[sdiy] Expo converters, lin log etc.

[Dave Manley]

Andre Majorel wrote:
> On 2007-01-31 21:05 +0100, Karl Ekdahl wrote:
> 
>> Basically, the response of a "normal" vco (for
>> example) is logarithmic, therefore an exponential
>> converter is needed to linearize the final response -
>> right?
> 
> No. As far as I know, VCOs are typically linear

This must be in the archives if not here on then on AH, but here are the 
fundamentals as I know them:

At the heart or core (as in 'ramp core' or 'triangle core') of a typical 
VCO is an integrator.

The control voltage is converted into a current, and this current is 
accumulated (integrated) on a capacitor.  Integrating a current on a 
capacitor gives a voltage that increases linearly in time (ie a ramp 
waveform).  When the voltage reaches a set value, the integrator is 
reset and the capacitor is emptied of charge.  Then the next cycle 
starts over.

The rate at which the capacitor charges is linearly related to the 
current, which in turn is directly proportional to the control voltage.
The charge rate determines the ramp slope, which determines the frequency.

Therefore a typical VCO 'core' is linear.

There are a number of system level issues with linear VCOs, so commonly 
an exponential converter is added in front of the linear VCO core to 
make an 'exponential VCO'.  The expo converter changes a linear voltage 
to an exponential current.

So the answer is VCOs at the system level can have either linear or 
exponential response.  But at the core, they are linear.

The reasons why to go linear or exponential are well documented in the 
archives, so I won't repeat them.

-Dave