Development of an Avalanche Output Stage

[BEP]

Yea. I had a computer monitor go on the fritz a couple of years ago. I knew I had a monitor in the attic.

After an hour of digging I found it. It was a 10" Hercules amber monochrome NIB   I didn't have anything to test it on.... Bah!

[Peterae]

Well it looks pretty enough 

but will it work.

Still a little to do yet.

[Grumpy]

Looking spiffy, Peter!

I'll be back up and running sometime today at about 2kv (give or take)

I'm using this driver circuit with an IRF510 (saving my 840's) and running it through a 12v transformer flipped around to get 120 out to a MOT - thus it will be adjustable and current-limited and much lower current than normal which was to high anyway.  Will run it off my adj power supply.  You can run other transformers off it as well and even flybacks.

[wattsup]

That looks really great.

At the risk of sounding really stupid (won't be the first time). Would this Avalanche circuit do about the same thing as a post I made some time ago about an idea I have concerning a Voltage Grabber Circuit animation I made located here;
http://purco.qc.ca/ftp/Wattsups%27%20stuff/voltage-grabber-circuit-VGC/wattsup-VGC1a.gif

Basically it would use several stages, charging small capacitors one by one, then putting them in series to discharge them all into a larger capacitor, then doing it again and discharging into another larger capacitor, etc.

Sorry if maybe this is off topic.

[poynt99]

Looks perdy Peter.

Hope it works as desired.

.99

[darkspeed]

Thats looks great!

I like the quick connect pins for the incendiary devices 

[Peterae]

Cheers Guys

ds  

Wattsup
Each stage forms part of a pulse shaping circuit and has charged capacitors which release their energy in a controlled fashion thus in theory giving me a 1-9 stage pulse width and energy controlled pulse, hopefully more controlled than a piece of coax, but this is new ground for my poor digital brain so it's a steep learning curve for me and it may not work as intended yet.

[darkspeed]

Ok I must have missed it  - what is the goal voltage / amperage / speed ?

[Peterae]

Impulse current maximun is 60A pulse settable in 9 stages from about 12 amps per stage.

Can use up to 5 smmt417's in a string each is capable of over 300volts each so max pulse voltage is 1500V-1800V

Pulse width if my guesstimate is right should be between 40-120nS again settable.

[darkspeed]

What are you using for a power supply?

[Peterae]

The power unit was documented earlier in this thread, you need to realize i realtively slow charge a string 0f 100pf caps and inductors, and when the avalanche transistor fires it dumps the energy into the coil at short high impulse amperage.

http://www.overunityresearch.com/index.php?topic=307.msg3878#msg3878
http://www.overunityresearch.com/index.php?topic=307.msg4025#msg4025

[BEP]

Peterae,

Just an observation of that fine looking layout...

Every time you pulse a coil it will induce a reverse current in the coils before and after it - maybe in the other run, as well.

To minimize this, you can only have 6 stages, unless you're going to fire in a sequence that allows each previous coil to aid the next fired?

As usual, very interesting designs!

[Peterae]

The whole point of the cap/inductor chain is to spread out the discharged energy over a longer period of time to form a square pulse of longer duration, just like G uses a length of coax to control his pulse.
I think the coils as they are should aid each other hopefully, but i maybe wrong.

[BEP]

If they don't, you could try having a decreasing pulse width for each newly fired coil. The idea there is for each previously fired coil to 'make way' for the next coil in the firing sequence. You should see maximum voltage output this way. Should be the same result as a saturated core pulse former.

Cannons firing cannons

[darkspeed]

Sorry Peterae - i didnt see that - so your doing this as a one shot.

If you had a lot of  hv in reserve capacity how fast could you cycle the output stage?

[Peterae]

Sorry Peterae - i didnt see that - so your doing this as a one shot.

If you had a lot of  hv in reserve capacity how fast could you cycle the output stage?

It depends on the resistor chain and how much juice the psu can throw into the cap bank.
I wont know for sure until i fire this up, but it will also be a function of how many stages i am using at the time.

[Grumpy]

I abandoned the MOT idea for now - too inefficient

4kv supply is up and running, but low on time until tomorrow

[Grumpy]

Boom!

big flash in my 0-60vdc 0-6A supply and then my 4kv supply lit on fire - arc and sparks everywhere - my kid standing right there "Wow Dad!  That's cool!

Variable supply is toasted.  It's a Lambda LK 345A FM - probably worth fixing.

Looks like just the transformer in the 4kv supply is dead  - variable supply was set at 2v and went supernova.

I was just starting to get the arcs to insulated objects when it went.  Have an avalanche transistor stack of 20 2N5551's.

One step forward, two steps back...but it was starting to work.

[darkspeed]

Wow.. sorry G....

Have you ever built a Terry Filter? Maybe a consideration? Do you have an overshoot gap to dump anything that is way over what you expect in voltage?

[Grumpy]

just the way it goes in the tall grass...ROFLMAO!!!!

[Peterae]

Dam Sorry to hear that G

Was this using the circuit i built with the 555 and the LM339 and 510 fets to drive the transformer.

[Grumpy]

Was using a 4kv neon supply (little one) running off my variable bench power supply.

I found that outlet is not grounded.  Anyway, something in the supply blew - saw a flash - voltage on supply pegged the meter (lost regulation) and smoked the transistor in the 4kv supply.

I have a couple of other supplies but they are 500ma and 300ma and I need 1A - Soooo - I pulled out an old PC pwr supply an it has one leg that is 12vdc at 1A - perfect.

Got to fix that outlet first though.

Hate to lose that supply, but I was getting the polarization current and that shows that it was working.  Also was able to arc after the oil so was getting some good voltage through the coil.

The 1000v limit happens to also be the minimum for detecting the polarization current and I recommend as high a voltage as you can get.  Took 20 2N5551'a to handle the 4kv.

Hey - quick question - the capacitors (.01uF 20kv ceramic - 4 of them) on the output of the 4kv supply were really "squeeling" and I am wondering "why"?

[darkspeed]

If it s squeeling it is most likley punched through

When you unloaded your supply - big volts - big amps - many turns -  the pulse on the output side may have been 40kv 50kv?

can you run your supply from a battery?

[Grumpy]

If it s squeeling it is most likley punched through

When you unloaded your supply - big volts - big amps - many turns -  the pulse on the output side may have been 40kv 50kv?

can you run your supply from a battery?

I thought one of them may be leaking, but they are rated at 20kv and I was only using 4kv and only about 15 or 20mA  I was running off my supply at 12v and 1A.  The caps are after a full bridge rectifier too.  Funny...

(I have a big hand-rolled .005uF "Ganga Mon!" cap with about 10 layers of poly in it that I bet won't squeal.)

Just back from the Shack and will replace the 2055 tranny in the 4kv supply and see if that is all that smoked..fingers crossed...got a 12v transformer too and regulator to build 12v 1A supply.

[BEP]

Your squealing could have been from coronal discharge outside the caps. If it squeals again (with a slow ramp up of volts), turn the lights off and let your eyes adjust. You should be able to see it.