I'm not an expert, but I do not believe that what you are describing is litz wire. As I understand it, litz is comprised of multiple, fine, individually insulated strands that are not just twisted, but woven (in patterns that can be proprietary), with the intention of reducing interaction in the EM fields from the differnt strnds and so that any given strand's total EMF "backforce" is the same as any other. I belive that Nikolai Tesla invented litz as a way of carrying large amounts of very high voltage wire without significant loss or radiation. Most litz that I know insulates the individual strands with varnish.

Litz (Litzendraht)wire is used in high-frequency inductive components to minimize losses due to skin effect. I don't see what use it would be in an audio-frequency transformer. Perhaps you're confusuing bifilar winding with Litz. Bifilar is a method whereby the primary and secondary windings are wound together (two wires side-by-side) at the same time, instead of one winding on top of the other. This improves inductive coupling between the primary and secondary. It's an expensive way to make a transformer, but it's found in some very high-quality transformers like those in McIntosh amps. I suppose the wires could be twisted together, rather than laid down side-by-side, although it seems to me that doing so would increase the leakage capacitance.

I was under the impression that api 2503s were wound with litz wire. this would not be the first time I was wrong...

dave

Bi-filar, litz, same effect achieved in two ways.
Bifilar is sometimes used when you have a large voltage between windings. Litz won't take the stress that bi-filar will.
Bi-filar will also lie flatter than litz. So if you needed a lot of turns...
We used to wind coils with fishing line inbetween the two wires to provide a uniform spacing for hv.
Winders hated it. Pain in the rear. Sometimes four or six seperate windings, each with the mono between them.
The API is designed so that the capacitive coupling starts to take place right when the inductive coupling lets loose. That's why you don't want to mess with turns or wire size when winding your own API.
Another good thing about the API is that there is very little phase shift.
I wonder what an API wound sound like with a different "normal" transformer, with all the peaks and valleys, phase shift, etc.

Yes, the API is Litz wound.

Bifilar and Litz are not the same.

Bifilar is a winding technique, and Litz is a type of wire.

Litz wire consists of several small conductors in parallel, whereas the two conductors in a bifilar winding are not connected together, since they form two galvanically separate (albeit physically interleaved and tightly inductively and capacitively coupled) windings.

The purpose of Litz wire is to reduce skin effect losses at very high frequencies. At these frequencies, the electrons tend to travel along the surface of a solid conductor; so even though the conductor may be fat, the resistance is increased because the signal will only travel across the outer portion of it. But if you take that conductor and break it up into several smaller wires connected in parallel, resistance (and losses) decrease.

Anyone who has ever tried to solder to an actual piece of Litz wire knows that it's a very different animal!

There is no reason to use Litz wire in audiofrequency transformers.

theres tons of spools of litz wire down at the famous sckycraft

sorry..

im just blabbering :D

Steve, if I give you a gauge number, can you look around for some?

In the API, there is one length of quad-Litz wire.
Unlike an RF winding, in which you correctly pointed out skin effect, the API uses one strand of the quad wire for the pri., and the rest are three individual sec. windings.
Sorry for the confusion. I just woke up!

what size do you want

4 naught? :D

you'll need to know the individual gauge size and either the equivalent gauge size or the number of individual conductors...

And this stuff is regular ole' litz, CJ, not colour-coded stoff?-YOu cool wid' dat?

Keith

cjenrick, post: 328296 wrote: Steve, if I give you a gauge number, can you look around for some?

In the API, there is one length of quad-Litz wire.
Unlike an RF winding, in which you correctly pointed out skin effect, the API uses one strand of the quad wire for the pri., and the rest are three individual sec. windings.
Sorry for the confusion. I just woke up!

The quad wire you are talking of is still not Litz wire as each wire has a seperate function. Just because you twist four wires together does not make it Litz wire - only if you short out all four at both ends does it become Litz wire. Windings can be bi-filar, tri-filar, quadri-filar etc. What you are describing here is a quadri-filar winding.

With bi-filar windings it makes very little difference electrically if the pairs are twisted or not, so it is common to see them untwisted as this makes the coils much easier to wind and more space efficient. With tri-filar upwards it is usual to twist them to maintain equal capacitance between all wires. To do the same with Litz wire would be pointless for audio but then you would use four lots of Litz wire to make the same windings.

The only use I have ever found for Litz wire at audio frequencies is to produce very high Q, high inductance coils for tuned circuits using Ferrites and not steel/iron. Even then the gains were marginal.