Hi Your temperature plots look like mine. I suspect the other unit has a broken temperature detection chip. I've seen that happen on TBolts.
Bob On Sep 9, 2013, at 10:44 PM, Arthur Dent <[email protected]> wrote: > I have a slightly earlier version, or a close cousin, of the NTBW50AA that > I modified and I've posted photos of it on this list before: > http://i906.photobucket.com/albums/ac262/rjb1998/NTPB15AA05.jpg > > The LH plot from the NTBW50AA by quartz55 doesn't look quite right. Here > is a LH plot from my NTPB15AA unit with most of the scale factors set the > same as the scale factor in the plot from the NTBW50AA to make it easier > to compare. > http://i906.photobucket.com/albums/ac262/rjb1998/NTPB15A_zps19b3bd33.jpg > > If you look at the temp plot from my unit it has small steps and is what I'd > expect to see. The plot from the NTBW50AA looks like it has some smaller > steps but it looks like it hits a limit at 36.750 and doesn't go beyond that > value, which isn't right. It looks like a higher order bit is being turned on > and off randomly causing the large apparent jump in temp, which > probably isn't really happening because some of the other traces would > be affected by any real jump that large. You could just ignore it because it > would probably only have an effect during carryover. > > The NTBW50AA oscillator probably hasn't settled down yet because the > DAC voltage is changing a lot more than mine and the 10Mhz doesn't look > anywhere near as stable. The design of the 2 units appears nearly identical > so I'd expect similar performance. Also the OSC ADEV at 10000 tau isn't close > to what mine appears to be. That may improve after the unit has been on > for a month or so but the temp plot just doesn't look right so the sensor may > be bad. > > -Arthur > _______________________________________________ > 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.
