NavList:

   

Reply

Discussion continues about Alex's interesting quote from a Russian
expedition to Japan in 1853/4.

I posted an answer, on 15th November, with I will repeat again at the end,
in case it didn't get around the list (though a copy was reflected back to
me).

A modern Times Atlas gives the longitude of Nagasaki as 129 deg 52 E, which
is exactly the same as  Raper (in 1864) gives for "Nangasaki". What is
remarkable is the agreement, with a minute, between that figure and the
long. quoted by Frank Reed from Kruzenschtern of 230 deg 07W, as early as
1805!

Frank wrote-

>It's interesting that the Russian chart, published in German, is  already
>referencing longitudes to Greenwich at this early date.

but that's not such a surprise, as Kruzenschtern had served in the British
Navy from 1793-7.

Frank made the interesting suggestion that a lunar occultation might have
been employed to get the "timing" error of the chronometer, and indeed, so
it could. That was a very accurate technique for determining Greenwich time,
and hence the longitude. A star extinguishes instantaneously behind the
Moon. Its precision was as good as were the predictions of the Moon's path,
which were getting much more accurate by the 1850s.

But useful occultations, from a particular location, don't happen that
often, averaging about 10 in a month, so the technique was more useful
to.geographers (who could wait) rather than to navigators. And that average
of about 10 per month presumes that all stars down to 6th magnitude are
considered, and are mostly of 5th and 6th magnitude stars. Occultations of
bright stars are very rare. So timing of an occultation needed a telescope
rather than the naked-eye. It was difficult to keep a telescope continously
trained on a star, even at anchor in harbour, except under very sheltered
conditions. To keep watch on a star, to time its disappearance, really
called for a telescope firmly planted on the ground; another reason for
going ashore on that rock..

But there were a lot of detailed corrections called for, worse than clearing
a lunar, which would be unfamiliar to an ordinary navigator,  though within
the slills of an astronomer, geographer, surveyor. Raper (1864) explains how
it's done. Were the Russians up to such sophisticated techniques, then? It
depends on the nature of the expedition. Was this an official
government-backed voyage of exploration, staffed by the Russian savants of
the day, I wonder? If so, longitude by occultation would likely have been
within their grasp. But if it was a trading voyage, with sea-navigators,
then probably not. Alex may know, from his reading.

===========================
===========================

Here's my posting of November 15, repeated, for what it's worth.

Alex wrote

>I am reading a book by a famous Russian XiX century
> author, Goncharov, on his voyage to Japan in 1853/4.
> ...The Russians asked for a
> permission to land on some rock between their ship place
> and the land. Apparently a small uninhabited island.
> The reason they clamed for this permission was the
> "Necessity to check chronometers".
> Apparently they thought that this was a good reason
> for landing on a rock).
>
> Unfortunately the author of the book was not interested
> in navigation and did not want to explain to the readers
> this point. WHY did the Russians think that this was a
> legitimate reason?
>
> Why exactly did they need land to check their chronometers?
> Or, more precisely, why did they think that "checking chronometers"
> is a legitimate reason for a landing?
>
> What was the proper procedure of "checking chronometers" they had in mind?
> Was this by the Lunars?
> Did they mean that a firm ground
> is needed for the Lunars? Or Jupiter satellites? to install a powerful
> telescope?

================================

Response from George.

There were things you could do on firm ground that couldn't be done from a
ship, even at anchor.

I will presume, for short, that Russians had then adopted Greenwich Time,
without being certain that they had.

In general, even the best chronometer will not keep perfect Greenwich Time.
There are two important things that need to be known. Unfortunately, there's
often confusion between them, because the words 'fast' and 'slow' are used
for both.

First, what is its error on GMT at the moment? It will be leading  or
lagging on GMT by an amount, which needs to be known to within a second or
so for good navigation.

Second, what is its rate of gaining or losing on GMT? Usually, it will not
run exactly 24 hours in a (mean Greenwich) day, and the mariner was not
expected to adjust it to do so; that was discouraged, even forbidden. The
rate will have been previously set, with some precision, by its maker or
adjuster, on land, as near to correct as possible. The mark of a good
chronometer was not that it kept perfect time, but that it had a CONSTANT
rate, whatever the weather, which would apply over a whole voyage. The
navigator's duty was to 'rate' it, to measure how much it was diverging from
GMT, whenever he could. Then he could apply that rate, over the weeks or
months of a forthcoming voyage, to predict what the time error on GMT would
be from  one day to the next. If the vessel would be returning to St
Petersburg, an immensely long voyage lay ahead. So it was necessary to
establish the daily rate. That rate will have to be multiplied by the number
of days into a voyage, so the best possible precision must be applied in
rating a chronometer, to measure its daily rate to within a tenth of a
second, or perhaps to a few hundredths. How is the mariner to do that job?
Deducing the moment of noon, from Sun altitudes measured during a day, isn't
precise enough to rate a chronometer over a period of a few days; it would
need a long stay.

The answer lies in the stars. If the navigator goes ashore, and sets up a
post with a clear view of the Southern sky behind it, and some sort of
peep-hole to position his eye well in front of it, he can time low-altitude
stars as they disappear, instantaneously, behind that post. The interval
between passages of the same star is exactly one sidereal day, which is just
3 minutes 55.91 seconds (and umpteen decimal places) short of a day by GMT.
The sightline doesn't need to point due South, the post doesn't need to be
precisely vertical. The chronometer doesn't need to be taken ashore, if some
signalling procedure, by sight or sound (gunshot?) can be devised. The main
error is in relating the observed star-time to the dial, and the ticks, of
the chronometer..

It was by such methods (using a window-frame and a nearby chimney) that
Harrison would test his clocks and chronometers, and how Cook would re-rate
his, during a stay in Queen Charlotte Sound, New Zealand.

Note that this technique finds the rate precisely, but doesn't determine the
chronometer's error, for that day, on Greenwich Mean Time. That error needs
to be known, but not to such high accuracy as the daily rate; to within a
second or two, ideally. However, such precision did not become possible over
great distances until the advent of cable, and later radio.

If that sightline had been set up to be precisely South (which cam be done
by methods that have already been discussed on this list) then that would
allow an accurate time-sight on the Sun, or better, a star. But that would
do no more than determine local time, not Greenwich Time. If the longitude
of the spot was already precisely known (but likely, it wasn't), that would
allow Greenwich Time to be determined.

Otherwise, the navigator had to determine GMT, as accurately as he could.
That could be done by lunars, averaging a long sequence of observations,
which could be taken on board ship. But the best and quickest method was to
time, by the chronometer, the disappearance of one of Jupiter's inner
satellites into the shadow of the planet. That requires a high-magnification
telescope, which had to be mounted on firm ground. It was not possible to
observe on-board, even at anchor. There remained a few seconds of
uncertainty, because the moment when the last glim disappeared depended
somewhat on the light collection of the telescope, and on the observer's
eye. It was the best method at times when Jupiter was being cooperative.
However, there were whole months when Jupiter was too near the Sun to be
seen, and even when it could, many of the satellite immersions occured in an
observer's daylight.

===========================

Alex adds-

> P.S. I am well aware that the first underwater transoceanic cable layings
> were made at about the same time. So the Lunars were doomed, already then,
> even as a
> mean
> of checking chronometers:-(

===========================

Not really, no. You couldn't pick up timing from a cable when at sea! Only a
limited number of ports became linked to the cable network, and others had
to wait for radio to come. It depended on the potential for lucrative
traffic. Lunars were doomed, when chronometers became cheap enough for
vessels to own a set of three, so if one went wrong it was obvious, and you
could still get the time from the other two. Before the days of radio, that
was an important matter.

Until the days of extensive cable networks, longitudes were compared by
sending a batch (sometimes as many as 20) of precise chronometers,
accurately rated and timed. They went to and fro between measurement
stations, often on many repeated trips, to compare with local time at each
end.

Fred mentioned cable-laying vessels. These needed their chronometry to be
particularly accurate, because they wanted to know exactly where their
cables would lie, in case (as often happened) they needed to be fished for
later if they failed.

George.

contact George Huxtable at george@huxtable.u-net.com
or at +44 1865 820222 (from UK, 01865 820222)
or at 1 Sandy Lane, Southmoor, Abingdon, Oxon OX13 5HX, UK.

   

Reply