Hi Steve, hey thanks!
microstrip brought up a couple good cases to look at.
The electrostatic speaker as a transducer is capacitive for the reasons I described and is driven by a direct drive amplifier or step up transformer to get the required signal voltage swing.
Now, the question of where the sound goes comes into play.
For a fixed size, a flat panel ess will have a radiation pattern that narrows as the frequency climbs. If it is producing a fixed power, then this would cause the SPL to rise as a fixed power is focused into a smaller and smaller area.
A first order slope can be applied to correct this by allowing the Rdc of the transformer primary and secondary to form an R/C low pass filter to counter directivity .
Anyway, the fact that the radiator is a “variable acoustic size ”effects the directivity, some speakers like maggies have a tweeter section of a more compatible acoustic size.
The ESL-63 was like no other speaker I had ever seen. My boss at Intersonics in the 90’s had a pair and kept popping the internal spark gaps. He knew I made ESS speakers and so he asked me to “fix” them so they wouldn’t spark, and to be clear in a lovely old English accent he said “tear the bloody things out’.
These were far more expensive than anything I owned and they were my boss’s to boot so I was really careful taking them apart. I got the grill off blinked twice and though WTF is this????
I had built bunch of simple flat and curve electrostatic speakers, this was NOT that at all.
These fought off the narrowing by radiated a portion of a sphere, constructed by sequentially driving a flat diaphragm in rings, rings of electrostatic speaker. See the picture below, see the rings, the signal is presented at the center first, then progressively outward as if an expanding spherical segment of sound passed through it.
http://www.hifiengine.com/manuals/quad/esl-63.shtml
In order to produce this effect each ring is driven separately, each signal is progressively delayed with a “delay line” which is made of capacitors and inductors. In the schematic one can see the delay line is the series of coils and capacitors with the “VA” through “VG” delayed taps. Each pair of L’s and C’s in the line may also be recognized as a 2nd order “all pass” filter, a filter which has flat frequency response but exhibits phase shift.
http://mark.rehorst.com/Quad_ESL-63/Quad ESL-63 Simulation Schematic.png
Anyway after removing the offending “protection” I listened to them but not loudly, for a weekend.
They had a hypnotic effect on me, when someone spoke, it sounded like the speaker was an opening into another space like window or something, not the source of the sound like every direct radiator I ever heard. .
They had no apparent flavor in fact underwhelming at first until I heard subtleties I had not heard before.
They were just a little bass shy for my taste but they disappeared like no speaker I had ever heard at the time or would ever hear for a long time to come.
Anyway, my point before daydreaming there was that unlike the other ESS, those distributed the C’s out with a transmission line, they are not like most.
Fwiw, I don’t know the model but the large Soundlabs I heard were luscious sounding and driven by some large futterman OTL’s, I didn’t want to leave.
Best,
Tom