multi-solenoid/electromagnetic muscle

Discussion in 'The Projects Forum' started by agentid36, Jan 18, 2010.

  1. agentid36

    Thread Starter New Member

    Jan 18, 2010
    2
    0
    I have an interesting project on my hands, and I need some input.

    I need an electrically powered, moderately high precision linear motor with roughly 8-11 inch travel distance with a low vertical+horizontal (10-15mm width and height would be optimal) profile (length of 8-11 inches). Cannot be pneumatic or hydraulic. Should be as low weight as possible. This is the best thing I can think up for my purposes:

    Effectively, what I'm trying to make is an electromagnetic muscle. Basically, a bunch of solenoids in a row, controlled by a microcontroller that then goes to either a transistor, or to the solenoids themselves, obviously depending on resistance/voltage and whatnot.

    So.
    Thin plastic sleeve to hold the steel plunger (or ball bearing or pill), lessen friction and hopefully mitigate sticking. (would a rare-earth magnet work better?)
    I would presume that a long, connected metal tube would be a bad idea, so would a bunch of small individual, separated steel tubes work better? (Will leakage inductance be an issue here?)
    Should I use 'metal washers' connected to the metal tube to decrease the reluctance? Will it make induction leakage worse? I'll probably separate each metal washer with a (thinner) plastic one.
    Each solenoid coil will probably be just 1 wire thickness wide. I want high precision here. 120g of force or higher would be optimal, but I highly doubt I'll even get that.
    For simplicity's sake, we can just say there are only 10 or 20 solenoids in the array for now.

    Does anyone know what might be the best way to control the plunger with the magnetic field? Drag, hold, push (if using a magnetic plunger), etc? Could even implement dragging and pushing together, etc.

    Any recommendation for the microcontroller and (if necessary) the transistors?

    Any recommendation as to the shape of the plunger? http://robots.freehostia.com/Solenoids/SolenoidsBody.html
    gives a good description of plunger shapes, but my circumstances aren't really the same, so I don't think that their information will apply 1:1.

    The solenoid array will be about 11 inches long, and what it will be driving will only be moving about a couple inches, so some mechanical advantage stuff can come into play, as long as it doesn't take up too much room. So. Any comments? Ideas?
    Has this ever existed before? It seems to me that someone must have thought of it before and either found some fatal flaw or something. I've looked around for a month now and found nothing like it.
     
    Last edited: Jan 18, 2010
  2. rjenkins

    AAC Fanatic!

    Nov 6, 2005
    1,015
    69
    Solenoids are very inefficient, I'd try other forms of actuator.

    The simplest would be a leadscrew, such as a piece of threaded rod with one end attached to a geared motor shaft and the nut being pulled along it.

    Another form of 'muscle' uses a highly geared motor, plus a tubular assembly of fibres or wires that is twisted to generate a pulling force.
    Imagine eg. six strands attached around the circumference of a disc at either end, and spacers at regular intervals from one end to the other.

    When one end disc is twisted, the whole thing winds up and gets shorter.

    You can get tiny motors with inline gearboxes, down to under 10 rpm at the output. With either the screw or spiral setup, you can generate considerable force with very fine control and low power consumption.
     
  3. retched

    AAC Fanatic!

    Dec 5, 2009
    5,201
    312
    How about muscle wire? It contracts with power change. More power shorter wire. The wire that lifts 150 grams is less than $20.


    Here: 100 micron wire will lift 150g its a 1 meter length. You can cut your own lengths. $18.95. The part number for the 150g is 32MT100LT

    http://elexp.com/kit_2mt3.htm
     
  4. agentid36

    Thread Starter New Member

    Jan 18, 2010
    2
    0
    thanks rjenkins and retched, I'm gonna take a look into the tubular assembly of wires and the muscle wire. thanks a lot.
     
Loading...