Hi,
I ve been using the MightyCore on a home designed 1284p platform for sometime now with very good results operating at 5V, 16MHz external Crystal.
Is there any reason why the same setup wont work if I just lower the supply voltage to 3V3?
BOD detection is disabled.
I tried to do exactly that by powering the system from a dual voltage (5V/3V3) FTDI module but the terminal no longer connects.
Any ideas?
This chart mentions a Nano clone (328p) but I believe same chart applies to all AVR chips including 1284p
Having said that, many forum members have reported AVR chips working well at 16MHz and 3.3V, even though it's not guaranteed to work. So it could be something else.
I'm guessing that the 5V comes from the USB 5V. Where does the 3.3V come from?
From memory there's a 3.3V output from the FTDI chip but it's rated from something small like 50mA.
I'm guessing that the 5V comes from the USB 5V. Where does the 3.3V come from?
From memory there's a 3.3V output from the FTDI chip but it's rated from something small like 50mA.
Yes the 5V comes from the USB. The FTDI module lowers the 5V to 3V3.
Not sure if that comes from the FTDI chip itself or from an external regulator.
But even if it comes from the FTDI chip, 50mA is enough for the MPU. I didnt detect any low voltage issues.
This chart mentions a Nano clone (328p) but I believe same chart applies to all AVR chips including 1284p
Yes, I am aware of this chart.
Previous experimentation with other MPUs 328, 32U4 indicates that they all worked fine at 3V3 even at 16MHz ( i know its out of spec) . In fact, when they didn't work it was because of the type bootloader burned, fuses etc
You asked for help and we are trying to suggest some things to consider. If there are things you have already tried or considered, but didn't mention them in your question, please mention them now so we don't waste time on them.
Understood.
I haven't tried changing the crystal to 8MHz in this platform, but will do and see if it makes any difference.
What I said was that from past experience with other MPUs it didn't affect operation,
Do you have some kind of variable voltage power supply, like a bench PSU or a variable linear regulator or buck converter? You could try gradually decreasing the voltage down from 5V to find what voltage it stops responding.
Yes. Will try that. Thanks
Just tested it with a variable PSU. MPU seems to stop working at 3.45V.
ie the connection with the terminal is lost and if the voltage is raised again by 0.5V, it reboots.
Then perhaps you have been unlucky this time and your particular 1284p won't run at 16MHz on 3.3V.
You could try setting the fuses to run at 8MHz using the internal oscillator (ignoring the external crystal) to see if it will run at 3.3V like that.
If that works, it might be possible to run the 1284p at 8MHz using the external crystal by setting the master clock divider to halve the main clock speed, so at least you get the benefit of the more stable and accurate crystal. I'm not sure if that's even possible with MightyCore without messing up millis(), delay() etc.
Did we check the BODLEVEL fuses that control the brown-out detector? The BOD can be set to generate a reset when Vcc drops below a programmable level. I don't know what the settings are in the MightyCore, or if they are in any way unusual. Reading out the fuses with AVRDude might be worth doing.
