# Building a Fluxgate Magnetometer - Part 1 (and NASA)

Today I want to discuss the first steps in building a simple fluxgate magnetometer for a classroom demonstrator.  Originally this post was going to be a wrap up of NASA work and the magnetometer would come later, but I'm still waiting on my presentation to clear export control so I can post it.  As soon as it does, I'll put it up along with a short article.

This semester I'll be the TA for 'Global Geophysics', mostly doing lab instruction/writing.  After some thought I decided that students need more hands-on classroom geophysics, which is difficult to do.  By its nature geophysics is an outdoor activity with normally expensive instruments.  The instruments are often viewed as a mysterious black box that spits out numbers used to make a map.  This must change.  With a proper understanding of the instruments students will better understand errors in the data, how to troubleshoot in the field, and know why certain hardware limits exist.

The concept of a fluxgate magnetometer is pretty simple.  Rather than go into detail I'll refer you to this wikipedia article.  This is mainly to chronicle the construction so others can reproduce this (assuming we get a working model).  My design came from a physics lab at Brown University.  The instructions were vague in parts and I'll be taking some liberties as we go along.  This first article will cover construction of the coil and the driver circuit.

The fluxgate coil consists of a driver coil surrounding a soft steel wire, and a secondary coil to pickup signal surrounding the primary coil.  First I took 16ga annealed steel wire from Lowes and cut it to about 1m long, cleaned it, and made it as straight as possible.  Afterwards I wrapped close to 2000 turns of 22ga magnet wire (Radio Shack #278-1345) tightly along its length.  This was then bent in half making a 'U' and that was wrapped with close to 1000 turns of 26ga magnet wire. I used large wire because it will be more durable and I used different gauge wire since the enamel insulation was a different color allowing students to easily see the windings.

That's all there is to the coil.  To increase durability I will probably clear coat the coil and place it into a small acrylic tube so its difficult to bend or break.  The next step is to build a driver for the primary coil.  The Brown lab used a function generator.  Currently I don't have one, nor have I found a suitable cheap unit.  This meant improvising, and luckily Velleman makes a signal generator kit that is just about right.  It operates at 1kHz (the desired frequency for this project) and produces sine, square, triangle, and integrator waves.  The kit was pretty easy to build in just about an hour and works well as seen by the oscilloscope output below, but frequency stability is not phenomenal (especially when then unit is cold).

Next a few amplifiers need to be designed and built.  The signal generator kit cannot pull the load of the coil, so a simple +/- 9V system will probably do.  The output will also need some kind of amplification.  The lab I found also uses a bandpass filter.  Once the amplifiers are working it will be time to decide if this is necessary and if I want to use an oscilloscope and hardware filters, or an ADC and display the waveform on a computer projector using software filters.