[sdiy] zener limiting of filter self-oscillation

David G. Dixon dixon at interchange.ubc.ca
Sun Jun 13 03:20:50 CEST 2010


Here's some questions for the design gurus, or anyone else who cares to
answer:

I'm just about to build the prototype of a 4-pole cascaded-VCA filter
designed to run on a +/-15V power supply.  Nominally, I would feed +/-5V
signals to it.  At self-oscillation, with no zener diodes to limit the
amplitude, the recycled stage 4 output would clip (relatively softly) at
roughly +/-13.5V, giving +/-10V sine waves out with relatively low THD.
However, I don't really want them to be twice as large as my normal signal
levels.  Hence, I will use a couple of zener diodes.

In simulation, I have found that 6.2V zeners give sine waves which are close
to the desired +/-5V output level.  However, these tend to cause fairly hard
clipping, which gives somewhat distorted sine waves, particularly from the
first stage.  Also, with these zeners, the initial ringing peaks on the
output of saw and square waveforms filtered at high Q are also being
clipped, which is bad.

I'm thinking that 9.1V zeners are a decent compromise, as ringing waveforms
don't seem to be clipped, while the sine waves are about 7V, which I can
live with.  What do you think?

Also, Jung's IC Opamp Cookbook (1st ed., 1974, Fig. 5-20, pg. 203) shows how
to use two PNP transistors (2N3906) with their bases and collectors
connected together instead of zeners, suggesting that these will clip at
about +/-7V.  However, simulation in Multisim shows no effect whatsoever.
So, has anyone tried this, and does it actually work?

Finally, that same reference (on the previous page) shows how a 5k trimmer
connected as a divider between the opamp output and ground with the zeners
connected to the tap allows the clipping voltage to be increased from the
nominal zener voltage to the rail (or the opamp clipping point).  This does
work in simulation.  I'm thinking of using 6.2V zeners and bringing this
control out to a panel pot, to allow the user to adjust the sine magnitudes
at will.  What do you think about that?  Has anyone here done this before?




More information about the Synth-diy mailing list