Flyback driver destroying MOSFETs / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

Flyback driver destroying MOSFETs

1 2 next

Move Thread

LAN_403

ScotchTapeLord

Thu Jan 29 2009, 11:59PM

Registered Member #1875 Joined: Sun Dec 21 2008, 06:36PM
Location:
Posts: 635

I have a 555+MOSFET flyback driver and it's reduced two 600V 15A MOSFETS to 5 ohms resistors from gate to source... How is this even possible?

101111

Fri Jan 30 2009, 12:24AM

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

Read

for a little more information. If you search here at the forum there should be a handful of threads about this.

ScotchTapeLord

Fri Jan 30 2009, 03:18AM

Registered Member #1875 Joined: Sun Dec 21 2008, 06:36PM
Location:
Posts: 635

Thanks for the link. The search results for "flyback" were a bit overwhelming!

But still, I'm curious as to how the CEMF found its way to the gate, especially with an ultra fast freewheel diode and a capacitor across the primary.

Dr. Dark Current

Fri Jan 30 2009, 08:43AM

Registered Member #152 Joined: Sun Feb 12 2006, 03:36PM
Location: Czech Rep.
Posts: 3384

Is the supply for the 555 well filtered and decoupled? Do you use a voltage regulator for it (you should use one)? The problem with dying 555's is usually the noise on its supply rails, caused by current spikes drawn from the supply by the transistor.
You should also have a ~0.1uF cap from pin 5 to ground.
And finally, aren't your MOSFETs overheating?

Antonio

Fri Jan 30 2009, 12:54PM

Registered Member #834 Joined: Tue Jun 12 2007, 10:57PM
Location: Brazil
Posts: 644

ScotchTapeLord wrote ...

Thanks for the link. The search results for "flyback" were a bit overwhelming!

But still, I'm curious as to how the CEMF found its way to the gate, especially with an ultra fast freewheel diode and a capacitor across the primary.

You don't use a diode across the primary in a flyback circuit. The usual connection is a high-voltage capacitor of a few nF across the primary coil and a fast, high voltage, diode in parallel with the transistor, anode grounded (most mosfets already have one). If anything gets so hot that you canÂ´t touch, something is seriously wrong.

ScotchTapeLord

Fri Jan 30 2009, 08:55PM

Registered Member #1875 Joined: Sun Dec 21 2008, 06:36PM
Location:
Posts: 635

I had it working perfectly at first; the MOSFET remained cool. I was using it and then I turned it off to replace the 10K pot with a 1M pot in order to try it at lower frequencies. When I turned it back on it was a MOSFET killing machine.

I would think that a diode across the primary would take care of all the CEMF.

I have decided that since I was only feeding the coil 12V, it must be a problem with the 555 and gate, simply because the CEMF couldn't be too much for collecter to emitter breakdown. I will try decoupling and a regulator now.

Thanks for the help.

Is it okay to run a flyback with an unloaded secondary for a long time? For my application it will require an unpredictable amount of open-circuit time, and that is why I have switched from ignition coils.

Dr. Dark Current

Fri Jan 30 2009, 09:02PM

Registered Member #152 Joined: Sun Feb 12 2006, 03:36PM
Location: Czech Rep.
Posts: 3384

ScotchTapeLord wrote ...

I had it working perfectly at first; the MOSFET remained cool. I was using it and then I turned it off to replace the 10K pot with a 1M pot in order to try it at lower frequencies. When I turned it back on it was a MOSFET killing machine.

So here is your problem. Lowering frequency increases primary current, with an ideal inductor the relation is linear, so if you reduced the frequency 100 times, you will have 100 times higher primary current. However, the core starts saturating with so much flux density, which further increases the current.

The usual rule of thumb is that if you can hear it 'whine', the frequency is too low (<16kHz). Don't use under 15kHz to drive TV transformers.

ScotchTapeLord

Fri Jan 30 2009, 09:23PM

Registered Member #1875 Joined: Sun Dec 21 2008, 06:36PM
Location:
Posts: 635

I thought that could have been the problem, but even at about 6kHz the MOSFET was cool... and I also tried going back to the higher frequency, but I must have had a defective 555 by that point. Whatever, I'll add the safety precautions and see what happens.

One other problem, that was my last MOSFET, so I'm down to four IGBTs. From other discussions, I see that they are very similar, but the drivers for IGBTs seem to involve gate-driving ICs or GDTs. Is driving an IGBT with a 555 bad? I've measured my IGBT's gate-collector capacitance at 2.4nF.

Antonio

Fri Jan 30 2009, 10:09PM

Registered Member #834 Joined: Tue Jun 12 2007, 10:57PM
Location: Brazil
Posts: 644

"I would think that a diode across the primary would take care of all the CEMF."
Would also impede the correct working of the flyback circuit. To have a high voltage pulse at the output when the transistor switches off, another high-voltage pulse (easily > 1000V) must appear in the primary winding. A diode would clamp it, and clamp the output voltage too.
If the circuit was working with the diode in place, you probably connected the primary winding inverted.
Look also at the ground connection of the circuit. Without ground, with positive high-voltage at the output the whole circuit tends to negative high-voltage, and a rupture will occur somewhere.
The lowering of the operating frequency really increases the current in the transistor, as mentioned. But it's perfectly safe to operate at audio frequencies, if the primary winding has sufficient inductance and the power supply voltage is not too high. Using the normal primary winding of the flyback, and a power supply of a few tens of Volts, it's really necessary to operate at a few kHz to have significant output power.

ScotchTapeLord

Fri Jan 30 2009, 10:30PM

Registered Member #1875 Joined: Sun Dec 21 2008, 06:36PM
Location:
Posts: 635

Oh, okay. I actually think I was using the diode across the transistor properly. In any case I am definitely doing it correctly now.

The post I was referred to says decoupling should be .1-1uF? Is there such a thing as too much decoupling? I have the decoupling capacitor before a 12V regulator, is that right or should it be after it?

1 2 next

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