Aurora Observatory

This project is something I've had on my mind for a while. Mike Szcys posted to Hackaday with 4 days to go to the deadline for the hackaday prize, convinced me to formalize the idea and begin work, now I have to work fast !

The prerequisite knowledge of the Aurora I already have due to my frequent photography expeditions. However the one area I'm lacking skills is magnetics / magnetometers, the last few days have been spent researching different types of magnetometers and the features and practicality for this application.

Many don't realize that most of the population in the world can see an Aurora 2-10 times during solar maximum, provided you know when and the best location to look from. Go further North (northern USA) and that becomes 50-100 times, We're around solar maximum right now. Yesterday there was another major storm (hit Purple (Extreme) on my aurora alerter), and it was seen as far south as Wisconsin, Michigan, Massachusetts, Idaho, Minnesota, South Dakota, Montana.

The variance in the earths field due to the Aurora is mainly in the Y direction and partially in the X, which means that the magnetometer needs to be able to monitor strength and direction of the field. This rules out many Magnetometer Designs and results in 2 "practical" possibilities for investigation (i.e. we don't want to be messing with super cooled magentics and caesium vapor type solutions).

( On a side note, Auroras are measured in strength by a K-Index (0-9): K-Index Primer )

Additionally, sensitivity is a concern, sub 1nT resolution is desired, and sub 10nT would be the limit for detection. This is because the quiet day curve (normal earth magnetic change during the day without an aurora) is around 50nT, and a K-Index of 2 (practical observation level for an aurora) is 10-20nT.

This puts the IC solutions out of the park. The most sensitive technology in IC's currently is AMR (Anisotropic Magnetoresistive) or GMR (Giant Magnetoresistive). The best state of the art commercially available sensor I could find is an AMR sensor with about a 200nT resolution (Honeywell HMC118X); this wouldn't detect anything but the strongest geomagnetic storms.

Considering that we want small size, low power usage, high resolution, high temperature stability and vector/scalar capability, this leaves essentially one choice a Fluxgate Magnetometer: Fluxgate Magnetometer

There is also a variation on the fluxgate which consists of one winding and the inductor is used to form a clamped oscillator where the frequency of oscillation is dependent on the earths field, this is called Magnetoinductive: New Magentic Field Sensing Technologies

Time to build some fluxgate magnetometer prototypes...