Screen/Plate Modulated 50W Input Mod Reactor Question

Discussion in 'Amplitude Modulation - AM Fans' started by W2WDX, Jan 4, 2018.

ad: L-HROutlet
ad: l-rl
ad: Subscribe
ad: l-assoc
ad: Left-2
ad: MessiPaoloni-1
ad: L-MFJ
ad: Left-3
  1. W2WDX

    W2WDX Subscriber QRZ Page

    Hi all,

    I have been thinking about building up small transmitter and have a question about adding a modulation reactor in order to remove the plate current off the Modulation transformer. Essentially it's for a 50W input 6146 based design, much like a Johnson Ranger; screen/plate modulated. I'm not talking about Heising modulation, just a simple choke reactor.

    I want to use available new parts, so the tube compliment will most likely be 6146, 12AX7/AU7, 6l6 or similar. (Still working out the oscillator, Digital or Analog?). The modulator will be a basic Class B PP type. I will be using all Hammond transformers throughout, including their new Modulation transformer (found here). However, I do want to remove the plate current from that mod transformer, so I figured I could use a choke reactor. Problem is I can't wrap my head around what the values or type would be usable for such a reactor. How many Henries, how much current rating, etc. Would a regular DC choke used for power supplies work? Should it be 20H, 30H or higher? Those sorts of questions.

    Also will this topology work properly with the typical screen modulation coming off the plate voltage B+ layout?

    Now don't reply to this with stuff like "just use grid modulation" or "use a Heising modulator" or "why not do this or that". It's not that I do not want to try those other modulation methods at this power level, it's just this is what I want to build. If the mod reactor isn't practical, I'll just use the basic screen/plate topology and run the B+ current through the mod tranny in typical fashion. I just want to explore this topology. So suggestions on how to calculate this choke is what I am asking for.

    Thanks in advance ...

    I should know this stuff already ... but ... ummm ....
    Last edited: Jan 4, 2018
  2. WZ5Q

    WZ5Q Ham Member QRZ Page

    It depends upon your modulation impedance as to the minimum modulation reactor inductance. The rule of thumb is 8H per every 1000 ohms modulation impedance.

    whatever your projected RF Final Current will be with some added headroom. Be aware that the rated inductance of the reactor is usually given at a specified current.

    Yes it will, but not as good as a reactor made for the purpose, but they are getting hard to find. Use whats available.

    Use the highest value in Henrys as practical.
    You can series string them together for more inductance if need be.

    yes, it will.
    Hard Driven Triodes are the best for modulation linearity in Plate Modulated service, but If you must use a combination of Plate and Screen modulation, then it's best to use the WA1KNX, Deano Modification where you find the sweet spot using both the modulated and un-modulated B+. Actually a screen tertiary winding off the modulation transformer is the best, but finding one these days isn't practical.
    Last edited: Jan 4, 2018
    W2VW likes this.
  3. N2DTS

    N2DTS Ham Member QRZ Page

    I tried the choke and oil filled cap setup once with a 50 h Dahl reactor.
    Sold it, as with a good mod transformer it did little or nothing for the sound.
    Do you really NEED clean sine waves at 40 Hz?
    If you are looking for a really clean output at 20 Hz, plate modulation is not the way to go.
    That mod transformer looks light for the job, only 100 ma leaves little headroom, maybe that is why you are doing modified Heising?
    Used mod iron is around, the lower power stuff is easy to find and cheap.
    If you run a mod transformer with half the current rating through it, it sounds MUCH better then at maximum current.

    I have found the ratio of modulated to unmodulated voltage feeding the screen important if you go that way.
    My 32V would barely make 100% stock but with a 50/50 mix of modulated/unmodulated voltage it does 130% or more positive.

    I like a separate adjustable screen supply with a choke, dropping resistor (large pot) and an overload relay.
    That may be over the top for a small low power rig.

    Think about using a 4D32 in place of the 6146, almost twice the power output and cost less.
    I could give you a tube and socket...
    Almost 50 watts carrier with super clean screen modulation...
  4. N2DTS

    N2DTS Ham Member QRZ Page

    Oh, and I can likely help you out with any parts you might need, tubes, resistors, sockets and so on.
    I am sort of swamped with homebrew stuff so I doubt I will need/use all the stuff I have.

    I don't think I have any high value chokes (40 to 50 H), they are very rare.
    But caps, tubes (4D32, 2E26, 5763 and so on I have lots of.
    Some small transformers and chokes, switches, lights, jacks, wire, etc.
  5. K4KYV

    K4KYV Subscriber QRZ Page

    I lucked out and found a 40H, 120 mA modulation reactor at a hamfest. The only one I have ever seen in over 50 years of parts accumulation. It's now on my Eico 730 modulator/ 720 transmitter (single 6146 @ 65 watts input), and and works beautifully; that's my 10m rig.

    One possibility, if you feel like trial-and-error engineering, would be to find a larger open frame smoothing choke, 15 to 20 Hy, rated for more current like 250 mA. Loosen the laminations and remove part of the gap spacing material. That will substantially increase the inductance at circa 120 mA. Experimentally find the gap spacing that gives the best performance at that current.

    Sometimes a power supply choke, due to internal capacitance in the winding or core material or lamination thickness will work OK at lower frequencies but shunt the highs. The solution for that is to find a smaller reactor, about 0.5 to 1 henry, and wire in series with the main reactor. The smaller reactor will likely have less internal capacitance and will serve the reactor function at higher frequencies, while the larger one takes over at lower frequencies. Sort of like a woofer and tweeter speaker system.

    For best performance with minimal phase shift distortion, the modulator should be flat to at least one octave above and one octave below the desired frequency range of the modulating audio. For example, if you are looking at voice-modulating over a range of 150~ to 3500~, the modulator should be close to flat over a 75-7000 cps range. That's why many BC transmitters tout something like a 30-15K frequency response; 60-7500~ is a pretty good audio response for AM broadcast audio (many BC receivers pass little more than slopbucket frequency response), but the additional flat response capability better preserves the audio waveform.

    Several power supply chokes in series works for the reactor, although this may take up more room than desirable. Still that would be a workable solution, and you could stay on the lookout for something more optimum to install later.
  6. WZ5Q

    WZ5Q Ham Member QRZ Page

    I did the same thing and lucked out and found 120H, 240mA Mod Reactor at a hamfest, and it's HEAVY!
    Check this one out Don...

    Reactor - Modulation 120H 250mA - 1.jpg

    Reactor - Modulation 120H 250mA - 2.jpg
  7. N2DTS

    N2DTS Ham Member QRZ Page

    Just what you need for a ranger!
  8. K4KYV

    K4KYV Subscriber QRZ Page

    Interesting that despite 120 henries of inductance, it's rated down to only 90~.
  9. N2DTS

    N2DTS Ham Member QRZ Page

    Why would it be rated at all?
    Voltage and current yes, but why frequency?
  10. AC0OB

    AC0OB Subscriber QRZ Page

    That's either the -1dB or -3dB frequency response points at the rated current and inductance.


Share This Page