About the output data

[Keyaoli1995]

Hi,
I ran this code yesterday, and I found that the output values of x, y, z is very large,
my input data unit is μT just like ‘-24.9,41.1,-40.2’， but the output in 'out.txt' like '-201751.784042 ,-4722.019151 ,91292.957875
'
where is wrong?
thanks for answer.

[nliaudat]

Hi,
Can you give me your MField value ?

class Magnetometer(object):

'''
To obtain Gravitation Field (raw format):
1) get the Total Field for your location from here:
   http://www.ngdc.noaa.gov/geomag-web (tab Magnetic Field)
   es. Total Field = 47,241.3 nT | my val :47'789.7
2) Convert this values to Gauss (1nT = 10E-5G)
   es. Total Field = 47,241.3 nT = 0.47241G
3) Convert Total Field to Raw value Total Field, which is the
   Raw Gravitation Field we are searching for
   Read your magnetometer datasheet and find your gain value,
   Which should be the same of the collected raw points
   es. on HMC5883L, given +_ 1.3 Ga as Sensor Field Range settings
       Gain (LSB/Gauss) = 1090 
       Raw Total Field = Gain * Total Field
       0.47241 * 1090 = ~515  |
       
    -----------------------------------------------
     gain (LSB/Gauss) values for HMC5883L
        0.88 Ga => 1370 
        1.3 Ga => 1090 
        1.9 Ga => 820
        2.5 Ga => 660 
        4.0 Ga => 440
        4.7 Ga => 390 
        5.6 Ga => 330
        8.1 Ga => 230 
    -----------------------------------------------


 references :
    -  https://teslabs.com/articles/magnetometer-calibration/      
    -  https://www.best-microcontroller-projects.com/hmc5883l.html

'''
MField = 1000

[Keyaoli1995]

yes,
i just change my MField
in my location the Total Field is 48994.4nT, so the first value is 48994.4E-5 Gauss
the gain of my magnetor is 3.333(uT/LSB), so the second value is 3.333 * 100,
MField = 48994.4E-5 * 333 = 163.15
and use the Mfield value , my output data is like '-87.07969318 -122.47257324 -68.21476464' whose sum of squares is still greater than the local magnetic induction

[nliaudat]

:)

The out.txt is generated to graph the pyplot.
Nothing to do with usable data .

All you need is the hard & soft iron matrix

[Keyaoli1995]

I think the data in 'out.txt' file is generated from the data in 'mag3_raw.csv' file after compensation.

[nliaudat]

line 104 :

np.savetxt('out.txt', result, fmt='%f', delimiter=' ,')