• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Transformers adding some "iron" to the sound

b1daly

Active Member
Forum Donor
Joined
Dec 15, 2018
Messages
210
Likes
358
The folks at Gearslutz are almost exclusively focused on music production. The reason they are interested in transformers is that producing good sounding recordings is very hard, and there is an endless search for anything that might help.

Most producers “mix in the computer” and apply insane amounts of signal processing to the various signals. There is a whole category of DSP distortion algorithms, which attempt to help “glue” multichannel recordings together, and add euphonious effects.

Of the many challenging aspects of producing recordings is that you have no control over the listening environments. The goal is to create a recording that “translate” across the myriad of playback environments. This is somewhat magical, but there are objective elements to this. (Relating primarily to the smoothness of dynamic levels across the frequency response.)

This presents a dilemma for both producer and listener. Producer tries to guess what a listener might be using. The listener needs to pick a playback system that is somewhat similar to the expectations of the producer.

As both a music producer and listener, this is a fascinating dilemma, which I am constantly learning.

If there was a product that, all things equal, made most recordings in genre the listener preferred sound “better” by introducing distortion, then this would be a desirable product.

I’m not sure whether this is possible, but nevertheless this is what the market for playback systems is producing on the consumer level. The ubiquitous small speaker systems that are so popular are using DSP to create a uniform and “exciting” sound. To my ear anyway. For example, the modern Bose speaker systems sound like they have all sorts of DSP trickery to make them sound “spacious.”

Frankly, it’s a mess.

I have one more comment related that fascinates me. When it comes to studio monitor speakers, they have a different task than regular listener speakers. They are intended to give the producer a listening environment to work in, which allows them to hear clearly, and hopefully results in an effective recording.

This is a hopelessly complicated subject, but one anecdotal observation I’ve had is that some monitors are “easy to please” meaning they tend to make for a pleasing listening experience, even when there are “flaws” in the mix. Generally, this is not a good thing. On the other hand, a speaker, or playback system, with this characteristic would likely be a better choice as a listening system.

Mostly I work on near field, bi amped, active monitors. These get the job done, but I find them very unpleasant to listen to. For example we have a Genelec nearfield system with active sub in our main studio, which is remarkably transparent. Meaning it reproduces the sound of an instrument in the recording space relatively accurately.

But for listening, I find the sound “clinical” and too controlled. For listening I prefer speakers that have a “coloration” that I happen to find pleasing. I generally like have a “boxy” sound, with some amount of resonance coming from the cabinet.

I’m not sure what accounts for this, but my current theory is that since I listen to mostly multitrack recordings, the speaker cabinet “integrates” the sound, providing the final “mix” to the sound. A strictly accurate speaker can result in a listening experience where the sounds are kind of “superimposed” over each other. This is an unnatural experience. Generally, if we hear sounds in a space, a collection of instruments, all of the sounds affect each other.

For example, in an orchestra every instrument generates resonance in its neighbors. Then the ensemble excites the room that they are in.

If you have captured this with a simple stereo mic setup, then it might be desirable to have a very accurate playback system. But in a case where the instruments are captured separately, in the controlled space of a studio, this interaction is lacking, and must be simulated by the mixer.

At the “mastering” stages further attempts are made to “integrate” the sounds, and the use of various pieces of analog gear with distinctive characteristics is practically a religious subject.

In this sense, listeners experimenting with tubes or other distortion producing devices like transformers are simply extending this process, and trying to create a system that sounds “good” to them.
 

Panelhead

Senior Member
Forum Donor
Joined
Apr 3, 2018
Messages
348
Likes
137
I use 1:1 input transformers in my preamp, and if they're adding harmonics/distortion, the AP doesn't seem to be able to detect it. Ditto the 1:10 transformers in my phono stage.
SIY,
What brand of line input transformer have you selected for your projects? Most of my use has been either microphone or moving coil step ups. Tried a lot of different ones, last was an AN UK. Worked well.
Have used Lundahl and Cinemag 4:1 line input for balanced to single ended and step down. Used Cinemag 15/15B in many preamps.
Got Cinemag 10K:600 on the input of current amplifier. Planning on converting an older project amp and have a another pair. But ordering a pair of Jensen or mumetal Lundahl is an option. Have a pair of the amorphous core Lundahl step downs. They do not measure as well as the Cinemag.
 

SIY

Grand Contributor
Technical Expert
Joined
Apr 6, 2018
Messages
10,500
Likes
25,315
Location
Alfred, NY
SIY,
What brand of line input transformer have you selected for your projects?

Cinemag 15/15B and Jensen JT11-P1. I also built a unit with some vintage Altec/Peerless 1:1 transformers.
 

Wombat

Master Contributor
Joined
Nov 5, 2017
Messages
6,722
Likes
6,464
Location
Australia
A bit of magnetic hysteresis love? ;)
 
Last edited:

Panelhead

Senior Member
Forum Donor
Joined
Apr 3, 2018
Messages
348
Likes
137
Cinemag 15/15B and Jensen JT11-P1. I also built a unit with some vintage Altec/Peerless 1:1 transformers.
Good enough, I will use my extra pair of Cinemag. The amp showed less harmonic distortion with them than it did with amorphous core Lundahl. Especially as the input level was ramped up.
 

MattHooper

Master Contributor
Forum Donor
Joined
Jan 27, 2019
Messages
7,310
Likes
12,255
But for listening, I find the sound “clinical” and too controlled. For listening I prefer speakers that have a “coloration” that I happen to find pleasing. I generally like have a “boxy” sound, with some amount of resonance coming from the cabinet.

I’m not sure what accounts for this, but my current theory is that since I listen to mostly multitrack recordings, the speaker cabinet “integrates” the sound, providing the final “mix” to the sound. A strictly accurate speaker can result in a listening experience where the sounds are kind of “superimposed” over each other. This is an unnatural experience. Generally, if we hear sounds in a space, a collection of instruments, all of the sounds affect each other.

I find that comment interesting because it so closely matches what I tend to hear from various speaker designs, especially when comparing the more "modern" lower coloration/highly damped speaker designs from the (perhaps older style) thinner-walled "let the cabinet ring a bit" school.

One overriding impression I have over and over whenever I set up my old Spendor speakers is how integrated the mix sounds on most music. Where the super-clean speakers separate out every instrument almost to a fault - on the Spendors even though they image like mad, there is some connecting tissue to everything, where it all sounds integrated in the same mix, rather than random sounds popping out here or there.

I had a similar experience when I owned the Harbeth SuperHL5plus speakers. I directly compared them to the Thiel 3.7 speakers I had at the time. The Harbeths showed all the things about the mix the Thiels did, but it was almost like the actual space between instruments itself had some sort of textural presence, and I constantly had the impression of a sense of balanced/coherence to tracks played through those speakers.

One example I"m thinking of is an album I've played forever from the Los Angelese Guitar Quartet, so 4 classical guitars playing spread right to left in the stereo image. On the Harbeths there was that sense of something xtra added in to the sound where they were all connected. The same track played on the Thiels and things seemed to "clear up" so now there was the sense that only sound was coming directly from the very tightly imaged guitars themselves, with the space between them losing that sense of "thereness." The effect to my ears was that now the guitars themselves sounded a bit more real on the Thiels. But it was also now more like 4 "more realistic" guitars playing individually vs "a recording of 4 different guitars all on the same recording" on the Harbeths. I don't know that it was true at all, but my intuition tried to explain this as my hearing the cannily incorporated thin-walled box sound of the Harbeths, that was "cleaned and cleared away" by the Thiels. (And I don't know how much other aspects had to do with it, e.g. the wider baffle-design Harbeths, frequency response differences or whatever).
 

egellings

Major Contributor
Joined
Feb 6, 2020
Messages
4,063
Likes
3,305
I built tube power amps based on the circlotron design, in which 2 floating plate supplies and two output tubes are arranged in a bridge (circlotron) setup. The OPT is the bridging element. There's no DC in the OPT when the tubes' plate current is balanced (the same!), and since the tubes are essentially in parallel (AC-wise) across the OPT primary, I get a 4 fold drop in required turns ratio compared to what I would get using a conventional push pull arrangement, which places the tubes in series with the primary. Low turns ratios are easier to implement and result in better OPTs. I use two small floating supplies for the screen grids so I don't burn the grids off at high volume levels. This also reduces distortion because screen to cathode voltages on the tubes are relatively constant, not swinging as they do in the classical circlotron design. The power xfmr has all the windings needed to supply all the floating voltages. The amps work like a charm.
 

mhardy6647

Grand Contributor
Joined
Dec 12, 2019
Messages
11,388
Likes
24,670
1581376151182.png

http://circlotron.tripod.com/
 

egellings

Major Contributor
Joined
Feb 6, 2020
Messages
4,063
Likes
3,305
That is what got me started on the circlotron idea. The problem with those EV amps is that the screen grid on one tube of the output stage got connected to the plate supply of the opposite tube. When played loudly, screen grids swung above their plate voltage and drew destructive currents. Besides distorting badly, the output tubes did not last long if played loudly. Adding floating screen supplies (low current, 20mA or so max) got around that. My nun handles are happy with the sound!
 

mhardy6647

Grand Contributor
Joined
Dec 12, 2019
Messages
11,388
Likes
24,670
some folks use transformers for all sorts of interesting things.

MCSUT2 by Mark Hardy, on Flickr
(repurposed Triad transformers for use as LOMC SUTs, courtesy of Gary Kaufman)

Y'all know of Dave Slagle, yes?
Nice article (column) 'bout him and one of his ventures (EMIA) in Stereophile this month, FWIW.

One of Dave's other ventures:
www.intactaudio.com

here, e.g., is a Slagle electromagnet(ic) phono cartridge.

DSC_5243.jpg


...and a couple of Slagle DC (low voltage, high power) power supplies (for field coil loudspeaker drivers) using Tungar tubes.

DSC_4319.JPG


I use one of Dave's early (and relatively cheap) autoformer volume controls on my 'real' hifi upstairs.

DSC_4219 (2) by Mark Hardy, on Flickr
 
Last edited:

Sharpi31

Active Member
Forum Donor
Joined
May 20, 2020
Messages
125
Likes
305
I’ve played around with transformers a lot over the years. I’ve built a number of DIY DACs using V-out chips (AK, Cirrus) and 600ohm 1:1 output transformers (like milw50717 above). It’s perfect DIY experimentation stuff - the output transformer enables both output phases of the DAC to be utilised and converted for single-ended amplifier inputs, isolates the amp from the DC offset present on both DAC outputs, and different transformers do appear to provide different colour/flavour (=distortion) - all in all transformers used in this way are a simple and elegant solution (but not the last word in accuracy, no doubt). I’ve also used pairs of input transformers to convert line level audio from Left/Right to Mid/Side (by combining output windings for sum & difference), then amplified and replayed the Mid/Side channels via a DIY single stereo loudspeaker.

I also still use a Music First Classic V2 transformer based passive preamp. That’s proven to be a very useful and (subjectively) transparent device - helpful with grounding issues with PC audio sources. Aside from galvanic isolation, the transformer preamp also avoids the impedance matching issues/compromises that can occur with potentiometer-based passive preamps.
 
Last edited:

egellings

Major Contributor
Joined
Feb 6, 2020
Messages
4,063
Likes
3,305
I have to admit that I sorta like the sound of transformers. I have Jensen SUTs and OPT and power transformers on my home-made tube amps. I like my iron. I go to a gym and move it around a bit, and I even ingest some in my multivitamin tablet.
 
Top Bottom