ad: Mircules-1

Mains Transformers as Modulation Transformers

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

ad: L-HROutlet
ad: l-rl
ad: Left-3
ad: Left-2
ad: L-MFJ
ad: abrind-2
ad: Radclub22-2
ad: l-BCInc
  1. G3YRO

    G3YRO Ham Member QRZ Page

    I haven't built an AM Transmitter for over 40 years . . .

    But reading comments on this section about the frequency response of different Modulation Transformers reminded me that back in the day I only ever used various Mains Transformers . . . yet those rigs always seemed to have good audio responses, both at the low and high end.

    I actually built an AM Broadcast Transmitter using one, which ran about 200 watts DC input, and the audio quality of the music was pretty good. (this had a QY4-400 in Class C, modulated by a pair of EL34s in push-pull)

    I also built a 100 watt audio amp which had a mains transformer in the output with a hefty 6.3V winding, so used that to drive the loudspeaker . . . again, it sounded pretty good !

    And I built several small stereo audio amps, single-ended Class A producing about 4 watts (using ECL86s), again using small Mains Transformers as the output transformers, which all sounded good.

    Roger G3YRO
    WA3JVJ likes this.
  2. W2VW

    W2VW Ham Member QRZ Page

    It depends of course on what the transformer is made from and how it's wound.

    The talented people I learned from used to insist on a frequency response sweep and a measurement of peak modulation capability as a starting point for an AM tx. There was a time when I thought they were just being difficult.

    Here's a link to a book which should address your post using newer components:

    Additionally, I know people who have used very old plate transformers as modulation transformers in autotransformer configuration.

    I have toroidal mains transformers used as modulation transformers on my main AM setups running legal limit.

    They work very well.

    I swapped one out with a modern 4.5kva mains distribution transformer and the high end was so bad I didn't bother measuring it.

    In 1995 I built a modulator with a pair of cathode driven 4-1000s as modulators. It worked almost as well as the later pair of 833As I switched to when I realized all that plate dissipation was not doing anything for me on the bands. The 4-1000 driver transformer was a 1:3 turns ratio low voltage center tapped mains type transformer. Drove it with a large solid state audio power amplifier. The 833s were driven by UTC CG output pp-parallel 6l6 transformer turned around backwards.

    I suppose a single ended class A amplifier would require quite an oversizing of output transformer to handle the unbalanced DC.

    There are a lot of fine amateurs on AM over there. They frequent the Hackgreen SDR and meet on 3612? khz.
  3. WB2GCR

    WB2GCR Ham Member QRZ Page

    Sounding "pretty good" and actually being "hifi" are not the same things.

    The problem with mains transformers is that due to their winding methods,
    simpler than audio transformers, they typically end up with a rolloff ~5kHz.
    Some may go a bit higher.

    But, two things work in one's favor in ham radio. One is that output much above
    5kHz isn't terribly valuable. Two, many people actually can't hear above the 5-10kHz
    range anyhow.

    So, I have used a mains to 6.3vac transformer as a 600ohm to "8 ohm" matching transformer
    out of the back of several different receivers. It sound fine. But, again, response above
    5-10kHz is not needed, and really not desirable.

    So, for ham use it may be ok in some situations.
    N2EY and AG5DB like this.
  4. SM0AOM

    SM0AOM Ham Member QRZ Page

    What has been mostly overlooked are the core DC component saturation aspects.

    In a push-pull class B circuit, the DC currents flowing to each modulator valve (tube),in the primary cancel but the anode (plate) current flowing in the secondary of a modulation transformer does not.

    There is a net DC magnetisation in the core, which may or may not lead to saturation and distorsion if the core area is too small.

    Also, as mains power transformers are not designed for frequencies much above 100 Hz, core losses may influence the results. On the plus side, as the lower frequency limit of "communications audio" is in the order of 200 or 300 Hz, the core area in especially 50 Hz transformers may permit some DC flux without peak saturation, as the flux density increases with decreasing frequency.

    Most transformers are also designed with some margins in these respects.

    AG5DB and KA9JLM like this.
  5. K4KYV

    K4KYV Premium Subscriber Volunteer Moderator QRZ Page

    Shunt feed (modulation reactor with blocking capacitor) will prevent core saturation. Many better quality modulation transformers are not designed to carry the dc to the final and require shunt feed. Several power supply filter chokes wired in series can make a satisfactory modulation reactor.
    N2EY likes this.
  6. W1BR

    W1BR Ham Member QRZ Page

    I'd wonder about using the CT high voltage secondary winding in a power transformer. SE
    class A modulator on one end, PA tube on the other end. B plus on CT. Wouldn't DC magnetic
    field cancel??
  7. K4KYV

    K4KYV Premium Subscriber Volunteer Moderator QRZ Page

    If the DC plate currents were adjusted to be equal. A modification of the "constant current" Heising circuit. It should work.

    Back in the mid 1930s UTC made a balanced mid-tapped audio choke just for that purpose.
  8. W2VW

    W2VW Ham Member QRZ Page

    Look up the schematic of almost any tube type CB radio from the mid sixties on. Most use one center tapped winding as you describe.

    Not sure about the modulator's conduction angle.
    W2BTK likes this.
  9. W1BR

    W1BR Ham Member QRZ Page

    I think those were generally tapped at 5k and 7k, basically Heising modulation. There would be no DC magnetic field cancellation since the B plus is feed at the end of the winding, not the
  10. W2VW

    W2VW Ham Member QRZ Page

    You're right. They aren't fed in the center and the center may not be the electrical center.
  11. K4KYV

    K4KYV Premium Subscriber Volunteer Moderator QRZ Page

    If the B+ is fed at the bottom end of the winding, the plate of the modulator tube at the lowest tap and the final at the highest tap, that will allow 100% modulation. Otherwise, with just a choke with no taps, the modulation will be limited to around 75-80%, unless a dropping resistor by-passed with a capacitor is inserted in the B+ line to the final. The extra higher taps will allow 100% modulation, but there will be no DC magnetic field cancellation.

    The mid-tapped choke or transformer winding can be used with a push-pull modulator and the +HV fed to the mid-tap, but that presents a 2:1 step-down turns ratio, which will not allow 100% modulation. "Modulation autotransformers" are designed to have some extra turns added to one end to present a normal modulation transformer step-down ratio. Several commercial rigs used those.

    Timtron says he has had good results by "turbo-connecting" the power transformer, so that the mid-tapped HV winding is used with the push-pull modulator, and the mains voltage primary winding is added in series at one end of the mid-tapped winding to give less step-down. Whether or not this works depends on the overall turns ratio, and whether or not the primary winding is insulated well enough to withstand the full modulated B+ voltage. This may work OK at the 100-watt class, but I think would be risky at higher power with the B+ in the thousands of volts. Plus, the amount of step-up achievable with the primary would likely not be enough for 100% modulation capability.
  12. W1BR

    W1BR Ham Member QRZ Page

    Yes, I had considered needing use the primary to increase turns, but that would also add to the DC biasing since windings would no longer be equal.
  13. K4KYV

    K4KYV Premium Subscriber Volunteer Moderator QRZ Page

    That's where shunt feed (blocking capacitor and modulation reactor) would help.
  14. W1BR

    W1BR Ham Member QRZ Page

    Unbalanced DC bias would be minimal using the primary winding to aid one side of the CT secondary, but adding a shunt feed on the PA side of the CT would completely mess it up.
    All the DC current would be on the SE modulator half of the winding with no field cancellation on the PA half. I was just thinking that an old
    1400V plate transformer might make for good cheap mod iron, even if the ratio doesn't allow for full 100% modulation depth.

    Last edited: Jun 9, 2020
  15. N2DTS

    N2DTS Ham Member QRZ Page

    Why not just use a mod transformer?

Share This Page