The TX2 4GB module cannot be self-started on the self-developed carrier board

Hi all,
We found that some TX2 4GB modules cannot be self-started on our self-developed carrier board. Other TX2 4GB modules in the same batch can be self-started on the same carrier board.
And,we have three self-developed carrier boards for TX2 series modules. The three carrier boards only have different external interfaces, but the circuit part conforms to the design standards. The TX2 8GB module can be directly self-started, and only Only some TX2 4GB modules cannot be automatically booted on two of the carrier boards. I would like to ask why, the same batch of TX2 4GB modules will have different phenomena on the same carrier board? The phenomenon is that some modules cannot realize self-starting.

Regarding the TX2 4GB module issues, may I know which SW version you’re using?
There was a BOM components changes, see Jetson TX2 4GB PCN 207862 Addition of Manufacturing BOM Components: SK Hynix Memory and eMMC Memory

As the PCN mentioned, you may receive mixed shipments, so you have to move to Jetpack 4.6 / BSP 32.6.1 (or later releases of L4T will include the required change)

We used Jetpack 4.4.1/L4T32.4.4 ,

Is the self-starting of the module system related to the memory?

Not sure, maybe other cause, but suggest to move on JetPack 4.6.1 first to remove the PCN cause.

I’m sorry to reply you so late. This module can be started by pressing the button on the board, but there is no log output from the debugging serial port during startup. However, other modules on this board are capable of self-boot and log output. I’ll try 4.6.1 later.

This piece of TX2 4GB module also cannot be self-started using the jetpack4.6.1 version

Then for those problme module, suggest to contact with the distributor or try to do the RMA:
Jetson FAQ | NVIDIA Developer

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.