I've heard of this before. I don't know why this is critical, but apparently it is.
On my boards, I've run wires between these pads and the backplane ground.
Color Classic motherboard woes
- Thread starter This Does Not Compute
- Start date
I've heard of this before. I don't know why this is critical, but apparently it is.
On my boards, I've run wires between these pads and the backplane ground.
ISTR that the thing about the grounding springs stems from a video by Adrian's Digital Basement in which he tried running an LC575 board in a Color Classic chassis without the RF shielding, resulting in audio issues. At any rate the springs are not critical on the original CC motherboard. My CC has been running without a hitch despite not having any of the springs; the board already ties them all together with no modifications necessary. Worst case scenario, I'd imagine you might experience some incredibly minor banding/interference on screen without the springs/shield. Any other problems that may appear to be solved by fiddling with the springs on an original CC board are almost certainly a fluke.
As far as I can tell the gray screen is in some way related to the DFAC IC, still not sure exactly how though. Broken connections to the pull-up resistors on the I2C lines between DFAC/CUDA and broken connections between CUDA/Spice both tend to cause the system to immediately death chime on power-up, whereas bigger CUDA problems usually cause the system to not power on at all. But removing the DFAC entirely reproduces the same gray screen symptom, at least on my two boards.
As far as I can tell the gray screen is in some way related to the DFAC IC, still not sure exactly how though. Broken connections to the pull-up resistors on the I2C lines between DFAC/CUDA and broken connections between CUDA/Spice both tend to cause the system to immediately death chime on power-up, whereas bigger CUDA problems usually cause the system to not power on at all. But removing the DFAC entirely reproduces the same gray screen symptom, at least on my two boards.
Last edited:
I know I have certainly seen the "no audio" issue (that Adrian also hit in that video) on the CC before when those grounding springs were making intermittent contact with the grounding shield. In that case a bit of a wriggle of the board seemed to fix it and the audio returned.
On the board last year when I was having the grey screen issue at that time, I also tried linking all those springs together via wires as I was seeing inconsistent grounding between them when testing with a multi-meter. It didn't help the grey screen problem though. :-(
I will pull my board out tonight and go have a looksy at that DFAC chip and lines to the CUDA on it as it's not one I've looked at before.
On the board last year when I was having the grey screen issue at that time, I also tried linking all those springs together via wires as I was seeing inconsistent grounding between them when testing with a multi-meter. It didn't help the grey screen problem though. :-(
I will pull my board out tonight and go have a looksy at that DFAC chip and lines to the CUDA on it as it's not one I've looked at before.
Got some time last night to try this out. I'm getting 4.85V from RST on both DFAC and CUDA (pins 24 and 15, respectively) when the machine is powered on. Nothing happens when I try an NMI though.
EDIT: Forgot to mention that I tried swapping the ROM chips between the two boards, and that didn't make a difference.
Last edited:
I have the same issue with a mac classic 2 board . Consensus is that it may be bad rom . But I am checking my recap to make sure.
Being able to rule out the ROMs too is nice. Last thing I'd try is replacing the 74LS245 at U10. The first things that come to mind for why you wouldn't hear death chimes after an NMI at this stage are a.) the CPU can't write to the Spice IC, where sound is generated before being filtered by DFAC, because the buffer between CPU/Spice (U10) is bad, or b.) the DFAC really is dead.
Only reason I'd check U10 first is because 'LS245s can still be ordered new for $1 from Mouser/Digikey/etc, easier than getting another DFAC. That being said, when I had open traces going to U10 on one of my CC boards (due to a battery bomb), it resulted in a black screen, not a gray one. Past that, I probably would look into finding a new DFAC from a scrap board or from UTSource.
So to be clear, by "grey screen" this is what I'm getting -- this is in a CC chassis with an analog board that I know needs to be recapped, and with a good board installed I still get flickering/changing levels of brightness. In its own chassis the display stays blank (but with a good board installed, the display looks perfect).
I've experience many (most?) of the problems of this thread, AFTER a full recap of both boards. The best way i'll show it is with some footage.
I also have a LC550 board but that didn't insulate me from having similar problems.
I also have a LC550 board but that didn't insulate me from having similar problems.
That looks like your G2 and sub-contrast are too high. See the Service Source guide for adjusting them. Agree with a previous poster you're not actually getting a gray screen. ROMs could be bad, not installed correctly, or you have some broken traces.
Have you checked the ROMs are in the right position? They should be in the rightmost side of their sockets.
If they're in the right spot, maybe try re-seatting them and cleaning with contact cleaner as an easy next step.
I swapped the ROM chips with my known-good CC board, no change. The originals (and the ones I swapped in) were seated just like the ones in your photo.
Like @Elemenoh said, looks like the G2 needs adjustment on that analog board, not the same kind of gray screen. My boards' DFAC-related symptoms looked much more like the photo @lfletche posted.
Since you're getting a black screen, you can probably rule out both the DFAC and CUDA at this point (even if both were damaged somehow, but not enough to affect the RST# line and soft-power functionality, you would still get some kind of raster on the screen).
Next thing I'd check would still be U10 (maybe even switch U10/U11 with each other and see if the symptoms change); if it fails it absolutely will result in zero video, zero sound, zilch. Next things after that would be the 31.3MHz crystal oscillator and the processor.
Since you're getting a black screen, you can probably rule out both the DFAC and CUDA at this point (even if both were damaged somehow, but not enough to affect the RST# line and soft-power functionality, you would still get some kind of raster on the screen).
Next thing I'd check would still be U10 (maybe even switch U10/U11 with each other and see if the symptoms change); if it fails it absolutely will result in zero video, zero sound, zilch. Next things after that would be the 31.3MHz crystal oscillator and the processor.
Another quick thing I will just add - I was doing some testing this evening between my grey screen board and the totally dead board by swapping the ROMs - similar to you it didn't change anything. But I decided to try running the board without the ROM chips installed at all and see what that does. It still powers on OK but then I get a similar black screen that Colin is seeing. Could his issue also be a trace issue to the ROM chips themselves?
).@This Does Not Compute during your vid there’s a shot of the caps removed where the traces underneath look okay at a glance. It made me think to ask if you inspected all of the traces running underneath the replaced caps carefully. Perhaps one is open or short causing your problem.
Yeah, that was the main reason why I decided to recap the board...but those traces really all did look fine. This board definitely wasn't a battery bomb victim, and I'm starting to suspect its original factory caps hadn't really even started leaking. The known-good board from my other Color Classic doesn't even look as clean as this one. This board just baffles me...if it isn't a faulty IC causing the problem, then it's a single, tiny instance of corrosion that I haven't been able to spot.
The next non-obvious class of spots to check after this are vias - do they make continuity to the other side as expected? Corrosion can sever that off and has been known to be a factor in other boards that are more traveled on in terms of repair like the SE/30. Things I picked up from Branchus Creations' videos.
Because you put in a known good board and the CC worked, that also rules out the large connector as a problem, something that I encountered in the past. Beyond @Mu0n‘s suggestion about vias, I’m not sure what else to try other than EGRET and/or sound chip replacement.
I may have a spare EGRET chip knocking around @This Does Not Compute - I gave @mitchkramez a few not long ago.