Menu navigation problems

[LogiK]

I finally got around to finishing the new 32-inputs analog board, and in preparation to attach it I re-programmed the Atmega with the 32-input firmware. Since I was upgrading from an old version, I cleared out the EEPROM at startup, and I ended up with the VU.
OK, great! Everything was working well, so I went to play around in the menus. No more than a few seconds would pass before I would get pushed into one of the channel configuration screens, regardless of where I was in the menu at the time (channels 31, 27 and 29 popped up the most).
Less frequently -- and this seemed to happen most regularly when I was pushing or holding a combination of buttons (i.e. to get to the very beginning and end of the menu) -- the MIDI activity light would turn on and the menu would go crazy by quickly moving in random directions through the channel configurations, much like the keys had gotten stuck. Power cycling the board is the only way I found to get out of this condition.
I double-checked the keyboard, and it appears to be constructed properly. Each button also works just fine until the keys "stick".
Has anybody else seen this behavior?

[slayer666]

I finally got around to finishing the new 32-inputs analog board, and in preparation to attach it I re-programmed the Atmega with the 32-input firmware. Since I was upgrading from an old version, I cleared out the EEPROM at startup, and I ended up with the VU.
OK, great! Everything was working well, so I went to play around in the menus. No more than a few seconds would pass before I would get pushed into one of the channel configuration screens, regardless of where I was in the menu at the time (channels 31, 27 and 29 popped up the most).

I guess you don't have the Hihat pedal connected? When the hihat pedal is closed, the menu jumps to the channel recieving a hit (or noise). Connect a hihat pedal and set it to full open and it will not jump in the menu.

Less frequently -- and this seemed to happen most regularly when I was pushing or holding a combination of buttons (i.e. to get to the very beginning and end of the menu) -- the MIDI activity light would turn on and the menu would go crazy by quickly moving in random directions through the channel configurations, much like the keys had gotten stuck. Power cycling the board is the only way I found to get out of this condition.
I double-checked the keyboard, and it appears to be constructed properly. Each button also works just fine until the keys "stick".
Has anybody else seen this behavior?

No never seen this.

[Synthex]

I finally got around to finishing the new 32-inputs analog board, and in preparation to attach it I re-programmed the Atmega with the 32-input firmware. Since I was upgrading from an old version, I cleared out the EEPROM at startup, and I ended up with the VU.
OK, great! Everything was working well, so I went to play around in the menus. No more than a few seconds would pass before I would get pushed into one of the channel configuration screens, regardless of where I was in the menu at the time (channels 31, 27 and 29 popped up the most).
Less frequently -- and this seemed to happen most regularly when I was pushing or holding a combination of buttons (i.e. to get to the very beginning and end of the menu) -- the MIDI activity light would turn on and the menu would go crazy by quickly moving in random directions through the channel configurations, much like the keys had gotten stuck. Power cycling the board is the only way I found to get out of this condition.
I double-checked the keyboard, and it appears to be constructed properly. Each button also works just fine until the keys "stick".
Has anybody else seen this behavior?

What is your analog board ?
Do you touch the circuit with your fingers during tests ?
Have you installed the 4 pull-up resistors for the keyboard ?

[LogiK]

I guess you don't have the Hihat pedal connected? When the hihat pedal is closed, the menu jumps to the channel recieving a hit (or noise). Connect a hihat pedal and set it to full open and it will not jump in the menu.

No, the analog board isn't attached currently, which makes the fact that it's detecting anything on channels very strange indeed! I guess it's possible I have a noisy ground or something? In any case, I'll just go ahead and attach the analog board.

What is your analog board ?
Do you touch the circuit with your fingers during tests ?
Have you installed the 4 pull-up resistors for the keyboard ?

The analog board is 32-inputs using the four 74HC4851s. I'm pretty sure there aren't any shorts happening while I'm testing things... The board is sitting on a non-conductive surface and I'm keeping all exposed wires away from other wires/the board itself. I didn't yet put in the pull-up resistors, because it seems I was having a slightly different problem than described elsewhere. Though, the foot controller activity described there could explain my problem with the random channel configurations popping up.
If I were to put these resistors in, would they have to be exactly 47k, or would it be OK to use a bigger one -- say 56k? I don't have much of a collection yet, and I can only get close to that resistance using many smaller resistors in series (four or five for each pin... there's not enough room on the board!)