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

[isim]

@TinselKoala
"Ok, that's expected. Now I believe that the true phase angle should be close to 90 degrees, probably around 87 degrees. A couple of different measurements have arrived at the 72 degree value, though, like this most recent one with the CSR. I would like to be able to "balance the books" if you know what I mean. So what can account for the reduced phase angle I seem to be measuring on the scope? How can I tease out the separate contributions to the phase angle measurement, without spending any money or developing special instrumentation buffer amps and filter networks? "

I do a simulation of your miniQEG, and I found that to have a Phase shift of 74° (approximately), I need to load the OC(oscillating circuit) with a serial resistor of approximately r=1.5ohm (plus R=0.25ohm). For me this resistor is:
1) an default in your circuit. (I don't thinks so)
2) an external load, your "receptor" for example.
3) an other parasitic load???
If this resistor is <0.5ohm the shift phase become 87°...
I will give you more information and pictures tomorrow evening.
Can you do a test with and without the "receptor"?
Thanks
@+

[TinselKoala]

@TinselKoala
"Ok, that's expected. Now I believe that the true phase angle should be close to 90 degrees, probably around 87 degrees. A couple of different measurements have arrived at the 72 degree value, though, like this most recent one with the CSR. I would like to be able to "balance the books" if you know what I mean. So what can account for the reduced phase angle I seem to be measuring on the scope? How can I tease out the separate contributions to the phase angle measurement, without spending any money or developing special instrumentation buffer amps and filter networks? "

I do a simulation of your miniQEG, and I found that to have a Phase shift of 74° (approximately), I need to load the OC(oscillating circuit) with a serial resistor of approximately r=1.5ohm (plus R=0.25ohm). For me this resistor is:
1) an default in your circuit. (I don't thinks so)
2) an external load, your "receptor" for example.
3) an other parasitic load???
If this resistor is <0.5ohm the shift phase become 87°...
I will give you more information and pictures tomorrow evening.
Can you do a test with and without the "receptor"?
Thanks
@+

I've tried to keep the tank circuit current paths as short and symmetrical as possible and I'm using heavy conductors and huge copper areas on the PCB. I am pretty sure there isn't any great resistance introduced in this current path by my construction.These most recent measurements were done with IRF830 mosfets though. Could the higher Rdss of these mosfets be the "missing resistance" in the tank circuit to account for the phase discrepancy? But the mosfets are in screw-clamp sockets for ease of replacement, these may be introducing a little bit of series resistance over what a soldered connection would do.

As far as I can tell by casually looking at the scope, the presence or absence of an external load, the "receptor" driving a lamp or motor, doesn't affect the phase angle, just the amplitudes of V and I and the frequency slightly. The phase measurements above were made with the receptor and light bulb in place but not driving the motor.

ETA: This turns out to be incorrect, there is a phase difference whether or not the motor load is being driven. See below.

Yes, later this evening I will be happy to do some actual phase angle measurements with and without the receptor driving a heavy load. But I am all out of really low Rdss mosfets and will have to stay with the 830s until tomorrow, when my replacement 3205s are scheduled to arrive.

In your simulation, I imagine you cannot simulate the Tesla Bifilar winding of the primary tank's loop. I will also make another loop coil with ordinary solenoidal winding but the same amount of wire, for comparison. But of course that will remove the "tesla magic" from my  apparatus.

[TinselKoala]

I suspect that any erroneous phase shift is caused by:

1) Induction loop between scope probe commons and the scope chassis, and/or
2) Parasitic loading of the voltage measurement.

I would try the following:

Make up a 501:1 50 Ohm resistor divider ( 25K Ohm / 50 Ohm ), with the common side of the 50 Ohm reistor to the current probe common and use that with straight coax to the scope set for 50 Ohms at the input.  The 25K load will be very light on the 75V pp signal, and you will still have 150mV pp to work with.

I think I'm following you here but it would be nice, just to be sure, if you could sketch up a little diagram of the divider and the connections to DUT and scope.

The Tek 2213a's inputs are hardwired to 1 megohm impedance, and so are the HP180a's and the RM503's. I think I can set the Link DSO's input impedance to 50 ohms, let me fire it up and check.

ETA: Nope, the Link inputs are also hardwired at 1 Meg, 5 pf.

[MarkE]

I think I'm following you here but it would be nice, just to be sure, if you could sketch up a little diagram of the divider and the connections to DUT and scope.

The Tek 2213a's inputs are hardwired to 1 megohm impedance, and so are the HP180a's and the RM503's. I think I can set the Link DSO's input impedance to 50 ohms, let me fire it up and check.

ETA: Nope, the Link inputs are also hardwired at 1 Meg, 5 pf.

At 50 Ohms, the divider impedance should be low enough that it is essentially unaffected at 300kHz by a normal 10X probe.  The 10 Meg resistor is optional.  Without it, the attenuation is 501V/V instead of 500V/V.

[TinselKoala]

OK.... thanks for making the diagram... I'll see what I can do.

Meanwhile the motor load in the "receptor" or TKTransverter does cause some phase shift from the non-loaded state. Here are two shots comparing "receptor with motor load" and "no receptor at all". The external loopstick lighting the three LEDs is operating for both shots though.

This is the best x-y display that the old Link DSO will give me. I can't make quantitative measurements of phase angle from that Lissajous figure, except to say that the "no load" one is almost perfectly circular indicating 90 degrees phase angle. But I may be able to do it with cursors and math on the time-domain screen. Let me fiddle around a bit.