|
|
|
|
|||||
|
||||||
Subject: Re: Fwd: Marine Electronics Issue #18: Fluxgate Compasses
From: George Huxtable (george@XXX.XXX)
Date: Sun Sep 22 2002 - 16:51:58 EDT
I hope it's not too late to add a postscript to this thread, about fluxgate
compasses, particularly the Autohelm design, and stir the pot a bit.
Such a fluxgate compass resolves two components of magnetic field, with
respect to the plane that the sensor is held in. To determine the direction
of the horizontal component of the Earth's field, it is essential that the
sensor is held in a horizontal plane, with no tilt. In parts of the world
with a large angle of dip (that is, where most listmembers live and sail),
the resulting bearing error from a residual tilt can be considerably
greater than the tilt error itself. It is therefore essential that some
sort of levelling or gimballing is provided.
As an aside, previous discussion on this list about fluxgate compasses
centred mainly on a different arrangement which contained three sensing
elements, which were not gimballed but "strapped-down" to the vessel, in
which case any tilt of the vessel had to be sensed separately and included
in a computation. We are not discussing that type here, but a much simpler
arrangement.
A traditional on-board needle-compass is usually mounted in gimbals, and a
further degree of self-levelling is achieved by allowing the card to tilt
on its point-bearing to maintain a horizontal plane. A gimballed fluxgate
compass has only the the first of these levelling devices, so it needs to
be gimballed more freely and damped better and balanced more accurately
than the gimbals on a traditional compass, to achieve the same level of
accuracy. A hand-held fluxgate such as the Autohelm, which has no internal
gimballing whatsoever, relies on the observer to act as an active gimbal
mounting, and he is expected to take positive steps to hold the plane of
the instrument parallel to the horizon, both in his side-to-side and his
up-and down direction, while the measurement is made.
This presents few problems in daylight, in calm conditions. But put
yourself in the position of a navigator going out on deck on a black night,
on a heeling vessel, when it's rough enough so he has to hang on and is
never really sure which way is up, to take a bearing of a 15-second light
flashing intermittently above the wave crests. A familiar situation to most
of us, I reckon. Somehow he has to hold his compass parallel (in both
directions) to a horizon he can't see. An impossible task, it seems to me.
I would welcome reports from users who have tried to take bearings with an
Autohelm under such conditions.
In my view, the design of the Autohelm makes matters worse in that its
display is hidden from the navigator while bearings is being taken. In that
way, the extreme sensitivity of the instrument to tilt error is concealed
from the user. Me, I would prefer to see what was going on. Wouldn't you?
Matters are improved somewhat in that the Autohelm compass allows a number
of bearings to be taken in quick succesion and then averaged, which will
certainly do something to reduce random scatter. However, in that "dark
night" scenario, there might well also be a systematic error, a tilt bias
the same way each time, which the averaging process will do nothing to fix.
I have tried the Autohelm personal compass, but for the reasons given above
do not own one. It's quite cheap, and convenient in other respects. So I
would be interested to learn from those that do, whether my prejudices
about it, as stated above, are based on reality or not.
George Huxtable.
------------------------------
george@XXX.XXX
George Huxtable, 1 Sandy Lane, Southmoor, Abingdon, Oxon OX13 5HX, UK.
Tel. 01865 820222 or (int.) +44 1865 820222.
------------------------------
|