Bifilar pancake coil overunity experiment

[ayeaye]

Look, you can poke around inside of just about any circuit and find two measurements that aren't the same. And if you call the smaller one "input" and the larger one "output"... well then, COP overunity!

No, again, oh no, i were not talking about whatever two measurements but measuring very well defined things, the coil's input and output power. And my experiment was only about measuring the overunity of the coil, or whether there is any. If you want to measure something else, this may be interesting for you, but not relevant at all, at least not for my research.

Make sure to measure the resistance between the oscilloscope ground and the circuit ground. I used a floating scope as i measured that there was no connection, but if there is, it can first show very weird things, but second it can damage your equipment. I think you know that, but that's a reminder like the map reader reminds things to the driver, one may always forget things.

There is very little effect from connecting or disconnecting the coaxial bifilar coil

This is very very weird, my oscilloscope traces showed clearly a typical voltage traces for a coil in channel 1. Your coil may have less inductance than mine, as i said mine was in effect two bifilar pancake coils in series, which together was 34 turns, but that it didn't show any exponential coil traces, that's very weird. Your channel 1 is zero, for whatever reason, there is nothing at all to see on your oscilloscope screen.

I have also tried connecting the transistor "incorrectly"

My transistor was connected correctly.

So far no useful results from you for any purpose, thank you for your effort though.

I also read the Rigol manual. It can multiply two channels yes. Then one can take manual measurements, which can then be written like to a spreadsheet and then do calculations, but this is rather a nuisance to do. More it cannot do, as it cannot do any analyze of the math trace, for that one has to select one of the real channels. I don't know what software comes with Rigol, but by what people tell, it seems that they can only save data from the serial interface. And no, this is not in the csv format, this is a series of bytes, each byte is for one sample, without a Python script i don't know how they can be read. I don't know though, there may be software that can convert it, but these people didn't know any, and one cannot search all life, something that is so easy to do with Python.

Depending on what Rigol you have, some can save csv data to usb stick yes.

[ayeaye]

Tinsel, Ok, do it then, import the samples from the oscilloscope to a spreadsheet if you so prefer and do calculations in the spreadsheet. It is not so easy to work with 600 numbers for both channels on a spreadsheet, but do it if you so prefer. Or make the oscilloscope to show the multiplication of two channels, cut it out of paper and weight the paper if you so like, however you prefer to do it, but do it.

Tinsel, please understand what is the coil's input power and the coil's output power. These both have to do with the voltage on the coil and the current going through the coil. The coil's input power is positive, it consumes power from an external source, coil's output power is negative, it generates power that doesn't come from any external source.

You can simulate the coil's input and output power in a circuit simulator. And it certainly will show unity, not an arbitrary values as you say. Because induction is a very symmetric process, and by its official equations it always has unity, circuit simulator never shows overunity. And induction is a very symmetric process indeed, the only matter is whether there is some other process also, like creating charge locally, that doesn't have a Lenz effect back, that may make it asymmetric.

Voltage on the coil multiplied by current going through the coil, this is the coil's power at any moment of time. The power is positive when the coil consumes power, and negative when the coil generates power (back-emf). Is it so difficult to understand that one is never able to understand it?

The voltage on the coil Vl, and the current through the coil. Il, at any moment of time the power in the coil

Pl = Vl * Il

It completely freaks me out that so simple thing is not understood.

[Void]

Voltage on the coil multiplied by current going through the coil, this is the coil's power at any moment of time. The power is positive when the coil consumes power, and negative when the coil generates power (back-emf). Is it so difficult to understand that one is never able to understand it?

The voltage on the coil Vl, and the current through the coil. Il, at any moment of time the power in the coil

Pl = Vl * Il

It completely freaks me out that so simple thing is not understood.

Hi ayeaye. Just dropped by this thread to see what was being discussed.
What you are saying above does not make sense. Except for the resistive losses in the wire
in a coil, coils do not consume power. You do not calculate power for coils like you are
suggesting here. In AC circuits, you would calculate the power being consumed by some real *load*
in the circuit (could be resistive losses in any components in the circuit or an actual external load),
and you must always take power factor into account when calculating or measuring power in AC circuits.
If you looking for over unity, you are concerned with input power versus output power across a load that is
consuming real power. 'Reactive power' has no bearing at all when trying to measure or calculate for over unity.

[ayeaye]

in a coil, coils do not consume power. You do not calculate power for coils like you are
suggesting here.

Oh yes they do. And at any moment of time the power is always calcuated in the same way, with everything. It doesn't matter where the power goes, to the losses in the coil's wire, or stored in the coil in some other way, this is how power is calculated. At any moment of time, and this is what matters, as i calculate energy for every small amount of time, and then add them up. Calculating power for some waveforms is different from that, as it doesn't calculate power at every moment of time, but for the whole waveform.

Unless you want me to learn wrongly from you, which i don't.

[partzman]

FWIW, here is an energy test of a bifilar coil in a similar arrangement to yours but idealized for digital scope measurements and the schematic is attached below for probe orientation.  The Math(red) channel is used to integrate the instantaneous measurements with the result displayed in mean or average.

The first scope pix shows the input power to the entire circuit and is seen to be 2.15 watts.  Since we will be comparing differing time intervals, we need to convert this to energy which will be 2.15*4.36e-6 = 9.374uJ.  In this same pix we can also calculate the energy dissipated in R1 from the average current shown in CH4 which will be UR1 = (109.1e-3)^2*100*4.36e-6 = 5.118uJ.

The next pix shows the dissipation in R2 which is seen to be 750.8mw.  This is calculated by using the square of the instantaneous differential between the supply voltage and the voltage across R2 and dividing by 100.  Therefore the energy dissipated in R2 = 750.8e-3*4.36e-6 = 3.273uJ.

The last pix shows the power dissipation in R2 from the discharge of energy stored in L1/L2 and is seen to be 24.54mw.  Therefore, the energy is UR4 = 24.54e-3*31.56e-6 = .774uJ.

So, the total energy consumed by the entire circuit is 9.374uJ and the total energy produced or dissipated by the circuit is 5.118uJ + 3.273uJ + .774uJ = 9.165uJ for a COP = 9.165/9.374 = .978 neglecting resistive losses in L1/L2 and the mosfet.

This does not mean that bifilar windings can't produce excess energy, it just means this particular example does not appear to.

Regards,
Pm