[sdiy] Linear/Log FM (was "Living VCOs" PCB in 2009 ?)

[Mattias Rickardsson]

2008/12/10 Ian Fritz <ijfritz at comcast.net>:
> At 05:50 AM 12/10/2008, Yves Usson wrote:
>>
>> OK on second thought I think I see what the problem is, the linear
>> modulation
>> is a linear frequency modulation but it's extent is depending on the
>> current
>> circulating in the servo circuit thus if one changes the CV at the V/ OCt
>> input
>> one changes the linear modulation amplitude and therefore a constant
>> offset applied to this input
>> will not be constant in frequency offset since it also depends on
>> what is on the V/OCT input.
>
> Yes, the control signals are multiplied. (Perhaps an easier way to think of
> it.)

If I remember correctly, an exponential converter could be seen as
doing something like the following mathematical job:

    y   =   A * e^(x + B)  +  C

Both A and C affect the result linerly, but in different ways.

* Modulate B, and you get exponential FM. (Normal V/oct inputs & pots)

* Modulate A, and you get linear FM scaled by the exponential factor
e^(x + B) ,  which makes the modulation index constant with respect to
exponential CV, i.e. with FM synthesis the waveform is intact when you
play different notes.
("Standard style" linear FM)

* Modulate C, and you get linear FM, unscaled! The frequency will
always be the same number of Hz away from the unmodulated frequency.
("JH *living* style" linear detune) ;-)

Having access to C on a detune knob seems very interesting indeed,
especially when using more than one VCO simultaneously. Two VCOs could
be tuned some Hz apart and the beating frequency would remain the same
regardless of pitch.
But how useful is it on a single VCO?


/mr