Home Up Search Mail
NEW

Calibrating the compass readings

If one measures with a compass (like a Suunto KB-20) the azimuth (azi_mag), one has to be aware that the magnetic north is not the same as the true (real) north. The differance between magnetic- and true north (magn-true north_x) has to be determined before getting the true azimuth (azi_c) of an aligment or direction, in the following way:

azi_c = azi_mag + magn-true north_x

Evaluation of below mentioned calibration methods

In all results of specific sites the following compensation has been used:

if magn-true north₁ is available and magn-true north₂ is almost the same; magn-true north₁ will be used.

if only magn-true north₂ is available, magn-true north₂ will be used.

if magn-true north₃ is totally different from magn-true₂; magn-true north₃ will be used.

method magn-true north₄.

Remark: the above errors in azi_c is with out any possible systemetic errors, e.g. due to iron contained rock, wrong measurement equipement, etc.

Calibration methods

So the compass reads not the true azimuth (azi_c) but the magnetic azimuth (azi_mag, all measured in 1996), thus the compass reading should be calibrated. To do this calibrating, some methods of obtaining magn-true north are being investigated:

magn-true north₁: use the suns azimuth (azi_sun) on a particular time of the day and compare this with the calculated azimuth (azi_cal).

Differance between suns azimuth measured and calculated
Date	Time (local time)	azi_sun	latitude	longitude	azi_cal	magn-true north₁
4/8/96	20:21	300^o30'	58^o32'	3^o32'	291^o41'	-8^o49'
4/8/96	21:11	312^o0'	58^o32'	3^o32'	302^o9'	-9^o51'
5/8/96	09:20	115^o0'	59^o0'	3^o14'	107^o44'	-7^o16'
6/8/96	17:44	265^o30'	59^o0'	3^o12'	258^o23'	-7^o7'
6/8/96	21:00	308^o0'	59^o0'	3^o14'	299^o51'	-8^o9'
7/8/96	11:36	151^o30'	59^o0'	3^o11'	145^o14'	-6^o16'
8/8/96	21:01	309^o0'	58^o12'	6^o45'	296^o54'	-12^o6'
9/8/96	09:02	108^o30'	58^o12'	6^o45'	100^o44'	-7^o46'
13/8/96	20:27	293^o30'	57^o28'	5^o22'	290^o25'	-3^o5'
14/8/96	11:17	140^o6'	57^o25'	5^o13'	137^o22'	-2^o44'
14/8/96	20:07	295^o6'	56^o50'	5^o4'	286^o28'	-8^o38'

magn-true north₂: determine the magn-true north₂ by means of the Ordnance Survey maps or by URL for Magn-grid and using the true-grid estimation of A. Thom (delta(longitude)*sin latitude).

Differance between magnetic and true north according to OS maps
		map
Location	magn-true north₂	year	magn-grid	delta/year	true-grid
Delfour	-6^o42'	1988	-6^o30'	10'	1^o32'
Achavanich	-6^o38'	1990	-6^o30'	10'	1^o8'
Corrimony	-7^o15'	1997	-5^o0'	7.5'	2^o23'
Ring of Brodgar	-6^o13'	1979	-8^o0'	10'	1^o3'
Maeshowe	-6^o13'	1979	-8^o0'	10'	1^o3'
Wideford Hill	-6^o13'	1979	-8^o0'	10'	1^o3'
Callanish I	-8^o26'	1988	-5^o30'	10'	4^o16'
Rudh' An Dunain	-8^o27'	1996	-5^o0'	7.5'	3^o27'

magn-true north₃: By determining azi_mag of distant objects and comparing them with azi_m. azi_m can be gotten from the OS maps and difference true-map north.

Differance between OS-map azimuth and magnetic azimuth
Location	azi_mag	azi_m	magn-true north₃
Maeshowe	226^o30'	219^o57'	-6^o33'
Rudh' An Dunain	163^o0'	146^o33'	-16^o27'

Remark: The magn-true north₃ of Rudh' An Dunain is quiet big! This could be due to the fact that there is much ferro rock in the neighbourhood (like the close by Cullin Hills at some 7 km).

magn-true north₄: not used during this research, but it seems to be the best method when using the compass:

determine of three distant object the azi_mag
determine the azi_m of these three objects (using the true-map north differance) .
determine the difference between azi_mag and azi_m and average it over the three samples. This is magn-true north₄

Some accuracy considerations

The magn-true north₁ seems to have an error of some 15'.

_sun

_cal

The magn-true north₂ seems to have an error of some 30'.

30' for magn-grid error

The magn-true north₃ seems to have an error of some 15'.

_mag

The error in the averaged magn-true north₄ is around 10'

₃

The error in the measured (magnetic) azimuth (azi_mag or azi_sun) is at least 15' (depending on type of compass).
The error in the latitude is around 1' (or in case *, 3').
The error in the calculated suns azimuth (azi_cal) should be within 5'.

some arcseconds

The error in the altitude of the horizon (alt_hor) is some 5'.
The error in the azi_m could be some 5'.

Disclaimer and Copyright
Home Up Search Mail

Last content related changes: July 18, 2000