Litz wire

[Tim Williams]

Who sells this stuff (without paying for miles at a time)?

I'm especially interested in stupid thick stuff, like, as large as 8AWG
equivalent. Nebraska Surplus for instance doesn't stock wire like this.

Tim

 

[Tim Williams]

We get Litz wire from MWS but nothing that big. If you don't need
that much have you thought of 'rolling your own'?

I do sometimes, but only for small things. I'm contemplating 10A at 1MHz,
so it needs to be pretty fine = way more strands than I'd want to deal with.

I actually have some copper rope, which is about 1/4" diameter and looks to
be made of 28AWG or so. I don't remember how many strands it is, but if I
guess the rope is wound from 7 strands of 31 strand twist, that's 7*31 =
217. If 28AWG is good for ~200mA, 217 strands should be good for 40A, which
sounds about right, I'd call it 8 or 10AWG equivalent. I salvaged this
stuff from some old motor driver, which used a spool of this stuff for
air-core inductors.

Tim

 

[Tim Williams]

I doubt you would find such a heavy gauge.
The litz wire I have used I recovered from IF transformers in old
radios.

Well, I've got the rope stuff I mentioned earlier, and I've taken apart Sony
transformers (which were wax impregnated, not varnished!) which used litz.
Just a twisted bundle, not properly woven, but I'm not that picky.

Tim

 

[Spehro Pefhany]

Who sells this stuff (without paying for miles at a time)?

I'm especially interested in stupid thick stuff, like, as large as 8AWG
equivalent. Nebraska Surplus for instance doesn't stock wire like this.

Tim

I did a web search last week for some for a power converter
transformer. 600+ strands of AWG 46 wire for something like $1/foot in
small quantities. You could braid about five of those together..
(I'm sure you can find it as easily as I can).


Best regards,
Spehro Pefhany

 

[life imitates life]

Who sells this stuff (without paying for miles at a time)?

I'm especially interested in stupid thick stuff, like, as large as 8AWG
equivalent. Nebraska Surplus for instance doesn't stock wire like this.

Tim

Pick a gauge, buy the mag wire, and make your own. It is not that hard
at all. Properly braiding out a custom length of Litz wire is no harder
that winding your own custom pot core transformer bobbin. One just has to
set one's mind to it, and accomplish a simple task without assigning
undue difficulty to it without even trying.

Can't be any harder than macrame. You can actually come up with better
stuff than you can even buy too.

The stuff you refer to sounds like it would not be a cheap per foot
price, even if you were buying a whole spool. Probably better to fashion
a double long length, then cut it in half when you are ready to use it.

What is the application , and the length of the run? Or is it needing
to be long enough to be a primary winding?

 

[life imitates life]

I actually have some copper rope, which is about 1/4" diameter and looks to
be made of 28AWG or so. I don't remember how many strands it is, but if I
guess the rope is wound from 7 strands of 31 strand twist, that's 7*31 =
217.

The problem is that it not being constructed from individual mag wires,
it is "seen" as a single strand to the current flow, and having only ONE
skin. The litz configuration requires the wire be discreet from each
other with the exception of the termination points.

 

[life imitates life]

Isn't there some magic braiding pattern for Litz wire?

No. The wire strands have to be mag wire, which segregates them from
each other, allowing the skin effect to be taken advantage of. Without
strand segregation, it becomes a single strand, from the POV of the
current flowing in it, with only one skin for the entire mass.

 

[life imitates life]

I doubt you would find such a heavy gauge.
The litz wire I have used I recovered from IF transformers in old
radios.
I have never found a commercial seller and doubt whether anyone would
still make it

jcdrisc

One can take multiple pieces of a smaller litz, and increase the
effective gauge. No need to unwrap the individual segments.

 

[Tim Williams]

See the beast of a transformer in this thing?
http://myweb.msoe.edu/williamstm/Images/Induction1201.jpg
That's one use of the rope I found (where the plastic covering is peeled
away, you can even see the cool ropey ropeness of it -- pretty!).

That's 20 turns, and each turn is maybe 8" long, or 160" = 13.3' total. If
it's actually 231 strand as I speculated earlier, that's 3kft, more than
half a mile of 28AWG.

Now, if I build more of these, I don't need quite this many strands since
it's only running at 20kHz. But the 1MHz thing I mentioned elsewhere needs
wire at least this fine.

http://www.bcae1.com/trnsfrmr.htm
Neet, this calculator suggests I want 242 x 36AWG to get something
resembling 12AWG.

Tim

 

[life imitates life]

I've seen power inductors wound from a bundle of smaller-gage enameled
wires just bunched together in parallel and gang soldered on the ends,
but not woven like proper Litz. That apparently cuts eddy currents a
bunch. MWS sells this style too, a bunch strands of smallish wire
loosely twisted together, with thermal-strip insulation.

John

The definition of Litz relies on the fact that the strands are
electrically segregated from each other between their terminus points.

It does NOT HAVE to be specially braided or woven. The thermal strip
wrapped stuff DOES qualify as litz, depending on length and intended
application, and I have bought the MWS product before.

The proof? I made several of my own Litz configuration primaries for
transformers and the proof is in the pudding of the observed performance
characteristics, even at lower frequency switcher supply cycle rates. The
fact is that it definitely works to segregate several strands of
conductor on short runs of AC pathways or in inductors and chokes and
transformer windings. Trial and error with gauges and strand count do
more to optimize the result than ANY specialized, presumed to be required
braiding or weaving. The segregation is the key to increasing surface
area within a given cumulative "gauge".

 

[life imitates life]

See the beast of a transformer in this thing?
http://myweb.msoe.edu/williamstm/Images/Induction1201.jpg
That's one use of the rope I found (where the plastic covering is peeled
away, you can even see the cool ropey ropeness of it -- pretty!).

That's 20 turns, and each turn is maybe 8" long, or 160" = 13.3' total. If
it's actually 231 strand as I speculated earlier, that's 3kft, more than
half a mile of 28AWG.

Now, if I build more of these, I don't need quite this many strands since
it's only running at 20kHz. But the 1MHz thing I mentioned elsewhere needs
wire at least this fine.

http://www.bcae1.com/trnsfrmr.htm
Neet, this calculator suggests I want 242 x 36AWG to get something
resembling 12AWG.

Tim

STOP top posting.

If it is non-segregated "rope" as you call it, it is no different than
welding cable and is merely being used as a flexible, high current link
to the winding or as the winding media itself, but there will be NO "Litz
function" if the individual strands are not isolated from each other
electrically along the length of the conductor.

All the strands of Litz type wire co-conduct *apart* from each other,
EXCEPT at the ends where they are ALL terminated at the same node.

Other wise there is no litz configuration and there will be no litz
effect or function taking place. The "rope" or cable as it is properly
called, will have a single "skin effect realm" that surrounds the entire
bundle and NONE of the inner wires will carry any current in high freq rf
conduction. With segregated wires, each has their own skin and the total
amount of conducting copper in an rf conduction setting is an order of
magnitude greater. Of course strand gauge has an effect on that and the
improvement will be less with larger strands, but they ALL MUST conduct
separate from each other except at their common endpoint nodes, or there
is no effect.

 

[Tim Williams]

I see you don't view links. The wire is clearly individually insulated,
which is what everyone else in this thread understands by "litz", except
yourself, as you have stated six times over. It's a good thing you're in
everyone's killfile.

Tim

 

[life imitates life]

I see you don't view links.

I looked. That is NOT #28 at 231 strand count, and your calculations
are wrong as well. You probably used bare copper diameters, when you
should be using single strength mag wire diameters.

The wire is clearly individually insulated,
which is what everyone else in this thread understands by "litz", except
yourself, as you have stated six times over. It's a good thing you're in
everyone's killfile.

Goddamned top posting retard. Six times over I stated that it had to
be individually insulated, dipshit, so it is not "except yourself" you
dippy little bitch.

It isn't fucking called rope either.

And your formula result is almost off by a factor of two! 12 Ga result
does not require the strand count you gave.

 

[life imitates life]

That's 20 turns, and each turn is maybe 8" long, or 160" = 13.3' total. If
it's actually 231 strand as I speculated earlier, that's 3kft, more than
half a mile of 28AWG.

231 strands of #28 would be bigger in diameter than that wire appears
to be. Even 231 strands of #38 or #43 would be pretty big. No way that
those are #28 strands, or if they are, there are not 231 of them.

Use some common sense. 231 strands of #28 single strength mag wire has
a bundle diameter of nearly 3/8"!

Your #12 ga arrival math is off too.

D=Diameter of a bundle of wires
d=diameter of a single strand in the bundle
N=Total number of wires.

D= 1.55 x d x (Sq rt of N)

The #12 is about 0.083" in diameter without a wrap. The 242 strands of
#36 you stated is 0.0056" in diameter, which is for single strength mag
wire, and comes out to 0.135"

So, the lesson is DO NOT forget the insulation.

 

[Ross Herbert]

On Fri, 19 Feb 2010 21:14:27 -0800, life imitates life

:On Fri, 19 Feb 2010 16:18:49 -0600, "Tim Williams"
:
:>I actually have some copper rope, which is about 1/4" diameter and looks to
:>be made of 28AWG or so. I don't remember how many strands it is, but if I
:>guess the rope is wound from 7 strands of 31 strand twist, that's 7*31 =
:>217.
:
:
: The problem is that it not being constructed from individual mag wires,
:it is "seen" as a single strand to the current flow, and having only ONE
:skin. The litz configuration requires the wire be discreet from each
ther with the exception of the termination points.


I hope you don't mind a little bit of layman's interpretation as an assistance
to the OP but what you mean by "mag wire" is that each individual strand which
makes up the total x-sectional area of the litz conductor MUST be insulated from
every other strand. If the strands aren't insulated then the skin effect will be
the same as if a single strand of the same overall x-sectional area had been
used - and this is no good for high frequencies.

 

[life imitates life]

On Fri, 19 Feb 2010 21:14:27 -0800, life imitates life

:On Fri, 19 Feb 2010 16:18:49 -0600, "Tim Williams"
:
:>I actually have some copper rope, which is about 1/4" diameter and looks to
:>be made of 28AWG or so. I don't remember how many strands it is, but if I
:>guess the rope is wound from 7 strands of 31 strand twist, that's 7*31 =
:>217.
:
:
: The problem is that it not being constructed from individual mag wires,
:it is "seen" as a single strand to the current flow, and having only ONE
:skin. The litz configuration requires the wire be discreet from each
ther with the exception of the termination points.


I hope you don't mind a little bit of layman's interpretation as an assistance
to the OP but what you mean by "mag wire" is that each individual strand which
makes up the total x-sectional area of the litz conductor MUST be insulated from
every other strand. If the strands aren't insulated then the skin effect will be
the same as if a single strand of the same overall x-sectional area had been
used - and this is no good for high frequencies.

Exactly. I do not word things well on the fly. I do far better as a
proofreader of other's works or even my own work, upon subsequent
re-examination

 

[amdx]

Trial and error with gauges and strand count do
more to optimize the result than ANY specialized, presumed to be required
braiding or weaving. The segregation is the key to increasing surface
area within a given cumulative "gauge".

No need to do trial and error, Here's a table that gives recommended gauge
with regard to frequency.
See Table B.
http://www.litz-wire.com/technical.html
Same thing in pdf.
http://www.litz-wire.com/New PDFs/Frequency_Chart_3.01.13.09.pdf

Regarding "ANY specialized, presumed to be required braiding or weaving":
This is used to equalize proximity effects which equalizes currents in the
individual wires.

Quote from; http://www.dartmouth.edu/~sullivan/litzwire/skin.html
"The objective of twisting or weaving litz wire, as opposed to just grouping
fine conductors together, is to ensure that the strand currents are equal.
Simple twisted bunched-conductor wire can accomplish this adequately in
situations where proximity effect would be the only significant problem with
solid wire. Where skin effect would also be a problem, more complex litz
constructions can be used to ensure equal strand currents. Thus, in a
well-designed construction, strand currents are very close to equal."

There is a great deal of information available about the use of litz wire,
if you have the math ability
you can become an expert in a couple of days!
Mike

 

[legg]

I do sometimes, but only for small things. I'm contemplating 10A at 1MHz,
so it needs to be pretty fine = way more strands than I'd want to deal with.

I actually have some copper rope, which is about 1/4" diameter and looks to
be made of 28AWG or so. I don't remember how many strands it is, but if I
guess the rope is wound from 7 strands of 31 strand twist, that's 7*31 =
217. If 28AWG is good for ~200mA, 217 strands should be good for 40A, which
sounds about right, I'd call it 8 or 10AWG equivalent. I salvaged this
stuff from some old motor driver, which used a spool of this stuff for
air-core inductors.

It might serve if you can decouple the strands.

Shake it loose and bake it to form an ~oxide layer on the individual
strands. Doesn't have to be an insulator as such, just a poor
conductor to adjacent wires.

The actual aim of the 'twist' is to have strands in any particular
bundle or strand at near-right angles to those on the opposing
surface, as the twist pattern is scaled up.

If the twist direction is reversed on each level of build-up, surface
strands will always appear to be pointed in the approximate direction
of current flow, and the structure itself will not 'bind'.

RL

 

[amdx]

Why, then, does it work to make up multi-strand wire bundles for SMPS
service where the wires are just lightly twisted into a bundle? Are
these getting one part of the way there, but not all?

I had heard about the Litz wire 'gotta be a magic braid pattern', but
then I've seen all these SMPS transformers that just have almost-parallel
strands.
Tim Wescott

It's just a matter of good, better, and best.
Best would be properly braided* and would have the least AC resistance.
Mike
*I'm not sure braided is the proper term. Each wire needs to have equal
exposure to all positions in the bundle.

 

[amdx]

I do sometimes, but only for small things. I'm contemplating 10A at 1MHz,
so it needs to be pretty fine = way more strands than I'd want to deal
with.

I actually have some copper rope, which is about 1/4" diameter and looks
to
be made of 28AWG or so. I don't remember how many strands it is, but if I
guess the rope is wound from 7 strands of 31 strand twist, that's 7*31 =
217. If 28AWG is good for ~200mA, 217 strands should be good for 40A,
which
sounds about right, I'd call it 8 or 10AWG equivalent. I salvaged this
stuff from some old motor driver, which used a spool of this stuff for
air-core inductors.

Tim

The copper rope is not long enough? We've got some discard litz wire,
but not nearly that thin or as many strands. We use it to make high Q
coils to detect the nuclear magnetic moments of protons spinning in
the Earth's B field. Frequencies a bit above 2kHz. And lots smaller
currents. Well the same coils polarize the spins, but that's 3 amps
at DC.

George H.

Hi George,
Is this used in a magnetometer?
There is a thread on rec.radio.amateur.antenna with the
subject: Carl and Jerry Magnetometer, that has a discussion
about about picking up the precession of the proton as it
returns back to alignment with the earths magnetic field.
I found it very interesting, probably because I knew zero
about the subject.
If you have anything to add to that thread, please do so.
Mike