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

Light reflection in length contracted system

  1. Nov 13, 2014 #1
    Light beam hits the mirror perpendicularly and returns to the source (left picture).

    The same system in horizontal motion (right picture) is skewed due to length contraction so that both beam and mirror are at higher angle. Because the beam returns to source in the system's rest frame, it must do that in any other frame. So the beam returns the same path that is drawn with solid red line, although it might be expected that the light is reflected to direction pointed by thin dotted red line.

    Contrastingly, let's have a system that has the shape of right picture in its own rest frame. In this case, the light reflects so that it follows the thin dotted line. I find this distinction strange, because light reflection should depend only on the geometry. The geometry is exactly the same in both cases.

  2. jcsd
  3. Nov 13, 2014 #2


    User Avatar
    Science Advisor
    Gold Member

    But relative to this second frame the source is moving, so the beam doesn't return on the same path.
  4. Nov 13, 2014 #3
    It's not the same path in space, but I meant that relative to the source the light travels at certain angle to certain distance, reflects and returns in the same angle. The whole system (source and mirror) are moving.
  5. Nov 14, 2014 #4


    User Avatar
    Science Advisor
    Gold Member

    If you want to see what a scenario looks like in a frame moving with respect to your defining frame, you need to use the Lorentz Transformation process, you can't just guess and hope to get it right.

    Also, although for your defining scenario, you can get away with a single diagram, because our minds can image a flash of light (you don't want to use a beam or it will be impossible to depict for the moving frame) propagating from the source to the mirror and back, you need to make a series of diagrams at equal increments of time, like the frames of a movie. This means that you can't just take all the events in the defining scenario at one particular time and transform them to the moving frame and show them on a diagram as a snapshot, instead, you have to find a set of events that all occur at the same time in each new successive transformed frame. It's kind of a complicated process but I have done it for you.

    To start with, I've taken your image of the frame where the apparatus is stationary, assigned coordinates to the light source, the ends of the mirror and the reflection point on the mirror. Then I made five snapshots showing the progress of the red light flash from its source to the mirror and back:






    I'm not going to show you all the transformation processes are the values of the events. Instead, I'm just going to show you an image with 5 diagrams that I have taken from your frame where the apparatus is moving. I hope this make sense to you:


    To interpret this correctly, you must imagine just one of the five mirrors and one of the five red flashes as occurring at the same time. For example, it looks like the second red flash is almost hitting the first mirror but that is not the case. It's the second contracted dot at the top (the light source), the second red dot which is part way down, and the second mirror which all go together. Once you figure it out, you can see how the red flash propagates in a straight line towards the center of the mirror and then propagates back up again in a straight line to its source.

    Admittedly, an animation would be a lot better but I hope this suffices.

    You can't just take the image of a moving apparatus that is length contracted, not to mention having relativity of simultaneity applied to it, and "construct" an "equivalent" apparatus at rest and think that it should behave in any particular manner.
  6. Nov 14, 2014 #5
    Thanks George for your thorough reply.

    Agreed. We don't even need continuous beam, because the angle is of interest. In fact, we could use some object e.g. ball that bounces back from wall and its speed doesn't need to be "relativistic" at all. I just wanted to use light and mirror for simplicity and familiarity.

    I think I got it. Here is the same idea drawn with 5 mirrors, which represent the system in different instances of time (looked from outside, the system is moving and length contracted).

    Why not? If we accept that the contracted object "really" has the shape it's calculated to have, we can visualize it and in principle even take a picture of it. Or construct an object at rest that has the same shape and size.

    Right, at least in this example the result doesn't seem to be equivalent. When we look the moving system from outside, it seems to me that the pulse reflection follows the shape that the object has in its own rest frame instead of the shape that it has in the frame where we are looking from. If all frames are equivalent, why rest frame should be preferred?
  7. Nov 15, 2014 #6


    User Avatar
    Science Advisor
    Gold Member

    You didn't draw the diagrams correctly. This one shows the path of the light flash in the frame in which the apparatus is moving:


    Keep in mind, that in none of these diagrams can an observer or camera see what we have drawn. Lengths of objects are established by a convention, not just by observation or measurement or appeal to reality. I had to do a lot of work to determine the correct positioning of the different "snapshots" of the apparatus, mainly to determine the simultaneity of their events.

    I challenge you to determine the coordinates of the events for the endpoints and midpoints of the snapshots of the mirrors and the coordinates of the light source and the light flash in the above diagram. Maybe then you will understand why you cannot claim that the shape of the apparatus that we see on a diagram is related to reality. Both the rest frame of the apparatus and the frame in which it is moving are equally valid in describing its shape but that shape is established by the frame we use to define our coordinates, not be a physical reality. Coordinates are not a physical reality.
  8. Nov 15, 2014 #7
    In principle (but hardly in practice) a CCD camera with extreme specifications that is set up very near to the mirror certainly should be able "see" an image like you have drawn, for the simple reason that the camera will be designed to interpret data coming from different pixels in accordance with the simultaneity convention. In such extreme cases, the picture produced by a camera thus includes interpretation. Regretfully I only hinted at this in an earlier thread:

    Also, nice drawings! :)
    Last edited: Nov 15, 2014
  9. Nov 15, 2014 #8
    George showed you how to construct correctly using the Lorentz transformations. However it's not needed to calculate the events. Your first drawing is erroneous because you used a rule of optics that is valid for non-moving mirrors only. In order to find the correct angle of reflection in an intuitive way, you can construct the angle of reflection with Huygens' method* - and you will find that it works out perfectly. Of course, you need to account for length contraction in the way you already did in your first drawings, as well as with motion of the mirror as explained by George. You only need a drafting compass and a ruler. :)

    * if you don't know it, see for an explanation https://en.wikipedia.org/wiki/Huygens–Fresnel_principle
    Last edited: Nov 15, 2014
  10. Nov 16, 2014 #9
    Now I hopefully get it. You mean the light flash actual path in this frame. I tried to draw only how pulse moves "relative" to the apparatus in this frame.

    I'm somewhat familiar with relativity of simultaneity and aware that if there are clocks in different parts of apparatus and the clocks are synced in the rest frame, they are not in sync in a frame where apparatus is moving. For example, if we set apparatus length in x-dimension 1 units in its own rest frame, and look in a frame where apparatus is moving 0.9c to right, the time difference between front and rear is 0.9 units. So if we take a picture e.g. using a camera that is wide enough to cover the whole apparatus that is moving, the different parts of the picture would have clocks showing different time. So the moving object is not "concrete" or "touchable" in the sense that objects we handle usually are.
  11. Nov 16, 2014 #10
    The intuitive understanding is exactly what I'm looking for. I'm convinced enough that the pulse returns to the source in every frame, because otherwise there would be a contradiction, but I'm just unable to see why it would reflect into this kind of surprising direction in a frame where the apparatus is moving.
  12. Nov 22, 2014 #11
    Here is a paper that illustrates the construction method: http://arxiv.org/abs/physics/0409014v3

    It is mostly discussing the Doppler effect. The same author published another paper that is more to the point, but I don't know if you can easily obtain it (you can write me a message): http://scitation.aip.org/content/aapt/journal/ajp/72/10/10.1119/1.1778390

    By similarly constructing with a length contracted mirror at an angle, you should be able to find the correct result for your particular question. If it is problematic, inputting your efforts here will likely trigger new responses. Just don't expect others to do all the work for you. ;)
    Last edited: Nov 22, 2014
  13. Nov 22, 2014 #12
    I'm a bit lazy sometimes, but I'm trying to improve ;)

    PDF downloaded successfully. Thanks!
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook