Quantum Energy Generator (QEG) Open Sourced (by HopeGirl)

[isim]

@synchro1
Cylindrique coils, bifilar or not, with same inductance and global overall geometry have the same effects... May be just a very small difference in theirs intrinsic capacitors due to their different voltage repartition!
@+

[TinselKoala]

The bifilar coils throws a larger spark across a greater spark gap distance than a single wire coil with the same Joules of discharge. Tinselkoala demonstrates just that "Larger spark" feature in his bifilar comparison video. The speed and magnitude of the magnetic field collapse determines the back spike voltage. This is how TK's reaching 500 volts in his capacitor.

Synchro, I have asked you several times to stop misrepresenting my work.



THERE IS NO COLLAPSE SPIKE HAPPENING ANY WHERE IN THE ENTIRE microQEG SYSTEM, including the 500 volt receptor. The system is _sinusoidal_ all the way through. The 500 volts + is a result, I believe, of simple transformer action, combined with the usual VRSWR: voltage rise due to standing wave resonance.

NO INDUCTIVE COLLAPSE SPIKES are used to pump up any capacitors in this system. I am simply half-wave rectifying a 300 -350 kHz AC sine wave with the UF4007 and feeding the resulting voltage pulses to the capacitor in the ordinary AC-to-DC method, except it is happening at high frequencies. Since I am not using metal cores, I don't experience magnetic saturation effects that limit the performance of the system.

ETA: That is Tesla's biggest and most neglected "open secret" by the way: the use of non-saturable cores in inductors. This allows magnetic fields to build without practical limit and allows full action from the highest frequency components of any signal.

[TinselKoala]

@synchro1
Cylindrique coils, bifilar or not, with same inductance and global overall geometry have the same effects... May be just a very small difference in theirs intrinsic capacitors due to their different voltage repartition!
@+

In this particular case, this is what I think too. The reason that I don't expect any effect due to the TBF winding is just that there aren't very many turns. In a larger cylindrical TBF coil the inter-turn capacitance can become large enough to cause detectable differences in certain applications. But in a six turn coil operating at 300 kHz I don't really expect to see much difference. However I am interested in the issue and of course I will be testing. I just haven't had a chance to make the "normal" test coil yet, have to set up the bandsaw to cut a piece of mailing tube for the form, pull out some more house wiring, etc. Maybe I'll get to it later this evening.
Some people believe the co-resonant tuning will be easier with the TBF coil, but it's pretty easy already so I may not be able to detect such an effect.

[TinselKoala]

@Farmhand: Looks pretty good already! You may get better results with fewer turns of heavier wire, and a larger capacitor to re-tune to the present frequency, in your transmitter section. I think there is some optimum "balance" between the inductance of the transmitting loop (should be low I think) and the capacitor in the tank. I don't know what the right relationship is though. I think of it as two reservoirs with fluid sloshing back and forth between them. If the reservoirs are unmatched--- one big shallow lake and one small deep pit -- the calculated resonant frequency will be the same as one with matched reservoirs of equal size, but I think the matched situation will result in better voltage rise in the resonant tank. Maybe this is the concept of "Q" working through my muddled brain. So the reservoirs of the inductance and the capacitance should be somehow matched in this way for best performance, even though an infinite number of L and C pairs will give the same resonant frequency.
Am I making sense? Maybe I need more coffee.

[TinselKoala]

As an aside: some other folks are using similar circuits (Royer-Mazilli-ZVS oscillators) for various purposes and are having trouble with mosfet heating and failing and other issues. It's clear that my mosfets are not heating significantly, but that's because I've destroyed  many dollars worth of mosfets finding out what makes ZVS circuits into non-zero switchers. Even though you might be operating at below 1 MHz, things that you might not expect will throw the switchpoints off the zero voltage, and this more than anything will make the mosfets heat up. That is, in the normal operation of these circuits, one mosfet turns on and the other one turns off at the zero-crossings of the oscillator's waveform. It doesn't take much asymmetry in components/construction/etc. to throw off this zero-voltage switching from the zero-crossings and cause the mosfets to heat up excessively.

In fact the uQEG is less symmetrical than I would usually build, because the board is designed to mount to the miniSlayer base and I had to be able to get to the screw connector block for the primary coil. Full symmetry would have caused this socket to be inaccessible once mounted to the miniSlayer base.

And obviously one should use low Rdss mosfets to begin with. I was kind of surprised to see the IRF830s work so well. But the IRF3205 works great, even though it is being operated close to its voltage limit in this circuit. With different mosfets the input voltage can be boosted to 24 volts without other changes in the circuit and then you will really see some powerful effects downstream. If you put in Gate Zener VR diodes you can go to 36 VDC input and start melting things down.