I find no mistake in what you wrote above. We think much the same way. I felt that capacitance had an important role and you are right, it was low, very low actually. I have a LC Meter and it registered about 2nF. It wasn't rational of me to think it would work at that low capacitance. Zinc, copper, it probably has no importance as long as the capacitance is high enough. Regarding Pavel Imris, he appears to have done similar research. I read the capacitive winding patent but unfortunately like most patents out there, the bigest problem is specifics. I think 99% of the patents I have seen in my entire life have no specifics! This is madness in my opinion because you cannot replicate something with acuracy if you do not have the exact configuration of the original. I know the patent holders want to present the general idea, but there should be a place in the patent, where a detailed scheme, with specifics of the device, must be present (a small scale device, it doesn't have to be big, just the point to be made, for everybody interested to replicate). Greed or paranoya of the patent holder also has a big place in this. All of these togheter, impede the majority of energy efficient seekers, like you and me and many others, to wonder years after years without finding much usefull information. It is like reading a book, for enjoyment and nothing more.
I understand Chriss remark about it being fools Gold, this device probably needs tuning for a specific load, but if you can use 1var to mantain 1W, I think that is still something.
All the best,
P.S.: I will try to scale up the capacitance.