[sdiy] Configurable Phase Shift Project

[scottnoanh at peoplepc.com]

The past couple of weeks I've been intermittently trying different cap
values for the phase shifter.  I changed the value from 560 pF to 1N, which
improved the bottom end quite a bit.  Then I put the original value I'd
started with (2N7) in, and that certainly did prove to be too high of a
value - the high end dropped considerably.  After that, I scaled back to 2N,
which improved the high end response and preserved a good chunk of the low
end response.

The higher value of capacitance always brought a certain amount of pitch
bend on the lower end of the sweep.  I started wondering if that was
because, as the VTL5C3/2 goes higher in resistance, the tolerance of
resistance begins to vary more greatly from device to device.  Taking that
into account, along with a suggestion from a list member, I lowered the LDR
parallel resistance from 1.5M to 1M, and that seems to have improved things
on the lower end quite a bit.

I doubt that this phase shifter could ever fully emulate the effect of the
Compact A - I think I'd specifically have to build one to get the full
impact of that wonderful phase shifting on 'Phaedra'.  I can't seem to get
the sharpness of the resonance without getting a fairly large 'boom' in the
bass.  Still, the configuration comes somewhat close to it, and creates a
timbre certainly worth hanging onto.  As an aside, it's interesting to
increase and decrease the number of shift stages and listen to the effect
while preserving the feedback filtering and routing of the Compact A.

Oh, yes, I've also created an 'even stage/odd stage' regen control by
inserting the feedback either to the first stage or the second stage.  That
certainly has added a lot of variation.  I've also increased the design goal
of two 12 stage sections to two 16 stage sections.  The control circuit can
handle up to 8 VTL5C3/2's without flattening out, so I figure chaining 32
stages might might a pretty tubular (literally) sounding phase shifter while
in that mode.

I've got some of the THAT RMS detectors and THAT 2180 VCA's from Small
Bear - I'm going to experiment with those just to see how a soft kneed
compressor/limiter will effect the balance of resonance.  All of that will
have to wait for a couple of weeks while I'm travelling for business.

I've found that sending signals from two different taps in stereo through
the Dim C creates a wonderful ambience.  I've been circling the Dimension D
thinking that I'd certainly like the increased headroom that it provides.
I've noticed how Mode 4 between the two devices seems to be derived in
different ways.  In the Dim C, the LFO frequency and depth of modulation are
altered to attain Mode 4, where it looks (at least from what I can tell) the
mix of direct delay to each channel in the Dimension D is what is actually
altered, using the same LFO frequency and depth of modulation that Mode 3
uses.  I guess I'll know more once it's breadboarded.

Samples of stereo tapped phase are at the bottom of the page here:

http://mypeoplepc.com/members/scottnoanh/birthofasynth/id18.html

Cheerio,
Scott