At the time, I was working with the guy who came up with this toroidal power amplifier, and he wasn't able to get any OU results with it. It does not use the "lenz utilization" principle that Matt talks about, which is key.
He is no longer working on it.
That guy also worked with me on testing the FNT, and he was the one who found COP around 19 when run at resonance, although I think that number is considerably too high.
As to the Jensen UDT, if you go up a couple of letters, I showed two files for a ferrite/metglas version of the UDT I called the flux null transformer. You definitely could do something with ferrite E and I core, but as I recall, Jensen himself said it would work better at higher frequencies. That is because of the gap: magnetic energy stored in the gap can be recovered better at higher frequencies by running at resonance, as I did with the FNT.
If you're interested, let me know what you have in terms of cores, and maybe we can come up with something using what you have.