A second bulb? Impossible!

[mystories]

This is a guest story, one of my dad's.

My dad was working on a problem with a satellite. Like most satellites, it was a complicated, expensive affair, but this particular model didn't last as long as it was supposed to, and my dad was asked to analyze the problem and suggest possible solutions.

The satellites generally failed when a particular optical encoder would stop working. This consisted of a light bulb, a slotted disc, and a light sensor. As the disc rotated, the sensor would see the light get brighter and dimmer as it passed through the slots. The problem was, this type of bulb had a filament that would evaporate, causing the glass to silver, blocking the light. There was an established procedure to deal with this when the light output got too low. They'd crank up the voltage to the bulb, which would then emit more light. Unfortunately, it would also evaporate faster, and they'd have to crank it up again. After a few cycles of this, the bulb would burn out and then the satellite didn't work any more.

My dad made a simple suggestion - add a second bulb. He was told that this would involve too much re-engineering, making a second hole in the casting to hold the bulb, switching circuitry and another driver transistor for the additional bulb, so the whole circuit board would have to be laid out again. Too much to design, test, and get approved before the next launch.

So my dad went and pulled the engineering drawings for the satellite. Lo and behold, there already was a second hole in the casting, in just the right place. And there was switching circuitry and a place for an extra driver transistor on the circuit board! A few phone calls confirmed that the original design had called for a second bulb, but it had been deleted for cost reasons.

Armed with this information, he was quickly able to get the second bulb approved and installed in time for the launch, effectively doubling the life of the satellites.