Multiplexed Sample and Hold Resonator? (was New Filter Designs)

[Grant Richter]

Having had a chance to experiment with the multiplexing capacitor
state variable, I sort of figured a few things out.

The design is not the same as a switched capacitor filter.
In that design the "charge pumping" of the switched capacitor
is into an integrator is used to simulate a variable resistor.

Using a 1000 pf capacitor value (which would be normal for a SV)
the effect of multplexing the capacitors disappears.
The effect only works with the control current at maximum
and 0.1 uf (100,000 pf) capacitors.

This makes me think that it is actually functioning as a series
of sample and holds. Apparently, if the input frequency is
equal to or an exact multiple of the multiplexing frequency,
the circuit will pass the signal through. Otherwise it is blocked.
If the capacitors were buffered and taken to a weighted
summing network it would seem to form a crude tapped delay
line filter. The main interest is the very low cost of 4051 multiplexers
and it is an easy build.

I am hoping to do some simulations under Electronic Workbench
because I do not believe the state variable form is needed at all.
If it is functioning as multiplexed sample and holds, then just a
resistor, multiplexer and eight caps should work the same way.
Something is nagging at the back of my mind that I saw this
design used for DTMF decoding at some point. Does this multiplexed
sample and hold topology ring a bell with anyone else?

Best,

Grant