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!

Use geometry -Redshift derivation?

  1. Jan 11, 2013 #1
    "use geometry"--Redshift derivation?

    1. Use geometry to derive z=v/c where c is the speed of light and z = [v(obs)-v(em)]/v(em).
    2. Relevant equations

    None given... Though I am assuming that I am constant.

    3. The attempt at a solution

    I believe it has to do with the Doppler effect which gives Δf = (Δv/c)f. I'm not entirely sure what it means by geometry though; am I looking at the geometry of the changes of light, cause that seems super bizarre to me.

    I feel so close to getting it but I feel like I'm missing something.
     
  2. jcsd
  3. Jan 11, 2013 #2
    Re: "use geometry"--Redshift derivation?

    What are v(obs) and v(em)? What about v? You cannot derive a formula without knowing what the different symbols mean
     
  4. Jan 11, 2013 #3
    Re: "use geometry"--Redshift derivation?

    v-observed and v-emitted, I'm assuming. It doesn't clarify much more than that
     
  5. Jan 11, 2013 #4

    cepheid

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Re: "use geometry"--Redshift derivation?

    First of all, you have to be careful with your symbols. Physicists often use the Greek letter "nu" (##\nu##) for frequency, rather than f. This is not the same thing as ##v##, the velocity of the source of emission. Secondly, your equation for redshift, in terms of frequency, should be ##z = (\nu_{em} - \nu_{obs})/\nu_{obs}##, as opposed to what you have there.

    As far as solving the problem, using "geometry" just involves thinking about the source of emission receding from the observer, and considering the observed vs. emitted frequency that results. To do this, consider two successive wavefronts of the EM wave. After all, the frequency is just how often wavefronts pass by. Have the first wavefront be emitted at t = 0, when the source is a distance d from the observer. Then the second wavefront is emitted at a time t = Δtem later, this being the period between wavefronts, (which is the reciprocal of the emitted frequency). Note that at this later time, the source is now farther from the observer, so the second wavefront has a longer distance to travel than the first one did. At what time t1 does the first wavefront arrive at the obsever? At what time t2 does the second one arrive? What is the difference Δtobs between them? How does this differ from Δtem? So what is the frequency shift?
     
  6. Jan 12, 2013 #5
    Re: "use geometry"--Redshift derivation?

    Thanks; I'll try to be more careful... I'm guessing I should be able to find it on the symbols haha.

    Anyway, here's a picture I drew (sorry, I'm on a tablet so it's kind of messy)


    So I'm guessing here that I can see that λ(obs)= λ(em) + x, and use that to figure out that z=x/cΔt = [itex]\nu[/itex] /c.

    Is there significance in the fact that the it's an isosceles triangle?
     

    Attached Files:

  7. Jan 12, 2013 #6

    cepheid

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Re: "use geometry"--Redshift derivation?

    I don't think you need a spacetime diagram.

    Did you try the method I suggested using wavefront arrival times?

    What is 'x' in your previous post ?
     
  8. Jan 12, 2013 #7
    Re: "use geometry"--Redshift derivation?

    Here's my attempt at what you described, but it's got me all mixed up again.

    I think t2 = t1+ some tΔx
    Meaning that Δtobs=t2-t1 = t1 + tΔx - t1 = tΔx

    Δtem=tem

    So if I replace z=[[itex]\nu[/itex]em-[itex]\nu[/itex]obs)/[itex]\nu[/itex]obs, we wind up with

    z=[itex]\nu[/itex]em/[itex]\nu[/itex][itex]\Delta[/itex]x

    and so [itex]\nu[/itex]=[itex]\lambda[/itex]/v?

    Is this the correct line of thinking so far?
     
  9. Jan 13, 2013 #8

    cepheid

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Re: "use geometry"--Redshift derivation?

    You're thinking along the right lines, you're just not quite getting the details. Say t = 0 is when wavefront 1 is emitted. Also, t2 is when wavefront 2 arrives. t1 is when wavefront 1 arrives. I'm assuming that Δt_x is the travel time required for wavefront 2. Therefore t2 is not t1 + Δt_x, because that would mean that wavefront 2 wasn't emitted until after wavefront 1 arrived. Instead, t2 = Δt_em + Δt_x, since Δt_em is the time interval between the emission of the two wavefronts.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook