Negative discharge effect

[ayeaye]

I think this and permanent magnet motors, is all about getting energy out of these fire wheels, which are called atoms.

How powerful should a free energy device be, well, a simple answer, 500w. Why, because the smallest electric radiators are 500w, such devices can give power to almost all house appliances. Also electric scooters 500w are useful things. But we can get only milliwatts, so a long way to go there. But it is only about research anyway.

[ayeaye]

I hope now i finally figured out, the only way how switching on the mosfet can cause any current in the opposite direction of the diode in my circuit. And indeed the current should be exactly such.

When the mosfet is closed, the capacitances of the diode and the mosfet will be charged so that the total voltage on them is equal but opposite to the voltage on the capacitor, because they are connected in parallel to the capacitor. Say some half of that voltage on one and half on the other, because both are some 50 pf.

This voltage stays on the diode, as the diode  is directed opposite. But this voltage does not stay on the mosfet, because its body diode is in the same direction. Yet the voltage that is equal to the threshold voltage of the body diode, i don't know how much it is, maybe 1 volt, stays.

Then when the mosfet switches on, the current starts to flow from the diode's capacitance to the + plate of the capacitor, and it rapidly grows. This current tries to increase the voltage on the diode's capacitance, to become equal to the voltage on the capacitor. Because opening the mosfet short circuited the voltage on its capacitance, and thus the capacitances in parallel to the capacitor no longer have the voltage equal to the voltage on the capacitor.

The diode's leakage current is extremely small, some 1 microampere, and it doesn't change, so it likely can be disregarded.

I had to consider the mosfet's and diode's capacitances, and even the threshold voltage of the mosfet's body diode, things which people usually don't consider when thinking about circuits.

Maybe i were lucky in using a large rectifying diode, that has a great junction capacitance, as it could not work with a smaller diode.

Now when using a relay instead of the mosfet in that circuit, some 50 pf capacitor may have to be connected in parallel to the relay's output, as weird as that may sound.

Neither may there be an escape from the self-oscillation, which happens with the frequency determined by all the capacitances in series, which is very small, thus the oscillation should have a quite high frequency. This self-oscillation should mostly only change the voltages on the diode's and mosfet's capacitances, which are very small, and almost not at all the voltage on the capacitor. Capacitors in series work that way, find that out if you don't believe. And because of that self-oscillation the effect would occur only on certain frequencies. Because the self-oscillation should start after the end of the voltage spike. Now when the periods of time from that to switching off and then to switching on are such that the voltages on the diode's and mosfet's capacitances are again such as they initially should be, then everything should work as described above.

Why it self-starts, i have no other explanation, than that the voltage on an electrolytic capacitor after short-circuiting it never goes completely to zero.

The input power of that circuit is that necessary for charging the mosfet's input capacitance, some 500 pf, to the voltage on the gate, that is 5 volts when using arduino, once in every switching cycle. I see no way how any of that charge goes to the capacitor in that circuit, all the mosfet's capacitances will be discharged and i think all their energy is just wasted. Yet formally for overunity, the power at the mosfet's gate in this circuit has to be considered the input power. And whether there can be any overunity also considering that, i'm not quite sure. But i don't exclude that either.

Now this all may sound complicated, but it may be the only way to achieve an asymmetric current.

Sorry for using this thread as a kind of blog, but i think this information is very important for understanding how that circuit works.

[TheComet]

You guys realise the power is coming from the microcontroller, right?

[ayeaye]

You guys realise the power is coming from the microcontroller, right?

Why asking in that tone, only want to find out things, knowingly not your enemy.

Cannot exclude anything. Have you realized that saying that there is overunity, is an offense? The same as saying that anything is true beyond doubt. Like Higgs boson has been found. It is an offense because it does not allow anyone to think differently, does not allow anyone to doubt. This is why it is an offense. All one can say is what evidence there is.

I see no way how any charge from the mosfet can go to the capacitor, the only way how the power can come from the microcontroller in that circuit. Because it can only happen when the voltage on the mosfet's capacitance is greater and opposite to the voltage on the capacitor. By the same direction i mean in the same direction in the closed loop, like in the Kirchoff's law. Like two capacitors in parallel, when can one increase the voltage on the other, think about it. But how can that happen there when the voltage on the gate-source capacitance, which is the greatest voltage there and the only voltage created by the microcontroller's output (the input voltage), is in the *same* direction as the voltage on the capacitor. If you disagree, what you can do, show how can the charge on the mosfet's input capacitance created by the input voltage, go to the capacitor in that circuit. Everyone has a burden of proof, otherwise some have no responsibility.

That but, i will also measure the currents there before the voltage spike and during the voltage spike. Whether the former is less than the latter. This shows whether there is overunity in the coil, when we look at the coil separately. Showing overunity in the coil is important, for theoretical reasons, and this is all what i'm really interested in.

Overunity in the coil though doesn't necessarily mean overunity in the circuit. Because the circuit has to be properly considered everything including the mosfet, including the current at the mosfet's gate. One may also go further and say that the input power is the power consumed by the microcontroller. The power of generating pulses has not been tried to be made minimal in that experiment, for lower power the pulses have to be generated by cmos gates, which is though more difficult to adjust. But anyway, whether it is is or is not considered overunity in that sense, is not really important, what is important is the research.

[TheComet]

Why asking in that tone, only want to find out things, knowingly not your enemy.

I apologise. Let me explain in better detail what's going on here.

The circuit can be simulated and analysed, there is a very logical explanation as to why the capacitor slowly charges. Attached you will find the simulation file which you can run yourself using the free program LT Spice from linear technology: http://www.linear.com/designtools/software/

I modelled the circuit according to attachment 0.png. As I mentioned earlier, the power is coming from the micro controller. Micro controllers have a push/pull output stage which can deliver up to 25mA typically. This means the pin can sink or source 25mA at whatever output voltage (typically 3.3V).

If you look at attachment 1.png you will see the relevant measurements. The micro controller is outputting a square wave at 10kHz with 4% duty cycle. If we measure the voltage over the coil, we see spikes. Why? This is basic boost converter theory and can be further researched here: https://en.wikipedia.org/wiki/Boost_converter

As to where the voltage is coming from in the first place, if you measure the current on all three pins of the MOSFET you will notice that current is being transferred from the the gate to the other pins via the MOSFET's parasitic capacitances, i.e. the current is coming from the micro controller. For a more accurate model of the MOSFET, see this image: http://powerelectronics.com/site-files/powerelectronics.com/files/archive/powerelectronics.com/images/Fig5-1-0515.jpg

If we look at the current going through the diode, we see that there is an average current of about 220nA. This means the capacitor is being charged with that current, and we are able to make a prediction about how much voltage will exist over the capacitor after a specific amount of time.

Is it overunity? No, it is not. We can prove this by impedance-matching the output of the circuit with the micro controller's power source and measuring the energies. We see that the micro controller has output way more energy than what was consumed by the load. Therefore, this device is not overunity.