3340 Sync - E&MM Spectrum Synth

[Martin Czech]

What type of JFET is T17 and T18 (discharge)?
What about leakage at low osc freq ?

> The soft-sync pin is not used in this schematic (Pitty)
> Nor is the hard sync pin #6. (Colin is this correct ? My crt has no good 
> resolution for this tif).
> This would be very interesting, because the tri wave would reset to some 
> small positive voltage in case of sync, but the charging state would not 
> necessarily change, rather stay where it is, thus the sawtooth and pulse 
> phase would NOT reset to a determined value in case of sync.
> This is not quite what a conventional sawtooth hard sync does.
> 
OK, I was partly wrong ,
the text explains, that in one mode the timing cap is pulled
below the lower threshold, so this will give always rising slopes.
Between triangle buffer out and timing cap pin there is considerable
dc offset I didn't knew about.
If you have hard sync, you can always get soft sync by allowing 
sync operation only in some specified window of the slave vco.
This kind of window comparator can always be added to a circuit,
even external.
So now we know how to achieve real hard sync with simple means.

>That's not what I thought - the pulse wave from the master oscillator is
>differentiated, but the pulse wave from the slave is dc-coupled. This
>means the slave will ignore the incoming sync pulses until its pulse
>ends, so the PW control on the slave allows sync to variable intervals
>above the master.

Hmmm, that's right. But I thought about the generall posibilities you have with
sync. These are:

1. Hard sync by pulling the timing cap to some level
   (which can be adjustable) if the slave is in some window
   (also adjustable).
   This includes soft sync (very low discharge, window narrow) 
   and conventional hard sync
   (very low discharge und wide window) and all between.
   By feeding the hard sync pin with proper pulses,
   the charging direction can be controlled, so that also syncing
   to any level will give controlled behaviour (rise/fall).
2. Reverse charging (triangle oscs only).
3. Gating or hold mode by constantly pulling the osc to some level,
   start/stop osc.

I think that this is all you can do with sync.
Doing all this makes sense, if you use the vco also in subaudio
ranges as vclfo.
Setting the vco to midlevel (0V with symetrical out of e.g. +-1V)
could make sense to avoid clicks with fast envelopes .

m.c.