Simple to build isolation transformer that consumes less power than it gives out

[Lynxsteam]

I'll give it a try.  I have two identical Torroids with isolated 1:1 windings.  I have a single phase alternator that puts out a nice sine wave.  I will use a 1 watt bulb as a load, and see what happens.  I am optimistically skeptical.

Update:  I spent a little time with it and it definitely performs like a joule thief.  Not with AC, but if you pulse the DC, it puts out enough power to do more than the DC alone.  The flyback will light an LED whereas the 1 volt DC alone can't.  With AC, there's no sudden collapse of the field.   I can see where the spikes could add.  I'll fool with it some more.

[Jack Noskills]

I did some more testing and I think I see what is happening. I try to explain.

I removed first trafo and connected directly to mains via 1 uf cap so it acts as current limiter.
For comparison test I put two lamps is series after cap and both lit at the same brightness, hardly visible. Earlier I have measured that I get about 80 mA via 1 uf cap.

Then I added second trafo as shown in figure, in this case bulb B is brighly lit while bulb A has no light. When I disconnect bulb B, I get flash of light in bulb A. So beware if you want to use signal generator directly as it shoots back. With two trafos I got flash at bulb A when load is connected so it is safer.

Next I added more caps, 7 uf so they should give 560 mA. Bulb B lights a bit more, bulb A still no light. Then I put more load at B and now I see bulb A getting lit.

My interpretation is that when load exceeds what core can produce only then it starts to suck more from the caps.

It is important to get the first trafo to operate in a sweet spot. The coil I used had very thin wire, must be over 10000 turns. If someone tests with some other trafo you can test it easily if it is suitable. Connect it to source without load, there should be very little idle current.

If you have ferrite and idle current is too high, then put parallel cap at source so you get tank circuit. Then exactly the same kind of trafo on output side and put same valued cap in series, like in the figure shown. Now find the sweet spot where current drops closest to zero on first trafo. When you see it, connect some load, what is the result ?

This should be a resonant trafo, I haven't tested this so it is merely a suggestion. I would try it if I had equipment.
I have only tested with pure sine wave, as that is what comes from the mains. No idea what happens with other waveforms.

[newton2]

HI , my kind greetings to honoured profiles of this ouforum....and thanks for this interesting suggestion for further discussing basics for OU made by mutual inductances....and by reactive networks..the hitherto in this newly thread of topic and answering comments though still is in the socalled state of prestudy and prestudying discussions circumstances...might it kindly be hinted, that if arranging surden researching concepts and if keeping experimental layout to a few electrotechnical reactive components,
then it is by established physics and technics possible to obtain OU...if named energy-generation though , then the circumstance of whether the term energy is auto-consistent in nature...or possible more aike a supporting-tern of conveniences in established elder physics...by solely own researches once upon years back was proven and demonstrated an OU-device capable by such electrotechnics reactive components layout...it is possible to calculate this interesting thread´s electro-technical foundation...yes,tricky if when more than one inductances-mutuality appear in a common circuit..the transfer by mutuality-coupling and alterations of impedances if when applying a "load" or removing a "load"....might it kindly be refered to elder book-paper-versions of elder, especially the elder versions , of electro-technics theses and formulas of mutuality-reactive-connectios......THANKS INDEED FOR SUCH IMPORTANT BASICAL DISCUSSING...really thanks to thread´s author plus answering authors.....have all Yourselves a nice day and fruitfull results from Yout various hardlabored deeds and workings concerning the honourable goal of OU....p.s.as a mere brief hint : profile zweiternewton in the other ou-forum/my profile zweiternewton/ouforum.de/OU bei Reaktanznetzwerk...   

[T-1000]

This should be a resonant trafo, I haven't tested this so it is merely a suggestion. I would try it if I had equipment.
I have only tested with pure sine wave, as that is what comes from the mains. No idea what happens with other waveforms.

If to push this even further, can you adjust resonant drequency of second trafo to frequency of AC source? Also if you have step up+step down trafos, you need spark gap between them so the losses of step down transformer would be compensated with power coming from step up transformer on oscilation peaks.
The missing link here is exactly this: you need resonant LC oscilations on output transformer for maximum output. This can be found in various circuits including N. Tesla and Don Smith.

If you will play with it, here is my modded circuit to start with:

[Jack Noskills]

Hey Jack

Are you sure of the circuit connections in the drawing?  I ran it on Falstad sim. The way you have it, Im getting little output as the load is across the transformer winding and the winding doesnt seem to want to give up anything to the resistor/load.

I tried switching the connections of one of the windings on the second transformer. and now there is output, but is is somewhat under the input.

Maybe Im missing something.  Im interested though.  ;]

Mags

Seems that simulation and real world do not match.
Resistance of 40 watt bulb is 1215 ohms (220 V grid). When current comes to upper junction from left, it wants to go through higher resistance bulb rather than against current that comes from the second coil, which the first coil created. Resistance of second coil was 165 ohms for iron trafo and about 45 ohms with nanoperm.

When current comes below bulb, it wants to go through lower coil as it has lower resistance than bulb. As this creates current in the upper coil current again is forced through bulb in the same way.

I you imagine this as water pipes, water does not go upstream if there is even a tiny hole elsewhere.
Is this is what really happens, I don't actually care since it works.

The very first picture missed the dot. In the later picture dot is in correct place.