Color Classic Missing Green

[superjer2000]

@Bolle You seemed to have solved this in an earlier thread - I couldn't quite tell what fixed it though...  (and it was only 10 years ago so I'm sure it's still top of mind!)

I recently recapped the analog board in one of my Color Classics.  When I started it back up I noticed the colours were off - after a bit of investigation, it seems that green is not working.

I hadn't powered the system up for literally years before so I can't say definitively if the system worked before the recap (from my notes it did back in 2018 apparently.)

There are two adjustment drive pots at the back of the analog board (blue and green).  The blue drive pot does change the intensity of the blue but the green drive pot doesn't change anything.  When I play with the convergence, I can see the red and blue but there isn't any green.  There three other adjustment pots (red background, green background and blue background).  When I crank the green pot up the screen background does go greens so the gun is working.

1. I reheated/added solder to all of the connections at the back of the video neck board.  No change.

2. I played around with the green drive pot while taking measurements with my multimeter.  It seems to behave the exact same as the blue pot so I think the pot itself is OK.  I can test for good continuity on the A/B from the green drive pot to the same components as to the blue drive pot.  There seems to be a lot of overlap between the green and blue drive pots with regards to what traces are connected to what - I was hoping that the logic would be more discrete so it'd be easier to test directly from the video connector on the A/B to the video neck board, going through the drive pot, but no such luck.

3. I checked for continuity between what I think are the R/G/B signals coming off the video chip on the mother board and the edge connector on the motherboard.

4. I checked for continuity between the edge connector and the video plug on the analog board (under the metal cover).  It seems good although there was one connection that didn't seem to beep right away).  Not sure if that was the cable from the edge connector, the solder joints, or me trying to balance the analog board and multimeter to test.

5. I reheated/added solder to the connections from the analog board that go to the video neck board. No change.

What I'm really hoping for is some thoughts as to what connectors might be responsible for sending the green gun information.  There aren't any analog board schematics as far as I can tell.  There are two bunches of wires going from the A/B to the video neck board.  There are some colored wires (pink/yellow/red/black/blue/orange) that are soldered directly into the A/B and there are a bundle of three white wires that are plugged into a harness near the adjustment pots.  I'm assuming it's the three white wires that are providing the R/G/B video signal but I'm hoping to get confirmation from that so I can spend more time poking around in that area...

Thanks!

 

[tattar8]

Do you have an oscilloscope by any chance?  You could probe the green signal from the logic board, just to make sure the board is actually sending that green signal.

 

[Bolle]

I ended up resoldering the whole analogboard and CRT neckboard back then - didn't change it.

I played around with the drive pots as well and could verify the green gun was firing, it was just not getting a video signal.

Cleaned logicboard and analogboard edge connectors, verified good connections afterwards - no change.

Full recap on the analogboard - problem still would come and go.

What seems to be critical for the video signal is that the grounding tabs on the bottom of your logicboard actually make good contact to the chassis. On my CC the sheet metal is slightly bend and wouldn't make contact to all the contact springs on the logicboard.

Since I bent it back into place as good as possible, cleaned the spots where the logicboard spring tabs should make contact and the tabs itself and made sure the springs are firmly attached to the logicboard which helped a fair bit.

The green signal would become a lot more stable after that treatment. I just had the CC out again last week and the green color drops out now and then and flickers so I might have to check again (and maybe put some kind of shim into the bottom of the back bucket to press the sheet metal against the LB springs)

 

[techknight]

The other thing is the video amplifier IC. Could check that. 

 

[superjer2000]

Thanks all - @Bolle That's very interesting on the grounding issue.  Right after recapping, I had started the machine and didn't notice any off colours, although I wasn't really paying much attention so I just had assumed the issue was there when I had first powered up the machine and I just didn't see it.  I shut the machine down for the night and then fired it up for longer testing the next day and that is when I had noticed the missing green.  Between the original boot and the second boot though I had pulled the logic board out of the chassis.  Without the computers back panel on the chassis seems to flex a bit when sliding the logic board in and out and so I wonder if I had messed up some of the grounding contacts.  I'll check that out.

Another area I want to investigate is the cable in the red box as when the shielding cover for the video area is on, it is definitely a bit stressed.  I was pushing the cable around with a chopstick when the unit was on and it didn't make a difference though.

I'm assuming the connector with the three white wires (in the blue box) are what is pushing through the RGB signal to the video board.  Any idea as to which wire would carry which signal? 

I had assumed that there might be three transistors on the video board and one might have gone bad but I didn't see any (albeit I haven't taken all of the shielding off).  Given that, I'm assuming the video amplifier IC @techknight is referring replaces those transistors and is the LM2418T that is mounted vertically by the white cable.  I just have a USB oscilloscope - but I should be able to probe the low voltage inputs coming into the LM2418T to see if a signal is getting through on all three inputs.  Assuming one of the input lines to the LM2418T is dead, I then know I can start working backwards from there.  From the BOMARC schematics, I think I know which lines are feeding the RGB signal to the logic board edge connector (30,26,22).  If I can trace those to the analog board edge connector (jury is still out on how I will do that) I should be able to start to probing to see if a signal is getting through to the unplugged end of the video cable in the image below.

Appreciate any other thoughts but I'll update as this unfolds.

 

[techknight]

You need to measure the output of the video amplifier IC first. If its good and matching, then everything is ok. 

but if your missing green, then work backwards until you pick it up again. 

 

[techknight]

Also that gray cable in the middle is what carries the RGB signal from the edge connector back to that motorola XC chip. 

 

[superjer2000]

Well the simplest solution...

I looked at the sample schematic for the video amplifier ic and was able to trace enough of the circuit to confirm good continuity from the edge connector of the a/b to the video chip to the amp to the crt neck board.  I didn't see any issues.  I figured out which line was green on the amp side of the video circuit but figured before I started testing that I'd look at @Bolles issue. 

Sure enough,  two of the grounding tabs under the logic board were not making contact.  I cleaned them and the mating surfaces on the logic board with contact cleaner and bent them to make contact.  I also cut away some metal from the video section shield area that was putting stress on the cable.  I cleaned the edge connector for the analog board and logic board, reassembled and presto. Green came back.  If I had to bet it was due to the grounding.  I'm not sure how you found that Bolle. I think I would have proved that there was no green signal coming from the logic board but I am not sure I would have traced it to a grounding problem. I wonder if it would be worthwhile creating a grounding strap for the motherboard.

I went through the crt adjustments but when I played with the focus pot the screen started to flicker a bit so I think I'll need to take it apart again and clean all of the pots with some nu-trol.

 

[techknight]

ive never lost green because of a logic board grounding issue with the chassis. Matter of fact, I think I broke the tabs on one of mine and still dont have a problem. 

Theres something else thats wrong and its trying to make ground through those tabs. 

 

[superjer2000]

ive never lost green because of a logic board grounding issue with the chassis. Matter of fact, I think I broke the tabs on one of mine and still dont have a problem. 

Theres something else thats wrong and its trying to make ground through those tabs. 

Would you assume it would be on the logic board rather than the analog board?  I was thinking that would be pretty marginal engineering on Apple's part.  

I'm assuming the edge connector on the logic board also ties to chassis ground and the tabs are just supplemental.  I'm not even sure where I would begin to try to identify the underlying cause. Any suggestions?

 

[techknight]

Yea the edge connector is supposed to supply the ground for the video signalling and other things. I think its even shown in the BOMARC schematics. 

Somethings off... I feel it. 

Maybe its like nintendo and the edge connector is starting to show problems? 

 

[IlikeTech]

My CC has problems that are related to the edge card I think.  Sometimes I loose green as well (jostling the machine seems to fix this), and I also seem to loose working SCSI until I reseat the logic board.

 

[SoundChances]

Yes, don't overlook the simple explanation that the edge connector is the problem.

I've had several Color Classics and the edge connector was a problem on all of them, check and clean those "pins" on the mainboard. I've seen some where the "pin" has lifted and causes a short, others where the "pin" caught on something being inserted and pushed the metal back along its plastic guide so it didn't make a connection.

But most of the time, just a little wiggle and jiggle will make it right.

 

[techknight]

But most of the time, just a little wiggle and jiggle will make it right.

That's what she said.

And yes, the flashlight trick looking into the slot should help identify if a pin is corroded or has broken. I assume the electrolytic caps in close proximity to the Edge connector could do this. 

 

[superjer2000]

Well I pulled apart the entire unit (i.e. the bottom part that holds the logic and analog boards, floppy and HD) to remove the metal shielding and get a better look at the edge connectors.  One observation is I was astonished at just how much more pliable the Color Classic's plastics are than the LC575's, but how much more fragile the actual workings of the Color Classic seem to be.

I didn't see any bent, damaged, or oxidized edge connector pins.  I did try to clean them the best I could with Nu-Trol and some paper towel but nothing really back on the paper towel.  I also cleaned the edge connectors on the analog board and the logic board.

I did try to work some Nu-Trol into the 15 or so pots on the analog board with minimal luck.  They are sealed up pretty good and I'm not sure where any of the gunk would drain to anyways.  That being said, I still tried to drip a bit in there and then work the pots back and forth.

End result:  Nothing really different.  The system is working well but every so often the display color flickers just a bit.  It doesn't really do it very often (and sometimes not at all for hours at a time).  I don't think I will take it apart again but I am curious to see how it contrasts against my other Color Classic when I recap that analog board.

As an aside, when I recapped it, I did replace the two diodes that cook up the analog board and I mounted them up off the analog board a bit to provide for some additional cooling.