Make 2000V at 100uA

Use the "boosting transformer" config I posted earlier and you could do it in 4 stages probably

Find yourself a used electric fence charger on ebay or elsewhere. They output around 5000v, (or less) but are safe as far as the current. THere are two kinds. One is a pulser. It puts out pulses of voltage about once per second. The other is a steady voltage (often called a "weed burner"). note: some of the pulse ones are called "weed burners" too.

You probably want the steady voltage type.

• Since you may not know about these things, they are connected to a bare electric wire that surrounds livestock. Animals quickly learn that touching that wire gives them an unpleasant shock, so they stay away. Weeds often grow and touch these wires, and too many can drain the voltage so it no longer is effective with the animals. The "weed burner" ones are intended to burn the weeds away, where they contact the fence wires.

These are pretty impressive, I can beleive they have 2W (or greater) output, the spark very loud I have not tried to measure the voltage

the output is strong strong enough to cause visible damage to static shielding bags

Sure, if you decide to go the C-W route just do the calculations to use caps just large enough to provide the required 100uA output current without too severe a voltage drop, and if anything that draws more than a couple mA gets connected (ur hand) the calcs say that the output ends up crowbarred down into the dirt.

Don't taze me, bro!

I'm not sure why I'd buy a product _designed_ to set my property on fire...

$10 on EBAY will get you a stun gun. Point being that the high voltage stuff is done. All you'd need to do is design in some regulation. Diode, cap, adjustable spark gap shunt regulator.

Den mandag den 20. marts 2017 kl. 01.14.44 UTC+1 skrev Tim Wescott:

a few of these, or mayby just a more multiplier stages

So, does a $10 stun gun from eBay generate 2000V unloaded?

I grew up in an area rural enough to have caught one of these in my back ducking under a fence. Very impressive.

I guess my dad got to watch while my grandfather peed on one -- he claims he saw it coming but not soon enough to say anything. He also reports thinking it was hilarious, so I'm not so sure about the first claim.

For 2000V at 100 uA, i.e. a quarter watt, that might actually work. Triangle wave for lowest output ripple, and a CCFL transformer for high ratio boost, you might only need a couple of CW stages. For low ripple, though, it has to be FULLWAVE C-W, so it's four diodes per stage instead of two.

Right, at first blush the numbers seem to work. I could use a ~2kV/ 100uA power supply, that worked off my function generator. (I wouldn't want to blow out my FG though.. transformer on the output... (I could start with more R and 10uA.) I've not done a lot of transformers, a pot core is easy.

Can't I do fullwave with a FG? (I've only made a C-W once, on my bench... with a function generator.) And I have no idea about the triangle wave, (well only a small idea about current into a cap), but that all makes it more fun!

George H.

I expect so. It's rated at 60,000V. For twice the price you can get one advertised as 500MV. ;-) If you can get an arc between the terminals, it's gotta be better than 2KV. IIRC it's 1KV/mm give or take an order of magnitude. I think I'd worry more about getting the voltage down to 2KV.

Point is...the transformer you'd want is probably in there. Motorcycle spark coil would also work.

A good HV transformer from CCFL surplus seems easiest:

Observe the multiple sections of the secondry winding... high voltage and distance are built into the (odd, elongated) design. Alas, not quick to ship, this item...

Some CCFL transformers are center-tapped (which makes it easy), but not all. Your secondary might have no center-tap, so for fullwave the easiest is a bridge rectifier (at a kilovolt, yet) with the negative bit grounded, and then build up three series strings of capacitors; one side is push, the other is pull, constantly while the triangle is ramping up, then the polarity swaps when the triangle ramps down, but there's always a diode forward biased attached to the output (at the end of the middle of the three series strings of capacitors...).

Schematic looks something like this:

It's a rough drawing, but better than ASCII

also.

That's what the big boys made us innocent juniors to do. It was an experience still remembered over more than 60 years ...

Not a diy solution, but I've had good experiences with Spellman modules

Pere

OK thanks... I don't quite get your circuit. (I need to stare at it a bit more.)

George H.

That's cute. I've come to prefer making current-mode blockers, if I'm going to make a blocker--that makes them a lot less sensitive to components' variability.

Your circuit reminds me very much of a camera's xenon strobe DC-DC converter. In fact one of those, surplus, would be a great starting point. Add a C-W multiplier to the transformer output, done.

Sam Goldwasser's s.e.repair FAQ has lots of 'em.

Cheers, James Arthur

Nuh uh! Not blocking - AFAIK blocking oscillators rely on xfmr saturation to cut them off at the end of each cycle.

This one looks to me like it's running regular ol' gain -> positive feedback -> oscillation, with a tuned tank in the collector and with regenerative feedback derived from the small secondary winding. The BJT spends most of its time in class C hard cutoff, popping out for a brief period every cycle to drive current into the tank to keep it "ringing."