SV: [sdiy] analogue physical modelling organ dream

[karl dalen]

I just did some analog PYMO this evening, i used two VCOs
a mixer two 12db VCF one as HP on as LP, a home made voltage
controlled digital delay two ADSR,Tvo vcas, a phase inverter
a multiplyer, a constant generator (a buffered pot really).
 
I was supriced how well it created, guitar and bass guitar sounds,
using noise as exitation for the resonator in both cases,
but by setting a fairly slow attac and decay times it 
created incredible pan flutes! Most impressive!

Using a square wave as exitation and varying the 
frequency gave the emideate effect of playing 
with a plectrum at different velocy's!

But in my experiments i came to the conclushion of the
absolute usage of a "fast non clipping limiter" situated 
either after the resonator (delay) in the feedback loop,
or post resonator after mixer stage (also in the return 
of the fedback loop).

My setup "squeaks" and then saturate 
into a harsh square wave quite easily! Ugh!

I wonder if a BBD would saturate more gently? 

I also have quite difficult to get the resonator to track 
KBD CV, im thinking of using a small micro to handle this.

In my experimenting i found a easy way to include harmonics
(sort of) run a HP filter in paralell with the LP filter
(used in the feedback loop) track HP fc with KBD CV, 
resonance adds a lot to the harmonics, to much, squeaks 
and saturate the whole. Place a note tracking VCA after
the HP to damp out the resonances, if held static it
will squeak when exitator pulse hits filters self 
resonace point!

I had no idea untill now that PYMO needed so delicate and
acurate controll of parameters to be able to play well! :-)

KD

--- "JH." <jhaible at debitel.net> skrev:
> I dream of building another fully polyphonic instrument right now.
> It's a dream, mind you, and very unlikely to be realised. But I said that
> about other dreams in the past as well.
> 
> I think of a small analogue physical modelling organ. It's easy to make
> quite convincing
> flute sounds from a delay line, a 1-pole filter, and some nonlinear
> circuitry. The big
> problem is making this track over some octaves without going out of tune.
> 
> So if I'd use the same approach as in the PE-1000 - using a lot of voices
> instead
> of ane tracking voice - the flute might become an array of flutes - a single
> register
> of a pipe organ - and all the tracking problems, plus the concept of a VCO
> itself - would be nonexistent.
> 
> I phantasize about using a single PUT for the oscillators, half of a CD4013
> to
> get complementary clock pulses, a BBD without any optimisation (no bias
> trim, no balance trim, to keep things simple), a simple LPF built from a R
> and
> a C (no more VCF needed ...), plus a handfull of transistors for the
> nonlinear
> part.
> 
> At the moment, I'm making experiments with a PM flute (i.e. tracking), to
> find out what results I get with a very simple nonlinear circuit.
> 
> Well, I'm aware that with something like 49 BBDs I would only have *one*
> register of an organ. The final goal would hardly be a huge church organ -
> more likely something like a 2 or 3 register portativ.

> 
> JH.
> 
>