induction heater stuff / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

induction heater stuff

1 2 next

Move Thread

LAN_403

tomonthemoon

Tue Jul 03 2012, 07:46PM

Registered Member #5548 Joined: Mon Jul 02 2012, 04:42AM
Location:
Posts: 6

hi, first time here, first time i'm facing any project of this kind... need some help...please

i'v been working on a simple induction heater for a while. (hoping to power it up to 5kW)
i took the schematics from

, changed some parts and built it.
i took it quite seriosely, buld some casing, prepeard a pump, lots of control switches. after some struggles, i managed to make the PLL work properly with a DC power supply up to 40V. it works fine, things are worming up (though it's very slow due to low power).Please watch my schematics here.

1341344015 5548 FT0 Induction Heater V05

I bought a used Variac, 20 amp max, and hooked it up.
Now, i start raising the voltage and everything works fine for a while, my steel bar turns red, i reach about 80V on the Variac (which are doubled to 160V by the circuit) and while spinning the Variac's dial, the variac fuse breaks, the inverter's Mosfets burn (2 out of 4 - one at each side of the bridge). When touching the Mosfets, i feel no heat at all. (i also measured only 1Amp coming out of the Variac at about 80VAC meaning, not any close to the Mosfets limit - 44Amp)
i did it several times, changed the mosfets again and again. it happens at different voltages, some time i reached about 150VAC on the Variac which are doubled to about 300VDC in the rectifiers capacitors... again, the mosfets are cold and they burn. it happens only while turning the Variacs dial. i suspect that maybe the sparks from the Variac coiles (i hear them while turning the dial) indicates some non continuous volage entering the rectifier. maybe some inductance of something in the circuit is sensitive to these voltage instantaneous changes. is there anything i can add to the circuit to smothen the voltage comming from the Variac? is the there anything else might be causing it?

i don't have any experience in that area, any help will be more than appreciated

Steve Conner

Wed Jul 04 2012, 10:06AM

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

I think you have some sort of system-level problem. For example, maybe the long unshielded wires to your pots are picking up interference. Use an oscilloscope to check the gate drive signals, and see if they change as you increase the voltage on the variac.

tomonthemoon

Wed Jul 04 2012, 11:53AM

Registered Member #5548 Joined: Mon Jul 02 2012, 04:42AM
Location:
Posts: 6

thanks Steve. i will try that as soon as receive some mosfets tomorrow (just burned a about 12 of them in the last week) assuming i understand what you're suggesting, don't you think the gates will be protected by the zeners?
i dont know much about it, but i think that if the Mosfets burn without getting hot, it should be due to spikes (momentary currents), no?

Steve Conner

Wed Jul 04 2012, 12:02PM

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

Yes, but spikes can be caused by bad gate waveforms that switch the MOSFETs at the wrong times, especially if your layout is loose with lots of stray inductance.

Sometimes when a device "burns without getting hot", what really happened is that it got hot so fast, there wasn't time for the heat to dissipate through the casing. For example, I once dealt with an IGBT switching circuit that went unstable and toggled the IGBT on and off at several MHz while it was passing maximum current. When this happened, switching losses destroyed the device pretty much instantly.

tomonthemoon

Wed Jul 04 2012, 05:24PM

Registered Member #5548 Joined: Mon Jul 02 2012, 04:42AM
Location:
Posts: 6

thanks again Steve, i see what you're saying, I'll replace the Mosfets and monitor the gates signal.

one more thing: my Mosfets are 65pq015pbf

are they any good?
is there any other reliable type you would recommend?

Dr. ISOTOP

Wed Jul 04 2012, 10:41PM

Registered Member #2919 Joined: Fri Jun 11 2010, 06:30PM
Location: Cambridge, MA
Posts: 652

tomonthemoon wrote ...

thanks again Steve, i see what you're saying, I'll replace the Mosfets and monitor the gates signal.

one more thing: my Mosfets are 65pq015pbf

are they any good?
is there any other reliable type you would recommend?

Hmm you linked to a dual Schottky diode there...

tomonthemoon

Thu Jul 05 2012, 07:39AM

Registered Member #5548 Joined: Mon Jul 02 2012, 04:42AM
Location:
Posts: 6

right, my bad...i meant FDH44N50

IamSmooth

Thu Jul 05 2012, 05:03PM

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

Overcurrent and overvoltage will blow them. More likely, you have noise on the gate as steve suggested. This, too, will blow the switches at low power levels. Once the mosfet goes your variac's fuse will blow. You need very short leads to the gate. I had mine off the board which metal shielding. The noise will get through zeners so you need to get a clean square wave. I also used hyperfast FWDs for the reverse currents. If you look at my detailed tutorial here I give the part numbers. For example, for the reverse currents I used FREDs DSE 120-12A (I got them on Ebay) with a Trr of 40ns. That's fast.

There are a few schematics so make sure you make it to the end for the high power unit.

I also found that an unclean signal from my voltage doubler caused problems. It was the capacitors. My doubler had about 1000uf of capacitance, I used an EMF filter and I had the power leads twisted to minimize stray inductance. All of these things were important as I was putting about +50A through my switches.

Also, when you get a clean signal you will need to cool your mosfets with air or liquid.

tomonthemoon

Thu Jul 05 2012, 08:32PM

Registered Member #5548 Joined: Mon Jul 02 2012, 04:42AM
Location:
Posts: 6

if you read my first post, you must have noticed that i did make it the the end at your tuturial. i've actuly been buildiing it for the last few weeks. I am greatfull, i've learnd from you a lot!!!

most of the the components are the ones you chose on your first setup. The active filter of the PLL wasent working for me so well so i made it passive and added a tuning level with a summing amplifier.
i also tried avoiding the variac by using a variable 40Amp SSR on the 220VAC source power

to raise the voltage gradualy. it did not work well and put a lot of noise to the voltage doubler. it also slaughtered some of my Mosfets....eventualy i gave up bought a variac and i'm still fighting to make things work

i will try to add some shielding to the gates leads, as you said, mine are off the board, coupled to a heatsink cooled by a fan. a 100 mm long leads....(mybe this is the sorce of the problem). i'll also try to get some faster diodes, i thought 115ns was fast enough, maybe its not.

IamSmooth

Fri Jul 06 2012, 01:59AM

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

tomonthemoon wrote ...

i will try to add some shielding to the gates leads, as you said, mine are off the board, coupled to a heatsink cooled by a fan. a 100 mm long leads....(mybe this is the sorce of the problem). i'll also try to get some faster diodes, i thought 115ns was fast enough, maybe its not.

100mm for an unshielded gate lead is definitely a problem. It may not be your only one, but this is no good. Look at my images and buy braided shielding and ground it. Have your resistor (10R) connected directly to the gate as this will damp the noise right before the gate.

I, too, tried a switched-mode voltage supply, but I got too much noise going to the unit. I was able to get it up to 30A before it blew the mosfets, but this was not good enough. I did not want to invest time and money in a power source when I could just use a variac.

1 2 next

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