Scream Modulated 4D32

Discussion in 'Technical' started by W5HRO, Apr 8, 2014.

  1. W5HRO

    W5HRO Administrator

    This is a spin off form another topic, but below is a really simple screen modulated design using a 4D32 which will produce a perfect AM modulation envelope with negative peak limiting. It only needs one plate transformer that yields +600V and one negative -100V bias supply. The 4D32's is one of those tubes which can handle a pretty high screen voltage and it makes a perfect tube for screen modulation. The below circuit would be really easy to do and the neat thing is no modulation transformer is required and it would sound just as good if not better than any plate modulation rig. The zeners keep everything stable and the modulator's load will be rock sold as a result unlike with a plate modulated rig that uses a transformer.

    What you do is apply a positive voltage with the AF drive to the grid of the 6L6 to control and adjust the voltage on the 4D32's screen. That could also be achieved from the 600V supply using another zener and a pot to ground with the wiper to control the voltage through a resistor or through the secondary of a small driver transformer. At this point I don't know what the exact voltage for the 6L6's control grid will be to place the 4D32's screen at the correct operating point and it would just need to be built and figured out from there.

    Anyway, this will be one of my upcoming projects because I want to build a compact transmitter for the higher HF bands. What I might do is use the Viking II for parts, mainly for the variables caps, roller inductor, knobs, dial bezel, etc. I could just use those components along with the 122 VFO components and build this simple transmitter with the VFO built in. Basically, I want something better than the Viking II with the 6146's and all of the problems it has. The below circuit would work much better and be way more reliable :icon_thumbup:

  2. W5HRO

    W5HRO Administrator

    After receiving a couple of PM's about this topic is a more detailed PRELIMINARY circuit and I stress the word PRELIMINARY!

    This is my first idea, but the basic concept is there. Using only two supplies, a -100V supply which is always on then the +600V supply which gets turned on during transmit with PTT control. The key is using the zeners and the electrolytic caps. This would be about one of the simplest and most stable reliable designs ever. The other key thing is finding about a 300mA to 350mA transformer that yields the +600V with a choke input.

  3. KM1H

    KM1H Guest

    I would think a slightly modified DX-60 circuit would do fine as similar voltages are used.

  4. W5HRO

    W5HRO Administrator

    My circuit is sort of a cross-breed between Steve's DX 60 modification and the circuit found in the old 1959 radio handbook.

    Anyway, the issue might be finding the 3:1 step-up transformer. In my case I'm probably going to use two 4D32's in push-pull with two 6L6 cathode followers and the zeners so I can eliminate the need for any audio coupling transformer and get more power out. I have the guts from a Johnson 250W matchbox to use for the RF push-pull circuit application.

    I think in Brian's case he has a Viking II, but was maybe considering using it for parts because of the condition. He would have most everything to make the screen modulated 4D32 rig, except for the 4D32 and 6L6. There's probably a way to get around using the 3:1 transformer, but maybe Hammond makes one or an existing step-down transformer can used and turned around backwards. He could probably use the one Viking II plate transformer with a 4D32, 6L6, and the two 6AU6's.

    Link to other topic:,56.msg134.html#msg134

  5. BBurns

    BBurns Member

    Good morning Brian,

    I did a bit of googling about for screen modulation, and came across this simple comparison with plate modulation.;wap2

    The basic point being that screen modulation is simpler, but has an efficiency of around 30% compared to 75% for plate modulation. Are you figuring that lower power to be adequate for a transmitter for the higher bands?

    And as for the Viking II, are you giving up on getting it back in operation?


    Brian B.
  6. W5HRO

    W5HRO Administrator

    The 30% efficiency info for screen modulation is complete BS. If done right you can easily exceed 100% positive peaks, but you have to limit the negative peak to get it there which is always the key anyway plus the carrier output level will need to be a little lower compared to plate modulation.

    I'm probably going to use two 4D32's in push-pull then screen modulate that setup to get enough output power. As far as the Viking II is concerned, it still works fine, but I never did finish that last grid bias supply issue to make it grid leak instead.

    Anyway, on the higher HF bands you never need more than about 200W max and anything higher is just wasted overkill. However, the issue for you might be the power output if only using a single 4D32 in a screen modulated setup because its going to be less than 100W out to get 100% or higher modulation. You would probably want to run it into a linear to get the power up. That's why if I decide to do it I'm going to use two tubes in push-pull instead of only one.
  7. KM1H

    KM1H Guest

    Simply put there is a difference between screen modulation of the 30's and controlled carrier screen modulation of the 50's and later.

    Final output efficiency IS around 30-35% for straight screen modulation and positive peaks limited to 90-95%

    With controlled carrier overall RF efficiency is on the order of 50-60% and positive peaks can go well above 100%. This is also a great way of driving a linear with a wimpy PS such as the SB series of Heathdrips.

    I can drive a SB-220 close to 1300W PEP with a modifed DX-60 running thru an attenuator as all it needs is 20-25W.

  8. W5HRO

    W5HRO Administrator

    You can do it without the typical controlled carrier that flops around and still get 50% to 60% efficiency with greater than 100% positive modulation and that's the point of this whole topic :icon_thumbup:

    You have to set it up with the zeners to keep the carrier stable then drive a linear or just use two tubes in push-pull to make up for the decrease in power. You can probably get around 50W to a little more though using a single 4D32 and then 100W or greater using two in push-pull.
  9. KM1H

    KM1H Guest


    Well, Im talking regular ole store bought linears here. The amount of carrier without audio is controllable but too much is too much for some amps and keeping it low gives the deep Class B modified ones a bit more overhead.
    Ive no problem with a widely variable carrier as long as the modulation is robust and low distortion which is not the norm with all the as designed commercial rigs Ive listened to. OTOH there are several on the bands Ive heard with the necessary mods that sound excellent and they are from all over the US.

  10. W5HRO

    W5HRO Administrator

    The whole issue I think is being able to build a decent AM transmitter without a modulation transformer since they are not just laying around anymore. You can find them, but they are usually very expensive and often need to be custom made.

    The answer to the problem is screen modulation using cathode followers. If done right it can actually work and sound better than plate modulation. My circuit in this topic can also be built using other tubes instead of a 4D32 for more power out and the 6L6 will still work perfect for the cathode follower. The 4D32 is a pretty good choice though.
  11. BBurns

    BBurns Member

    Hello Brian and Carl,

    Quite by coincidence there is now a thread going, on screen modulation on the Glowbugs list. You may think that I'm spending too much time on ham forums, but you're right (;->)...Anyway Al, AE6CM, just listed a bunch of references on the topic, and I thought that there might be something of interest here:

    "The problem with using power pentodes in high-mu configuration (drive
    applied to screens rather than to G1's) is that much more drive power is
    required than with a conventional medium-mu configuration. That tends to
    offset the circuit's advantages.

    Probably more than you want to know:

    Hanchett, Zero-Bias Sweep-Tube Modulators, QST Feb 1962, 6GW6s or 6GJ5's, G1
    tied to cathodes, cathode follower drive to screens.

    Jones, Plate Modulated 60 Watt Transmitter, CQ Jan 1947, 6F6's, G1 tied to

    Jones, VHF for the Radio Amateur, 1961, Three modulators with G1 tied to
    screen through current-limiting resistors -- drive applied to screens.
    6AV5, 6W6, 6V6

    Brown, A High Power Mobile Modulator, CQ Feb 1950, also in CQ "Command Sets"
    1957, also "The Command Phone", CQ New Mobile Handbook 1956. RCA Siebold
    design but with cathode follower driver

    Brown, A High-Power Modulator for Mobile Operation, CQ Feb 1950 -- same as

    Orr, The Command Phone, CQ New Mobile Handbook 1956, Seybold design but with
    cathode follower drive.

    Levy, Class B Modulators, 73 Sep 1961, 807s -- similar to "Ham Tips" design

    McCoy, Plate Modulation for the 150-Watter, QST Jul 1962, 6GJ5's, G1 tied
    to cathodes, knock-off of Hanchett?

    Seybold, New Modulator Circuit Utilizes 807's in Class B with Zero Bias, RCA
    Ham Tips Jun 1947 -- G1 to screen through resistors, 10 watts to drive
    screens, article describes method of choosing resistor values. Schematic
    also in RCA TT-4.

    Stock Gonsett G-76 uses 6DQ6's with G1 tied to cathodes.

    Early RCA tube manuals: 46, 49, and 59 tubes were specifically designed so
    they could be used as class A or B power amplifiers depending on how their
    grid connections were arranged.

    73, Al


    Brian B.
  12. W5HRO

    W5HRO Administrator

    I wouldn't place too much weight on that other board's material or information. Those articles and circuits won't provide the type of improvements you can make today.

    Limit the negative peak to around 90% and use the zeners to make everything stable including the carrier. You will be surprised what you can actually make screen modulation do when you do it right.
  13. BBurns

    BBurns Member

    Hello Brian,

    I should learn to look before I leap. I went to a couple of the referenced sites, and they all seem to be about plate modulation.

    So, taking your advice about the 300 watts output power necessary to do Ok with AM on the lower bands, and wanting to build only one AM transmitter, what tubes would you use for that scheme?


  14. W5HRO

    W5HRO Administrator

    For that power level then either plate modulation using a modulation transformer or screen modulation driving a linear.

    What you could do it use the 4D32 in the screen modulated setup and use that to drive a bigger linear tube all mounted within the same cabinet or rack. If the 4D32's power is too hot then a 6146 would drop it down to about half. Maybe do the 6146 screen modulation driving a single 3-500Z in grounded grid or even the 4D32 driving the bigger 3-1000Z in grounded grid for super screen scream linear modulation :surprised: :eek: :icon_thumbup:

    Attached Files:

  15. W5HRO

    W5HRO Administrator

    Actually, after thinking about it some more the 3-1000Z would be the better choice because if using a 3-500ZA instead it will still need around +2500V on the plate to yield 300W (+/-) of AM carrier output power with around 600W (+/-) on modulation envelope peaks.

    It may be bigger than what you were thinking of using, but a 3-1000Z at +2500V in grounded grid being driven by a screen modulated 4D32 would work great and you can make the 4D32 modulate 125% to 150% on the positive side easy. Running that into a 3-1000Z linear tube would be awesome :icon_thumbup: Just find a 1A plate transformer.

  16. BBurns

    BBurns Member

    Hi Brian,

    I just pinged Wayne about the old diathermy plate transformer, and he is going to bring it and some other likely stuff for an amp in the fall when he and his wife come out to Fort Bragg. The diathermy transformer is huge, but I don't know how high in voltage it goes. The tubes that it was powering were a pair of triodes about 10 inches tall and 2-1/2 inches in diameter.


    Brian B.
  17. W5HRO

    W5HRO Administrator

    If it's anywhere between 2500V to 4000V you are good to go :icon_thumbup:

    Another alternative would be to just screen modulate a 4-1000A, but it would be nice to use the 4D32 for lower power op then switch the 3-1000Z inline for high power op.
  18. KM1H

    KM1H Guest

    The 3-1000Z is becoming very scarce and expensive and even those using it in a commercial amp are replacing with something else.

    One of those Russian metal/ceramic tubes with a 2500W plate dissipation would make an excellent choice and at say a 400W carrier would require less air (noise) than if run at full power.

    A triode can be GG and a tetrode swamped grid to burn some power. A Viking I or II would still have excessive drive and need an attenuator before the amp. Reducing drive or loading would not be a good idea for a plate modulated rig. I even needed to swamp the input a bit on the pair of Class C plate modulated 250TH's with my V I.

    The LK-500ZC linear with a pair of 3-500Z's requires only 20-25W of AM drive for a 350W carrier, the 3-1000Z is also a high gain tube. A Viking Ranger, AF-67/68 or similar would be more than sufficient as a driver.

  19. W5HRO

    W5HRO Administrator

    That is true, you can still find them new in the box, but they run between $400 to $500 :eek:

    You can still get 4-1000A's really cheap for $100 and less so that might be the better choice. You could use it as an amp tube or just screen modulate it. The 6L6 cathode follower would need to be bumped up to around 600V on its plate which is fine and it will still handle it in the configuration that it is in with the 100K resistor. I remember Bacon once had a screen modulated 4-1000A rig.
  20. W5HRO

    W5HRO Administrator

    I found Bacon's 4-1000A rig.

    His notes are below...

    Here are the operating parameters as I run mine @ somewhere around 400w carrier output.

    Ep = 3450v
    Eg2 = 290v+
    Eg1 = -150v
    Ig1 = 35-40mA
    Ig2 = somewhere around 50mA at resting carrier
    Ik = somewhere around 400mA (and watch the munky swing with the audio!)

    Just a last note, the screen supply should be regulated and adjustable, and well "stiffened" !!


  21. KM1H

    KM1H Guest

    That looks like a recipe for a great oscillator
  22. W5HRO

    W5HRO Administrator

    :smile: What I'm not sure about is in his attached pic it says "don not use any grid leak resistor" "grid and screen must be in the same phase" However, he was using a small modulation transformer to modulate both the control grid and the screen. Using a 6L6 cathode follower at 600V to modulate the screen would work better than that anyway, but I need to research about the control grid and phase issue. It may have just been because of the way he was doing it, but then again maybe not. My 4D32 circuit has grid leak.
  23. W5HRO

    W5HRO Administrator

    I was working on another project using this same screen modulation technique and attached below is an updated and improved design using the 4D32. I figured out an easy way to eliminate the interstage transformer. This would work fantastic and it also includes an adjustable negative peak limiter. Most of the component values have also been added and are correct. Still only 4 tubes :icon_thumbup:

  24. W5HRO

    W5HRO Administrator

    Here's a final update below which solves all of the issues.

  25. W8EJO

    W8EJO Member

    Did you ever buil
    Did you ever build this transmitter? What was final config. How did it work?