sound like not worth it. My idea of 'cooling' was say keeping the insides at something like 25C. Just having a 2nd TBolt as a spare would be easier
-pete On Tue, Jan 18, 2011 at 9:41 AM, Bob Camp <[email protected]> wrote: > Hi > > There's pretty much nothing in a TBolt that wears out. The heat rise on the > parts on the pc board is modest and the only heated part is the OCXO. > Cooling down the OCXO actually increases the stress on it (more heater power > pulled => more stress on the heater). > > Obviously you can get it to hot. The OCXO does have an upper temperature > limit. I don't think that extreme measures are needed to keep it at a > rational temperature. Simple fans and boxes seem to be working pretty well > for people. Any setup that results in sub 60C temperatures at the OCXO is > likely to have very little impact on reliability or performance. > > The more important question is how you package it to reduce gradients. If > you use the internal sensor, it's location is a bit sub-optimal for tracking > the OCXO. You could put all sorts of sensors throughout the package. An > easier thing to watch is the correlation between external temperature and > EFC voltage. As long as EFC correlates (positive or negative) with outside > temperature, you have room for improvement. > > Bob > > > > -----Original Message----- > From: [email protected] [mailto:[email protected]] On > Behalf Of Pete Lancashire > Sent: Tuesday, January 18, 2011 12:16 PM > To: Discussion of precise time and frequency measurement > Subject: [time-nuts] Thunderbolt cooling vs heating for stability > > Been getting ready to build a box for the t'bolt and instead of it all > getting hotter to maintain a stable temp > is the idea of cooling with a Peltier cooler, or a mix of a Peltier > and a heater ? > > Extending life time is the major concern > > -pete > > _______________________________________________ > 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. > _______________________________________________ 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.
