Heising Reactor vs Modulation Reactor

Discussion in 'Amplitude Modulation - AM Fans' started by N1BCG, Mar 4, 2019.

ad: L-HROutlet
ad: l-rl
ad: Left-3
ad: L-MFJ
ad: Subscribe
ad: DLSpec-1
ad: Left-2
ad: MessiPaoloni-1
  1. N1BCG

    N1BCG Ham Member QRZ Page

    Setting the record straight on these terms:

    "It is important to remember Heising modulation does NOT use a modulation transformer. Contrary to popular misuse on AM phone forums and on the air, a dc current bypass choke or modulation reactor used in conjunction with a transformer is NOT a Heising modulation system! The modulation choke or reactor in a transformer coupled systems prevents power amplifier steady-state current from flowing in the modulation transformer and biasing the modulation transformer core. While this is often called "Heising modulation" and the choke is sometimes referred to as a "Heising reactor", the use of "Heising" is incorrect."

    https://www.w8ji.com/Heising modulation.htm

    From the 1976 Amateur Radio Handbook:

  2. AC0OB

    AC0OB Subscriber QRZ Page

    And, W8JI never identifies his Point “C” nor voltages at any nodes which would have helped future modifiers and experimenters. In addition, all L1’s in the Globe Scouts I have seen are 7H. He also swapped R101 and R106 in his discussion.

    Heising's original paper: Heising, R.A., Modulation in Radio Telephony, Presented before The Institute of Radio Engineers, New York, December 1, 1920.

    Raymond A. Heising was born August 10, 1888, at Albert Lea, Minn. He received a degree in electrical engineering from the University of North Dakota in 1912 and the master's degree from the University of Wisconsin in 1914. After graduation, Heising began his work with the Western Electric Company and then Bell Telephone Laboratories. He played a major role in the development of radio-telephone systems for military use in World War I, and for transoceanic and ship-to-shore public service use. He invented several modulation systems which are still in wide use today: the grid modulation system for radio, the rectifier modulation system used in carrier telephony, the Class C stage, and in particular, the constant-current or Heising modulation system. He was a member of the I.R.E. (forerunner of the IEEE) and the American Physical Society. (See: Physics Today. 18 (4): 105. April 1965).


  3. W2NBC

    W2NBC Ham Member QRZ Page

    Seems like "modified Heising" just might be a little easier to say and quickly understood on the air than " a parallel-feed modulator circuit with modulation reactor "
    A gold standard reference :

    AC0OB likes this.
  4. N1BCG

    N1BCG Ham Member QRZ Page

    There seems to be a renewed interest in building AM rigs, even among younger amateurs, which is a fantastic trend so the design discussions should use terms that accurately describe the circuits.

    The definition of Heising modulation and the application of an inductor to provide a constant current form of modulation render any other reference as inaccurate and misleading.

    The term "modified Heising" as it's commonly used suggests the role of the inductor as well outside its function as a Heising reactor. Modulation reactors are used to take the D.C. supply off the secondary of a modulation transformer and do not carry current for the modulator supply. This is likely what is meant in the majority of discussions.

    Heising reactors carry the D.C. supply for both a single ended modulator and the modulated stage. The current supplied to the circuit remains constant due to the modulating signal causing it to shift between the modulator and modulated RF stage through the Heising reactor.

    Easiest way to remember: There's either a mod transformer, possibly with a reactor -or- a Heising reactor.

    "I use a modified dipole and feed it at one end". Nope.
    Last edited: Mar 4, 2019
  5. W7TFO

    W7TFO Subscriber QRZ Page

    Yo'all just don't choke on it...:)

  6. K4KYV

    K4KYV Subscriber QRZ Page

    An intermediate form of "modified Heising" would be to use a mid-tapped choke (reactor), with the +HV applied to the mid-tap, and a second modulator tube in push-pull with the normal modulator tube, with its plate connected to the other end of the mid-tapped reactor. The two tubes in push-pull could be class A, AB or B. You would have essentially a push-pull plate modulator, using a modulation autotransformer with a 2:1 plate-to-plate step-down ratio, or a 4:1 impedance ratio. This would allow for cleaner audio, since a push-pull stage in any class of service is more linear than a single-ended stage. The series resistor shunted by a by-pass capacitor (C110 and R106 in the W8JI schematic) would still be necessary to achieve 100% modulation. UTC offered such a mid-tapped reactor in their early 1930s catalogue, with at least one version rated up to 500 watts of audio to modulate a kilowatt input.

    Hams have successfully used this circuit when no modulation transformer was available, by employing the HV mid-tapped winding of a power transformer. However, to avoid core saturation and allow satisfactory performance, a separate modulation reactor is needed to take the DC current flow off one side of the mid-tapped winding, or else the transformer could be taken apart and the laminations re-stacked to form a gap in the core to accommodate the unbalanced DC. The separate modulation reactor could be a power supply filter choke, at least 10 henries or so, rated at the plate current the final amplifier stage is drawing. Not broadcast quality, but using readily available power supply iron, at least "ham radio quality" on par with a DX-100 or Viking I/II.

    If a real modulation transformer eventually becomes available, it could easily replace the mid-tapped choke, and the need for C11o/R106 eliminated. This demonstrates how Heising modulation eventually morphed into regular plate modulation, but it took Loy Barton's treatise on class-B audio amplification circa 1930 to make this scheme widely known.
    W2VW, AC0OB, W2NBC and 1 other person like this.
  7. AC0OB

    AC0OB Subscriber QRZ Page

    1) I have Heising's original paper and he never stated the reactor could not be used with other modulator/modulation circuits.

    2) The modulation reactor does divert dc away from the modulation transformer's secondary to reduce core saturation, but the modulation reactor itself still carries DC current as well as an ac component.

    So the Heising inductor still carries dc current, whether by itself in a class A modulator, or as a capacitive-coupled inductor in a combination with a Class B modulation circuit.

    I.e., I think a delineation is being attempted that is not necessary.

  8. AC0OB

    AC0OB Subscriber QRZ Page

  9. SM0GLD

    SM0GLD Ham Member QRZ Page

    Or maybe not in push-pull, just single ended like this.
    How would this set up perform, is it of any use?
    It would be possible to use an old audio transformer and have the dc magnetizing current canceled out.
  10. K4KYV

    K4KYV Subscriber QRZ Page

    The PA plate loading to the antenna would have to be adjusted so that the PA plate current is identical to the total modulator plate current, to fully cancel the magnetizing current in the transformer winding; this would be consistent with the classic case of "constant current" Heising modulation in which modulator and PA plate currents are the same. A high power audio output transformer, or a power transformer could be used.

    The tap on the reactor could be located slightly to one side of the mid-point, towards the modulator plate(s), to boost the peak audio voltage output from the modulator, thus eliminating the need for the 470Ω/47mfd R-C combination in the PA plate lead. This would, in effect, turn the tapped reactor into an autotransformer.

    What exactly is the purpose of the tap on the 470Ω cathode resistor and the 1N4007? The DC screen voltage on the 6146 would be much higher than the voltage at the tap on the cathode resistor of the 6L6s, so it would conduct only on negative peaks near 100%, thus brick-wall limiting negative peak modulation of the screen. Wouldn't this produce distortion right at the crest of negative peaks approaching 100%, since the plate would still be getting the full negative peak with no similar limiting, introducing a discontinuity to the ratio of plate and screen modulation on negative peaks?
    AC0OB likes this.

Share This Page