Also guys, been researching for the past 2 months, somehow i already got the gist, but what are the particular specific issues or notes i should know about? like...
-RST point is very sensitive so make sure to clean it properly without flux leaving behind and covering some tape...
-Adding additional resistors if it doesnt work...
-20k resistor for RST if it does not work..
anymore?
The mosfet -> APU soldering must be as good as possible, to reduce resistance and improve glitching. So 0.2mm diameter wires, good flux, low temp soldering paste pre-applied onto the caps
20k is not working anymore, that was for 1.x fw only
there are some consoles in the wild that give glitch error (cyan LED) but work fine after pressing reset on the rp2040, hopefully that will be fixed
no resistors on the eMMC lines can lead to data corruption
single mosfet is enough, it is 76A capable. but it's better to connect it to both capacitors
you can fit rp2040-zero without desoldering parts, but you must cut the cover then
in total now 5 boards are supported (ws2040-zero/one, xiao2040, itsybitsy, RP Pico), the waveshare is a default one, for others you must bridge detection pins
Unpatched v1 also works, but people keep asking that
with really low battery you can see the IPL screen, but further load may fail
putting .uf2 firmware file over USB is enough, no need to erase or reset it in any way
update through the toolbox is safe (unless there is a critical bug in the new firmware you are writing). it completes only after a successful boot with the new firmware. otherwise the old firmware starts on the next boot.
picofly automatically resets (and reflashes sdloader into the BOOT0) in case of system update detected, multiple unsuccessful boots, pico firmware update, and even putting into different console.
purple LED might be also caused by bad RST connection
long thin CMD/CLK/D0 wires may also cause "slow eMMC mode" and data corruption
despite of detection pins, all possible mosfet gate pins (6, 22, 15, 24) are pulled down to prevent accidental CPU burn by the high current
there is a possibility of unsupported eMMC, for example, it has never been tested on Micron ones
rp2040 consumes a bit of power after the glitch, now it's 0.6 mA, can't go lower, hardware limits.