For several years I have been enjoying this forum. I have constructed many nice amplifiers from schematics and layouts published here. One thing I find missing for us as diyers is actual output transformer winding details.
I have produced several nice power transformers in the last year for my current project, but, find detailed layouts for output transformers quite scarce. I have managed to learn the math for this endeavor and sourced many a good core supplier. I also admit that the actual winding is tedious and extremely time consuming. Time and tedious endeavor are what makes for diy.
SO, why so little input on the minutae which makes for a truly good output tranny. I have noticed that some of these units employ reverse winding rotation and multiple split layers. I have yet to find anything on the net regarding why or when to reverse the winding direction. The use of multiple layers to make for better coupling is explained.
I am using many non standard output tubes , READ, inexpensive, for my projects and off the shelf iron will not work.
Is this a true art won by experimentation and failure to master?
Tad
I have produced several nice power transformers in the last year for my current project, but, find detailed layouts for output transformers quite scarce. I have managed to learn the math for this endeavor and sourced many a good core supplier. I also admit that the actual winding is tedious and extremely time consuming. Time and tedious endeavor are what makes for diy.
SO, why so little input on the minutae which makes for a truly good output tranny. I have noticed that some of these units employ reverse winding rotation and multiple split layers. I have yet to find anything on the net regarding why or when to reverse the winding direction. The use of multiple layers to make for better coupling is explained.
I am using many non standard output tubes , READ, inexpensive, for my projects and off the shelf iron will not work.
Is this a true art won by experimentation and failure to master?
Tad
Here is a small tutorial about winding an output transformer on a C core.
I've often wondered the same thing as you about the little details of winding. Right now I have a bunch of EI-150 M6 laminations and plastic bobbins. Wire is readily available on ebay and we have several coil winding machines at my job. The only thing stopping me is time.
If you do some searching you'll find a thread about a guy winding his own transformers for an 833. I always wanted to wind for the 6C33C because of the lower impedance and larger wire.
I've often wondered the same thing as you about the little details of winding. Right now I have a bunch of EI-150 M6 laminations and plastic bobbins. Wire is readily available on ebay and we have several coil winding machines at my job. The only thing stopping me is time.
If you do some searching you'll find a thread about a guy winding his own transformers for an 833. I always wanted to wind for the 6C33C because of the lower impedance and larger wire.
David Hafler, a founder of Dynaco and Acrosound, patented the output transformer coil-placement optimizations used by both of those companies and licensed to some others in patent 2,815,408. I'd suggest reading it. Aside from interleaving coils and making choices to get tight coupling while minimizing leakage inductance, capacitive currents were minimized by avoiding placing ends of layers with a large AC voltage difference next to each other. To do that and still get the needed phase for the circuit, some need to be wound backwards. For instance if you have a segment that is two layers, completing one layer then continuing to wind the second going back the other way on top of it would leave the highest voltage difference ends next to each other. If the second layer had not been connected to the first and was wound backwards, the other end of it would get connected to the first to properly add up in phase, and then the highest voltage points would not be adjacent. Choices also relate to close magnetic coupling between primary and secondary layers while keeping the lower AC voltage portions (like center tap region in push pull) of the primary closest to the secondary. Of course it gets more complicated when dealing with a number of layers. David Hafler certainly deserves credit for a major contribution in any event, and if the patent was extended a license should be obtained from the current holder for manufacture (I don't know the status). The performance difference is significant and it would be well worthwhile to encourage all audio transformer vendors to do whatever is needed, if anything, to legally offer the superior transformers. Note that optimal designs with the correct turns ratios may have slightly difference DC resistances on either side of a center tap. That isn't a problem in the plate circuit, but can upset cathode bias for circuits taking the output from there instead. In that case a small equalizing resistor should be added on the lower resistance side.
The Dynaco transformer brochure TO668 had the patent number (as well as specs on transformers they offered and both suggested circuits and those of their products that used them).
http://www.the-planet.org/dynaco/Misc/Transformers.pdf
another thread else where on transformer winding -
DIY Simple Output Transformer
The Dynaco transformer brochure TO668 had the patent number (as well as specs on transformers they offered and both suggested circuits and those of their products that used them).
http://www.the-planet.org/dynaco/Misc/Transformers.pdf
another thread else where on transformer winding -
DIY Simple Output Transformer
Last edited:
Winding details are a mystery to me too. For example, if I would like to arrange a c-core transformer with parallel-connected coils on "both straight bits", and have wires coming out on same end/side on this assembly, would I
a) wind both coils identically and then flip the other coil
b) wind the other coil to different direction and then flip the other coil
c) Other arrangement, what? Any ideas?
a) wind both coils identically and then flip the other coil
b) wind the other coil to different direction and then flip the other coil
c) Other arrangement, what? Any ideas?
You can
a) wind identical coils, and feed outcoming wires of one coil straight back to the other side to have all wires at the same side; or
b) wind the second coil in the other direction.
First remark: there is no single best winding scheme.
Second remark: there is not that much "black art" involved in winding audio transformers; with "knowing what you're doing", "accurate winding" and "patience" you will reach your goal.
It all depends on impedances, capacitance, leakage inductance, dc resistance and more.
About 15 years ago I started winding audio transformers by trial and error, and this is by far the best way to learn (but it takes time and patience). An output transformer for a 6C33 is different from an output transformer for a 300B or whatever, and by the trial and error process you will learn how to optimize.
I started to roll transfos by 1999 and, as pieter t I learned by trial and errors.
If finding the best compromise between a dozen of incompatible variables is an art, then designing an audio transformer is one.
If possible read RDH4 chapter 5 more than once, you will save considerable amount of time !
Yves.
If finding the best compromise between a dozen of incompatible variables is an art, then designing an audio transformer is one.
If possible read RDH4 chapter 5 more than once, you will save considerable amount of time !
Yves.
"Money don't smell"
I should patent my university projects in 1970'th. One of them used rotary encoder (could be a source of wellness today) for example (pitch control in synthesizer), another was SMPS for an audio amp (governments push now in this direction), and so on...
Right, but what about patenting of optimized design of transformer? Advertisement? Probably. I should consider this in the future.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.