Hall effect sensors

satchmo

Does anyone have any experience using Hall effect sensors? Any idea what sort of range they have, and how accurate a reading they can get of a magnetic field?

Chet T16

As with most things, it depends. Theres a million different options out there. How accurate do you need it?

Check out this video, the accuracy needed for levitation would be pretty high

https://www.youtube.com/watch?v=LaGv2FHS5zg

satchmo

Interesting video, thanks. I'm looking to use a hall effect sensor for measuring the distance between the sensor and a magnet - I need a maximum range of about 15-20cm or so. The higher the accuracy the better of course, but at the moment I just want to know if this sort of range is even realistic for a hall effect sensor, or would it need a massive magnet?

satchmo

Or I guess coming from the other direction, what sort of range could I expect for an accuracy of ~1mm? Does it depend on the strength of the magnet?

Chet T16

20cm would require a huge magnet and even then if you were wanting to measure close to the sensor it would probably become saturated as it came close.

1mm is quite an accuracy over those distances. What are you trying to do?

satchmo

Right, maybe hall effect sensors aren't the way to go then.

I'm trying to figure out a good way to accurately detect the movement of a fingertip by using sensors. RF, ultrasound and optical tracking are out because of occlusion (and because I don't want to limit the hand movement to a particular area), and because of the relatively small distances I thought I could use a hall effect sensor along with a magnet on the back of the hand - but I haven't been able to get any good idea of range vs accuracy.

What about using a series of sensors placed along the back of the finger each tripped in turn by electromagnet on the previous joint, so the maximum range would only have to be ~5cm? Even then, I suspect the size of the electromagnet would have to be too big (or unsafe?) to comfortably place on the finger. Of course then there's the problem that the sensor will rotate with respect to the magnet, so multiple orthogonal sensors would probably be needed, but that's a separate problem. For the moment I want to know if even a linear distance is possible.

As you may have gathered, I don't have a lot of electronics experience (I'm a software guy!), but I'm looking to learn more and this is a project that has been on my mind for a while.

InvisibleBadger

Microchip recently anounced some new chip that could detect nearby hand gestures via changes in electric field:
http://www.microchip.com/pagehandler/en-us/technology/gestic/technology/home.html

Chet T16

So you're talking about tracking the movement of the fingers relative to the hand, not a fixed point?

satchmo

Right. Plan B was electromagnetic tracking (like this) - I've seen it used for trackers in a few other places too, so I know it will at least work. I'm not sure where I'll get the coils, but I haven't looked into it much yet.