Hello Dr. Stellmach,
Great info! Thanks so much for the detailed analysis. Some comments: > 27MHz can be realized by fundamental crystal, but it's on edge of > technology because the crystal is pretty thin already. Therefore, a > tighter specification is critical and costly, so use PLL design with > 9MHz or less instead! Instead of a PLL, it's also possible to use 3rd overtone crystal with an LC tank circuit on the output side of the inverter (after the series resistor). This allows the crystal to be about 3x thicker. You are absolutely right, fundamental crystals get very thin, brittle, and shock-sensitive above 24MHz. These days, there is very little price difference between a fundamental 27MHz or a 3rd OT 27MHz though. Thus the LC tank may actually increase the system cost by $0.05 - $0.1 versus a fundamtental 27MHz part; I have had different opinions from different crystal vendors in Asia about this (depending on their capabilities). > We often have discussions between the (digitally minded) designers of > the inverter and the crystal supplier: For us, an additional critical > aspect for series production are oscillation start margin (which is > very crudely checked by our hardware designers) This is the biggest problem in my opinion. Infineon (_www.infineon.com_ (http://www.infineon.com) ) has a great application note on this on their website (how to guarantee start margin in mass production etc). You mentioned that you have "clock trimming by software correction": How does this work in mass production? Do you measure the frequency during production at room temp, then add/subtract a number of seconds from the real time clock-time once a day or so to compensate for the offset? Also, what kind of failure rates and failures do you see with such a large sample set in the field :)? Do you see problems with aging, mechanical vibration, the large temperature changes during day/night etc? Thanks, Bye, Said _______________________________________________ time-nuts mailing list time-nuts@febo.com https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts