Long version
hey Kach,
I bought the cells from a friend who had replaced the cells in a quest because the connection on one of them went south. I do modify almost everything, but it is for the result, not because I like to tinker.
Checking out the ML bulletin board, I see that discarded cells are fairly common, and I may need to run separate voltages on each to get them to perform at the same efficiency, even if they are not damaged in some way as yet unknown. I have two Quad polarizing voltage units (and output xformers) so the voltages could be set individually.
A diy magazine, Amateur Audio, had articles by Roger Sanders years ago on building estat cells. I built some and they had a passive eq shelf to bring the treble up. All stats are big caps and their impedance goes low at high freqs, so they are hard to drive there. At that time I had a TVA 10 amp, with a pair of EL34's. Although a great design by a master (IMO) Tim de Paravincini, the 4000pf of each of the two cells was "tough" to drive.
What "tough" means is, as the charge on a capacitive load goes up from the electrons that the amp has already pushed in, less charge flows from a given output voltage of the amp. Tough to follow the input waveform.
A resistive load doesn't build up charge, so its "easy" to drive.
An inductive load, like a speaker voice coil, resists change, or flow of charge, so it takes more force to push charge through.
Re breakdown,at high altitudes: At 7800' there is only 3/4 of the air at sea level. I'm a retired high school physics teacher, and a standard demo we used to do is put an electric bell in a vacuum pump and evacuate it. The sound goes softer, but you also see the spark (where the contacts are making and breaking) get larger.
The Sander's estats were big flat panels of .062 aluminum, and you would real quickly see where the stators were the closest by the sparks. You could take a little allen wrench and pull the stator out there, but it provided a limit as to how loud they could get.
(High altitude also messes up bass alignments because the air mass in the port is less and so resonates at a higher frequency, while the speaker mass stays the same. The air damping the suspension resonance of the driver is also less.)
The 2kv2 power supply is freaking dangerous, and I decided I wasn't fastidious enough to be safe (avoid accidental contact) over the long, long course of developing a new amp.
Re active load, a tube works by allowing electrons to flow from the cathode, through the grid (which is tracking the music signal) and on to the anode/plate, drawn by the positive voltage of the plate.
In a transformer load, the high voltage on one end of the winding (or the center tap on a push/pull transformer) attracts the electrons flowing through its winds from the tube. The flow induces a current on the secondary, which drives the speaker.
On a resistor or inductor load, the resistor or inductor only pass so many electrons per second to the positive supply, so the amount of charge accumulated at the anode varies as the audio signal dumps more or less in and only a fixed amount leave. It is this force of these charges which drives the next device.
However, both resistors and chokes vary in the amount of electrons they accept/pass depending of the voltage difference at their terminals, so they are not perfect sinks/sources, (depending on whether you think in terms of classic positive current flow or electrons, I'm talking electron flow here for clarity).
Even more linear behavior can be had by using a tube as a load, hooked up to actively adjust to the conditions. The stacked arrangement also allows tubes with a given high voltage limit to operate in higher voltage supplies. (The heater, if it is separate from the cathode, also needs needs accommodations.)
Re direct drive: A tube amp has a step down transformer in its output, and an estat has a step up xfmr, so a direct drive at high voltage swings can obviate two transformers and their limitations of linearity and bandwidth, if the design gets it right.
Any less opaque?
Sorry you asked?
Dan