filtering it through his bank of capacitors, the AC signal really isn't working very well, and it's causing degradation in the electrolyte (at least that what Frank said). It seems that the best power for the equipment would be flat DC (see the top graph in the picture), but that would take a whole pile of batteries. AC power is much easier to produce and transmit, so despite Edison's efforts to promote DC, we are a nation, no a world, of Alternating Currents.
Yes, there are methods to convert AC to DC, but both half-wave (see middle graph in the picture) and full-wave (bottom graph) have bumps. In other words, they still aren't flat. Even adding a few filter caps to the output only gets close. It still has wigglies. The best bet would still be to run the equipment directly from a full DCsource such as a battery, but as that's not feasible, we're just going to keep on with what we have.
There was this rather useful circuit that Frank came across when he was trying to reverse engineer the equipment, but he's been unable to reproduce it (mostly due to the language barrier). He says that it must have been a Unimolecular Rectifier. Unfortunately, the technology available to folks these days isn't fully functional, so all we can do is work with what we have. (He has been pretty good at re-building stuff just by analysing the equipment, in spite of the objections, and this stumbling block really irks him.)
If anyone knows of a better DC source, please let us know, and I'll report on it next week.