I've always read that schmitt gates were poor candidates for crystal oscillators. I just decided that, since I could try 'em in the same fixture, I'd give all the ones I had on hand a try. I didn't try the 74C04/40106 types because they're famously underpowered and I figured I'd never want to use one of them anyway. They are common in clock oscillators of very low frequency.

I suspect that there would be problems with biasing, etc, if one wanted really reliable oscillators. However, I was curious whether I could get by with a schmitt gate oscillator in a circuit where I needed one schmitt trigger inverter and had six in the package.

My small sample is not a basis for a production design, BTW, and I'd not use one in a production application, primarily due to conventional prejudice.

RE