who can suggest a switch supply controll chip? frequence >500Khz?

[DreamCat]

I want to DIY a High-frequency heating electric soldering iron, I'm very interesting it.

there is many this type handle selling, so I think it is easy.

Reguards!
DreamCat.

[solutions]

I took the same attitude as you appear to have when I first started looking into switchers. 

While higher frequency makes for smaller inductors, it makes for huge losses in the MOSFETS and their drivers, and in some cases the core losses in the inductors get awfully huge as well.  Despite the fact that you can buy switchers all the way out to 5MHz, the sweet spot still seems to be a to be about 100kHz or so if you are pushing any kind of current.

Of course, in your application you actually WANT heat, so you may not have the same issues that those of us who do not want self soldering circuits do 

[DreamCat]

yeah,  you are right. higher frequency will make huge loss for MOSFET.

I should consider the frequency.

[phasetek]

I think he is trying to design an induction heated element at high (RF) frequencies?  Similar to what metcal is using, theirs operates at approx 16mhz, where a coil is excited with RF that induces heat in the tip.  Several companies have tried this at lower frequencies that do not work as well.  The advantage is that there is no mechanical contact between the tip and the heating coil required (a common failure point in many irons), the energy is induced in the tip.  The alternative is a standard PID type control of a conventional resistance heating element.

[solutions]

An induction heated tip would be interesting, though I'd think pretty inefficient.

He did say "switcher", versus "RF power amplifier" which is what you need for an induction heater - a big one.

[DreamCat]

phasetek is right, but not very high frequency.
I heard someone use ATX supply's 3.3V output to modified Induction heating, just a few turn coil, and connect to 3.3V before the rectifier.
ATX supply only close to 30kHz. So I think I no need too high frequency for it.

here is some link:

http://www.penguinslab.com/induction.htm

http://hackedgadgets.com/2007/06/22/induction-heating-demonstration/

I also found there is a patent:
http://www.freepatentsonline.com/5408072.html


please search the "High-frequency induction heating electric soldering iron".

[carbontracks]

500KHz isn't that fast... even at 1MHz and beyond, good (>%80) efficiency is easily obtainable, even with switching losses, so long as your devices are fast.  If power density is a concern, then you'll want to go as fast as possible; otherwise, might as well go at 100KHz or something slower.

But if you want to do induction heating, then standard switching controllers probably won't help you.  You're likely looking at a class E amp running open loop.  Your main concern will probably be coupling the energy to the tip (if you're designing that as well).

[DreamCat]

I decide use pwm controll the power supply to adjust the max energy.
a fixed oscillator works at 100khz use to provide pulse, the temperature control by the senser in handle.

btw, there is a importent problem for me,  as a given handle, how can I choice the suit value of capacitor for the resonance on induction in handle and capcitor ??

sorry for my bad english.

[carbontracks]

Why does the capacitor have to be in the hand piece?

And the value depends on what type of driver you're using.  You'll need to be more specific as to how you're exciting your coil.

Also, if you're going for a system like metcal's smartheat system, then you should know that they don't have any temperature feedback at all.

[DreamCat]

Why does the capacitor have to be in the hand piece?

And the value depends on what type of driver you're using.  You'll need to be more specific as to how you're exciting your coil.

Also, if you're going for a system like metcal's smartheat system, then you should know that they don't have any temperature feedback at all.

maybe you are wrong, the smartheat system has a temperature sensor in iron.
the handle have four line , one pair connect to the induction, and the other pair connect to a temperature sensor, maybe it is a RTD.

the capacitor and coil induction form a resonant circuit. but the capacitor isn't in handle.

[carbontracks]

No, I'm quite sure there is no feedback.  The handle in the SP200 series has three lines (but I'm not sure what the third does).  The MX500 series has only two lines.  The output temperature is set by setting the curie point of the tip material.  There is no feedback, at least not in the typical sense, but the output is still regulated.

But I doubt you'll be able to choose the curie point of your tip material, so I doubt this really matters for your application.

[DreamCat]

“The output temperature is set by setting the curie point of the tip material.  ”

this is nice idea.  thanks! let me think how to build the curie.

-----------------------------------------------------------------------------------

I'm wrong, the method of heating you talked  isn't induction heating,  that is PTC heating.

[solutions]

The curie point is set by choosing the magnetic material composition.  Temperatures higher than the curie point cause the material to lose its magnetism (without magnetism, no eddy currents, no heat).  There's probably a handbook or web page that has a table of curie points for different magnetic alloys and materials if you Google hard enough or someone here has been there, done that.

I'm guessing from basics is that material resistance (skin effect - depth and resistivity with frequency) has a fair bit to do with your design choices as well.

Sounds like a fun project - I already have, ahem, "too many irons in the fire", so I'm adding an induction furnace to my projects to-do list, so please keep us posted on findings and progress.

[DreamCat]

ok!
I will write here every step that I done once I bought the material all my needed. but now I only can consider and do the simulate of some part of circuit.
The Induction Heating base on skin effect. there are many working projects on net.

There is a big problem, the handle in market all have two pair lines, one of them use to te temperature feedback and control. so I only can use this method control and adjust temperature.