Monthly Archives: November 2009

It’s Alive!

So, after some confusion around disk image transfer shenanigans related to the way Microsoft changed UAC in Windows 7, the Apple III is in good, running condition once again.  As I was digging through the cases, I noticed something I’d missed the first time around: holes.  Specifically, a series of holes Apple drilled through the pressed aluminum case of the two later models I have:

Cooling holes present in later 5V Apple III and Apple III Plus models are also seen here in this 12V machine.

Cooling holes present in later 5V Apple III and Apple III Plus models are also seen here in this 12V machine.

These are not present in the earlier Apple III I have:

The case from an early 12V model, serial number 018052

The case from an early 12V model, serial number 018052

What’s interesting to me about this is that these holes exist in the case of one of the two 12V models, indicating that Apple knew about and was trying to address design flaws even before they officially started the replacement and upgrade program. Remember that before Apple released the III Plus, they first rolled out a revised Apple III which featured some significant changes to the machine’s design, including the aforementioned holes, as well as a new more efficient 5V power system. But the presence of the cooling holes in this 12V case (serial number 050756) shows that Apple were aware early on of the troubles plaguing the machine.

Two 12V Apple IIIs, one with the cooling holes drilled in the cast iron case.

Two 12V Apple IIIs, one with the cooling holes drilled in the cast iron case.

At any rate, after a handful of successful disk image transfers, I loaded up the Apple /// Dealer Diagnostic disk.  I’m not sure if the obvious issues with the color video modes are the fault of the Apple III or the monitor it’s plugged in to. I’ll need to dig further into that:

Color video mode test pattern.

Color video mode test pattern.

But, it was very nice to see this at the end of the rather extensive diagnostic:

The Apple /// Dealer Diagnostic: complete and successful.

The Apple /// Dealer Diagnostic: complete and successful.

Maybe not so bad after all

The Apple III, considered to be Apple Computer's first commercial failure.

The Apple III, considered to be Apple Computer’s first commercial failure.

If you’re like me, you’ve come to accept the popular wisdom that the Apple III was a deeply flawed machine, crippled from the beginning and doomed to failure before it left the design stage. We’ve all heard the rumors of the fanless case causing excessive heat build-up and leading to warped PCBs that dislodged chips from their sockets.  It’€™s a popular tale in Apple lore that one of the manuals shipped with the Apple III actually suggested lifting the main unit six inches off the table and dropping it to reseat loosened chips; and perhaps it’€™s true – I’€™ve never seen it myself.  Excessive price and a late start into a business market already cornered by IBM and its legion of clones further limited sales; by the time the III was quietly pulled from Apple’€™s product list in September 1985, it had sold only 65,000 units (or 75,000.  Or 125,000, depending on who you talk to).  Apple’s attempts to address the flaws (real or perceived) were too little, too late and the updated Apple III Plus sold even worse than the original.

But was the final product really as bad as everyone said?

While searching for Apple III technical documents and information to help the restoration of my Apple III, I happened across this article at AppleLogic, in which an electrical engineer with experience designing hardware for the Apple II takes a second look at the so-called design flaws in the Apple III. If you’€™re like me, the conclusions he draws from his analysis might surprise you.

Apple III restoration cont’d.

Okay, so more research is never a bad thing.  I dug around a little and came across a copy of Sun Remarketing’€™s Do-It-Yourself Guide for the Apple III.  You can download it in PDF from 1000bit’€™s documentation archive here, and if you ever have to do any hardware troubleshooting on a pre-€Plus version of an Apple III, I highly recommended consulting this guide.

Serial #018052, which I’d decided to try to bring back to life first, is one of the early 12V 128K Apple III units, released just after the recall, but before the introduction of the ‘revised’€ 5V Apple III.  This isn’t a III Plus – those came much later, in 1984. Apparently the III Plus was only on the market for a short time, the damage in the public’€™s perception of anything with an Apple III label having already been done, so it sold in very low numbers and while functionally not much different, today it’s a much rarer commodity and more highly desired by collectors.

At any rate, #050756 is also a 12V unit, so I’m thinking I could scavenge it for working parts. Serial #051780 is a 5V machine, and for the purposes of fixing #018052, of little use.

I ran the in-built memory diagnostic mentioned in Sun’€™s guide and discovered that two of the RAM chips on the Apple III Main Memory board were bad.

Bad RAM chips, as indicated by inverse blocks in the memory map.

Bad RAM chips, as indicated by inverse blocks in the memory map

I borrowed two chips from 050’s board…


A healthy RAM board passes the diagnostic

As an aside, I also discovered that the memory board in 050 was installed backward and may have been at least partially behind the unit’s earlier complete boot failure. Getting 050 up and running again may not be as daunting as I’d thought.

Now to address the 018’€™s floppy issue. Again, an easier fix than I’d expected. The floppy cable was plugged into the wrong port on the analog card. Moving it to the correct one brought it back to life, at least to the point where it tries to read a disk at power on.  It may have other issues (timing is apparently a big one with these drives), but until I fire up ADTPro and transfer over some disk images to real floppies, I won’€™t know.


The drive tends work better when it has been properly cabled.

 Finally, there’€™s the issue of the case. This is mostly a cosmetic thing and one I’€™m not as worried about, though I am looking forward to experimenting with Retr0bright.  The cases haven€™’t been too badly abused, suffering only light scratching, dirt and the requisite yellowing that comes with age.  I picked what I thought was the best of the three and did a little basic cleaning with some 409 and paper towels.  The final result?  Not too bad, I think.


Not as pretty as the day it was manufactured, but closer than it was a few days ago.

Still to do: dig out the Apple Monitor /// that’€™s supposed to go with this thing; whip up a batch of Retr0bright and get to transferring disk images.

Given the condition of 050, that will probably be my next Apple III project.  I suspect tracking down the replacement parts needed will be the hardest task. Anyone have a good source for Apple III keycaps and 12V RAM chips?



The culprits

Apple III Restoration: Hey, It’s Progress

The information about these machines was at least partially incorrect, and that’s a good thing.  All three units did in fact power up, though none of them were completely successful.  The first Apple III I plugged in, serial #051780, showed some good signs of life with both the keyboard power lamp and the internal LED (CR7) lit up.

The floppy LED didn’t blink and I didn’t get any video output, though.  The attempt to reach the MONITOR prompt returned a screen full of flashing garbage graphics. Not a good sign.

Power light - 051780

Power light – 051780

The second system (050756) also responded with warmly glowing power lights, and again, no video.  This time, I got no response when I tried to invoke the diagnostic.  Gently wiggling the cable to the composite monitor gave brief flashes of static on the screen and little else.  Perhaps a failed video ROM? Again, no drive activity was seen.  My sense of impending doom was growing.

ROM test - 051780

ROM test – 051780

The final III, oldest of the trio (serial #018052), is also the one to have best weathered the ravages of time.  All keycaps are present on the keyboard, and it’s got the least build up of dust, dirt and what appear to be wood shavings.  On the other hand, several of the keys became the final resting place of partially digested worms and whatever else bird eat and then poop out (see below).

Fingers crossed, I hit the power switch and said a little prayer to the silicon gods.  This machine got furthest of all, reaching a flashing cursor prompt (always a good sign) under the word RETRY.  According to this article, this may in fact be expected behavior.  The diagnostic appears to complete without error, and a CTRL-RESET got me to the MONITOR prompt.  Still no drive activity, however.

So now, it’s time to dig into the plethora of Apple III literature available around these nets and maybe learn a thing or two.  If all this takes is a new internal floppy mechanism and a dose of Retrobrite, I’ll be a happy computist.

RETRY - 018052

RETRY – 018052

Apple III Restoration Project

One of the great things about the retrocomputing hobby is that once your friends and family learn about your (admittedly slightly odd) habit, you will probably end up inheriting most or all of their computer stuff over the years.  While it can also be a bit of a curse – no, Aunt May, I’m still not interested in that 486 that’s been mouldering in your closet for the past decade – if you advertise properly, you can end up with some neat stuff in your collection.  This has definitely been the case for me, as donations from family, friends and friends of friends have padded out my storage space nicely.

Take, for instance, this set of three Apple /// systems, recently given to me for free, on the condition that their owner never sees them again.  They’re all in pretty bad shape – none of them power up, keys are missing and they’re generally filthy – but I love a challenge.  Now I know what I’ll be doing with whatever spare time the wife gives me over the next few days (or weeks, if they’re really bad).  My intention is to cannibalize the three into a single, working unit.

So let’s get started, shall we?  Here’s some pictures of the initial donation.  Along with the CPUs, I was also given a milk crate full of external Apple /// floppy drives, an AppleColor Composite Monitor IIe and, strangely, a Microsoft Premium SoftCard //e.  While the Z80 card won’t work in the ///, it should fit nicely in one of my needy IIe systems.


This /// has not aged well.

This /// has not aged well.

A milk crate full of Disk /// Drives

A milk crate full of Disk /// Drives

Internal case corrosion

Internal case corrosion

Yep, that's bird poop.

Yep, that’s bird poop.

A trio of IIIs

A trio of IIIs