• Updated 2023-07-12: Hello, Guest! Welcome back, and be sure to check out this follow-up post about our outage a week or so ago.

Another Ailing Color Classic: Timeline so far + what next?

sambapati87

Well-known member
I'm looking for some advice on a Color Classic I acquired over the summer. This is a long post, but I wanted to provide enough detail to answer some of the first questions that seem to be asked about these problems, and hopefully to serve as a reference/resource for others by consolidating some common advice I've read into one post.

The machine is in excellent cosmetic shape (including original box, "accessory kit", floppies, and even receipt!). Except, it won't boot. I've read everything I can find on this forum and elsewhere about boot issues and am looking for advice on what the best next step is of the options I understand. I would never have gotten this far without the great info shared here and other sites.

Here's how it behaved when I first tried it:

  • Switching the hard power switch on gives a good clunk. Things seem normal to that point, at least in-so-far as I don't hear anything obviously weird.
  • Powering on from an ADB keyboard originally did nothing at all.
  • Switching the hard power on with the logic board removed results in a normal-sounding CRT energizing/static noise, the fan running, and the hard drive spinning up for a few seconds, then spinning down.


    Attempting a "jump start" by sliding the logic board in once in this state repeated the temporary hard drive spin up for the same amount of time.

[*]No video, chime, or disk access at any point. 





Naturally I assumed bad logic board capacitors. Visual inspection showed what looked like minor leakage and just the tiniest hints of very minor corrosion. I cleaned the board with several cycles of distilled water, IPA, a soft toothbrush, etc. in the hopes of getting some life out of it without a recap. This cleaning didn't change the behavior at all.

I checked continuity across the documented points in this thread (https://68kmla.org/forums/topic/24740-color-classic-no-power-issue-debunked/page/2/), as well as the traces I could figure out visually around the area of the EGRET chip. Everything I checked had the continuity I would expect, but this has not been 100% extensive by any means.

Months passed as I put off attempting a recap. I've always been bad at soldering and didn't have the right equipment. Last week I decided to give it a go, purchased a hot air rework station, flux, braid, etc., watched several excellent Branchus Creations videos (https://www.youtube.com/watch?v=LPrCkKgelco), and went at it after practicing removal on some junk boards. I was surprised by how well it *seems* to have gone. The joints look mostly good and clean to me. I had no lifted traces and the pads cleaned easily. I swabbed off under the old caps with IPA before adding the new ones. I have triple checked polarity and value. As an aside it was a huge confidence boost for me as I've always written off soldering of any kind as something I just can't do, let alone SMD components.

Here are photos of the board/caps as they sit today: https://www.icloud.com/sharedalbum/#B0tGrq0zwGcsslY One of the 100uf guys is a little suspect, but I think it has a solid connection?. There is a little more corrosion evidence visible here than I couldn't see before with the naked eye. I'll attach a couple of the more seemingly relevant photos to this post but that album has details of everything.

The recap did change things! Now, pressing the power key on the keyboard results in the same behavior as the "jump start" from before. The fan and light turn on, the HD spins up for a few seconds, the CRT energizes, then... nothing. No chime, no video. But it's something. The soft power signal from ADB gets to the EGRET, right?

So, I cleaned the board again, especially focusing on the area around the all-important EGRET chip. I've read the advice that capacitor goo accumulates here and mucks with things. I cleaned under, between, and around all the pins, the oscillator, the back of the board in this area, etc with IPA. No change.

I tried one last, more aggressive cleaning by spraying the whole board with contact cleaner and rinsing with distilled water (twice), letting that dry, spraying it down with IPA, letting that dry, and warming at ~180F for about 20 minutes to be sure it's dry. No change.

I think I know what you're thinking: "what about the analog board?" I've taken the case off and inspected things (I have not removed the board). Here's what I know:

  • I don't *see* any evidence of capacitor problems here based on what I've seen in other posts and examples. There are dark areas around some diodes but that seems relatively "normal". I know caps don't have to be bad enough to show visual evidence to cause issues, though. It was incredibly dusty, which I remedied with compressed air.
  • With the case off, I can hear what sounds like a slight "buzzing" (?) coming from the area of the flyback transformer after attempting soft power. I don't know if it's normal, but it seems like that would cause a CRT/high voltage issue, not a boot issue, and I haven't gotten that far yet.
  • I checked voltages at the HD power cable.


    5V line: ~5.05V
  • 12V line: ~11.58V (sometimes more like 11.55, sometimes more like 11.7)
  • The voltages seem consistent whether or not the HD and/or logic board are connected. They don't bounce around when I attempt boot, nor do they slowly creep in either direction if I let it sit "turned on".





The common knowledge I've found says that I should do one of (or both) of the following next:

  1. Remove, clean below, and resolder the EGRET chip, and the smaller one nearby (U8).


    • Does it seem based on the photos I've provided that goo might still be lurking under there, or that the solder joints are suspect?
    • Does anyone see any obvious issues with my recap?
  2. AND/OR Remove and recap the analog board.


    • Is ~11.58V too low for any sort of boot, with the 5V reading dead on? 
    • What other things could I test? I've no idea what else I could check on the analog board with a multimeter, and I'm not keen to go poking around without knowing exactly where and why I'm poking!



Both of those things are daunting for me, so I'm asking for advice on which to do first. Does this sound like it's still a logic board issue, an analog/PSU issue, or both?

Thank you for reading and for any advice. I hope that my timeline and description of failures might help someone in the future if they stumble across it. The Color Classic has been at the top of my list for a long time, and this example is so good that I am very invested in getting it to run.

---

An appendix of other things I've tried based on common advice:

  • I've tried leaving it switched on for hours/days before attempting soft boot.
  • I've tried cleaning the logic board contacts and socket inside the chassis by spraying the board-side connector with IPA and sliding it back and forth in the socket, then letting it all dry.
  • I've tried with and without a fresh PRAM battery as opinions seem to be mixed on whether it's required. (My board does not have a CUDA switch).
  • I tried a boot floppy to see if it did anything, just in case this was a HD, video, and sound issue, not a boot issue. No disk activity.
  • I've tried booting with the HD disconnected, in case that load was enough to keep a weak PSU from allowing boot.
  • I've tried a PRAM reset, even though I'm 99% sure it never gets far enough for that to matter whatsoever. As you'd expect nothing happened when I did this.
  • There are no VRAM/RAM modules, FPU, nor PDS stuff installed. This is fine because the machine doesn't need extra RAM or VRAM to boot, but wanted to clarify that here as well because it seems like a common question/misconception.
  • I've double checked that the ROM chips are installed correctly.

2B74A8F0-96EA-46BC-9E6E-AA9AEE96CAC9.jpeg

81202D36-2C64-4BC6-BA15-C9FC13FFBD30.jpeg

51BFA7E9-0392-4722-85D9-8CD6211ED2DC.jpeg

882D6836-377C-48FD-B12D-8FDDAAE3F32F.jpeg

8C79638B-2AF3-4E27-85CE-50E8F00A9949.jpeg

21504A12-AA14-4306-8FA3-1550655AB448.jpeg

D46522E5-8E99-4FA7-857E-86786DB45D4A.jpeg

D1517D98-34F1-4A9C-A101-A8E2B6112F23.jpeg

 
Last edited by a moderator:

bibilit

Well-known member
Job done on capacitors is fine, don’t see anything obvious. 

Egret doesn’t look heavily contaminated, but I have seen many ones looking fine and only working after removal, a thorough cleaning and nice and clean soldering (joints are looking a bit dull in my opinion)

At this point you have got nothing to lose. 

 
Last edited by a moderator:

Plasticoat

Well-known member
Although I am no expert, I have 3 colour classics, all undergoing resto... stick with the logic board for now,

if you can power the machine from the keyboard (soft power) the analogue board is functioning for now.

the empty slots do not need to be filled to enable your board to work, you do not need a battery to power up either, so the fault Lies somewhere else.

removing the egret chip, and replacing after a clean is no easy task!

you need some serious skill to remove the chip without pulling pads, and resoldering requires a degree of skill, unless your using paste (I prefer not to use paste)

I’ll let the real experts answer other questions,

but I wish you luck, these machines need a certain amount nowadays ...

 

sambapati87

Well-known member
 I have seen many ones looking fine and only working after removal, a thorough cleaning and nice and clean soldering (joints are looking a bit dull in my opinion)
if you can power the machine from the keyboard (soft power) the analogue board is functioning for now.


This is exactly the kind of info I needed, thank you both. I had a hunch that the analog board was doing its job (so far, at least—who knows what video will look like), so logic it is

you need some serious skill to remove the chip without pulling pads, and resoldering requires a degree of skill, unless your using paste (I prefer not to use paste)


I will definitely be practicing first! I was planning to use paste for those, but I'll try several methods on some other junked boards first.

 

bibilit

Well-known member
To be honest is not so hard. 

Old boards were relying on lead solder, with a pretty low temperature melting point. 

Take your time, don’t leave the hot air at the same place for too long and don’t use any tool while lifting the chip, should come easily as soon as the right temperature is reached. 

 

sambapati87

Well-known member
An update with progress!

TL;DR, I now boot to immediate chimes of death (still no video display). I take this to mean a RAM issue and have begun checking traces to the built in DRAM, as I have no SIMMs installed. I've never been this happy to hear those chimes! Ironically a lot has to go right to get to that point.

---

Before removing any chips, I obtained the Bomarc schematics (these are the same sold on maccaps.com) and tediously checked continuity across a bunch of things. The schematics are very helpful, but a warning: at least on my board, there are some inaccuracies (see below). I think some are a result of the scanning process, and some either are just a mistake or from a different revision of the board (???).

I found and repaired one broken trace (from CUDA/EGRET pin 7) that was near C6, and 6 others near the sound circuitry and the smaller cap near the CPU. Five of these were data bits on their way from the CPU to the FPU (and from there, on to the RAM, etc.), and one was an address line not making it to ROM. The data bits are patched with jumpers from the J12 test pads to the appropriate pins for the PDS connector on the bottom of the board. The address line was taken from a SCSI chip pin and also to a PDS pin. My repairs aren't the prettiest but continuity checks out and I can't find any shorts (I'm also not afraid of soldering anymore).

After repairing the traces I still didn't have boot, so I took a deep breath and removed/cleaned/resoldered C7 and C8. It wasn't noticeably dirty underneath and I didn't see anything broken, but it was nice to clean them up. I used a lot of flux and drag soldered and it went surprisingly well. Re-checked continuity and still no boot.

This leads me to a schematic issue. In the schematic, the the connection going from the CUDA/EGRET chip pin 16 is shown as going to R28 then on to the ASIC (Spice) chip's pin 16. This is not accurate. Pin 16 on the ASIC is A23 (address bit 23). Excited to have found a bad trace, I added a jumper to restore continuity to A23 (and thus, ASIC pin 16) via a PDS pin on the back of the board. However, after fixing all the other traces I could identify (and resoldering C7 and C8), and still not getting a boot, I grew skeptical of this patch because I couldn't work out why CUDA would need a single address pin. After painstakingly following the actual trace through various VIAs (a flashlight helps here), I found that EGRET pin 16 is actually meant for pin 145 on the ASIC chip, and I already had continuity. Removing this patch wire is what let me get to the death chimes.

 

sambapati87

Well-known member
I am also thinking that the jumper wire method for fixing data bit traces may not really work as it changes the electrical properties of the connection, quite possibly enough to cause it not to work as the CPU expects, which I imagine could also cause the chimes. I may need to just repair those traces in-line on the front of the board. They're just sooo close together that I was hoping to avoid it. May be no choice, though!

 

JC8080

Well-known member
Did you ever get the machine working? I just picked up a CC I haven't dug into yet, I'm trying to figure out what issues I might face.
 
Top