Ensemble frequency response
2008-01-31 by dark_november2000
As much as I like the sound of the CS-80 Chorus, even the "darkness" of it, I was wondering why is sounds *so* dark (low pass filtered), compared to your usual BBD-based chorus devices. All right, you need a anti-aliasing filter and a restauration filter for any good BBD-based effect, but usually you only need dramatic low pass filtering of *long* delays (echo devices), where you find corner frequencies of 5kHz, or even 3kHz, sometimes. But on a chorus, you normally get away with "almost" full audio bandwidth; the sound may get *slighty* dark, but not much. I've been curious about what dramatic filtering would go on in the CS80 chorus, and what I've seen so far was a big surprise: * Anti-Aliasing Filter: 16kHz, 2nd order * Restauration filter: 16kHz, 2nd order Hey - that's pretty much "full audio bandwidth" ! Even the two combined may result in something like a 12kHz corner frequency, and 24dB/Oct slope - not explaining the dark sound at all. Curiously looking further, I found a combination of 27k shunt resistor and 15nF shunt capacitor at the output of each VCA that passes the signal from the BBD for panning effects. Together with a 100k resistor that feeds an opamp summing node from the VCAs, we can calculate the corner frequency of the resulting low pass filter function: R = 27k // 100k = 21k (The VCA has a current output which doesn't contribute an R) C = 15n f = 1 / (2 * Pi * R * C) = 500 Hz That means, the signal is conditioned for the BBDs with two 16kHz filters, only to be filtered with 500 Hz (!) in the VCAs ?! Very strange. Ok, it's just a one-pole filter (6dB / octave), but 500Hz is damn low. Maybe there's a typo in the schemos, and it's 1.5nF capacitors, not 15nF. Then we'd have a 5kHz LPF instead. But the same "0.015" is printed at each of the four capacitors on the four VCAs. Or is this a deemphase circuit? But I can't spot any preemphase part anywhere ... JH.