Quote from: TinselKoala on June 03, 2012, 03:06:00 PM
Actually, what I said was that the LED can be made to light in _either_ direction, depending on where and how the additional inductances are added. You can't just take the LED out, turn it around and put it in and have it light up the same. But with an inductor in parallel you sometimes can. And again, during the oscillations, there is little or no real DC current visible at the main CVR.

I'm surprised that no one caught this. I thought it was relatively obvious, so I didn't mention it. TK, the average current through the circuit is from the battery. When you install an inductor in parallel to the reversed LED, it lights up only because of the energy release of that inductor due to the flyback effect.

You could prove this by installing a forward-oriented 1N4007 in series with this arrangement, and the reversed LED should still illuminate.

Quote from: Rosemary Ainslie on June 03, 2012, 12:19:24 AM
Guys - POST 2692 - upload that waveform PLEASE to keep record.  It's highly significant.  I have NOT been able to repost it as there's some kind of 'control' applied.

Meanwhile here's the post that I wrote.  For some reason my previous post did not take.  And for this contribution I needs must break my silence with respect to TK's work. 

Guys it is simply the most significant result that I have ever seen.  Please scroll  back there and take a long hard look.  We've aimed at this result but never achieved it.  Only smaller less significant excursions into that battery negative voltage.  Amusingly it has not even drawn comment from TK's usual acolytes.  I'm only posting about this because we'll be including this in our own demonstrations.  I freely admit that this was PRECISELY the definitive battery voltage value that we've been looking for and only managed in very small moments and at much smaller values.

Well done indeed.  I realise that this is FAR from your intention - but you've assisted us all greatly in this result - from a circuit DESIGNED by Poynty and - as ever - BRILLIANTLY defined in all its essentials - by Groundloop.  We are all of us always deeply indebted to Open Source.  And while there is NOTHING else of value in this thread related to your work TK  - this experimental evidence is GOLD. 

Thank you TK. 
Kindest regards

Rosemary,

The circuit diagram you referenced bares little resemblance to the circuit I posted and that TK built. If you want to obtain the wave forms TK posted, you need to build it similar to the way shown below.

If you build the other circuit, chances are that it won't oscillate because there won't be enough inductance in the circuit. My design calls for significant values of inductance, purposely installed in strategic places.

Yes. Those 2 vids are exactly what I believe. Sorry I missed those. Thanks for reposting.

So to get a reverse led to light in the circuit, is it the inductor producing the reverse current to do so? When it is switched off that is. ;] And when in the proper position, as you say.

Or is it as you said where the "led is across an inductor", and the reversed led is just taking on the continued forward current from the collapse during mosfet off time? In this case it looks as if the led is in reverse in the loop of the circuit, but really does nothing other than recycle the inductors collapse.

Mags

Quote from: poynt99 on June 03, 2012, 03:58:14 PM
I'm surprised that no one caught this. I thought it was relatively obvious, so I didn't mention it. TK, the average current through the circuit is from the battery. When you install an inductor in parallel to the reversed LED, it lights up only because of the energy release of that inductor due to the flyback effect.

You could prove this by installing a forward-oriented 1N4007 in series with this arrangement, and the reversed LED should still illuminate.

I hear ya. Just posted about the same thing. ;] When you say flyback effect, do you mean the continued forward current produced by the inductor? Fly"back" could be a confusing term.  If that reverse led or diode isnt across that inductor, the inductors forward emf will try and a lot of times succeed in breaking down the switch(reed, etc) gap or transistor barrier and disperse through the circuit, in forward motion. Like in a switching supply, if there were no freewheel diode to a cap, the inductors forward current is absorbed safely as a charge in the cap, instead of causing breakdown of the switch.

But if the collapse current doesnt break down the switch, the inductor will produce a reverse current, giving another go at breaking down the switch, but using another polarity to do so.



According to the original circuits showing leds in the loop, I believe there were 2 in parallel, one reversed. in the loop. But I guess we have veered away from that way for some time now.

Mags

I've made a quick video illustrating Altoid's behaviour using the single LED load and an extra 2.2 uH inductor. It's nice to have a "pushbutton" overunity demonstrator on my bench! (Even if it is really a slide switch instead of a pushbutton.) Now if we could just figure out a reactionless inertial drive that fits in an Altoid tin, and combine that with an Altoid buoyancy drive.... we could start building our Altoid UFO.

Video uploading now, it will be half an hour before it's ready, probably.