A Technological Appendix to Postblogging Technology, October 1953: Transducer Days

 

With the very low key add for the IBM 650,  "the first mass--produced computer in the world,"we take another big step in the direction of home computing. With the bizarre use of a Mark 14 bombsight as a frame to describe the workings of the bellows in a modern pneumatic system points us towards AIRPASS. With all that going on, Aviation Week has a pictorial for us showing how transistors are made, and everyone in avionics seems to have a transducer on ad this month. And something strange is going on. By this I do not mean the Wikipedia article illustration.

According to the Wikipedia Commons credit, is from the Cushing Memorial Library and Archives, and I am guessing that it is from the early Sixties. The obvious sociological question I have here is why the two operators in our retrospective view of early computing are male, while at the time operators of complex technological systems consistently code as female. It's a very striking change that I've worn out the electrons commenting on, because I want some smart person who isn't me to do all the hard work of coming up with an analysis of it. 

What I mean, rather, is the default assumption that a "transistor" is made of germanium, in a month in which piezoelectric transducers are being pushed heavily in the advertising space. Only 40 tons of germanium were mined "by the end of the Fifties," per Wikipedia. (Or, in 1998, germanium cost $800/kg, silicon, $10/kg. We are not getting to the Information Age using the 50th most abundant element in missile guidance systems.Crystals of various germanium do appear to have piezoelectric properties, but I'm not sure anyone knew that in 1954, and in any event quartz is a satisfactory piezoelectric material and is as common as dirt, so that would be what we would use here.  

But first, before the jump, something for 1954 from the Paul/Ford studio, although not obviously electronica, which word I apparently can't use because "electronica" is a 1990s music genre and holy shit look at this Wikipedia listicle. 

Silicon (Element 14), is the eighth most abundant element of the universe, with "more than 90% of the Earth's crust consisting of silicate minerals," so it's just darn lucky that we found a use for it in the vital field of watching Youtube videos at home. (Also, some other stuff.) In 1940, Russell Ohl discovered silicon's photovoltaic properties, which have come up in recent (ie in 1953) discussions as well. Immediate interest focussed on its use as a radar detector rather than as an "electric eye," but it certainly had potential there, and some attention from the solar power enthusiasts who were not convinced by the prospects of blue green algae. Per Wikipedia's potted history, Morris Tannenbaum made a silicon junction transistor at Bell Labs in 1954, and in 1958, two more Bell Lab workers discovered a way to grow a layer of silicon dioxide on a pure silicon chip  to prevent contamination during diffusion processes.

This is evidently an oblique way of referring to some common doping processes. Again, credit is given to Bell lab workers who discovered that the silicon dioxide layer mediated the passage of various dopants into the crystal, but the silicon dioxide layer has a functional purpose in "passivating" silicon transistors. Some of this is in a patent emanating from Sperry Gyroscope, credited to Robert Woodyard and issued in 1950. Sperry Rand being much harder to patent pirate than some sad European worker (to include French Westinghouse, not to be confused with Westinghouse France, which was probably convinced that it invented the transistor for as long as it existed as a corporate entity). If all of this sounds a bit loosey-goosey, it is. I'm trying not to push my very limited knowledge of the techniques of modern chip production here. I've got quite enough on my plate trying to understand the Dominion of New England this week, and the point here is a historical discussion. The issue seems to be that the evolution of the silicon transistor specifically is bound up with the advance of theory and methods in the field of transistor doping. Silicon crystals do not need to be doped to be used as semiconducting devices, specifically as transducers, but do need to be doped to be  used as transistors. The step from using silicon crystals to produce sound under electrical stimulation, and vice versa, to using solid state silicon transistors as amplifiers on telephone circuits seems obvious, leading to frustration with making the amplification reliable with actually existing silicon devices, leading to the use of doping to make it reliable; leading to silicon chips supplanting germanium ones. 

Is that a story? Probably: The Wikipedia article on the history of the transistor jumps all-in on the transistor radio with products on the market from IDEA of Indianapolis, Sony, and Chrysler/Philco (a car radio) in 1954--55. I'm obviously getting a bit ahead of myself here; we will probably hear all about it in the next twelve months or so, and the reader does not need to hear me repeating what Wikipedia has to say. At some point in the next few years the silicon chip is going to be the latest thing, but not quite yet! So I will shut this discussion down at this point and move on to the edifying story of James II's best men versus the "Puritans of Massachusetts."