semiconductor industry building crystal growth reactors
Synthesizing new semiconductor material (e.g. Zn3P2) is more fun. An occasional explosion here and there.
Our crystal growth systems at the time consisted of a vertical furnace and a very slow moving timing motor. So, in one of my first real jobs I had to learn some rudimentary glass blowing and put a glass tube with a tiny hole in it with some synthesized material and then into a quartz larger tube that had been necked into a smaller sized. Sealing was done under high vacuum using a hydrogen-oxygen torch for purity. I learned in about 6 weeks, did it for a while and trained a replacement, then I went on to do other stuff in the lab. Which was to make what someone else bought work (a MINC-11) and it required programming,
The ya-hoo;s that built the furnace lab, put in "water to air heat pumps" in the room with TWO thermostats. Some of the other yahoo's that speced some of the other lab sections, but in 60 A 208 single phase 3 wire power for 4 wire power. Then the same yahoo's tried to use the heat pump water, usually about 90F, to cool mostly diffusion pumps that required 60F water, by boosting the pressure of water in a 6" pipe, so it would fit in a 3/8 tube. Total disaster except for the fact that they had the forethought of running a drain.
It wasn't unusual for one piece of equipment to have a 60 or 90 A 208 3 phase service. One piece that was added very late took 200 A, 208 single phase.
They put timers on heat pumps and the building leaked from the roof and the wind. They forgot that the AC drains had to be pitched and they ran them into the sanitary sewer/
The isolated slab was uniquely built too. They poured the entire slab for the building, and then broke out the middle (about 100' x 50') isolated it and re-poured it.
Sometime after the building was built, I discovered a possible defect that could exist in up to 420 electrical outlets. One had destroyed a computer because the ground could let go if you plugged two items in the outlet. They did fun stuff like put the hall receptacles used for things like floor buffers on critical lab circuits.
One piece of nonsense was to see if a real re-circulating cooling loop could be put in, because we had so many pieces of equipment that ran potable water for cooling. Every piece needed a water filter. It was determined that it was cost effective to put a system in, BUT we didn't pay for water we used. In fact, our building didn't even have a water meter until many years later.