Litz wire for induction heater 2

[asimpson]

I need to make up flexible cable for induction heater coil and I would like to work out best configuration of Litz Wire to minimize losses.

1 kHz
100 A
1000 mm cable length.
copper wire

Worked out skin depth is 0.21 mm for Cu at 100 kHz should I then use wire of maximum 0.4 mm diameter?

Where could I easily obtain small quantity of suitable wire? Suppliers I have talked only do large orders of custom product.


Thanks

 

[IRstuff]

Skin depth would seem to account for only one dimension of a two dimensional problem? Does a 0.4mm diameter wire provide sufficient cross-sectional area to achieve the desired oeprating conditions?

TTFN

FAQ731-376
Chinese prisoner wins Nobel Peace Prize

 

[VEBill]

Litz wire is many strands isolated from each other. So N wires at 0.4 mm diameter each where N is chosen to meet the 100A capability.

The list shows "1 kHz". Should be 100 kHz?

100A in a flexible cable is a bit of a challenge to start with. By the time one uses isolated wires it's going to be a pretty thick cable. And that make flex an issue.

What about wide flat conductors?

What about putting a transformer on the other side, so that the current is reduced (and the voltage is increased)?

 

[waross]

Have you worked out the losses for a length of 3/0 or 4/0 AWG building wire? If that shows improvement but not enough, go a size or so larger. Then compare the cost of Litz wire to oversized off the shelf wire.
Consider also 1/2 inch type K copper tubing. Dimensions:
Diameter 0.625 inches
Thickness 0.049 inches
The area of the copper is about 0.076 square inches.
At the old rule of thumb of 2000 amps per square inch that gives you about 150 Amps ampacity at 60 Hz. I'll leave you to work out your losses.
3/4" is:
Diameter 0.875 inches
Thickness 0.065 inches
The area of the copper is about 0.17 square inches.
That will give a 60 Hz ampacity of about 340 Amps.
At this current level, your Litz wire may not have a linear ampacity. The current in the outer layers may well choke the current in the inner layers enough to skew your calculations.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[IRstuff]

I'm not sure who's reading what or doing what, but at 100A and 0.21mm skin depth, you'd need a 6-in circumference to get meet the 2000A/in^2:

(100A/0.21mm) / (2000A/in^2) = 6.04 in, which would be a 2-in diameter wire.

TTFN

FAQ731-376
Chinese prisoner wins Nobel Peace Prize

 

[VEBill]

That's where Litz wire provides the way forward. By breaking the conductor into many isolated strands, correctly woven, it can make more efficient use of the copper.

It's still going to be a thick cable due to the added insulation layers.

 

[sreid]

Try MWS.

http://www.mwswire.com/litzmain.htm

 

[waross]

Skin effect is dependent on both frequency and current. Electronic folk tend to dismiss current and focus on frequency.
As the currents rise above about 100 Amps, 60 Hz, skin effect, and eddy, (both in the conductor and surrounding iron) become significant. Another factor with higher currents is heat rejection.
At higher currents the power folk have found the most economical solutions to be;
laminated flat strips,
multiple separated conductors, (mostly for heat rejection),
and
hollow conductors.
I.P.S. bus is copper or aluminum pipe manufactured to standard Iron Pipe Sizes.
At 100 amps, you are near the threshold for issues at 60 Hz.
You may save some money by considering a power distribution solution and checking out the usability of various dimensions of copper pipe.

Bill
--------------------
"Why not the best?"
Jimmy Carter