user TURBO?s replication of Steven Mark?s TPU ?

[pese]

it give a lot of 400-800 volt types
(are used in power switching supplies monitors tv !)

[tosky]

Hi otto,
Though your output 170V, the Drain of the MOSFET should still below the applied voltage at 12V. If the back emf happen at control coil it should be directed by the internal diode of the MOSFET to the input capacitor that will store to 12V power. So it should be safe to use 30V MOSFET.

[MeggerMan]

Hi Otto,
Yes you would think that this might be a problem but I do not think it is.

What you do is put a blocking diode across the drain and source together with a resistor and small value capacitor (the capacitor and resistor are in series).
You pick a resistor and capacitor value that is enough to quench the large back emf spike just after the pulse goes off.

I have suggested this as this is what the PWM circuits use. I have tried them on a circuit simulator and they seem to do the trick.
Now it is possible that these 3 components will upset the TPU in some way in which case you will need to use a very high voltage MOSFET, these are more expensive and the "on" resistance is much higher, so more power will be lost.

See page 8.
http://focus.ti.com/lit/ds/symlink/uc28025.pdf

If you can try these 3 components I think you will be surprised how effective they are.

Hi Tosky,
When the back emf occurs I think you will find that the drain to source voltage will now be:
D,S = back emf + supply voltage

e.g. back emf = -170V, supply = 12V, D,S = 182V
I know it sounds odd because we are all use to seeing the voltage across the coil.

You see it go +ve during the pulse and a big -ve spike when the pulse goes off.
But if you put the scope probes across the drain and source of the MOSFET, things are a whole lot different, you see zero volts (or just a small voltage) while the pulse is on, and a big spike as if someone had just swapped the 12V battery for a 182V battery.
Then it will slowly balance out to 12V when the spike bounces die away.

I think the internal diode only protects against a reverse voltage across the drain to source, plus I don't think all mosfets have one.

Regards

Rob

[bluedemon]

@everyone
Here is a little experiment I did on Tesla pancake coil using a 555 timer and a darlington pair. SM releated? You be the judge.

http://cmnet.ca/projects/flatcoil1.wmv   

Have you tried connecting the coils as in tesla's 0512340 patent?

[Grumpy]

PULSES