Induction heater / General Science and Electronics / Forums

Forums

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

« Previous topic | Next topic »

Induction heater

previous 1 2

Move Thread

LAN_403

IamSmooth

Sun May 06 2012, 06:14PM

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

ben123324 wrote ...

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. thanks for the expertise

When I was running off of 120v and I doubled it to 240vac my caps were rated for 750vac/3000vdc. The rise can be quite high and I did not want to take any chances, so I had a huge safety margin. I measured the rise at the time, but I can not remember the value.

My larger unit used a 1000A/710peak (500vrms) volt celem capacitor. I was running this one off of 240vac doubled to 480vac. I was probably pushing this one close to the limit. However, as the current would increase I would throttle the voltage back so I was never at maximum input voltage when the current was high.

Ben Solon

Sun May 06 2012, 06:49PM

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

ok, so my eurofarads have a 580vrms rating, so they should be fine in 120v operation, and for 240, i will just see what is happening in 120, then scale up preportionaly. they have (most likely) a 800v peak, just using the rms value and converting it. so if i am carefull, i wont have any problems.

Microwave

Tue May 08 2012, 08:23AM

Registered Member #4544 Joined: Sun Apr 22 2012, 09:14PM
Location: Switzerland
Posts: 3

Haha thanks to you, IamSmooth I got my IH setup finally to work.
Fuckin nice site you created... I hadn't ever seen before what a real supply CT has to look like!
But now, it's running quite well. :)

I found that series resonant tank circuits can be very easy in handling, although mine is still not running on direct european mains due to crappy gate signals over 180VDC bus voltage.
So for beginning I would propose something like the 3kW version of IamSmooths induction heater.

I myself wouldn't know how to calculate those LCLR circuits correctly. ;)

Anyways, Good luck!

Ben Solon

Mon May 14 2012, 09:04PM

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

I'm sorry... I just had to do it

Ben Solon

Thu May 17 2012, 05:08PM

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

Just a couple more things to add. Using a pll circuit any and all delays in between the output and the tank do not matter because the pll will compensate for the shift, correct? I thought of this because I thought about controlling tank current by holding the frequency at resonance under all conditions, then altering the duty cycle of the bridge into the impedence matching. This is instead of detuning the system. A seperate microcontroller circuit would monitor peak current in the tank and controll on time.

The pll output would feed into one buffer and one inverter. Both lines would feed into seperate monostable fast 555's which generate a pulse on the beginning of each transition of the pll square wave. Each 555 would have a digital potentiometer controlled by a microcontroller. That uC would give a value based on both the measured peak current in the tank and an operator's potentiometer. The he end result: 0-50% duty to each igbt and current level controll.

Steve Conner

Thu May 17 2012, 06:23PM

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

A reasonable idea, except IGBTs don't like PWM because it messes up the zero current switching.

My driver boards work by missing out whole cycles of drive whenever the current goes over the limit. This is easier to do than PWM and seems to work great. I've had no failures in Tesla coil service. My induction heater blew up, but I think it was because I had crappy heatsinking and turned the current limit up too high.

IamSmooth

Thu May 17 2012, 06:57PM

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

ben123324 wrote ...

Just a couple more things to add. Using a pll circuit any and all delays in between the output and the tank do not matter because the pll will compensate for the shift, correct?

After all the delays the two waveforms will not line up as theory dictates. You will need to empirically determine the correct phase shift for optimal performance. Then, you need to raise the frequency above resonance to prevent reverse current through the switches. You will sacrifice some current, but have a larger margin of safety for blowing switches.

Ben Solon

Thu May 17 2012, 07:05PM

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

Ok, so the frequency is fres+constant. But using large delay circuitry in between the output and the tank will not matter as long as I can measure the delay, then compensate?

Edit: Hmm. Steve didnâ€™t see your comment there! How do you achieve the missing cycles? I would prefer to run it through some kind of and gate controlled half by the uC and half by the pll. I donâ€™t like flip-flops and such mainly because they will do their own thing any you just sit back and watch XD it would be easy enough just to drive the input low right when the over current is detected, but then you cut the current off mid-wave.

previous 1 2

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