[LEAPSECS] When did computer timekeeping get good enough for leap seconds to matter?

Poul-Henning Kamp phk at phk.freebsd.dk
Fri Jan 10 04:27:06 EST 2014


In message <20140110050534.E4388406063 at ip-64-139-1-69.sjc.megapath.net>, Hal Mu
rray writes:


>Do you have data or references for that? If I heard it from anybody less

>credible, I'd guess it was an urban legend? Was it accurate 50 years ago?


Several people deeply involved in the Danish electricity grid have told
me the same thing: The utilities prided themselves by their frequency
precision and used a lots of synchronous motor clocks internalll.

At the local powerplant, they would call up the speaking clock every
day at noon, compare it to the clock in the control-room and log the
difference.

(I have seen this myself as a kid.)


>The winter/summer variation doesn't make sense to me. If the PLL can cover

>daily changes, anything lower frequency would be covered for free.


The winter/summer variation has also puzzled me, but the consenus
seems to be that it was caused by electrical heating running through
the night, preventing them for regaining the lost cycles due to
business load during the day. (The danish grid was surprisinly
marginally provisioned until everybody started conserving electricity
after the OPEC-stunt in the winter 1973-74.)


>It's easy to collect data. Take an AC wall wart type transformer and connect

>it to a modem control pin that the kernel is setup to use for NTP's PPS

>signals. (Contact me off list if you want the software.)


I've done that many years ago, right when the deregulation too effect
in Denmark, and the average frequency dropped by 0.05Hz because they
could save fuel that way. Since then the frequency has just wandered
aimlessly around 50Hz, leaving the phase to do a random walk.

--
Poul-Henning Kamp | UNIX since Zilog Zeus 3.20
phk at FreeBSD.ORG | TCP/IP since RFC 956
FreeBSD committer | BSD since 4.3-tahoe
Never attribute to malice what can adequately be explained by incompetence.


More information about the LEAPSECS mailing list