induction heating faq? beginners' guide? / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

induction heating faq? beginners' guide?

1 2 3 next

Move Thread

LAN_403

jshine

Tue Oct 28 2008, 02:49AM

Registered Member #1784 Joined: Tue Oct 28 2008, 02:30AM
Location: Rochester, MN, USA
Posts: 14

I've been looking into induction heating as a part of a larger project. While I don't have much experience with solid state inverters, my needs are modest -- 500 watts of output should be plenty. I've been doing some reading on the subject, particularly at

, and the part of the circuit that confuses me most -- the power supply & oscillator circuitry driving the MOSFETs -- is missing.

I was hoping that someone would be kind enough to point me towards a modest and simple HF power supply that can drive an LCLR work coil. Based on my understanding, about 200 kHz is ideal, and a higher-voltage/lower-current supply simplifies construction (that's what the above website claims, in any case).

Also, I was wondering if anyone has ever attempted adapting an existing switching power supply for this task? Computer power supplies are cheap, readily available, and capable of handling a few hundred watts. If they operate at a decent frequency, is there any reason that they could not be used to drive an LCLR work coil?

Any pointers on this subject will be greatly appreciated!

Thanks in advance,
-Jon

jshine

Fri Oct 31 2008, 02:53PM

Registered Member #1784 Joined: Tue Oct 28 2008, 02:30AM
Location: Rochester, MN, USA
Posts: 14

After some further digging, I ran across the following web page (in Czech):

http://danyk.wz.cz/induk2.html

The schematic is based on a commercial driver IC which greatly simplifies things, but I was wondering if someone would be kind enough to explain what's going on between the output of the MOSFETs and the resonant tank part of the circuit.

There's a network of 4 2.2uF/400V caps that doesn't appear in the example circuit shown in this application note (goes with the driver chip, IR2153):

http://www.irf.com/product-info/datasheets/data/ir2153.pdf

Also, just downstream from there, the 4.7 ohm/4W resistor in series with the 4.7 nF/1kV cap is something I don't *really* don't get -- why are those parts across the output? And why does each MOSFET have a 10 ohm/2W resistor and a 2.2 nF cap across it?

If someone with more experience in this work could shed some light on this, I'd be much more comfortable building this. I ordered a few of the IR2153 driver chips off of eBay, but they're coming from Hong Kong, so I've go a few days to think about this...

Thanks!
-Jon

Arcstarter

Fri Oct 31 2008, 05:41PM

Registered Member #1225 Joined: Sat Jan 12 2008, 01:24AM
Location: Beaumont, Texas, USA
Posts: 2253

Well, the 2.2uf 400 volt caps are for the halfbridge i do believe.

101111

Fri Oct 31 2008, 06:13PM

Registered Member #575 Joined: Sun Mar 11 2007, 04:00AM
Location: Norway
Posts: 263

The 2.2ÂµF capacitors are for DC blocking, and to make sure you wont blow up your MOSFETs.

To me it seems like the IR2153 has a inbuilt oscillator and gate drive to the transistors.

Uzzors have made two very sucsessfully projects about induction heating.

the last one is controlled by a PLL circuit that makes it lock on the right resonant frequency. Quite genius. I am currently trying to build the last one myself.

MRacerxdl

Sat Nov 01 2008, 01:10AM

Registered Member #989 Joined: Sat Sept 08 2007, 02:15AM
Location: SÃ£o Paulo, Brazil
Posts: 476

I ever wanted to know how this driver works, it doesnt have any GDT but the reference "floats", I dont know how the IR does it.

I played arround with a induction heater with IGBT's here, but I didnt use any ressoant schemes, only the coil drived by the two igbts, I was get a red screw driver on it.

Arcstarter

Sat Nov 01 2008, 03:35AM

Registered Member #1225 Joined: Sat Jan 12 2008, 01:24AM
Location: Beaumont, Texas, USA
Posts: 2253

101111 wrote ...

The 2.2ÂµF capacitors are for DC blocking, and to make sure you wont blow up your MOSFETs.

Ah, thanks... Still have alot to learn, don't i? LIKE THE NEW AVATAR?!??!?!?!?!!?!

Adam Barkley

Sat Nov 01 2008, 12:48PM

Registered Member #1743 Joined: Mon Oct 06 2008, 03:47PM
Location:
Posts: 1

The IR2153 is one of a family of self-oscillating dual High-side / Low-side mosfet gate drivers. The self-oscillating part works very much like an astable 555 circuit. The pot adjusts the switching frequency.

Several of you asked about the lack of a gate drive transformer (GDT). First of all, driving the low-side gate is easy since the source of the low side mosfet is ground. The high side mosfet has a floating source, which requires either a gate-drive transformer to provide isolation or a floating power supply for the gate driver IC to use. This particular driver creates a floating power supply using a bootstrap diode (BA159) and capacitor (100u / 16V) arrangement.

The bootstrap supply works in the following way:
1) You must first turn the low-side mosfet on (which is easy).
2) When the low-side mosfet is on, its drain-to-source voltage is very low (Rdson*Id ~ 0V), so the source of the high-side mosfet is at 0V.
3) Now diode BA159 (which should be ultrafast or shottky) turns on and charges the bootstrap capacitor (100u / 16V) to approximately 15V.
4) Turn the low-side mosfet off. Diode BA159 turns off and leaves the capacitor charged to 15V. Notice that both terminals of the bootstrap capacitor connect back to the gate driver IC. This is the floating power supply that can be used to drive the high-side mosfet. As the high-side mosfet turns on, its source voltage rises to the DC bus voltage. Since the capacitor is connected to the high-side source terminal, it floats up also. So the gate drive IC has the DC bus voltage +15V available to keep the mosfet on.

This solution is pretty elegant and you get many benefits. You don't need a gate drive transformer. You get dead-time built in to the driver. There are some under-voltage protections to shut the IC down if your bootstrap circuit stops working.

Practically, there are a few problems. You need devices inside the IC rated to the full H-bridge voltage (+15 V). Also, you must turn the low-side mosfet on first or the bootstrap capacitor never charges. To avoid this problem, some ICs have built in minimum and maximum duty cycle limits.

jshine

Mon Nov 03 2008, 04:14AM

Registered Member #1784 Joined: Tue Oct 28 2008, 02:30AM
Location: Rochester, MN, USA
Posts: 14

Thanks again to all who replied for the clarification. I've got another quick question, if someone has the time to answer:

I was able to get a hold of a nice tank capacitor designed for induction heating (the capacitor even has fittings for water cooling!), but it's a smaller capacity than the one used in the schematic that I mentioned above, leading to a higher resonant frequency.

mods washed ur capacitor and it shrunk lol

My question is: do I also need to change the values of the caps in the output stages (the 2.2 nF caps across the MOSFETs, the 2.2 uF filter/isolation caps, or the 4.7 nF cap across the final output -- I've got no idea what that one is supposed to do)? The IR2153 should handle the new frequency, but without fully understanding what the output caps are doing, I'm reluctant to start making changes. On the other hand, as the frequency increases, the 2.2 nF and 4.7 nF output caps could pass more current than the author of this schematic intended...

Also, how critical is the quality of the 2.2 uF caps? These are expensive on Mouser:

Original Schematic (translated from Czech):
http://74.125.93.104/translate_c?hl=en&sl=cs&u=http://www.danyk.wz.cz/induk2.html&prev=/search%3Fq%3Dhttp://danyk.wz.cz/induk2.html%26hl%3Den%26client%3Dfirefox-a%26rls%3Dorg.mozilla:en-US:official%26hs%3DRLB&usg=ALkJrhgcck83UNYV4KrWiQ9jAw_ruP-UZA

Billybobjoe

Mon Nov 03 2008, 10:49PM

Registered Member #396 Joined: Wed Apr 19 2006, 12:55AM
Location: Pittsburgh, PA
Posts: 176

jshine wrote ...

Thanks again to all who replied for the clarification. I've got another quick question, if someone has the time to answer:

Also, how critical is the quality of the 2.2 uF caps? These are expensive on Mouser:

Where did you get that capacitor?!?

Mouser has DC link capacitors which are somewhat cheaper (look under 2.2uf film caps). From the schematic it appears that the caps closest to the work coil will need to actually carry resonant current, so you should be able to get away with cheaper 2.2uf capacitors (I assume . . .)

jshine

Tue Nov 04 2008, 01:24AM

Registered Member #1784 Joined: Tue Oct 28 2008, 02:30AM
Location: Rochester, MN, USA
Posts: 14

That cap came from a friend who has been collecting electronic components for decades -- I'm really not sure where it's from originally. My guess would be that at some point he took apart a real induction heating machine that was being scrapped. I'd rather it was a little larger capacity, but 0.4 uF isn't terrible. Also, I have no idea what the ESR/Q value is -- the thing does look old, after all...

Thankfully, this isn't one that I had to pay for -- I'm sure it would be pricey...

1 2 3 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob