Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

2DOF Arms

  1. Apr 6, 2005 #1

    I have a question regarding creating a "mechanical" arm which would have the ability to move freely in 2DOF ( ie. lift up and back down , kind of like our human arms, but when it goes down, it can go down a lot further than we can ).
    I saw bots like ASIMO use those umm... looks like a round thingie, what are those and how do they rotate them??? I dun see any motors near them.

    -Thanks and I'm just curious...
  2. jcsd
  3. Apr 6, 2005 #2


    User Avatar
    Science Advisor
    Gold Member

    The just look like pin (hinge) joints to me, encased in a kind of turret which allows rotational movement. They look like they're probably actuated just by a couple of motors, with a pinion/spur kind of arrangement. One motor would extend the upper arm, another would rotate the turret. Probably a lot easier to engineer than a ball and socket joint (as we have).
  4. Apr 6, 2005 #3
    Would you like to show me some graphical representations, illustrations of your socket joints? and how to arrange motors inside so that it can rotate the joints?

  5. Apr 6, 2005 #4


    User Avatar
    Science Advisor
    Gold Member

    When I said "ball and socket joints (as we have)" I meant we as humans!

    Just think of your shoulder :-)
  6. Apr 7, 2005 #5
    how do I attach motors on to it so that I can move the joint in the 2DOF?
  7. Apr 7, 2005 #6


    User Avatar
    Science Advisor
    Gold Member

    Prudens, as I suggested, you could have one motor with a pinion driving a gear to raise and lower the arm, and another motor with a pinion driving a gear to rotate the turret containing the pin (hinge) joint. You've seen ASIMO, you can picture it.
  8. Apr 7, 2005 #7


    User Avatar
    Science Advisor

    Your terminology is off - a DOF is the abilty to rotate and both directions are covered by it.

    So up and down on one joint is 1 DOF.

    And up and down with a second joint that allows left and right is 2DOF.

    There are other combinations of joints to achieve different levels of articulation but a ball and socket offers more like 3DOF in one joint but now how do you move such an arrangement? A hexapod comes to mind.

    For moving an arm like you're describing, a very optimal solution is the worm gear arrangement since it offers great mechanical advantage and when the electric motor is not spinning the arm will be locked automatically.
  9. Apr 7, 2005 #8
    I am just a curious being... I am not engineering student. So Please bear with me,

    How do you set up a worm gear correctly? Is there graphical tutorials?
  10. Apr 7, 2005 #9
    The small motors that controls ASIMO aren't really motors at all
    they are powerful servos that very accurately control the rotation of the shaft.
    they also have limits to prevent the them from fully rotating
  11. Apr 7, 2005 #10
    how do you determine whether one is a powerful motor/servo? what specifications do you look for?

    Also, I wonder where do they situate the motor/servos in ASIMO? What kind of gear train would they use?
  12. Apr 8, 2005 #11


    User Avatar
    Science Advisor

    Not really a point of distinction. A servo is an electric motor with a feedback mechanism. You can take pretty much any DC electric motor and attach a feedback mechanism and electronics to make it a servo.
  13. Apr 8, 2005 #12
    You guys/gals are great,

    How do you calculate the amount of force needed by the motor to lift the arm up? Suppose the arm has mass of M and the object the arm is holding is mass m. And what specification in a motor/servo do you look for to see what is its max power capable of output?
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?

Similar Discussions: 2DOF Arms
  1. Robot Arm (Replies: 2)

  2. Robotic arm (Replies: 2)

  3. Robotic arm control (Replies: 0)