Designing the biggest, most euphonic set of transformers?

This one is for CJ! Though it might be an intersting discussion.

I'd love to learn about designing and specifying transformers...
Instead of just buying off-the-shelf designs it seems people like Sowter and Cinemag are able to do customs for reasonable money.

But where would one start?
Are there any good books on transformer design?

I found some good PDFs from 1950s military and academic journals on transformers but most are in regards to valve designs.

Is it impossible to do without a highlevel of mathematics?

If one wanted to spec the biggest, most euphonic, colourful set of transformers...something that had open high freq response >50kHz and could take good levels but still produced nice low freq, harmonic distortion...

Where would one start? Can open top end be achieved if going for vibe?

Why aren't there new transformers made that are really colourful...Lundahl/Jensen are all very good performers. Sowters Neve replacements are smoother/cleaner more open than the originals?

If one wanted the coolest 10k:10k 1:1 line input TX, a 1:3 mic input and a 600:600 output where to start?

Steel cores?
Winding techniques?

I'm intrigued by what goes into specifying a custom TX?

Thanks Tom

Check out Groner's mic preamp META. There have been discussions in here about books on transformers, some of which may be in there.

Books that come to mind include Grossner, Transformers for Electronic Circuits, and Flanagan, HB of Transformer Design & Applications.

To a certain extent the details of successful commercial transformers are often trade secrets, and although some details can be deduced from measurements sometimes the device must be CJed :razz:

While this may be a personal problem, I have a philosophical problem with trying to perfect a circuit component that can be eliminated.

I guess the bottom line is what is you target or end goal? The title of this thread "biggest" is not a typical design goal, and "most euphoric" (euphonic?) sounds like code words for intentional nonlinearity.

Since the perception of coloration is likely quite subjective this may not be well covered in a technical discussion.

Have fun... BTW I don't know how much he published on the subject but Deane Jensen did some work in the metallurgy of materials used and some related factors perhaps less explored.

JR

I thought there was a huge selection of transformer info here. have you exhausted the search here already?

Euphoric...

He wants happy transformers?

[quote author="bcarso"]Check out Groner's mic preamp META. There have been discussions in here about books on transformers, some of which may be in there.

Books that come to mind include Grossner, Transformers for Electronic Circuits, and Flanagan, HB of Transformer Design & Applications.

To a certain extent the details of successful commercial transformers are often trade secrets, and although some details can be deduced from measurements sometimes the device must be CJed :razz:[/quote]

Thanks Brad. Am going throught the preamp META again, as well as CJs transformer META.

I did spot some of the books in the library META and will try and get a few asap. I have some interesting material, there is a Sowter article on soft magnetic materials and some of the useful old texts...

I figured somethings must be trade secrets. Like how did Rupert end up at his torroidal design for Amek? Or his new designs for Portico?
As far as I can see most manufacturers are using off-the-shelf transformers these days. Rupert and Paul Wolf seem to be the only ones to have some custom stuff.

I realise that the Triad, UTC iron back in the day was likely off-the-shelf stuff too, but you see so many people hold these in high regard compared to new stuff.

It appears to me that the modern transformers are cleaner than the vintages ones, and there aren't really that many really colourful ones available currently...maybe I am wrong. I was just wondering. I'm not really talking from experience as I have never had any UTC iron - just observation on trends.

[quote author="JohnRoberts"]While this may be a personal problem, I have a philosophical problem with trying to perfect a circuit component that can be eliminated.[/quote]

John, are you saying get rid of the transformers and associated colouration?

[quote author="JohnRoberts"]The title of this thread "biggest" is not a typical design goal, and "most euphoric" (euphonic?) sounds like code words for intentional nonlinearity. [/quote]

Hahahaha yeah that was a mental brain fart I meant euphonic...sorry I guss I'm having a 'gearslutz' moment!!! Its hard to describe sound with words. But yeah something that intentionally has a 'sound'...

[quote author="amorris"]I thought there was a huge selection of transformer info here. have you exhausted the search here already?[/quote]

Oh ya, there is, no doubt - I just didn't stumble across a thread on the specifics on specc'ing a design - laying it out from the get go.

You know - do designers have a 'tonal' goal when choosing custom transformers? You would think they do. It seems most just buy a bunch of them ad listen until they find one they like, that fits with their aesthetic, totally subjective I know.

Could you do that and then say to Per Lundahl..."I love the LL15xx...but could we extend its high frequency response and made it sound more 'weighty'?" and have him understand what you mean?

It seems transformers are almost as big a part of the signal path as every amplifier stage following put together...so why not 'custom' voice them so to speak?

I wonder if there are concepts above the usual stuff presented by current manufacturers, such as nickel cores for lower low frequency distortion and steel for more...or hybrid 50/50 for inbetween.

I need to read about winding configs I think...why would you choose bifilar over quadfilar?

[quote author="NewYorkDave"]Euphoric...

He wants happy transformers?[/quote]

Hahah Dave - absolutely. Who'd have a Decepticon when you can have an Autobot?

I just want a transformer with a smily faced frequency response dammit!!

I changed the typo...

Cheers Tom

chapter 5 from radiotron I think it's the radiotron handbook comes to mind. Isn't that the one on transformers?

Yeah I just downloaded that the other day...am reading now.

Just read some good threads from here...PRR and CJ got the goods.

CJ mentioned something interesting...Cobalt a slower core material good for bass but no good at high freqs so mixed with Nickel alloy or similar.

I guess like the Wunder Audio transformers..from the Allotrope. They look like Sowters to me....a mix of Cobalt, Iron and Ni

Also just read that Nickel works well at low signal levels like mic inputs due to less loss of low level detail but can saturate oddly at higher line levels. And there is a good thread from Edcor which seems to suggest Nickel being a more expensive option and not really any better than M6 steel for price/performance...if I understood correctly.

Sowter require the following info for a custom part:

Output impedance of device driving the TX
Input impedance of the device driven from the TX
Turns ratio
Quiescent DC applied to Primary
Max input level in dBu or volts
Lowest freq at which max level is desired
-3dB points top and bottom
Electrostatic screening
Mumetal shields
Packaging and other requirements.

This is interesting. I wish I had looked at this sooner.
I feel this thread is almost premature....I'm playing catch up with myself!!

-T

And when someone tells you that transformers are noiseless (other than thermal noise in the winding resistances), remind them of Barkhausen noise. I got my audiophile friend going with that one :razz:

I am thinking about to designing an amp around Lundahl LL7905-

It seems like the 'biggest' badest off the shelf input transformer out there, FWIW.

:guinness:

Brad I just googled Barkhausen...holy hell thats another world of info...lol

That LL7905 is huge - I looked at it and the 7903 before. Think I might order some.

Going to get a few from different manufacturers to shoot out.

Wired 1:2.8 mic input it has crazy headroom. Seems to lose top end response when wired as balanced to unbalance...although -1dB @ 30Hz is still pretty good! lol

-T

I built the c800g type microphone with a 7903.

[quote author="TomWaterman"]
[quote author="JohnRoberts"]While this may be a personal problem, I have a philosophical problem with trying to perfect a circuit component that can be eliminated.[/quote]

John, are you saying get rid of the transformers and associated colouration?

Cheers Tom[/quote]

In a nutshell... yes, but YMMV.

As a product designer my philosophy for all but effects devices was to attempt the "straight wire with gain" design target. Any attempt to add euphonious coloration becomes problematic a) because good/bad can be personally subjective, and b) when designing just one link in a multiple component audio chain, how do you account for either the same coloration added in another part of the chain, or multiple colorations that don't play nicely together.

My personal opinion is if you identify a nonlinearity that you believe sounds so good that it is merchantable (people will pay for it), put it in it's own black box, with a bypass switch.

I appreciate that some players in the more esoteric high end are playing coloration games with presumably linear gear, and probably do anticipate what will be used in the rest of their typical customer's audio chain. A little too slippery for my taste.. I lean toward accuracy. I am not opposed to whatever it takes to improve sound, just give me a bypass switch in case I don't agree.

I recall when they first came out with the vocal exciters, and I kept trying to correct the station tuning on my FM radio... (remember when those were adjustable?).

JR

> the biggest, most euphonic, colourful set of transformers

Behind my building are three transformers. Estimating the windings, I'd bet you could load a couple power tubes into a speaker with two of them. Is 5 tons big enough? They are ugly-green, but you could paint them any colorful colors you like.

hihi prr :grin:

[quote author="JohnRoberts"] I lean toward accuracy.[/quote]
i think everybody does. it´s a matter of definition, some believe in measurements for accuracy, others believe in what they hear.
if linearity (the way we measure it) was everything that counts for sonic / euphonic reproduction, wouldnt we all take measurement microphones for recording?

imo designing recording equipment is more than science, art/taste is involved, similar to building instruments. at least with tube microphones, the circuits are rather simple and so every part has a big impact on the sound.
and every part has it´s own unique character, putting them together is like cooking.
the transformer of a mic is working very close together with the tube, there´s no point designing a transformer without the intended use in it´s circuit.

audio is a complex field, looking for the best mixture of balance and character.

excuse my english...

-max

[quote author="ioaudio"]hihi prr :grin:

[quote author="JohnRoberts"] I lean toward accuracy.[/quote]
i think everybody does. it´s a matter of definition, some believe in measurements for accuracy, others believe in what they hear.
if linearity (the way we measure it) was everything that counts for sonic / euphonic reproduction, wouldnt we all take measurement microphones for recording?

imo designing recording equipment is more than science, art/taste is involved, similar to building instruments. at least with tube microphones, the circuits are rather simple and so every part has a big impact on the sound.
and every part has it´s own unique character, putting them together is like cooking.
the transformer of a mic is working very close together with the tube, there´s no point designing a transformer without the intended use in it´s circuit.

audio is a complex field, looking for the best mixture of balance and character.

excuse my english...

-max[/quote]

I tried to draw the distinction between linearity/accuracy for reproduction and flexibility wrt effects, but perhaps I wasn't clear.

Not to feed this topic drift but a friend of mine used to offer classes from his recording studio and I sat in on the session he taught on microphones. One thing he used to get a kick out of was lining up 3 or 4 popular microphones and then EQ'ing them to pretty much sound the same. It was a powerful demonstration.

I don't deny that there are differences in polar pattern and subtle differences between transient accuracy of small vs. large diaphragm mics, etc. the far larger difference is simple frequency response.

Just like every console channel comes with EQ that gets well used, mic selection can often be broad strokes EQ. An artist that gravitates to a certain microphone that is complementary to his/her voice, is probably not listening for the transient accuracy. Pattern may make a bigger difference for live SR than recording but it still matters as different acoustic instruments have sundry propagation patterns, proximity effect for vocals, etc.

Making music involves doing whatever it takes, reproducing music means accurately passing some one else's decisions of how to alter the original product.

YMMV

JR

[quote author="JohnRoberts"]I tried to draw the distinction between linearity/accuracy for reproduction and flexibility wrt effects, but perhaps I wasn't clear.[/quote]

hello john,

you were perfectly clear about that - but i think, that every circuit has it´s own "signature" , in my book there is no "wire with gain". preamps feature different charateristics, thats why many recording studios have more than one type.
nonlinearities involve more than frequency response at a fixed level, especially regarding transformers (thats why i think this discussion is still on topic)
subtle resonances which only occour under certain conditions, under certain levels at certain frequencies, how does the core take bass, intermodulation distortion etc

-max

At the risk of appearing cynical, preamps are probably 10x or 100x more similar to each other than microphones are. Further, some esoteric preamps in an attempt to differentiate may play games with input termination.

While there appears to be a defacto standard termination impedance around 2k ohms (10x nominal source impedance of mic), some products deviate from that enough to perhaps audibly interact with some mic's reactive source impedance.

Just like microphones are often selected for some crude EQ, preamps that either hype or diminish a microphones signature may be part of that selection process. If I was operating a recording studio I would have some "money channel" preamps just for the psychological reasons. There are a few prejudices regarding preamps. #1 having more than 2 channels in a single chassis (such as in consoles) is too much of an opportunity for the manufacturer to cut corners so console preamps will be compromised or sub standard, and #2 more expensive is better.

To get the best performance out of an artist, it's good for them to think their vocal is getting special treatment. IMO that money is better spent on a quality mic but YMMV.

I have never heard IM distortion that I found remotely euphonious. Harmonic distortion that mimics natural musical overtones can in some cases be good sounding, but that is subjective and depends on the content of the original source, not something I would be comfortable making intrinsic to any accurate audio path.

JR

[quote author="JohnRoberts"]
I have never heard IM distortion that I found remotely euphonious.
JR[/quote]

Is it Arthur Benade in his little book Horns, Strings, and Harmony that shows an example of make-your-own IM in a Telemann duet for two recorders? It's kinda cool---in essence the composer, we assume with knowledge aforethought, contrived the difference tones to be about of proper pitch to add a phantom bass line! It works---I tried it with a friend. It's most audible to the performers, since it's mostly the nonlinearity of the ears that does the heterodyning. :green:

[quote author="bcarso"][quote author="JohnRoberts"]
I have never heard IM distortion that I found remotely euphonious.
JR[/quote]

Is it Arthur Benade in his little book Horns, Strings, and Harmony that shows an example of make-your-own IM in a Telemann duet for two recorders? It's kinda cool---in essence the composer, we assume with knowledge aforethought, contrived the difference tones to be about of proper pitch to add a phantom bass line! It works---I tried it with a friend. It's most audible to the performers, since it's mostly the nonlinearity of the ears that does the heterodyning. :green:[/quote]

Yes, I have "heard of" that phenomenon and there may be a few isolated examples where very specific program material could beneficially utilize IM products as phantom notes.

Again if ever desired that could be a separate black box effect with a bypass switch, so we're not burdened with the distortion the rest of the time when it's undesirable.

JR