I have a system right here, a real one, not a simulation, that runs on 120 volts AC, draws about 10 amps, has a 20 amp fastblow fuse on the line input. It produces an output that is adjustable, between 15 and 25 KILOVOLTS at currents of as much as 1.5 KILOAMPS. This power is sufficient to shatter almost any container, to literally vaporize wiring, to puncture holes in 1/4 inch thick sheet metal. The current is so strong that it will turn a 19" equipment rack into a giant one-turn transformer and literally weld itself together at the seams. By the logic expressed in this thread, this device is WAY overunity. Waaaaaayyyyyy OU.




Thank goodness its duty cycle is short. It can only produce about 10 pulses like that per second, and each pulse is only microseconds long, representing the ringdown of a 3 uF 30 kV capacitor bank. But if I approximated that ringdown waveform with a square wave, according to my own private model...and then computed the RMS power, whatever that is, and equated that with energy, then I too could claim waaaaayyy overunity, and with a real device, too, not just a simulation.


(This is called a reductio ad absurdum.)

I actually agree with you.

 

The output of the simulation was made up of pulses.
When pulses are involved then you are trading time for power.
Charge a cap over time and then release that power as a pulse.
This tends to mess up measuring devices too.

In all fairness, I think he's attempted to account for duty cycle--at least his concept of it. In fact his claim of OU depends rather sensitively on "duty cycle", and in my mind a big problem, as I have tried to show, is that his output isn't square enough to justify his assumptions. Particularly the duty cycle calculation. Hence, even on the face of it, using his own data, the claim of OU fails.

All this, of course, ignores the issue that he can simply have LTSpice integrate the instantaneous power waveforms over a given amount of time to give the actual input and output ENERGY--as was done in the original posting of the info in this thread--by, was it .99 ? I forget, and since the original thread has vanished down the memory hole, I can't check and see.

Dear Friend. Here is another LTSpice circuit by some friend I have forgotten his name. I implemented it in  LTspice, optimized it by slightly modifying and changing values of the components. I am attaching it here for study and analysis. I will appreciate comments by all on it. Thanks in advance. I am also attaching the actual cct by that unknown friend here, which I optimized to get more power output. You can drive the same with a 5V DC also. It was basically enlisted in Joule Thief category.