Induction heater / General Science and Electronics / Forums

Forums

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

« Previous topic | Next topic »

Induction heater

1 2 next

Move Thread

LAN_403

Ben Solon

Fri May 04 2012, 04:32PM

Registered Member #3900 Joined: Thu May 19 2011, 08:28PM
Location:
Posts: 600

A friend asked me if I wanted to build a large induction heater with him after seeing my small zvs version. I would like to use my igbts as I have a lot of them. I don't see many people using them though. Igbts have switching losses and assuming 50a and 80khz the losses in the 30n60 equal roughly 300w. If you compare that to mosfets on resistance losses assuming 50a and 170v, the losses are enormous at over a kw. Am I going about this correctly? My aim is to have something in between marko's zvs and Tim Williams monster. 5kw maybe?

The other question is how should I drive it? Igbts don't parrallel well like mosfets do, so should we drive to impedence matching xmfr with a full bridge?

Inducktion

Sat May 05 2012, 04:56AM

Registered Member #3637 Joined: Fri Jan 21 2011, 11:07PM
Location: Buffalo, NY
Posts: 1068

You have to remember that an induction heater's tank isn't ever directly coupled to what flows through your switches.

The switches only have to conduct for a short amount of time to achieve resonance. With impedence matching, and inductive matching you can easily achieve ZVS or ZCS and prevent your mosfets from getting excessively toasty.

My induction heater used mosfets; but because of the inductor the actual current the mosfets will see is very low.

Ben Solon

Sat May 05 2012, 03:08PM

Registered Member #3900 Joined: Thu May 19 2011, 08:28PM
Location:
Posts: 600

how low exactly? and what power was your ih? there is very little info of peoples work with induction heaters beyond 1kw on the internet, and it always helps me to see what works and what doesn't before building something of my own.

edit: i know generally where i want to resonate unloaded, but is there any way to know in advance the amount of current in the tank based on the capacitance and frequency? i know that more capacitance gives a higher current. i am looking at something in between 6.6 and 8.8 microfarads. but i don't want to spend money on a large tank cap, then blow it up...

Steve Conner

Sat May 05 2012, 07:41PM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

To melt a small piece of steel, you need in the region of 1kW input power and maybe 100 or 200A of work coil current. Search the forum for a thread called "OMG Induction Heater".

IamSmooth

Sat May 05 2012, 08:15PM

Registered Member #190 Joined: Fri Feb 17 2006, 12:00AM
Location:
Posts: 1567

When you get above 20-30khz the IGBTs will have very large losses and you should use mosfets.

Read this for a tutorial on how to build a unit from 2-12kw. It includes theory and schematics.
You will need between 1-2kw to melt a small steel nut.

Ben Solon

Sat May 05 2012, 09:09PM

Registered Member #3900 Joined: Thu May 19 2011, 08:28PM
Location:
Posts: 600

steve: thanks for the thread. i read it a long time ago, then forgot the name and couldn't find it again.

iamsmooth: that actually is what i ahve been reeding a lot lately, and while it is the best theory site out there(aside from maybe richies, but its too close to say), it doesnt have too much on the circuit other than the one schematic.

iamsmooth uses fets, tim williams uses igbt's and the guy from the video in "omg induction heater" uses igbt's. i think i may be able to get by using them. i have driven my 30n60's all the way to 1.8mhz and they worked just fine in a class e tc.

i intend to heat some large chunks of iron and steel, so i am going with a low resosnant frequency. 25-60khz. if i am shooting for 3-6kw, what should my tank capacitance be? i am going to buy one of those eurofarads off of ebay.

IamSmooth

Sat May 05 2012, 11:24PM

Registered Member #190 Joined: Fri Feb 17 2006, 12:00AM
Location:
Posts: 1567

Your coil is going to be a few uH. Using this and your intended frequency range you can figure out what capacitor you need. The inductance of the coil will change with the workpiece, so the frequency is just a ballpark number. Once you settle on a capacitance value you will see what the general frequency range of your tank will be. If it is too high use fets; if it is low enough use the IGBTs.

Ben Solon

Sun May 06 2012, 12:40AM

Registered Member #3900 Joined: Thu May 19 2011, 08:28PM
Location:
Posts: 600

dont take this the wrong way, but i know this already. its why i said 25-60khz the coil is going to be 1uH and the cap will be 7.5uF. this means it is self resonant at ~60khz and will drop with the added inductance of the work piece. i assume 308A rated tank will be adequate?

one other question though. in your ih what was the resonant rise? my caps are rated for 580V, and i plan on using either 170 or 340 vdc bus. i know that in a mazzilli circuit that the voltage can rise over 3X the supply voltage, but what about in an lclr? what did you use as a bus voltage? your caps could take 750v. i can always use over-voltage cutout, but i want to design a sytem that wont be using that feature often.

thanks for the expertise

Storken

Sun May 06 2012, 12:03PM

Registered Member #1960 Joined: Thu Feb 05 2009, 11:57PM
Location:
Posts: 48

The mazilli driver raises the voltage by Pi, just to be totally clear on that (I see alot of misunderstanding here).

The lclr setup has constant voltage with sinusoidal (at resonance) current flowing through the tank. I'm building one myself, but I've not gotten it to work yet. Mine uses a microcontroller and a half bridge driver - driving two 80V fets. Hope to be able to work in the 1,5 to 2kW area myself.

Ben Solon

Sun May 06 2012, 05:06PM

Registered Member #3900 Joined: Thu May 19 2011, 08:28PM
Location:
Posts: 600

yes, but i assumed it was resonant rise that raised the voltege. the lclr is also at resonance, so doesnt that mean that it will also have rise? and what do you mean by constant, because that could mean constant at line input voltage or constant at 20kv.
thanks

1 2 next

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