Hey nut the significant digits are those of LTspice in its display and so, what the hell do you think you are doing with all of this?

A Square Wave in electronics does not have to be perfect or don't you know that?  So long as the 95% of it is within the square wave form its square enough or don't you know that?

As for a little ripple across the top.  Thats common is switching power supplies and means little in the end.

You think that everything in electronics has to be precise.  No circuits ever has to be as precise as the model  and never is,  just close enough.  Builders seldom are as precise in the real model as the design of the circuit diagram.

And so, why the thing about all the significant figures?  In electronics we usually round the units off to three figures as I do in my post.

Who cares if the software computes that the wattage has 7 of more significant figures?  We don't add them all up, we can if we want.

And most anyone who uses LTspice can and most likely will know how to measure ever crest and trough of the wave form ripple and then average up the power that way.  And its simple to do, and the way the software is made to analyze the wave forms of the power output since after all it is a power supply design software.

I suppose I might ask you what you been smoking but you probably wouldn't admit it.

If done correctly, power measurements are valid for determining COP.

In the case presented here where the claimed COP is 8.33W/5.29W = 1.57:1, an error has been made.

With pulsed DC power, the RMS power is the average power over one cycle (assuming the wave form is periodic).

Input power (RMS) = 5.29W RMS (no dispute here, will assume correct)

Output power (RMS) = 8.23Wp x .64 (duty cycle) = 5.27W RMS

COP = Po/Pi = 5.27/5.29 = 0.996:1

Conclusion: The circuit is clearly underunity, and LTSpice is running correctly. Interpretation of the output data is in error.

Two oversights:

1) Duty Cycle must be taken into account for any measurement, and for pulsed DC the RMS power is the average power. If the wave form is periodic (i.e. repeating consistently) one need only compute one cycle. The RMS power is: Po(peak) [Ton/period].

2) From the above it was noted that the COP was nearly 1:1. In theory it should have been exactly 1:1, but due to rounding errors etc, it was 0.996:1. This is not a big concern, but the point that should be noted, is in the real world this figure would have been significantly less, perhaps 0.8:1 or worse. The reason the circuit presented here exhibits a COP of 1:1 is because no real world finite resistances are present in the circuit. Every inductor or transformer has a real world DC resistance, and if introduced in the circuit, the simulation would clearly show a reduced COP, well below 1:1.

Regards,
.99