1. Not finding help here? Sign up for a free 30min tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Chain in a tube

  1. Mar 16, 2008 #1


    User Avatar
    Science Advisor
    Homework Helper

    1. The problem statement, all variables and given/known data
    A uniform chain of length L and mass M is constrained to move in a frictionless tube. Initially a fraction (1-f0) of the chain rests in a horizontal section of the tube. The remaining fraction f rests in a section of the tube that is inclined downward from the horizontal at an angle theta. At t = 0, the chain is released and begins to slide down the tube. Find the time for the remainder of the chain to pass through the horizontal section of the tube.

    2. Relevant equations
    Lagrange equation

    3. The attempt at a solution
    We want to characterize the system completely by the parameter f.

    [tex] T = \frac{1}{2}M\dot{l}^2[/tex]
    where l = f * L
    [tex] T = \frac{1}{2}ML^2\dot{f}^2[/tex]

    [tex]V = -\int{mg dh}[/tex]
    where m = f * M
    [tex]V = -\int{mfg\sin{\theta}dl}[/tex]
    [tex]V = -\int{mfg\sin{\theta}Ldf}[/tex]
    [tex]V = \frac{-Mgsin{\theta}Lf^2}{2}[/tex]

    Note we take the horizontal part of the chain to be at a gravitational potential of 0, thus it does not contribute.

    [tex]L = T - V = \frac{1}{2}M\dot{f}^2L^2 + \frac{1}{2}Mg\sin{\theta}Lf^2[/tex]

    Then I can plug into the Lagrange equation and solve. Does this solution seem to be correct? If I am not mistaken this Lagrangian gives trigonometric functions as the solution, which seems somewhat odd for this problem.
    Last edited: Mar 16, 2008
  2. jcsd
  3. Mar 16, 2008 #2


    User Avatar
    Science Advisor
    Homework Helper

    … negative: trigonometric … positive: exponential

    Hi nicksauce! :smile:

    For negative parameter, it's trigonometric, for positive parameter it's exponential …

    which btw agrees with the T + V = 0 method (I take it you're not allowed to use that in this question?). :smile:
  4. Mar 6, 2011 #3
    Hi! I got the same equation for L. Solving for the equation of motion I get

    [tex] f(t)'' - f(t)*c^2 = 0 [/tex]

    which has as a solution

    [tex] f(t) = c_1 * e^{c*t} + c_2 * e^{-c*t} [/tex]

    where c is a constant. This doesn't seem right, since I have to solve f(t)=1, but I don't know c1 or c2. Help plz? :s
    Last edited: Mar 6, 2011
  5. Mar 7, 2011 #4


    User Avatar
    Science Advisor
    Homework Helper

    hi mathman44! :smile:

    (try using the X2 and X2 icons just above the Reply box :wink:)

    you can find c1 and c2 from the initial conditions :wink:

    (usually they're given as f(0) = f'(0) = 0)
  6. Mar 7, 2011 #5
    Thanks for the response tiny_tim!

    I get f(0) = f_initial = c1 + c2. Also from f'(0)=0 I get that c1 = c2. Hence c1 = f_ini/2 = c2. But that doesn't work, I'm still unable to solve for "t" in the equation f(t)=1, I'm getting complex solutions. Help please! (could my solution to the ODE be incorrect? Doesn't the solution have to be an exponential since f is positive at all times?).
    Last edited: Mar 7, 2011
  7. Mar 7, 2011 #6


    User Avatar
    Science Advisor
    Homework Helper

    hi mathman44! :smile:

    in this case the initial condition is f(0) = f0 :wink:

    (or is it 1 - f0 ? :confused:)
  8. Mar 7, 2011 #7
    Maybe this helps :s


    "f" is definitely the proportion of the chain hanging on the incline!
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?

Similar Discussions: Chain in a tube
  1. Particles in a tube (Replies: 20)

  2. Spring and Chain (Replies: 1)

  3. Spring and a Chain (Replies: 0)

  4. Venturi tube (Replies: 3)