Oscillator idea come's unstuck

[jhaible]

> Just before christmas a few of us discussed an idea of mine to use
analogue
> switches to change the polarity of a cap in an oscillator circuit instead
of
> discharging the cap - to allow a glitch free triangle wave, and to remove
> the problems of using a reset circuit.
> 
> I built a simple proto-type before christmas, which I have now added an
expo
> convertor to, to try out with a keyboard cv.
> Unfortunately, I can't get the oscillator to scale over more than an
octave.
> As far as I can see there is nothing wrong with the expo current source,
so
> I must assume that the analogue switches (4066) in the discharge path of
the
> capacitor are affecting the current in some way...
> I don't know exactly what's going inside the 4066 so I can't really tell
if
> the problem is solvable, but it looks like the idea is no use for a
musical
> vco.
> 
> Oh well, it makes a good LFO at least...
> 
> Colin f

Hi Colin, and list, 

I was one of those who were excited about your idea, and I wasn't idle,
too.
I have tried some simulations with the scheme of the 4046 that Don has
pointed us to. 
All was fine as
long as I built the switches, current source etc. around the cap, and
forced
the switches on and off with an external clock. (Just to see the
waveforms).
But when I tried a real VCO configuaration with FETs as switches, the
waveforms
weren't smooth anymore. I have stopped at that point, but I think it has
something
to do with the way the switches are implemented, and the control signals of
the 
switches. What you want to avoid is two switches conducting at the same
time,
so maybe you need a short gap between the complementary control signals.
As I said, I stopped at this point but just because of lack of time.
(Finished
the HiFli, meanwhile ...)

Has anybody checked the signals on a 4046? Is the difference voltage
between
the two capacitor pins a true triangle or not ?

JH.