Using finger hall sensors to control Arduino & leds. Has it been done?

[Drae]

Hey all

In an effort to avoid unnecessarily reinventing the wheel, are there any projects out there which use hall sensors & magnetic strip attached to a glove to communicate with the arduino?

Tried a google search for various terms, but sadly the search function on here is sadly offline.

- How it works:
I'm planning to use 4 hall sensors per hand, with a magnetic strip attached to the palm.

The right hand are the 'option groups'
The left hand sets the value for that group. To switch all leds to strobe for example, you hold finger 3 of the right hand against your palm (R3), then hold finger 3 (L3) of your left palm. To quickly restore to default, hold R3 then all left fingers (L*).

R1 = Colour control
 L1/L2/L3/L4/L* = Red, green, blue, white, all-fingers = default colour mixes.
R2 = Speed/fade control
L1/L2/L3 = slow, medium, fast
R3 = Mode
L1/L2/L3/L4/L* = torch, multi-fade, strobe, flash, reset to default
R4 = reserved

I'd appreciate any links or hints to existing projects and stuff. If not, time to get soldering and coding!

[Erdin]

No google search ? How can you develop an Arduino sketch without it ?
http://lmgtfy.com/?q=magnet+hall+fingertip+palm+glove

This is with a magnet on the thumb:
http://www.sonami.net/LS-reviews/rev_EM6_98.htm

It is pretty straightforward, so it should not be hard.
Detecting the proper finger and detecting a double 'click' would need some programming skills, but you can always ask on this forum.

[Drae]

Thanks for the link Erdin, though the lmgtfy link really wasn't necessary ;p

I was referring to having already tried various google search terms for completed low-profile projects and libraries, rather than available research/theory. The Ardu forum search was offline.

Most of the available projects out there seem to rely on USB or are very bulky. I'm hoping for one built into thin spandex gloves, for example.

Edit, there's one interesting project funded two years ago: http://www.kickstarter.com/projects/jrowberg/keyglove-wearable-input-device - Seems to rely on electrical contact than hall sensors (which can't be worn under another glove)



Indeed, the programming won't be too hard, I'll manage it fine. Lots of testing and debugging.  Looks like it's time to cook up a new library.

Seems I'll need an ardu mega too, for the 8 analog inputs.

[Erdin]

Sorry for the misunderstanding.

The code is very specific. I don't think there is a library or example sketch for this.

The Keyglove uses contacts and conductive wires. The contacts could get greasy and it won't work under a normal glove as you wrote.

Some hall sensors are switches, some are analog. I guess you want to use the analog hall sensors.
You can place the sensors under the fingertip, but also on the other side on top of the nail. I think in your sketch you can sense that just as well.
It is even possible to use a flat coil under the fingertip instead of a hall sensor.

Some Arduino Pro Mini clones have 8 analog inputs.
The Leonardo has 12 analog inputs.

Do you know what kind of wires you want to use ? I want to buy conductive carbon flexible wire, but I haven't found what I want yet.

[Grumpy_Mike]

Be aware there are two types of digital hall effect switch bipolar and unipolar.
One switches in the presence / absence of a magnetic field.
The other switches one way when it sees a north pole and the other way when it sees a south pole. This is not the type you want although it is more common. Read the data sheet carefully for the one you choose.
I used some very small ones in this project:-
http://www.thebox.myzen.co.uk/Hardware/Dice_Game.html 

[Drae]

@ Erdin: Thanks for the info, especially on the Pro Mini & 12 port Leonardo, that might just be what I'm looking for! You're right about the hall sensor above the fingernail, these are sensitive enough for a~13-20cm detection distance, dependng on the magnet strength (or further with analog port oversampling). I just ordered some magnetic fridge-tape, I'm hoping the flexibility and magnetic strength is significant enough to pick up. Otherwise the coil idea may work better.

I just checked the project box I bought to house the pcb & arduino, a mega would need one of the screw risers to be chopped off to fit, but can work if push comes to shove.

The wire I'll be using is just some 1mm stranded copper. I'll try figure out an efficient gnd/+5v routing plan so there's not too many wires going along my arm.

As for hall sensors, you both raise good points, I have a handful of A1302 analog ratiometric sensors, so I can make a calibration routine to determine initial tolerances for each sensor, to help prevent false positives/negitives and such.

And interesting project there by the way, mike!