• Please review our updated Terms and Rules here

Macintosh classic with full grey screen (no mousse pointor) nor startup song

N3rd

Member
Joined
Jun 21, 2020
Messages
34
Location
France
Hi everyone,
I am currently trying to save a Macintosh Classic logic board. Currently, the computer only display a full grey with no mouse pointor (see linked picture) and don't produce the startup song.
This is a macintosh classic sold "for parts" that I am trying to ressurect. When I received it, I was suprised to discover that the capacitors, the y1 oscillator and the two reset switch was missing. I now have reccaped the logic board with tantalum capacitor and the y1 oscillator with a brand new 32,768 kHz one. The two reset switch are still missing but I don't think it could be a problem for testing since they are normally open, if I'm not mistaken. I have looked for cut tracks but for the monent, I did not seen any. I have also tried to swap the rom ic with one issued from a dead logic board due to a leaked battery but the result is still the same.
Can you confirm me that the problem could not come from the analog board and the value of the y1 oscillator just to be sure (It appears to be used by the rtc, so it should be a 32,768 kHz).
Do you have any idea or advice ?
Thank you for your help.
 

Attachments

  • photo61000.jpg
    photo61000.jpg
    248.9 KB · Views: 2
  • photo61001.jpg
    photo61001.jpg
    250.6 KB · Views: 2
If it's booting well enough to show a grey screen, I'd think that the logic board is mostly working. Maybe one of the ICs isn't 100%, or has a disconnected leg from a bad solder joint?
 
Yes, I also think that this logic board do not have a big problem but It could be difficult to find, unfortunateley. The only IC mounted on support is the rom that I have already tried to swap without success. Almost all the componenents are SMD on this board, so it is very difficult to detect a bad solder joint. Must I try a blind reflow (hot air gun and flux) until It's work ?
 
I have found the problem, the via under the battery holder are destroyed by corrosion. The tracks seems pretty ok but the via are damaged. I am not sure this could be repair. An other victim of a leaked battery...
 

Attachments

  • photo61009.jpg
    photo61009.jpg
    234 KB · Views: 2
  • photo61010.jpg
    photo61010.jpg
    187.5 KB · Views: 1
Yeah those VIAs and traces don't look very good. There are schematics for the Macintosh Classic available if you want to take a shot at repairing the board. I'm pretty sure that the inner layers are mostly power and ground planes, so you may get lucky and those VIAs can be traced pretty easily.

https://archive.org/details/Macintosh68kSchematics

And even if you don't have luck trying to repair them, there is an effort underway to make new logic boards available for several compact macs. The Macintosh SE/30 has a working logic board replacement, and the regular SE is not far behind. The Classic is coming next and is probably about halfway done to the point of getting a prototype board up. The guy doing the reverse engineering has the top and bottom layers done so far and they look really good. The goal is to use all new components that are generic and readily available (caps, resistors, most connectors, RAM) and transfer over the proprietary chips. Though there is a reverse engineering effort on some of those too since they're hard to come by and expensive.

Classic_Top1.jpg

Classic_Bottom1.jpg
 
Yes, I would like try to repair this board. There is three vias that I am not able to trace to rewire the resistor R88, R89 and R90 (see the arrows on my linked pictures). Could you help me with this ?
The reengineering of compact mac logic boards is a very interresting news. I was not aware of it. Is it allready availlable to sell for some models or is it just in the study step for now ?
 

Attachments

  • photo61023.jpg
    photo61023.jpg
    89.7 KB · Views: 1
  • photo61024.jpg
    photo61024.jpg
    83.2 KB · Views: 1
The schematics I linked to show you what those resistors are on "sheet 2 of 5" ...

They pull lines up to the +5v rail. All three VIAs should connect to the +5v power plane inside the board. The pulled up lines go to the serial ports in the back of the machine and probably don't have anything to do with it not booting. You should be able to get continuity to any +5v source on the board, but the best place would be the power connector in the middle of the board.

You should concentrate on repairing all of those green traces and what looks like a broken trace above them. Those are important and would cause problems. I would recommend carefully scraping away the solder mask and using something to neutralize the battery goo. Many people like vinegar, but I prefer Deoxit Gold G5, or Krud Kutter rust converter liquid on a cotton swab.
 
I have rewired all the tracks and vias damaged that I have detected with the exception of the six vias directly linked to the resistor R88, R89, and R90 since it seems to be used only for the serial connectors. I have also cleaned the board with a vinegar bath (the only product that I have at home). I have now a glitched sad mac screen with random error codes.
 

Attachments

  • photo61032.jpg
    photo61032.jpg
    211.1 KB · Views: 2
  • photo61033.jpg
    photo61033.jpg
    190.7 KB · Views: 2
  • photo61034.jpg
    photo61034.jpg
    187.3 KB · Views: 2
  • photo61035.jpg
    photo61035.jpg
    211.3 KB · Views: 2
You MUST fix the bad traces, you can't just use bodge wires around them. When a trace is as bad as those on your board, they become resistive and will cause signal integrity issues, even if you put a bodge wire to bypass them because they're still in-circuit.

Scrape off the solder mask on the traces and treat them with something like Krud Kutter rust conversion liquid: https://www.walmart.com/ip/Krud-Kutter-Must-for-Rust-Remover-Inhibitor-8-oz/17619958 Only use this on a cotton swab/Q-tip, don't dump it on the board, or soak the board in it because it is very aggressive. This stuff will make quick work of corroded traces if you rub the cotton swab soaked in it back and forth a few times. Don't use vinegar because it will eat the trace away.

Once you clean up the traces, you'll want to run a soldering iron over them with solder to tin them and prevent oxidation, as well as improve their conductivity. If there are any breaks in the traces, you can fix them by soldering a thin wire on top of them. I would avoid using bodge wires around the CPU due to timing bugs they can introduce. If you must use them, you need to cut the original trace, but do this as a last resort. You also need to make the bodge wires as short as possible to not introduce noise into the line, as well as trying to avoid running them by other signal lines for the same reason.

It's good you got something on the screen at least, but I'm betting those bad traces are the source of your grief due to the random nature of them. "00000002" is a memory bank A error "00000004" is memory bank B error, "0000000E" means something on the data bus failed. With the random nature of the errors and corrupt garbage on the screen, it's pretty certain the bus lines are screwed up.
 
It seems to be a very difficult operation with my skills / meterial but I can try. However, I don't understand how to deal with detroyed vias in this case, knowing that the cpu lines traces are originally propaged between the boards faces thanks to theses vias. Is there a way to repair those vias and avoid the need of using bodge wire to bypass them.
 
You can try to use the same Krud Kutter I recommended on the traces and then follow it up with soldering them. VIAs are tiny copper tubes or rivets that go between layers, sometimes filled with solder. If adding solder results in garbage floating to the surface of hot solder, try removing it with a vacuum pump or solder wick and add more solder. You can repeat this process a few times until it looks better, just be quick and don't use a ton of heat so you don't stress them too much.

The problem with bodging bus lines in particular is it alters the amount of time it takes for signals to propagate through them. If you have a 16 bit bus and an arbitrary number of the lines are much longer than other lines, you're going to have timing issues. This compounds with the fact the nasty corroded bus lines are still in circuit, which could be dragging the signal down or distorting it. In circuit board design, bus lines are usually routed in such a way that they're all as close to the same length as possible to keep data arriving on time on each trace.
 
The problem is that in my case the vias are not just corroded, the copper is actually simply missing. I have also begin to scrap the solder mask and the copper is missing in large portion of the traces.
 
I'd recommend fixing the traces and then jumping bodge wires from the last part of the fixed trace to the next VIA where the trace links to.

aL13YYo.png


For traces that have missing copper, you can strip a length of wire as long as the bad part and solder it over the top of the trace. You can anchor it on one end and stretch it out over the trace to make it easier. If there's no copper at the end of the bad section of trace, try to instead run the bodge wire along the original path of the trace and then jump it over. If the bad traces start to fall apart, you'll probably just want to do it this way. Just make sure you remove the bad traces from the circuit by cutting them.
 
I will take a little break on this board. Working on this board make me realized that my other Macintosh classic's logic board that I was considering as dead in reason of leaked battery damages and its general corrosion state (espacially back connectors) was finally maybe in better condition that the one I was trying to resurect.
The traces are globally correct on it. Only one CPU line trace is cut. I have for now bypassed it with wire. Since it's seem to be working I would prefer for now not trying to make a "real repair" of the trace to be sure to not make more damaged if something goes wrong during the operation. The traces going to one of the serial connector are also dammaged but for now it is a minor problem.
After having recapped the board, I was able to succefully boot from a System 7.5 floppy.
I have to do more test but It's seem pretty functional. However, I am worry about a point, the Mac doesn't produce the startup "bong". Maybe the problem is coming from the analog board ?
There is also a little zone of the underside of the board which don't look quite good especialy the C74 capacitor.
 

Attachments

  • photo61046.jpg
    photo61046.jpg
    169.2 KB · Views: 2
  • photo61048.jpg
    photo61048.jpg
    179.9 KB · Views: 2
  • photo61049.jpg
    photo61049.jpg
    215.1 KB · Views: 2
  • photo61050.jpg
    photo61050.jpg
    172.9 KB · Views: 2
  • photo61051.jpg
    photo61051.jpg
    121.5 KB · Views: 2
Yes, I Know that but in my case all the electrolytic capacitors were replaced by brand new tantalum capacitors. So except if it can be caused by a damaged ceramic capacitor like the C74, the "no bong" problem should not be caused by bad capacitors.
I will try to unsolder the extremly corroded headphone connector. Maybe the bong is simply muted in reason of this connector. By the way, if you have a link or reference for a remplacement part for this connector, it would be great. The pinout of this jack connector seems pretty specific and for now I am not able to find a corresponding one.
 
I have to do more test but It's seem pretty functional. However, I am worry about a point, the Mac doesn't produce the startup "bong". Maybe the problem is coming from the analog board ?

I would say it has more to do with the audio jack being a pile of corrosion. The audio jack has an internal bypass switch that disables the internal speaker when an external sound source is plugged in. If that switch is damaged (and it most likely is), then it would prevent the internal speaker from working. You'll need to replace that jack, as simply deslodering it from the board will likely yield no sound, but you can try.

I'd also recommend removing the serial ports and SCSI port due to how damaged they are. You can see green corrosion coming out of the holes of the serial ports, and SCSI has the 5v rail on it, it wouldn't be good if that got shorted out to something.
 
I have removed the jack connector and indeed there is still no sound. Unfortunately, I am for now not able to find a replacement part with the correct pinout to replace it.
I will replace the external SCSI connector that is clearly too much corroded to be saved and probably the serial connectors.
 
The audio jack on the Mac Classic should be a bog standard part, it's not like the later 68k Macs with the deeper audio socket for the plaintalk mic and special speakers. I can't see the link you posted though.

The serial ports are 8 pin S-Video connectors and you can find them all day on Ebay and Amazon.
 
Back
Top