On Fri, 06 Jan 2012 16:33:19 -0500, John Ackermann N8UR <j...@febo.com> wrote:
>On 1/6/2012 4:14 PM, Rick Karlquist wrote: > >> Before making this into a science project, consider this data >> point: We converted the oscillator in a 10811 to run in mode >> B at 10.95 MHz. The tempco in mode B is about 30 ppm per degree >> C. Needless to say, the converted 10811 was extremely sensitive >> to crystal temperature. I could wave my hands around it and notice >> the temperature change from the air currents. However, simply >> putting the 10811 into a small empty cardboard box completely >> solved this problem. No rocket science necessary. So in the >> short term, this simple box was like a double oven. Of course, >> in the long run, the 10811 thermal gain is only 1000 or so and >> you can see the effect of the HVAC shutting down at 7PM, etc. > >That's interesting, Rick. I was also thinking that just still air was >most of the battle, but since I have the opportunity to do it "right" >within reason and this is a fairly major project, I thought that adding >some additional thermal isolation couldn't hurt, so why not? >(time-nuttery, etc.) Still air certainly matters for thermocouple effects which can be difficult to distinguish from 1/f noise but that is what Dixie cups are for. I have seen voltage noise improvements of more than 10 times in some cases but I am not sure how that would apply to the voltage control signal of an OCXO or the OCXO itself yet. There must be some improvement to be gained since double oven OCXOs do exist. Like you say, the added insulation can not hurt. I suspect it would be more effective either mounted directly around the OCXO (probably difficult) where the surface area would be smaller or mounted on the outside of the enclosure (impractical) so the added thermal mass of the enclosure itself would buffer the temperature changes. The inside of the enclosure seems to be an easy compromise and would dampen convection. _______________________________________________ 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.