Buying Litz wire / Sale and Trade / Forums

Forums

4hv.org :: Forums :: Sale and Trade

« Previous topic | Next topic »

Buying Litz wire

previous 1 2 3 next

Move Thread

LAN_403

Forty

Mon Jan 23 2012, 06:17PM

Registered Member #3888 Joined: Sun May 15 2011, 09:50PM
Location: Erie, PA
Posts: 649

you could also cut through a large degaussing coil from a crt and use it as a cable. But for properly twisted litz wire i'd unwind the thing and twist it together in strands.

Inducktion

Tue Jan 24 2012, 05:18AM

Registered Member #3637 Joined: Fri Jan 21 2011, 11:07PM
Location: Buffalo, NY
Posts: 1068

Why not use copper piping? That way you can water cool it and it'd be a lot cheaper too.

m4ge123

Tue Jan 24 2012, 06:10AM

Registered Member #4118 Joined: Mon Oct 03 2011, 04:50PM
Location: MD
Posts: 140

At the frequency I'll be using, copper piping won't be very effective. Most of the thickness would be wasted because the skin depth is so small.

I made some litz wire from some magnet wire I had lying around, it's 12 feet of 60x26 or 28 awg. I tinned the strands on one end one by one, but now the wire's all tangled so I have to separate them all and somehow twist them together without making a huge mess. I don't know how to do this yet without finding 2 others willing to hold wires for an hour (not gonna happen...), but I'll probably figure something out.

Forty, what difference does twisting the wire make?

klugesmith

Tue Jan 24 2012, 08:14AM

Registered Member #2099 Joined: Wed Apr 29 2009, 12:22AM
Location: Los Altos, California
Posts: 1716

Simply bundling many strands of magnet wire won't have the desired "litz wire" effect. Straight or twisted, the surface strands carry more current than the central strands, if the bundle diameter is much larger than the skin depth.
Here's a non-wikipedia reference explaining that the bundle must be formed so each strand alternates between core and surface locations.

I imagine you could get close by twisting strands into a yarn, then twisting yarns into a rope.

Don't use needlessly fine stranding. Skin depth in Cu at 0.25 MHz is about 0.13 mm or 5 mils, right? About equal to the radius of round 30 AWG wire, which would have perhaps 1.03 times its DC resistance. About half the radius of round 24 AWG wire, which would have about 1.2 times its DC resistance. (according to Figure 4-5a in Ramo-Whinnery-VanDuzer, Fields and Waves in Communication Electronics).

But these guys

seem to know what they're talking about, and recommend 42 AWG for that frequency range. I can't explain the discrepancy. Also check out this, at a subtly different domain name:

Their 6 AWG equivalent for 200-350 kHz has 4200 strands of 42 AWG, in a Type 5 construction organized as 6(5X5/28/42).

In a transformer winding, I think proximity effect may be as important as skin effect. Not sure of the consequence when many tightly packed turns have equal currents in the same direction. This reference gets down to the nitty-gritty for practical litz wire construction and flat-strip windings.

m4ge123

Tue Jan 24 2012, 11:29AM

Registered Member #4118 Joined: Mon Oct 03 2011, 04:50PM
Location: MD
Posts: 140

Thanks for the info, kluge.
According to my calculations, 42AWG has 1.351x the dc resistance at 250KHz. However, I changed my plan to use a lower frequency, about 130KHz. Should I make smaller litz bundles and then twist them together, like those sites' type 5, or is it sufficient to just twist them all together?
As for the proximity effect, it should be fine if I separate each turn by about a diameter of the wire, right?

Sulaiman

Tue Jan 24 2012, 01:26PM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3140

From the little experimenting that I have done in this area,
at 250kHz 'skin effect' is important,
and because the calculations are simple it's much discussed
BUT
proximity effect is more significant than skin effect, it's not discussed so much because
there's no simple mathematical formula to plug values in and get an answer.

For a single strand wire formed into a single layer coil spacing the wires by one wire diameter is about optimum, with Litz wire each strand in the bundle is subject to the other strands with respect to proximity effect!
Normal multi-strand wire is actually almost as good as Litz due to surface oxidation of the individual strands, so you could try welding cable etc. which may require an overall larger csa but would be MUCH cheaper than proper Litz wire.

Steve Conner

Tue Jan 24 2012, 02:01PM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

I've always used the fancy wire sold by Maplin as speaker cable, for wiring up Tesla coil tank circuits. It has hundreds of fine strands and seems to work OK. But I've heard from others who melted the PVC insulation off it with excessive RF current.

klugesmith

Tue Jan 24 2012, 03:43PM

Registered Member #2099 Joined: Wed Apr 29 2009, 12:22AM
Location: Los Altos, California
Posts: 1716

Back to what Sualiman said early on, have you considered a flat copper strip with 1 turn per layer?

Can someone please help me make the numbers match?
litzwire.com and litz-wire.com both recommend 42 AWG strands for 250 kHz.
m4ge figured the AC/DC resistance for that gauge to be more than 1.3, which IMHO means that gauge is not fine enough.
But I figure the skin depth (1/8 mm) to be about four times the wire radius (1/32 mm), so the AC/DC res. ratio is obviously 1.00 and the gauge is finer than necessary.

Perhaps the litzers (or their uber-sophisticated customers) are derating the wire
to allow for the current at higher-harmonic frequencies in SMPS's. Then why don't they say so?

thanks

Patrick

Tue Jan 24 2012, 04:08PM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

Klugesmith and sulaiman are right to point out the proximity effect, this is sooper--dooper important int SMPS transformers and non-isolated inductors.

that may explain some of the different numbers your seeing quoted in strands, resistance and awg.

Forty

Tue Jan 24 2012, 07:02PM

Registered Member #3888 Joined: Sun May 15 2011, 09:50PM
Location: Erie, PA
Posts: 649

The twisting helps even out the amount of exposed surface area of the wires (properly made litz wire has approx the same amount per strand.) When I go about making it, I go outside with a drill, the wire, and some wire cutters. loop the wire back and forth (or tie each strand off if it's already in pieces) between two fence posts or trees or something. when that's done, grab the bundle at one end, cut it off of the post, tie it in a knot, and stick the knot in the drill. when you get the desired number of twists per length, set the drill down, walk to the other end, cut it off of the post. If you need to use >10 strands in your final cable, I would probably repeat this process several times with smaller bunches, and then twist the bunches together in the opposite direction (as you would with rope.) of course if you only need ~6ft of the stuff you could stay inside and use bed posts or a vise or something for anchoring the wires.
This is certainly not the optimum way of making litz wire, but it is way cheaper than buying it.

previous 1 2 3 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob