• Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Induction Furnace 1 - 3 kW

Not open for further replies.


New Member
Induction Furnace 1 - 3 Kw Information Please !

Hello ElectroTech Community

I haven't had any luck with my post about a 10 - 12.5 kW Induction Furnace, so lets scale it down a bit to 1 - 3 kW.

There are plenty of people placing their achievements on YOU TUBE, but they don't give the circuits.

May be if I build a smaller one first, this may help in the overall picture

My Mentor is an Electronics Engineer ex Royal Australian Navy who did his Honours Degree part time [ Hard Yakka]. He is letting me do the ground work myself as this will be of more benefit to me.

Note: This is NOT a Commercial project. I am nearly 63 years of age and I want to start doing some things for myself for a change.

Current information.

1. Incoming Power 415V 3 Phase 50 Hz 16 Sq mm Mains.

2. I have collected an number of components so far eg

1000V 75A IGBTs,
40 A 3 Phase Bridge Rectifier,
High Current Diodes
High Current SCRs

Refrigeration components for the cooling of the heat sunk electronics and Induction Coil and many more smaller components.

3. A foundary friend of mine runs a new commercial 125kW [Cast Iron] and 75 kW [Non Ferrous] Induction Furnace for a Technical College. The operating frequencies vary between 2.5 -3.5 kHz.

Question ?

Has anyone else built a unit in the 1 - 3 kW Range ? as I would like to contact them.

Kind regards



A basic induction circuit is very simple really. Rectify the mains to DC, Use the IGBT'S as switches driven from a frequency generator, like a 555 timer. If you could link a PDF to the IGBT's you'll be using that's a good place to start, as most of the worrisome details are in the switching section. At 415volts you'll only be conducting 2.5amps of current for 1000 watts, which is perfectly reasonable for the rough specs you've listed for the components you have on hand.

Also exactly what you're trying to DO will shape how you proceed quiet a bit, are you trying to do a mini foundry, or just obtain heat treat temperatures on a target part?


New Member
Hello Sceadwian

Thankyou for your reply and information.

What do I want to do with this ?

1. I initially wanted to build a 10 - 12.5 kW Induction Furnace as I laid out in my original posts, however I didn't receive any replies in nearly a month.

So then I thought I make have the wording wrong, but was advised by another member in a PM that it all seemed OK to him.

So then I decided. well, I'll scale it down a bit and build a smaller one first to get some experience.

I can use a smaller Unit for specific heat treatment of components as you said, but the larger Unit I want to melt Ferrous & Non ferrous materials for Castings.

As I said in my Post "Note: This is NOT a Commercial project. I am nearly 63 years of age and I want to start doing some things for myself for a change".

I have a lathe, universal Milling Machine & a Cincinatti Tool & Cutter Grinder for making anything I want.

Thankyou once again for your reply.

Kind Regards



New Member
Hello rfranzk

Thankyou for your reply and the link you provided.

As you will see in my reply to Sceadwian, You will see what what I am up to.

I really want a 10 - 12.5 kW Induction Furnace eventually.

All the components I have scrounged so far are High Voltage High Current components that I have acquired from a scrap metal yard.

Any once I have my Ground Work done, I will be posting my progress.

Kind Regards



At 415volts you'll only be conducting 2.5amps of current for 1000 watts, which is perfectly reasonable for the rough specs you've listed for the components you have on hand.
This is not true at all.
within the tank circuit there is 5-30 times as much current or voltage as there is flowing in the work piece, and the inverter has to supply the losses as well as the load.
50% efficiency used to be pretty good. I can get better than that if i use a lot of copper in the work coil, but then there's a lot more VARs circulating in the tank.
Within the inverter there will be at least 1.41 times as much peak current as there is RMS current, and that's assuming a transformer driven series resonant circuit that stays in resonance.

In theory a 1KW inverter is easy, and indeed it is, but perhaps not with scrap IGBTs.
If you can give me a part number i could tell you if they will work or not.

meanu, what are you going to use for the capacitors?


New Member
Hello Johansen

Thankyou for your Reply and Information.

As I put in the post to Sceadwian,

I really want to build a 10 - 12.5 kW Induction Furnace, however, I then thought it might be better if I start off smaller first and get some experience.

The Capacitors I haven't any at the moment, but am open to suggestions for the type required.

The IGBTs are out of X Ray Machines and can I send you the Numbers in another post as I have a few and will have to round them up in my shed.

Thankyou for your reply once again.

Kind Regards



@meanu there is a seller on ebay that has capacitors sufficient for a 10kw induction heater and it will probably cost you about 45$ just for the capacitors.
however i can't recommend them until either i buy one and test it (i'm intending to within about 6 weeks) or the seller replies to my inquiry.

in any case, a 10kw induction heater isn't a trivial project, neither is a 1kw unit.

here is a few things i've run into:
1)if you make the whole system too efficient then its too easy to blow up.
This means you're forced to build a PLL to keep it happy, as well as peak current detectors and glitch detectors to keep it happy.

2)I am already looking to sensing the tank current with a rogowski coil rather than a current transformer, because methods of peak current detection are sensitive to noise, (and putting a current transformer on the tank isn't acceptable). RC snubbers sufficient to dampen the current waveform (the square wave out of the inverter has to drive the parasitic capacitance of the transformer and that's essentially hard switching) introduce a phase shift and this means its not kept in resonance. If you provide extra inductance between the inverter and the transformer, then you increase the inverter KVA required, and that inductance resonates with the parasitic capacitance of the transformer. -it ends up being a mess.

3) A wide range of loads (iron, or did you want copper melted?) require a variable voltage power supply set to constant current mode (as limited by the inverter)

4) it is often desirable to be able to switch the transformer ratio at will, unless you wish to just over engineer the transformer and use a 30kva transformer for a 10kva tx.

regarding the last point, the only way i would even consider building an induction heater today is the transformer fed series resonant tank.

take the work coil, capacitor and transformer in series. change the turns ratio to match the current and voltage of your inverter.
at resonance, the inverter current is a sine wave, at 100% power factor.
turn off and turn on losses are thus mostly negligible.

regarding the comment that it is easy to blow up:
in theory, if you drive the inverter from an ideal current source, as long as the switches can handle hard switching at leading or lagging power factors, you Cannot blow anything up.

the problem is if its too efficient the energy stored in the tank will be dumped out back into the power supply if the frequency changes, and it may be sufficient energy to blow up the inverter.
to solve that problem you can over engineer the inverter or just connect more capacitors across the inverter.. and all of a sudden, an ideal current source is buffered by rather large capacitors, and it is no longer ideal.

another problem:
assuming your inverter is a two level square wave inverter, supposing there is a glitch and the high side switches don't turn on (high side driver doesn't have enough voltage on the bootstrap capacitors?) then you get a dc bias in the transformer.
there's two ways to solve that problem.
use a transformer to drive all the switches properly. or. connect the inverter to the transformer with a capacitor in series.

you are looking to build a 10kw system, this means you need to build the inverter for 20kw, and a capacitor rated for the dc voltage feeding the inverter, and enough microfarads to provide a relatively low impedance aren't cheap, unless you buy them on ebay for $20. again, i would provide links but i only know of a few sellers and i'm not about to go scrolling through 200,000 capacitors looking for coke can sized film caps. I can go into detail on how to identify the capacitors you need for the tank coil, the dc link filter, and the ac coupling between the transformer and the inverter if you need more clarity.

that said, if you are willing to spend three hours soldering lots of small capacitors together, don't throw out the idea of using a lot of small ones.

here is a photo of a 1 kw unit, capacitors, transformer and work coil
look in the parent directory for build photos.

i'm about to build a proper tutorial and throw it all on my website, so don't take any of this personally....
let me know if you need any more help.
I am looking to throw together a decent PLL and gate driver into the open source community/provide boards if group orders persist.


New Member
Hello Johansen

Thankyou for your second Reply and Information.

You seem to have gone to a lot of effort to help me. Thankyou kindly for that.

Your knowledge of electronics is far greater than mine, I have to start somewhere.

I managed to acquire a basic circuit for a 75 kW Unit using SCRs for switching.

I am going to see my Foundary friend on Wednesday [EST Australia] and he is giving me access to the complete Manual for the new 75 - 125 kW furnace.

I will let you know how I go, here, or is it better to send you a PM on the non technical Information ?

Thankyou once again for your efforts.

Kind Regards



New Member
Hello Johansen

I've got a couple of Pictures [Basic schematics of a SCR 75kW Furnace and will post them as soon as I can work out how to do it.




New Member
Hello Johansen

I have a couple of schematics of a 75kW SCR Furnace and will post them as soon as I can work out how to do it.




Well-Known Member
Most Helpful Member
will post them as soon as I can work out how to do it.
Got to Forum/Site Issues & Feedback and read the second Sticky.


interesting. i can't say i would even attempt to build this circuit today.
IGBTs are cheap, and the scr inverter above Requires a controlled rectifier because if for some reason the inverter scr's both trigger on, you'll either blow the circuit breaker, or kill the scr's.

i have a feeling that a thyristor based inverter would be great for induction heaters intended to keep say, 1 ton of steel liquid in a steel furnace but not much else. (like heating a 4 foot diameter oil pipeline so you can bend it?)
note that the inverter in the schematic you posted has to handle the full reactive power of the resonant coil.
if you're heating anything other than iron then you will have to overbuild the induction heater significantly if you try to drive the coil directly.
its a lot easier to couple the resonant coil to the inverter and only pass real power, the tradeoff being you need a transformer to do it.
occasionally these transformers can be found on ebay, but you can make them yourself for a lot less if funding is a problem.
Not open for further replies.

Latest threads

EE World Online Articles