[sdiy] VCO Jitter, Slop ... (was: Re: Smith's Evolver Desktop's spikey VCO waveforms)

[Ian Fritz]

For another look at that old Moogy Boogy jitter/drift question, I have just 
done a test on one of my analog VCOs.  This is the Dial-a-Tempco VCO, which 
has trimmer adjustments to tune out both exponential and linear temperature 
drift.

Recently someone posted a question as to why analog VCOs drift so 
much.  Pressed on this, he said that he could hear an obvious short term 
drift of ~2 cents on a module he was using.  A 2-cent drift is about 0.1Hz 
on a 1kHz signal.  This much change can easily be resolved on the frequency 
meter of a good DVM.

So I measured the frequency output of the DaT VCO at 1kHz for two hours 
using a B&K 2890A multimeter.  The update rate on the frequency counter is 
1.2sec, so any variations on the time scale of 2sec to 2hr would be 
observable. First I watched the meter constantly for a few minutes, then I 
took readings every 10min.

The result?  The frequency counter *always* read *exactly* 1.0132kHz.  I 
never saw even a flicker in the last digit. Since the last digit represent 
0.2 cents, it is clear that there was nothing like a 2 cent variation (this 
would be a digit in the 4th place of the 5 place display).

So, just one result on one VCO.  Conclusions?  I'll let you draw your own.  :-)

Ian