I thought the flyback might be bad at first too, but I think it came down to the grid of the output tube not being tied to the correct negative voltage source. Once the tube was biased correctly, HV came up to 20kv and stayed there. Which brings up an interesting point: This chassis gives us the opportunity to document the operation of the pulse regulator circuit, I would very much like to convert another set to operate this way since it takes a little stress off the flyback.
We already discussed that the cabinet will be the TC-168 with doors, but I'm open to making a clear display case for it. Such a case would make the guts easier to see, though if you ask me it's pretty compared to production units. It definitely has a face only an engineer could love, if you know what I mean. In that respect, it is indeed a thing of beauty.