Problem with USB type C controller/GPU/USB Mux

Hello,

I have been developing a carrier board very similar to the Xavier dev kit and have been having considerable problems with the USB C subcircuit. Essentially, I have no problems programming the Cypress 4226 chip itself using the PSoC programmer (programs, verifies flash and checksum) but whenever I power cycle my carrier board, it seems that my 3.0 device attached on the type C port will not enumerate again as 3.0 device (i.e. it reverts back to 2.10 device after being 3.0 before power cycling and switches USB buses) until I program the Cypress again.

I realize this isn’t directly related to the GPU or the dev kit board but my type C circuit is essentially exactly same as the dev kit with the exception of not using the 2nd port functionality. Any help would be appreciated as this is very confusing. I’ll be glad to further elaborate if necessary.

Hi,
It sounds like the signal is not stable. May need to do compliance test:
https://developer.nvidia.com/jetson-agx-xavier-series-tuning-and-compliance-guide-application-note

And check if you need to update firmware binary:

1 Like

Appreciate the help. Going to get a compliance tester and see how it goes.

I was actually curious; would nVidia be able to recommend to me the compliance tester (if any) that was used to validate the dev kit. We’d like to use the same one to test our design.

If not, any recommendation is welcome.

Thank you!

What does the compliance tester mean? All setup and test procedure are in guide: https://developer.nvidia.com/jetson-agx-xavier-series-tuning-and-compliance-guide-application-note

Something like this from Lecroy:

https://teledynelecroy.com/protocolanalyzer/usb/voyager-m310p?gclid=CjwKCAjwsan5BRAOEiwALzomX7W_qfDha731Ml7JN6z9jXUAikNVEUFn42D50CTY_JSBODyvdmXy7xoCX_QQAvD_BwE

I’m just going to assume you guys didn’t use a compliance tester then. Thanks for your help

No, we don’t use such tester.