New project: Calibrated Gaussmeter / EMF detector

I've seen several threads on DETECTING EMF with the Arduino, but I've been asked to create an EMF detector / Gaussmeter with an LCD display that will actually report results in Milligauss as opposed to a relative value compared to an uncalibrated baseline.

I've looked into several methods of doing this, with the use of a linear continuous hall effect sensor being the seemingly most accurate option.

The Allegro A1302 seems to be accurate enough, though at higher ranges. It will output 2.5V + 1.3mV / gauss given a 5V input. Given that I want to read down to as little as 0.01 MILLIGAUSS, it's gonna take a pretty damn serious ADC to begin to be accurate with it.

The device will need to be portable, but will usually be plugged in to power and stationary.

Does anyone have any other ideas for a project like this?

thanks!

Hall effect sensors are only useful at larger field strengths. If you want to work down that low you are going to have to implement a flux gate magnetometer design.

Oh dear. It appears that this might complicate things a bit too much.

While we don't mind working on the extra hardware, calibration is a problem. We want to ensure that the reported numbers are meaningful and accurate, but when designing new hardware that none of us have worked with, I'm not sure the confidence would be there, nor do we have the expertise in calibrating such instruments.

This is the unit closest to what we intended to create:
http://www.tenmars.com/cetacean/front/bin/ptdetail.phtml?Part=TM-192&Category=115311

We CAN Just buy these, it's simply a matter of them being limited in output / logging. While there is a connection there, the interface leaves a lot to be desired.

That instrument is an order of magnitude less sensitive than you want.

Yes calibration is tricky that is why these instruments cost a lot of money.

Understood on all points - I was pointing to that particular device because of the display and USB connectivity aspects, not the sensitivity. There are other units out there that use different detection methods (I assume), that will report in different ranges.

My boss now reports back that we're looking almost exclusively for sensitivity in the 1-10 milligauss range, with sensitivity outside that range being broadly unimportant. That would seem to relax our requirements quite a bit, though I'm not sure it relaxes them enough to make our task MUCH easier.

Not sure why you are being asked to make one. Is this an exercise for you or do you want to use one at the end of the day. If the latter it might be much better to buy one rather than pay you to make one, it will be much cheaper.
That relaxation in the sensitivity spec is going to make it easier to calibrate.
One way of calibrating it is to wind two coils.
One is given enough current to cancel out the Earth's field and the other is given enough current so you can calculate what field is generated.

You might want to give member dc42 a PM as he is a bit of an expert with magnetic fields.

We have a combination of "Hey, lets see what we can do with this extra time and money we have" and "Lots of people use the gear we're measuring. I bet if we could design a product to replace the hacked up solutions we've all been using, we could make a fortune"

I shall!

Thank you very much!

Hi,

I see this topic is now 9 years old.

I am playing with EMF and Arduino.

Did this go anyway in terms of successful calibration please? I’m guessing not as all I have now done extensive web searches and reading and all the Arduino based meters I can find are uncalibrated.

Seems to me best option is either to not bother or perhaps build a circuit (such as in the photo) with an Arduino replacing lm386 if this is possible....then using Arduino to drive lcd/serial monitor etc