|
|
|
|
|||||
|
||||||
From: Paul Hirose (no email)
Date: Wed Jun 01 2005 - 23:22:17 EDT
Fred Hebard wrote:
> There are ways of transmitting signals both ways
> to account for the various delays. Paul Hirose was kind enough to tell
> us how this was done a few years ago; unfortunately, I didn't
> understand the mechanism well enough to reproduce it here.
I probably explained it badly. To illustrate the principle, suppose
the clocks at either end of the line are in perfect synchronism. If I
tap my telegraph key precisely at the top of a minute (according to my
clock), your telegraph sounder clicks, say, 0.1 s later. From your
viewpoint, your clock is 0.1 s *ahead* of mine.
But if we exchange roles, I hear your ticks coming late by
0.1 s. From my viewpoint, your clock is 0.1 s *behind* mine. By taking
the mean of these results, the delay is cancelled and
the true time offset determined. As a byproduct, the discrepancy between
the two measurements, divided by two, equals the propagation delay.
In practice, break-circuit clocks automatically generated the seconds
pulses, which were recorded as helical ink traces on paper by rotating-drum
chronographs. The equipment and techniques were developed with amazing
rapidity. By the time of this report in 1858, telegraphic longitudes
were routine:
http://www.history.noaa.gov/stories_tales/cs1858_6.html
The great length of the trans-Atantic cable did create technical
challenges, but they were were quickly overcome:
http://www.lib.noaa.gov/edocs/BACHE2.htm#CARDINAL
("A Cardinal Point for Longitude")
According to footnote 26 at the bottom of that page, "The seconds in
time of longitude of the Cambridge Observatory [i.e., the seconds
portion of the Harvard College observatory time offset with respect to
Greenwich] as determined with the 1927 North American Datum was 30.928
and as determined in the 1983 North American Datum was 30.802. The
original telegraphic determination of longitude fell approximately
half-way between these values demonstrating beyond a doubt the accuracy
of the 'American Method' of longitude determination."
I recommend science writer Trudy Bell's article on telegraphic
longitude, "The Victorian Global Positioning System" (about 840 k PDF):
http://www.tbp.org/pages/publications/BENTFeatures/BellSp02.pdf
|