We have designed a custom carrier board for Jetson Orin NX/Orin Nano SOMs. The board is working fine with all interfaces up.
In our current design, We have 1x USB Type C port and 2x USB Type A ports (all direct USB2.0+USB3.0 in the 3 ports). However, we’ve received a new requirement to add an additional USB 2.0 lane to support Bluetooth functionality over the M.2 E-Key slot.
So for the next design revision, We are considering to use the USB2512B (USB 2.0) Hub to get additional USB2.0 ports. We needed some clarification on whether it is possible to map a USB2.0 downstream port from hub with a direct USB3.0 Superspeed lines from the Orin NX/Nano SOM. (please check the attached block diagram)
We have tried to implement this setup by rewiring the USB2 D+/D- signals from SOM to the upstream port of an external USB hub, and it’s downstream port signals to our USB Type A connector and M2 E connector. (same setup indicated in the block diagram)
We observed that the USB2.0 and USB3.0 functions work without any issues for the USB2 port. The M.2 port USB is also working fine and we are able to establish Bluetooth connectivity via M.2 module.
but we have observed something strange with the USB1 type A port. If we directly connect a USB3.0 device to USB1 port, it is not detected. but when we connect a USB2.0 device to USB2 port, the USB1 port detects the 3.0 device.
We have checked the device tree for the lane mapping for USB2.0 and USB3.0, there is no swap between the mapping of USB2.0 with USB3.0 between the lanes.
So with the current setup, USB2.0 is working fine in all 3 ports, USB0 and USB2 Superspeed lanes are also working. but only this particular USB1 Superspeed lane we are observing the issue.
Our understanding is, the individual USB ports operate without any dependency on adjacent USB ports. Could you please let us know your thoughts on this observation?
