My return to Classic Cmp - San Diego, software archive, etc.

On 18 Dec 2008 at 6:05, Dave McGuire wrote: Almost all were for process temperature control--from ingot soaking pits to solutions on the continuous galvanizing line. "Rugged" was probably more important than "stable", but I think it was more to avoid the issue of once-monthly replacement of dry cells and maintenance of the associated calibration mechanism. The big square black 1.5v cells (made by National Union?) were used extensively, instead of the smaller round No. 6 ones. On an open mill floor, temperatures could go from sub-freezing to Saharan. Dirt, of course, was everywhere. Right about this time, I recall that IBM had installed a computer for process control in one of the new plants. Sadly, I don't recall any details. Perhaps a 1710 used with the BOF. Like a lot of American heavy industrial operations of the time, the mill was a wild mix of period technology, much of it pre-war and much running from 25 Hz power. Up until that computer, it seems to me that the most sophisticated stuff was pneumatic--full analog computers, with P-V and V-P converters, pneumatic amplifiers, square- root extractors, differentiators, integrators, etc. "Wiring" done with quarter-inch copper tubing and compression fittings. If the air supply was kept clean, the pneumatics were pretty much indestructible. Earlier controls were mostly hydrostatic types of the diaphragm-and- jet type. Keep the tank filled with Pydraul and you were good for decades. There were very few electrical-only process control systems, mostly restricted to non-critical processes. Cheers, Chuck (I hope there's enough "computer" content here not to irritate)