Kapanadze Cousin - DALLY FREE ENERGY

[Grumage]

The IGBTs will be OK if you do not open one of them longer than for several tens of microseconds.

The IR2153 has one major flaw. Its gate driver circuits are incapable of sourcing/sinking enough current to charge/discharge the gates of big IGBTs quickly. Thus your transistors will be between the ON and OFF state for a long time and they will heat up. Because of that deficiency, their switching frequency will be limited, too.

Dear Verpies.

Thank you for your, as usual, full and precise help with my project. I am attaching one of Nick's beautifully enhanced images of the IGBT and transformer.

This is a quote from a freind of mine who is "UP" on electronics and I think it might have a bearing on what you, dear Verpies have been talking about.

"Big diode is freewheel diode as that transistor does not have one built in.
The parallel diode cap resistor is in series with the gate driving current, when switching off the diode conducts the current so it switches off very fast, when switching on it goes through the resistor and is slower and can help reduce unwanted ringing."

He also suggested this link.........http://www.ti.com/lit/ml/slup169/slup169.pdf

I hope this will help all the constructors?

Cheers Grum.

Addendum. Having closer studied Nick's photo I have come to the conclusion that the circuit is shown in the above reference Fig 12 Page 14. Grum.

[verpies]

Grum,

Your friend is giving you a good advice. The Texas Instrument Balogh application note is a good one.
I use the attached Chen application note even more frequently.

I think that there are two ways of building these exotic devices.
1) Blindly replicating everything to the dot from a very exhaustive documentation (without understanding and without ego improvements).
2) Building a custom optimized design with the understanding the operational mechanisms.

These application notes will not be much help to people who just like to solder components without understanding exactly how they operate.

Note, that the freewheeling diodes are not always good.  Yes, they protect transistors from high voltage spikes but they also  prevent the current in the windings of a transformer or inductor from disappearing quickly.  (the same goes for magnetic flux).   Most of the time it does not matter, however if your operational principle depends on fast current fall times, then the freewheeling diodes are detrimental.

[a.king21]

Absolutely. I have made this mistake loads of times. You protect the transistor and the spike disappears.
I use TESLA switches now. No transistors to blow up - ever.
BTW Bedini guys: Make the device spin at 3000 rpm and then look at your results.
That's all Magnacoaster did.
NB Kapanadze told me personally that he went through bucketfulls of transistors before he found the right ones.

[NickZ]

I'm on vacation,  be back in two weeks or so
   ps. I'm following along

[T-1000]

Absolutely. I have made this mistake loads of times. You protect the transistor and the spike disappears.

If you still need to use transistor, attach diode in series (just make them 10+ in paralel to lower resistance) between coil and transistor. The mechanical break will still produce the best spikes but sometimes the electronic counterpart is OK too.