hardware vs software serial

[neilger]

I'm delighted to find pyupdi; I'm looking at options for making embedded interfaces. I've got it working through an FTDI FT230X, but not through the preferred solution of an ATSAMD11C running a USB-UART bridge. I'd like to review what pyupdi needs to be able to configure in the serial interface that's being missed by the simple byte bridge code I tested.

[ClundXIII]

@neilger I am sorry to hijack your Thread but what was your exact circuit? Does 4.7kOhms work as well or did you go higher or lower? Did you use the 3.3V as VCCIO?

[neilger]

Hijack away.

This D11 board:

http://academy.cba.mit.edu/classes/embedded_programming/D11C/hello.D11C.serial.3V3.png

works to do a serial echo:

http://academy.cba.mit.edu/classes/embedded_programming/D11C/hello.D11C.serial.3V3.mp4

to a t1614:

http://academy.cba.mit.edu/classes/embedded_programming/t1614/hello.t1614.echo.png

This FT230XS board:

http://academy.cba.mit.edu/classes/embedded_programming/FTDI/USB-FT230XS-UPDI.png

works to program the t1614:

http://academy.cba.mit.edu/classes/embedded_programming/t1614/hello.t1614.echo.mp4

as does this adapter with an FTDI cable:

http://academy.cba.mit.edu/classes/embedded_programming/UPDI/FTDI-UPDI.png

but not with the D11C bridge.

Looking through the pyupdi code I see it's dynamically reconfiguring the port; short of working through the UPDI protocol, my question is what the bridge code needs to support.

[mraardvark]

UPDI uses UART with even parity and 2 stop bits. Baud rate is not that important since the target will sync with it. In order to recover from an error state, a sequence of two break characters is sent. This is currently done by reconfiguring the UART to a lower baud and sending slow 0-characters.

[neilger]

Thanks, that's what I needed, I'll test.