Historic Papers

Historic Papers

AM

• About Diode Demodulators, Jim Tonne, 2009. Though not historic, this paper reviews the operation of diode based envelope detectors for AM. It reviews several envelope detector designs and looks at the causes of distortion of the demodulated waveform on the negative peaks. The paper references "the slew rate problem" where the capacitor of the low pass filter used to remove residual RF after the diode has a low resistance charge path (through the diode) and a high resistance discharge path (typically a resistor to ground).

Another perspective, not mentioned in the paper, is that an envelope detector should work without distortion if the output resistance of the demodulator (before the low pass filter) is zero, or at least very low when compared to the input resistance of the low pass filter. If, for example, the RF driving the diode has a source resistance of 0 ohms, which then drives an ideal diode followed by a resistor to ground of 1 ohm, driving a low pass filter with an input resistance of 100 ohms or greater, the capacitors in the low pass filter would "see" equal resistances for charge and discharge. It is interesting that none of the sample circuits take this approach. It is interesting to compare the envelope detector to a synchronous detector.

A synchronous detector, of course, multiplies the incoming modulated waveform by a sine wave that is in phase with the original carrier. If, instead, the incoming modulation were multiplied by a square wave instead of a sine wave, and that square wave is biased up so that the low side is at zero volts, the resulting output is a half wave rectification of the input waveform. This, of course, is the first part of envelope detection. Moving to the frequency domain, multiplication of the incoming modulated waveform by a sine wave that is in phase with the carrier results in a DC component proportional the the product of the two waveforms plus a sine wave at twice the carrier frequency (difference frequency of zero, sum frequency of twice carrier). The low pass filter removes this twice carrier component, leaving the DC proportional to the amplitude of the incoming waveform. As this amplitude varies, the "DC" varies. This comes out of our low pass filter as the demodulated waveform.

Use of a square wave instead of a sine wave results in higher harmonics of the "difference" frequency. If the carrier is 1MHz, and we use a square wave, it has components at 1MHz, 3MHz, 5MHz, etc. Using the "sum and difference" result of analog multiplication, we find the "sum" results coming out of the analog multiplier as 2MHz, 4MHz, 6MHz, etc. The "difference" frequencies (other than zero) land on the same frequencies. Each of these are removed by the low pass filter. (comments by Harold Hallikainen)

• AM Transmitter Performance Without Filter Chokes, Wallace W. Wahlgren, Electro Engineering Works, San Leandro, CA. From 1964 National Elecronics Conference Proceedings. Scanned by John Lyles. 1.6MB
• The WLW 500kW Transmitter, Proceedings of the Institute of Radio Engineers, Volume 22, Number 10, October 1934. Describes the design and installation of the transmitter. Among the unique features are use of high level plate modulation with a class B modulator (with modulation reactor), a series fed vertical radiator, ability to operate at reduced power with portions of the transmitter removed from operation for repair, and use of coax between the transmitter and the antenna tuning unit. Another intersting feature is that the tower lighting AC wiring is enclosed in the output inductor of the antenna tuning network. A similar technique has been used relatively recently to get a variety of signals (AC and RF) across the base insulator of series fed towers. File provided by Stanley Adams. File size 5.4MB.
• A New High-Efficiency Power Amplifier for Modulated Waves - by W. H. Doherty, Bell Telephone Laboratories, Presented at the annual convention of the Institute of Radio Engineers, Cleveland, Ohio, May 1936.
• Ground Systems As A Factor In Antenna Efficiency - by G.H. Brown, R.F. Lewis, and J. Epstein of RCA, Proceedings of IRE, June 1937.
• How Radio Grew Up from Radio Broadcast, December 1925. From http://www.coutant.org/radiogrow . 27.5MB.
• The Pulse Duration Modulator: A New Method of High-Level Modulation in Broadcast Transmitters, H. Swanson, Gates Radio Company, IEEE Transactions on Broadcasting, December 1971. Scanned by John T. M. Lyles.
• Gates PDM Transmitters - High Level Plate Modulation without Modulation Transformers or Reactors, Hilmer Swanson, Gates Radio Company - 1972 NAB. Scanned by John T. M. Lyles.
• Modulation Systems for Amplitude Modulation - From the Edmund LaPort Archive (RCA 1920-1950) of the Smithsonian Museum of American History. Describes AM modulation systems including low level, class B high level, Doherty, and "out phasing" (AmpliPhase). Scanned by John T. M. Lyles.
• Powell Cathanode Modulation System, R. J. Rockwell, 1966. A system for high level, high efficiency amplitude modulation is described which avoids the use of a modulation transformer, thereby permitting the application of wideband feedback, resulting in very low distortion and a full power pass band of 20 to 20 000 Hz. File size: 2.6MB
• Technical Evolution of American Broadcast Transmitters - From the Edmund LaPort Archive (RCA 1920-1950) of the Smithsonian Museum of American History. A brief description of the history of AM broadcast transmitters and their manufacture. Scanned by John T. M. Lyles.
• A Plate Modulation Transformer for Broadcasting Stations, Loy E. Barton, Associate Professor of Electrical Engineering, University of Arkansas. Bulletin Number 8, Engineering Experiment Station, Fayetteville, AR, May 1930. Describes the design and advantages of the use of a modulation transformer for high level plate modulation. Includes transformer design for a class B modulator driving a class C RF amplifier. Does not include the use of a modulation reactor to remove DC magnetic flux from the transformer core. Scanned by John T. M. Lyles.
• Amplitude Modulation - 1973, Ronald Graiff, Allocations and RF Systems Engineer, American Broadcasting Company. Reviews the effects of AM modulation where positive peak is higher than negative (to FCC limit of +125%), carrier shift, etc. Scanned by John T. M. Lyles.
• Enhancing AM Signal Quality Through Improved Modulation Techniques, Brian C. Cox, Harris Corporation. Demonstrates how a PDM transmitter handles nonsinusoidal waveforms better than the typical plate modulated transmitter. Scanned by John T. M. Lyles.
• Performance Measurements and Test Techniques For Modern AM Broadcast Transmitters, Jeffrey Malec, Harris Corporation, Broadcast Division. IEEE Transactions on Broadcasting, June 1989. Discusses test techniques for AM broadcast transmitters including audio frequency response, thtal harmonic distortion, AM signal to noise, positive peak modulation capability, carrier shift, squarewave tilt and overshoot, audio intermodulation, transient intermodulation, out of band emissions, incidental quadrature modulation, CQAM L-R frequency response, efficiency, conducted emissions, temperature, humidity, altitude, line voltage variations and transients, electrostatic discharge, AC line phase imbalance and loss of phase, VSWR tolerance, and high voltage discharge into output network. Scanned by John T. M. Lyles.
• The Dependence of AM Stereo Performance on Transmitter Load Phase, Jerry M. Westberg, Westberg Consulting. This gives results of stereo performance done at 590kHz where a non-ideal load was rotated by use of a line stretcher. Scanned by John T. M. Lyles.
• Improving the Efficiency and Reliability of AM Broadcast Transmitters Through Class-E Power, David W. Cripe, Broadcast Electronics. This paper compares the operation of Class-D and Class-E power amplifiers and predicts how the performance and reliability of each is affected by the non-ideal conditions likely to occur in an AM broadcast transmitter. Scanned by John T. M. Lyles.
• Radio Antenna Engineering, Edmund A. Laport, chief engineer RCA, Fellow IRE. 1952. This treatise deals with antennas made of wires, masts, and towers up to about 30 megacycles. This book attempts to compile a sufficient amount of engineering information to enable nonspecialists to handle many of the ordinary antenna problems that arise in point-to-point, ground-to-air, and military communications, and broadcasting. 36.7MB
• Second Generation Techniques For AM Stereo Exciter Design, Edward J. Anthony, Broadcast Electronics. This paper reviews some of the improvements and new approaches developed by Broadcast Electronics during the design effort for the AX-10 exciter. Scanned by John T. M. Lyles.
• Some Developments in Common Frequency Broadcasting, G. D. Gillett, Presented at the Sixth Annual Convention of the Institute of Radio Engineers, June 4-6, 1931. Published in the Proceedings of the I.R.E. 19, 1347-1369, August 1931. Describes techniques used by WHO and WOC to simultaneously broadcast the same program on the same frequency while minimizing interference through frequency synchronization. Scanned and contributed by Stanley Adams. 1.4MB.
• Spectrum Analysis - Amplitude & Frequency Modulation, local copy, HP Test & Measurement Application Note 150-1, 1996. Provides a nice explanation of why AM and FM generate sidebands, and how to use a spectrum analyzer to analyze AM and FM signals. 1.4M. Reproduced as authorized by Agilent Technologies. Contributed by Stanley Adams.
• Audio Signal Peak Energy Equalization - Patent 3,060,389 by L. R. Kahn, October 23, 1962. Basis for the Kahn SymmetraPeak
• Noise Free Radio - A proposal to run narrow band FM on the medium wave (AM broadcast) band by George W. Yazell, P.E.. Does not address relationship between FM bandwidth and demodulated signal snr. Uses relatively complex method of modulation and demodulation.
• Western Radio Catalog from 1929. 23.6MB. From http://www.coutant.org/western .

FM

• FM Patents

  • 1941066, Edwin Armstrong, December 26, 1933. Describes a method of demodulating frequency modulated signals.
  • 1941068, Edwin Armstrong, December 26, 1933. Describes the use of quadrature AM to phase modulate the signal created by a crystal oscillator. To keep phase linear with relation to input voltage, phase deviation needs to be 30 degrees or less, since below this angle, the sine of the angle and the angle itself are nearly proportional. Further, since this phase modulates the signal, instead of frequency modulates, it is necessary to drive the system with a modulating signal with where voltage is inversely proportional to frequency (provide a -6dB per octave preemphasis to convert phase modulation to frequency modulation). The resulting FM is relatively narrow deviation, so it must be multiplied many times to get the desired deviation. The patent gives an example with frequency multiplication of 192 times. When higher modulating frequencies are used (such as for television as opposed to voice), a higher master oscillator frequency is desirable. It then becomes difficult to get the desired deviation through frequency multiplication. This can be resolved by heterodyning the multiplied frequency down for transmission or further multiplication.
  • 1941069, Edwin Armstrong, December 26, 1933. Describes a method of increasing the distance of transmission on very short waves. Notes that at high frequencies, there is low atmospheric static, so tube noise of the first amplifier in the receiver becomes the limiting factor in transmission distance. The use of wide deviation frequency modulation overcomes this tube noise.
• A Method of Reducing Disturbances in Radio Signaling by a System of Frequency Modulation, by Edwin H. Armstrong, Proceedings of the Institute of Radio Engineers, Volume 24, Number 5, May 1936. This paper, for the first time, demonstrates the noise reduction advantages of wide band FM over AM. Prior analysis explained that FM required a wider RF bandwidth with no advantage. The frequency modulation was created by generating a phase modulated carrier at a relatively low RF frequency, then running it through a series of frequency multipliers to arrive at the broadcast frequency (about 40MHz). A later design included a hetrodyne circuit so the broadcast frequency could be changed without having to retune all those frequency multipliers. The phase modulation is generated by adding the unmodulated carrier to the output of a balanced modulator shifted 90 degrees (quadrature amplitude modulation). The paper also discusses the use of both AM and FM subcarriers on the FM carrier. A suggested use of the subcarrier is facsimilie broadcasting (a predecessor to today's data broadcasting proposals).
• The Building and Tests of Empire State Building and Alpine between 1933 and 1940] includes the received image of the front page of the November 24, 1934 New York Times. File contributed by Stanley Adams. File size 2.7MB. The paper Electrical Engineering, December 1940. This is perhaps the fullest treatment of his FM tests and early drive for FM. Radio Electronic Laboratories of Long Island enter the development at this point in time.
• A Receiver For Frequency Modulation, J. R. Day, Columbia University, as published in Electronics, June 1939. This is one of the very first of the commerical receivers, it rivals the REL and GE in performance and in circuit design. This type of receiver was built primarily for commerical use for those pioneer broadcasters. It was designed and described by a worker of the MHL in which EH Armstrong was director at this time and inventor of the System of FM Communications. Contributed by Stanley B Adams. 871k
• The Composite Signal - Key to Quality FM Broadcasting by Geoffrey N. Mendenhall, P.E., Broadcast Electronics, a tecnical paper oriented toward helping the station engineer better understand how the transmission system affects the composite signal. Scanned by John T. M. Lyles.
• Fine Tuning FM Final Stages, by Geoffrey N. Mendenhall, P.E., Broadcast Electronics, how tuning an FM transmitter affects FM sidebands and synchronous AM. Scanned by John T. M. Lyles.
• Early FM - Several New York Times articles on early FM including transition to the existing band in the late 1940s.
• FM Stereo Report and Order, 1961. Contributed by Mark Humphrey. File Size: 171kB. This R&O established the standard and rules for FM stereo operation in the US. The system numbers referenced in the document are:
  1 - Crosby-Teletronics Corporation
  2A - Calbest Electronics
  2B - Multiplex Development Corporation
  3 - Electrical and Musical Industries (EMI Percival)
  4 - Zenith
  4A - General Electric (identical to Zenith)
  5 - GE Alternate Proposal
  6 - Philco