I have the DSO150 Shell kit scope, which uses the STM32F103C8 processor at 3.3V. JYE-Tech provides a way to update the firmware, but it requires opening the case, connecting leads from an external USB-to-Serial module to the GND, RX and TX through-holes, and solder-bridging Boot0 (to 3.3V) and Boot1(to GND). Then after flashing you have to undo all of that. We're on about the sixth update now, and this is getting old.
I would like to embed a CP2102 USB-to-Serial module in the existing circuit so all I would need to do to flash firmware is connect a USB cable. The GND, TX and RX connections are straightforward, and it appears that Boot1 can be tied to ground and left that way during normal operation.
That leaves Boot0, which needs to be high on boot to enter the bootstrap mode. I had thought I could tie Boot0 to the 3.3V output of the module. So I could plug in the live USB cable, then power up the scope. But initially that would mean that since Vdd is at ground when the scope power is off, I would be applying a low-impedance Vdd + 3.3V to the Boot0 pin.
I can't tell from the datasheet whether doing it this way would cause too much current to flow into the pin - either through the protection diode or otherwise. Boot0 appears to be a 5V tolerant pin, but I assume that's with Vdd at 3.3V. So I don't know.
Does anyone know for sure if this would work? If not, how would I go about testing it non-destructively?
I should say that another approach would be to use the module's 3.3V output to power the ST, and leave the scope power switched off. Then everything would come up at the same time. But I don't know if the module can power all of the things powered from the 3.3V rail. Also, these modules don't behave well when they are powered down (when USB isn't connected). They can sink a good bit of current. Of course I could insert a diode or p-channel mosfet to prevent that, but that just complicates things more. So the original plan would be best if it won't damage anything.