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

Any mechanical device allowing free lateral shaft movement?

  1. Jun 26, 2012 #1
    Hello everyone!

    I am designing a sort of automatic spring gun to use with projectiles the size of normal tennis balls - sort of like a bowling machine.

    I have a horizontal spring, attached at one end to a square sheet of metal, and fixed at the other end. The square sheet of metal is supposed to hit the 'bullet' when the spring is freed after having been compressed. As a constraint to allow for movement in only one axis, there are four holes in the square sheet around the spring, and four fixed shafts to go in them so the sheet can only move forward or backwards.

    The only problem is, I need something which will connect the square sheet with the shafts in such a way so as to allow lateral movement of the square sheet caused by the spring, with minimum friction. A picture is attached:


    Are ball bearings alright? Is there anything less complex? Preferably I need something that I can mount onto the sheet. I'm a complete beginner when it comes to mechanical components...

    Any help would be appreciated.

    Attached Files:

  2. jcsd
  3. Jun 26, 2012 #2
    You don't need any component really, just no misalignment (ie good tolerances). A press in rubber/plastic bush would allow for some small misalignement caused by the spring action.
  4. Jun 26, 2012 #3
    The sheet would move really quickly once the spring is pulled back and released. Won't the high friction between it and the rods slow it down (which isn't what I want since it is supposed to be a 'gun')?

    I need some way to avoid this friction....
  5. Jun 26, 2012 #4


    User Avatar
    Science Advisor
    Gold Member

  6. Jun 26, 2012 #5
    Flange-mounted linear bearings! Just the thing! Thanks a lot!
  7. Jun 27, 2012 #6
    You could also use a single shaft inside the spring.
  8. Jun 27, 2012 #7
    A good fit, surface finish and grease. A bush is probably the best option, misalignment is taken by the stiffness of the rubber/plastic. A PTFE bush seems sensible for low friction.
    Bearings seem a bit overkill of this.

    The problem you are going to have isn't friction per se. It's misalignment (side loads).
    The 4 rod design is flawed because it's really hard to control the alignment of the plate relative to 4 long slender rods. When it's drawn back the plate tilts, giving a poor sliding action.

    Why not just make it a spring powered piston in a cylinder?
  9. Jun 27, 2012 #8
    I think you're probably right xxChrisxx. But won't a piston in a cylinder experience a similar tilt? Probably much less I'm assuming.

    How about making the holes in the sheet a little bigger than required to accommodate for misalignment? Or perhaps even bigger and fix a ring in each hole made of soft/slippery material like acetal, or similar?

    The thing is, while the piston idea seems to be reliable, I am reluctant to introduce a very big change unless it is wholly necessary...

    Coming back to the piston idea, the tilt could be solved easily by increasing the thickness of the piston, but what I'm worried about in this design is the constant wear and tear the piston might experience, especially if it is made of some soft, low-friction material.
    Last edited: Jun 27, 2012
  10. Jun 28, 2012 #9
    "Oilite" bushings would seem to be the most cost-effective method here. Unless we are talking HIGH-speed and VERY tight tolerances required. But for the system it looks like you have, I would suggest drilling bearing bores thru the top & bottom capture plates while they are clamped together to ensure maximum alignment, then simple oil-impregnated bearings (SAE 841) press fit in the plates will work.

    I've created similar systems, but with jack screws for transition.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook