NavList:

----- Original Message -----

From: George Huxtable

To: navlist@fer3.com

Sent: Thursday, December 10, 2009 10:29 AM

Subject: Re: [NavList 11064] Automatic deviation calculation by electronic compasses

Lu Abel asked a fair question, which I will do my best to answer-

"Why do you assume that an area of low deviation is required as a
location on a steel vessel for a mounting place for a fluxgate compass?

It seems to me that ANY spot, regardless of how esoteric its deviation
table or Napier diagram, would be a suitable mounting place so long as
the deviation (table) does not change with vessel heading or activation
of machinery. "

But first, let's dispose of this one-
"(before you object, I'll admit that I'm assuming that a proper
constant-angular-velocity turn can be executed to compensate the compass) ".
I  ask Lu to examine my posting [10799] of 23 November. There, I pointed out
(after belatedly realising it) that there was no need at all for constant
angular velocity. As long as the vessel passed through every point of the
compass, and didn't turn too fast, all the necessary information was there
to deduce deviation, at ANY inconstant speed-of-turn.

Now, back to Lu's question. First, lets deal with the deviation of the
compass as it affects most small-boat-owners: those with fibreglass or
wooden hulls, in which the deviation is the result of odd bits of iron and
steel in its vicinity; particularly an engine. In that case, deviations are
small, both the sine(heading) or "hard-iron" component, and the sine(2 x
heading) or "soft iron" component. It can be handled by the algorithm we
have discussed, simply making a 360-degree turn. That would also apply to
the rare stainless-steel hull, and perhaps even to a construction of "very,
very, special steel", if it was a Nickel alloy of low magnetic permeability.

But an ordinary ferrous hull is a different matter. Its magnetic effects can
be enormous, and there's nowhere within the hull they can be escaped. The
history is informative here. In the 19th century, it was a powerful and
valid argument against iron and steel construction; that it wasn't possible
to correct a compass properly. Mariners knew how to make a deviation table,
but that wasn't sufficient. Deviation varied, not just with heading, but
with (magnetic) latitude, with heel-angle, and with cargo, and with time, in
ways which were not understood. There were countless accidents as a result
of compass errors. The work of Airy and (particularly) Kelvin helped to sort
it out. It resulted in the familiar polished binnacle, placed as high as
possible above the hull, in solitary splendour on the bridge so there could
be no undetected interfering objects. That carried permanent magnets to
correct for the ships "hard-iron" component, soft-iron hollow balls placed
thwartwise to correct for the "soft-iron" component of induced lengthwise
magnetism, and the Flinders vertical iron bar to correct for vertical
induced magnetism. Even after all that, there was the need to create a
deviation curve, and a prosperous trade of compass-adjuster to bring it
about.

Kelvin (then William Thomson) wrote a very readable account of this in
"Terrestrial magnetism and the mariner's compass", collected into vol 3 of
his Popular lectures and addresses" of 1891. This is the volume on
"Nivigational Affairs", which I thoroughly recommend. Kelvin owned a
fully-crewed 130-ton schooner, and did a lot of sailing aboard her.

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

Let's try a thought-experiment to demonstrate the sort of problem that can
arise, as simply as possible. We will be at the (magnetic) equator, so
there's no dip, and we can choose a spot with no magnetic variation, so the
field points North. Represent our ship by a length of board, with a compass
mounted a foot above it.

As we turn our ship, the card will ride over the lubber-line
correspondingly, and indicate true heading with no errors. But next, attach
a little bar-magnet to the board, directly below the compass, pointing
lengthwise, such that when the ship points North its field enhances that of
the Earth. That magnet represents the "hard-iron" component of a ship's
hull; we are ignoring the "soft-iron" component here, for simplicity. It's
clear that it causes no compass-error on a Northerly heading, nor (as long
as it's effect is weaker than the Earth's) on a Southerly heading, and the
maximum error will be on Westerly or Easterly courses. The error will vary,
more or less sinusoidally, as long as the field from the magnet is small
compared with that of the Earth. However, in a steel ship, that can no
longer be assumed. If the magnet happened to produce a stronger field than
that of the Earth, it would override that of the Earth when the heading was
South, so the compass card would turn with the ship and still indicate a
North heading. It would then be impossible to create a deviation table. All
you could say, if the compass read North, is that the ship was pointing due
North or due South!

That may seem an extreme case, but the effect of a steel hull can be so
great that it is by no means far-fetched. The binnacle components compensate
so as to reduce the hull effects to manageable proportions; single-valued at
least, that can then be handled by a deviation card.

George.

contact George Huxtable, at  george@hux.me.uk
or at +44 1865 820222 (from UK, 01865 820222)
or at 1 Sandy Lane, Southmoor, Abingdon, Oxon OX13 5HX, UK.

--
NavList message boards: www.fer3.com/arc
Or post by email to: NavList@fer3.com
To , email NavList+@fer3.com