Radar altimeter

The underlying concept of the radar altimeter was developed independent of the wider radar field, and originates in a study of long-distance telephony at Bell Labs. During the 1910s, Bell Telephone was struggling with the reflection of signals caused by changes in impedance in telephone lines, typically where equipment connected to the wires. This was especially significant at repeater stations, where poorly matched impedances would reflect large amounts of the signal and made long-distance telephony difficult.[6]

Engineers noticed that the reflections appeared to have a "humpy" pattern to them; for any given signal frequency, the problem would only be significant if the devices were located at specific points in the line. This led to the idea of sending a test signal into the line and then changing its frequency until significant echos were seen, and then determining the distance to that device so it could be identified and fixed.[6]

Lloyd Espenschied was working at Bell Labs when he struck upon the idea of using this same phenomenon as a way to measure distances in wire in a more general fashion. One of his first developments in this field was a 1919 patent (granted 1924)[7] on the idea of sending a signal into railway tracks and measuring the distance to discontinuities. These could be used to look for broken tracks, or if the distance was changing more rapidly than the speed of the train, other trains on the same line.[6]

During this same period there was a great debate in physics over the nature of radio propagation. Guglielmo Marconi's successful trans-Atlantic transmissions appeared to be impossible; studies of radio signals demonstrated they travelled in straight lines, at least over long distances, so the broadcast from Cornwall should have disappeared into space instead of being received in Newfoundland. In 1902, Oliver Heaviside in the UK and Arthur Kennelly in the USA independently postulated the existence of an ionized layer in the upper atmosphere that was bouncing the signal back down to the ground so it could be received. This became known as the Heaviside layer.[8]

While an attractive idea, direct evidence was lacking. In 1924, Edward Appleton and Miles Barnett were able to demonstrate the existence of such a layer in a series of experiments carried out in partnership with the BBC. After scheduled transmissions had ended for the day, a BBC transmitter in Bournemouth sent out a signal that slowly increased in frequency. This was picked up by Appleton's receiver in Oxford, where two signals appeared. One was the direct signal from the station, the groundwave, while the other was received later in time after it travelled to the Heaviside layer and back again, the skywave.[8]

The trick was how to accurately measure the distance travelled by the skywave to demonstrate it was actually in the sky. This was the purpose of the changing frequency. Since the ground signal travelled a shorter distance, it was more recent and thus closer to the frequency being sent at that instant. The skywave, having to travel a longer distance, was delayed, and was thus the frequency as it was some time ago. By mixing the two in a frequency mixer, a third signal is produced that has its own unique frequency that encodes the difference in the two inputs. Since in this case the difference is due to the longer path, the resulting frequency directly reveals the path length. Although technically more challenging, this was ultimately the same basic technique being used by Bell to measure the distance to the reflectors in the wire.[8]

In 1929, William Littell Everitt, a professor at Ohio State University, began considering the use of Appleton's basic technique as the basis for an altimeter system. He assigned the work to two seniors, Russell Conwell Newhouse and M. W. Havel. Their experimental system was more in common with the earlier work at Bell, using changes in frequency to measure the distance to the end of wires. The two used it as the basis for a joint senior thesis in 1929.[9]

Everitt disclosed the concept to the US Patent Office, but did not file a patent at that time. He then approached the Daniel Guggenheim Fund for the Promotion of Aeronautics Guggenheim Foundation for development funding. Jimmy Doolittle James Doolittle, secretary of the Foundation, approached Vannevar Bush of Bell Labs to pass judgement. Bush was skeptical that the system could be developed at that time, but nevertheless suggested the Foundation fund development of a working model. This allowed Newhouse to build an experimental machine which formed the basis of his 1930 Master's thesis, in partnership with J. D. Corley.[9][10]

The device was taken to Wright Field where it was tested by Albert Francis Helgenberger, a noted expert in aircraft navigation. Hegenberger found that the system worked as advertised, but stated that it would have to work at higher frequencies in order to be practical.[9][lower-alpha 1]

Espenschied has also been considering the use of Appleton's idea for altitude measurement. In 1926 he suggested the idea both as a way to measure altitude as well as a forward-looking system for terrain avoidance and collision detection. However, at that time the frequency of available radio systems even in what was known as shortwave was calculated to be fifty times lower than what would be needed for a practical system.[6][10]

Espenschied eventually filed a patent on the idea in 1930.[10] By this time, Newhouse had left Ohio State and taken a position at Bell Labs. Here he met Peter Sandretto, who was also interested in radio navigation topics. Sandretto left Bell in 1932 to become the Superintendent of Communications at United Air Lines (UAL), where he led the development of commercial radio systems.[9]

Espenschied's patent was not granted until 1936,[11] and its publication generated intense interest. Around the same time, Bell Labs had been working on new tube designs that were capable of delivering between 5 and 10 Watts at up to 500 MHz, perfect for the role.[10] This led Sandretto to contact Bell about the idea, and in 1937 a partnership between Bell Labs and UAL was formed to build a practical version. Led by Newhouse, a team had a working model in testing in early 1938, and Western Electric (Bell's manufacturing division) was already gearing up for a production model. Newhouse also filed several patents on improvements in technique based on this work.[12]

The system was publicly announced on 8 and 9 October 1938.[13] During World War II, mass production was taken up by RCA, who produced them under the names ABY-1 and RC-24. In the post-war era, many companies took up production and it became a standard instrument on many aircraft as blind landing became commonplace.[12]

A paper describing the system was published jointly by Espenschied and Newhouse the next year. The paper explores sources of error and concludes that the worst-case built-in scenario was on the order of 9%,[14] but this might be as high as 10% when flying over rough terrain like the built-up areas of cities.[14]

During early flights of the system, it was noticed that the pattern of the returns as seen on an oscilloscope was distinct for different types of terrain below the aircraft. This opened the possibility of all sorts of other uses for the same technology, including ground-scanning and navigation. However, these concepts were not able to be explored by Bell at the time.[13]

It had been known since the late 1800s that metal and water made excellent reflectors of radio signals, and there had been a number of attempts to build ship, train and iceberg detectors over the years since that time. Most of these had significant practical limitations, especially the use of low-frequency signals that demanded large antennas in order to provide reasonable performance. The Bell unit, operating at a base frequency of 450 MHz, was among the highest frequency systems of its era.[14][lower-alpha 2]

In Canada, the National Research Council began working on an airborne radar system using the altimeter as its basis. This came as a great surprise to British researchers when they visited in October 1940 as part of the Tizard Mission, as the British believed at that time that they were the only ones working on the concept. However, the Canadian design was ultimately abandoned in favour of building the fully developed British ASV Mark II design, which operated at much higher power levels.[15]

In France, researchers at IT&T's French division were carrying out similar experiments when the German invasion approached the labs in Paris. The labs were deliberately destroyed to prevent the research falling into German hands, but German teams found the antennas in the rubble and demanded an explanation. The IT&T director of research deflected suspicion by showing them the unit on the cover of a magazine and admonishing them for not being up-to-date on the latest navigation techniques.[12]

• Espenschied, Lloyd; Newhouse, Russell (January 1939). "A Terrain Clearance Indicator". The Bell System Technical Journal. 18 (1): 222–234. doi:10.1002/j.1538-7305.1939.tb00813.x.CS1 maint: ref=harv (link)
• "Historic Firsts: Radio Altimeter" (PDF). Bell Labs: 18–19. January 1948.
• Colin, Robert (July 1967). "1967 Pioneer Award: Lloyd Espenschied and Russell C. Newhouse". IEEE Transactions on Aerospace and Electronic Systems. AES-3 (4): 736–742. doi:10.1109/TAES.1967.5408855.CS1 maint: ref=harv (link)