Cyber Toast

It all started at 430am Thursday, when the dogs suggested it was a good time to go out in the backyard. This is a ritual that occurs a few times every night. Jan and I split the task 50-50 by waking each other up to the words, "It's your turn I did it last time." This is followed by the ritual question, "Oh ya, I just did it, when did you last go?" On Thursday, her 315am beat my 145am, so it was, indeed, my turn. Before I got to the back door I was greeted with the acrid smell of burning circuit board.

The silence from the hard drives was deafening. The room wasn't filled with smoke but it was pretty clear something had cooked and I have a problem. I clicked off the switch on the power bar and let the dogs do their business. One problem per night only please.

So, Rubik VI, hits the bench, the early diagnostic information suggests a power supply problem (burning and no power). The Rubik VI computer lives without a case cover to improve his ventilation (as his Pentium sized noodle does not have a cooling fan). All that is required to inspect a power supply is to pull the power wires from all the components (including two from the main board; items P8 & P9), and the higher voltage wires from the main power switch (four of them plus a grounding wire). Then remove the support screws from the power supply housing, break all the seals that say "No user serviceable parts inside (often a mis-truth) and you are in.

Rubik IV, a laptop 386 I bought in 1992, fizzled out the same way one day in 1994. After sawing through the welded plastic case of the transformer assembly, I repaired the unit by installing a replacement 15 cent fuse I bought at Radio Shack. I think that "No user serviceable parts" is more often a wish than a promise. That 386 cost an arm and a leg in those days! It still works for me today at the office as a primary word processor. Though last week the LCD back light failed and I am going slowly blind trying to keep up with my notes. But this is another story.

Rubik VI certainly did have a problem there in the power supply. Two of four voltage regulators had cooked themselves right through the perf board. Fortunately, I noticed, that the fuse assembly had blown as well. I might still have a computer. What caused this, I don't know. Granted the cooling fan in the supply housing wasn't working, but it had seized up over a year ago. Maybe it is a long term effect of overheating, maybe the regulators have a life of five years. Who knows?

In the strange hope that when a main component goes down, it leaves the rest of the system intact, I set out to pick up a replacement power supply. An after work walk of Friday put Jan and I on College Street just west of Spadina. It's a great neighborhood. Hundreds of wondrous little shops with everything you could imagine. I even found a stained glass supply store that stocked the 1/2 inch caming I needed to complete a window right beside the store that sold me the power supply. If there had been a bread makers supply store next to a do-it-yourself wine store, next to a sailing riggers shop, next to a philosophical bookstore, next to a Home Depot, then I would guess they built the whole street for me!

Back on the test bench on Saturday things were not going well. No I did not even try to fix it on Friday night and yes that was out of character. It seems a low pressure zone from the West, Northwest brought one of those fast moving spring migraine headaches with it on Friday afternoon. I was able to find the power supply and the caming while highly medicated, slurring my words and making very little sense to anyone (according to Jan). For some reason she wanted to stop at the Clarke Institute of Psychiatry on the way back to the car. I think she might have been trying to see if they would keep me. All that happened though, was a nice man gave me a free coffee (Mochachino) and we were on our way. Really, I'm not making this up.

Things were going so poorly, in fact, that most of the afternoon I tried to distract myself from the untimely demise and seemingly over shocked state of Rubik's recovery. Strangely, I felt alone and disconnected from the world. I was being past-by, e-mail collecting, web robots searching out information for me, people visiting my web pages, and me sitting in a cave looking at the high tech obelisk like those cave dwellers in 2001, not knowing about the flood of conservations and life just beyond my grasp.

Could Ken have neglected to make a comprehensive wiring diagram when he pulled off all the wires? He didn't even make a rough sketch, did not even glance at the wiring in some areas. What was he thinking? He thought that after he so successfully completed the last major upgrade (new motherboard, etc) that everything would be a piece of cake. He didn't know which way the wires were supposed to go! The best way to describe my reaction is PTSD (Post-Traumatic Stress Disorder). It goes way back to Rubik II.

Rubik II was a rebuild of RUBIK I, an original Radio Shack Colour Computer in 1980. The F-series mother board was one of the best of the series and the computer spawned a cult following of publications from "What is a computer" to "Circuit board re-designs to support bubble memory." (What ever happened to bubble memory?)

I had just finished rewiring the display circuits to by-pass the RF modulator so I could run the machine on a real computer monitor instead of the television set. To effect the final adjustment I had to solder a piece of hobby wire from the video display take off over top of the mother board and solder it to a land on a little chip called a quad-inverted-nor-gate. I was fly'in. It was a remarkable success. I now had a computer that would work with a monitor and I could actually make out the letters of the words I typed into the work processor. This was high-tech!

Then the trauma. The next project was to wire a new full-stroke keyboard into the whole system. The Rubik I had a membrane keyboard of little square buttons, nothing like a type writer, most like a calculator. I wanted to separate the keyboard from the computer so the box could sit under the desk and be out of the way, while I worked on the keyboard and looked at the monitor. Just like all systems are built today. This meant sawing off the thirty five wires connecting the keyboard to the main board on an piece of ribbon cable. I thought five feet of thirty-three wire ribbon cable would be a little too much (it would have been about five inches wide), so I twisted thirty-five hobby wires together and soldered them to the connector on the keyboard via some rs-232 connectors and directly soldered the wires to the main board with the monastic patience that you can imagine this required. To permit the whole thing from just falling apart in my hands, I decided to set the whole main board wire assembly in epoxy. This was a fine idea and I was proud of my art student ingenuity beating out the typical engineering of the time. If only the monastic patience had stopped me from pushing the power button before the epoxy was completely dry.

It took me five weeks to trace how far back onto the board things had died before I ever saw the green phosphorescence of Rubik I's noodle again. Two PIA's burned, one Proprietary Radio-Shack ROM (replaced by mail-order from Texas), and a variety of little exotic chips that I found by going to a strange micro chip dispensary in Hamilton. A hole in the wall on a side street and filled with the pocket protector crowd, a single man stood on a raised platform in front of hundreds of bins. It reminded me of a pharmacy. Geeks would call out chip numbers, and he would reach behind him (sometimes without looking) and hand out the high tech bug. Mine was a P456033-a or something like that. With a knowing smile he dropped it into my hand from the platform. Maybe it was because I was a kid, or the way I over-pronounced the P456033-a, but he knew and said "Here you go" with just enough disdain to say "Art students shouldn't play with electronics, son."

I'm sitting on the carpet looking at the scattered boards of Rubik VI, seventeen years later and I feel worse than the kid who didn't let the epoxy dry. My last gasp this afternoon was to take off the mother board power supplies and wire the main on-off switch in complete contradiction to the usual wire-colouring protocols and the wiring diagram on the power supply unit, throw the switch and see what happens. It was a gesture of hopelessness. The first wiring produced nothing. Not a pop, not a sound, if something was happening, it was not electric. The second wiring, still conventional, but seemingly a less logical reading of the wiring diagram blows circuit breaker in the power bar. Thinking that perhaps I have a bad power unit, a $69.00 lemon, I wire it to the switch and hope for the emotional satisfaction of a bang, a puff of smoke and lights out. Out with a bang, forget the whimpering! I stand well back and use my umbrella to throw the switch. What happens? The fan starts.

Fifteen minutes later I do it again and the whole thing winds up, lives and boots so fast that I can even read the POST results on the screen. Windows 95 just drops me back where I left off. This is not supposed to happen. I should be out for the next few weeks buying a new mother board, then a new processor, then a new video card until the only burning plastic I can smell is coming from my Visa card. But instead I'm just sitting here typing e-mail. Am I happy or what! Hope you have a good week, it's the future now and things are better than they were.