[sdiy] AC coupling caps on MS20 clone

[David G. Dixon]

Concerning AC coupling caps on my Korgasmatron 2164 MS-20 VCF clone, I
realized that I could simply add a DC bias to the input signal on the
simulation to see how the hypothetical circuit would handle it.  Here's what
I found:

Adding a small DC bias to the input signal (e.g., 2V, or raising a 10Vp-p
square wave input to +7V/-3V) has no discernible effect on the output.

Adding a large DC bias (e.g., 8V, or raising that same square wave input to
+13V/+3V) causes the higher resonance peaks from the HP filter to clip
slightly on every other half cycle.  Neither results in a DC bias on the HP
output.

Hence, I see no pressing need to apply an AC coupling cap on the input.

Adding a DC bias to the first LP filter summer (i.e., the first VCA) causes
the LP output to retain that same DC bias (albeit inverted).  However, since
the HP output evidently will never have a DC bias, I see no need to put an
AC coupling cap between the two filters.

Ergo, I see no need for AC coupling caps in this filter circuit.

I suppose my final step will be to simulate the original (OTA-based) circuit
and see what effect the AC coupling caps have there (unless someone here has
already done so, and can save me the trouble...?)


> When are AC coupling caps necessary on filters?  Should they be on the
> input
> and the output?  Should they be on all outputs of multimode filters?
> Should
> they be switchable?  What's the best value and type?
> 
> I haven't used them on my lowpass filters, but I'm unsure how to handle
> this
> aspect of my 2164-based MS20 VCF clone design.  Any suggestions would be
> most appreciated!