1. Limited time only! Sign up for a free 30min personal 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!

Homework Help: Timelike Geodesic and Christoffel Symbols

  1. May 18, 2010 #1
    1. The problem statement, all variables and given/known data

    How do I show the following metric have time-like geodesics, if [tex]\theta[/tex] and [tex]R[/tex] are constants

    [tex]ds^{2} = R^{2} (-dt^{2} + (cosh(t))^{2} d\theta^{2}) [/tex]

    2. Relevant equations

    [tex]v^{a}v_{a} = -1[/tex] for time-like geodesic, where [tex]v^{a}[/tex] is the tangent vector along the curve

    3. The attempt at a solution

    First, I write it as the Lagrangian

    [tex] L = -R^{2}\dot{t}^{2} + (cosh(t))^{2} \dot{\theta}^{2} = -R^{2}\dot{t}^{2} [/tex]

    as [tex]\theta[/tex] is a constant.

    How do I proceed to show that this indeed gives us a time-like geodesic.

    Could someone also tell me if I have computed the Christoffel symbol components correctly? My result is

    [tex] \Gamma^{t}_{\theta \theta} = 0-sinh(t) \times cosh(t) [/tex]

    [tex] \Gamma^{\theta}_{t \theta} = tanh(t) [/tex]

    and all other components vanish.

    Cheers!

    P.S. How do I type minus sign? It doesn't seem to work if I have left the 0 out at above.
     
    Last edited: May 18, 2010
  2. jcsd
  3. May 18, 2010 #2
    Maybe I'm not understanding the problem correctly, but if you're holding [itex]\theta[/itex] constant, couldn't you just use [itex]d\theta = 0[/itex]?
     
  4. May 18, 2010 #3

    diazona

    User Avatar
    Homework Helper

    [tex] \Gamma^{t}_{\theta \theta} = -sinh(t) \times cosh(t) [/tex]
    Works fine for me :confused:
    Anyway, I get the same thing as you for the Christoffel symbols except without the minus sign. I could have made a mistake but you might want to recheck your calculations.
     
  5. May 20, 2010 #4

    gabbagabbahey

    User Avatar
    Homework Helper
    Gold Member

    I don't see the need to compute the Christoffel coefficients at all. The solution to the Euler-Lagrange equations will be a geodesic, so if [itex]R[/itex] and [itex]\theta[/itex] are constants, you want to solve

    [tex]\frac{\partial L}{\partial t}-\frac{d}{d\tau}\frac{\partial L}{\partial \dot{t}}=0[/tex]

    (Where I'm using [itex]\tau[/itex] to represent your affine parameter...i.e. [tex]\dot{t}=\frac{dt}{d\tau}[/tex] )
     
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook