The little Classic that could(n't)

[Rusty33]

Hi folks,

Seeking some guidance re: my little mac that could…but, seemingly can’t hang on any longer…

The back story: for the past couple of years, I have been using an old Mac Classic (M0420) purely as a writing machine. I absolutely adore using it for this purpose.

The computer itself is, of course, 25 years old…but the board has been recapped within the past few years.

In recent weeks I have begun to encounter some funny business when booting the machine up. A few odd things are happening:

• the initial image on the screen is garbled
• the image on the screen vibrates radically
• the hard disk does not power up

(see Image 1 below)

After about 2-3 minutes minutes of continuing in this state, the image vibrations stabalize, and the hard disk inevitably powers up. About 1 minute later the system chimes, and then after about another minute of the grey boot screen (see image 2 below). the computer boots up perfectly normally (see image 3 below).

Upon booting, the system functions perfectly. A system restart (from an on position) works flawlessly. It isn’t until I leave the system sitting off for a few minutes that the above cycle repeats itself.

Can anyone advise what I might be likely dealing with here?

• More leaky caps?
• A failing Hard disk?
• Motherboard trolls?

Any advice will be greatly appreciated!

 

[MacTech68]

...Any advice will be greatly appreciated!

Based on that behaviour, I'm going with caps. Could be mother board or power-sweep board.

When it does it, does a hefty whack change it's behaviour?

 

[Rusty33]

Based on that behaviour, I'm going with caps. Could be mother board or power-sweep board.

When it does it, does a hefty whack change it's behaviour?

Hmmm, I am somewhat inclined to speculate that I have a power supply issue on my hands here...I've been doing a little reading this evening and learned that wavy screens can often be attributed to insufficient voltage being fed into the CRT...this might also explain why the HD is not powering up?

But where on earth would one find a new power/sweep board for this unit?

I have not yet had an opportunity to give it a solid whack yet...but shall report back with my findings!

 

[keysersoze]

Hmmm, I am somewhat inclined to speculate that I have a power supply issue on my hands here...I've been doing a little reading this evening and learned that wavy screens can often be attributed to insufficient voltage being fed into the CRT...this might also explain why the HD is not powering up?

But where on earth would one find a new power/sweep board for this unit?

I have not yet had an opportunity to give it a solid whack yet...but shall report back with my findings!

I would guess it's an issue with the analog board caps and/or solder joints... have you tried posting over at 68kmla.org? There are some folks over there who really know there stuff with respect to the compact macs. Also a couple from Aus!

 

[MacTech68]

Hmmm, I am somewhat inclined to speculate that I have a power supply issue on my hands here...I've been doing a little reading this evening and learned that wavy screens can often be attributed to insufficient voltage being fed into the CRT...this might also explain why the HD is not powering up?

But where on earth would one find a new power/sweep board for this unit?

I have not yet had an opportunity to give it a solid whack yet...but shall report back with my findings!

Hard to tell from static pictures, but yes, I'd be looking at the power/sweep board first. If it works fine after a short warm up, bad capacitors exhibit that kind of failure. A bad solder joint could do the same, (arcing for a while and basically welding itself together) but a good whack will often break them loose again even when warm.

I hadn't noticed you are from Australia. As you can see, I'm in Perth if that helps.

 

[Rusty33]

Hard to tell from static pictures, but yes, I'd be looking at the power/sweep board first. If it works fine after a short warm up, bad capacitors exhibit that kind of failure. A bad solder joint could do the same, (arcing for a while and basically welding itself together) but a good whack will often break them loose again even when warm.

I hadn't noticed you are from Australia. As you can see, I'm in Perth if that helps.

Unfortunately my whack may have been a little too bold - I not only shorted the system (and its fuse)...but also blew my home's main circuit breaker...oops!

I now have myself a wonderful vintage beige doorstop.

I'm tipping that we may now be outside the realm of repairability...although i've yet to open it up and have a gander. Given that Adelaide's vintage mac part offerings are meagre at the best of times, I might be hard pressed to bring her back from oblivion...

 

[MacTech68]

Ouch. I'm almost sorry I suggested it.

It may not be a complete loss, but would need diagnosis.

How does freight to and from Perth sound, given that the box would need to be larger than the machine with lots of buffer around the machine?

 

[Rusty33]

Ouch. I'm almost sorry I suggested it.

It may not be a complete loss, but would need diagnosis.

How does freight to and from Perth sound, given that the box would need to be larger than the machine with lots of buffer around the machine?

Oh absolutely not a worry at all - I very much appreciate all of your advice, and the machine was, after all, on its way out!

It had a good run there for a while...but as they say, all good things must come to an end!

 

[MacTech68]

An update to this saga - the machine has now been repaired and returned to it's owner.

The whack to the side of the machine resulted in the rectified 354 volts DC being fed directly to the Switch Mode Power Supply's controller IC ( a TDA4605). Sadly, these ICs are no longer manufactured and are no longer available from the more 'reputable' component suppliers. I sacrificed a replacement from one of my own Classic IIs which has suffered a rather bad Maxell battery leak.

The tragic tale is that the inside of the rear case is sprayed with a conductive coating which acts as an RFI shield. Normally, the Analog Board which sits vertically adjacent to the left side of the machine is protected by a polymer insulating sheet. At some point in it's life, this sheet was removed and not re-fitted (before the current owner bought the machine) - hence the ability of the inside of the case to short out the Analog Board. Be aware, folks, those sheets are there for a reason.

Aside from the above damage, the capacitors on the output of the power supply had leaked all over the analog board. One particular fast rectifier diode's leads (DP7) were so badly corroded that I decided to replace it - an MUR420 appeared to be a suitable replacement since the original EGP30D seems to be no longer available.

After removing those and other components (including the transformer and a 79L12 TO-92 regulator) to clean the corrosive electrolyte away, replacing the capacitors with new ones and scraping away the corrosion present on the tracks on the solder side and applying solder to the resulting bare copper and replacing the insulator sheet from an old machine, she powered up and booted to a lovely clear bright focused display. A tiny tweak of the voltage adjustment, and all was well.

Some pics to follow soon.

 

[MacTech68]

First pic is what greeted me once the rear case was removed - this black mark was the result of the rectified AC shorting to the TDA4605 - normally fed via RP14 and RP44 (220k ohm and 22k ohm resistors), so instead of about 2 volts, the poor IC got 354 volts DC to Pin 2.

The bad capacitors as a group on the output of the power supply section. What I found disturbing is how far the leaked electrolyte travels across the board. Whilst not clearly visible in this picture, it had completely covered all three fast rectifier diodes - DP5, DP7 and DP6 which is mounted on a heatsink (upper right of the picture). Both DP5 and DP7 had their painted markings etched away, pins on the transformer were showing copper oxide as was the 79L12 (negative 12 volt regulator extreme bottom right corner in the pic). The metal mounting rail on the right side of the pic was covered too and had seeped underneath it.

Overview of the power supply section showing the poor TDA4605 (roughly center of pic). This picture really doesn't do justice to what it really looked like. Everything in the left half of this picture was covered in the oily looking leaked electrolyte.

More to follow
[doublepost=1483160609][/doublepost]A deep breath later and the disgusting smell of rotting fish (I knew that was coming) and the bad caps were removed. You can see the oily electrolyte pooled under the removed capacitors. After looking at the board in this state, I removed more components for individual cleaning and inspection.

After removing the transformer (ZP-1), the rectifier on the heatsink (DP-6) and the AC filter cap (CP-1), which surprisingly checked OK, I vigorously cleaned both sides of the board and then rinsed the board a few times. The tracks on the solder side then revealed their damage. Corrosion had begun underneath the green lacquer. The lacquer flaked away from the copper track revealing a very dirty copper with very dark spots in places. Apologies that the close-up view is out of focus. By this stage, I had already scraped one track back to copper and cleaned it with solder and desolder braid

The component side of the board after the first cleaning cycle. At this stage, I still hadn't removed the choke (LP4) or the 79L12 regulator (IP3).

More to come
[doublepost=1483161341][/doublepost]After taking a short break, I returned to find the following. The choke (LP-4) which is encased in heatshrink tubing was leaking electrolyte
I removed the choke and flushed it several times back and forth (or top to bottom) purging it of the dreaded electrolyte and left it removed to dry. At this stage I also had a closer look at surrounding components to see what else might need special attention. The 79L12 regulator (IP-3) in the bottom left of the picture was the only one that looked bad. I didn't have a replacement, so I scraped it's leads with a scalpel and inspected it with a jeweler's loupe - it appeared to be ok.

more to come
[doublepost=1483161925][/doublepost]Next was the part I hate doing, but I believe if it isn't done, the corrosion will just continue to eat the tracks. I used a small, blunt, flat-bladed screw driver to scrape the tracks where the copper was dirty and anywhere black spots were visible. This required removing a fair amount of the green lacquer in some places.

After getting the tracks back to a nice shiny copper color as best I could, I applied solder to those areas and then used de-solder braid to attempt to remove all the solder whilst using the weave of the de-solder braid to lightly scrub the exposed tracks.

This first pic shows the result BEFORE cleaning the solder flux away. The second pic shows the board cleaned yet again, removing said flux.

One more to come

 

[MacTech68]

By this stage I was feeling reasonably confident that I'd done the best I could to prevent further damage from the electrolyte. All that remained was to re-check the individually cleaned components, refit them and add the replacement parts.

The leads of DP-7 looked so bad that I decided to replace it - but with what ? The electrolyte had done such a good job of etching the silver painted markings from the component that I had no way of knowing it's original spec. Luckily, I had a circuit diagram that revealed DP-7 to be an EGP30D which I couldn't find in stock anywhere. A comparison of the data sheet allowed me to hone in on an MUR420 which I found at element14 (Stock #1625175). The original DP-7 is mounted well above the board so requires pre-bending the leads and soldering so it sits away from the PCB.

Finally, everything was refitted. Following all this work, I held my breath and powered in on. It chimed, booted from it's internal HD and produced a lovely, bright, focused screen. I checked the voltages at the hard drive power connector, and tweaked it a little higher to 4.95 and a little over 12 volts. Silly me, however, forgot to take a picture of the working machine - so you'll just have to believe me.

Fin -

 

[Rusty33]

The little Mac that could(n't) arrived home safe and sound! A happy end to this story - made possible by one of the amazing members of our community. Looking forward to another few decades of 'Classic' computing. Thanks again MacTech!!!

 

[keysersoze]

Awesome job! Great pics too, of the process.