Date:         Wed, 1 Mar 2000 00:34:48 -0500
Reply-To:     David Beierl <dbeierl@IBM.NET>
Sender:       Vanagon Mailing List <vanagon@gerry.vanagon.com>
From:         David Beierl <dbeierl@IBM.NET>
Subject:      Re: Precision Navigation compass
Comments: To: Mark McCulley <transporter99@hotmail.com>
In-Reply-To:  <20000229213954.20278.qmail@hotmail.com>
Content-Type: text/plain; charset="us-ascii"; format=flowed

At 04:39 PM 2/29/2000 , Mark McCulley wrote:
>Not sure exactly what you mean by "adjust compass deviation". You first
>calibrate the compass by putting it in CAL mode and driving in two
>complete circles.

That's what I meant.  Deviation is the error introduced by magnetic
influences in the vicinity -- on our (plastic) boat we can get within a
degree on most headings with just the crossed magnets in the compass mount,
but on an iron vessel it gets a whole bunch more complicated, with "heeling
magnets" and "flinders bars" and whatever they call the big balls of soft
iron mounted on either side of the compass (I used to know...).

My car compass uses a Hall-effect detector which drives a motorized
display, and it's compensated electronically with a couple of pots instead
of magnetically.  The sensor sticks on the windshield which gets it a
little farther from the iron than otherwise.  But the Hall sensor has to be
precisely perpendicular to the Earth to be accurate, so for boat or other
compasses where they actually care, the Hall element has to be mounted in
fancy gimbals which makes the whole rig quite expensive.  I suspect that
mine gets a fifteen degree error by going up or down a steep hill, aside
from the compensation errors.  I can get it within +/- 15 fairly easily.

I'd be really interested to know what algorithm they use to figure out
deviation by driving in circles.  I can see them plotting a deviation curve
('cause deviation generally makes one complete cycle from none to max
easterly to none to max westerly in 360 degrees -- but I don't see how they
figure out where the zero-deviation points are to anchor the curve.

Be really interesting to know how they decide their reading is out of
whack, too -- field strength?

Almost worth getting one just to try to figger it out... :)

thanks
david


David Beierl - Providence, RI
  http://pws.prserv.net/synergy/Vanagon/
'84 Westy "Dutiful Passage"
'85 GL "Poor Relation"