Finger Position System

Discussion in 'The Projects Forum' started by Mazaag, Aug 18, 2006.

  1. Mazaag

    Thread Starter Senior Member

    Oct 23, 2004
    Hi guys,

    I wanted to design a system that would be able to detect the movement on my finger. I only need it to be able to detect the "bending" of my finger ( joint in th emiddle of my finger ) as well as at my knuckle.

    I was thinking of having a rigid mechanical system where I could place my finger into , and have potentionmeters at the joints connected to a voltage divider circuit, which would therefore give different voltages for different "angles" and use that along with a lookup table of some sort to identify the fingers position.

    My question is whether or not I will get significant voltages changes when I alter the angle of rigid system, and what are the ways to get maximum change in voltage per angle change in the potentionmeter ( eg: high resistance values, larger voltage source , etc etc )

    I also wanted to know if there are other ways of detecting such a motion, note that I will be extending this idea for all 4 fingers.

    Any if there are any mechanical dudes out there , is there a way to keep the rigid system from " collapsing" at the joints but still be able to move it light force by my finger?

    Thanks guys
  2. n9352527

    AAC Fanatic!

    Oct 14, 2005
    How accurate do you need the system to be? I've seen a similar system that uses pressure rubber strips (change resistance with pressure, stretching or compressing). The advantage of that system are it is flexible and can be quite discreet (no rigid mechanical arms). However, I imagine that the accuracy probably not that great and there would be some problem with calibration as well.
  3. Mazaag

    Thread Starter Senior Member

    Oct 23, 2004
    I am intending to make this system as accurate as possible ( high resolution ) , since it will be used for mimicing such movement on a robotic arm.

    Now, would you be able to post links on such strips you mentioned? are they like strain gauges? , it would definetly help if i were to use that, since it will allow the structure of the system to be much more discrete and smaller in size.

  4. Chris Wright

    Well-Known Member

    Jul 26, 2006
  5. Mazaag

    Thread Starter Senior Member

    Oct 23, 2004
    I found this on a website about Strain Gauges..

    "The resistance of an electrically conductive material changes with dimensional changes that take place when the conductor is deformed elastically. When such a material is stretched, the conductors become longer and narrower, which causes an increase in resistance. A Wheatstone bridge then converts this change in resistance to an absolute voltage."

    Now, i'm not quite sure how a Wheatstone bridge works, but I believe you have to change the resistance of one of the 4 resistors (call it R3) to be able to determine the resistance value of the Strain Gauge Rx as a voltage reading.

    My question is as follows: The Strain Gauge will have a varying resistance depending on deformation, would I have to actively readjust R3 everytime the resistance changes to get a voltage reading? or how would that work ? Do I even need a Wheatstone Bridge ? or could I just use a Voltage Divider Circuit like i mentioned above, and measure teh voltage across either the SG or the other resistor in the divider , and use that value as a measure of the position of my finger.. ?

    Thanks guys
  6. microextremo

    New Member

    Aug 29, 2006
    Strain Gauges may work, if u can buid it, send a picture !!
  7. Søren

    Senior Member

    Sep 2, 2006

    No, the Strain Gauge is just like a variable resistor (electrically speaking).

    Strain Gauges only vary very little, so to get any kind of range you need the W. B.
    Another thing is, that you will still need a clever mechanical arrangement, since the S.G. can only be stretched a fraction - they're used on a piece of eg. metal bending as little as you can hardly see by eye, or the S.G. will be ruined.
    IMO, S.G's will be very hard to use for such a system, if you want your fingers to have your full travel, as opposed to just put a varying pressure on a static "finger rest".

    You would probably be better off with the (rubber) resistance strips mentioned and calibration should be easy, making the controller do the work... Long term precision is probably not very good, but you could have a calibration routine at the start of each session - nothing more than the controller telling you to stretch your finger all the way, press a button, bend the finger all the way and press the button once more (a couple of bends/stretches ahead will increase precision).