Anyone built an SS version of McIntosh unity gain transformer coupled output stage?
I think its a brilliant solution and Im thinking of winding a transformer to try it.
Any thoughts, ideas or experience is appreciated!
I think its a brilliant solution and Im thinking of winding a transformer to try it.
Any thoughts, ideas or experience is appreciated!
The tube McIntosh designs used a unique output transformer that had extra taps on it for feedback to earlier stages. Their solid state designs don't use transformers in the usual sense; they're autoformers. To the best of my knowledge, no other company uses such a thing at this time.
Can an autoformer be done on a DIY basis. Absolutely. And that's pretty much the only way it'll happen. I'm not aware of any transformer manufacturer that offers such a thing in their standard product line. McIntosh used to (and I believe they still do) build their own.
Grey
Can an autoformer be done on a DIY basis. Absolutely. And that's pretty much the only way it'll happen. I'm not aware of any transformer manufacturer that offers such a thing in their standard product line. McIntosh used to (and I believe they still do) build their own.
Grey
I don't think that a solid state version would be useful. You must try to understand why the unity coupled circuit was invented in the first place.
GR, if I'm not mistaken, Atmasphere offers an autotransformer as an accessory. And as John intimated, notch distortion is not an issue.
Now that you mention it, I think I remember seeing something about that. I bet they roll their own, too.
Grey
Grey
Thanks for answer!
Im not intrested in the autotransformer, but in their original pentode with unitygain coupling and crosscoupled screengrids. When thinking about it it seems to work as a cascoded cathodefollower. The Cathode-Screengrid voltage being constant, and 100 % local feedback.
Im just fooling around trying to learn all about balanced circuits. I found this one very intresting, and I think its a worthwhile expirment translating it to solidstate.
Johannes
Im not intrested in the autotransformer, but in their original pentode with unitygain coupling and crosscoupled screengrids. When thinking about it it seems to work as a cascoded cathodefollower. The Cathode-Screengrid voltage being constant, and 100 % local feedback.
Im just fooling around trying to learn all about balanced circuits. I found this one very intresting, and I think its a worthwhile expirment translating it to solidstate.
Johannes
for tubes it makes a lot of sense, but for solid state, what for? speaker impedance levels are within reach of solid state state circuitry anyway...
The exact same reason im building amps in the first place: Its fun!
I know there is no real good reason to experiment with all kinds of strange circuits. Its better and much simpler buying a Krell or PassLabs X-350.
I like transformers in the output section of amps. They make the amps safe and easy to build. Dc offset is no big problem.No loudspeaker protection needed and they seem to sound very good. Se Susan Parkers Zeus!
I like the concept of connecting both source and drain to the output transformer, even though i cant realy explain why. Maybe its just cool.😉
However its a fun expirement and im rewinding a transformer at the moment to try it out.
Anyone having any thoughts or ideas are very welcome to tell me about them.
Most of this is unknown to me so Im in desperate need of information about the circuits and possible variations, pros and cons and so forth.
Johannes with fever and a bad cough.
🙁
I know there is no real good reason to experiment with all kinds of strange circuits. Its better and much simpler buying a Krell or PassLabs X-350.
I like transformers in the output section of amps. They make the amps safe and easy to build. Dc offset is no big problem.No loudspeaker protection needed and they seem to sound very good. Se Susan Parkers Zeus!
I like the concept of connecting both source and drain to the output transformer, even though i cant realy explain why. Maybe its just cool.😉
However its a fun expirement and im rewinding a transformer at the moment to try it out.
Anyone having any thoughts or ideas are very welcome to tell me about them.
Most of this is unknown to me so Im in desperate need of information about the circuits and possible variations, pros and cons and so forth.
Johannes with fever and a bad cough.
🙁
a valid reason i aggree, not that i need to. diy is all about fun!
btw, i made a desgn for output traffo for the tube mc270 amp for a friend but he insists to =see the original data, can you help? thanks
The info I have is in Ir.Menno van der Veens book, Modern highend valve amplifiers.
A very nice book by the way.
What does "btw" mean?
Johannes.
A very nice book by the way.
What does "btw" mean?
Johannes.
"By the way"
The main purpose of the unity coupling was to allow the amp to be run near class B without causing notch distortion. (NB: Notch distortion is NOT the same thing as crossover distortion!)
The main purpose of the unity coupling was to allow the amp to be run near class B without causing notch distortion. (NB: Notch distortion is NOT the same thing as crossover distortion!)
Resources for SS transformer design
I have an interest in transformers in SS design also but am having a difficult time finding any resources on the subject. Any suggested reading (beyond Susan's thread) would be appreciated.
mike
I have an interest in transformers in SS design also but am having a difficult time finding any resources on the subject. Any suggested reading (beyond Susan's thread) would be appreciated.
mike
Any and all litterature about tube-amps is good. Some candy at www.tubecad.com.
The above mentioned book is the best one so far for me. Old litterature about transformers in general, non-audio related.
Start experimenting. Actually building the transformer, or rewinding one as you see fit, is a very good way to learn. So far I have winded two transformers, with normal plastic isolated wire, and the work very good. Dont ever be too afraid to try something out just because such a simple thing *** people saying it cant be done, or it wont be perfect.
I would rather bend some cheap ironbars into a core and wind them with some thin skitty wire than not build tha transformer at all. Who knows? It might sound wonderfull despite disapprovement from authorities and "experts".
Johannes ,Having problems with authorities!😀
I have just winded my bobin with 32 turns quadfilare primary and secondary windings,and 16 turns quadfilare windings for sourcefeedback. I know its way to few turns but I only had thick plastic isolated wire at home and there is not enough room in my EI-150 lamintionstack for any more. This is a test to se what happens.
Im deviating a bit from McIntosh Unity gain transformer coupling since I want voltagegain. McIntosh uses the same turns ratio on both anod and the cathodfeedback windings to get unitygain, and I will have 4 times voltagegain. The feedbackwindings will be in parallel for a 8 to 32 windingratio.
I hope to make it down to lower midrange at least.
Im deviating a bit from McIntosh Unity gain transformer coupling since I want voltagegain. McIntosh uses the same turns ratio on both anod and the cathodfeedback windings to get unitygain, and I will have 4 times voltagegain. The feedbackwindings will be in parallel for a 8 to 32 windingratio.
I hope to make it down to lower midrange at least.
Actually, there are some good reasons for using the McIntosh topology in solid state. Years ago, I wanted a decent amplifier for a car, but with an HT of only 13.8V, choices were a bit limited. Using a transformer immediately doubles the available swing and it allows a true push-pull stage to be made (NPN and PNP aren't really complementary, and as for FETs). I wound a PP transformer and was going to use a pair of IRF131 but the car disappeared so the project bit the dust...
In a desperate effort to apply some local feedback to the horrible non-linearity of the FETs. Bear in mind that I wouldn't have been able to apply much global feedback on account of adding a nasty horrible output transformer...
- Status
- Not open for further replies.