How did you do your analysis? Walk me through it.

I'm estimating about 0.266 W average input power from the 20V source, and about 0.231 W average power dissipated in the 50R load resistor.

TK,

OK, here is my analysis as you requested. There are some minor changes to my original post that is, C4 is now .0024ufd and the stabilization run time is 9ms instead of 4.5ms so there may be slight result differences from the original.

The first pix is the plot and math for the V(Ecap) portion of a cycle. I wish I had removed the extra traces for clarity but the results are the same. For those who may not know, the conventional current flow (positive to negative) in LtSpice is toward the dotted end of an inductor. So, we see 406.52mW (average power) entering node Etap.

The power across L4 is calculated by (V(etap)-V(ecap2))*I(L4) and results in -404.87mW. Since current flow in an inductor in LtSpice is measured toward the dot, the negative result means that the current flow is in the opposite direction or away from the dot. Therefore, there is actually 404.87mW leaving node Etap.

The power across L2 is calculated by (V(etap)-V(ecap))*I(L2) and results in -1.4297mW. This value does not satisfy the power balance at node Etap due to slight errors in the sim's math calcs IMO, but the amount is small enough to ignore for now.

The basic point from the first test is that we have ~405mW average across the primary from the dot to non-dot polarity.

The second pix is the same basic test done on the V(Ecap2) portion of the cycle. Without all the explanation, we can see that in general we have a near balance 398.95mW leaving node Etap. This ~400mw however in this cycle is across the primary from non-dot to dot polarity which is opposite the V(Ecap) measurement.

The results from above are what I used to conclude that the average reactive power in the primary of this device is near zero. The above measurements do not reflect the entire cycle however and that is done below.

The third pix is a full cycle plot stripped of everything except the power averages for DC input, reactive input, and resistive output and indicates an approximate -10mW net reactive input.

I welcome any and all corrections and/or comments.

pm