100VoltTube

Interesting amp. The amp section is pretty straightforward and by-the-book. I am curious, though, why did you choose the tubes you did? Was it purely for rolling purposes? I would like to point out that the output stage can't swing nearly as much voltage as most electrostatic headphone amps, but if it's enough for you, then there's no problem. Also, I notice the NFB loop is connected to only one side of the output stage, and it presents a low enough impedance to change the performance of that stage significantly, which would lead to an imbalance between the top and bottom portions of the amp. Have you considered loading the top portion similarly to restore the balance? The case and internal build quality is amazing.

I was under the impression that it was a problem, or possible problem with electrostatic headphones. Sony took pains to avoid symmetrical reflections (same reflections from both sides) in their electrostatic headphones. That may have just been a marketing move, though. Aranaud expressed that he sees reflections as a possible phenomenon in the head-fi shangra-la thread. That entire conversation between him, JaZZ and astrostar59 is interesting, and somewhat relevant to this headphone. If he has run simulations since then, I would be very interested in seeing the results. I do believe, though that this one small plus (if it is an audible phenomenon) is not worth all of the problems created by going single-ended, unless the problems are eliminated properly, which they may not have been in this implementation.

I haven't heard these headphones, so I can't comment on their sound. The driver tech looks promising to me. No front stator means reduced reflections in the driver capsule, which is a good thing, but comes at the price of the linearity of a push-pull driver, as Gilmore stated earlier. Electrostatic force over distance follows an inverse square law; the force between two objects of steady charge decreases over the square of the distance. DSP is, by far. the easiest way to deal with the nonlinearities. Doing it in the analog domain with the same precision as digital would be next to impossible. An R2R DAC with custom-chosen, nonlinear resistor values would also work, and would be my method of choice if I were making a system like this one. I doubt the amp section is up to par with any Gilmore designs. What is in there is anyone's guess.