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

[NEED SOME TIPS] Robot Arm Designing

  1. Aug 2, 2010 #1
    hi everyone,
    I'm a mechatronics student.
    I'm sorry i posted this question here. I hope i can get some tips/answers/designs(rough design)


    I'm having this Robot building project.
    And faced a designing problem,
    This design is passed to me by my senior...
    And the parts is fabricated out...
    The problem is the whole robot arm is not rigid, it tends to move when it's assembled.
    Is there any designs/ways to make the whole robot arm to be rigid ?
    (If you don't mind, provide your rough design to me) I will appreciate it alot.
  2. jcsd
  3. Aug 2, 2010 #2
    Are you clamping two flat faces to each other with screws and nuts?

    You could use some pins on one side that mate with precision holes on the other. These would resist motion in the x-z plane and the screws would be used only in the y direction. You could also use some type of locking washers or screws with locking threads to prevent them from loosening.

    Any of that help?
  4. Aug 2, 2010 #3
    Hmmm.... Do you mean using dowel pins ?

    Sorry, I din give u the more detailed drawing...
    This is another exploded view of the parts...
    Hope this will help u to see clearer.


    This is the exploded view... Is there any suggestions or other designs to make this whole arm to be rigid ?

    Actually i thought of another idea but i'm not sure whether it will work...
    Can you give comments on my idea ?
    This is the picture...
    My idea is by clamping the neck of the arm part...

    Exploded view:

    Assembled view:

    Thanks for your advice =D
  5. Aug 2, 2010 #4
    Yes, I meant dowel pins. Screws don't work well to align parts. There is almost always too much clearance involved. Precision dowel pins can be press fit into holes slightly smaller than they are. The pins can then mate with holes in the other piece that are only very slightly larger than they are. The holes must be machined with a high level of precision.

    If you have the resources you could also redesign those two flat plates in a way that they lock into each other. That would make for a more rigid coupling than just using pins, but would obviously take more time and money.
  6. Aug 2, 2010 #5
    Here's a useful site for projects like this:


    Even if you don't intend on buying anything from that site you can still download 3D models for many of the parts and fasteners listed.
  7. Aug 3, 2010 #6
    Thank you very much, Skrambles...
    It helps me alot...
    Thanks for the idea =D
  8. Aug 3, 2010 #7
    Btw will the clamping method be useful ??
    If u saw the exploded and assembled view....
  9. Aug 3, 2010 #8
    You posted the same picture twice, and I can't tell much about the holes from your pictures. I don't know if they are threaded, or smooth, or go through the part all the way. Can you put some fastener models into your assembly and exploded views to make it easier to figure out?
  10. Aug 4, 2010 #9
    Opps sry I posted the wrong picture...
    This is the new pictures i've generated for you...

    The whole arm assemble:
    ( The highlighted part is the part that we're dicussing about ) [Red Circle]

    For Magnified part view that I'm changing:

    The clamping method (Assembled) with screws:
    Screws: M3x20

    Clamping method (Exploded) with screws:

    Clamping method (Eploded) with screws + Arrow indicators:

    Red arrow indicator: Holes is threaded for M3 screw
    Orange arrow indicator : Through hole [ diameter 3.4 ]
    Yellow arrow indicator : M3x20 Screws
  11. Aug 4, 2010 #10
    Thanks for ur time....
    Sorry to take u so much time for helping me.
  12. Aug 5, 2010 #11
    It should work if the parts fit almost perfectly together. It seems like the design is using several parts where one would work. If you can use some CNC machine tools you could use more complicated geometry in a single part to mount the motors more rigidly.
  13. Aug 13, 2010 #12
    Check out the attached image.

    My assumption is that if you can lengthen the piece I have shown, it will allow for a little more clearance and remain rigid.

    Attached Files:

  14. Aug 17, 2010 #13
    oh thanks... that really helps =D Thanks for your advice =D
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook