Induction heating DIY

[Peter.]

Has anyone here dabbled in DIY induction-heating? I'd like to hear of anyone who has successfully (or un-successfully) tried this.

I have an idea that requires the heating of the end of thin-walled steel tube to silver soldering temperatures. Typically the last 25mm of a 100mm tube of 2-3mm wall thickness but diameters will vary. Needs to be a fairly portable rig and temperature control can be as crude as an on-off foot switch. If it would be possible with 3KW of power then all the better.

[lazlo]

It's relatively straightforward to heat up a part with an oscillating magnetic flux -- there are dozens of them on Youtube.

It's much, much harder to do it practically (with an adaptive resonant circuit). This guy used a PLL chip, but as you can see, the setup is not trivial:

http://www.youtube.com/user/kynanisk.../4/-y9whTDieTk
http://www.youtube.com/watch?v=yXq2yls8IRY

Buried in one of the video descriptions he's got a link to his web page where he describes the circuit design.

[small.planes]

try posting a ping on usenet group uk.rec.models.engineering for Andrew Mawson. He has an induction setup that he built. He is somewhere down south in the uk.
Dave

[Herm Williams]

I am interested in a small unit too. I would like to use hi power mosfet driving scr for the power but lack knowledge to design the circuit. I have spare 20 amp 1 kv mosfet's if someone would care to share (trade info for hardware) diagrams.
re

[Peter.]

It's relatively straightforward to heat up a part with an oscillating magnetic flux -- there are dozens of them on Youtube.

It's much, much harder to do it practically (with an adaptive resonant circuit). This guy used a PLL chip, but as you can see, the setup is not trivial:

http://www.youtube.com/user/kynanisk.../4/-y9whTDieTk
.

You're not kidding - looks like enough gear to split the atom. Surely the same must be possible by a much more basic path - even if it's less efficient.

Dave thanks for the reference I'll do a search on Google groups.

[Willy]

Peter I posted a thread a couple of years ago looking for info on the feasibility of a DIY induction heater.

http://bbs.homeshopmachinist.net/sho...6941&highlight

There seems to be a lot of interest in the subject as the thread has well over 12,000 views to date but not too much for viable solutions.
I'll be watching your thread with interest as I'm sure others will as well.
Hopefully someone has a workable unit that can be built without a degree in high frequency electronics.

[Ries]

This comes up from time to time, and the answer is that to do it right, its not easy or cheap.

It is comparable to building your own Tig welder with High Frequency, or making your own overhead cam V8 engine.

Both are certainly POSSIBLE in a home shop, and both have rarely, very rarely, been done by individuals on a shoestring budget.

here is a long discussion of the subject, with a variety of links to working models and maybe not-so-working ones.

http://bbs.homeshopmachinist.net/showthread.php?t=37861

[Rich Carlstedt]

You fellows should go visit your local auto repair shop, or a muffler shop.
Snap-on Tools already has such a unit.

Its portable , about the size of a Saws-All and is used for stuck nuts.
The induction coil (different sizes) is placed around the nut like you would a wrench, and the trigger pulled. After 10 to 15 seconds the nut will glow red to yellow and you spin it off. They use to torch them off, but that is dangerous around a gas tank...

Rich

[RWO]

It would be much easier to simply resistance heat it. Almost any welding machine should be able to put out enough current to heat it up. Simple spring-loaded copper block contacts machined to fit a short section of the OD of the tube, one on each side, and turn on the current.

RWO

[Peter.]

I've clocked one of them on eBay Rich, looks like the perfect tool for bolt heating but I think I need more power than that unit would produce. There's also a nice 4KW unit on canadian eBay which would be handy because they use the same mains voltage as us but at £2.5k it's a bit much for an experiment.