Trying to understand why op amp voltage follower goes high when fed negative input

Question

I have a project where I want to take the positive and negative outputs of a passive differentiator and produce only positive pulses. E.g. when the switch feeding the differentiator is closed I get a positive pulse which I want to merely pass along; when the switch is opened I get a negative pulse which I want to invert and also pass along as a positive pulse. I got this to work by feeding the output into an active rectifier circuit. But the interesting thing is that while experimenting with Ltspice I found that I could achieve the same result with the voltage follower configuration shown below, and sure enough it worked on the breadboard, too.

I understand that when you exceed the ratings for a particular device you can get bad or unexpected results. (In this case my negative going pulses approach minus nine volts whereas the LM358 VICR is -.3V to 32 volts.) Still, if anyone can explain exactly why this works, I would be glad to know.

Also, is the op amp being damaged? Is abusing an op amp in this way a bad practice?

^{simulate this circuit – Schematic created using CircuitLab}

Answer 1

The current into the input during the negative pulses is small, so no damage occurs. But leaving the input common mode voltage range can cause phase reversal leading to a high output. This often happens for old opamps.