• 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.

Macintosh Classic Logic Board Revival

Garrett B

Well-known member
I bought this computer a few months ago, and have been gradually working toward a full repair. Unfortunately, this computer was the victim of a Maxell battery explosion (didn't know it was a common issue when I bought it). Regardless, I've made a lot of progress from where things started. Here is a list of what I've done:
  • Cleaned battery area and removed all acid/corrosion
  • Visually inspected and repaired damaged vias with magnet wire
  • Replaced all electrolytic capacitors and confirmed continuity of all nearby vias
  • Checked continuity of almost all traces going to and from 68000 CPU
  • Cleaned with IPA to remove any traces of flux or debris
After all this, when I go to boot, the problems begin. I see the "RAM check" pattern for a few seconds, the screen pattern changes, I get a mouse pointer, and that's it. No floppy disk, no happy Mac with a good SCSI HD. While the HD powers up, based on what I can hear, the system doesn't attempt to read data from it. Additionally, the mouse pointer behaves very erratically (i.e., it doesn't move easily across the screen or respond how I think it should). Pressing one of the side buttons gives me the sad mac, and the other seems to restart it.

My analog board has not been recapped, but it successfully powers a Classic II board, HD, and floppy without issue, so I am thinking that's less of a suspect. Rail voltages are 11.49, 4.75, and -11.88 for the +12, 5, and -12V systems. If I am wrong in thinking that the voltages are enough then please tell me otherwise. Any advice on things to check next would be appreciated!
 

Attachments

  • IMG_5248.JPEG
    IMG_5248.JPEG
    176.1 KB · Views: 26
  • IMG_5247.JPEG
    IMG_5247.JPEG
    877.2 KB · Views: 32

Garrett B

Well-known member
Photos attached. Let me know if I need to get a closer shot of anything!
 

Attachments

  • a.jpg
    a.jpg
    3.3 MB · Views: 55
  • b.jpg
    b.jpg
    3.4 MB · Views: 58

joshc

Well-known member
This board is not clean enough to operate properly yet in my opinion, from what I can see in the photos. There's a lot of crud around various ICs still, which IPA alone won't shift. A wash with soapy water and scrubbing with an old toothbrush, a really good rinse, and an IPA soak and then a thorough drying for a day or two in the sun would be how I'd deal with it.

When you say you see a RAM check pattern for a few seconds, do you mean the Sad Mac and then it resets or? Does it always behave the same on each restart?

Some of the legs on the SCSI chip look a tiny bit corroded, might need more cleaning and worth checking the continuity around that. C5,C6,C8,C9 leak into the SCSI and SWIM chips.
 

Garrett B

Well-known member
I've got a hot air station - is it worth removing those chips, cleaning off old solder, and replacing? I agree that it seems like it's waiting or hung up on something SCSI related. I attached a video of the boot process to help explain the issue. I have 4MB RAM installed so the check takes a while.
 

Attachments

  • Booting.mov
    11.3 MB

bdurbrow

Well-known member
I'm not sure but the lower bodge wire on the back side of the board near R84 might be going to the wrong place?
 

Garrett B

Well-known member
I'm not sure but the lower bodge wire on the back side of the board near R84 might be going to the wrong place?
Just checked it over - all is well there. It's a little hard to see with the silkscreening in that area but the connections are good. These board recreation photos from the Development forum have been super helpful. I only wish I knew how the inner layers were connected. That would tell me if I am missing any power signals on any of the destroyed vias.

If I'm correct in thinking the inner layers of the board carry the 12V/5V power to the chips, it'll be very difficult for me to determine how things should be connected. The "simplicity" of these boards is great, but I really wish I could use Openboardview to see all the connections like I can with newer Macs!
 

Attachments

  • macclassic-bottom-1.JPG
    macclassic-bottom-1.JPG
    3 MB · Views: 42
  • macclassic-top-2.JPG
    macclassic-top-2.JPG
    2.5 MB · Views: 43

davidg5678

Well-known member
The classic is only a 4 layer board, signals on the outside, +5V/GND on the inside
I worked with @Kai Robinson to sand through the classic logic board layers, and I took scans of all the inner layers a few months ago. All of the related posts were lost in the 68kmla crash, but I still have the scans if you need them for this project.
 

Garrett B

Well-known member
I worked with @Kai Robinson to sand through the classic logic board layers, and I took scans of all the inner layers a few months ago. All of the related posts were lost in the 68kmla crash, but I still have the scans if you need them for this project.
If you would be so kind, that would be extremely helpful. Really pulling my hair out on this one! I am thankful it's only four layers though, that should simplify things (I hope). I might start removing chips, cleaning, and replacing tonight. RTC chip looks the worst, so maybe I'll start there.
 

davidg5678

Well-known member
Here is a scan showing the split power planes from the top of the logic board. View attachment ClassicTopGroundPlaneFiberglassOutline1 copy.jpg

This picture shows where the power plane splits on the bottom of the logic board. (Please excuse the uneven sanding in this photograph --I eventually figured out how to sand more evenly, but this is one of the only remaining WIP scans I could find that was useable.)

View attachment ClassicLBScanPartialLayers.jpg


I believe @Kai Robinson may have some newer versions of the Classic logic board diagrams (that account for these internal layer scans) backed up somewhere. I don't know if he's willing to share them/what their status is, but if they are available, I think that they could perhaps be of use for your project as well.
 

Kai Robinson

Well-known member
Here is a scan showing the split power planes from the top of the logic board. View attachment 32713

This picture shows where the power plane splits on the bottom of the logic board. (Please excuse the uneven sanding in this photograph --I eventually figured out how to sand more evenly, but this is one of the only remaining WIP scans I could find that was useable.)

View attachment 32714


I believe @Kai Robinson may have some newer versions of the Classic logic board diagrams (that account for these internal layer scans) backed up somewhere. I don't know if he's willing to share them/what their status is, but if they are available, I think that they could perhaps be of use for your project as well.
I still have the sprint layout for the Classic.

It's as-yet untested. I'm not going to share them out just yet, i'm on long-term hiatus from electronics now.

I have had a small batch of boards produced for testing, and that's just arrived at the person who's testing them out. If they pan out, then i'll put the info up on the github page for the project and the wiki.
 

Kai Robinson

Well-known member
If you would be so kind, that would be extremely helpful. Really pulling my hair out on this one! I am thankful it's only four layers though, that should simplify things (I hope). I might start removing chips, cleaning, and replacing tonight. RTC chip looks the worst, so maybe I'll start there.
For cleaning the actual chips - toothbrush and IPA soak is good, but you can also get great results for cleaning the oxidised pins using a red scotchbrite wheel on a dremel. Sometimes it reveals the copper of the pins themselves, but keep them clean and they'll re-tin easily enough. Here's a before and after - took about 5 mins to clean up this SWIM chip.
 

Attachments

  • DSC_0772.JPG
    DSC_0772.JPG
    306 KB · Views: 42
  • DSC_0773.JPG
    DSC_0773.JPG
    295.1 KB · Views: 43

Garrett B

Well-known member
I wanted to give an update on the status of this board. Long story short, it's working again as it should. Turns out I failed to notice one corroded trace under some silkscreen. Oops. With that repair completed, I did a little conformal coating for protectant, and then soldered on a new battery holder. No it's not the prettiest board, but it's much better than where I started! Huge thanks to everyone for their help, especially Kai for the board scans which helped me to repair the traces. I probably spent more time on this board than I should have, but it's another happy Mac nonetheless!
 

Attachments

  • IMG_4430.jpeg
    IMG_4430.jpeg
    1.4 MB · Views: 41
  • IMG_5459.JPEG
    IMG_5459.JPEG
    830.4 KB · Views: 38
  • IMG_5460.JPEG
    IMG_5460.JPEG
    892.9 KB · Views: 44
  • IMG_5462.JPEG
    IMG_5462.JPEG
    915.4 KB · Views: 38
Top