I remember a while back that I had an op-amp where if the op-amp's output came close to the rail, the output would invert. I'm not an op-amp expert so I need to ask a couple of questions:
How do I know if an op-amp exhibits these characteristics? If it is not specified as Rail-to-Rail?
This is the op-amp I'm planning on using simply because it is the cheapest SMT I could find on Digi-Key (still its almost a buck, what the heck?) It will be powered from +12v and GND. I only need to know when the signal is passing between 5.5 and 6.5v. I'll have the op-amp set up as a voltage follower and then use a resistor divider to step the voltage down. My micro will see 4.3v and 3.6v at those voltages. If the op-amp inverts when it is too high or too low, I'll simply ignore those readings.
I'd love to use a resistor divider and save a buck and a half (and I'm not thrilled about running an op-amp from the car's 12v accessory line) but I can't load the line. I'll have to settle for filtering the power line unless someone can suggest a reasonably priced op-amp suited for automotive uses. The op-amp will only draw about 3 mA, so R9 could probably be much higher to help with any noise and give D2 additional time to turn on when needed.
The TSM100 is an old slow opamp and a comparator like an LM (I can't remember its number). The datasheet doesn't have enough information.
Its inputs work at the 0V supply so it probably does not have the phase-inversion problem.
Thanks, I'm not worried about the speed. I'm going to be measuring the voltage across a thermistor and cycling power to an AC unit in a car. Turn on at 5.5v and off at 6.5v, and it'll take several seconds to vary that much.