Control revisited

Some personal recollections of automatic control before 1947. My first contact with a process control system occurred when I went to work in the Physics Laboratory of Dow Chemical Company, Midland, Michigan, in the summer of 1936. This laboratory was then headed by Dr. John Grebe. I was assigned as an operator of a very small pilot plant which was reacting chlorine and acetylene to make what was called acetylene tetrachloride and, I guess, would now be called 1, 1, 2, 2-tetra-chloro-ethane. The automatic control on the system consisted of a pressure reducing controller on each of the tanks containing acetylene and chlorine. Glass U-tube manometers containing liquid acetylene tetrachloride, measuring the differential pressure across orifices in each of the supply lines, were used to measure the two supply flows. A glass tube served as a liquid level gauge in the reactor and my job on the 12:00 midnight to 8:00 a.m. shift was to maintain the supply flows at the specified values and to withdraw the acetylene tetrachloride product when the level in the reactor exceeded a certain value. The top of the reactor was connected to a stack pipe going out the roof of the building and from time to time a sample of effluent gas was drawn into an orsat gas analyzer which I operated by turning stopcocks and raising and lowering mercury filled flasks.