USB VBUS_SENSE on STM32U5A9NJ Discovery Kit

[jcmoreau]

I am experimenting with the USB HS peripheral on the STM32U5A9NJ Discovery Kit, and noticed that there is a signal called VBUS_SENSE which is routed to PG1 (ADC4 IN8). I've configured ADC4 to sample from this pin, and notice that the sensed voltage is ~0.14V with the USB connected, and 0.0 with the USB disconnected. Is this the expected behavior? I would have expected to see a higher voltage with the USB cable connected. What is the maximum possible voltage for VBUS_SENSE? My ADC resolution is set to 12 bits, sampling time is 814.5, and Vref+ is 1.8.

[Mike_ST · Accepted solution]

Hello,

>> Is this the expected behavior ?

>> What is the maximum possible voltage for VBUS_SENSE?

Expect around 0.38V when USB is connected.

Try to set PE9 as output to 1 (DBn signal), so the TCPP01 will trigger the Q1 transistor.

Check if you have 5V on the 5V_UCPD net (JP4 Pin 5).

[jcmoreau]

Hello Mike,

Thanks for your answer. I measure 5.1V on JP4 Pin 5 using a DMM. I still see the same ~0.14V on PG1 with the USBC cable connected when I set PE9 high. Just to make sure I had ADC4 configured correctly, I verified that I was able to accurately measure the 1V8 output by connecting it to PA4 on CN9. Is there anything else which may be causing this discrepancy? 

[Mike_ST]

Hello,

I attached a .ioc file and a main.c file  to open with the STM32CubeIDE.

Please compile the project, and check whether you get a value close to the expected one.

If it's working, it can help to compare the settings.

[jcmoreau]

I wasn't able to see any output from the program defined by the .ioc, looks like maybe the UART pins weren't configured. 

However, when I use the voltage divider values implied by R1 and R2 in main.c, I get readings much closer to 1.8V with the cable connected, and 0 with it disconnected. For reference, my (new) formula for calculating voltage using raw ADC readings is 

result = (1.8f / 4095) * raw_result * ((27 + 330) / 27);

Maybe there's another divider somewhere upstream.

[jcmoreau]

My ADC4 was not configured correctly, I missed this note in the data sheet:

Note:
The ADC4 analog block clock frequency must be between 140 kHz and 55 MHz.

After correcting this (applied 8x prescaler in ADC4 CCR), I read ~3.7V (651 / 4095 = ~0.286V) with the USB connected.

Edit 2: Found a second error in my ADC4 configuration which caused my sampling period to be set to 1.5 ADC cycles instead of 814.5. After correcting this I am reading 5V (after adjusting for voltage divider).