John.k1. Here's a thought. What if the magic frequency for utilizing a direct earth ground connection is the
same or similar magic frequency that gives over unity on devices that don't have an earth ground connection? 
In previous testing I did with earth ground connections in the last couple of years, I noticed that there
seemed to sometimes be something different occurring around 1.2MHz to 1.3 MHz, and also around 1.6MHz to 1.7 MHz.
I was testing with two different earth ground connection points, but those same frequencies seemed to hold regardless
of the earth ground connection point or earth ground wire length. The frequency in the 1.2MHz to 1.3 MHz range seemed
to give the most notable results when using an earth ground connection, but this same frequency range also seemed to
sometimes have an effect even when I didn't have earth ground connected into whatever circuit setup I was testing with.
I notice that in the Lithuania experiment, they initially found 1.2 MHz to be a magic frequency, but this frequency apparently began to drift upwards.
Things that make you go hmmm...

Update. 4days ago I was about to record the vid to show you guyz the effects and loading with some readings.
After the accident which happend right after unexpectedly . My TL494/ Ir2110 Pushpull got smoked.

That's a very easy yoke circuit to replicate. but, to get it to produce the same results, is something else.

I have asked similar questions a long time ago and based on answers to them, I drew the schematic diagram of the Lithuanian Experiment #1 (sparkless, low-power version) according to all EE conventions including the dot convention and winding numbering, below.  I skipped turn counts.
P.S.
IMO the answer to Q1 is "Y", to Q2 is "underneath W2 & W3, while W4 remains on the other core half", to Q3 is "Y", to Q4 is as depicted on the schematic below ("non-cancelling mode").

Hi Void.   Yes, the frequency Romanov showed in his movie (1.36MHz) and I got the same. No mater what coil you use. I was speaking about it here long time ago and nobody paid attention to it. https://www.youtube.com/watch?v=PI1dSZ_dxRM

That may be a test we each need to run in our respective locations since we don't know for sure what is responsible for setting this particular frequency.  It also means if we get different values, our devices must be built and tuned to our specific location.

Yoke core - copper foil applied around one half of smaller opening end.
It is easier to wrap the copper foil around that part because it is more vertical,

   
And each new version is much more advanced and complicated, with higher output, longer 40 meter long earth ground lines, and needing special components. Like HV, HF diodes the will rectify the output from the 168t coil, to DC, and not go up in smoke, like half my Mazilli crt did, last week. 
 

AFAIR the original also had the copper foil on the "flat part" but that part was wider on the original yoke core.
However, I am not certain whether the copper foil was allowed to be in the gap between the core halves - instead there might have been a wire bridge soldered to two separate pieces of the foil before the gap.  I remember a diagram of it on this forum (see here) and Wesley talking, that there was only paper in the gap.  The thickness of this gap is evidently very critical, so the thickness of any materials within the gap must be critical, too.
I don't think the core needs to be wrapped in an insulating plastic tape, under the copper foil, because the original black core had a very high resistivity, as far as I remember.
Actually, the unimpeded contact of the copper tape with the naked ferrite might be responsible for some effect.
These points should be consulted with the people who worked on this device ...and settled once and for all.