5
\$\begingroup\$

I'm looking for a sensor that can detect large magnetic fields. Basically, I want a 1 when the field is greater than 75 Gauss, and a 0 when it's less than 75 Gauss.

I believed (EDIT: Wrongly!) Hall Effect sensors rely on moving through a magnetic field. Unfortunately, my application does not permit me to move through the field, and the field is not changing. It's since been pointed out that the Hall Effect does not rely on a change in the magnetic field. However, I still do not have any control over the orientation of the field.

Does anyone have any suggestions on how to detect such a magnetic field?

\$\endgroup\$
  • 1
    \$\begingroup\$ as i put, 2 sensors will detect a field in any direction. \$\endgroup\$ – Kortuk Nov 5 '10 at 17:03
5
\$\begingroup\$

The Allegro A1301 is a cheap analogue hall effect sensor. It's a 3 pin device: 5V GND and OUT. It outputs 2.5mV per Gauss detected.

It's very easy to use, here's a project I used it in.

\$\endgroup\$
  • \$\begingroup\$ Thanks! That's perfect, and will hook up just fine to an ADC input. It looks like Allegro has a nice collection of sensors...they even have a PWM one that might work well with a PIC. \$\endgroup\$ – ajs410 Nov 5 '10 at 19:47
  • \$\begingroup\$ FYI, I settled on the A1362; programmable sensitivity up to 16 mV/Gauss (I guess 75 Gauss isn't as large as I thought it would be!), and a filter pin to constrain bandwidth/noise \$\endgroup\$ – ajs410 Nov 5 '10 at 22:28
6
\$\begingroup\$

The Hall effect doesn't depend on variations in the magnetic field (i.e., motion relative to the field or time-varying field), as far as I know, it only depends on the strength of the field, and the orientation relative to the sensor.

Alternately you could look into 'magnetic reed switches'.

\$\endgroup\$
  • \$\begingroup\$ hall effect does not need motion. \$\endgroup\$ – Tim Williscroft Nov 4 '10 at 22:52
3
\$\begingroup\$

A hall effect sensor can detect large magnetic fields relatively easily. This is done by measuring the variance in resistance. You allow a DC current to pass, the Magnetic field will deflect electrons, increasing the DC resistance.

This gives you a magnetic field dependent resistor. These will be able to measure the current in 2 planes, the path of the current will be unaffected by a magnetic field in it's direction.

If you need to measure in 3 directions, I would suggest 2 to detect magnitude of field, and a general idea of direction. With 3 you can get a rather precise idea of direction.

There is no motion of the sensor required.

\$\endgroup\$
  • \$\begingroup\$ You're right, I will need more than one to calculate the field intensity. I may be able to make an assumption about how the field won't be oriented, so I may be able to do with two axes anyway. \$\endgroup\$ – ajs410 Nov 5 '10 at 19:45
3
\$\begingroup\$

Are you sure that you want it to both turn off and on at 75 Gauss? Magnetic fields are variable, and you might get a lot of noise at that edge. Most Hall effect switches are designed with Hysteresis, or a Schmitt Trigger, for this reason.

With this knowledge, you can start looking at Digikey's selection of magnetic sensors.

The Allegro A3295 is an example of such a device. It turns on at 75 Gauss, and turns off once the field decays to 5 Gauss.

\$\endgroup\$
  • \$\begingroup\$ You're right, hysteresis is good, but I think I actually want a linear hall effect sensor so I can do it in software. \$\endgroup\$ – ajs410 Nov 5 '10 at 19:43
-2
\$\begingroup\$

LOL 75 Gauss means never having to say your circuit failed; the electrons in there aren't going in any loop the magnet didn't pick first! Pick an optical sensor (e.g. a mineral anvil or mirror, or more easily a simple F/O loop) and remotely detect changes in light transmission and polarization to confirm the field characteristics. Are you making bets with http://www.magnet.fsu.edu/ , installing your first maglev train, or just in charge of field cleaning COTS naval artillery? I knew One Piece was popular, but not maglev trains....

\$\endgroup\$
  • \$\begingroup\$ Er what? Are you confusing Gauss with Tesla? 75 Tesla would be a lot. E.G. the Magnet Lab managed to get 100 Tesla which was an amazing achievement. Achieving 75 Gauss means buying a couple of fridge magnets. \$\endgroup\$ – Rocketmagnet Apr 13 '12 at 21:06

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.