the trick with the Janzen's (iirc) is that they used graphite for the coating. this works well for a long time, but when it doesn't, the cells cease to function.
The ones I had did NOT want to come apart cleanly.
When I could still hear above 20kHz, I found these panels annoying up there, but that may have been a function of the amplification, not the cells. I don't really see any reason that they could not be done with thinner membrane, but there may or may not be a real advantage. Perhaps the thing to do is to repair one, but
clamp it together for testing, and measure to see what the change is. Doesn't need to be pretty.
They have a rather noticeable limit for both SPL and HV bias. Above a certain bias voltage they start to "sizzle". Maybe less in a dry climate? Also they have a definite LF limit.
It's not obvious from the outside, but they are actually a series of edge-to-edge vertical cells, not little square cells all together. The HF dispersion leaves a bit to be desired.
They were my first ESLs!
Had a tweeter array whose mfr name I can not presently recall, I think 4 of them horizontally, looking a bit like the front of a small horn. Great fun back in the day!
_-_-
The Infinity Servo Statiks used in house made mid panels and half cells (at first they cut them in half) of the Janzens, later they used the blue RTR tweeters, which are the identical technology.