COP 20.00 (2000%) Times, Reactive Power Energy Source Generator,

[TinselKoala]

Illusions strike deep.

I've made another version of my own TKTickler Reactive Power Generator. Below you can see the images of the device. It uses a 6-turn Tesla Bifilar output coil, two IRF3205 mosfets, and runs on DC input at 12.5 V and about 0.6 Amps. It draws less current than the wireless power transmitter version, I think because of the coil's impedance.

I measured the impedance of the coil in isolation on my Pro'sKit meter and it read about 5 microHenry. I measured the capacitance of the capacitor stack and got a measurement of 64.5 nanoFarad. I measured the oscillation frequency directly across the output coil with the scope and frequency counter and got 303.4 kHz.

Running these numbers through the Resonant Frequency Calculator I get very close to the theoretical prediction. The coil computes to about 4.26 microHenry, taking the frequency measurement on the Philips counter as correct to 6 sig digs.

Now... the output power. I measured directly across the coil and get a pure sinus waveform at 303.4 kHz and 82 v p-p. But what about current? Well.... the coil itself is actually dropping that voltage across it, it is acting as its own "current viewing resistor". But its DC resistance is too low to measure. However, at 303.4 kHz the 4.26 microHenry of inductance produces a reactance of about 8.186 ohms, if I did the math right. So I can state that the coil drops the full 82 V over the 8.186 ohms impedance of the coil. So the current in the coil is (82/8.186) = about 10.02 A p-p.   Right?

So the output power then is 1/8 x V p-p x I p-p = (82)(82/8.186)/8 = about 102.7 Watts. Right?
The input power is 12.5 x 0.6 = about 7.5 W. Right?

So the COP is 102.7/7.5 = only about 13.7. Damn, I was really shooting for COP 20.00. Back to the drawing board, I guess.

(Surely I must put a cosine theta correction in there somewhere, but what's the phase angle of a signal with itself?)

By the way, the mosfets stay cool. The only components that heat are the 100R resistors and the chokes. I'm starting to think that mosfet overheating in these circuits is caused by too much inductance in the load (other causes having been eliminated one by one). I think that if the load inductance goes up, the choke values also have to go up, because the operating frequency will go way down and this means the chokes aren't choking like they should. I fried one 3205 mosfet when the cheap RadioShack choke overheated and shorted. The insulation on the wire they use is a joke. So I rewound both chokes with some high-temperature magnet wire to keep that from happening again. The chokes are critical in this circuit.

I can haz cheezburger now?

[TinselKoala]

By the way... note that the above images are 1024 pixels wide. I did this on purpose to show that this is _wide enough_ and doesn't play havok with the page width. There used to be a notice on the attachment types list that said this was the max pixel dimension. If everyone would please keep their images to 1024 pixels wide and less, the pages would display more better.

[wayne49s]

Surely we can agree on those simple points.

1) On the charge phase the power dissipated by the resistors cannot charge the capacitors.

2) On the discharge phase the power dissipated by the resistors cannot be returned to the supply.

If we can not all agree on those two points, we have strange things to discuss. 

If anyone disagrees with the two points above please say so and explain why and how it can be different.

..

P.S. Basically these people are claiming that they can draw power from the supply and dissipate almost all of that power in
the light bulbs and then return that same power to the supply.

Our job as experimenters is to determine what is actually happening because what I just described cannot be what is going on.
Can't have your cake and eat it too.

..

I don't think they are saying that. Jim Murray is just showing experimentally that the average positive and negative VA averages close to 0 (1watt).  If the power was produced by a generator/motor, half the cycle it is a generator/ half a motor , so average energy consumed is low. The current flowing through the resistance is always consumed in either current flow direction.


The effect of the resistor on the capacitor charge/ discharge is the RC time constant, so charge/discharge is slower with a larger resistance.  It effects the final charging voltage attained in the given charging (also discharging) interval.


So the bottom line is, do we know the circuitry to duplicate the experimental data that is shown?

[G4RR3ττ]

@TinselKoala, there's nothing weird with having a larger power out than in... gigawatt lasers do it all the time. For instance, having 1w in and 1GW out can be done, but with obvious duration and repetition limitations. What matters is finding the average power and multiplying it by the time duration to obtain energy. If E_in was less than E_out + E_stored you would have something novel...

[TinselKoala]

@Garrett
The measurements I'm citing are steady-state; you can measure that same output power and input power until the battery goes dead. The output in the coil is a pure sine wave at 303 kHz (roughly) and the input is straight non-ripply DC, so there aren't any weird spiky signals that are controversial and difficult to integrate.


But... you are preaching to the choir here. My point is that it's not hard to come up with measurements on certain apparatus that can look like OU. The recent hoo-hah over QEG's "overunity in VARs" and the topic of this present thread motivated me to produce some "overunity" demonstrations of my own, using simple electronic means that demonstrate the same principles.

I will say again: If anyone has a genuine electrical OU device with a COP of at least 1.3, much less 20,  I can make it self-loop. I don't care if the input is 20 kV at 30 mA RF and the output is 3 volts and a hundred amperes AC or whatever. If the OU COP is genuine and the inputs and outputs are electrical, I can handle the conversions necessary for self looping. The _fact_ that these reactive OU systems cannot be self looped and in fact can't even power loads efficiently means that they are not, in any real useful sense, overunity at all.

And just as with my circuit which has all that power circulating in the coil and caps, if I take it out it will either collapse the resonant condition and the system will die, or I can take it out just as fast as I am supplying it from the battery. The fact that it's a lot of power circulating, or residing in a capacitor as stored energy, is just peachy keen, but it won't run your microwave oven or your electric scooter any better than the supply source will all on its own.

And of course what I (laughingly) and others (more seriously) call "output" in this kind of situation isn't actually output at all. It's not "output" until it actually gets put out of the system! Thanks Farmhand for making that little bit of trickery explicit.

ETA: Scopeshot below. This is directly across the "output" coil with a 10x attenuated probe, API 510-10-1-A. The horizontal scale is 1 microsecond/division and the vertical scale is 20 volts/division, and the center graticule marker is the baseline. Input is 12.4 V at 0.64 A.