I programmed the unit to give 20V output and connected it to a supply with a USB meter in between. It properly raised the output to 20V. I unplugged the fpx from the meter but since the meter was pulling power, it kept the USB rail at 20V. I then reconnected the fpx and it immediately smoked and the STUSB4500 chip burnt up. The main power draw occurred at Vdd (pin 24).

I am not sure why this happened since the chip is tolerant of 20V but maybe only when its expecting 20V will it be tolerant to it. Maybe the fact that the USB rail was 20V to begin with was unexpected although it should be able to handle that.

I had it working perfectly up till this point and it would program and output voltages properly.

Any ideas of what could have gone wrong? Have you seen anything like this before in your testing?

I was surprised there is no back-to-back MOSFET in the power path - I know this is meant to be a power sink, but if enough bulk capacitance is connected to it, someone might plug it into a port while the caps are still charged.

This is not really an issue.

Hi, I found you project digging the internet and I find it amazing and it suits perfectly for my project!
The saddest part is that I cannot buy anymore an assembled one from tindie.

My question is where I could buy one on sites like tindie?
I tried to make one, ordering the board from OSHPark and soldering every component by myself, but probably my soldering skills are still not that good!

https://au.element14.com/c/semiconductors-ics/power-management-ics-pmic/power-distribution-switches

https://www.digikey.com/products/en/integrated-circuits-ics/pmic-power-distribution-switches-load-drivers/726

https://www.digikey.com/product-detail/en/stmicroelectronics/VN7140ASTR/497-15994-6-ND/5431502

Example circuit

Hey!

I was looking through PD controllers and I saw the fpx and I thought this would be a great fit for my project.

I was wondering if it's possible to expose which profile was selected, e.g. through two GPIO pins.

For context, I have some high power portions of my design that has an enable pin that's controlled by a microcontroller and needs a PD controller to negotiate additional power from the USB. If we couldn't negotiate the necessary power, I'd keep these high power portions off and just keep the microcontroller running.

If two pins were exposed (11 for profile 1, 10 for profile 2, and 01 for fallback, for example), the microcontroller would easily be able to decide to power on the high power circuits.