Washington Apple Pi Labs is best known for its top-notch testing expertise, honed to a fine edge over the past 20 years. Our motto is: "If we can't break it, it must be OK." Proof of this expertise will be for sale at the December 1998 Washington Apple Pi Computer Show and Sale, better known as the Garage Sale. We broke some of this stuff a couple decades ago…
But we have other skills, aside from writing in the first person plural: we are also world-renown experts at solving all manner of arcane technical problems. Occasionally, we even handle the mundane and ordinary.
A case in point: one day, a distressed damsel came to us with a computer that wouldn't boot. Nor would it glove, belt, or verbize any other piece of clothing. In short: when you pressed the Power On key on the keyboard, no power came on. This change happened suddenly: after working perfectly since purchase, the machine simply wouldn't turn on; there was no warning or transition. This alone was strange: Macs tend to die tragic, movie-style deaths over several months or years and then, just like in the cartoons, magically spring to life again in a new movie (or after repairs with Disk First Aid).
All the normal checks were fruitless: the machine was plugged in, the power switch was on, the keyboard was plugged in, etc. This particular machine was a Power Computing Power 100, one of the very first Power Macintosh clones, and functionally something of a cross between a Power Mac 7100 (it is a desktop machine) and a Power Mac 8100 (the motherboard has many of the AV features of the 8100). They are not known for strange, unexplained failures, so we did a comprehensive check:
Nothing worked. Despite the best efforts of a coterie** of highly trained, experienced people, representing almost 15 years of post-graduate education, the Power Computing 100 just sat there. While we don't like computers to mock us, a little anthropomorphic mocking would have been appreciated: we were desperate, and looking for any signs of life. Even abusive signs. ("If a Mac can sit," we observed, "it should be able to mock.")
At this point the distressed damsel looked surprisingly cheery. She observed that, if the Power 100 really was dead, she would be forced to replace it with a Power Macintosh G3. Oddly enough, she didn't seem too upset at the thought of being "forced" to get a G3.
We, on the other hand, decided to take desperate measures:
In fact, the Power 100 was incredibly clean inside, with none of the usual dust bunnies. (In some machine, dust bunnies are the size of small horses.) There were no scorch marks indicating a short, no smell of burned electrical components, nothing out of the ordinary. The Washington Apple Pi Labs team was getting a bit miffed: we expect this kind of behavior from those machines but Macs, and Mac clones, are supposed to behave better.
A conversation on the TCS (the Pi's computer bulletin board) alerted us to the possible need to reset the "Cuda." Exactly what the "Cuda" might be (aside from a nickname for a 1960s Dodge sports car) was a mystery. Fortunately, the TCS also had a file called the PowerGuide MLB 1.0.2 (File Transfer Area 3, File 704) that is an electronic guide to the main logic boards on Power Computing computers. The guide didn't explain what the "Cuda" was, but it did show us where to find the Cuda reset button on the Power 100.
So we pressed the red Cuda button to reset it. It isn't clear if this resets it -- the PowerGuide just shows where it is, not what to do with it. We pressed the Power On key and -- nothing. We were getting annoyed (but the distressed damsel was looking happier and happier).
Knowing that the Power 100 uses many standard parts also used by those other computers, we decided to rip out the machine's power supply and see if we could get a replacement. As soon as we removed the power supply we noticed: a battery. A purple battery. A purple 3.6 volt lithium battery.
One thing that can cause a Macintosh to fake its own death is: a dead battery. This is most often seen in old "compact" Macs (128K and 512K Macs, Mac Pluses, Mac SEs, etc.) or in certain desktop models (particularly the Mac IIsi, Centris 610, Quadra 610 and Power Mac 6100). We'd never heard of someone having a dead battery problem in a Power Computing machine, but we decided to test the battery with our multitester.
It was dead. Now, "dead" is a relative term; a 3.6 volt battery reading 2.5 volts will prevent a Power Mac 6100 from booting. But this battery didn't read 2.5 volts; it read: nothing. The Washington Apple Pi Labs team has never seen a recently-working Mac with a completely drained battery; we were impressed, and privately started to refer to the machine as Bunnicula, the battery-sucking vampire.
The formerly distressed damsel was less impressed: after a quick trip to a Radio Shack store and the purchase of Part 23-026, "Lithium Battery: for Apple Macintosh Computers and Wireless Security Systems," a brand-new purple battery was installed, the Power On key was pressed on the keyboard and --
Apple suffered a delay in the purchase of a Power Macintosh G3. The Power Computing Power 100 instantly sprang to life, somewhat confused about the date and location, but otherwise in robust good health. A quick trip to the Date & Time Control Panel, and the Map Control Panel, eliminated the confusion over date and location, and the Power 100 was off and running.
Nothing else was wrong. Nothing else at all: the hard disk drive was in perfect health, all the peripherals worked, the UPS was fully functional. In short: the only problem was a dead $10.95 battery.
Few things in life are as sad as a formerly distressed damsel who will not be (immediately) purchasing a Power Macintosh G3.
* A multimeter is used for testing multis (preferably when they are asleep, though the infants are kinda cute and not as dangerous). It can also be used for checking electrical outlets and batteries.
** We've always wanted to use "coterie" in an article, and this looked like a good opportunity.
Revised Saturday, August 28, 1998 Lawrence I.
Washington Apple Pi