Macintosh Performa 550 - No Boot

[HB1024]

I picked up (or, rather, had shipped, which I'd have preferred not to but local pickup wasn't an option for the seller) a Performa/LC 550 today that was billed as working when it was packaged up. It did not arrive working.

When plugged in, the Trinitron screen turns on (very audible degauss coil). Turning the computer on results in speaker pops, power LED coming on, the hard drive spinning up, and the fan coming on, but nothing else happens. The keyboard does initialize but only does so once. There's no drive activity, and no chime.

I have tried re-seating the main logic board several times, and I have also tried removing the installed RAM SIMM (I know that it has 4 MB of soldered memory) and re-seated the ROM SIMM as well. What concerns me is that there was a replacement battery installed: someone removed the 3.6V holder and soldered in a pin header and attached one of those 4.5V lithium boxes. Even though this looks like it was done fairly well, there's spots around where the battery holder used to be that look a little corroded.

I did notice when I first got it that moving the machine side to side resulted in the speakers crackling, but it's not doing that anymore.

I am obviously going to have to go further than the main logic board, any ideas how to do this without incurring the wrath of seriously brittle plastic?

 

[GiGaBiTe]

Some Macs won't power up with a dead PRAM battery.

I would remove that 4.5v battery box and reinstall an original 3.6v Lithium battery holder with a 3.6v battery. 4.5v is too high and may cause damage to the board.

 

[HB1024]

Small correction: it's a 4.5V alkaline box, not lithium. One of those Rayovac cube things once common on PCs. I also believe what I thought was a ROM SIMM is actually VRAM on these, but I don't know. It's a 68-pin SIMM bearing an Apple copyright, although 68 pin memory was also an Apple thing so it might not mean much.

The seller sent another chassis (from an LC 575 but it's the same basic idea). The packaging for the second chassis was sub-par and predictably it came smashed. Somehow the tube did not get destroyed, but it confirms I do have a main board issue, not a chassis issue as I suspected initially. Further confirmation came in the form of a Macintosh TV logic board that boots and works fine in the original 550 chassis.

I've heard that reflowing boards like this in the oven is a bad idea for varying reasons but I may just try that because it's already dead. It was billed and photographed working before it shipped so I assume that even though the chassis is unharmed it got bounced and a solder joint cracked.

 

[GiGaBiTe]

I also believe what I thought was a ROM SIMM is actually VRAM on these, but I don't know. It's a 68-pin SIMM bearing an Apple copyright, although 68 pin memory was also an Apple thing so it might not mean much.

Yes, 68 pin SIMMs are VRAM. Assuming this is your logic board:

Then the second slot from the right is the VRAM slot. The logic board has built-in VRAM and should work without it, the extra 68 pin VRAM just gives you more color and resolution options.

I've heard that reflowing boards like this in the oven is a bad idea for varying reasons but I may just try that because it's already dead. It was billed and photographed working before it shipped so I assume that even though the chassis is unharmed it got bounced and a solder joint cracked.

DO NOT DO THIS, EVER.

Your oven does not get hot enough to melt solder, and you will destroy everything on it.

The most likely culprit is that your PRAM battery is dead/missing. You said a pin header was installed, where exactly? This board originally took a 1/2AA Lithium cell in the top right corner above the RAM slot. If there's a pin header somewhere else on the board, that is NOT a battery header, and you may have damaged the board if you plugged a battery into it. As I said in the last post, there are several Mac models that won't boot if the PRAM battery is dead/missing. If there's corrosion around the pads where the battery was, there's no telling if the board is getting power or not.

The second most likely culprit is failed capacitors. If the seller didn't say the board was fully recapped, expect all of the capacitors to be bad, because they are. All Macs made from 2005 and older will have failed SMD capacitors that need to be replaced. The board won't work properly at all with dead capacitors. You may get lucky and have it boot, but it won't be stable, and will accelerate the degradation of the failed caps, resulting in possible logic board damage.

Check around the base of each capacitor for a sticky liquid, which would be leaked electrolyte. If the electrolyte has gotten under the solder mask, traces may appear darker than those around them, which is an indication of trace damage. This will require repair if the trace is sufficiently corroded or broken.

 

[HB1024]

PRAM battery missing is absolutely not the problem. Again, it worked without a battery (well, with an assumed dead battery) before it got here. For the same reason, I am fairly confident this is not going to be fixed by blanket-replacing capacitors; I can do that after I am sure I am not wasting my time.

Here is the board in question, looks almost if not the same as the one in your picture (which has the same battery modification):


Here is the battery mod up close:

Old 1/2AA holder removed (there are silk-screened outlines where the holder would have been), wire soldered to positive and negative, and a support wire in the middle that's a little suspicious.
There is corrosion by the VRAM slot, some by the positive battery terminal, and a little bit in the traces above U8. I don't know if this is just surface corrosion from improper storage or something worse.

Little but of the same going on by the ROMs:

 

[GiGaBiTe]

PRAM battery missing is absolutely not the problem. Again, it worked without a battery (well, with an assumed dead battery) before it got here.

If you never saw it running with your own eyeballs, you can't make that call. Put a battery in it, the CORRECT battery. You can get 1/2AA Tadiran batteries on Mouser for $6-7, and the holder for another $2-3.

For the same reason, I am fairly confident this is not going to be fixed by blanket-replacing capacitors; I can do that after I am sure I am not wasting my time.

You're wasting your time by not replacing the capacitors. It is not something that is optional, those capacitors are bad, I can see that they're leaking, and two of them look physically broken. The logic board will not work properly with failed capacitors, period. You're also going to have to recap the power supply, because those fail just as badly as the caps on the logic board.

Vintage computers are a pay to play hobby, especially old Macs. Any Mac with SMD capacitors like this is going to have leaked and will need to be replaced. I just finished recapping two Quadra 605s that I bought, and every single capacitor had leaked on the logic board, floppy drive and power supply. This has been true for the dozens and dozens of Macs that I've worked on, including all of my own Macs.

 

[AndyO]

It is also worth adding that capacitors that have leaked can also impact the functionality of the board in quite unexpected ways. Recapping isn't the only thing that will need doing, but also a very thorough cleaning of the board, which may require some components to be removed first.

Neither my Classic or Classic II would boot, neither even getting to a blank screen or Sad Mac, until they'd been ultrasonically cleaned - now they both run well.

 

[GiGaBiTe]

Yep, the electrolyte is conductive and can short ICs out, and bridge VIAs, causing erratic operation.

It also causes electrolysis damage when power is applied, much the same way putting iron bars in water and applying DC across them, corrosion will start. Component legs can be eaten clean off, traces can be eaten and if it seeps inside the board, VIAs and internal traces can also be destroyed.