Thermomechanical fatigue can significantly reduce the lifetime of Peltier devices if the ripple current flowing in the Peltier device is too high. This can become an issue with switchmode drive to a Peltier cooler.
Bruce > > On 03 June 2017 at 11:02 jimlux <jim...@earthlink.net> wrote: > > On 6/2/17 2:51 PM, Donald E. Pauly wrote: > > > > > > This is an improvement of 476 to 1. You apparently have not thought > > thru what improvements are possible with thermal coolers/heaters. > > Among these is near instant warm up and greatly reduced power for > > thermal management. > > > > > > without getting into the whole crystal issue, one of the advantages of a > heater is that it can be VERY simple (and hence reliable, just on a > parts count basis). With a decent package, once it's hot, the power > required to keep it hot can be quite low. > > With a heat/cool, you need to be able to have a bipolar supply to the > peltier device, and they're not particularly efficient (that is, to > extract 1 Watt of heat, you're putting in significantly more than 1 watt > of DC, and rejecting 1+X watts to the outside world. > > And then, if you use a linear power supply/amplifier to drive the > device, that is probably a class A device, and somewhat lossy. A > switcher would be more efficient, but then you have the problem of > switching noise, in close proximity to the crystal. You could put a big > low pass filter in, but now you're adding even more components. > > There are undoubtedly some cases where the thermoelectric scheme would > work better - for instance, you have a system with a TCXO and it's > really set up for the TCXO to be at 25C, and you want to regulate that. > > _______________________________________________ > 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.
