Building the Broadcast Band

Building the Broadcast Band

Thomas H. White -- February 12, 2004
The history of AM broadcast band (mediumwave) in the United States spans eighty years. This is a review of its first decade -- how it was established, initially evolved, suffered through a chaotic period when government regulation collapsed, and finally was reconstructed by the newly formed Federal Radio Commission, along lines that are still visible today.

Sections

Technical Antecedents
Guglielmo Marconi's pioneering wireless work, begun in the late 1800's, developed an important principle which more than twenty years later would help determine which wavelengths would be available for broadcasting. Marconi's most significant early discovery was of the "groundwave" radio signal. This was a key development, which made longrange signaling using electromagnetic radiation practical for the first time. Prior to Marconi, all electromagnetic radiation was thought to act similarly. Like light, it was believed to normally travel through the air in a straight line until absorbed or reflected. What Marconi stumbled across was that, for longer wavelengths with a properly constructed antenna, some of the radio waves, instead of just "going through space", actually "traveled along the ground", following the contour of the Earth. Thus, the Earth could be used as a guide, carrying signals over the horizon to distant points. Moreover, it turned out the ocean was an even better conductor than soil for transporting radio waves to distant points.

It was found that the longer the radio wavelength, the better the Earth acts as a conductor, and the greater the range for a transmission of a given power. For this and other early work, Marconi shared the 1909 Nobel Prize for physics. And for 25 years following his pioneering work the groundwave signal was the most important factor in determining the desirability of a given radio wavelength.
United States Government Regulations
In the United States the use of wireless initially was unregulated -- anyone could operate a radio transmitter anywhere, at any time, on any wavelength. And most utilized the longwave signals that traveled so well across land and sea. Naturally severe interference occurred with everyone trying to use the same wavelengths. Eventually it was decided to do something about this, and because the individuals involved were the United States government, the action took the form of An Act to Regulate Radio Communication, passed on August 13, 1912.

A year earlier a Radio Service had been established in the Department of Commerce and Labor's Bureau of Navigation. It was initially charged with making sure ships carried wireless equipment, as required by a June, 1910 act. With the passage of the 1912 Act, the job of licencing stations and operators was added to the Radio Service's duties. The country was divided into nine radio inspection districts, with a district headquarters for a Radio Inspector set up at a major port within each district. Initially radio was dominated by ship-to-ship and ship-to-shore stations, plus amateurs who comprised the bulk of the land stations.

As far as government control goes the 1912 Act was fairly liberal, since some, particularly the Navy, had wanted to nationalize radio altogether. Unfortunately, the Act's language wasn't always very clear, and was geared toward two-way communication between stations that were permitted, and even expected, to use various wavelengths of their own choosing. Fourteen years later these flaws would help cause a breakdown in the regulation of broadcast stations.

The 1912 Act essentially divided the radio spectrum into four parts. Following the standard set by the Service Regulations of the 1912 London International Radiotelegraph Convention, a choice band of wavelengths, from 600 to 1600 meters (500 to 187.5 khz) was appropriated primarily for government use. This band was selected due to the superior groundwave coverage these wavelengths provided. Two additional bands, available for commercial use, were designated on either side of the government band. The first group, consisting of wavelengths greater than 1600 meters (frequencies less than 187.5 khz), actually had groundwave coverage superior to that of the government band. Here were found the huge transoceanic stations. The other commercial band ranged from 600 meters to 200 meters (500 khz to 1500 khz). Groundwave coverage provided by these wavelengths rapidly diminished as the wavelength decreased. This band was used by commercial stations with more limited service areas, and for other special purposes, such as 300 and 220 meters (1000 and 1365 khz), set aside because ship antennas were too short for effective use on longer wavelengths.

The final "band" was really a single wavelength -- 200 meters (1500 khz). Although they were not mentioned by name, this wavelength was assigned to amateur stations. Because of its poor groundwave coverage, it was considered to be all but useless, and was far removed from the wavelengths amateurs had used prior to 1912. Still, this limited allocation was better than being completely eliminated, which some, again particularly in the Navy, had favored. The Act also allowed individual amateurs to receive "special" licences to use longer wavelengths, and a number were issued within the 200 to 600 meter band, in order to support communication between amateurs doing "relay" work. (According to the Bureau of Navigation's September 28, 1912 edition of Regulations Governing Radio Communication, "...a special license will be granted only if some substantial benefit to the art or to commerce apart from individual amusement seems probable".) Still, the Act was a major setback for amateurs, and severely restricted their activities.
The Rise of Voice Broadcasting
All early radio work used telegraphic signaling, in most cases using spark transmitters. However, following the example of the wire telegraph, which would lead to the telephone, many worked to transmit sound by radio. As this work progressed hundreds, perhaps thousands, of experimental and publicity broadcasts were made. Some were even conducted on regular schedules. However, the first technologies used -- high-frequency spark, alternator and arc transmitters -- turned out to be dead-ends in the attempt to provide reliable, high quality, and cost effective voice service. Only with the development of vacuum tube continuous wave transmitters, just before the start of World War I, did broadcasting became practical. During the war all radio equipment -- both sending and receiving -- was either shut down or taken over by the United States government, so broadcasting experimentation ceased. However, the new vacuum tube transmitters were perfected under government supervision. In late 1919, with the end of the wartime restrictions on transmitting, numerous commercial, experimental, government and amateur stations renewed dabbling with broadcasting, using the new vacuum tube transmitter designs. By its September, 1920 issue, QST magazine would note that "it is the rare evening that the human voice and strains of music do not come in over the air".
The Westinghouse Stations
Of all the players involved with broadcasting experimentation and development, it was the Westinghouse Electric and Manufacturing Company, headquartered in East Pittsburgh, Pennsylvania, which would finally spark the transformation of radio broadcasting from an experiment into a national institution. Westinghouse was a relative newcomer to radio work. Its post-war efforts arose out of wartime contracts, combined with the broadcasts of Westinghouse engineer Frank Conrad's experimental station, 8XK. Westinghouse was to become the first concern to have the vision, commitment, financial stability, and clout to propel broadcasting into the national consciousness.

Previously the person most associated with broadcasting had been Lee DeForest, who was behind a number of efforts by various companies on both coasts, beginning before the war. However, these activities always seemed to eventually evaporate. In particular, DeForest had a knack for getting stations shut down for violating regulations. With a well established firm like Westinghouse there was no doubt their broadcast activities were a stable and on-going service, that would be funded in part by profits from the sale of Westinghouse radios to the general public. In contrast, with the DeForest efforts there was always the nagging suspicion that a station's main purpose was to promote the sale of watered stock, or that the company responsible, along with the broadcasts, might soon disappear, as had so many of the previous efforts. By 1921, when Westinghouse's work began to bear fruit, DeForest had left radio research, and was concentrating on work on a sound-on-film system for talking movies.

Westinghouse inaugurated its new broadcast service from East Pittsburgh with presidential election returns on November 2, 1920. Most accounts simplify things by crediting this historic broadcast to KDKA, operating on 360 meters. Actually, either due to a delay in the delivery of KDKA's Limited Commercial licence, or more likely indecision about the proper classification for the station's entertainment offerings, the election night broadcast went out under the temporarily assigned Special Amateur call of "8ZZ". Moreover, it wasn't until the fall of 1921 that KDKA moved to 360 meters.

Westinghouse's broadcast was hardly unique, as a number of other stations sent out election returns at the same time, and some had also broadcast results during previous elections. Nor were there historic numbers of listeners to the broadcast, since contemporary estimates put the audience at about 100 receivers, and it attracted little attention outside of the immediate Pittsburgh area. However, Westinghouse differentiated itself from the others which had made broadcasts by launching a regular daily schedule, with plans to establish additional stations if the Pittsburgh station proved successful.

Westinghouse understandably sought good coverage for KDKA and its later broadcast stations. However, the commercial longwave band beyond 1600 meters was too congested to be usable, while the 600 to 1600 band was reserved for government stations. Thus, KDKA's home would have to be somewhere within the 200 to 600 meter band -- the only wavelengths remaining after earlier radio settlers had claimed the longer wavelengths with their superior groundwave coverage.

Information is sketchy, but contemporary reports state that the election night broadcast, using the callsign 8ZZ, was transmitted on a wavelength of 550 meters (545 kilohertz) while later publicity places KDKA's broadcasts on 330 meters (909 khz). There is evidence of shifting around, as some later reports list one or more of the Westinghouse stations on 375 meters (800 khz). With the success of KDKA, the fall of 1921 saw the establishment of three additional Westinghouse stations -- WJZ Newark, NJ (now WABC, New York), WBZ Springfield, MA (now in Boston), and KYW Chicago, IL (now in Philadelphia, PA). At this time Westinghouse officials lobbied for a special wavelength for their stations, and after negotiating with Commerce officials, 360 meters (833 khz) was selected. (Unlike DeForest, Westinghouse seems to have had good relations with government regulators). Louis R. Krumm of Westinghouse later claimed credit for proposing 360 meters as the standard. The first station to receive a license that explicitly specified 360 meters was WBZ on September 15, 1921. Licences for 360 meters for WJZ, KDKA, and KYW soon followed.
Establishment of a Broadcast Service
Westinghouse apparently thought only its stations would be assigned to 360 meters. However, the Commerce Department had no intention of giving Westinghouse a wavelength monopoly. Officials began assigning 360 meters to broadcast stations that other companies set up beginning in the fall of 1921. Unwittingly, Westinghouse's suggestion for itself instead became the seed wavelength which would flower into the broadcast band.

By late 1921 enthusiasm for broadcasting had started to develop nationwide, and the Bureau of Navigation decided to formally designate standards and wavelengths for a specific broadcast service. Moreover, in addition to entertainment broadcasts, it saw the need to provide for broadcasts of official government reports. On December 1, 1921 two wavelengths were formally set aside for broadcasting, set up as a service category within the already existing "Limited Commercial" class of stations. A clause was added to the Limited Commercial regulations, reading: "Licences of this class are required for all transmitting radio stations used for broadcasting news, concerts, lectures, and such matter. A wave length of 360 meters is authorized for such service, and a wave length of 485 meters is authorized for broadcasting crop reports and weather services, provided the use of such wave lengths does not interfere with ship to shore or ship to ship service".

Thus, broadcasting was formally introduced using just two wavelengths -- 360 and 485 -- in the 200 to 600 meter band. However, it would rapidly expand, until it ended up occupying almost all of this band, plus some of the "useless" territory beyond 200 meters. In addition, it would also drive out the ship-to-shore and ship-to-ship services it initially was required to protect. At this time there were few limitations on who could get a broadcast station licence. Generally all you needed was the desire, the equipment, and American citizenship--plus an on-duty technician holding at least a commercial second-grade operator's licence.
"Crop Reports and Weather Services"
Having a separate wavelength -- 485 meters -- for government market and weather reports made theoretical sense, but ultimately proved impractical. After the Navy Department, the Agriculture Department had been the government agency most involved in pioneering radio work. In particular, it wanted to speed weather and market information to isolated farmers, at that time dependent on mailed daily newspapers. (The August, 1913 Monthly Catalogue of United States Documents noted that the Weather Bureau had begun a daily radiotelegraphic "broadcast" of weather reports, which it explained as follows: "'Broadcast', as the term is used in the Radio Service, means that the message is fired out into the illimitable ether to be picked up and made use of by anybody who has the will and the apparatus to possess himself thereof".)

Beginning with international conventions preceding the 1912 Act, it was the practice to set aside certain wavelengths for special purposes. So, it was natural to set aside a special wavelength for broadcasting market reports and weather forecasts. Then a radio could be tuned to a single wavelength and receive service from a number of stations. If the reports had instead gone out on 360 meters, farmers would have risked having distant reports drowned out by nearby stations broadcasting at the same time.

The 485 wavelength--with its better groundwave coverage--was probably seen as the more important development, and a greater public service, than the mere entertainment being sent out on 360. On many occasions the Bureau of Navigation's Radio Service Bulletin listed stations and schedules of weather and market broadcasts, but it never featured the latest listing of stations carrying the Chase and Sanborn Hour. Any broadcast station could get 360 just for the asking, and most did. However, before the Bureau of Navigation would issue an authorization for 485 meters the station had to first submit a written authorization from the Chief of the Bureau of Markets and Crop Estimates or the Chief of the Weather Bureau. (In its 1922 annual report, the Agriculture Department reported it was limiting 485 authorizations to just two stations per community.) Although the number of broadcast stations authorized to use 485 meters rose from 15 to 137 in the year ending March, 1923, there were few problems with interference. The two Bureaus strictly regulated dissemination of government reports. They also controlled the schedules for the broadcasts, so that stations sending out reports on 485 meters would not interfere with each other.

From the government's point of view the dual-wavelength system worked pretty well. For example, in late 1922 the Weather Bureau Office in Springfield, Illinois announced that, using a good receiver, a daily schedule of thirteen weather and market reports, from seven different broadcast stations, could be heard in central Illinois on 485 meters.

Unfortunately, individual stations were not as impressed, especially since most concentrated on the entertainment side of their offerings. Credo Fitch Harris, in "Microphone Memoirs", a history of the "Horse and Buggy Days" of WHAS in Louisville, Kentucky, wrote:

What logic gave rise to that mandate to tune a transmitter suddenly from its normal operation of 360 meters to 485 for the weather reports, and then quickly back to 360 for the continuance of a program, has never been explained and it still remains one of the most profound departmental enigmas. Practically none but farmers yearned passionately for news of tomorrow's weather, and crystal sets were incapable of serving distant areas. There were a few, though quite exceptional, instances of longer range receivers, -- using earphones of course. These were homemade affairs built from published diagrams and strung out from mother's parlor table to the kitchen, but so imperfect and confusing to tune that usually we had sent the forecast on 485, and were back again on 360, before the tyro had emerged from his wilderness of tangled wires, knobs, rheostats and other gadgets. The rulings were so patently absurd that the chief of the Louisville Meteorological Bureau personally appealed to Washington and had it changed. Parenthetically, for fifteen years I have tried to discover the father of it. None will confess.

In defense of the Weather and Market Bureaus, it's doubtful they expected a station to jump back and forth between 360 and 485 meters like WHAS did. Most likely they expected the station to set aside, and publicize, a fixed period each day for the broadcasts on 485, after which it would sign off. Then, after a decent interval, it would start up operations on 360. In any event, as reviewed later the split wavelength operations ended in May of 1923, not because of the intervention of the Louisville Meteorological Bureau, but as a result of the expansion of the frequencies allocated to broadcasting. (The concept of broadcast frequencies reserved exclusively for public weather reports continues into the present, via the NOAA Weather Radio frequencies.)
"News, Concerts, Lectures, and Like Matter"
The government, viewing broadcasting as a public service, may have thought that 485 meters was the more important development. However, the general public saw 485 meters as only a sideshow. The main attraction was the entertainment offered on 360 meters. However, in contrast to the carefully controlled activities on 485 meters, the situation on 360 meters eventually became badly congested, especially in the larger cities. In the year ending March, 1923 the number of stations authorized for 360 meters jumped from 65 to 524. Moreover, it was up to the stations themselves to come up with equitable timesharing agreements when more than one station was located in the same area. Although most stations only wanted to broadcast a few hours per day or week, most coveted the prime early evening hours. In the New York City area, Westinghouse thought that WJZ, which began broadcasting in October, 1921, was going to be the only station there on 360 meters. Certainly it didn't see a need for additional ones. However, by the middle of 1922 nine more stations had been licenced for 360 meters in the region, requiring a complicated and hard-fought timesharing agreement for the New York City area. Other cities had similar problems. San Francisco had been an early broadcast center, with a number of experimental stations operating on various wavelengths, some of which pre-dated KDKA. However, when the new policies required them to be converted to broadcast stations, they congregated on 360 meters, requiring a timesharing agreement.

In a few cases talks came to an impasse, and two stations would start to transmit at the same time, drowning each other out. Officials at the Commerce Department normally refused to get involved in these disputes. Eventually the stations, which looked pretty silly, would bow to public pressure and work out some sort of compromise. (No doubt it also was difficult to lure talent with the opportunity to participate in "broadcasts" that were completely drowned out by another station).
Meters and Kilohertz
The initial broadcast service allocations referred to the "wavelengths" that stations would use. This practice dated back to early radio work, when the length of the antenna had a strong influence on the wavelength of the radio signals that were transmitted and received. However, for technical reasons, beginning in 1923 the Bureau of Navigation switched to specifying a station's "frequency", as measured in "kilocycles per second" (later recast as "kilohertz"). Frequency and wavelength are reciprocals -- to convert one to the other you just divide the value into the speed of light. So, how many kilohertz is 360 meters? Suddenly the simple division is not so simple, because the speed of light was only roughly known in the early 1920s. In some early Department of Commerce references 360 meters was stated to be 834 khz. In other cases the rounded figure of 300,000 kilometers/second was used for the speed of light, so depending on how many decimal places were calculated the answer became 833 or 833.3 or 833.333. Sometimes a more precise estimate, 299,820, was used for the speed of light, which gives a result of 832.8 khz. And if you use the even more precise modern estimate of 299,792.458, the answer becomes 832.757 khz. (485 meters is equivalent to either 618 or 619 khz, depending on the value used for the speed of light.) All this leads to a question -- if you could go back to 1922 with a modern radio with a digital frequency readout, and you wanted the radio tuned to the exact frequency equivalent for a station operating on 360 meters, what you punch in? The following excerpt from "Microphone Memoirs" gives a clue:

The way a transmitter was complacently assumed to be kept on its required 360 in those days could be amusing now, or horrifying. A government inspector arrived every four or five months to 'measure' us. In front of the main panel was a large aluminum disk with a center knob, devised by the manufacturer to vary its emitted frequency. The supervisor would gravely and thoughtfully turn that knob back and forth, watching his meter betimes. He would then take a pencil and make a thin mark on the disk's circumference, announcing solemnly: '360'. Another mark: '485 for the weather'. If those pencil strokes escaped being rubbed off by an over-zealous janitor some early morning, we probably retained an accuracy of five or ten meters, above or under par. Or if they remained long enough for the supervisor's next visit, it was interesting to observe that he invariably rubbed them out himself and put on new ones"

A ten-meter swing each way for a station at 360 meters translates to a frequency drift from about 810 to 855 khz. Obviously WHAS' setup wasn't very precise. But its transmitter was no homebrew concoction -- it was an expensive top-of-the-line 500 watt Western Electric, the best that money could buy. Government regulators would struggle for a decade with keeping stations on their assigned frequencies. [Kilohertz to Meters Conversion Charts].
Restrictions
By the end of 1921, 29 broadcast station authorizations had been issued for 360 and 485 meters. In early 1922 the broadcasting bandwagon rapidly gained momentum. On board, in addition to formally recognized broadcast stations, were government, experimental, technical and training school, plus regular and special amateur stations, each operating on their own wavelengths. Government stations were outside the control of the Bureau of Navigation, so nothing could be done about them. In any event, many of their broadcasts were speeches by elected officials, so it probably wouldn't have been wise to try. However, the rest were required to conform to the new regulations, and convert to formal broadcast stations, if they wanted to continue broadcasting to the general public.

Broadcasts by amateur stations were explicitly prohibited beginning in January, 1922. The Bureau of Navigation regarded most of the broadcasts coming from these stations as frivolous -- in most cases the best they could offer were scratchy phonograph records. Since most people already had phonographs there didn't seem to be a pressing public need to fill the airwaves with recorded songs. (According to the June 30, 1929 Annual Report of the Chief of the Radio Division of the Department of Commerce: "During the early days the programs of a majority of stations consisted almost entirely of phonograph records. The announcers had favorite records which they repeated numerous times during a program. The Secretary of Commerce foresaw the danger of the station losing public interest if a change were not made in the programs.") Amateur broadcasts were said to only be "temporarily" banned, pending new regulations. Eighty years later amateurs are still waiting for the ban to expire. In the meantime, some amateur stations were converted into broadcast stations, helping to swell the broadcasting ranks.
First National Radio Conference
By early 1922 it was clear that broadcasting was an important, and probably permanent, development. It was also beginning to tax the ingenuity of its regulators. In order to receive advice on a number of pressing issues, Commerce Secretary Herbert Hoover convened a Conference on Radio Telephony, composed of representatives of various government agencies and radio groups. The conference met in Washington from February 27th to March 2nd, and again from April 17th to the 19th. The resulting conference report proposed that major portions of the 200 to 600 meter band be set aside for broadcasting. In fact, it suggested separate bands for Government and Public, Private and Toll, and City and State Public broadcasting stations. It favored a total ban on "direct" advertising, and even suggested rules governing broadcasting by private detective agencies. The report also favored legislation strengthening the Commerce Secretary's regulatory authority.

Secretary Hoover, while lauding the efforts of the conference, moved cautiously, partly because Congress failed to pass any new legislation. Only a single new wavelength, 400 meters (750 khz) was added, as a second entertainment wavelength. This was designated the "Class B" wavelength, with 360 meters now referred to as the "Class A" entertainment wavelength. Although 400 meters was envisioned for the use of "better quality" stations, in order to avoid the appearance of censorship only technical requirements had to be met in order to be assigned to the new wavelength. The maximum power permitted was 1000 watts, and "mechanically reproduced" programs were prohibited. As on 360 meters, stations in the same locality had to devise timesharing agreements.
Class B Stations on 400 Meters
In most cases there are about a dozen claimants when you try to identify "the first station" in one category or another. Surprisingly, there seems to be universal agreement that the first Class B station was KSD, the Saint Louis Post Dispatch station in Saint Louis, Missouri, beginning in late September, 1922 (now KTRS-550). Eventually around thirty stations nationwide qualified to use 400 meters.

Although most stations that met the new standards welcomed the chance to move to the less congested 400 meter wavelength, for some it caused problems. The March, 1923 edition of Radio News carried the following report: "One big broadcasting station after trying out the Class B licence on 400 meters for a short time has returned to the 360 wave. The Department of Commerce has just relicenced WHAS, The Louisville Courier Journal, on 360 meters. That paper believes the 360-meter wavelength is better suited for broadcasting, and more popular with the fans".

In fact, the order to move to 400 meters had caused an odd crisis at WHAS. As recorded in "Microphone Memoirs", the following exchange took place between station manager Harris and his technician:

'Now what?' I asked. 'Can you put us on 400?' 'I can try,' he said. 'When the supervisor measured us last September he marked 360 and 485, but the 485 got rubbed off. Let's see. The 400 meter change would be -- ' (out came the slide rule). 'Well, it would be about a third up from where we are to where 485 is if 485 was there, which it isn't. We can't move a third up to nowhere. Maybe I can guess it, within about ten or fifteen meters'

This technical problem, plus fear that their listeners would find it as hard to retune their sets to 400 meters as WHAS did, prompted Harris to get permission to stay on 360 meters.
Station Wavelength Assignments
With the addition of 400 meters, it was now possible for a broadcast station to be licenced to 360-only, 400-only, 485-only, 360/485, or 400/485, where 360 and 400 were Class A and B entertainment wavelengths and 485 continued as the Market and Weather wavelength. Below is a chart reviewing the authorizations on these wavelengths, compiled from official station lists issued for selected dates from March 10, 1922 to March 1, 1923:

Date	Total Stations	Station Wavelength Assignments					Wavelength Totals
Date	Total Stations	360-only	360/485	485-only	400-only	400/485	360	485	400
3/10/1922	67	52	13	2	--	--	65	15	--
3/31/1922	137	105	26	6	--	--	131	32	--
4/30/1922	223	178	38	7	--	--	216	45	--
5/31/1922	312	254	50	8	--	--	304	58	--
6/30/1922	378	307	63	8	--	--	370	71	--
3/1/1923	556	409	115	5	10	17	524	137	27

(Links to on-line copies of these stations lists are available at Early Radio Station Lists Issued by the U.S. Government).
Dawn of the Skywave
Because the stations on 400 meters had superior equipment, they did a better job of staying on their assigned wavelengths. Surprisingly, in some cases this resulted in more interference between stations. A letter from Murfreesboro, Tennessee, appearing in the February, 1923 issue of Radio News, in part complained: "Can't you start some kind of a campaign among your thousands of Radio fans and readers to get Washington to do something about this wave-length question? Since all the good stations have gone to 400 meters it is worse than ever, as they are square on 400 meters and all come in together... while before they were scattered below and over 360 meters". This letter reflects a new problem which was being encountered during nighttime hours. It was the result of the development of better radio receivers, combined with the existence of long ignored "skywave" radio signals.

Until the early twenties, most radio receivers used by both the public and commercial companies had been primitive. The majority were crystal sets, limited to picking up strong signals, which in practice usually meant only groundwave signals. The spread, in the early twenties, of receiving sets using vacuum tube amplification meant radios were now thousands of times more sensitive. The wavelengths assigned to broadcast stations had relatively poor groundwave coverage, and the stations used relatively low power, with few rated at more than 500 watts. So, considering only the groundwave signal, stations could be packed fairly close together on the same wavelength without unduly interfering with each other. However, with the introduction of the better receivers, at night during the prime listening hours people were beginning to receive stations from far beyond the range of the groundwave signal. This would have profound effects on how to deal with interference between stations operating on the same wavelength.

At this point it's valuable to return to Marconi's original work. Like many scientific discoveries, his discovery of the groundwave signal both advanced and hindered the art, because it lead to a single-minded pursuit of good groundwave coverage. Huge spark stations of tremendous power were developed, using giant antennas. By later standards these early stations were absurdly overpowered -- in fact they were so powerful that their signals were probably traveling around the world more than once. However, because receivers were so insensitive, these transmitting behemoths were needed in order to insure quality service.

Forgotten in the "cult of the groundwave" was the fact that not all of a station's signal is groundwave -- some of it does indeed travel "through the air". Originally it was thought that these "skywave" signals merely fled into the cosmos, never to be heard again. However, soon there was evidence that something strange was happening, especially at night. Somehow, some of the signals were coming back to Earth at distant points. English physicist Oliver Heaviside did pioneering work on the subject, and found evidence that high above the Earth there is an encircling layer of charged particles. This was originally called the Heaviside Layer, but is now known as the ionosphere, and is the cause of the reflected signals. At first it was mainly viewed as a curiosity, responsible for "freak" reception. Unlike the groundwave signal, which is unaffected by the sun, and has the same strength day and night, the strength of the reflected skywave signal is variable, and usually was too weak to be readily detected by the primitive receivers then in use. Also, on the wavelengths then in use there normally wasn't any skywave signal during daylight hours, so daytime reception was completely dependent on the groundwave signal. In fact the skywave signal was seen mainly as a nuisance, since it interacted with the groundwave signal, causing fading.

With the introduction of broadcasting, information about skywave signals suddenly became important. However, a full understanding of what was taking place did not exist in the early twenties. It was obvious the sun was involved, since in most cases skywave signals appeared only at night. Eventually it was determined that the ionosphere is composed of layers, each with distinctive characteristics. What became known as the "E" and "F" layers are responsible for reflecting radio signals back to Earth. (Unlike groundwave signals, the strength of reflected skywave signals are essentially the same across the entire 200 meter to 600 meter band.)

Due to the ionizing effect of the sun, these reflecting layers actually are more concentrated, thus more effective at reflecting radio signals, in daylight hours than at night. Therefore, in theory skywave signals should be even stronger during the daytime than at night. However, it turned out that a inner "D Layer" also existed. And the D Layer absorbs signals in the wavelengths that happened to be assigned for broadcasting, blocking them before they have a chance to reach the reflective outer layers. But unlike the E and F Layers, the D Layer only exists during daylight hours, which is why skywave signals disappear during the day but return at night. An analogy is that, when talking about the wavelengths assigned to broadcast stations, the E and F Layers act as a mirror reflecting signals back to Earth, while the D Layer is a curtain drawn in front of the mirroring layers during daylight hours.

(It is popularly believed that old "Amos and Andy" shows are winging their way through the cosmos. Unfortunately for old radio buffs on alpha Centauri, in most cases these signals actually were snuffed out by the absorbing D and reflecting E and F layers a fraction of a second after they left the radio station. In the mid-twenties amateurs began experimenting with frequencies higher than the traditional 1500 khz. As expected, the higher they got the worse the groundwave signal. However, unknown to the amateurs, when you get above a certain frequency the D layer no longer absorbs the signals, but they continue to be reflected back to Earth. Thus, they stumbled upon the shortwave frequencies, which have almost no groundwave capabilities -- thus are "worthless" under the old view -- but also have globe-spanning skywave coverage, sometimes even better during the day than at night. As you continue to go up in frequency, you eventually reach frequencies which pass through the entire ionosphere, both day and night. Therefore, unlike AM band and shortwave signals, FM and TV signals are indeed spreading throughout the cosmos.)

The greater nighttime coverage on broadcast wavelengths meant it was now possible, at night, for stations to interfere with each other over great distances. In some cases this meant, as reported in the Murfreesboro letter, hearing more than one program at the same time. However, there was an even worse problem. When two stations are close in frequency their signals interact, creating a piercing "heterodyne" tone, which was estimated to extend ten times as far as the audio interference. (For example, if one station were on 833 khz, and the other on 830 khz, the resulting heterodyne tone would be 3 khz, which is the difference between the two station frequencies.) If stations stay within about .05 kilohertz of each other the tone disappears. However, as seen by the earlier WHAS quote on frequency control, with early 1920s technology any such convergence would have only been a fleeting coincidence. (At this time many stations drifted in frequency both in response to what was being transmitted and whenever their antennas swung in the wind. The "flattop" antennas in use at this time had stronger skywave signals, and weaker groundwave, than the modern "vertical" antennas that supplanted the flattops beginning in the 1930s)

Until the development of affordable precise frequency control, plus directional antennas suitable for use on the broadcast band frequencies -- both a full decade away -- the only tools for preventing heterodyning on a common wavelength were wide separation of stations, timesharing, plus reduced nighttime powers and daytime-only operation.
Second National Radio Conference
By early 1923 it had become clear that a major overhaul of the broadcast service was needed. The most critical problem was that two entertainment wavelengths were not nearly enough. Ideally each station should be given its own wavelength, but that was impractical. Secretary Hoover convened a second conference of government and industry representatives, beginning on March 20th.

Once more the conference proposed increasing the number of broadcast frequencies. This time the Commerce Department acted quickly, announcing in early April a sweeping expansion of the broadcast allocation. Over a period of time broadcasting was to be assigned, in 10 khz steps, all the frequencies from 550 to 1350 khz (545 to 222 meters). Stations would still be divided into Class A and B, but this now would refer to two bands of frequencies. Class A stations would be limited to 500 watts, while Class B's would use 500 to 1000 watts of power. Although a few new Class A stations were assigned to the new frequencies beginning in April, the full plan did not start to go into effect until noon on May 15th.

Under the plan, none of the multitude of stations operating on 360 meters would be forced to change to a new frequency -- they could stay on 360 meters, as "Class C" stations, if they wished. However, no new stations would be assigned to 360 meters, and it was hoped that all the current 360 meter residents would soon voluntarily switch to the new, less congested, Class A and Class B frequencies. Once the stations on 360 meters disappeared, the new band would consist of 50 Class B frequencies running from 550 to 1040 khz, plus 31 Class A frequencies, from 1050 to 1350 khz.

The Class A frequencies consisted of lower power stations -- some using a little as 5 watts -- which were located relatively close together. The initial plan specified that about two-thirds of the frequencies could be used in all nine of the radio inspection districts, while the rest would be used in at most three assigned districts. Under this setup, nighttime heterodynes were unavoidable on the Class A frequencies. The upper limit of 1350 khz available for Class A stations apparently was set by the existing ship wavelength at 220 meters (1365 khz).

There were more Class B frequencies available than stations qualified to use them, which was a good thing since a number of the frequencies were not immediately usable. The clump of Class C stations on 833 khz were pretty shaky in the frequency control department, so initially no Class B stations were assigned from 810 to 860 khz, giving the Class C's a little wobbling room. Also, 1000 khz (300 meters) was an international ship frequency, so broadcasters stayed clear of 980 through 1040 until the ships could be reallocated to other frequencies.

The old Class B entertainment wavelength at 400 meters became just another Class B frequency, now known as 750 khz. (Ironically, this frequency was assigned to WHAS, which apparently had finally figured out how to tune its transmitter to 400 meters). The separate Market and Weather wavelength on 485 meters disappeared. To the relief of stations like WHAS, broadcasters now sent out their entire program on their one assigned frequency. However, the government still maintained strict control over the use of official government reports and forecasts. The handful of stations which had no entertainment offerings, and thus were licenced only for 485 meters, were moved to 360 meters.

The Commerce Department made a special effort to assign the showcase Class B frequencies equitably. The United States was divided into five zones, and each zone was assigned at least ten Class B frequencies. Because of the relatively low powers then in use, Zones 1 and 5, on opposite coasts, were far enough apart to permit simultaneous use without nighttime heterodyning interference. However, all the other zones required exclusive use of their frequencies to avoid heterodyning problems. Below is a review of the fifty Class B frequencies, and their zone assignments, as initially announced by the Bureau of Navigation:

550-3      630-4      710-5      790-1      870-2      950-3     1030-4
560-5      640-1,5    720-2      800-3      880-4      960-5     1040-1
570-4      650-3      730-4      810-5      890-1      970-2  
580-2      660-1,5    740-1      820-2      900-3      980-4  
590-1,5    670-2      750-3      830-4      910-5      990-1  
600-3      680-4      760-1,5    840-1      920-2     1000-3  
610-1,5    690-1      770-2      850-3      930-4     1010-5  
620-2      700-3      780-4      860-5      940-1     1020-2

Within each zone, frequencies were assigned for use by specific localities. Commerce was careful to state that frequencies were allocated to jurisdictions, not to individual stations. But they obviously had taken a close look at the 400 meter roster when deciding the initial allocations. One standard was that there be a minimum 50 khz separation between stations in a given locality. This was viewed as the smallest spacing that an average radio could discriminate between when near two stations. There was also a minimum 20 khz spacing within zones.

The final step was to assign stations to the new frequencies. Since there were more frequencies assignments than qualified stations, some Class B frequencies were reserved for later use within specific zones. In some of the more congested cities frequencies were shared by two or three stations.

Below is a review of the initial May 15th Class B allocation, plus the stations that were assigned to them by the end of July, 1923. Seventy-seven years later many of these stations are among the most prominent in the nation. Others, with owners who couldn't afford the expense, later became lesser stations or were deleted altogether. In fact, three stations, WDT (Ship Owners Radio Service), WGM (Atlanta Constitution) and KFDB (Mercantile Trust Company) would be deleted before the end of 1923. Amazingly, given all the changes in the succeeding seven decades, three stations have continuously stayed on the frequencies they received under the May 15, 1923 plan: WMAQ-670 Chicago (now WSCR), KFI-640 Los Angeles, and KSD-550 Saint Louis (now KTRS).

Allocations Announced for May 15, 1923			Station Assignments as of July 31, 1923
Zone	Location	Freq.	Station Assignments as of July 31, 1923
1	Springfield/Wellesley Hills, MA	890	WBZ Springfield, MA
	Schenectady/Troy, NY	790	WGY Schenectady, NY & WHAZ Troy, NY
	New York, NY/Newark, NJ	660	WJZ Newark, NJ
	" "	610	WBAY/WEAF New York, NY
	" "	740	WJY/WOR New York & WDT Stapleton, NY
	Philadelphia, PA	590	WOO/WIP Philadelphia, PA
	" "	760	WFI/WDAR Philadelphia, PA
	Washington, DC	690	NAA Arlington, VA
	Reserved	640	WRC/WCAP Washington, DC
	Reserved: 840, 940, 990, 1040
2	Pittsburgh, PA	920	KDKA East Pittsburgh, PA
	Chicago, IL	670	WMAQ/WJAZ Chicago, IL
	Davenport/Des Moines, IA	620	WOC Davenport, IA
	Detroit/Dearborn, MI	580	WWJ/WCX Detroit, MI
	Cleveland/Toledo, OH	770	WBAV Columbus, OH & WJAX Cleveland, OH
	Cincinnati, OH	970	WLW/WSAI Cincinnati, OH
	Madison, WI/Minneapolis, MN	720	WLAG Minneapolis, MN
	Reserved	870	KYW Chicago/WCBD Zion, IL
	Reserved: 820, 1020
3	Atlanta, GA	700	WSB/WGM Atlanta, GA
	Louisville, KY	750	WHAS Louisville, KY
	Memphis, TN	600	WMC Memphis, TN
	Saint Louis, MO	550	KSD Saint Louis, MO
	Reserved	650	WCAE Pittsburgh, PA
	Reserved: 800, 850, 900, 950, 1000
4	Lincoln, NE	880	---
	Kansas City, MO	730	WDAF/WHB Kansas City, MO
	Jefferson City, MO	680	WOS Jefferson City, MO
	Dallas/Fort Worth, TX	630	WFAA Dallas/WBAP Fort Worth
	San Antonio, TX	780	WOAI San Antonio, TX
	Denver, CO	930	---
	Omaha, NE	570	WOAW Omaha, NE
	Reserved: 830, 980, 1030
5	Seattle, WA	610	KGW Portland, OR
	Portland, OR	660	KDZE Seattle, WA
	Salt Lake City, UT	960	---
	San Francisco, CA	590	KFDB San Francisco, CA
	" "	710	KPO San Francisco, CA
	Los Angeles, CA	640	KFI Los Angeles, CA
	" "	760	KHJ Los Angeles, CA
	San Diego, CA	560	---
	Reserved: 810, 860, 910, 1010

The Commerce Department made a tentative step in establishing frequency control standards by "suggesting" that stations stay within 2 khz of their assigned frequencies. This did nothing to reduce heterodyning interference between stations on the same frequency, but at least it would keep stations from drifting into neighboring frequencies. In spite of the suggestion, there would continue to be reports of stations straying far beyond the 2 khz standard.

Although stations were now being assigned in neat 10 khz frequency steps, the public generally clung to the older, and less precise, wavelength nomenclature, usually stated to the nearest meter or tenth of a meter for the corresponding frequency. It would be more than a decade before wavelength references completely disappeared in the United States, and many in Europe (where AM stations are now allocated in 9 khz steps) still use the older terminology.
Continued Expansion and the Third National Radio Conference
In the year following the May 15, 1923 reallocation the number of Class C stations on 360 meters declined, so the gap of unused Class B frequencies around 833 khz also shrank. Also, with the reduction, and then elimination, of ship transmissions on 300 meters Class B stations were assigned to the frequencies around 1000 khz. However, problems continued, including a shortage of Class A frequencies. Hoover announced a third industry conference, beginning October 6, 1924.

One of the conference recommendations was to increase the number of Class A frequencies. Under the May 15th allocation amateurs had gotten a little more breathing room, as Special Amateurs were permitted to move below the traditional 1500 khz (200 meters) to 1350 khz (222 meters). However, this expansion would prove short-lived in the face of broadcasting's appetite for additional frequencies. In July, 1924 the lower limit for amateurs had been shifted back to 1500 khz. Then, following the recommendations of the Third Conference, starting in November, 1924 Class A broadcast stations were assigned to fifteen additional frequencies from 1360 to 1500. Not that very many stations wanted to go there. Along with low powers, poor groundwave coverage, and interference from the nearby amateurs, these stations were faced with the fact that many radios didn't tune this high.

Following the conference Class B stations were allowed to experiment with powers of up to 5 kilowatts, to be attained in 500 watt steps. (RCA's proposal that stations be allowed to use up to 50 kilowatts was met with shock and a promise to study the matter further).

By April, 1925 the elimination of the Class C stations on 360 meters was essentially complete, and the Class B stations filled in the freed-up frequencies. Thus, from the initial footholds at 360 and 485 meters, broadcasting had expanded in both directions, and now occupied all but the first 50 khz of the 200 to 600 meter band. (Broadcasting's low-end expansion ended at 550 khz due to the need to protect 500 khz -- 600 meters -- from interference. 500 khz was, and still is, an international distress frequency). The three Class A frequencies adjacent to the Class B band had been converted to Class B use, so the broadcast frequencies now consisted of 53 Class B (550 to 1070) plus 43 Class A (1080 to 1500), for a total of 96.

Class B Complexities

Throughout the mid-twenties there was a tremendous turnover of stations. However, whenever one disappeared another popped up to take its place. The overall number of stations fluctuated between 500 and 600. However, powers steadily increased, along with the resulting interference, especially at night. A major problem developed because of a lack of Class B frequencies.

Although Class B radio stations were expensive to operate (and generally there was no direct financial return, as commercial sponsorship was only just beginning to appear) the prestige was great enough that more and more companies wanted one. The crush was exacerbated when the United States, realizing that an entire country was located to its north, informally set aside six Class B frequencies -- 690, 730, 840, 910, 1010, 1030 -- for exclusive Canadian use. (Recognition that other countries, such as Mexico, also existed would not come until 1940 with the NARBA agreements).

As a partial solution, some Class B stations were placed on Class A frequencies, but this didn't do much to satisfy their owners. In 1925 the Commerce Department had experimented with shrinking the spacing between the Class B frequencies from 10 khz to 7.5 khz, but this proved unsuccessful. Finally, in October, 1925, the Commerce Department announced it would generally cease licencing new stations, because the broadcast frequencies were filled beyond capacity.

Legal Actions
Secretary Hoover knew the embargo was on shaky legal ground. For years he had pleaded with Congress to pass a new law, giving him clearer control of radio. However, the two branches of Congress had never come to an agreement, so radio remained under the increasingly creaky control of the 1912 Act.

Moreover, station licencing was not the only area of legal challenge. The Zenith Radio Corporation operated WJAZ, a Class B station in Chicago, which it thought of as a showcase for the firm. Unfortunately, due to the Class B frequency shortage the station was assigned a grand total of two hours per week of air time, on 930 khz. Zenith found it's showcase wasn't very visible. So it moved to 910 khz, which had been one of the exclusive Canadian frequencies, and challenged Secretary Hoover to do something about it.

Ironically, Zenith had no intention of diminishing Hoover's overall regulatory powers. It only claimed it found a small loophole which permitted frequency shifts for a handful of stations which, like WJAZ, had been granted "Developmental" licences. However, earlier challenges had not been favorable to the Commerce Department, and the effects of the WJAZ case instead would be sweeping.

The Commerce Department challenged Zenith's move, and the case ended up in Federal Court in Chicago. In his April 16, 1926 decision, Judge James H. Wilkerson sided with WJAZ on its right to choose its own frequency. However, Wilkerson's ruling mainly addressed the legality of WJAZ's frequency shift, and did not delineate exactly what Hoover could and could not do. The Commerce Department debated whether it should appeal the WJAZ ruling. In the meantime, everyone looked to Congress to pass a new law to stabilize the situation. Congress promptly dropped the ball. Although both branches passed new laws, they were significantly different, and Congress adjourned in early July before the differences could be worked out in committee. Congress would return in session on December 8th, after the elections. Until then Hoover was on his own.

Hoover's next step was to ask Acting Attorney General William J. Donovan for advice on what powers Hoover held under the 1912 Act. Donovan had a difficult task in trying to make sense of the Act and how it related to broadcasting. The bill's language was obscure at times, and some important sections were widely removed from each other, so that their exact relationship was unclear.

The Act was oriented toward to regulating two-way communication, and allowed stations a great degree of flexibility. A key problem was in frequency assignments. The Act stated that stations were to be assigned a "normal wavelength", but they also were allowed to use additional wavelengths of their own choosing, as long as they fell outside of the 600 to 1600 meter government band. In fact, in keeping with standard practice, the first few broadcast licences were actually issued stating that the station's "normal" wavelength was 600 meters -- not that any broadcast station actually ever used this wavelength. Thus, their broadcast authorizations for 360 and 485 meters fell under the category of "additional" wavelengths. These early authorizations, following guidelines set by the Act, also required the stations be capable of communicating with ships on 300 meters, when needed. Not that it ever was.

The Act was also ambiguous whether the Commerce Department could withhold licences from qualified applicants, or could regulate powers and hours of operations outside of the 600 to 1600 meter government band. Given the ambiguity of the Act, various opinions ranged from the extremes that Hoover either had complete authority to regulate broadcasting, or he had virtually none at all.

Donovan released an opinion on July 8, 1926. It wasn't legally binding, but did give the Commerce Department an idea whether it should pursue an appeal of the WJAZ case. As it turned out, Donovan's opinion matched Hoover's worst fears. In Donovan's opinion, except for the government band Hoover not only had to issue licences to all upon request, but he also had no right to restrict frequencies used, hours of operation, or powers. Broadcasting had become a free-for-all. The only thing Hoover could do was ask stations for restraint and try to keep track of things until a new law was passed. Just before the breakdown of regulation Canada had complained that its six exclusive Class B frequencies were not enough. In the "wave jumping" by U.S. stations that followed, it would watch this number drop to zero.
A Little Bit of Anarchy
Because of the new state of affairs, the station list appearing in the December 31, 1926 issue of the Radio Service Bulletin included the following rueful disclaimer: "The power and wavelengths given in this table were compiled from applications for licenses furnished the department by the owners of the stations. Since the department does not make assignments in either respect, this list is not necessarily in conformity with wavelengths or power actually used".

Although the first few months saw relatively few changes, eventually a torrent of new stations and frequency changes developed. In an eight month period around 200 new stations flooded the airwaves. Many stations jumped from Class A to Class B frequencies. Some broke new ground, such as WOBB in Chicago, which was reported to be on 540 khz, a step below the former 550 khz lower boundary of broadcast frequencies. WHAP in New York City decided there was just enough room between WJZ-660 and WOR-740 for another Class B station, and it settled on the unorthodox new frequency of 697 khz. Another case where a station headed for a split frequency was KFKB in Milford, Kansas, which began operating on 695 khz. KFKB was owned by J. R. Brinkley, M.D., the infamous "Goat Gland" doctor. His later stations, on the other side of the Mexican border, would continue this affection for split frequencies.

In some cases it's hard to determine exactly what frequency a station was operating at, because many were still reporting station wavelengths rather than frequencies. Thus, when KEX in Portland announced it was operating on 447 meters, it probably was specifying the nearest whole-meter equivalent for 670 khz. However, the Commerce Department dutifully divided 447 into 299,820, and reported that KEX was now operating at exactly 670.7 kilohertz.

New York and Chicago were worst hit by the increase in stations and congestion, but the effects were felt nationwide, especially with an increase in nighttime heterodynes. In the West, one group of stations staged a novel demonstration in support of the restoration of government controls. According to the June, 1927 Radio Broadcast "Between the hours of eight and nine February 11, KFI, and ten other Pacific Coast stations presented what they termed an Interference Hour. The stations were paired off and so changed their wavelengths as to interfere seriously with one another. After an hour of squeals, howls, indistinguishable announcements, and distorted music, the stipulated wavelengths were resumed, following which pleas were made from each of the stations in support of the radio bill before the senate".

Stations turned to the courts to clear things up. Eventually the courts would have stabilized the situation, as a series of rulings generally gave established stations priority and relief from interference from newcomers. However, these rulings were getting dangerously close to giving stations property rights to their radio frequencies, something the government desperately wanted to avoid.
Re-regulation
Congress reconvened in December, and work slowly began on the radio crisis. Although all agreed that something needed to be done, a controversy broke out whether to strengthen the powers of the Commerce Department, or form an independent commission, modeled after the Interstate Commerce Commission. Finally, on February 23, 1927, President Coolidge signed the newly passed Radio Act of 1927. A compromise, it set up a temporary independent Federal Radio Commission, which would have one year to settle the radio mess. After that most of its powers would revert to the Commerce Department. Most of provisions of the 1927 law were based on the recommendations made by the various Radio Conferences beginning in 1922.

The United States was divided into five regions and five commissioners -- one to represent each region -- were appointed. Two promptly died. (Credo Harris of WHAS turned down the offer of a Commission appointment). It was a high pressure assignment -- radio broadcasting, although only six years old, was seen as a national resource. With the chaos radio sales had declined, and there was a sense that radio was being wasted. The whole country was watching.
Initial FRC Work
The FRC had to act carefully -- every decision was a potential court case. There were a total of 732 broadcasting stations when it took over, far more than could comfortably fit into the broadcast frequencies. The Commission was given the power to delete stations not found to be in the public "Convenience, Interest, or Necessity", but that didn't give it the right to arbitrarily delete stations in bulk. However, it did halt new station grants, except in a few underserved regions of the country.

The Radio Act of 1927 explicitly protected stations from deletion for 60 days following the enactment of the new legislation. When this ban expired on April 25, 1927, the FRC made no move to start culling the broadcasting ranks. Instead, all existing stations were given "temporary" operating extensions. A series of 30 to 60 day extensions followed, eventually dragging out for more than a year.

Ultimately stations would be required to formally apply for licences, which would give the FRC a chance to winnow the ones that didn't meet standards. But first the standards had to be developed. Until then, it was hoped that time would see an attrition in the number of stations. Meanwhile, information was collected on the stations, and various technical tests and studies were conducted in order to get an idea on what could be done with all of them.

Although it was strongly hinted that the broadcast band would be extended by adding 50 broadcast frequencies from 1510 to 2000 khz, in the end the frequencies assigned to broadcasting remained unchanged. (The International Radio Convention of 1927, which met in Washington, DC, specifically set aside 550 to 1500 khz for broadcasting purposes). Among the first actions the FRC did take was to clear out the Canadian frequencies and get all stations back to 10 khz frequencies from 550 to 1500. This produced something roughly like the old Class A and Class B bands, but with a lot of shoehorning in of extra stations. Although they had done nothing illegal, most "wave jumpers" and stations that had popped up in the preceding few months did relatively poorly under the reassignments.

Every few weeks or months new refinements were announced, and stations were shuffled to new spots on the radio dial. The commissioners made visits to the regions they represented, to consult with station owners and evaluate the situation. On their return stations within the region were juggled once again. WEBC in Superior, WI was allowed to increase its power from 250 to 1000 watts "in order to make certain that President Coolidge would have good radio reception at his summer home". Although the initial standards were fairly generous, the overall trend was to reduce interference by reducing the number of stations broadcasting simultaneously. This meant an increase in the number of stations forced to share time, or limited to daytime-only operation.

The Commission made a special effort to clear the key frequencies of 600 to 1000 khz of "heterodyne and other interference", in order to give the listening public an island of better reception while the band was being reconstructed. The FRC applied pressure to get recalcitrant stations to cooperate, proclaiming "Broadcasters who are parties to placing annoying interference, instead of programs, on their respective channels are not looked upon as serving public interest, convenience, or necessity. Instead of creating good will for themselves certain radio stations have become extremely unpopular due either to blanketing or heterodyning interference, complaining letters indicate". It added "Regarding divisions of time requested, the commission feels a distinct service is rendered to any station which is encouraged to broadcast fewer hours under clear reception conditions rather than full time with its signals at most points utterly valueless". However, the clearing effort met with only limited success.

The FRC set a new standard that stations would have to stay within .5 khz of their assigned frequencies. But this was still about twenty times the limit needed to avoid heterodyning other stations on the same frequency. And even this liberal standard proved difficult for most stations to meet. The key objective in the evolving FRC reallocation came to be the reduction of heterodyne interference, especially during the prime nighttime hours.

It became clear the FRC was not going to finish its task in the year allocated by the 1927 Act. On March 28, 1928 Congress approved a one-year extension for the FRC, until March 16, 1929. Many wondered why the process was taking so long. Radio Broadcast informed its readers that, contrary to popular belief, "The Commission is not incompetent; it is impotent".
Portable Stations
The FRC did move aggressively against one class of stations that was a particular annoyance. The Department of Commerce had licenced "portable" stations, usually to transmitter manufacturers, who could move the stations from place to place for demonstrations. The FRC decided it wasn't required to regulate moving targets, so in April, 1927 it restricted portable licences to two frequencies -- 1470 and 1490 -- and announced that eventually all would be eliminated. As of early 1928 there were still about a dozen portable stations, but all were gone by July 1, 1928. Not all were deleted, however. A few were allowed to become permanent stations in underserved areas of the country.
Refining Standards
In March and April, 1928 the FRC, along with industry engineers, worked to finalize the new broadcasting band structure, choosing from among a number of plans submitted by various public and industry representatives. However, in addition to technical concerns, there was also a political one. The legislation continuing the FRC included a clause that came to be known as the Davis Amendment. It required that station allocations be equitably made between the states. The commissioners were divided whether the provisions of the Davis Amendment could be instituted over time or had to be implemented immediately.

Finally the FRC started to pull everything together. All stations were required to formally apply for licences by January 15, 1928. The FRC reviewed the applications, identifying stations which appeared to fall short of meeting the new Convenience, Interest, or Necessity standard. On May 11, 1928 the FRC issued General Order 32. It targeted 164 stations that the FRC felt had failed to meet the new public standard. Hearings would be held July 9, 1928, with the stations to be deleted on August 1st if they were unable to sway the Commission. Most of the stations contested their fate, and a majority survived, with the FRC actually complimenting the work of some of the challenged stations. Figures vary, but between fifty and ninety stations eventually disappeared, many by default or surrendering their licences rather than deletion, and many of the survivors had their powers and hours of operation reduced.

Some of the deleted stations had been found to be no longer operational. Others had served as little more than platforms for their owners, used to fill the airwaves with personal opinions and attacks. Perhaps the oddest case was KFQA, licenced to The Principia in Saint Louis, Missouri. The FRC reported that "During the hearing, held on July 9, the representative of the station urged that all the applicant wanted was to maintain a licence from the commission but did not care about the transmitter". In other words, they wanted a licence, but didn't want to actually operate a station, preferring to broadcast through KWK's facilities. In deleting KFQA, the FRC noted: "This case is a good illustration for a direct application of the principle previously announced by the commission that it is not in the public interest, convenience, nor necessity to continue to licence a station which is not putting its transmitter to any use". (A year later KFQA got its wish, and it became a special callsign for KMOX when broadcasting Principia programming).
New Broadcasting Structure
With the broadcasting ranks now reduced to about 585 stations, the FRC finally announced the long awaited restructuring of the broadcast band. On August 30, 1928, General Order 40 described the new setup. It had taken more than a year for the FRC to come up with a definitive broadcasting reorganization, which was scheduled to take effect at 3:00 AM on November 11th. The Commission itself reported significant disagreement between the commissioners, and the best the final plan could muster was a four to one vote in its favor. The holdout was Commissioner Ira E. Robinson, who reportedly felt the commission was acting rashly, and had favored high-powered stations, to the detriment of the low-powered ones. Nor could Robinson be called a "good loser". After the new plan was announced, he released the following statement: "Having opposed and voted against the plan and the allocations made thereunder, I deem it unethical and improper to take part in hearings for the modification of same".

Using legal language best described as "tortured", it was formally announced "That a band of frequencies extending from 550 to 1500 kilocycles, both inclusive, be, and the same is hereby, assigned to and for the use of broadcasting stations, said band of frequencies being hereinafter referred to as the broadcast band". The new plan organized the broadcast band in a more complicated manner than the previous Class B/Class A setup. Most noticeable was that, instead of two adjacent groupings, blocks of high and low power frequencies were placed at various locations within the band. Also, stations were now divided into three categories, which in time would become known as "Clear", "Regional", and "Local".

Six of the 96 frequencies were off-limits for United States stations, as 690, 730, 840, 910, 960, 1030 were set aside exclusively for Canadian use. The United States was divided into five zones, and forty frequencies -- eight per zone -- from within the range of 640 through 1190 khz were assigned for the primary use of individual zones. These "Clear Channel" frequencies were the successors to the old Class B authorizations, and stations on them would eventually have powers up to 50 kilowatts. Forty regional frequencies were allocated, for stations using a maximum of 1000 watts, to be used concurrently in two to five zones. These were the successors to the old Class A band. Four additional regional frequencies were permitted to use a maximum of 5 kilowatts, as an incentive to get stations to accept the unpopular high-end frequencies of 1460 to 1490. (These frequencies would eventually be converted to Clear channels.) The final six frequencies effectively marked the reappearance of the old Class C 360-meter wavelength. These were to be used by "local" stations nationwide, with a 100 watt power limit. The overall structure of the November 11th reallocation has been modified over the years, but today's AM band strongly reflects this historic restructuring.

Following is the frequency setup that took effect on November 11, 1928, from 550 to 1500 khz. Numbers in parentheses are the zones assigned dominant use of individual Clear Channel frequencies:


550 - 630: REGIONAL
640 (5), 650 (3), 660 (1), 670 (4), 680 (5): CLEAR  
690: CANADA (exclusive)
700 (2), 710 (1), 720 (4): CLEAR
730: CANADA (exclusive)
740 (3), 750 (2), 760 (1), 770 (4): CLEAR
780: REGIONAL  
790 (5), 800 (3), 810 (4), 820 (2), 830 (5): CLEAR
840: CANADA (exclusive)
850 (3), 860 (1), 870 (4): CLEAR 
880 - 900: REGIONAL 
910: CANADA (exclusive)
920 - 950: REGIONAL
960: CANADA (exclusive)
970 (5), 980 (2), 990 (1), 1000 (4): CLEAR
1010: REGIONAL
1020 (2): CLEAR
1030: CANADA (exclusive)
1040 (3), 1050 (5), 1060 (1), 1070 (2), 1080 (3), 1090 (4),
1100 (1), 1110 (2): CLEAR
1120: REGIONAL
1130 (5), 1140 (3), 1150 (1), 1160 (4), 1170 (2), 1180 (5),
1190 (3): CLEAR
1200 - 1210: LOCAL
1220 - 1300: REGIONAL  
1310: LOCAL  
1320 - 1360: REGIONAL 
1370: LOCAL 
1380 - 1410: REGIONAL
1420: LOCAL 
1430 - 1450: REGIONAL
1460 - 1490: REGIONAL (high power)
1500: LOCAL

Radio Broadcast cautiously hailed the new plan. It noted that "We hesitate to praise any constructive step announced by the Commission because, up to this time, it has always reversed itself before promised reforms have been put into operation. It proposed to eliminate all stations persistently wandering from their channels, but backwatered before the echo of its brave statements had died out. It called a host of stations before it to prove they were operating in the public interest, necessity and convenience, and with great fanfare to the effect that they would be weeded out, but the actual result of the hearings was negligible. From past evidence, we cannot avoid fearing a complete reversal of form and a repudiation of the meritorious broadcast allocation plan".

In spite of the fears of Radio Broadcast, the FRC moved forward. Its next hurdle was to assign stations to frequencies for their November 11th debut. There were still signs of tentativeness, as assignments were announced September 10th but then modified on three occasions in October. The Commission also made an unsuccessful effort to rationalize network operations. Chains had started to gain prominence, and the Commission was worried all its hard work would be devalued if all the strongest stations ended up carrying the same programs. However, the FRC eventually gave up its effort to reduce network broadcast duplication, and announced that instead the issue would ultimately be part of a comprehensive review of chain programming.
Effects of the November 11, 1928 Allocation
By all accounts the November 11, 1928 allocation was successful in greatly reducing interference. And the FRC was proud how few stations it had to delete along the way. However, many stations were unhappy with the new allocation, and some headed to the courts to get relief. Most were unsuccessful.

Because of its emphasis on reducing heterodyning interference, the Commission had adopted a very conservative approach, assigning low powers and limited frequency slots. And although they hadn't been deleted, scores of stations had in effect been given death sentences. On the regional frequencies the FRC limited the number of stations operating concurrently to two to five nationwide. And, in major population areas the states were over-represented under the guidelines of the Davis Amendment. Thus, in major metropolitan areas, particularly New York and Chicago, the FRC in some cases required four, and occasionally five, stations to share the same frequency. It was impossible for a station to survive economically on a ration of a quarter or a fifth of a broadcast day, especially with the coming of the Depression in late 1929. Fierce legal battles broke out, as stations used the FRC and the courts to wrest broadcast hours from -- or kill off -- the stations they were partnered with. Some of these legal battles lasted years and gained legendary status within the broadcast industry, and were credited with financing the college educations of numerous legal counsel's children. (Ironically, many educational stations were paired with commercial stations, which often lead to the demise of the educational stations. This was one of the main reasons educational channels were set aside when the FM band was created.)

The final timesharing agreement in the New York City area wasn't consolidated until 1985, and the last timesharing arrangement that dated back to November 11, 1928 -- WEDC/WCRW/WSBC on 1240 khz in Chicago, IL -- only recently ended. (The owners of WSBC purchased WCRW, which stopped broadcasting in July, 1996, then bought out WEDC, which made its final broadcast June 12, 1997, ending 68½ years of time-sharing).
Consequences and Conclusions
The November 11, 1928 reallocation was a major achievement, as government regulators finally regained control over the broadcast band, lost a year and a half earlier. But there was still plenty of work to be done. The Commission had to refine the equalization of station grants, as required by the Davis Amendment. The early thirties saw the development of "vertical" antennas, which replaced the old "flattop" antennas. The new antennas had better groundwave coverage, at the expense of reduced nighttime skywave service. They also could be set up as directional antennas, which, combined with better frequency control that finally eliminated audible heterodyning, allowed closer placement of stations with less interference. Despite the FRC's "temporary" status, and court challenges by disgruntled stations over its constitutionality, the Radio Commission survived until 1934, when it was replaced by the Federal Communications Commission. (In contrast, Radio Broadcast expired in 1930).

In the early forties the North American Regional Broadcasting Agreements extended the broadcast band to 1600 khz. However, the overall November 11, 1928 structure remained intact. The lower frequencies were unaffected, and in most cases where stations were moved to a new frequency, all the stations on a given frequency moved to a new dial position as a group.

After World War II there was an easing of interference standards, and thousands of stations were added to the AM band. Still, even today, on many frequencies there is a core group of pioneer stations that have shared a common frequency since 1928. One change has been an increase in power limits -- to 50,000 watts on the old Clear and Regional frequencies (now known as Class A and B respectively), and from 100 to 1000 watts on the Local frequencies, now known as Class C.

It's an overused phrase, but the best description of the November 11, 1928 reallocation is that it "brought order out of chaos". And nearly seventy years later this historic work still provides the underpinning for the AM broadcast band. From its tentative beginnings on 360 and 485 meters, and through its descent into chaos, broadcasting had finally been given a stable and secure foundation.

Allocation Overview

Kilohertz =>	500	540	550	619	666	750	833	870	1000	1050	1070	1365	1500	>1500
Mid-1921	Ship				Relay				Ship			Ship	Amateur
Dec. 1, 1921	Ship			M/W	Relay		Ent.		Ship			Ship	Amateur
Late Sep 1922	Ship			M/W	Relay	B	A		Ship			Ship	Amateur
May 15, 1923	Ship		===Class B====				=C=	Class B	Ship	=====Class A=====		Amateur
Nov 1924	Ship		===Class B====				C	==Class B==		==========Class A==========				Amateur
April 1925	Ship		===============Class B===============								======Class A======			Amateur
7/1926-3/1927	Ship	Anarchy												Amateur
Nov 11, 1928	Ship		FRC Reorganized Band: '''''''''\|\|\|\|\|[\|\|\|[\|\|\|\|'\|\|\|\|\|[\|\|\|''[''''[\|\|\|\|'\|[\|\|\|\|\|\|\|\|'\|\|\|\|\|\|\|*'''''''''''''''''''''!!!!*

Meters ==>	600			485	450	400	360		300			220	200	<200

The above chart is a general overview of the evolution of the broadcast band, and selected wavelength and frequency allocations from 1921 to 1928. Wavelengths are listed horizontally across the top of the chart, with the kilohertz equivalents directly below. Individual wavelength assignments are marked with a single entry, explained below. Bands of frequencies are marked with double lines. The entries include:

M/W: "Market & Weather" (485 meters/619 khz) -- broadcasting wavelength used from December, 1921 to May 15, 1923 for official government reports, including market reports and weather forecasts. Discontinued after the May 15, 1923 expansion.

"Ent.", A, C: Entertainment wavelength (360 meters/833 khz) -- broadcasting wavelength used for entertainment offerings beginning in September, 1921 and formally assigned December 1, 1921. In September, 1922, with the creation of the "Class B" entertainment wavelength, 360 meters became known as the "Class A" entertainment wavelength. On May 15, 1923, with the creation of "Class A" and "Class B" frequency bands, it became known as the "Class C" wavelength. It quietly disappeared in mid-1925 when the final holdouts were moved to Class A and B frequencies.

B: Entertainment wavelength (400 meters/750 khz) -- created late September, 1922 for better quality stations. Expanded to a band of Class B frequencies on May 15, 1923.

Ship: International ship wavelengths. 300 meters and 220 meters were quickly absorbed by the expanding Broadcast Band, while 600 meters (500 khz) remains an international distress frequency to this day, thus a barrier for any expansion of the AM band to lower frequencies.

Relay: Special Amateur Relay (450 meters/800 khz) -- One of the wavelengths set aside for relay work by Special Amateurs. Special Amateur work was moved to the 1350 to 1500 band in the May 15, 1923 reallocation, and later discontinued altogether.

Amateur: Standard amateur wavelengths.

FRC Reorganization: Graphical representation of the 96 frequencies assignments, from 550 to 1500 kilohertz, under the November 11, 1928 plan. The following symbols are used:

Regional (40)	'
U.S. Clear (40)	\|
Canadian-only (6)	[
Local (6)	*
High-power Regional (4)	!

BIBLIOGRAPHY Following are the major sources for this work: DeSoto, Clinton B. "Two Hundred Meters and Down". The American Radio Relay League, Inc., 1936. Harris, Credo Fitch. "Microphone Memoirs". The Bobbs-Merrill Company, 1937. Pejza, Father Jack. "A Beginner's Guide To The Ionosphere". DX Monitor, International Radio Club of America, March 25, 1972. "Commercial and Government Radio Stations of the United States". Annual list issued as of June 30th for 1920 through 1931 by the Department of Commerce. QST. Selected issues from 1920 to 1922. Radio Broadcast. Selected issues from 1922 to 1927. "Radio Communications Laws of the United States and the International Radiotelegraphic Convention". August 15, 1919 edition. Issued by the Bureau of Navigation, Department of Commerce. Radio News. Selected issues from 1920 to 1927, especially The Development of Radiophone Broadcasting by L. R. Krumm, September, 1922, p. 467. Radio Service Bulletin. Issued monthly, beginning in January, 1915 by the Bureau of Navigation, Department of Commerce. Continued in various formats until 1952. Included occasional broadcast station lists plus changes in regulations, including FRC General Orders. "Regulations Governing Radio Communication". September 28, 1912, February 20, 1913, and July 1, 1913 editions. Issued by the Bureau of Navigation, Department of Commerce. "Report of the Federal Radio Commission". Annual reports, 1927 through 1933.

THOMAS WHITE'S SITE: