracted) with Balena Etcher; Etcher reported clean writes and passed verification

  • Tried multiple SD cards

  • Confirmed the card is fully seated

  • Replaced Pi

  • Replaced Navigator

Power

  • Tested with a proper 5V/3A USB-C wall supply on the bench (off the rover) — same behavior: runs for minutes, then resets

  • Confirmed the rover’s converter (suspected) runs visibly/tactually hot

  • Have NOT yet replaced the converter

Pi/hardware

  • Tried a known-good replacement Pi 4 — same boot-loop behavior

  • Replaced the old navigator

  • Inspected for corrosion/bent pins from prior condensation exposure (no obvious damage found)

LEDs observed

  • Pi green ACT LED, on the rover: regular repeating blink pattern (suggests boot files not read / partition issue), but resolved to power instability under load

  • Pi green ACT LED, on bench wall supply: blinks busily then goes solid → flickers → blank → returns (consistent with Pi resetting)

Any Help would be much appreciated (I attached 2 videos that could be of help)

Hi @moshi -

Welcome to the forums!

That is a very strange issue. I’d recommend unplugging any auxillary 5V loads, like your exploreHD and any camera servo or external servo being used, to make sure they aren’t contributing to your power problem.

Replacing the 5V supply also seems necessary, given the behavior of the LEDs.

Can you measure the resistance between the 5V input and ground on the Navigator?

Plugging a computer monitor in via hdmi adapter can also shed more light on what exactly is happening in the Pi startup process…

Powering via external 5V supply is a good idea to troubleshoot- doing so via USB is not the same and could cause issues as this bypasses the Navigator power handling…

Hey, Thanks so much for getting back to me. We ended up finding that the converter was the issue, along with the claw pulling way too much power. Do you have any advice for avoiding issues like this in the future?

Thanks so much for your help

Hi @moshi -

A multimeter can tell you both the output value of the converter, and how much current a payload is pulling (if properly configured / connected.) This is the best tool for troubleshooting power issues!