The problem with measuring the input to the SG from the battery,is that the SG uses 32ma at 12.5 volts befor anything is hooked up to the output side.So where already useing more power with out a load hooked to it.

Each element in the system needs to be measured and accounted for, in order to properly sort out any suspicion that there may be excess energy entering the system somehow.

It's also helpful to treat everything beyond the battery as part of the total load. Although the FG is supplying energy to the rest of the circuit, it too uses energy. As such, it can be likened to a resistor that passes on energy from the battery, while dissipating some power of its own. Therefore, it really is a load. The only true source in the circuit is the battery. Hence; PBAT - (PFG + Pcircuit) = 0

I did quickly try two other air core coils,one from a MOT and one from an old project-niether worked.
Im guessing this is because they have no capacitance.

Perhaps, but all coils exhibit some capacitance. In the case of a Tesla bifilar however, that capacitance is significantly increased.

Do you get the same effect from your sim .99 if you remove the 50p cap?

I did not try it as I am waiting for the DC measurement of your coil.

I also must admit that i am unsure how you can disipate more power on the output than you are consuming on the input?
To do this,wouldnt the components have to be over 100% efficient themself?

It's a matter of perspective actually. What you are calling the "input", isn't really the input at all; the FG is. The battery is the source, and everything else beyond it dissipates power and therefore is a load on the battery.

I'm sure my coil has capacitance... how could it not? I have no idea why my meters aren't reading it. They do read down into the low pF range when tested with actual capacitors.

My coil has inductance of 670-695 microHenry, measured three ways ( ProsKit MT5210 and Arduino inductometer, and resonating a tank circuit and calculating manually) and a DC resistance of 3.9 ohms (also measured with three different meters, ProsKit, Fluke 83 and Simpson 464). 60+60 turns of #27 enamelled magnet wire.

But I can't get a capacitance reading with either capacitance meter (Fluke, ProsKit). If I put a small cap (22 pF) in parallel with the coil, I read a short. If I put the small cap in series with the coil, I get the value of the small cap. Ditto with the other bifilar coil (the one used in the "coreless" JT video from a few days ago.) No capacitance reading, inductance about 125 microHenry, DC resistance 1.4 Ohms, 23+23 turns of plastic insulated #26 stranded wire.

I just made another matching bifilar coil to play with, and I'm changing the 1n914 diodes to 1n4004, changing the green LEDs to some other ones, and will resume testing after I've rested up some.

TK,

Did you remove the connection between the two coils to make your capacitance measurement with the meters?

You would then be measuring between two open-ended coils. I assume that is what tinman did.

I had strange effects measuring a spool bifi capacitance. Measuring capacitance between the coils(coils not connected electrically, open ends) like 11nanofarad, but when I connected the cap meter 'across' one coil, as if to measure resistance, it measured near 1uf. Either coil.

I was impressed with this amount of capacitance 'between' the coils. till I realized I was measuring across 1 coil.

Very strange.

Mags

Winding a bifilar pancake coil onto a CD-ROM blank that has been sprayed with 3M Super77 adhesive. The wires are coming off two spools and are gently wound and pressed into the adhesive as I go around, with a wooden popsickle stick.

Even stranger for me. If I select high ranges on my cap meters and measure across the total coil, connected at center tap, I can see _negative_  capacitance values on the meter.
There should be a buzzer "wwwoooonkkkk" that goes off when that happens, I think.
 

Lol, they are strange puppies, thats for sure. I tried many things with different coils to try and find some standard to use when determining absolute values and, well, here we are. Many years after its invention, and still no bifi standard calculations.  Weird. Maybe Tesla wrote something up on it somewhere.

Mags

Sorry, Ill stop here if these things are not relevant.  1 more thing.

It just came to me from my last post.

Here is a Tesla pat  Apparatus for Transmission of Electrical Energy

http://www.free-energy-info.com/TeslaPatents/US0649621.pdf

A 2 turn primary on the left and a 2 turn secondary on the right. Bifi coils by definition.

Ehh. Maybe Im nuts.

Mags

Well TK, at the very least it is an interesting effect.

So i took the day off today,as i wasnt feeling well.
But i managed to drag myself into the work shop (ofcourse)and do some measureing on the coil.
Following TK's video tutorial,this is what i came up with.

I used a 1uf ploy film cap-same as TK's
To get the two o volt gaps in the wave form to sit behind the 0 volt line,and have the high and low side of the wave at peak-the frequency was 2.450KHz
This is what i have apparently
0.00422 henrys
4.22 milliHenrys
4.2200e+3 MicroHenrys
I tried the test useing half the square wave(used a diode on output of FG)and also useing the full square wave(no diode inline)
Both results came back the same.

The coil resistance is 6.5 ohms-which i think is quite high for such a small amount of wire.

Now the next size cap down from the 10nf cap i had was 2.2nf.
So i decided to give it a try any way-so out with the 10nf and in with the 2.2nf.
Now the highest frequency i can get is 72.45KHz.
I think i need to go even higher again,as the input meters are still dropping when i hit the highest frequency.
I will let you see the result's so far in my next video(which im just about to do) But on the first run,i blew the green LED befor i got to the peak frequency.
So now im back to two super white 10mm 3.5 volt LED's

Unlike TK,the brightness gets clearly higher as i go up in frequency.
So i am going to use a small solar pannel over the LED's hooked to an ammeter,and see if the current in the cell keeps rising throughout the lift in frequency.

After reading TK's post about not seeing much diference in the light output of the LED's,i decided to place a solar pannel with a 1k resistor across it and test the voltage output from the pannel as we lifted the frequency.

The setup is now running at a top frequency of 74KHz.
The diference between the input and output meters is very large now.
But it seems that that is what the ma draw would be with those LED's running at 4 volt's.

I tried my analog meter ,and that read about 2ma less than the DMM.
Another thing i noticed when i changed the 10nf cap to a 2.2nf cap on the SG,is it now draws 32ma from the battery with nothing conected to the output of the SG ?.
The SG as it is seems to be a fairly inefficient unit.
Maybe a better one could be made.Something that covers from 30KHz to 100KHz.
https://www.youtube.com/watch?v=v7MbkgMehNQ