Hello. I joined here to look for some insight ... I am a hobby blacksmith, and, as such, my electronics knowledge is not the greatest. However, since we are in the future, I have discovered there is such a thing as induction heating and the use of this technology by other blacksmiths. The sort of induction heater often used is the Chinese-made 15kW, 30-80kHz type - LH-15A is the model name, I believe. I can provide a link for reference. Buying exotic commercial machines from China is out of my price range, however, and I'm looking into alternatives.

One that I have found is a little board with a coil described as "flyback driver, ZVS - 1000w". I understand the schematic to this type somewhat, but I do not know how robust this could be. From what I understand, it's probably a much higher frequency (probably ok for my purpose), has a much smaller tank and probably less mosfets. I don't need this thing to run 24/7, but I would probably put many heats through this (4-5 hours of use a week) and heat stock no larger than 2-3" in diameter. I just need it to suffice. The one that I am looking at has no power supply - I am capable of rigging up some sort of DC power supply and have AC220v/60a readily available in my shop. I can make a water-cooling and air-cooling system.

I'm curious if this would work for my purposes... and, if not, might it be modified to suit? This would be a huge money and time saver for me. Plus, 3000F coal fires in the summer aren't that comfortable!

Thank you,

Kevin

 

IMO, a 1kW unit will be vastly under-powered for heating 2-3" stock (or even 1/4" stock!) .

 

Here is a link of a small one you maybe able to increase power for http://kaizerpowerelectronics.dk/general-electronics/royer-induction-heater/
For continuous use, you need to water cool everything including the capacitors, which should be good grade HF style.

 

Alec_t - what size do you think would be sufficient/practical? If I can heat something 1" in 30 seconds or so then that would satisfy my needs. I don't think I would need the feature to change frequency, either. I mean, clearly, I don't really know what I'm doing but if I were to build one I'd only like to spend $1-200 on parts. I have a ton of junk, so heat sinks, transformers etc. common stuff I have available already.

Max - I was planning on water cooling. Thanks for the link - trying to read as much as I can but wrapping my head around the abstract is very intensive at times - as I'm sure you understand!

 

The heater Max linked to claims to heat a M10 20mm bolt to red/white hot using 650W, but it's not stated how long that takes. Scaling up from M10 to 1" diam stock (but still 20mm length) implies a power increase of at least (1"/10mm)^2, i.e. ~6.25. That brings the power up to > 4kW. I guess for stock longer than 20mm there would be more heat loss through conduction away from the heated zone, so you could be looking at >10kW needed. I must emphasise that's only my opinion. I've no practical experience with induction heaters.

 

There are several examples on utube of the Royer HF units, here is one rated at 3kw using a concentrated coil.

Max.

 

I'm usually trying to cool things rather than heat them, but the physics are the same. The material properties (specific heat, melting point, etc.) should give you enough information to calculate the bare minimum power required to liquify it. That number must be increased to account for coupling efficiency, radiative and conductive heat loss, and anything else that keeps the heat from reaching the part, but the theoretical limit is a cheap and reliable starting point to qualify potential solutions.

ak

 

The internal construction of MOSFETs with a higher voltage rating makes them unsuitable for use in a self oscillating circuit like this Royer oscillator.

Utter nonsense! --- While devices specified for V(br)ds>200V exhibit higher Rds(on) characteristics -- there is no intrinsic inapplicability to Royer/'Royer-esk' topologies! -- I, for instance, have employed a pair of (Infineon) IPW60R126c6 (V[br]ds=600V, Rds[on]=125milliohms) in my radiography 'generator' (implemented via a 'modified' resonant Royer oscillator operating at a sustained input current of 50 Amperes) for over a year (average 100Hr/Week operation) with excellent results...

In fairness to the author, I have not evaluated the potential effects of the increased Rds(on) figure upon his application -- My point is merely that such devices are applicable to self-oscillating power conversion systems, and, indeed, power oscillators in general!...

Please forgive the "T crossing" and "I dotting" --- 'Tis merely that I have little patience with 'declarative ignorance' and, sadly, there's Waaaay to much of it in the HV/'Power Conversion' community! ---

Best regards
HP

PS -- Lest this 'goes astray' (owing to language barriers, my own bad grammar or whatever) -- Please note: the above diatribe is not aimed at anyone here! -- But, rather, at the irresponsible, offhanded remarks of the author on the linked site..

 

One thing that the T/S (thread starter) should keep in mind is that one heating coil doesn't work for every size part. There is a ratio involved with the coil diameter to part diameter. Too large of a part in too small of a coil changes the induction and efficiency of the coil. If you look at the many online and 'how it's made' shows, you'll see that the coils match the size of the part to be heated.

And no I can't find my source of information at the moment, just a remembrance from when I was looking into building one, years ago.

 

I used to maintain HF heating equipment and one application had a 10 turn coil but that was the primary for a one turn secondary one turn work coil for very local heating used in Silver solder app.
Max.

 

One thing that the T/S (thread starter) should keep in mind is that one heating coil doesn't work for every size part. There is a ratio involved with the coil diameter to part diameter. Too large of a part in too small of a coil changes the induction and efficiency of the coil. If you look at the many online and 'how it's made' shows, you'll see that the coils match the size of the part to be heated.

And no I can't find my source of information at the moment, just a remembrance from when I was looking into building one, years ago.

Yes, I do understand that. I went to school to be a plumber, so forming copper pipe, soldering and thinking up joints isn't new to me. This is something I'll have to experiment with later on - but I need to find something practical first.

The 3kW unit that Max linked to seems pretty suitable to my needs. 60v DC/50 amps? I do not want to melt my work - rather, bring it up to a white heat consistently and quickly. That's about, I dunno, 2000F? 2100F? Somewhere around that temperature. I should note that most of my work will .5" - 1.5", but I would prefer the capability of 2-3" (with specific coils, yes).

 

When thinking about this kind of thing for myself, the thoughts of modifying an old Lincoln tombstone welder was where I was headed. Then saw a story about DIY gas forges and never looked back. You can heat anything that will fit in the opening no muss, no fuss. Not quite ass fast as induction but much more versatile.

 

Propane is roughly 8 times the cost of induction and gives off significant waste heat.