Hi Murray an interesting solution. What worries me is the thermal resistance to the outside world and the thermal inertia should have been considered in the control equation for designing the temp control system. If you apply too much insulation to the outside world them you are prone to overshoot and undershoot. It could be the oven in the 10881 is a bang-bang controller, but I would imagine something a little more sophisticated than that for milli-degree control. In which case should you not then adjust the control parameters?? Maybe they were not fully optimised??
Alan G3NYK ----- Original Message ----- From: "Murray Greenman" <[email protected]> To: <[email protected]> Sent: Tuesday, September 20, 2011 7:19 PM Subject: [time-nuts] Better option than oil > Folk looking for alternatives to dunking things in oil might consider trying > a variant of the following technique which I have used with success: > > The thermal stability of TCXOs and XOs can be improved considerably for > experimental purposes by slowing down the thermal time constant between the > device and ambient. A simple technique which I have used consists of a > recycled polystyrene box (picnic box, frozen food or dry ice transporter) > filled with dry sand. I place the device under test in a plastic bag or > other small container sealed as far as possible to keep sand out, and bury > it in the middle of the sand. The sand has a high thermal mass, and that > combined with the polystyrene insulation gives the device a thermal time > constant to ambient of many HOURS. > > While I'd not suggest this is the right solution for an OCXO intended for > use in free air, it is a way of achieving impressive short and medium term > stability with simple unheated devices, especially if operated in a modestly > controlled (e.g. air-conditioned) ambient environment, since the time > constant is significantly longer than the ambient temperature cycling. The > technique is almost good enough to remove diurnal ambient temperature > variation. > > 73, > Murray ZL1BPU > > > _______________________________________________ > time-nuts mailing list -- [email protected] > To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. _______________________________________________ time-nuts mailing list -- [email protected] To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
