I had not spotted the huge electrolytic on the output of rail splitter OPA. This is indeed commonly advised against.
The R/C in front plus the OPA's driving stage take care of the purity of this rail.
My unit exhibited quite a bit of switching noise, which may partially due to the cheap footswitch they delivered with the kit. If I were to build a new one, I would order a Gorva - much higher quality.
I remembered Mr Black's tips against switch pops (to be found on the internet) and added a 4.7k in series with the input, 2.2M against ground, 1k in series with the output and 100k to ground right at the board. With those resistance dimensions the order is of little importance.
Mr. Black thinks the input grounding resistance can be omitted by the grounding action of the true bypass switch, but since he is nit-picking about DC voltages building up we might as well consider the switch when it is in transition during the change of state. (I noticed on some recently available kits some of those resistors are already included).
This took care of the switching pops.
The Tri-Vibe works very well after dirt, unlike the uni-vibe. I wish the whirl position (the one opposite of the phaser sound-alike) were a bit more pronounced. It reminds me of a rotating speaker.