Digital FX troubleshooting

i soldered a piece of wire to the pin, as i couldn’t reach the pad easily (lcd)
i’ve swapped the avr, uploaded new firmware, and still the same result…
from what i’ve seen on the scope i was sure it’s the pin, maybe there’s more than one issue, maybe more than one avr with fried TX pin 17.
both channels were looking the same though.

there was also no change related to whether the green LED was on or off.

i’ll try another AVR later, and if that doesn’t change anything, i’ll record those scope screens :smiley:

so it should actually look like the midi signal…
that was definitely different. i had to go to the smallest range and it was a jumping distorted triangle

ok i managed to dig through my avrdude flashing notes (have to re-learn that procedure everytime i do it)
and flashed firmware 0.93 with bootloader on a brand new AVR.

still the same, audio just from the 328’s input.

i have one question about the AVR type. avrdude always gives the following message so i have to end every command with -F

avrdude: Device signature = 0x1e9609
avrdude: Expected signature for ATMEGA644P is 1E 96 0A

maybe that’s related to that problem, but i guess no.

at first i though well, maybe the nbr is different because the avr is from pollins, but the reichelt one and also the one i originally purchased preprogrammed in november all show the same signature complaint.

scope time again, i guess?

could there be something else wrong on that particular ctrl board that would cause a malfunction of the AVR TX pin, maybe some bad soldering somewhere else?

on the one hand i think it’s strange that all my AVR’s should be damaged somehow, on the other i wasn’t getting a good signal on the scope (directly from the pin).
i’ll try it with a different ctrl board next, if that doesn’t help too, you’ll probably see me talking to the scope soon ;D

Strange, the signature (1e9609) is that of an ATMega644, not an ATMega644p.

oh so i have the wrong AVR !


stupid stupid…
so i have 2x ATMega644 and 1x ATMega644p, which should be the one from you.
but that didn’t work too…

crap, i’ll need to look closer at part nbrs it seems

I had a look at the differences between the 644 and the 644P and here it is: the 644 has only one UART! The link with the digital FX board is done through the 644P’s second UART, so this could very well be the cause of the problem.

Are you sure it doesn’t work with the 644P too?

yep just tried again.
i’ve now soldered a small wire in the via that links the trace from both sides of the pcb (near the pot) so i can scope those guys in a more comfortable way.
but no time atm, will do that later.

so does that mean i could use the avrs i have for say the cem3379 or smr4? or would you recommend getting the right part after all?

I haven’t dug into the datasheet in details but it looks like the only new feature of the 644p I use is the second UART for talking to the digital FX board. So it should be fine to keep those for the simpler filter boards.