Odyssey, dual VCF
As mentioned before, this seemed like a must-have mod. During the years Odyssey was manufactured, the filter module evolved from a 12dB/oct version to a 24dB/oct version and the design was changed from a bi-quad OTA version to a Moog-style transistor ladder version. Having both of these version incorporated in the same synth would give a larger “palette” to work with, especially if/when external signal/control inputs are added.
So yeah.. Nothing too overly difficult with this board (and once I had all the parts) besides the few spots yours truly forgot to solder and then wasted time on measuring and scoping the obvious.
First, thanks to the self-made board (no through-hole plating) solder pads connecting the upper/lower sides of the board need to be soldered individually for each side. So on my board, these solder “faults” resulted in the 24dB filter feedback chain (resonance) not working at all.. Audio signal was passing through the circuit nicely though! Second, I forgot to solder in a few wire jumpers since these weren’t marked out too clearly in the documentation. Having few empty holes on the board should’ve served as a clear sign of “hey don’t you think there should be something here?”.. Then again, it’s me were talking about ;)
I didn’t have some of the required resistors in stock, so I used two of lower value paired up in series. It’s just the same as using a single resistor and since I had some that fit the purpose I could avoid a run to the parts supply shop. The keyboard CV scaling calibration might’ve been easier with multi-turn trimmers, but eventually I got it somewhat right.
There was just enough room between the power supply board and the pedal jack for this module so I mounted it there using epoxied standoffs (no case drilling required, yey!).. With cabling and connecting the 40235 board to the rest of the synth, I (for some reason) managed to miss the note about filter types for quite some time (pg.2 on the service manual) ALTHOUGH I had looked at it beforehand and tried memorizing “DEFINITELY DO NOT FORGET ABOUT THIS MOD WHEN INSTALLING THE NEW BOARD” at that point. And because of this, I once again wasted and hour or so poking around the filter board with a scope and wondering why does the resonance slider work in reverse and ADSR envelope modulation doesn’t seem to be doing much ;).. Funny when I think of it now, but not that I was amused about this at that point.
So.. I swapped R1-R3 to 680k, rerouted the output of A1 through a 100k resistor to the junction of R10 and R11 and reversed the GND/+15V connections for the resonance slider. T4 on Board C (VCF Bal trimmer) wasn’t needed for the new module so I pulled that out from the board altogether.
And off it goes \o/ .. I gotta admit that the 24dB filter is a nice addition, as much expected. On my module, the 12dB filter resonance starts self-oscillation a bit too early (resonance slider about halfway up), so I think I’ll try adding a trimmer in place of the extra 15k resistor at some point and see if the range could be adjusted more precisely. Not that this bothers me much, but I’d rather have increased slider resolution instead of excess squealing.