Toroids as output transformers.

[boater sam]

Just a brainstorm idea, but has anyone tried using toroidal mains transformers as valve output transformers?

Occurred to me  that as the secondary is usually  the outer winding it would be an easy job to match the turns ratio to the speaker.

Sam.

[ppppenguin]

I haven't done it myself but the received wisdom is that it can work. In a push-pull circuit there is no DC flowing so a mains transformer can work well. In a single ended circuit you need to de-rate substantially to avoid saturation.

I'm a afraid to say it's trial and error.

[Mark Hennessy]

Velleman did it many years ago:

http://www.velleman.co.uk/contents/en-uk/p230.html

Can't remember for sure, but I'm thinking early '90s perhaps...

[boater sam]

And I thought I was being original.
Surprised that Velleman mounted them all the same way round, perhaps the magnetic coupling is less.
May try it next time I need an output tranny.

[pamphonica]

The magnetic flux is contained fully within the core in a toroidal transformer so there's no coupling between adjacent toroids. Co-planar mounting would be fine.
Jeremy

[Nick]

This has been done for years. Remember that mains transformers are designed for no DC and to operate at 50/60Hz - for guitar use its an acceptable approach, but not for any sort of quality audio - they are poor.

Having said that, there are manufacturers who produce toroidal transformers specifically for audio use - Amplimo springs to mind in Europe - http://www.amplimo.nl/en/ - their own website is pretty rubbish on their own product range, but you can find them at http://www.toroidal-transformer.com/shop...rmers.html

The "VDV" range are designed by Menno van der Veen - a well-known valve audio designer and proponent of toroidal output transformers... his book "Modern High-End Valve Amplifiers (based on toroidal output transformers)" is published by Elektor (ISBN 0-905705-63-7) in various languages - I have a copy of the English version and its a pretty good read.

[ppppenguin]

Very much somethng to try if you're stuck for a replacement output transformer in a radio etc. Then if the sound is acceptable you're OK. For a high quality amplifier you'd need to do proper measurements, not the lash-up I'm about to suggest.

No reason in theory why you can't measure a transformer. A low amplitude frequency response test is easy. So is adding some DC. A full power test is a bit harder as you'd need a signal generator with a very high output voltage. With care you could measure in reverse, using a modern amplifier to drive the low voltage winding, loading the mains winding with Ra of your proposed output valve. Put the DC in the mains side, it'll be easier.

With a scope you'll only see gross distortion but there's always the old tricks of winding up the amplitude and/or the DC until you can see the waveform distorting.

[tiv48]

Hifi world also produced a design using toroids which was sold as a kit .
But i notice in later issues the kit was sold with std transformers .
Rob T

[Kalee20]

Toroidal transformers tend to saturate more easily with DC flowing than conventional types. That's because the geometry of the core is such that the magnetic field is always directed along the core's grain structure. Permeability (and inductance) is highest, but the down side is that DC has most effect.

So, for a single-ended stage it's a no-no. Although, if you try it, you will get something!

As a push-pull stage, however, as long as balance is good, there should be decent performance to be obtained. The killers in output transformers are primary capacitance, and leakage inductance (hurting high-frequency response); and primary inductance (hurting low-frequency response).

If you have a 0-120 / 0-120V dual primary, you effectively have a centre-tapped primary, and you can strip off the secondary and rewind to suit your loudspeaker. Being designed to operate at 120V each half secondary, at 50Hz, you should be OK with bass response down to 74Hz with 250V HT before saturation-induced distortion sets in. (Calculation: 250V HT gives at most 250V peak, or 177V RMS drive to the transformer; and if the transformer does not sturate with 120V RMS at 50Hz it will equally be happy with 177V at 74Hz). At the HF end, the primary self-capacitance may limit response, if the winding extends over more than one layer (modern enamel will easily withstand 120V AC of overlapped ends). And leakage inductance will not be mega low, as there is no sectioning of primary and secondary, though it won't be unduly high either. So treble response ought to be at least reasonable!

One tip - if rewinding the secondary to suit your loudspeaker, ensure the winding i9e evenly spread around the toroid circumference. Bunching turns in one place hurts leakage inductance, and also spoild the zero-external-field property of an ideal toroidal transformer.