Simple Gauss Meter (Magnetometer)

Magnets are complicated. You are taught about them at school and you think you understand them but they're way more complicated than that. How "strong" is a magnet? How big is the force between a magnet and a paper clip? Or between two magnets? Is repulsion as big as attraction? (No!) I've got a box full of magnets - what do I know about them? I was playing with a hall sensor and decided to make a simple Gauss meter.

This meter will take you about 15 minutes to make. Making a nice case for it will take a lot longer.

The meter has a range of +/- 1000 Gauss. For comparison, the Earth's magnetic field where I live is 0.5 Gauss.

You probably already have most of what you need:

A linear Hall effect sensor: I used an A1301 it gives an output that varies linearly with the strength of the magnetic field. With a 5V supply and no magnetic field, its output is at 2.5V. The output varies between 0V (with -1000gauss) and 5V (with +1000gauss). The output current is up to 10mA.

When I look on eBay I see that the A3503 and SS41 are more common. The SS41 is "bipolar" - it gives a logical on/off signal so you don't want that one.

The A3503 will probably work. The data sheet says it's linear (but it's schematic shows a "bipolar" output - maybe a mistake). With a 5V supply, the output varies between 0V (with -1666gauss) and 5V (with +1666gauss). The output current is up to 1mA. That's what I think the data sheet says - I haven't tried it.

A Panel meter: A 1mA FSD meter would be ideal but whatever you happen to have will probably work as well. 

I had an old "VU meter" that is 0.5mA FSD. Be careful with "VU meters" - the movement might not be linear. Try it with, say, 1mA and then with 0.5mA; does the needle drop to half? I found with a different "VU meter" where it didn't. A nice new 1mA meter bought from eBay will be fine.

A 100uA meter will work but I find they are more delicate. A 10mA meter would be OK with the A1301 sensor but the A3503 can't provide enough current.

A resistor: choose a resistor so that the meter shows half-scale when there's no magnetic field. If it's a 5V supply and a 1mA meter then that's 5k. You may need to adjust the value slightly so maybe use a 4k7 with a few hundred ohms in series. (With my 0.5mA meter, I used a 10k resistor.)

9V PP3 battery and clip: that's the easiest way to power it.

A 7802 5V regulator: I'm assuming you'll use a PP3 battery. You could power it with 3 or 4 AA cells and have some way of adjusting the "zero" as the cells go flat.

A push-button: turn it on to make a measurement. Then the battery will last a long time.

The circuit is very simple. The 9V from the battery goes to the 7805 regulator input. The 5V output from the 7805 powers the A1301. The output from the A1301 goes though a resistor to the meter.

I printed a scale showing 1000-0-1000 gauss and glued it on to the meter. I labelled the "0mA" end of the scale "N" to mean that the sensor is near a North pole of a magnet. And I labelled the "FSD" end of the scale "S". That was with the curved face of the A1301 facing outwards towards the magnet. You could check how yours works with a compass.

People express magnetic fields in several different ways. Some useful numbers are:

• 1 Gauss = 100 uTesla
• 1 Gauss = 100,000 gamma
• 1 Oersted = 1 Gauss in a vacuum
• one ampere per metre = 4pi mOersted
• Earth's magnetic field = 25uT to 65uT
• Solar magnetic field = around 20nT
• fridge magnet = 100G
• Neodymium magnet = 1000G

Different kinds of sensors have different sensitivity. Of the ones you can build yourself, a Hall sensor can measure down to maybe 10,000nT, a Fluxgate sensor can measure maybe 1nT and a proton precession sensor can measure maybe 0.1nT. So a Fluxgate sensor is perhaps not quite sensitive enough for an archaeologist to do a geomagnetic survey.

It took far longer to design and 3D-print the case than to build the circuit. I quite enjoy CAD design work but printing, sanding and finishing things is much more annoying. Maybe I should buy a better quality printer.

The "probe" is the barrel of a ballpoint pen. The sensor is hot-glued into the end with the curved side outwards.

The STL files and source files are here. I don't imagine they'll be much use to you as they're specific to the meter I used.