Industries / Media, arts and recreation / Media and entertainment / Radio broadcasting
Transmitters
Transmitter work is getting the signal out: the transmitters, antennas and towers for AM, FM, long wave and shortwave, and the choice of power and frequency.
4
Milestones achieved, from 1934 to 2026.
1
To watch for: announced, not yet confirmed outside.
2
Open deliverables. Nobody has shipped them.
The road
AchievedWatch forOpen
Achieved
1934
Half a million watts
WLW in Cincinnati, owned by Powel Crosley Jr.'s Crosley Radio Corporation, began broadcasting at 500 kW, ten times the 50 kW it had used before, with an amplifier of 12 water-cooled vacuum tubes rated at 100 kW each. Its chief engineer, Joseph Chambers, reported in an October 1934 paper that the higher power was the "most effective static eliminator yet devised," but WLW also interfered with CFRB in Toronto, 10 kHz away. After FCC hearings on superpower broadcasting that began in November 1936, the FCC limited US stations to 50 kW and WLW had to cut back, by the IEEE Cincinnati Section's account.
LessonOne very strong station reached distant rural listeners but spilled into its neighbors' channels, and the regulator chose many smaller stations over a few giants.
Credit
Crosley Radio Corporation (Powel Crosley Jr., Joseph A. Chambers)
1945
FM moves up the dial
FM (frequency modulation) carries sound by varying the frequency of the radio wave rather than its strength. Edwin Armstrong turned on the carrier of his FM station W2XMN at Alpine, New Jersey, on 10 April 1938, and commercial FM began on 42–50 MHz on 1 January 1941. On 27 June 1945 the FCC moved FM to 88–108 MHz, keeping 88–92 MHz for noncommercial stations, and a July 1946 order eliminated the lower band. NAB Reports in 1945 described the "old FM band between 42-50 megacycles" and the "new FM band between 88-108 megacycles," and Armstrong was still in court in January 1949 to stay on 44.1 MHz.
LessonThe band was moved after stations had launched, and they had to rebuild on new frequencies.
Credit
Edwin H. Armstrong; the FCC
1961
FM goes stereo
In the late 1950s the FCC considered 14 proposed systems for stereo FM, and in April 1961 it approved a joint General Electric–Zenith system. It keeps left plus right as the main channel and adds the difference on a 38 kHz subcarrier, with a 19 kHz pilot tone that receivers use to recover it. Regular stereo FM began on 1 June 1961: GE's WGFM in Schenectady, New York, went first at midnight Eastern time, and WEFM in Chicago followed an hour later at local midnight.
Credit
General Electric; Zenith; the FCC
2026
Long wave goes dark
The BBC's long-wave service on 198 kHz, carrying BBC Radio 4, closed at 1 am BST on 27 June 2026, shutting the transmitters at Droitwich (Wychbold), Westerglen and Burghead. The Wychbold masts had broadcast since 1934, letting one site cover the whole UK, and carried Neville Chamberlain's declaration of war in 1939 and, later, the Shipping Forecast and ball-by-ball cricket on Test Match Special.
Credit
the BBC
Watch for
Japan leaves AM
44 of Japan's 47 commercial AM stations have announced plans to stop AM broadcasting by autumn 2028 and move to FM. In an April 2026 article for mediumwave.info, Shinobu Unakami gives two causes: the spread of Wide FM, which puts AM stations' programs on FM, and ageing equipment. Japan's Ministry of Internal Affairs and Communications (MIC) issued a policy in March 2023, amended in December 2024, that lets stations suspend AM for a period before closing it, and on 30 March 2026 the public broadcaster NHK cut its three radio services to two, taking Radio 2 off the air.
Achieved whenthe stations' AM transmitters are off the air after autumn 2028, as MIC licence records show, and listening figures from before and after are published.
Claimed by
Japan's commercial AM broadcasters and the MIC; latest report April 2026
Open
01
Map the AM noise floor
Fixed receivers in several cities that log the noise level on empty AM channels around the clock for a year, with a public map and dataset, so that engineers and regulators can see whether electrical noise from lights, chargers and power supplies is rising, and where.
The briefCredit · unclaimed02
Redraw AM coverage for cities
Measured coverage maps for AM stations in dense cities, made from drive tests with a calibrated receiver, set beside the coverage that the FCC's prediction methods give.
The briefCredit · unclaimed
Open 01 · the brief
Map the AM noise floor
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
Fixed receivers in several cities that log the noise level on empty AM channels around the clock for a year, with a public map and dataset, so that engineers and regulators can see whether electrical noise from lights, chargers and power supplies is rising, and where.
Why it is still open
In 2016 the FCC's Technological Advisory Council opened an inquiry (ET Docket 16-191) asking whether radio noise was increasing and for quantitative evidence. Radio World reported in 2017 that respondents were "virtually unanimous in calling for an official noise study." The most recent study found, NRSC-R102 (January 2025), measured noise on three empty AM channels along one 50-mile drive from Germantown to Baltimore, Maryland. That is a snapshot of one area, not a trend over time.
Built with what exists
Software-defined radios with HF coverage, such as KiwiSDR or Airspy HF+ receivers, with active loop antennas; a Raspberry Pi at each site; Python for logging; NRSC-R102's method and levels as the calibration reference.
First step
Set up one receiver and loop antenna at an urban site, log three empty channels for a month, and compare the readings with NRSC-R102's urban figures.
Done looks like
10 sites in at least three cities logged for 12 months; hourly noise levels on at least three clear AM channels per site, calibrated against a field-strength meter, published as open data with the month-to-month change and its uncertainty.
Open 02 · the brief
Redraw AM coverage for cities
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
Measured coverage maps for AM stations in dense cities, made from drive tests with a calibrated receiver, set beside the coverage that the FCC's prediction methods give.
Why it is still open
NRSC-R102 found that building clutter can cut AM field strength by a factor of up to 100 (40 dB), "which are not accounted for in signal coverage maps," and that measured fields can be well below what the usual ground-wave models predict. Its measurements cover three 50 kW stations on one route in the Baltimore and Washington area.
Built with what exists
A software-defined radio and calibrated active antenna, checked against a field-strength meter such as the Potomac Instruments FIM-41 used in the NRSC study; a GPS logger; the FCC's published AM station data for the predictions; QGIS for the maps.
First step
Drive one station's predicted strong-signal area through one dense downtown and one suburb with an SDR and GPS logger, and plot measured against predicted.
Done looks like
20 stations in three cities driven on a street grid; for each station, the area where the measured signal is above a stated level (NRSC-R102 found 3 to 5 mV/m generally good for mobile listening) compared with the predicted area, published with the raw data.