N.R.M.R.E. An investigation.

[padova]

TK,
there was confusion about the direction of that transformer from the start, i expected too that it would be connected to the collector via the secondary (24V) to get a HV (radiant) signal from the primary (240V), but Graham just asked Nelson the other day in this thread:
http://overunity.com/17186/the-bifilar-pancake-coil-at-its-resonant-frequency/msg503543/#msg503543
So primary (240V) connected to the collector.
Concerning the used 3N247 FWBR, that was what i had laying around, not sure what Nelson used.
Itsu

Then you can't get HV radiant output if transformer is connected like so.
I believe it's a misunderstanding in communication.
I'm using audio output transformer, connected so that low impedance is in collector circuit, and high impedance side is producing
nice HV over 1000V. 

[padova]

When you get something slapped together Tinsel, see if you can replicate then explain this...

You apply a few volts and you will see a typical oscillator, nothing special, drawing a few milliamps as would be expected.

If you slowly increase PSU voltage, you will get to a point (between 7v & 9v) where the circuit behaves much different.  Milliamp draw will go to practically zero; the waveform will go berserk quite literally, voltages on the scope will exceed 10x probe settings or nearly so.


I built this goofy thing about a year ago and still don't comprehend what it is doing or how.  It certainly didn't appear to me to be normal.  Maybe it is.  Hopefully you can get similar results and clue us in.
Whatever Nelson was on to, he chased something most of us wouldn't think to follow.

One other little note:
  Pay particular attention to how the circuit operates when it first goes into "berserk mode", because after just tens of seconds, you will never see this again quite the same.  I think the voltages actually damage the electrolytic capacitors.  However, you cannot substitute these electrolytic capacitors with something of higher voltage rating.  The way they are designed as rolled flat plates is essential to the operation of this circuit, as though their real purpose is some sort of coil instead of capacitor.  Probably why Nelson has been doing extensive research with pancake coils--he knows what he is looking for now.

I had similar effects.
Also with capacitors 100u to base . I've cooked a couple of these.
One small 100u/16v even exploded.

But that second transformer when connected in my setup, somehow everything works better.
It's pretty obvious. As if HV has a some feedback effect on the rest of the circuit.
Just my speculations.

[Dog-One]

My TIP122 Darlingtons are supposed to arrive Thursday, hope they
are genuine and not some recycled rebadged Chinese fakes.

Mine are Fairchild.  I'm not certain with the TIP122s, but with the TIP120s different manufacturers use different value internal bias resistors.

What you've described reminded me of a video I made some time ago with a very simple circuit from TinMan, using a Tesla Bifilar pancake coil.
I posted a link to it when I first made it and I'm kind of reluctant to post it again, what with the flame wars and all, and I don't want to
take this thread off-track.

Grum won't allow flame wars, so speak up if you know something associated with this gizmo.  After a year of thinking about this, I'm still in WTF mode.

In your build did you have the transformer and the FWBs installed when you saw the weirdness, or just what you have on the breadboard?

I didn't connect the "low voltage" bridge, diode and cap, but I did use the transformer.
http://www.mouser.com/ds/2/410/media-1067443.pdf


The "feeling" I have is somehow voltage is turned loose, pure voltage or nearly so.  There's no current you can equate to power.  This gizmo triggers only voltage, so it's kind of like a flyback or boost convertor of sorts without the normal step-up.  That in and of itself is probably no big deal, there's plenty of contraptions able to do it.  What's weird (to me) is the way this circuit does it, where it comes from.  Longitudinal Magneto Dielectric waves?   Who the hell knows.  It makes voltage though, quite a bit of it for how it's wired up.  I'm sure with enough attention to detail you'll figure out what it's doing and why.  It would be kind of nice to me to have that mystery solved.

[itsu]

Thanks for the clarification. Now I have a couple more questions (of course!)

Does that HV radiant 3n247 still work normally when pulled out of Nelson's circuit and tested within its limits as indicated on the data sheet?

Good question, but as i seem to not have it anymore, i can't answer. but i have no doubt it would still work normally as it never was working outside
its limits as far i had measured (except from the frequency).

And how does the HV radiant show itself? Corona, sparks, lighting up neons or CFLs, making people feel ill, or just how?

I personally never saw any of those things (Corona, sparks, lighting up neons or CFLs, making people feel ill,) but did not try the CFL's or neons.

That transformer primary-secondary thing still puzzles me. I'm wondering how the wire insulation withstands the HV stress.

The caps on the base puzzeled me as C4 has its + connected to the base, while C3 has its - connected, i changed C4 to be a bipolar elco (3th video).

The going into "berserk mode", i have noticed too, often resulting in damaging the Tip122 internally so that it still works, but not producing the initial oscillations.

It has severall states of operations depending on the supply voltage, like a continues oscillation, a packaged oscillation (where at around 8V you suddenly see the amps
go down) and again a (high amp / high frequency) continues mode which could cause some damage to transistor and/or elco's.

Itsu

[TinselKoala]

So your build did not produce "HV radiant"?  Maybe that is down to the transformer connection, then (step down vs step up) as padova commented.

Yes, I noticed the cap polarity thing too. Connecting an electrolytic the wrong way around can result in exploding the cap, that's no mystery! It seems that this circuit is hazardous to components, so it's no wonder that results are not entirely repeatable on demand.

Good thing I ordered a batch of 5 TIP122 transistors, I guess.