Hello again all

I am asking humbly once again for your assistance. I would like to create a spark generator, and havent found many specs online for something like this. The pulses need to be DC and higher voltage is better. I was thinking a DC pulse transformer is a good opion, but the real requirement is that I would like the sparks to be as SHORT as possible, the smaller the time the better. Any ideas?

 

I wouldnt think a pulse transformer would be a good option. They usually do not have very high ratio's and are not usually designed for high voltage operation either.

Use a flyback xformer from a TV set and see how short you can make your pulse.

 

This is my area of specialty....

An ignition coil will give you about 150us of spark per discharge. You apply 12 volts and once the current through the primary reaches its maximum, you switch off the primary current. You will get about 15kV on the secondary for about 150us with a decent coil. There will be some good kickback voltage on the primary, up to a few hundred volts (200V is what I was getting). A good way to do it is to connect 12V to the + side of the primary, and use a FET as a low side switch (high voltage, high current FET). Then use something to drive the FET that will get you a real good Vgs as power FET's have high gate capacitance and need a high turn on voltage (usually 8-10 volts). A 555 might be able to drive it directly but you may need a driver FET between the 555 and the power FET for more current. You may want to have the 555 on a different isolated supply and use isolators to connect it to the HV side (driver fet and/or power FET) so you won't blow the 555. You will also need to tweak the frequency and duty cycle of the 555 so you don't waste power in the primary. Once current hits its max in the primary you should discharge it, otherwise you are just burning up power.

Anyway, this will get you a pulse that looks like an exponential discharge, with some ringing on the tail end. A good ignition coil can be had for <$50.

 

This sounds like a good option, could this give me a pulse on the order of 50 nano seconds? also I need voltages around 3000, I think this is outside the range of most mosfets.

 

What might you be working on?

 

I am doing a couple of expiraments that are similar to tesla coil technology. I need unidirectional dc pulses across a spark gap. I need to quench the spark, or controll its durration to a time below 100ns. If any of you are familiar with Edwin Gray, it is something along those lines.

 

There's a circuit for a very-fast-risetime (250 ps) pulse generator that's described in AN61, at linear.com, that might give you some general ideas that could be helpful.

- Tom Gootee

http://www.fullnet.com/~tomg/index.html

-

 

I have read that people have successfully used vacume triodes to accomplish this feat, however when I look at their schematics, it looks as though they only have two connections to them, I thought that triodes needed three leads to function properly, any ideas as to why?

 

What about charging up a capacitor using a constant current source (a current limited HV supply, such as a neon sign transformer transfromer with a rectifier on the output) and discharging it though a spark gap switch?

__________________
I also post at the following sites:
http://www.stop-microsoft.org http://www.heated-debates.com
Screen name: Aloone_Jonez
And http://www.silicontronics.com, same screen name as here.

 

Hi,
Have you looked at rotary spark gap generators?

http://www.richieburnett.co.uk/rotary.html

__________________
Eric
"Good enough is Perfect"

PIC tutorials:
Gramo's: www.digital-diy.net/
Bill's: www.blueroomelectronics.com/