Friday, July 08, 2005

Engineering Precursors

As I look back at my youth, I am amused at the little things I did that sparked my interest in engineering, and even gave me some insights into the workings of computers -- even though I hadn't the foggiest notion what a computer was back then. Each time I learned a new physical principle, it fascinated me, and I just had to explore how it could be used.

At one point, I learned how to make an electromagnet. You could wrap thinly-insulated wire many times around a big nail, connect the wire to a battery and the big nail turned into a magnet and could pick up little nails. Disconnect the battery, and the little nails would fall to the floor. I bought the wire and battery, and demonstrated the magic to my younger brothers.

Then I learned about the telegraph and the Morse code, and the history of how these were used to send messages over great distances. Now I was cutting up 'tin' cans to get strips of steel. I mounted an electromagnet and a strip of steel on a block of wood, so that when the electromagnet was energized, the strip would click down on the electromagnet. Another steel strip, wood block, and nails was used to make a switch to turn the electromagnet on and off.

.---- switch -------------------------------- electromagnet
'---- battery ------------------------------- and click-strip

The switch and battery would be 50 feet away from the electromagnet, connected by a pair of wires. When you tapped on the switch, the electromagnet would click 50 feet away. Now all I had to do is teach my younger brothers Morse code, and we could have loads of fun. Well, they didn't think that memorizing a code was fun, so I made a chart for them. That was a little easier, but still they resisted. It was hard to get a consistent rhythm, else a 'dit' and 'dah' could be confused. So I modified the telegraph with a double switch, three connecting wires, and two electromagnets, so that a 'dit' and 'dah' were signalled by separate electromagnets. Years later, I learned that some historic telegraphs were actually constructed in a similar manner.

Then I learned about 'relays' -- the metal strip pulled by the electromagnet could function as a switch. Now, turning on a switch here could turn on a switch over there. Or, you could make it so that turning on the first switch would turn off the second switch, and vice versa. That opened up a bunch of new possibilites.

What if you connected the relay so that when it was on, it would turn itself off, and when it was off, it would turn itself on? The cycle of cause-and-effect would repeat itself, wouldn't it? Well, I built one to see what would happen, and sure enough, I had a buzzer -- the relay couldn't decide whether it should be on or off, so it turned on and off, on and off, as fast as it could.

What if you connected two relays so that relay 1 would try to do the same as relay 2 (on if on, and off if off), but relay 2 would try to do the opposite of relay 1 (off if on, and on if off)?
Then the relays would go through a cycle like this:
relay1 .. relay2
off . . . . . off
off . . . . . on
on . . . . . on
on . . . . . off
off . . . . . off
off . . . . . on
on . . . . . on
on . . . . . off
... etc.

That made an even louder buzz! (With a slower cycle, the relays had more time to turn fully on and fully off.) I was having so much fun that I had to buy more batteries.

Next, I learned about serial and parallel connection of switches. If switches were connected in series, like this --

======== switch1 ----- switch2 ----- switch3 ========

-- then the wire pathway was on if switch1 AND switch2 AND switch3 were on. And if switches were connected in parallel, like this --

=======,--- switch1 ---,
. . . . . . |--- switch2 ---|
. . . . . . '--- switch3 ----'======

-- then the wire pathway was on if switch1 OR switch2 OR switch3 were on. Since relays could be substituted for the switches, endless possibilities lay before me. I struggled to construct interesting and useful machinery with these ideas, but I was overwhelmed. My trial-and-error methods didn't work because there were too many possiblities.

I didn't know it at the time, but I was learning some of the principles of computer logic. But I didn't even know what a computer was, and I didn't have all the tools. Later, in college, I learned about Boolean logic, Karnough maps, DeMorgan's theorem, Venn diagrams -- tools that a designer of computer logic needs. And when I got out of college and into ITT, the first thing I did was to design part of a computer. But the switches and relays were now replaced by transistors.

1 comment:

Rich said...

Ah, I remember doing the relay buzzer as a kid with an electronics kit - though in that case it was a single relay (that had switch contacts for both the open and closed position).

I managed to give myself a shock touching the contacts and was impressed when an oscilloscope found that it was putting out ~260 volts with only a 9V power source. I never really did learn enough EE to fully understand it, but I'm sure the oscillating EM field in the relay has all kinds of propensity for creating high voltage.