Friday, July 19, 2024

A Technological Appendx to Postblogging Technology, March 1954: Noise About the Type 2001

 

For me, there's something infinitely cosy about the first generation of professors at western Canadian universities. I've lived more than half my life near the campus of UBC, and routinely bike by the old neighbourhoods where these gentlemen built their homes, way back in the Twenties, and there lived reasonably comfortable lives during the upheavals of the Thirties. My case example is usually UBC's Garnet Sedgewick, an English professor for whom the kernel of the modern Koerner library is named, which gives me an excuse to insert the video below, filmed in front of it, with two-for-one Grace Park. This week, however, we're on about Robert Boyle (1883--1955) of the University of Alberta, born in Carbonear, Newfoundland, just in case anyone ever asks you what happened to the Beothuk. Boyle headed the physics department at UofA from 1912 to 1919, and was dean of Applied Science from 1919 to 1929, after which he went to the NRC in Ottawa, retiring in 1948. No family is mentioned, so my "cozy" image of a Boyle family settling into a quiet neighbourhood in Edmonton and growing happily old is completely off-base. 




Oh! And he invented sonar. Like a lot of other people, sure, but I wanted to share that photo and talk about cycling by old houses up  near campus. 
 
So according to this site, upon which this post depends heavily, the Sonar Type 2001, apart from being given a four-digit vice three-digit numbering to indicate how revolutionary it was, was supposed to:
"In summary, under good asdic conditions, the set was expected to detect an 8-knot snorting A-class submarine in deep sea state 2 at 30 miles. At own ship's speed of 10 knots the range was 17 miles and at 20 knots, 6 miles.249" The tag around here for that kind of thing is "Magic Aeroplane," and apologies to Republic Aviation, but the Rainbow is pretty magic. 

So the story is that the Admiralty having been fiddling with rafting and propellers and rubber mountings and anechoic coatings, it was thought that long-distance passive detection was not on. The lads at the Admiralty were very proud of their sonar, notwithstanding it never being accepted into service formally, and going to sea in various ad hoc experimental set-ups for its entire career, not that, in retrospect, this was at all a bad thing! Submarines would shoot the 2001 into crowded target areas with all the electricity in the world, identify targets, and then shoot them at long range with Tigerfish. It sounds like a plant, and a bit above the part of the site that describes the 2001, I learn that RN submarine commanders experimented extensively before and during(!?) the war with jamming enemy arrays with active transmissions. So I'm getting a slight lack of realism here. 

The selling point of Type 2001 as it entered development in 1955 was that, in order to achieve 240 degrees of coverage vice the 40 degrees of the well-liked Type 183, an incremental improvement of a WWII equipment, it used "about 16,000 transistors," Type 2001 incorporated passive detection with a low frequency band able to detect noisy targets out to unspecified long ranges to be compared with the HF band detection of quiet targets at short ranges of "say, 10 kyds." 

The active end of 2001 was a bow-mounted array of hydrophones inspired by German and American work but in absolutely crucial ways completely original and better. (It says here.) The underwater sound team was very smug about making the weapons team move their torpedoes, and the designer about receiving in the 6.5kHz range instead of the American 3.2.  I've found a more lucid account on Reddit which describes how the 2001 leapfrogged the electromechanical beamforming of the German GHG and American BQS-6 by adopting a digital beamform by imposing sequential delays. In layman's terms as far as I can make out, this is just a way of controlling a hydrophone's natural all-direction sensitivity so that it only "listens" in a narrow band, with all the hydrophones in the array pointed in the same direction. The beam is then rotated, and, in particular, can make bank shots off the ocean floor.  

Type 2001 fed to 27 "cabinets," which might be idiosyncratic language for servers. The Sound Room Cabinet, which employed electromechanical analog computing devices, fed output devices including a PPI, four pen recorders, and an Initial Display Device which recorded off a CRT. Plessy Ilford got the contract to build all of this and made almost enough money not to be taken over by a series of competitors that finally became Ericsson. As befit the operator of such expensive equipment, it is related that Valiant's first commander liked to run at 500ft, 28kns, spraying sound around while drinking a large glass of brandy and ginger ale, whilst the more abstemious Sandy Fieldhouse just had a tumbler of sherry. 

 
The Sixties, everybody. 

Not surprisingly in a series of 1973 exercises some South African Daphnes so completely pantsed the nukes that the "transmit" key was blocked off with duct tape and never used again. (It didn't hurt that the Admiralty had some American towed arrays to trial.) Type 2010 gave way to improvements, and the world moved on. This is perhaps inevitable given that materials scientists have discovered how to make magnetostrictives out of alloys of terbium(!), dysprosium(!), and even boring old iron and gallium. This all comes well beyond the ultimate limits of this blog, but what the even heck, my dudes? "Little is known about the toxicity of Terbium. Handling of terbium, like other lanthanides, should be done with care." I can't say that that "little is known" fills me with confidence about the state of research! 

So that would be all, short and sweet, as the best things are, were it not for a chance comment in the Wikipedia article on the history of Plessey electronics leading me to this neighbourhood in a working class district in Liverpool. It's not quite Leeds (since I doubt the contemporary reference will age well), but close enough. This is from a nice website maintained by a retired employee from the Old Days 




And this is Wikipedia's account:

"In 1959 AT&E, later Plessey, became the prime contractor for a new UK air defence system, known by the company under the name Plan Ahead and, from 1961, as Project Linesman.[8] To enable the system to be designed and built without too much information becoming public knowledge, a new factory called "Exchange Works" was built in Cheapside in Liverpool city centre, where young employees were granted exemption from conscription.

At the heart of the system, installed in a huge building in the middle of a council housing estate in West Drayton, was the computer room, occupying an area of around 300 by 150 feet (91 m × 46 m) and filled with around 1,000 7-foot-high (2.1 m) racks of electronics, including mainly the XL4 computer, based entirely on germanium transistors and using a computer language developed at Exchange Works in the 1950s and 1960s.[8] The secure status of the factory attracted many other secret contracts and led to it becoming one of the major designers and manufacturers of cryptographic equipment. Exchange Works is now luxury flats."

So now I say, with some emotional force, what the fuck? There you go, that's a snapshot of an unspecified "crypto" equipment designed and built in the Seventies in a secret factory hidden in a working class neighbourhood of Liverpool. I sure hope that you had to step into a telephone booth and ring the secret number to get to the timeclock! Okay, it obviously wasn't terribly secret by the time that the factory was sold off, but still. Anyway, the important thing is that the whole enterprise was sold off in the defence dividend era and is now, like the Wiki says, "luxury flats." I mean, producing defence equipment for the government in a distressed era instead of gentrifying it and selling it to offshore investors? What were they even thinking? 

No comments:

Post a Comment