If the correct method to compute wattage is in the instantaneous analysis of the shunt and the battery voltage then school classical has a big problem. I want to be very clear what I'm referring to here. I take the first one sample of the voltage across the shunt and then the first one sample of the voltage across the battery. Then I multiply those two values together and divide that product with the Ohm's value of the shunt. Then. I copy that equation to each of those 500 000 plus samples. Then I take the sum of those samples and divide it by that sample number. That's what I do, and that, I believe, is in line with what the math function does and what it shows in the math trace.
Then we have ALWAYS have a negative value - indicating that nothing is being discharged at the battery. Then too, the battery voltage first drops - dramatically - and then it consistently climbs back to a steady high voltage. So MileHigh. If this is the correct analysis then - again - school classical has a problem.
Regarding that 'short' positive spike at the drain. There is no question that this is not breaching the resistance at the MOSFET where, I presume, it would be blocked. It's the fact that it still generates those really high oscillations at all - after the discharge of this small spike - that is of interest. Surely? Because what is also self-evident is that the voltages at the battery and the voltages at the resistor - ramp up to a higher and higher value until the point that it 'levels out'. How does that energy influence the battery voltage? It clearly goes through the battery and through the load as it's also evident at the drain. And the voltage at the drain shows a waveform that is consistent with the battery voltage.
And again. The kicker. One can adjust the offset or the duty cycle - or both - and one can then get the clean 'on time' that you're all looking for. At which stage - depending on the level it's tuned to - one sees the voltage rise, correspondingly, across the shunt. I've shown this but will post this again - later today. Then here's what happens. The mean average and the cycle mean average MAY SOMETIMES default to a positive value. But the product shown by the math trace ALWAYS STAYS NEGATIVE. And this is born out in the close analysis of the instantaneous wattage that you all have determined is the CORRECT analysis. And it certainly does not result in any evident loss of charge to the battery supply source.
THEN. We have the negative oscillation persisting - during the off time. No matter what. There has been prior evidence of a negative triggering - evidenced by Aaron Murakami. If it is the result of stray capacitance then so what? I understand that stray capacitance is seen as a kind of residual charge. On my side, I see it as an induced voltage over circuit material. But stray or spurious oscillations are not expected to be that strong that they can be returned to both the supply source and then back to the load - repeatedly. And every return ADDS to the charge conservation from that supply - a little more with each osciallation.
That it has not been evident before is due to a variation of the circuit. What is enabled is that there is sufficient path made available to the circuit to ensure that the full benefit of the current induced by that negative spike is able to flow. I suspect that all prior circuit configurations blocked this courtesy some resistance in that Zener diode. Access the full range of it's value and it most certainly returns a net energy gain to the system.
This is what we intend showing. However. It is absolutely NOT the only way to 'skin this cat'. One can achieve precisely the same thing as has been shown on previous test replications. But the net return is then more modest. That negative spike invariably rang 'down' not up. COP >1 rather than COP infinity.
And for those who have read it - the explanation is only in line with known Inductive Laws. I keep saying this. What is evident is almost prosaic in it's essence. All that aether energy - and all it turns out to be is the full and proper use of the negative potentials in induced voltages. Which does not minimise this application. The implications are mind bending. It points to the possibility that there is far more potential locked in inductive/conductive material - than has, heretofore, been fully exploited. And that points to the 'thinking' that initiated this circuit design in the first place.