Editing Chain Home (section)

===Into production===
[[File:Watson Radar.jpg|thumb|The first working<!--workable? really?--> radar unit constructed by Watson-Watt and his team. The four widely separated NT46 valves can be seen. Production units were largely identical.|alt=]]

The system was deliberately developed using existing commercially available technology to speed introduction.<ref>{{cite web|last=Waligorski |first=Martin |url=http://spitfiresite.com/2010/04/from-peace-to-war-royal-air-force-rearmament-programme-1934-1940.html |title=From Peace to War – Royal Air Force Rearmament Programme, 1934–1940 |publisher=Spitfiresite.com |date=10 April 2010 |access-date=10 February 2013}}</ref> The development team could not afford the time to develop and debug new technology. Watt, a pragmatic engineer, believed "third-best" would do if "second-best" would not be available in time and "best" never available at all.<ref name="Vectorsite.net">{{cite web |title=Longwave Radar at War / Early American Radar Efforts |publisher=Vectorsite.net |url=http://www.vectorsite.net/ttwiz_02.html |access-date=10 February 2013 |archive-date=17 February 2013 |archive-url=https://web.archive.org/web/20130217041924/http://www.vectorsite.net/ttwiz_02.html |url-status=usurped}}</ref> This led to the use of the 50&nbsp;m wavelength (around 6&nbsp;MHz), which Wilkins suggested would resonate in a bomber's wings and improve the signal. Unfortunately, this also meant that the system was increasingly blanketed by noise as new commercial broadcasts began taking up this formerly [[Shortwave radio|high-frequency spectrum]]. The team responded by reducing their own wavelength to 26&nbsp;m (around 11&nbsp;MHz) to get clear spectrum. To everyone's delight, and contrary to Wilkins' 1935 calculations, the shorter wavelength produced no loss of performance.{{sfn|Bowen|1998|p=20}} This led to a further reduction to 13&nbsp;m, and finally the ability to tune between 10 and 13&nbsp;m, (roughly 30-20&nbsp;MHz) to provide some [[frequency agility]] to help avoid jamming.{{sfn|Bowen|1998|p=21}}

Wilkins' method of height-finding was added in 1937. He had originally developed this system as a way to measure the vertical angle of transatlantic broadcasts while working at the RRS. The system consisted of several parallel dipoles separated vertically on the receiver masts. Normally the RDF [[Radiogoniometer|goniometer]] was connected to two crossed dipoles at the same height and used to determine the bearing to a target return. For height finding, the operator instead connected two antennas at different heights and carried out the same basic operation to determine the vertical angle. Because the transmitter antenna was deliberately focused vertically to improve gain, a single pair of such antennas would only cover a thin vertical angle. A series of such antennas was used, each pair with a different centre angle, providing continuous coverage from about 2.5 degrees over the horizon to as much as 40 degrees above it. With this addition, the final remaining piece of Watt's original memo was accomplished and the system was ready to go into production.{{sfn|Gough|1993|p=5}}{{sfn|Bowen|1998|p=21}}

Industry partners were canvassed in early 1937, and a production network was organized covering many companies. [[Metropolitan-Vickers]] took over design and production of the transmitters, [[AC Cossor]] did the same for the receivers, the Radio Transmission Equipment Company worked on the goniometers, and the antennas were designed by a joint AMES-GPO group. The Treasury gave approval for full-scale deployment in August, and the first production contracts were sent out for 20 sets in November, at a total cost of £380,000.{{sfn|Gough|1993|p=5}} Installation of 15 of these sets was carried out in 1937 and 1938. In June 1938 a London headquarters was established to organize the rapidly growing force. This became the Directorate of Communications Development (DCD), with Watt named as the director. Wilkins followed him to the DCD, and [[A. P. Rowe]] took over AMES at Bawdsey. In August 1938, the first five stations were declared operational and entered service during the [[Munich Agreement|Munich crisis]], starting full-time operation in September.{{sfn|Gough|1993|p=6}}