[sdiy] Noninverting precision half-wave rectifier?

[David G. Dixon]

It's not that big of a deal, Barry.  It's just that there are two ways to
get precision half-wave rectification with an opamp: a (non-inverting)
precision diode and an (inverting) precision half-wave rectifier.  The first
has high input impedance but suffers from slew-rate distortion at high
frequencies (depending on the opamp used).  The second is distortion-free
but has finite input impedance.  This can complicate matters if you're
feeding a signal through an attenuator, as you'll get a non-linear
attenuation response.  Of course, in that case, you simply make the
attenuation pot into a variable feedback resistor, and Bob's your uncle.

Anyway, I thought I had found a non-inverting precision half-wave rectifier,
which would be distortion-free and have high input impedance.  However, it
was too good to be true.  That's all.  It's a "don't believe everything you
read" story, I think.


> Ok I'll admit my ignorance.  I don't get what you are trying to achieve
> or what the problems are exactly.  Is it your simulator that has the
> issue?
> Did you build a circuit on a breadboard to see if it behaves the same?
> Why not just invert the input signal?
> You can also use 2 diodes in the opamp feedback loop tied to a 3rd that is
> switched to
> ground or whatever that will clamp the output to a desired reference level
> (not V- for example).