Problem booting the VIM4 - stuck in boot loop with purple OOWOW screen

Which system do you use? Android, Ubuntu, OOWOW or others?

Ubuntu (eMMC install). Also tested OOWOW recovery and a freshly written Ubuntu SD card during troubleshooting.

Which version of system do you use? Please provide the version of the system here:

vim4-ubuntu-22.04-server-linux-5.15-fenix-1.6.9-240618-emmc (this is also the image I used when writing a fresh SD card for testing)

Please describe your issue below:

Board received as a gift, no purchase receipt available, so I can’t file a warranty claim and I’m trying to diagnose this myself.

The board does not complete booting. The sequence repeats continuously:

  1. Khadas boot logo appears
  2. Screen changes to the purple OOWOW splash screen
  3. Screen goes black
  4. Fan spins up briefly, then stops
  5. LED cycles on/off
  6. Back to step 1, repeats indefinitely, never reaches a usable OS

This happens whether booting from eMMC (originally had Ubuntu 22.04 installed and working before this started) or from a freshly written SD card with the verified image above, so eMMC corruption alone doesn’t explain it.

What I’ve already ruled out:

  • Power supply: tested with two different USB-C PD chargers, an Anker 317 and a Belkin 65W multi port charger (single port, its own cable). Same failure pattern on both.
  • Peripherals: tested with everything disconnected except power and display (no keyboard, mouse, SD card, USB devices). Same failure pattern.
  • SD card boot: wrote the fresh image above to a separate, known good SD card and booted from it directly, bypassing eMMC entirely. Same failure pattern.
  • Display cable: confirmed correct cable orientation (full size HDMI into VIM4, HDMI to mini HDMI adapter cable into monitor’s mini HDMI port), verified working with another source first.

What I’ve tried for recovery:

  • OOWOW recovery (Function button, held while plugging in power): reaches the purple OOWOW splash screen but crashes/resets before fully loading, every time.
  • USB burn mode via aml-flash-tool on Linux:
    • TST mode (3x rapid Function button presses within 2 seconds): LED blinks red as expected, but the board falls back into the normal boot loop instead of staying in Upgrade Mode. lsusb never shows an Amlogic device (vendor ID 1b8e).
    • Key mode (hold Power, tap Reset, release Power after 2 to 3 seconds): same result, no Amlogic device detected via lsusb, board continues looping.
    • Tried with the USB cable connected throughout (not unplugging/replugging) to avoid racing the board’s auto boot on power detect behavior. Still no success entering a detectable USB burn mode.

Given that the identical failure occurs regardless of power supply, storage (eMMC vs fresh SD), and peripherals, and that I can’t get the board into a detectable USB burn/maskrom mode via the documented button sequences, I suspect a hardware fault (SoC, PMIC/power sequencing, or RAM) rather than a software or image corruption issue, but I’d welcome being wrong about that.

I’m also looking for guidance on getting a UART/serial console connected (debug header location and correct baud rate) so I can capture an actual boot log instead of guessing from LED/fan behavior, since I don’t have one yet (see below).

Happy to provide photos of the board or any other test results, just let me know what would help.

Post a console log of your issue below:

No UART/serial console log available yet. I don't currently have a confirmed debug header location or baud rate for the VIM4, so I haven't been able to capture boot output. If someone can point me to the correct UART pins and settings, I'll capture and post a full log.

Please use latest 1.7.5-250925 version to check again.