Mains Transformers as Modulation Transformers

Discussion in 'Amplitude Modulation' started by G3YRO, Jun 7, 2020.

ad: L-HROutlet
ad: l-rl
ad: Left-2
ad: L-Geochron
ad: HRDLLC-2
ad: MessiPaoloni-1
ad: L-MFJ
ad: abrind-2
ad: Left-3
  1. WD5JKO

    WD5JKO Ham Member QRZ Page

    Sure....here is the full text. Also added the rest of the schematic.

    "I decided to go a little further with my G-50, so I replaced the single ended modulation transformer with a push pull transformer. This is still single ended class A Heising. This made a huge difference. From the Mod adj pot onward, the modulator is flat from 50-5000 Hz with little distortion up to 85% (sine wave), and capable of 100% positive peaks on voice. See scope plot of YEA_AM.jpg where it hits +100% modulation.

    I used the P-P transformer in a unusual way where the RF class C plate current (120ma), and the TX Modulator current (120ma) balance out the DC modulation transformer magnetic field (120 - 120 = 0) such that I get around the saturation issue common with Heising modulation. This allows the use of full transformer lamination interleaving of the E & I lamination pieces such that the primary inductance is at maximum. The way it was, the currents were additive, and the E & I laminations were grouped together and the core gaped. Getting good modulation below 300 hz at full depth was a big problem with the stock iron.

    I added a parallel RC in the modulated B+ line such that it drops the B+ about 50 volts to the 6146, but the capacitor bypasses the resistor for AC (mid range and up) such that instantaneous voice peaks at + 100% modulation are possible. The reduced B+ is offset by solid stating the 5U4G, so the Modulated B+ cut and Power supply B+ gain pretty much cancel each other out.

    With sine wave drive, the cathode bias rises as you approach 100% modulation, and the screen voltage goes down. This makes the modulator with over-driven voice audio somewhat limiting, or compressed as in the Gonset G50 tradition, yet with these modifications, the audio is very clean with voice modulation up to 100% (+) modulation. Another part of this is where I limited the (-) downward modulation with a diode clamp on the 6146 G2 grid such that the G2 over the audio swing cannot dip below about +30 volts. This makes the rig incapable of over modulating downward and splattering. The +30v is from the modulator cathode bias. If I tapped down the Rk resistor, then it would be possible to Taylor the maximum (-) modulation percentage to to 90 or 95%. What was unexpected is that a hard clip to the screen voltage minimum only causes a soft clip to the modulated envelope.

    In order to get linear plate + screen modulation out of the 6146 over the audio range, I had to do a lot of work. The work included increasing the grid1 leak resistor, increasing the G2 dropping resistor, and making a capacitive AC voltage divider for G2 to set the % of G2 modulation to a value less than before, and a value that was more constant over the audio bandwidth. This is an area for improvement in many HF plate modulated rigs using 6146 RF tubes. I was modulating with triangle waves to look for RF linearity; alternatively I could have configured scope in X:Y mode for trapezoid waveform.

    The frequency response is Tailored in the first audio stage (12AT7) to suit your voice, microphone, and individual preferences. I increased R17 from 270K to 1 M to boost the base a bit (D-104), and I added a R-C preemphasis network going to the Mod adj pot too boost the midrange in the hope of boosting weak signal intelligibility. Depending on your voice, and chosen microphone, preemphasis can be a very useful tool. Do you really want to sound like a "Woofer" on the receive end?

    Notice I did change out the 12AX7 for a 12AT7. I cannot recall why, but The 12AT7 does a great job with plenty of reserve gain when using a Crystal Mic D-104..

    Let me know what you think..."

    Jim
    Wd5JKO
     

    Attached Files:

    W1BR and SM0GLD like this.
  2. SM0GLD

    SM0GLD Ham Member QRZ Page

    Jim, thank you for the explanation.
    I have been looking at this schematic for a some time now and like to implement the design in an upcoming qrp-tx for 160m.
     
  3. WA3JVJ

    WA3JVJ Ham Member QRZ Page


    Hi Roger:



    Good post, I am also interested in using power transformers in audio applications. More specifically I have done some experimentation using AnTek torrid power transformers as modulation transformers, output transformers, and as driver transformers with excellent results.

    Your comment with respect to using EL-34’s is interesting. I have done an experiment taking a tube stereo amplifier, using EL-34’s in each channel, and substituting an AnTek AS-1206 for a standard Hi-Fi output transformer in one channel; leaving the other channel unchanged. The sound was unexpectedly excellent, with a good low and high end. The 115 vac windings were in “series”, (the center connection went to the B+), with the 6 vac windings connected in parallel. Using my 4 ohm AR3a’s the plate to plate impedance was about 5900 ohms; not a bad match for EL-34’s! Output power was equal to the standard Hammond output transformer in the other channel.

    I have also used a torrid power transformer to drive my push-pull-parallel 805 modulator; again excellent results. Good frequency response with low distortion.

    When it comes to using torrid power transformers as modulation transformers things get really interesting. Brett, N2DTS, gave me some AnTek AN-4T800 power transformers to experiment with. They were hooked up in a balanced configuration, (see fig 3 using push-pull-parallel 805’s at 1275 vdc on the plates), modulating my Globe King 400B. I successfully used this modulator for a few days and one of my life long dreams came true ….. one horsepower of audio. Of course there was a flash over and I had to back thing down, and reinforce the GK RF deck.

    The end result is that torrid power transformers can be successfully used in a variety of audio applications; one must be careful of voltage excursions, turns ratios and core considerations.


    73,
    Bruce – WA3JVJ
     

    Attached Files:

    W3SLK likes this.
  4. W3SLK

    W3SLK Ham Member QRZ Page

    Lamont, (ex)KK4BO, told me that you should use triodes for audio and tetrodes for RF. I suspect you could use the torroids for winding a choke for the screens on tetrodes. I always wanted to make a rig with 4CX250B's but I understand they don't play well unless you modulate the screen. Chokes are available I guess, but I never found any suitable. I wouldn't mind getting into winding some transformers to keep my current gear running.
     
  5. G3YRO

    G3YRO Ham Member QRZ Page

    Intersting that you found Mains Transformers sounded as good as Audio output transformers too Bruce . . . although I have never tried Torroidal ones.

    (By the way, Torrid means "rather difficult" . . . or "hot & dry" . . . I think you mean Torroid LOL!)

    Roger G3YRO
     
  6. W2VW

    W2VW Ham Member QRZ Page

  7. K5UJ

    K5UJ Ham Member QRZ Page

    The higher end and pro audio gear that came out in the 1980s started using toroidal power supply transformers. CRL used them although it might have been mostly because a flatter toroid could fit in a 1 RU cabinet.
     

Share This Page

ad: elecraft