[sdiy] Mind blown -- TL074 saturated outputs different

David G Dixon dixon at mail.ubc.ca
Wed Nov 28 21:01:24 CET 2018

> A warning based on previous experience: Zeners do not make 
> very good small-signal clamps. Because they do not have a 
> "hard knee" to their characteristic curve, when you run them 
> at low currents the voltage at which they start conducting 
> can be very different to that marked on the packet. To get 
> the advertised voltage, you need to drive them with a fair 
> bit of current.

Well, I'm just limiting a comparator's voltage, and I don't expect this
device to be operated under conditions where the comparators will be driven
at audio rates.  Mostly it will be manually controlled, or with a slow CV
signal.  Also, I can drive the zeners as hard as I want by selecting the
comparator output resistor.

> To get real precision and repeatability, I'd be looking at an active
> (opamp-based) clamp circuit.

I don't want to make anything that complicated.  I need three of these
comparators, and their sole function is to act as a 3-bit AD to control some
digital switches.

> If, however, it is only bipolar symmetry that you are after 
> rather than an exact level, consider putting a single Zener 
> inside a diode bridge. 
> That makes the positive and negative clip levels identical.

That would increase the number of diodes on the board in this area from 6 to
15.  I don't think it's worth it for what I need.  There is a bit of slop in
the voltages I can tolerate and the +/-5% tolerances of zeners should be
well within my tolerable range, and in any case, I can even measure the
output and adjust my voltage dividers accordingly, or hand-select the zeners
(I'm only building one of these things).

Another thing I like about the output resistor plus zeners solution is that
it will act as a bit of a lowpass filter, and soften the comparators rise
and fall slightly, which is also somewhat desirable for me.

In any case, I'm going to test all these concepts in the next two hours, so
I'll let y'all know how it all worked out.

More information about the Synth-diy mailing list