• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Musings on the design of “the big one(s)”

My experience is quite the opposite. I've been playing with two different pairs of mains toroids the last couple of years, 50VA and 400-500VA, and both had really decent highs (35k and 50k respectively) but could only handle 1/10 of their VA rating at 30Hz. These transformers where discussed in a thread here and the conclusion was that the LF power rating is severely limited by the low number of turns in the primary windings, while the HF probably depends mostly on winding geometry.
 
In order to get full power response to 30 Hz with a power toroid, you have to limit the applied voltage to half the design value. They’re designed to have a “reasonable” magnetization current (proportional to 1/L) at 60 Hz. Half the frequency, half the voltage for the same flux. Applying more than 240V RMS to a 240V primary is a recipe for lost low frequency performance. When applying 700 volts RMS to a winding rated for 1600 at 60 Hz, there ought to be bass response. Giving about 53 volts to the speaker, out of a winding rated for 120.

How good are the highs? Don’t know - will have to test. The big 800+ watt units tend to have fairly low leakage reactances. Regulation no load to full load is better than on the little ones, and the voltage drop even at 60 Hz is dominated more by reactance than resistance. Running at twice the normal turns per volt however, goes against this. So the highs will take a hit. If I would have gotten 20 KHz at design flux density, I’d only get 10. Good enough for a high power PA rig? Sure worth a try. I’ve had decent results using the little 50-100 VA units, when keeping the applied voltage down to around 240V. Bass tends to be “good enough”, and so are the highs even running open loop.
 
It will have some capacitance to ground. But that doesn’t dominate the HF roll off. The leakage reactance usually does. Self-resonance of the primary winding is also something to watch out for, as it can cause PEAKING (and a sharp roll off above it). The good news is that SRF tends to be higher in toroids (not hard and fast rule, but generally).

The shield’s main function is to prevent magnetic coupling to/from OUTSIDE - lowering the effective leakage reactance but adding a small LOAD to the transformer. It is a shorted turn around the outside, capturing nearly all the stray or asymmetrical flux which isn’t high to begin with. If the stray flux were high it would get HOT, and it doesn’t. Unless it’s shown to cause trouble, leave it there. If you want to go adding turns to the secondary, add feedback windings, or otherwise modify it of course it must be removed first.

Those flux bands do make the Antek toroids ugly because they are never quite round. If I want a given amplifier to win a beauty contest I’ll spend the money and buy EI’s. Got one (not this big) in the works and made the EI power trafo myself. Planning on painting it Edcor blue - and will use the 4.2k 60 watt
 
I have used power (mains) toroids for OPT's in lots of projects. Some work good, and some just suck, often at both ends of the spectrum. Any toroid will be much more sensitive to core saturation due to DC imbalance than a E/I stack of equal power capability. This is true even with my Plitron toroidal OPT's. As stated, you need a power toroid that's rated for at least 4 X the power (VA) capability to get to half its specified frequency rating (30 / 25 Hz). You must also not stuff more voltage into any winding than it was rated for. You can't expect to use a pair of 120 volt 50 / 60 Hz windings for the primary in a push pull amp that runs from more than 150 volts. I did use an Antek 50 VA toroid successfully in a 20 watt guitar amp that ran on 340 volts, but it never sees lower than 80 Hz. The same amp made a bit more power with a real OPT though.

You can wire two identical toroids in series for more voltage handling capability provided that you connect and phase them such that there are equal and opposing DC currents flowing in both transformers. I have even stacked two 120/120 to 120/120 volt isolation transformers on top of each other used all 4 windings and wound a common speaker secondary through both cores with success.

Many years ago I stacked 4 power toroids with one common secondary, drove them with a SPP board which was fed by a guitar preamp for a 50 watt guitar amp. I ran some different colored EL wires through the core with the secondary which were driven with a DIY controller. I called the monster glowing OPT the "Tubelab Sonic Reactor" but it suffered a "core meltdown" when the EL wires shorted out the mosfet based controller. I still have it all in a box somewhere, but it has been robbed of its working SPP boards and transformers.
 
I always keep an eye open for cheap mains toroids with dual primaries and multipe secondaries. Both pairs I've tested so far have extra windings that can be used for CFB, which also adds to the primary inductance/voltage handling, which of course is most welcome.
The smaller 50VA pair is now put to good use in a PL36 (25E5) PP amp that doesn't sound half bad , in fact it was one of my favourite amps until a couple of days ago when I finished my PP DHT amp with much, much more expensive iron.

I'm still trying to make up my mind about what to do with the bigger pair. They are 2k:8R with 24% CFB and good for maybe 30W @ 30Hz, so one pair of big or two pairs of smaller sweep tubes per channel in operating in class A would probably make sense.