[ntp:questions] oscillations in ntp clock synchronization
unruh-spam at physics.ubc.ca
Thu Jan 17 04:53:36 UTC 2008
ibuprofin at painkiller.example.tld (Moe Trin) writes:
>On Tue, 15 Jan 2008, in the Usenet newsgroup comp.protocols.time.ntp, in article
><SE6jj.4473$vp3.3451 at edtnps90>, Unruh wrote:
>>"David L. Mills" <mills at udel.edu> writes:
>>>Assuming the residuals are in the low microseconds and the variation
>>>period is much longer than the time constant and the same in both NTP
>>>and chrony, it would seem very likely that what you see is due to small
>>>temperature changes. This is the same thing seen in our primary servers
>>>synchronized by GPS and PPS. They hold typically within a few
>>>microseconds but occasionally budge twice that. I have traced this to
>>>small variations in machine room temperature as the air conditioning
>Servers as in ordinary computers of some kind? Depending, those are
>usually either cheap plug-in oscillators (under two bucks in onezies)
>that typically spec out at +/- 50ppm over the range 0-50C (or sometimes
Yup ordinary computers. They are between 0-100PPM out-- across about 8 of
them. But the cyclic variation is strange.
>+/- 100ppm over 0-70C), or just an HC-49 crystal. Good oscillators cost
>way to much money for this market, though I have seen systems where
>someone stuck a chunk of styrofoam over the oscillator to improve the
>>Maybe I will have to put in a digital thermometer into the case. The
>>chrony oscillations are 10's of milliseconds. and the oscillations in
>>the rate of the RTC vs the system time are larger than that.
>Are they moving in more/less the same direction at similar times?
>>Of course those might also be temperature oscillations-- different on
>>different parts of the motherboard (I assume that the rtc has a different
>>crystal to the one that runs the cpu clock).
>Most of the time - but "that depends". Are you talking PCs? The RTC
>is usually something low frequency (originally 1.8432 MHz, but now just
>as easily a 32768 Hz tuning fork). They could also be part of the
>CMOS RAM (usually a "Dallas Semiconductor" device - 24 to 34 pin DIP
>that also has the crystal and lithium battery in the same package).
>There are temperature compensated versions, but I've never seen one in
>a commodity type of computer.
What is strange is the about 1.5 hour timescale. There is nothing I know of
in any of the rooms that has that time scale.
Is there any way of getting the temperature from one of those CPU
thermometers (onboard) to better than 1C? I can see some major variations
in teh temp causing rate changes in the clock, but the small oscillations,
if they are caused by thermal effects, are much smaller temp variations
than are visible with a 1C accuracy.
> Old guy
More information about the questions