STM32F0x2 USB and VDDIO2

[michele77]

Posted on January 05, 2016 at 09:58

On STM32F0x2 (USB line) PA11 and PA12 are used for USB. However, the datasheet states these are supplied by VDDIO2 which can go down to 1.62V. So one cannot use the USB functionality when using the STM32 for logic level translation (from VDDIO2 at 1.8V to VDD at 3.3V). However, one could use VDD at 2V and VDDIO2 at 3.3V I guess.. or I am overlooking something?

#stm32f0x2-vddio2-usb

[re.wolff9]

Posted on January 05, 2016 at 12:40Sounds like it should be possible. It would be interesting to hear back from you if it works. 

[Tesla DeLorean]

Posted on January 05, 2016 at 14:20

The GPS Guy?

I suspect the intent is to permit the use of the USB pin drivers within specification, which can't happen with the lower voltages to the IO drivers. I can't say I've looked at it in the F0 context, but it's more explicit in the L4 where you have VDDUSB. Here the expectation is that the device is battery operated, and when USB connected you can use that supply to power the interface.

http://www.st.com/web/en/resource/technical/document/datasheet/DM00108832.pdf

[Nesrine M_O]

Posted on January 05, 2016 at 14:24

Hi mbavaro,

Please refer to the USB electrical characteristics table in your STM32F0x2 datasheet, it is indicated that the minimal VDDIO2 value is 3V to ensure USB functionality.  -Syrine-

[michele77]

Posted on January 07, 2016 at 14:19

Yes.. designing a RaspberryPi shield with a 3G modem (UG95) and a RTK capable GPS module (S2525F8-RTK). So I was thinking to use the STM32F072 as USB composite to multiple UART. RPI and GPS are at 3.3V, but most modems are at 1.8V so having the VDDIO2 functionality decoupled from the USB would have been useful.

Thanks for mentioning the STM32L476, seems like indeed improves on that. Yes it's a little more expensive and not mainstream.. but it is not important here. I shall examine how time consuming is to adapt the design from F0x2 (USB crystal-less) to the L4.