Mac Classic -- horizontal bars after recap, -12V problem

[zachlapidus]

Hey there, definitely a newbie to this. I had a Mac classic with a good bit of electrolyte damage. No sound, couldn't boot, but came on with picture and blinking disk -- it threw a bus error when trying to boot to the rom disk. Looking at the board, it was clear that the electrolyte did some damage, wrecking the 74LS174 and 7905 regulators. I recapped the board, replaced those ICs and repaired the lifted pads from the corrosion. IN MY INFINITE WISDOM I decided that it would a good idea to recap the analog board since it was already out, so I replaced every single cap on the board with the list from Amiga of Rochester. There was a good amount of corrosion under the caps so it seemed like a good idea.

I reassembled the machine and turned it on. No smoke -- a good sign. No bong, I get a picture, but it's just alternating black and white horizontal lines, one pixel each. The picture is at least very clear, so it seems like the high voltage for the CRT is working. This happens every time I reboot it. I tried to measure the voltages through the floppy connector -- I see 4.5V on the 5V line, 10.8 on the 12V line, and at first I see -7V on the -12V line by the time I measure it, but this voltage keeps steadily climbing. Uh-oh.

Anybody have any tips on how I should proceed? Does it seem like the problem is probably on the analog board? Since there are all these high voltages involved, not sure exactly how to safely probe anything on the analog board or even the logic board. Thanks in advance!

 

[bibilit]

HI, low voltages or High voltages can be a sign of a bad optoisolator, so i will replace it first with the two diodes that are in the neighborhood DP3 and DP4 (1N4148)

Don't have the reference right now but probably CNY17G

Keep in mind that the 12 volts line is important but the 5 volts is the key, 4.85 or above is required to get a boot.

 

[zachlapidus]

Hi there, thanks for this! Just measured directly off the PSU's connector -- got a stable 10.57V on the 12V rail, -10.7 on -12V rail, 4.6V on the 5V rail. Seems like nearly everything is 10% off. I will check out that opto and diodes. I know there's also an analog board trimmer for the voltages, should I play around with that before I replace those parts?

Since I get a changing (and too high) voltage on -12V measured through the floppy connector when the logic board is connected suggests that perhaps there's a short on the logic board on the -12V line...I was a little suspicious of the 7905 regulator that I replaced. In fact, I just tested it with a multimeter and...pin 1 and pin 2 are shorted! Not sure how this happened -- there is a bodge wire that maybe is connecting to an exposed trace or something. I didn't see anything on the microscope on that thing when I was working on it, but noticed some weird junk around pin 1 and 2 just now. I wonder if the chip cooked internally from the short. Sigh... Anybody know if the computer can boot without a -5V rail? I was trying to figure out what it's used for from the schematic and it seems like it's just for the printer port? I'm no expert but I'd be nearly sure that UB7, the 26LS30 probably didn't appreciate the over voltage.

 

[zachlapidus]

Okay, maybe a minor breakthrough:

Haven't done too much, but noticed one thing -- not a dead short, but only 6 ohms of resistance between pin 1 (-5V out) and pin 2 (-12V in) on the 7905. That seems super low to me! I have around 0.3ohms of resistance between pin 1 of the 7905 and pin 10 (-5V input) of UB7, and the same 6 ohms between pin 2 of 7905 and that pin. Could anybody tell me if this is this typical of a Classic mobo? I measured a spare 7905 and there are several megohms of resistance between pin 2 and 1.

When I removed the initial 7905 from the board when I bought the non-working Mac, I noticed that pins 2, 3, 6 and 7 of the 7905 were dark colored. I thought this was just corrosion, but maybe there was a short somewhere else on the board before I even removed it. I'm wondering if the junk between pins 2 and 1 of the 7905 when it (maybe) died was something under the chip (flux residue or something) being instantly cooked if the chip got way too hot. Hard to tell without removing chips and etc. I probably should have just tried to fix the logic board before I fixed the analog board so I was only troubleshooting one thing at a time, but here we are.

 

[zachlapidus]

I am waiting on a parts order for the opto and diodes, but I tried adjusting the analog board’s trim pot. Measuring from the analog board’s connector I’ve got 12.1, -12.1 and 5.1V. I took the 7905 voltage regulator completely off the logic board to remove the short, and connected the logic board. Got identical behavior, with the no chime and horizontal black and white one pixel lines. Measuring from the floppy connector, I get nearly bang on 5V and 12V, but I saw the -12V rail was at -2V and climbing. What could be going on here? There must be some hidden short on the logic board, right?

Weirdly, in the Dead Mac Scrolls PDF, it mentions this horizontal line thing as a symptom of installing the programmer switch incorrectly in the Mac plus. Is this a red herring here?

 

[zachlapidus]

Hey everyone -- it turns out that the reason the -12V was measuring so wrong at the floppy port was because of a rotted VIA -- perhaps it was just really high resistance and preventing the voltage from reaching there. With an ATX extension cable, I can properly verify the -12V on the J10 edge connector.

With the benefit of some time off and a bit of research, I decided to go back to square one and remove the variable of the analog board. I'm now powering the Mac Classic logic board off of a bench power supply with 5 and 12V, which is allowing me to probe the board with a multimeter and scope. What I'm seeing is that the /RST signal is being held low at the UA5 SND/RESET chip. Pressing the reset button does nothing. Voltage at pin 7 is pulled up to 5V -- pressing the reset button drops it to 3.3V. C3, the 1uF capacitor, has kind of a weird voltage on it... 0.89V. Pushing the reset button drops it to .61V. I noticed that sometimes when I was measuring it on my scope I was discharging it slightly, which seemed suspicious to me.

Measuring a good 5V on C2 and 12V on C1.

I'm wondering if next step needs to be removing the sound chip from the board, but I am kind of nervous because it's a custom chip. I have a hot air station but have never lifted a chip like this one before. Any tips here? Just hit it with 300C and a medium airflow and carefully lift after awhile?

Also a bit of a question that may be related to this...having a strange effect measuring the voltage on C3. Is it typical for multimeter and scope probes to measure a voltage when close to, but not touching, the pin? This seems strange to me, but I noticed it am when I was probe close to the anode of C3 touching the PCB, I pick up a voltage as I slide it around the PCB. I experienced this around C1 as well. The explanation that I'm favoring (and dreading) is that perhaps some of this conductive electrolyte has persisted on the board and is offering a high resistance path between my probe and the terminal. I am at a loss, though -- I washed this board with IPA probably 5 times and did a huge wash with dish soap and water. Would love any tips or next steps.

 

[zachlapidus]

Made some progress. After deciding to reflow the caps, I noticed the C3 pads lifted immediately when heat was applied. Electrolyte corrosion was pretty nasty, apparently. Cleaned thoroughly, ran a bodge, and tried to boot the machine.

Now, the machine makes sound for the first time ever! I think it might be trying to bong? Sounds like a demon bong. Something's obviously getting distorted. There is the zebra stripes simasimac pattern for awhile, the "bong," and then eventually the screen changes. Is this trying to boot?

Not sure where to check yet. All the address line and data lines on the rom chip beep out for continuity, and I'm seeing activity on the ram address lines from the BBU (not sure what I'm looking for). I've tried testing with the ram expansion board inserted, and get the exact same results.

 

[zachlapidus]

Just in case it helps anyone else in the future, the ROM chip was in fact bad! I bought a for parts logic board from eBay, swapped the ROM, and I got the bong! Not out of the woods yet, though. I tried to boot to the ROM disk, and while I get past the Welcome to Macintosh screen, I never get to the desktop. The blank menu bar appears, and then the “a system error occurred” bomb dialog box occurs. It give an error id=02 which seems like a bus or addressing error. Does this point to bad ram? I’d assume so, except that the startup ram test apparently passes. Any ideas from anyone? I’ll keep updating this repair diary.

 

[davidg5678]

Sounds like you've been making some good progress! I think this machine in its current state would be a good candidate for the serial terminal self test mode: