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

[TinselKoala]

You took apart a precision Simpson current transformer just for the cores?

This makes my heart hurt.

http://www.surplussales.com/inductors/FerToro/FerToro-3.html

http://www.surplussales.com/Inductors/Inductors.html

[gyulasun]

....
I measured the resistances as follows:

Primary:  21.3 ohms
Secondary:  35.9 ohms, which is a bit surprising since these are 1:1 transformers, but evidently a smaller-gauge wire is used on the secondary.

I have an inductance meter, but it only goes up to 20 Henry -- and both the primary and the secondary exceed this value.  I doubt the cores have high permeability; but have found no data on this.

....

Hi Steven,

The higher number of turns in the secondary coil(s) comes from keeping the 1:1 voltage ratio at full load and compensate for copper and core losses. I wrote about this for you here:
http://www.overunity.com/12487/simple-to-build-isolation-transformer-that-consumes-less-power-than-it-gives-out/msg328931/#msg328931  probably you did not notice it.  However, the difference between the primary and the secondary coils in their DC resistances seems rather high in your trafos case, do the secondary coils in parallel provide the normal 117-120V AC output when loaded (and the primaries are also in parallel).

To measure the coils inductance, try to connect the primary coils in parallel and also the secondary coils in parallel, this should bring the inductance below the 20 Henry range on your L meter.

rgds,  Gyula

[wattsup]

You took apart a precision Simpson current transformer just for the cores?

This makes my heart hurt.

http://www.surplussales.com/inductors/FerToro/FerToro-3.html

http://www.surplussales.com/Inductors/Inductors.html

@TK

No I did not take them apart yet since I have them. I just opened the cover. That's why I am posting it here to see if it is a smart move or not to risk. I get what you mean about the precision.

Otherwise, let's cut through the chase and get some NanoPerms. hehe

Now when I search their web site.......
http://www.magnetec.de/en/nanopermr-products/

I am puzzled as to why there is no mention of their cores being used in isolated transformers. Mainly for chokes and current transformers but no mention of isolated or step/up/down transformers.

What model choice is the other question???
Something around the 4" (100mm) diameter range.

wattsup

Added:

My standard transformer X coils are 2.2 ohms and the H coils are 1.5 ohms. I had taken for granted that all the coils were identical. Seeing that, I connected the H side to the AC, since in my last trials the X side was on the AC. No difference in final results. So measuring the resistance or asking what the resistance of the coils are before buying them is a smart way to try and find the right ones.

[Jack Noskills]

Yesterday I got me a watt meter, I wanted to see how much that 150 watt halogen uses. I realised I can do comparison so that I measure only what I take from the mains and just check out brightness of light.

First I measured the 2*400 meter nanoperm in normal trafo mode. Meter showed 87 watts in primary side and I got some light out. I estimate the efficiency of the trafo to be atmost 90 % to be on the safe side. In generator mode I got 37 watts going in primary side and halogen is significantly brighter, somewhere between 100 - 120 watts. When I used smaller 40 watt load the watt meter does not act realiably, it can only measure above 5 watts. Based on my earlier testing I know power consumption can be get close to zero, milliwatt range.

I also tried 300 watt halogen. In normal trafo mode primary used 92 watts and halogen had just a faint glow. In generator mode I got plenty of light, enough to burn the damn desk again and watt meter showed 76 watts.

Seems that this core can create about 70 watts by itself, just an estimate and not exact figure. How I would love to feed this bugger with higher frequency and see how much more efficient it would be. Idle power of one coil is way too high, I think it was close to 60 watts. If I could get this down below one watt using for example 500 Hz then I would get 10 times more power out with better efficiency too.

wattsup, I think it is better if you get metglass instead of nanoperm unless you have access to toroid winding machine. magnetec says they have cores with permeability of 200000 but when I tried to get them I did not even get a decent reply if they are for sale or not. If metglass has one million permeability cores then that is worth a try.

[Jack Noskills]

I found nice DC/DC controller, it outputs squarewave if smoothing caps are removed and this could be fed in first trafo to make sine out of it. You can set oscillations from 1 kHz to 1000 kHz, there is a design example that shows how to make output 24 volts at 2 amps from 12 volt input at 600 kHz with schematics and equations how values are obtained.

Device costs less than one dollar, maybe it could be used as a driver.

http://www.ti.com/product/tps40210#feature

There are many others and did not check them out. Just made a search for a device that can output 220 volts and only this one came up. 100/220 volts would be nice to have directly as you dont need third core to step it down. Or maybe a proper voltage divider circuit could be used, dunno.