Jetson Orin Nano 8GB Developer Kit stuck in APX mode after accidental drop - Flash always fails during blob transfer

Hello,

I have a Jetson Orin Nano 8GB Developer Kit.

The board was working perfectly before. It was booting Ubuntu from an NVMe SSD and I had successfully flashed this exact board multiple times using the same laptop.

Unfortunately, the board accidentally fell from approximately desk height while the power adapter was NOT connected. The NVMe SSD was installed at the time of the fall.

After that incident, the board has never booted normally again.

Current behavior

Every time I power the board:

  • The power LED turns on normally.

  • The fan spins for about 2-3 seconds.

  • The fan stops and repeats this cycle 2-3 times.

  • The board never boots Linux.

  • It is always detected by the host PC as 0955:7523 NVIDIA Corp. APX.

The Recovery jumper is NOT installed, yet the board always appears in APX mode.

The behavior is exactly the same whether the NVMe SSD is installed or completely removed.

Physical inspection

I carefully inspected the board after the fall.

I do not see any obvious physical damage such as:

  • PCB cracks

  • Broken components

  • Bent pins

  • Loose connectors

  • Visible damage around the USB-C connector

  • Visible damage around the M.2 connector

I have attached clear photos of the module and carrier board (top and bottom), including the USB-C port and M.2 slot.

photos are in this zip:

jetson images.zip (655.2 KB)

Troubleshooting already performed

I have already tried all of the following:

  • Multiple fresh Ubuntu Live USB sessions

  • Completely reinstalling SDK Manager

  • Rebuilding Linux_for_Tegra

  • Flashing with sdk Manager

  • Manual command-line flashing using sdk manager files

  • Flashing with the NVMe SSD installed

  • Removing all external USB devices

  • Removing every peripheral except power and USB-C

  • Different USB ports on the host PC

  • A different USB-C cable

  • Multiple recovery attempts

  • Multiple complete power cycles

Nothing changes.

Flashing behavior

SDK Manager correctly detects the board.

It successfully reads:

  • EEPROM

  • ECID

  • Board information

However, flashing always fails during blob transfer.

The SDK Manager log shows:

Sending bct_mem
Sending blob

ERROR: might be timeout in USB write.
Error: Return value 3

log file here:

NV_L4T_JETSON_LINUX_FLASH_COMP.log (429.2 KB)

Sometimes SDK Manager reaches approximately 99% and remains there indefinitely.

Other times it fails immediately after “Sending blob”.

During flashing, the device repeatedly disconnects and reconnects as APX.

Manual flashing

I also tried command-line flashing.

The process fails with:

Stat for blob_boot0.imgimg failed
Error: Return value 19

The generated command references:

kernel boot0.imgimg

instead of

kernel boot0.img

which looks incorrect.

Power supply

I am using the original NVIDIA-recommended power supply for the Jetson Orin Nano Developer Kit.

Also this is result of lsusb command when I got when recovery jumper was not installed

ubuntu@ubuntu:~$ lsusb
Bus 004 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub
Bus 003 Device 014: ID 0955:7523 NVIDIA Corp. APX
Bus 003 Device 003: ID 04f2:b7ba Chicony Electronics Co., Ltd Integrated Camera
Bus 003 Device 005: ID 8087:0aaa Intel Corp.
Bus 003 Device 002: ID 346d:5678 USB Disk 2.0
Bus 003 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 002 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub
Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub

My question

Since this problem started immediately after the accidental drop, could this indicate hardware damage?

Could a damaged USB-C interface, carrier board, module connector, QSPI, RAM, PMIC, or another hardware component allow the board to remain permanently in APX mode while consistently failing during blob transfer? (It didn’t fall very hard.)

Any advice or additional diagnostic steps would be greatly appreciated.

Thank you.

jetson images.zip (655.2 KB)

Hello @beym75121!

Based on the title and content of your topic, it looks like it may receive better visibility and feedback in a different category. We took the liberty of moving it for you.

If this was an incorrect assessment, please send me a direct message.

Disclaimer: this moderation suggestion and message were generated with AI assistance.

Hi @beym75121,

Thanks for the detailed information and the log.

From the attached SDK Manager log, the host can still communicate with the module in RCM: it reads the ECID, EEPROM, board information, and identifies the board as Jetson Orin Nano 8GB Developer Kit / jetson-orin-nano-devkit-super on JetPack 6.2.2 / Jetson Linux R36.5. The failure happens later during the RCM boot stage:

Sending bct_mem
Sending blob
ERROR: might be timeout in USB write.
Error: Return value 3

Also, in the SDK Manager log you attached, the generated command uses kernel boot0.img, not kernel boot0.imgimg. So the boot0.imgimg error from the manual attempt looks separate from the SDK Manager blob-transfer failure.

The important point is that 0955:7523 means the Jetson Orin Nano module is being detected in Force Recovery Mode. According to the Jetson Orin Nano Developer Kit User Guide, Force Recovery Mode is normally entered by connecting Button Header pins 9 and 10 while powering on or resetting the board.

Since you see 0955:7523 even when the recovery jumper is not installed, and this started immediately after the fall, hardware damage is possible. It could be related to the recovery strap path, USB-C data path, module-to-carrier connection, or power/PMIC behavior. The photos do not show obvious gross damage, but photos cannot rule out connector damage, cracked solder joints, or a marginal power/USB path.

Recommendation:

  1. Power off the board and disconnect NVMe and all peripherals.
  2. Check the Button Header area carefully, especially pins 9 and 10. Make sure there is no jumper, debris, bent pin, or accidental short between those pins.
  3. With power removed, if you have a multimeter, check whether pins 9 and 10 are shorted when no jumper is installed. They should not be shorted in normal use.
  4. Reseat the Jetson module in the SODIMM connector and inspect the connector/latch area.
  5. Power the board with only the DC adapter and USB-C connected to the host, then check lsusb again.

If it still shows 0955:7523 with no recovery jumper installed, I would treat this as a likely board/module hardware issue rather than continuing to reinstall SDK Manager or retry the same flash procedure.

For reference:

Thanks.

Thank you for your reply and for the suggestions.

I completed the additional checks you requested and here are the results:

  • I tested a different USB-C cable.

  • I tested different USB ports on the host PC.

  • I also tested with a different NVMe SSD.

  • I tested with the original NVMe removed completely.

  • I disconnected all external peripherals.

  • I used multiple fresh Ubuntu Live USB sessions.

  • I completely removed and reinstalled SDK Manager and JetPack.

I inspected the carrier board and the module carefully. I do not see any visible physical damage, cracked components, bent pins, or damaged connectors.

I also checked the Force Recovery pins (FC_REC and GND) with a multimeter while the board was completely powered off and with no recovery jumper installed. There is no continuity (no short circuit) between FC_REC and GND.

I removed the module, inspected the SODIMM connector and the module contacts, then reinstalled the module. Everything appears normal visually.

Unfortunately, the behavior is still exactly the same.

The board always enters APX mode even though the recovery jumper is NOT installed.

The fan spins for about 2-3 seconds, stops, repeats a few times, and the board never boots normally.

SDK Manager always detects the board correctly, reads the EEPROM and ECID successfully, but flashing always fails at exactly the same point:

Sending bct_mem
Sending blob
ERROR: might be timeout in USB write.
Error: Return value 3

I also tested with another SSD and obtained exactly the same result.

Since the problem started immediately after the accidental drop, do you think this behavior is more consistent with a hardware failure of the module or the carrier board?

Is there any additional diagnostic procedure (for example UART logs or other hardware tests) that could help determine whether the module or the carrier board is damaged before replacing any hardware?

Thank you very much for your help.

Hi @beym75121,

Thanks for checking and confirming the results.

Based on the updated tests, this is now much more consistent with a hardware failure than a software or SDK Manager issue. The key points are:

  1. The board still enters APX / Force Recovery Mode even with no recovery jumper installed.
  2. FC_REC and GND are not shorted.
  3. Reseating the module, removing NVMe/peripherals, trying other USB-C cables/ports/SSD, and reinstalling SDK Manager / JetPack did not change the behavior.
  4. RCM communication starts, but the transfer still fails at the same point:
Sending bct_mem
Sending blob
ERROR: might be timeout in USB write.
Error: Return value 3

At this point, I do not think additional SDK Manager reinstallations or repeated flash attempts will recover the board. A UART log might show where boot stops if the board outputs anything, but it would not reliably identify whether the module or the carrier board is damaged.

Recommendation:

Please treat this as a hardware failure. The practical way to isolate module vs. carrier board is to test this module on a known-good compatible carrier board, or test a known-good module on this carrier board. If that is not available, please proceed with hardware service or replacement.

Thanks.