We use the above inverter in one of our products. It is hooked up in the conventional cap & feedback resistor to square up a sine wave coming off a sensor tuned near resonance (either 500k or 1Mhz). There are several performance criteria that we'd like to know, but are never listed in the manufacturer's specs, because, what the heck, they don't spec analog use of a digital chip.
For example: What is the variability of the switching level from part to part? Consistency is very important to us and if the switching point varies too greatly, we'd have to toss the board. Also on the switching point, how does it vary with temperature? Plus, what are the consequences when the gate is over-driven? We are dealing with resonance, so input voltages above the supply of the inverter are possible.
We had one system where merely putting one's finger on the inverter caused a huge change in voltage. We wrote this off as a 'bad chip', but was it?
Finally, what is the jitter in the switching point (apart from the signal we're feeding it!)? The application is non-contact measurement and every bit of jitter shows up on the output as noise. Yeah, we have lots of fun designing oscillators!
Right now we have only two alternatives: Close our eyes and hope it all works out or Sit down and do a lot of chip measurements which can be obsoleted at a moment's notice as soon as a mask or process change occurs. We got caught using a competitor's opamp that went through a process change (which we found out later) which renderd it unusable for us. That company never admitted that the process change had any effect.
Is any of this information available or someone have some knowledge on it?
Thanks....Steve
