[Roger A. Modjeski]

I often get asked about Toroid output transformers, which make no sense at all due to the fact that a very small DC imbalance in output tube current will saturate them. I design and wind my own output transformers using EI lams and copper wire. Below is a quote from Tim de Paravicini, with whom I completely agree in this and many other things. He's one of the few designers I respect in this field.

Tim de Paravicini is blunt about it: output transformers determine the
performance of an amplifier. He describes himself as ‘very anti-toroidal’, and
goes so far as to judge amplifiers by their weight – a clue to the mass of their
transformers. ‘I lift these lightweight amps from China with 845s and 2A3s and
I know they’ll have no bass. There are no short-cuts. If those amplifiers don’t
weigh 70lbs, they aren’t gonna work.’ It explains not just the heft of EAR-Yoshino
models, but also the amps he conceived for Quad. Then ask Tim to describe his
hand-wound, in-house transformers and he sounds like a chef asked to name the
spices in his secret brisket dry-rub: ‘Basic EI construction, with special layering
and winding, with copper wire – not silver. The layering and how you arrange the
windings are the key. Most manufacturers just don’t get it.’

[Ericus Rex]

Are toroids acceptable as power transformers?

[Clio09]

I have been watching Jack Elliano of Electra Print wind transformers for nearly five years now. He has taught me many things regarding the importance of output transformers. Tim de Pavaracini is spot on in his comments and I suspect Jack would agree as well. We recently sold an amp to a local customer. It weighed in excess of 75 lbs. mainly due to the output transformers.

[JoshK]

I have seen some really good empirical data with use of the big 400w Plitron Toroid OPT from George of Tubelab.com.    However, George isn't a novice and takes cares of the details.  I'd imagine balance in a toroid OPT is uber critical.   

In terms of iron and weight elsewhere, wholly agree.

[Roger A. Modjeski]

Are toroids acceptable as power transformers?

They appear to be though I have not employed them as yet. One horribly overlooked aspect is the fact that they have more coupling to the power line (less isolation) than EI transformers. This coupling also extends to higher frequencies. On one end of the spectrum is the "ultra isolation transformer" where the windings have a great degree of physical separation (half an inch or more) and on the other is the toroid. Toroids have such good coupling that the DIY electrostatic builders are using them for the audio step-up. If one measures the 3 dB down coupling, an ultra isolation transformer goes out to just a few hundred cycles. A split bobbin to about 400 Hz and a standard concentric wound (secondary over primary) goes out to about 2 KHz. This effects of this coupling is to conduct line noise into the ground circuit of an amplifier. A torroid can go out to 10 KHz. The coupling is a combination of both capacitive and leakage inductance and its effects on individual systems is difficult to predict.

I employ concentric wound transformers in all of my designs so far and currently we wind many of them here along with all of our output transformers. I feel these power transformers have the best balance of characteristics and to my knowledge line induced noise has never been a problem in any of my designs.

This is a very short presentation of a very deep topic that has held my interest for almost 30 years. The books only got me started, the rest has been learned the hard way.

[Roger A. Modjeski]

I have seen some really good empirical data with use of the big 400w Plitron Toroid OPT from George of Tubelab.com.    However, George isn't a novice and takes cares of the details.  I'd imagine balance in a toroid OPT is uber critical.   

In terms of iron and weight elsewhere, wholly agree.

I looked at his web site and saw only Single Ended designs. Where is he using the toroids?

[pehare]

sure glad my RM10 MKII doesn't weigh 70lbs

cheers, patrick

[Roger A. Modjeski]

In only one thing do I disagree with Tim. Weight is not necessarily a measure of a good transformer. As I have stated here before, if one can make an output transformer smaller and still achieve the same low end then I believe the smaller unit will sound better as it is "carrying along" less iron. No matter how good the iron is it still has to be magnetized up and down with the music (mainly the lowest bass). Each type of iron has a defined loss and magnetizing current per pound. More pounds, more loss.

However, I do believe that Tim has seen some very bad Chinese amps as have I. They are likely to make a transformer small to save money and unlikely to understand how to make that small transformer good. They just make it small.

Output transformer design is a balance between iron loss, copper loss, magnetic saturation, winding geometry and care in winding. Certain things affect the top end only and certain things the bottom end. Torroids as output transformers need to have a gap and perhaps some do. I have read the papers published by Plitron and disagree with some of their theories but best of luck to them and those who want to fight the problem of plate current imbalance. Even complex bias circuits will not hold balance under dynamic conditions. The cons here outweigh the pros.

[Roger A. Modjeski]

sure glad my RM10 MKII doesn't weigh 70lbs

Thanks Peter, I hear that heavy amps are loosing their glory by owners who are tired of dealing with the problems and costs of moving, shipping and having these things. Shipping rates are continually increasing and heavy amps are both difficult to pack safely and deal with. Who needs hundreds of pounds of amplifier to listen to music. When the 99% of tube amp buyers finally get done with judging amps by the pound I will make the lightest amps around, perhaps I already do but I could go farther.

I amp pleased to note that the RM-10 is likely the only 70 watt amp that can be picked up with one hand by grasping the transformer cover. This is how I generally carry one. Makes it much easier to open the door with my free hand. I could make it lighter still but I am waiting for the 99% to come around.

I am currently building a small run of RM-10 amps what will be even lighter than the current one. The power is slightly reduced to 25 watts per channel from 35. The weight by several pounds. It's main virtue however is making it friendly to the user who does not want to be bothered with bias adjustments and meters. In this amp each output tube has an LED to monitor its tube current and one can see at a glance if all tubes are operating properly. Unlike many self biased amps the tube matching is not critical as it is in the Dynaco ST 35 and other designs that use a common cathode resistor for all four output tubes. This point is largely overlooked or unknown by most Dynaco and owners of current designs that use a common cathode resistor. I will be happy to expand on this common fault if desired.

[JoshK]

I looked at his web site and saw only Single Ended designs. Where is he using the toroids?

Yeah, he has a lot of SE designs on his site.  But he is an avid experimenter/scientist and most of his data in on diyaudio in various threads.  I don't think he ever used those OPTs in a finished amp as much as he uses them in breadboard experiments.

He has done a number of PP amps with those in breadboard.   That is what I was referring to.

As a hobbiest/diy'er, I wouldn't pick a toroid OPT for similar reasons as you.  I'd probably drive myself insane.