Precise magnetic sensor

Hi everyone, recently I am working on a project recently that require magnetic sensors.

These sensors need to detect 3mmx1mm neodymium magnets which will be placed about 5 centimeters above the sensor, but these sensors will be placed only about 1 cm apart from each other because I need to know exactly where the magnet is.

So they have to detect the magnets either only when it’s directly above it or the reading difference between sensors has to be different enough for me to tell where the magnet is. Do such sensor exist? Sorry I am not familiar with any kind of magnetic sensors.

Also, I need 72 sensors for the project,so it will be good if these sensors cost around $1 each.

Thanks!

Hall effect sensors will detect a nearby magnet, but there are many different types, with different sensitivities, and the sensitivity will depend on the precise relative orientation of the sensor and magnet.

Unless you know the details of the magnetic field strength surrounding the magnets, you will have to experiment with a number of sensors.

I need to know exactly where the magnet is.

Define "exactly". I doubt you will be able to locate the magnet to within 2 cm, with the arrangement you describe.

jremington:
Define "exactly". I doubt you will be able to locate the magnet to within 2 cm, with the arrangement you describe.

The magnets will only move on a circular path, where the path will be covered with a magnetic sensor every 1cm.
So I only need to know if the magnet is above a particular sensor, that is "exactly" for me. 1cm accuracy is enough for me.

you have

  1. mechanical(passive) magnetic sensor, aka reed sensor

https://www.aliexpress.com/wholesale?catId=0&initiative_id=SB_20200104213553&SearchText=reed+sensor

  1. hall effect sensor A3144 ( or similar)

https://www.aliexpress.com/wholesale?catId=0&initiative_id=AS_20200104213555&SearchText=a3144

  1. ultra low power hall effect sensor SL353LT ( for battery apps )

well they are normally rated in Gauss ( or mT ), the larger the number the lower the sensitivity

i have couple of A3144s with me rated 20 mT , the one deployed in flow sensor like this

https://www.aliexpress.com/wholesale?catId=0&initiative_id=SB_20200104214123&origin=y&SearchText=flow+sensor

Thanks for the advice but reed switches are way too dull and I think I could only use ones with analog output since the distance is too far for a3144 or similar digital output ones to trigger(I saw some vids about them).

I just found someone in my country selling AH3503 for 10 cents each. I think that might work. Anyone have experience with it?

Wish-Lin:
Thanks for the advice but reed switches are way too dull and I think I could only use ones with analog output since the distance is too far for a3144 or similar digital output ones to trigger(I saw some vids about them).

I just found someone in my country selling AH3503 for 10 cents each. I think that might work. Anyone have experience with it?

you might want to take a look at this

Thanks! I think AH3503 is the current way to go!

Wish-Lin:
The magnets will only move on a circular path, where the path will be covered with a magnetic sensor every 1cm.
So I only need to know if the magnet is above a particular sensor, that is "exactly" for me. 1cm accuracy is enough for me.

That totally doesn't make sense. A 1 cm error in detecting magnets 1 cm apart means you're under the next magnet.

Do post images of what you have in mind. It's going to be really hard if not simply impossible to detect where a magnet is when they're 1 cm apart from one another, and five times as much from the sensor. Or maybe I misunderstood something.

Also wondering why you need that ridiculous number of sensors (too many even for a Mega to handle, especially if you're going to have sensors with analog output).

wvmarle:
That totally doesn't make sense. A 1 cm error in detecting magnets 1 cm apart means you're under the next magnet.

Sorry. I mean that the precision of 1cm-aparted sensors is enough for me.

wvmarle:
Do post images of what you have in mind. It's going to be really hard if not simply impossible to detect where a magnet is when they're 1 cm apart from one another, and five times as much from the sensor. Or maybe I misunderstood something.

Also wondering why you need that ridiculous number of sensors (too many even for a Mega to handle, especially if you're going to have sensors with analog output).

Speaking of the sensor, I just found that I only need 24 for it to work.

The magnets will be attached on a big Rubiks cube(18cm) so that on each face's 4 edges, there will be a magnet facing out. Then, I will arrange the sensors on 6 ring tracks about 5 cm from each face. Each face will have 12(I found 4 can work just as well) sensors arranged like the numbers of the clock so that when the Rubiks cube turns, I can detect the magnet's movement thus detecting the turns.

mmm... Some image would be very useful here, I'm sure.

Considering the limited degrees of freedom a rubik's cube has, you may be able to reduce that number of sensors even more.