If the oscillator doesn’t work stable, it could be that not the complete code (MIOS and/or application) has been loaded into the PIC. And this can result to such unexplainable effects - a lot of random instructions are executed.
So, first thing which has to be solved is to get the crystal running. Unfortunately I cannot give you that much tips here - does anybody use the same crystal and caps from Mouser?
Best Regards, Thorsten.
Ihave used the same parts with succes.
Ihope you get it figured out.
By the way i have been working on my sid for over a year, I have one good core/sid combo, and recently converted it to an 8580 and optimized psu.
There are six points together on the core, the two each for the two caps and the crystal. If i touch those six with my finger the flood of midiout messages stops, then if i let go it gives the correct message then soon after it goes back to the flood, very strange.
How long it takes to “midi flood” to start? (I’m thinking of midibox boot process here)
What should your PIC contain? Bootstrap loader? (+MIOS?) (+MIOS & Application? (SID application?))
Did you actually _fully_ erase PIC before burning bootstrap loader again?
I’d say, that shorting XTAL caps to ground or touching pins around there causes PIC to reset. Then it seems to start normal boot process (“the correct message”).
“Random” midi flood is known to be caused by “not properly” terminated input pins.. with SID application that would probably mean R9 missing (or not properly connected) from core board.
If you’re thinking that XTAL/caps are causing the trouble, try reflowing solder joints connected caps, crystal and pic pins where crystal is connected.
..Bye, Moebius
So, first thing which has to be solved is to get the crystal running. Unfortunately I cannot give you that much tips here - does anybody use the same crystal and caps from Mouser?
Many, many times over - both the Vishay xtal and Xicon caps from Mouser. I built a working core from the parts on the list before I released the list, and the Vishay xtal has shipped with around half of all of the core kits I have sold. The others I use (abracon) are just as good, and perform slightly better in extreme high EMI/ interference environments (the only good way to test a crystal,in the shadow of one of USA’s most powerfull AM transmitters!) It’s purely an availability issue that dictates which I use of the two manufacturers…
If you’re thinking that XTAL/caps are causing the trouble, try reflowing solder joints connected caps, crystal and pic pins where crystal is connected.
Absolutely. Reflow not just look. Is there soldering flux around these connections? I have seen flux conduct enough to cause high resistance short and keep the crystal from starting or other unusual behavior.
Remember the caps are really only there to start the xtal oscillating, and don’t really help things after it does.
If these caps are too far different in value either naturally (from their 5% tolerance) or if one has changed drastically due to soldering overheat the xtal won’t act right.
Also if the xtal was overheated during soldering it can act like this, but they usually just die completely instead.
As always, I’ll burn the bootloader for anyone anytime, for free, even if you don’t buy the pic or anything else from me.
Also describe your power supply please, I helped a builder with xtal issues like this that completely went away when he changed power supplies, he was trying to use a switch-mode supply from and old computer, and it apparently had enough AC ripple on the 5V line to mess things up.
BTW: I haven’t forgot about the issue that got you to this point (Mac MIOS upload), I’ll help you get that going after the core is happy. 
Best!
Smash
I will be trying out these suggestions as soon as i have time hopefully today, thank you for all of your great suggestions I will let you know. - kBud