PowerPak + CV3 (jpn) + audio mod = intermittent cycle reset?

This is an archive of a topic from NESdev BBS, taken in mid-October 2019 before a server upgrade.
View original topic
PowerPak + CV3 (jpn) + audio mod = intermittent cycle reset?
by on (#60012)
I just got a PowerPak a couple days ago and decided to do the audio mod on my NES so I could hear VRC6 audio in Castlevania 3 on my TV.

Everything works fine, graphics and audio, but after about 15 minutes of playing Castlevania 3 (or leaving it doing the title demo for about 15 minutes), my NES will start to reset over and over again like it has bad contacts (the contacts are brand new and work great). Everytime it resets, the game is starting up, as I can hear the first note or two of the intro, and see the graphics start to slide onto the screen, but then it resets and repeats. If I press reset on the console, the game resets and works like a charm... then 15 minutes later it does the same thing again.

I ran the same game, CV3 (jpn), on an untouched NES for around 30 minutes and, with it not playing the extra sound channels and sounding like crap, it never did the cycle reset.

I put maybe an hour or two into playing Dragon Warrior IV when I first got the PowerPak and had no issues at all, nor when my nephews were playing SMB for quite a while. I played Dragon Warrior IV for about 20 minutes just a while ago and it did the cycle reset thing. Right now the modified NES has been running a game of Anticipation with 3 computers for about 45 minutes and has not reset.

I've double and triple checked the soldering job on the 47k ohm resistor and it's a clean install and there are no extra parts in contact with each other. I'm considering putting the mod on the untouched NES just to see what happens.

So just to recap:
    - NES works fine with all games, played Dragon Warrior 4 for a couple hours (did not test CV3 before this)
    - Decided to mod NES for sound for VRC6 games
    - Castlevania 3 causes NES to cycle reset after 15 minutes or so
    - Pressing reset during this causes the problem to stop, but of course I lose my position in the game
    - Dragon Warrior 4 suddenly does the same thing after 20 minutes or so (only tried once)
    - Tried CV3 on untouched NES and the intro played for 30 minutes with no issue
    - Ran Anticipation game for over 45 minutes on modified NES with no issue


Has anyone else had an issue like this or maybe know what I could look at or try? I've exhausted the extent of my troubleshooting skills with this kind of stuff.

Oh, and on a second note, does the PowerPak not work is Madara or Esper Dream II? With both games, PowerPak reports it cannot find the files when trying to load (ran chkdsk, files system sorter, etc, too). These are the only games that have that the file not found issue.

Thanks for the help. :)
Re: PowerPak + CV3 (jpn) + audio mod = intermittent cycle re
by on (#60015)
beigemore wrote:
Has anyone else had an issue like this or maybe know what I could look at or try? I've exhausted the extent of my troubleshooting skills with this kind of stuff.

I can't think of an exact reason if you can consistently make this happen after 15 minutes. But if you want to get rid of the problem and don't care to find the cause, disable the lockout chip in your NES (it's easier than doing the audio mod, search the forums).
beigemore wrote:
Oh, and on a second note, does the PowerPak not work is Madara or Esper Dream II? With both games, PowerPak reports it cannot find the files when trying to load (ran chkdsk, files system sorter, etc, too). These are the only games that have that the file not found issue.

"File not found" most probably means it didn't find the MAPPER file. Madara uses a different iNES mapper code than Akumajou Densetsu because it has some address lines swapped. Madara is iNES #26. So if you don't have a file named MAP1A on your CF it's not going to work. I'm not sure if it's included in the official mapper distribution or not.

by on (#60025)
Thanks. Disabling the lockout chip did the trick.

by on (#60041)
Yeah this was probably a problem with the CIClone I think. Maybe it got out of sync after a while and the data it sent to the lock CIC didn't match what it was supposed to be.

by on (#60052)
I played Battle Kid for hours on end, so I'm surprised that the ciclone failed.

by on (#60054)
Dwedit wrote:
I played Battle Kid for hours on end, so I'm surprised that the ciclone failed.

It's very, very, VERY unlikely that CIClone was the culprit here.

by on (#60057)
I don't see any other way it can be explained.

by on (#60058)
Drifty lock CIC oscillator?

by on (#60060)
Yeah clock drift is what it sounds like. For something where a high reliability is needed (a game cart doesn't first come to mind, think safety/medical/industrial systems) a burn-in test would be done that would catch problems that only show up after it's been running for a while.

Maybe not in this PIC, but I know in some other PICs there is a register to somehow tune the internal oscillator.

by on (#60072)
Isn't the CIClone clocked by lines from the lock CIC ? Yes if I remember well PICs have to be tuned with their internal oscillator, which I belive is unused for the CIClone (but I might be wrong).

by on (#60077)
Yeah you're right, it does take the clock from the NES edge connector.