What transformers to run your mix through

[JohnX]

What other options would you recommend? .

The truth is that I have only begun to scratch the surface of the utility of color boxes. I currently have six of them here, each with its own characteristic. Unlike many of the people in this forum, I will also be happy to admit that even though I have been a professional recording engineer/mixer/produce for many years, when it comes to tech design, DIY projects, I barely know what I’m doing.

The truth is that I have only begun to scratch the surface of the utility of color boxes. I currently have six of them here, each with their own tonal characteristics.
For color, I generally veer towards two specific ones, and each of them has an impedance switch, which give them two colors each:

One of them I have had for a while, and it is made by a company called Big Noise - Audio Magic XFB9071 (1:1 & 1:2) - it’s got Carnhills inside of it and I would describe this as my beefiest one. It’s pretty rare for me to put this one on my own mixes but I have done some remastering to my friend’s mixes that were very thin sounding and it really helped bring them to life. I had purchased this one before I became brave enough to start building my own.

The other more colorful one that I have is a box with a pair of Lundahl 1521s inside of it. I haven’t been able to get much information about those particular transformers, but I took a secondary output which has the opposite coloration and wipes out most of the upper frequencies. I love that one on my synths and use it pretty regularly. The main switch position seems to boost the treble and after reading a previous post from someone else, I’m going to experiment with updating to a three position switch and adding a (600ohm?) resistor to one position to see if I can add yet a third option.

Although there are numerous articles that are constantly referencing incredibly expensive, vintage transformers, I am really trying to take a more “ghetto” approach - one of my boxes has a $9 pair of transformers that I just randomly bought off the Internet. Those do a nasty square wave distortion that I also use pretty often on my synthesizers.

Once I learned how to use breadboard and alligator clips (duh), I kind of stopped looking for suggestions and just started experimenting on my own - and I am finding the results to be quite satisfying

 

[Mr. Moscode]

I’m pretty new at this and I’m gonna guess that this is a total noob question but I have heard this term “ringing” with regard to transformers before, but I don’t actually understand what it is referring to.
Would you please be so kind as to explain it?

Hi JohnX,
All transformers resonate. If you hit a bell with a stick it rings - resonating at its natural frequency. If you damp the ringing with you hand, rubber or a cloth the energy will dissipate as heat.

If you hit a transformer with a short electrical pulse, the analogue of a stick, the transformer will resonate at it's resonant frequency, i.e. it "rings".
With a load - a resistor and/or capacitor, on the transformer the resonate ringing energy will be dissipated as heat in the load. Ringing is not part of the original signal so it needs to suppressed, unless it is useful for "coloring" the sound.

The resistor-capacitor values to suppress ringing are chosen to absorb energy at the resonant frequency of the transformer, or to shift it to another frequency if desired.

Your xfmers you bought probably aren't audio xfmers or are low quality ones that overload easily which is a another kind of distortion. You can see this on a oscilloscope, which an be had cheaply online. Every engineer should have one. Seeing what is happening is a hugely valuable tool.

 

[JohnX]

Ringing is not part of the original signal so it needs to suppressed, unless it is useful for "coloring" the sound.

Your xfmers you bought probably aren't audio xfmers or are low quality ones that overload easily which is a another kind of distortion. You can see this on a oscilloscope, which an be had cheaply online. Every engineer should have one. Seeing what is happening is a hugely valuable tool.

thanks so much for the clarification on this. Someone else had chimed in earlier regarding my query, but I think you’ve given me even more clarity.

One of these days, I will definitely get around to purchasing an oscilloscope and start doing this “properly” - but in the meanwhile, I am really enjoying my little breadboard rig and testing my experiments by ear. I’ve already got a couple of “clean” ones, but I do find myself gravitating towards the lower quality, colorful models.

 

[AlgoRhythms]

Hi AlgoRhythms and kags,

thank a lot for these useful informations ! We have a UREI 1178 and the manual says : " If the Model 1178 output is connected to a high impedance circuit,
we recommend shunting the +- and com output terminals with a 620 ohm, 1 / 2 watt resistor. This assures the optimal loading. "
Maybe we will try the recommended bypass switch here as well.

Kind regards, Alien 25

Sure, why not? It's a very easy mod and gives you another sound

 

[JohnX]

Great article. Simple and clear. Thanks for posting.

This article was my introduction to color boxes. I had initially thought about building such a device, but at the time I didn’t know that they already existed until I read this article.

I will be joining Barry at his Tech Breakfast in about an hour.

 

[Nen'O]

You’re the second person that I have seen mention using a direct box as a color box.

Are you just patching a balanced line out mix straight into the direct box and then bouncing the output of the direct box as your mix?

It’s an active DI with transformers with 30db of gain...it takes a low line level.

Since I got phoenix audio preamp with EQ I sold Neve 5042 and other colour/ tube gear and use only that with Neve 5043 compressor on master bus plus it’s killller preamp.

But if some di boxes can’t handle line level you can use some sort of attenuation before hitting the transformer.

 

[Cqwet Dbdfte]

The purpose of a line transformer is one or two: galvanic isolation (stop ground loops) and voltage stepping.
Incidentally it also functions as a band pass filter. The later could of course be done without isolation with R, L and C.
Transformer core size determine min freq at a given voltage level. Higher freq do not use the core as much.
The core material has some magnetic hysteresis that may affect the signal.
A low source impedance driving a high impedance makes the driving easier.
600 ohm lines sets an expectation for how much current can be drawn by the load, while meeting its specs for dBu levels. An even ratio means just a filter/isolator with a power limit, no transformation is done, i.e. just a magnetic coupling device.
If the source material has offending freq content the transformer would help as much as any band limiting filter.
Every (magnetic) transformer has some self resonant frequency, the larger the core, the lower that freq generally.
Breaking up ground loops is good. If the sound is affected by a transformer in any way it would point to a some issue, somewhere, which may be worth investigating. Grounding issues, HF noise, loading sensitivities, whatever.
It is hard to think core saturation would add anything meaningful to sound reproduction, maybe OK for a sound effect stomp box.
Using an oscilloscope and a wide band spectrum analyzer would shed some light on whats going on.
A low cost Rigol DSA8.. series would work, down to 10Hz up to GHz.
A built in tracking gen would ease freq response tests, to include unexpected out of band fluff, and a good tool for tracking RFI.

 

[Newmarket]

You can see this on a oscilloscope, which an be had cheaply online. Every engineer should have one. Seeing what is happening is a hugely valuable tool.

+1

 

[JohnX]

The purpose of a line transformer is one or two:

Thanks for that.
If you’re interested in reading about another purpose, you might enjoy this article by Barry Rudolph about the color boxes that we have been discussing.
https://www.sonarworks.com/blog/learn/transformers

 

[Graham Nystrom]

There is only one choice. It’s the secret of many mastering studios and a few big studios in Nashville. Western Electronic 111c. You have to strap them for 600/600. The Westwrn Electric web site has documentation on these.

I've racked a couple pairs up for mastering engineers. Can I ask why you've found the 600:600 strapping is the best?

 

[Cqwet Dbdfte]

What is "best" ? Objectively or subjectively?
Strapping would be to change transformation ratio, i e. Voltage gain, or loss.
How does the inputs and sources react with higher/lower voltages?
Unless you can measure an effect it would be subjective, and semi useless.

 

[AlgoRhythms]

A subjective observation isn't necessarily semi-useless. Not all knowledge is quantifiable.

 

[Cqwet Dbdfte]

It is, if there is quantifiable out of band HF noise getting into your signal.
A supposed improvement may just be filtering out some junk.
HF intermod products from SMPS', dimmers, routers, cellphones, whatnot.
Blindly applying solutions to a problem when better methods are available may not be optimal.

 

[JohnX]

add a 600 ohm resistor to the secondary terminals which help flatten frequency response and ringing.

Sorry to dredge this up again, but I’m getting ready to make a new box and I would like to implement your suggestion, utilizing a switch so that I will have the option to engage or not - If it’s OK, I would like a little bit more clarity.
Do you add one 600 ohm resistor to each of the secondary terminals to (in this case) pins 2 & 3? Or is it just one across both secondaries?

 

[AlgoRhythms]

Sorry to dredge this up again, but I’m getting ready to make a new box and I would like to implement your suggestion, utilizing a switch so that I will have the option to engage or not - If it’s OK, I would like a little bit more clarity.
Do you add one 600 ohm resistor to each of the secondary terminals to (in this case) pins 2 & 3? Or is it just one across both secondaries?

Add one 600 ohm resistor across the two output terminals. See an earlier message of mine about really dialing in the perfect termination resistance.

 

[JohnX]

Add one 600 ohm resistor across the two output terminals. See an earlier message of mine about really dialing in the perfect termination resistance.

Thanks so much. I do remember reading your instructions for dialing in the perfect resistance, but as I do not have a scope here, I just have to wing it.
What I may do is just have a few other resistors on hand and test each one out first. The only way I can test is by ear at this point.

 

[Cqwet Dbdfte]

The perfect termination resistance is one that can produce a square wave without overshoot and ringing and have flat tops.
600 ohm is good number for 1mW 0dBu, keeping those VU meters happy.
Outside of that I see no point in 600 ohm lines.

 

[MidnightArrakis]

The only way I can test is by ear at this point.

>> Be careful that your test-probes don't accidentally poke your ear!!!

/

 

[JohnX]

>> Be careful that your test-probes don't accidentally poke your ear!!!

Considering how much abuse I have put my ears and the rest of my body through, I’m not even sure if they would notice

 

[MidnightArrakis]

Considering how much abuse I have put my ears and the rest of my body through, I’m not even sure if they would notice

>> Haven't we all!?!?!?!?!?!? Now...............where did I put my rolling papers???

/