It all depends on how accurate the frequency has to be. If you only need 2%, I would use a C8051F300 microcontroller's built-in oscillator (24.5MHz +/-2%) and divide it down using the processor itself. The chip comes in a 11 pin QFN that is 3x3 mm, a little bigger than you need, but it does not require any external component. The frequency can be trimmed to better than 2% on many Silabs chips but I am not 100% sure that is the case on this one.
On Fri, Nov 1, 2013 at 11:30 AM, Dan Kemppainen <d...@irtelemetrics.com>wrote: > Hi all, > > I've got a project upcoming that will require a frequency of 32.768 Khz > in a harsh environment (Imagine a thousand G's at 100Hz with 150 Deg C > temperatures). Also, this thing needs to be small, 2mmx1.5x1.5mm or so. > It also has to be low power. Frequency stability is probably less of a > concern than just surviving (some frequency error vs. temp can be > trimmed out with other smarts in the design. > > > In the past, crystals just haven't liked surviving due to the > construction. Have had good luck with ceramic resonators at higher > frequencies (50Mhz and up). I've also looked at silicon oscillators, > which will work in the application, however with the chip and associated > resistors/caps they get a little bigger that what I was hoping for. > > Is anyone aware of a frequency source (crystal/resonator or other) in a > small package that is robustly mounted? > > Or are there any ceramic resonators available that are in small packages > in those low frequencies? I checked the big distributors, and did not > have any luck. > > > Dan > > > _______________________________________________ > time-nuts mailing list -- time-nuts@febo.com > To unsubscribe, go to > https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. > _______________________________________________ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.