Poor man's transformer volume control?

[mikelm]

If you have a very low source impedance and you use very few turns you can quite easily wind your own with good results, be it an autoformer or a true transformer (place the tapped secondary between two layers of 1/2 primary each), however the nominal impedance will be low, usually in the region of a few 100 Ohm and level handeling tends to be shitty.

Hi Kuei

This point is very pertinant to a project that I am working on right now.

I was under the impression that generally speaking transformers work better in a high current / low impedance environment.

I this light I was thinking that if the o/p impedance and the termination impedances are both lowish then fewer turns can be used without loss of low level quality given that the magnetic flux in the core for a given i/p signal is equal to that of a higher impedance system with more turns.

Am I mistaken ? or is this understanding sound ?

thanks

mike

 

[ThorstenL]

Konnichiwa,

I was under the impression that generally speaking transformers work better in a high current / low impedance environment.

Not neccesarily. But it is easier to design/wind transformers that work in said conditions.

Am I mistaken ? or is this understanding sound ?

In principle you seem to be right. I have found however that things are rarely as cut & dried....

Sayonara

 

[Bill F.]

Some more out-loud thinking on TVC triamping:

Another benefit of the relaxed bandwidth requirements might be reduced low-level distortion through use of air gaps while still maintaining sufficiently high inductance.

As I understand it, TVCs without air gaps can allow a little bit of low-signal distortion to creep in (the bottom of the S-shaped initial-magnetization curve). But of course, gapping lowers inductance.

I don't know for sure, but I imagine it wouldn't be a terrible challenge to wind a bass TVC with bandwidth to >5kHz, inductance around 100H, and a gapped core.

The mid TVC wouldn't need more than 10-15H, and I doubt the treble would need as much as 3H, so again, gapped cores all around.

Perhaps even a ferrite core on the treble?

 

[ThorstenL]

Konnichiwa,

As I understand it, TVCs without air gaps can allow a little bit of low-signal distortion to creep in (the bottom of the S-shaped initial-magnetization curve). But of course, gapping lowers inductance.

Well, the S&B TX-102 manages a midband distortion of -105db (0.0005%) for 5V RMS input and much lower at lower levels. Only below 100Hz does distortion rise to -75db (0.02%) @ 20Hz/5V with 5V input from a 600 Ohm Source (mneasurements from HiFi news review).

I don't know for sure, but I imagine it wouldn't be a terrible challenge to wind a bass TVC with bandwidth to >5kHz, inductance around 100H, and a gapped core.

Maybe. But the TX-102 has 400H with an ungapped core and is still rather linear.

Perhaps even a ferrite core on the treble?

Maybe. Why not try it?

Sayonara

 

[Bill F.]

I believe I will try it. Thanks for the encouragment. Gapped silicon steel cores for bass and mid should be far less expensive than mumetal, and ferrite cores are a dime a dozen. I'd like to get the whole six-channel volume control done for $400 or less. (The most expensive part may end up being the 6-deck switch.)

Do you think split windings on UI humbucker cores would be worthwhile? Shielding might then be far less of an issue.

 

[ThorstenL]

Konnichiwa,

I believe I will try it. Thanks for the encouragment. Gapped silicon steel cores for bass and mid should be far less expensive than mumetal,

Yes, they are. But steel cores also sound different. You want Nickel especially for the bass.

Do you think split windings on UI humbucker cores would be worthwhile? Shielding might then be far less of an issue.

Depends how it is done. I'm afraid you are on your own in making your own. You know how the ones I had design input in are made. Had I felt that there would been benefit from applying other techniques I would have suggested them.

Sayonara

 

[paba]

I always wanted to try one of these units before spending big bucks a proper TVC. Granted some models don't do 20-20KHz but a few do.

www.hometech.com/audio/volume.html

But, I ended-up skipping this step and order Tx-102s if only I could find a nice chassis to put them in I would get started.
Seems hard to find a metal box with wood sides. All wood, or metal sides with wood top.

I would prefer avoiding to build the chassis as my tools and skills are more electrical than wood working or metal working.

cheers
paba

 

[Bill F.]

Yes, they are. But steel cores also sound different. You want Nickel especially for the bass.

I'm curious--in the bass range, what properties of nickel make it sound better? Do these differences manifest themselves when inductance is the same?

It also seems to me that, in transformers of equal inductance, ungapped vs. gapped steel cores might sound quite different. In one, the initial magnetization curve of the steel predominates. In the other, the characteristics of the air in the gap. Any info or experiences on this?

On an only semi-related point, I wonder what sonic changes might be had by including a small coil to carry a little DC current to effectively bias the core into its linear region. Obviously, the design would have to account for the reduction in available inductance. I imagine it wouldn't benefit the bass, but maybe higher frequencies...

 

[ThorstenL]

Konnichiwa,

I'm curious--in the bass range, what properties of nickel make it sound better? Do these differences manifest themselves when inductance is the same?

Higher permeability, more linear BH curves.

It also seems to me that, in transformers of equal inductance, ungapped vs. gapped steel cores might sound quite different.

They might, but size also goes up. If you don't mind big transformers, why not use an aircore TVC for midrange/treble?

Sayonara

 

[Bill F.]

why not use an aircore TVC for midrange/treble?

Now there's an interesting idea! It crossed my mind too a couple days ago, but I thought it would take too long a winding with too much capacitance. (I was figuring on treble bandwidth from 500Hz up wound for 2kOhms, so about 0.63H.

Fact is, -3db @ 500Hz for the treble section is probably overkill since I'd rarely cross below 4k. If I raise the cutoff to 1khz, that'll need half the inductance.

Think it's really possible with an air core?

 

[Bill F.]

Sample aircore primary winding from an online calculator:

Inductance 300.11 mH
DC Resistance 32.8 Ohms
Wire Gauge 24 AWG (pretty hefty, but thinnest option)
Wire Diameter 20.1 mils (1 mil = .001 in)
Coil Length 0.8 in
Coil Inner Diameter 1.5 in
Coil Outer Diameter 3.47 in
Average Turn Diameter 2.46 in
Wire Length 1253.5 feet
Copper Weight 1.53 pounds
Turns 1945
Levels 48.87
Turns/Level 39.8

Add a multitapped secondary, and it's a big transformer indeed.

 

[Steve Eddy]

So, in electrical terms at least the EP arrangement achieves the exact and polar opposite of what the classic TVC achieves. Instead of loading down the source minally while keeping output impedance low as the classic TVC does the EP arrangement maximally loads down the source and creates a fairly high output impedance.

Yes. But then its primary intended purpose wasn't to provide attenuation as a TVC does but to provide voltage gain. They wanted a truly passive "passive preamp" without the "preamp" part being an oxymoron as it typically is.

Of course the consequence of having a 1:8 step-up with reasonably low output impedance is a crushingly low 150 ohm input impedance.

Elliano wrote up a piece about this preamp for AudioXpress some months back and conveniently left out any mention of the 150 ohm input impedance which quite a few source components won't get along well with.

se

 

[ThorstenL]

Konnichiwa,

Yes. But then its primary intended purpose wasn't to provide attenuation as a TVC does but to provide voltage gain. They wanted a truly passive "passive preamp" without the "preamp" part being an oxymoron as it typically is.

Using a TX-102 does the same and with a sensible level of gain (6db).

Of course the consequence of having a 1:8 step-up with reasonably low output impedance is a crushingly low 150 ohm input impedance.

And between 12 - 18db too much gain for a linestage... ;-)

Sayonara

 

[Steve Eddy]

Using a TX-102 does the same and with a sensible level of gain (6db).

Yeah. But I thought they stopped making that version.

And between 12 - 18db too much gain for a linestage... ;-)

Not if it's feeding a power buffer.

se

 

[ThorstenL]

Konnichiwa,

Yeah. But I thought they stopped making that version.

No, why would you think that?

Ciao T

 

[Steve Eddy]

No, why would you think that?

I wish I could remember what it was that I'd read that gave me that notion but it was nearly a year ago and I don't remember the specifics.

Since they're still making it I'll just go ahead and get rid of that notion.

se

 

[fdegrove]

Hi,

I wish I could remember what it was that I'd read that gave me that notion but it was nearly a year ago and I don't remember the specifics.

Same here and I'm pretty sure it was Grandmaster Flash TL himself saying so too.
Maybe the super duper local search crusher could dig it up?

Cheers,

Edit:

Rewound my magnetic tape:
Could it be that the original version, the one TL specced for S&B to make is no longer manufactured and S&B settled for an easier to manufacture but somewhat inferior version? >> FF.........Buzzzzzzzzzzzzz.

 

[jeff mai]

Sample aircore primary winding from an online calculator:

Add a multitapped secondary, and it's a big transformer indeed.

A multitapped autotransformer will be half the size and benefits from no interwinding capacitance / coupling issues. 30ga wire helps reduce the size too.

 

[Steve Eddy]

Same here and I'm pretty sure it was Grandmaster Flash TL himself saying so too.
Maybe the super duper local search crusher could dig it up?

I'm afraid my super duper local search crusher is at the cleaners.

Rewound my magnetic tape:
Could it be that the original version, the one TL specced for S&B to make is no longer manufactured and S&B settled for an easier to manufacture but somewhat inferior version? >> FF.........Buzzzzzzzzzzzzz.

Mmmm. Don't think so.

I seem to recall something about wanting to keep the primary a single winding rather than two.

se

 

[pooge]

But a transformer will break ground loops, and easily allow for a mix of single-ended and balanced inputs and single-ended or balanced output (the best feature, IMO), with no additional conversion circuitry.