Building a fluxgate magnetometer
Magnetic materials, like iron and ferrite, limit how much magnetic flux can pass through them. Normally, they concentrate nearby magnetic fields, but at the limit — in saturation — they exclude them.
When a coil is wound around an iron rod, running a large current will saturate the rod and kick out any external magnetic field. Doing this with alternating current will chop the external field into a pseudo-alternating field, which can be detected using another coil:
A problem with this design is that the sense coil will also pick up the strong magnetic field of the driven coil, drowning out the signal. This is usually prevented using two magnetic cores driven in opposite directions so that the induced voltages cancel out, but simply putting the two coils at 90 degrees to each other also works:
I could never quite get the cancellation to work right in the parallel version, but the orthogonal design worked every time. The best results I got were using an orthogonal setup made from small (3.5 mm) binocular ferrites:
