D-Lev PWBs Are A Go
The PWBs arrived on Friday, 3 days early - JLCPCB makes an excellent product quite quickly and relatively painlessly, and DHL really shags ass with international shipping (though one pays semi-dearly for it!):
Boards (clockwise from lower left): main, large tuner, tiny tuner, 4x encoder, tuner driver, and (center) AFE.
I spent Friday evening examining the physical stuff like overall dimensions, hole locations, and parts footprints, and did some electrical buzzing out comparisons with Roger's boards. Saturday I built up a main board bit by careful bit, checking functions as I went, until it was complete and fully functional (unsoldering is difficult, and you don't want to waste parts on a dud). Then I built up both large and tiny tuners, a couple of tuner driver boards, and a two AFE boards, also checking things carefully in steps.
Mounted them in printed white PETG panels:
The controls are hooked up to a tiny tuner and I haven't seen anything show-stopping so far; I hooked one AFE to a random coil and rod and it's behaving like it should. TOSLINK out works, MIDI out works, the footswitch inputs work. The only thing I haven't tested yet are the SPDIF outputs, but they're pretty trivial.
I've been holding off on the coil winding because it felt premature. Indeed, yesterday I looked into 1-1/2" polypropylene sink drain pipe instead of 1" schedule 40 PVC (slightly larger OD) and it looks like it may be a better fit. The relative permittivity of 1.5 is about as low as it gets (minimum self capacitance) and the weight is less as it's a thin wall pipe (any significant mass experiencing sudden deceleration can cause trouble, so the less the better) though the coefficient of thermal expansion is likely a bit higher (which is probably one of the most important factors for long-term stability / drift). I wound a pitch side 1mH coil this morning and am currently testing it, so far so good:
So things are definitely moving in a directionally correct direction!