induction heater resonance / General Science and Electronics / Forums

Forums

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

« Previous topic | Next topic »

induction heater resonance

previous 1 2 3 4 next

Move Thread

LAN_403

uzzors2k

Wed Aug 06 2008, 09:50AM

Registered Member #95 Joined: Thu Feb 09 2006, 04:57PM
Location: Norway
Posts: 1308

Myke wrote ...

Also try to minimize wire length as that introduces stray inductance.

Stray inductance isn't much of a problem with LCLR since the matching inductor will totally swamp it, and because hard-switching doesn't occur either.

Hernan, looks like your setup is suitable for some real power. You should scope the inverter current to see what the switches are actually working with, at 10A input from mains your fets my very well be circulating more than three times that much current! In my heater the inverter current is kept around 18A, but I would estimate the mains input to only a few amps. For all we know your gate driver might have been alright but the fets died and killed it first. Regardless I would use a GDT with a damping resistor, as this would save the logic circuit in the event of a failure.

I just finished my new IH driver, you might be interested in trying it out?

GeordieBoy

Wed Aug 06 2008, 11:01AM

Registered Member #1232 Joined: Wed Jan 16 2008, 10:53PM
Location: Doon tha Toon!
Posts: 881

Hernan: From your pics I can't see if you have a DC blocking capacitor between your half-bridge and the matching network? It looks as if you have taken the return wire from the tank circuit back to the mid-point on the electrolytic bank in your picture. I would not recommend this practice for two reasons: Firstly this puts half the DC bus voltage on the work-coil which may be a significant shock hazard. And secondly, if a MOSFET were to fail short-circuit it would result in the full supply voltage being applied across one half of your bank of seriesed electrolytics, perhaps leading to venting. In a half-bridge arrangement, it is safer to take the load return back to the midpoint of a capacitive divider of poly-prop or mylar caps than to take it back to the mid-point of the electrolytics in a voltage-doubler. Or alternatively use a DC blocking capacitor in series with the L-match inductor like Uzzors did.

Uzzors: That looks good! I'm glad you got the controller working properly. You will probably get that bolt a lot hotter when you experiment with other work coils using less turns and fitting the workpiece more snuggly.

Hernan

Thu Aug 07 2008, 04:30PM

Registered Member #1614 Joined: Wed Jul 30 2008, 03:08PM
Location: Argentina
Posts: 52

Geordieboy, I'm not using DC bloking caps because between the inverter and the matching inductor I placed a large ferrite transformer with 1:1 relation to isolate the tank from mains voltage . I atached a picture below. one terminal of the primary winding is conected to the mid point of the electrolitics caps and the other terminal to the mid point of the mosfets, so in the secondary I have a square wave with an amplitude Vdc/2.I used electrolitics to Keep Vdc/2 constant but I will switch to full Bridge were is not critical the value and quality of the filter caps.thanks for your advices!

Uzzors, I have seen your PLL control circuit , I will try it when I get better Mosfets,it looks very robust design.I will use GDT's to isolate the control. I blew a couple of IR 2110 which aren't cheap here ....Ouch!
here are some pictures

thanks !

Hernan

Tue Aug 26 2008, 07:12PM

Registered Member #1614 Joined: Wed Jul 30 2008, 03:08PM
Location: Argentina
Posts: 52

Hi all, I've been looking for large rating mosfets but I 'll have to parallel three IRFP460 because it was the larger rating avaliable here. the input capacitance would be three times larger , I hope that wont be a problem at 35 KHZ.

Is convenient to split the matching inductor to L/2 before to drive the work coil and the power caps? because I 'm using one inductor between the isolating transformer and the parallel resonant tanK.

101111

Tue Aug 26 2008, 07:36PM

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

ST have some ISOTOP MOSFET's: 45nk80zd. but I would also try IGBT's.

Hernan

Wed Aug 27 2008, 07:15PM

Registered Member #1614 Joined: Wed Jul 30 2008, 03:08PM
Location: Argentina
Posts: 52

Hi all ,
I 've made some changes in the circuit and the workpiece (0.4 Kgs steel cilinder) got hotter than last times I esatimate it was at 950ÂºC , I changed the isolation transformer relation to 2:1 to boost the current input to the LCLR tank and I used less turns in the matching inductor ( six turns) around a gaped ferrite core. the mains input was 8 Amps and tank voltage was 155 V. I would like to measure coil current , what would be the best metod to measure it???

uzzors2k

Wed Aug 27 2008, 07:45PM

Registered Member #95 Joined: Thu Feb 09 2006, 04:57PM
Location: Norway
Posts: 1308

Place a small (low capacitance compared to the tank cap) capacitor in parallel with the tank capacitor. Then put a CT on it, more specifically put one of the smaller capacitor's legs through the CT so you can sense the current flowing through it. The sense capacitor to tank cap current should be proportional to the sense/tank cap capacitance ratio if you use a low ESR capacitor.

Hernan

Fri Aug 29 2008, 01:34PM

Registered Member #1614 Joined: Wed Jul 30 2008, 03:08PM
Location: Argentina
Posts: 52

Uzzors, should I use many turns or single turn as a primary and secondary??? will be Ok with 1 ohm resistor as load?
I used a small ferrite core to make CT
the tank cap is 14.5uF , I used 100nF sense cap and got hot fast so I 'll have to reduce its value.

uzzors2k

Sun Aug 31 2008, 01:50PM

Registered Member #95 Joined: Thu Feb 09 2006, 04:57PM
Location: Norway
Posts: 1308

One primary turn for the CT is ok. Just use some math to get the correct values. For example; 1 pri turn , 33 sec turns and a 33 ohm load resistor. The current will be stepped down 33 times, only to pass through a load of 33 ohms. U = I*R so 1A will equal 1V across the burden resistor.

Experimentonomen

Sun Aug 31 2008, 02:00PM

Registered Member #941 Joined: Sun Aug 05 2007, 10:09AM
Location: in a swedish junk pile
Posts: 497

I prefer 1:10, 1:100, 1:1000, much easier.

previous 1 2 3 4 next

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