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

[a.king21]

Jack Noskills:
Just to be clear: Are you saying you wound your isolation transformer with the primary winding clockwise and the secondary anticlockwise?

[e2matrix]

T-1000 or anyone with knowledge on this - a question on using a Variac for trying this.  I've got a heavy duty Variac and wonder if that would be like a 1:1 isolation transformer when it is adjusted all the way to full voltage?  In other words 120VAC in and 120VAC out.  I think I also have a second isolation tranformer out of an old power supply. 

[baroutologos]

Seems that simulation and real world do not match..

Is this is what really happens, I don't actually care since it works.

the problem is, i am afraid,  that it does not.

[T-1000]

T-1000 or anyone with knowledge on this - a question on using a Variac for trying this.  I've got a heavy duty Variac and wonder if that would be like a 1:1 isolation transformer when it is adjusted all the way to full voltage?  In other words 120VAC in and 120VAC out.  I think I also have a second isolation tranformer out of an old power supply.

http://en.wikipedia.org/wiki/Autotransformer would not fit. It has single winding while it needs to be 2 coils on same core. The setup seems is quite simple from first look: TR1 gives power source and TR2 is connected as resistive load to TR1. But because of oscillation in TR2, it gives induction in its secondary and that increases impedance for TR1 while TR2 is under load...

[Jack Noskills]

T-1000, I tried parallel caps as you suggested and it helped. Also serial caps worked, right after upper coil before junction. If one would need to stop current flow without high self inductance then parallel tank circuit is the way to go. Earlier I had the opinion that series caps would be best but I take it back.

Those two trafos are not magnetically coupled and that is why tank circuits on both sides is the correct way. In Don Smith circuits you have aircored trafo magnetically coupled and there you need parallel tank in primary as to minimise current flow in there and series resonance in secondary as to maximise current flow.

In this setup however, you don't need any tuning caps. You minimise current flow by using high self inductance only.

a.king, winding direction matters at the secondary side but in case you have it wrong way no worries. If you connect the lower coil in wrong way then you don't get anything at load and all the power is used from the source. In this case power that is created by the first coil goes pass the load. If this occurs, just swap connections in either coil. We need to push from both coils to the middle at the same time. As they can run at full speed there is no delay of action, the source gets the same sinewave what it creates so it does not need to push more to compensate. In normal trafo operation the source always gets delayed version which then allows more current to pass.

One test I did was to take power directly from mains via 10 nF cap. I got no light at all. This confirms that we need a push, before we can push back. This pushing back then creates power that can be used. Also the power from the source gets used. Understand that there are three currents running: one current from the source, it goes via upper coil which creates current in lower coil. Lower coil runs and goes through the load, this then induces back EMF in the upper coil which is exactly in phase with source.

Same setup should work with higher frequencies so here is a proposal:
Take any ferrite core, wound wire on it until you get self inductance high enough at the frequency you want to use so almost no current flows. On the secondary side of this first trafo you put wire to meet your desired output voltage. For the second trafo, you can use bigger core and again enough wire on it so you get high self inductance at the same frequency. For the secondary side you then put the same amount of wire so it is exactly 1:1. Why bigger ? I noticed with nanoperm secondary that current dropped in the first trafo when load was taken. So stronger second trafo kicked back more towards first, hmm inductive kick anyone ?
If you are lucky enough to have high perm core, then you might find that you cannot get the frequency up as self inductance blocks current flow too soon. To solve this, you can add series cap, or even better try bifilar winding which adds capacitance. The more windings you can use the more flux you will get, of course.
To make a self runner, system needs to push back more and this you get with inductive kickback as I explained. Safety SG will be needed at the source so excess can be dumped. Self runner wont run too far unless there is always some load, so there needs to be a resistor at the load so you get higher kicks back, or a stepdown trafo in case you want to use higher voltage. For stepdown need to watch out so it does not block due to high self inductance, so use thicker wire there. When SG starts to fire at the source you can disconnect it from the starter battery. When you take power, kick back will be stronger and SG fires more often.
Hmm, I wonder if I just described what TK is using ? Did not realize it until I got to end of this thinking.

Here is my master plan. Anyone who does the above and makes a working prototype should get the second gold medal from Doc, I get the first one as one memento is quite enough. Your device need to give out more than 35 Watts, because that is what I am getting now. Now you will have your name on it and with help of Mr. Jones we go after bigger prizes. This OU prize here first, developer gets 80 % so he can develop it further: bigger, better, badder, well you get the idea. I might take 20 % to cover my own blown up meters and stuff I have spent for this. For bigger prizes I want equal share, just in case I happen to lose my nice job mysteriously. No greed, just that I got 7 mouths to feed. If someone then improves the prototype significantly then we make 40-40-20 split. This 20 percent will then be splitted again later in case someone makes another significant improvement and so on. Then we let anyone do the manufacturing, without anything coming to us unless they want to give. I need no hassles.

So boys, guys and fellow Duudson's, the race is on !