Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP Path: utzoo!linus!decvax!decwrl!amd!fortune!wdl1!jbn From: jbn@wdl1.UUCP (jbn ) Newsgroups: net.unix-wizards Subject: Re: RE: Almost Accurrate Clock Message-ID: <461@wdl1.UUCP> Date: Wed, 3-Oct-84 21:24:28 EDT Article-I.D.: wdl1.461 Posted: Wed Oct 3 21:24:28 1984 Date-Received: Fri, 5-Oct-84 08:28:31 EDT Lines: 27 Nf-ID: #R:rna:-29700:wdl1:17100010:000:1744 Nf-From: wdl1!jbn Oct 3 12:42:00 1984 The Heath WWV receiver is not just counting cycles of a waveform; WWV is actually sending the time and date in a suitable format. So this is an absolute reference, not a relative one. For those of you who do not know what WWV is, a brief explaination is in order. WWV is a radio station operated by the US Government National Bureau of Standards. WWV broadcasts various time signals on different frequencies. The time broadcast is controlled directly by multiple cross-checked cesium-beam atomic clocks at the WWV site in Boulder, Colorado. These are in turn checked against the United States Master Clock at the U.S. Naval Observatory in Washington (composed of over a dozen cesium-beam atomic clocks cross-checking each other). Cumulative error in WWV signals is under 300 microseconds per century. WWV time is defined relative to the Earth's rotation, and the rotation of the earth is slowing down; the Naval Observatory measures this and every few years WWV inserts a ``leap second'' to compensate. When this occurs, the time sent by WWV jumps one second exactly. These adjustments are made only at the beginning of a year. (Exactly at the beginning of a year, 0000Z UT, 1 January). Receivers which obtain WWV signals on bands reflected off the ionosphere (these are the short wave bands) are subject to some drift due to ionospheric height variations; this drift is measured in tens of milliseconds but the long-term drift is zero. There are also satellite signals which can be used for timekeeping and navigation; these are also very tightly controlled. Unless you have some very exotic timekeeping application requiring millisecond-level syncronization at distant points, the Heath clock is fine.