Typical 32KHz clock crystals are very stable in frequency if you can keep them 
close to the turnover temp. If you can hold 1 degC it is .04 ppm. 
That is 40ppb - very good. If you can hold .1 deg C it is .0004 ppm. 
That is .4 ppb.  Very expensive.  (it goes as the square of the 
difference in temperature from the turn over temp - which varies from crystal 
to crystal. ). The turn over temp 
is around 25 degC. 

I saw some 40mm on a side thermoelectric heaters/coolers  units for $15 at 
Spark Fun. I thought I might 
have some fun. Two units and a power supply. Plus a honking current amp.  
Although with milliwatts for the oscillator it might not take a lot of 
heat pumping to keep the osc at a constant temp. And +/- 10C to 20C is probably 
enough range for experiments. 
 
Be interesting to see if this could be engineered to compete with OCXOs. A 
VariCap would be included in later experiments for tuning. If the initial 
experiments show promise.  6 wires out to start: 2 pwr, 2 thermistor, and 2 
Freq out. with the oscillator in the center of the coolers the material 
arrangement: Osc potted in high thermal conductivity. Void. Outer Insulation. 
With another thermistor or temp sensor in the outer insulation. Another in the 
vicinity on the outside.  I'm open to suggestion. And also possible proof I'm 
barking up the wrong tree. The idea is to keep the whole package as thin as 
possible to minimize heat flow to or from the outside - except through the 
coolers/heaters. 

Some uP to run a PID loop for temp. An update rate of  100 times a second for 
the PID should make it fine grained enough. DACs and ADCs good to 16 bits. 
Initial exploration eqpt need not be so good. Unless it is cheap. And then once 
you have the contraption running, disciplining it with 1 PPS might be of 
further interest. 

I currently have no method for testing such a rig for stability. 

Simon
Engineering is the art of making what you want from what you can get at a 
profit.
_______________________________________________
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.

Reply via email to