1
\$\begingroup\$

In an induction heater one of the main problems is that the coil carries a large current. This causes the coil to become hot. There is a limit to how large diameter the wire can be, because it needs to have multiple turns (I think). The standard solution to the problem is to use hollow tubing, then run water through the tubing, thereby cooling it.

Is there a better option for making the coil? The overall goal is to make a coil that can transmit large amounts of current without getting hot.

One obvious improvement I can think of is to make the coil out of silver. Silver is both more electrically conductive than copper and also more thermodynamically conductive, so it will be able to transmit electricity better and be less susceptible to heating up.

Is there an even better way, maybe superconductors?

\$\endgroup\$
  • \$\begingroup\$ Gold is even better. Superconductors need to be kept cold so using them rather defeats the objective, you might as well go back to copper and use your liquid nitrogen or helium to cool the coil. \$\endgroup\$ – JIm Dearden Aug 1 '16 at 16:43
  • \$\begingroup\$ I think you would be surprised how slowly heat can move through metal via simple conduction. Since the water is physically moving the heat out (via actual material movement) you can get rid of a lot of heat quickly. The only thing that I can think of that might be superior is to increase the surface area by embedding the coil in thermally-conducting material like mineral oil or something. \$\endgroup\$ – Daniel Aug 1 '16 at 16:54
  • 4
    \$\begingroup\$ @JImDearden silver is a better conductor than gold last time I checked! \$\endgroup\$ – Andy aka Aug 1 '16 at 16:55
  • \$\begingroup\$ Define "better". Copper and water are both cheap and easy to work with. So what dimension do you want to improve? \$\endgroup\$ – Dave Tweed Aug 1 '16 at 16:56
  • \$\begingroup\$ @DaveTweed Energy efficiency. \$\endgroup\$ – Tyler Durden Aug 1 '16 at 17:26
5
\$\begingroup\$

Commercial Induction Heating Equipment is all about the economics of the equipment and the process of heating.

There are commercial units using super conductors Seen in this post and in This Post

Also, many uses of Induction Heaters are one of a kind for a specific product. Fabricating exotic induction coils for a one time sale might not make sense.

Induction Heating metals of low conductivity (aluminum and copper) would be candidates for using super conduction coils. Induction heating of copper components using copper coils results in efficiency less than 50%.

Heating ferrous objects (below the curie temperature) is probably not a good candidate for exotic induction coils since the efficiency could be quite high (85% to 90% efficient). However in multi-megawatt systems running continuously could save lots of power and $.

The efficiency gain of using silver rather than copper would likely still require water cooling.

Flux concentrators , see Fluxtrol also help with efficiency of the heating process. (not an endorsment for fluxtrol).

Flux Concentrator taken from Induction Presentation enter image description here

The wall thickness of the copper tubing can be made thinner, so as to increase cooling. But, decreasing wall thickness begins to harm the efficiency (skin depth limitation).

\$\endgroup\$
  • \$\begingroup\$ That is pretty interesting. What about using Lintz wire as Andy Aka suggests below? I guess the issue there is that the wire insulator would have to be heat resistant. \$\endgroup\$ – Tyler Durden Aug 1 '16 at 17:33
  • \$\begingroup\$ As Andy Aka shows, efficiency can be improved by using Litz wire. If water cooling was still required, some type of hose or jacket would be required around the Litz wire to contain water, and possibly increasing separation between the heated load and Litz wire (coupling to load). However, the thinness of the Litz would help in heat removal. \$\endgroup\$ – Marla Aug 1 '16 at 17:53
1
\$\begingroup\$

One obvious improvement I can think of is to make the coil out of silver.

First of all conductivity tables: -

enter image description here

Silver conducts about 5% better than copper so not much of an improvement here. The better option is Litz wire because this counteracts something called skin effect.

Skin effect happens when AC current passes down a conductor - magnetic fields induced in the conductor tend to force current to flow at the periphery of the wire and so individually insulated multi-strand wire improves things. Skin effect: -

enter image description here

As you can see, most of the AC current flows at the periphery - this is why copper pipes work so reasonably well - the centre of the copper would hardly conduct anything so it can be filled with water for cooling purposes. Here's a table of skin depth versus material type versus operating frequency: -

enter image description here

This table is the backbone for designing Litz wire - depending on operating frequency, the individual strand diameters can be determined i.e. at 10 kHz there is little point in choosing a copper conductor that has a radius bigger than ~0.6mm.

Here's a good example of how increasing the number of strands is a benefit: -

enter image description here

So for the higher graph the construction is 16 strands of 0.2mm diameter. This is a cross section of copper of 16 x pi x 0.2 x 0.2 / 4 = 0.502 sq mm. For the next lower cable the cross sectional are of copper is the same at 25 x pi x 0.16 x 0.16 / 4 = 0.502 sq mm and the same all the way down to the wire made of 255 strands of 0.05 mm diameter wire (0.501 sq mm).

In conclusion, Litz wire is probably the best way to go but by how much depends on the operating frequency.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.