Op amp with diode and resistor in parallel in feedback

[injb]

What kind of amplifier is this? It's an inverting op amp with a diode and a resistor in parallel in the feedback path. I haven't been able to find any examples like this, but I'm seeing it in a circuit. I've seen explanations of op amps that have just a diode in the feedback path (log amps), but how does the resistor change that? The input to this is AC coupled - I'm not sure if that makes any difference in how to understand the feedback path?

[schmitt trigger]

A precision half wave rectifier with gain.

This circuit is identical to one in the book Stout and Kaufmann's "Handbook of Operational Amplifier Circuit Design", Chapter 16.

[injb]

Thanks, that's a big help. I don't have that book, but I've been reading about precision rectifiers, and comparing what I have to examples. The closest I've found is a version that uses another diode in the opposite direction, like the one shown here:

https://www.circuitlib.com/index.php/lessons/94-precision-rectifiers

So if I understand this correctly, the only difference with my example is that when the output goes low and forward biases the diode, some current can still flow from the input through the resistor, but it would be extremely low because the diode would have almost no resistance - is that right? In that case, why would you ever bother with the second diode (from the example in that link)?

[Vovk_Z]

Doesn't it logarighmic amplifier?

[schmitt trigger]

As clearly shown in the attached LTSpice capture, the single diode amp does provide rectification to the waveform, but its output drops slightly below zero until the diode's starts conducting.

That is the reason for the second diode, if you want to have a rectified waveform that starts precisely from zero.

[nigelwright7557]

What kind of amplifier is this? It's an inverting op amp with a diode and a resistor in parallel in the feedback path. I haven't been able to find any examples like this, but I'm seeing it in a circuit. I've seen explanations of op amps that have just a diode in the feedback path (log amps), but how does the resistor change that? The input to this is AC coupled - I'm not sure if that makes any difference in how to understand the feedback path?

I use a similar circuit on an op amp just before an A2D converter to stop input to A2D going negative. Works a treat.

[injb]

I use a similar circuit on an op amp just before an A2D converter to stop input to A2D going negative. Works a treat.

Interesting - that's what it's used for in the circuit I'm looking at. But what confuses me is that this circuit has 2 outputs that go to the ADC, and one of them uses this circuit, but the other one uses the 2 diode version described in the link I posted above. Now I'm pretty sure that of these 2 outputs, one is used for diagnostic,s to detect things like the sensor missing from the input, and the other one is actually used for important measurements. The 2-diode version is the one used for measurements. Would they really have decided that the diagnostic output could get by without the second diode, just to save a few cents? Or would there be some more practical reason?

[schmitt trigger]

Without seeing the schematic of the circuit which are fed these signals, is difficult to speculate.

I have used the single diode precision rectifier when I want to detect precisely a zero voltage crossing with a comparator.

The ubiquitous LM393 has an input common mode range which includes -300mV, even with a single supply. Thus by feeding a signal that slightly goes below zero, you can be certain that the exact zero volts, plus or minus the small offset voltage, is detected.