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

[avalon]

Avalon.

Using TK's unit, I would suspect that standing (or at least colliding) waves may be had
with 2 TX loops connected in series placed apart, with the RX loop in between.

I agree. I also suspect that a properly constructed waveguide might increase the effect significantly.

[Jack Noskills]

Nice to see some experiments going on.

Look at the two trafo experiment. It showed that input in the first trafo does not change when load is added. If you have watt meter then put it in the mains. No matter what stuff is connected after it that consumes power should not affect. If the coil in the second trafo is connected in wrong way there will be power consumed in the input while output has no power. If you cannot see this effect then something is different compared to my test setup.

Capacitor is needed only if idle current should be bring further down. It can be difficult to get it right and if C is off then your results will go down. I had some caps and I was lucky to tune the trafo closer to optimum.

Do not use two different primaries in same core connected in series, only one. I tested with two cores so that the coils of the second trafo were across two toroids and efficiency dropped. The nanoperm cores made some funny rattling noise which means that there were created new frequencies which varied all the time. This noise I learned from my previous testing when I still had working meter. This makes me think that two primaries in the same core is also not a good idea.

Use sine wave, any other waveform will not replicate my experiment. We need to replace the signal that comes in with one generated in the second trafo and nature is not very good at producing square waves.
Using ratio other than 1:1 will not replicate my experiment.
Using coils with low self inductance will not replicate my experiment. There has to be lots and lots of wire when using grid frequency, or you need to have good luck with C.

My nanoperm cores were small, 63 mm outer diameter and permeability of 80000. They had 280 meters of 0.31 mm wire for primary and secondary, I used 63 strand Litz wire so very easy to make. When used as normal trafo it could deliver about 100 watts. The idle current was too high though, around 80 mA when 280 meter coils were connected together. So I started to wind new one, this time I try to put 400 meters of wire on both sides using plain wire. First unwind the litz wire, then wind 10 meters at a time. Very very  time consuming. Been doing that for the past two weeks now, little by little, almost complete.

I see there are some experimenters here with access to signal generator, I want to propose a simple test:
1. Wind two sets of wires of same length around ferrite toroid.
2. Feed pure sine wave into one coil and increase the frequency until current drops to zero. This will be the operating frequency. If you cannot reach this level then add more wire until this condition is reached.
3. Connect the coils in bucking mode as shown, use the frequency found in step 2 and begin from low voltage.
4. Put some load and observe the current going on in primary. If possible use same load as current limiter on primary side in case something goes wrong. You dont want to burn your equipment.
5. Crank up the voltage, carefully.

Understand, output does not affect input. What does it mean ? On the input side you have charge banging back and forth which is reduced only by coil resistance, the resonance condition. Each bang creates copy of it on the output side which you can use. Since output does not affect input, input is not slowed down and power is not consumed. Input side is happily running at full speed while output side is giving out power. Now I am using 50 Hz frequency and I suspect that with 400+400 meter toroid I can light up 150 watt halogen lamp without consuming any real power, I just need the pressure. Now, imagine frequency is cranked up to 1000 times. Will I get the charge 50000 * 2 times per second and 150 * 1000 watts out ? I think so but I dont have the equipment to move to this level.

I do not write this stuff to keep me warm. I am looking for help to explain/replicate why output does not affect input and then improve on that. Also, why the power is close to zero at N line while there is power in the output of second trafo ? If you connect two lamps in series between L and N they are both lit to same brightness. When connected as shown, why there is power in the lamp at output of trafo but there is no power in the lamp near the N line ? I dont think this is normal.

I do not ask for people to go and buy lots of expensive stuff, there are already others that have correct equipment to do a simple test that will take atmost one hour of their time. If you can do this test and it does not work then lets debug it and improve this thing together, this should be R&D thread.

[Jack Noskills]

Steve/wattsup, how did you measure the idle current in primary ? It occurred to me that maybe this is the difference if watt meter is used. It shows 2 watts consumed by coil resistance, not what is going through the coil. I measured it with light bulb like this:

L - primary coil of trafo - light bulb - N

There was very little light on the bulb, current is below 40 mA. When two coils were connected to get it in OU mode then this went further down. This happened with iron core and with nanoperm core. I need over 2*280 meters of 0.31 mm wire when using 80000 perm core to get it in this mode. When using iron core which has 5000 permeability there needs to be more wire. This makes me think that maybe trafos that you guys are using are not suitable for replication without tuning capacitor.

[baroutologos]

Jack,


its not about peoples' trafos and your trafos. its just what you say does not work.
One major mistake is that idle current - choked by impedance - could not be similar or even less when the bulbs is run per your configuration. No way per my view.


And yes, the idea of running a choke with an inverter (or signal generator sine wave) of adjustable large frequency till achieving choking inductance (almost null current) for given voltage, makes sense assuming the basic concept worked in plain 50 Hz iron trafos.


we have not seen anything so far contradicting enstablished electrical views my friend

[a.king21]

Jack, I personally think you may be on to something which is why I am on this topic. Can you be a bit clearer in your instructions please? For example you ask us to wind two coils on a toroid. I am sure you know what you mean, but there are many different ways to wind a coil. ie bifilar? cw-ccw, twisted pair, litz etc etc. Really, a picture of what you mean would help us enormously.
I am not using pure sign wave, but I noticed in one experiment that the transistor producing ac started to smoke! lol. Fortunately it recovered later.
So something is happening, I am not sure what yet. Please.... clearer instructions, it will save us all time.