Jan,

I don't recall EVER having an oscillator fail to start or have it fail in some other way. In fact, the only IC's with which I frequently had problems were RS232 receivers, and 'LS244's. Over the years, I've probably replaced about 100 of each.

Actually, I've had the sense that the quality of commercially available crystals has, if anything, gone down rather than up over those 30 years.

Yes, once in a while one's best effort, even mine, turns out to be boardering on crap. Fortunately, I'm normally the only one who gets to see those cases.

I'm going to move my comments along these lines to the BBRAM thread, which you wisely started because the main topic of interest to me relates more to that than to oscillator startup.

If you think it desirable, I could put it in yet another thread, but I believe the business of managing RESET, managing Vcc rise and fall times, and managing the oscillator are all related to the matter of managing runaway MCU. Managing runaway MCU is, in turn, tightly coupled, IMHO, to corrupted BBRAM, among other problems.

RE