Omega 2S+ VDD_FLASH pin
On the Omega 2S+ can I connect the VDD_FLASH pin directly to 3V3? Or is the series diode and 50 ohm resistor required? I don't need hardware reset.
@SimonPR If the datasheet specifies a configuration, it is best to follow it
In both configurations they add the 50 ohm resistance, I would think that it is limiting the current that reaches the VDD flash pin, but it may have other critical reasons related to the correct functioning of the Omega
crispyoz last edited by
@crispyoz @jossval Yes I totally agree re following the recommendations, and it's what I'll do for now.
Was hoping to get a bit more detail on the exact implications of just connecting it to 3V3 (eg backflow and excess supply current risks).
I just really don't want to add more components if I can help it!
crispyoz last edited by
@SimonPR this would be a question for one of the Onion engineers, however the Onion team are pretty fastitious engineers, so if they say you need it, then you need it. I prefer to keep the puff of blue smoke inside my devices rather than release them
luz last edited by luz
You can directly connect VDD_FLASH to +3.3V. I have >1000 24/7 working devices in the field that are wired like this, so I'm quite confident
The reason for having that pin separate is that if you find yourself in a situation where you need to directly re-flash the SPI flash chip (e.g. after a failed attempt to update uboot), you can apply power via VDD_FLASH to the SPI flash chip without powering the SoC (MT7688).
This allows to re-program the flash using an in-circuit SPI flash programmer. The diode allows this without changing anything on the board (no jumpers required). I guess the 50Ω resistor is there for noise reduction.
If you wire VDD_FLASH to 3.3V directly, there's no way (except cutting traces on the PCB or unsoldering) to get a ruined uboot re-flashed. For my devices, I decided I will never touch uboot once they are produced…
The HW reset thing is another story - this extra circuit is needed to make the SPI flash reset (by interrupting its power supply during HW_RST) to ensure the address mode is right for the MT7688's startup.
If I recall the details correctly, address mode needs to be 3-byte during boot, but the firmware might later switch it to 4-byte. If then a reset occurs, the SPI chip is in 4-byte mode and the MT7688 tries to boot with 3-byte, which fails and leaves the device inoperable until power-cycled.