STM32F746G Discovery board as ST-link programmer for external STM32 MCUs

[petr2]

Hi everybody,

does anybody experience at using ST-link in STM32F746G Discovery board for external microcontroller?

There is a ST-link connector (CN8) on the board with 4 signals - 3V3, CLK, GND, DIO. Problem is that there is not any possibility to disconnect the ST-link from main M7 MCU on the board and therefore programming data from ST-link would go to both MCUs which would interract together on DIO.

My idea is to turn off power supply to M7 MCU via JP and then communication will be running only between ST-link and external MCU but I am afraid of damaging of main M7 MCU because its Vdd = 0V and DIO = CLK = 3,3V (from ST-link) = Vdd + 3,3V.

At absolute maximum ratings I have found that Vin max on 5V tolerant pins is Vdd+4.0V and for other pins just 4.0V so from this point of view it should be OK for 3,3V from ST-link but before that I would like to see your experince/opinions. I really do not want to damage my new discovery board.

Thank you for your answers.

Regards

Peter J.

[S.Ma]

A nucleo board allows separation of target user mcu from on board stlink. With discovery, it was not. Nucleo or stlinkv2 would be better bet.

[Tesla DeLorean]

And is a $10 board vs $56 one. Even cheaper ST-LINK clones on eBay

[Andreas Bolsch]

I'd strongly support the suggestion regrading nucleo or separate ST-Link.

AND: The signals on CN8 are SWD access to the embedded ST-Link's STM32F103, *NOT* to the target!

There are some solder bridges SB14-17 by which the target's SWD pins could be isolated from the embedded ST-Link, and then one could attach wires to those pads, but that's far from being a good idea ...

[petr2]

Thank you for your answer. I will probably buy some Nucleo.

[MikeDB]

I am sure some of the early Discovery boards allowed the programmer to be broken off. The F407 one in front of me doesn't but I think it was one of the M0 processor ones that did.