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

Relativity, time and the speed of light

  1. Jul 10, 2003 #1


    User Avatar

    It states that nothing can travel faster than light. But all the books that I've read on introduction to relatvity use the train and light pulse to demonstrate length contraction and time dilation. And finally, they say that nothing can travel faster than light. How can this claim be made when they have just only consider light clock. What about mechanical clock, biological clock etc?
  2. jcsd
  3. Jul 10, 2003 #2


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    It all stems from the principle of relativity.

    Even before Einstein, physicists postulated that the laws of physics are the same in all inertial reference frames. (in classical mechanics, we call this Gallilean relativity) This is why you can be in a speeding locomotive, but feel like you're sitting still!

    Then Maxwell came along and collected the most important laws of electromagnetism, including some contributions of his own, into Maxwell's equations. The problem with Maxwell's equations is that they don't work with Gallilean relativity!

    Through the thought experiments with light beams, we see that the needed correction is to allow time dilation and length contraction (I shall now call it Einsteinian relativity). Remember that the governing principle is that the laws of physics are still supposed to be the same in all reference frames.

    The laws of physics include the fact that we can synchronize ordinary clocks (of all kinds) with light clocks... so if Einsteinian relativity is correct, then the time dilation we observed with light clocks must apply to all means of measuring time. A similar argument works with measuring distances.

    This, of course, was very disconcerting to physicists in the earlier part of last century (and even to some in the latter part)... but they have since confirmed the predictions of special relativity in every experiment performed. To my knowledge, there is literally not a single shred of experimental evidence that special relativity is flawed.
  4. Jul 10, 2003 #3


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    All such clocks must remain synchronized with the light clock that is in the same frame. Otherwise, if you were to bring clocks that have experienced relative velocities back together in the same frame, there would be disagreement as what each clock read. Observer A might say that clock B reads 10:00, but Observer B would insist that it reads 10:10. Remember, this would be after the clocks and observers had been brought back together again.
  5. Jul 12, 2003 #4
    Janus' response is only partially right. If the other clock is moved arbitrarily slowly it will show the same time indication when it comes back to the original clock and set to rest in relation to that clock.

    It should be mentioned that there are 2 interpretations of relativity which in all practical situations yield the same results. For Einstein the speed of light c is a constant by itself. In motion the space shrinks and the time is dilated.

    In the other interpretation called the Lorentzian one (which is not the exact original version of Lorentz) the speed of light is *measured* to be the same in all moving frames. Not the space contracts but the physical objects do and not the time is dilated but all physical time measurements. As a result the speed of light c which is *not* the same in different moving frames seems to be the same by measurent.

    As a result in the world of Einstein there is no speed possible greater than c because this is inhibited by our fundamental space properties. In the world of Lorentz the fields and the particles have this speed c as a limit, but it is not fundamentally forbidden that a speed > c can occur. This is an important point in the view of recent experiments which seem to prove speeds much greater that light.
  6. Jul 12, 2003 #5


    User Avatar
    Staff Emeritus
    Gold Member

    None of those experiments are incompatible with relativity. Information, or any other physical influence, is not transmitted in any of them.
  7. Jul 12, 2003 #6
    I know that there are controverse discussions about this point. I have personally observed one of the known experiments (of Guenter Nimtz, Cologne) during which by my understanding information was transfered with a speed much greater than c. This experiment was observed by a lot of physicists who are very critical about it. But none of them could explain all aspects of the experiment without admitting a superluminous signal.
  8. Jul 12, 2003 #7
    Please show the source of the information that information is transmitted with a speed much faster than c.
  9. Jul 12, 2003 #8


    User Avatar
    Staff Emeritus
    Gold Member

    The effect does not pose a problem for SR, and is well understood and explained. No information is transmitted faster than light.

    Imagine a long street, with light posts every 10 meters. You can easily set things so that all of them turn on at the same time: you just have to send the signals at different times (the "turn-on" commands for the farther posts depart earlier, to compensate for the propagation time).

    Similarly, you can also set this up so that the lightning of the bulbs appears to "travel" through the street at, say, 10 times the speed of light.

    The front end of your "bulb-wave" travels faster than light, but no physical influence was transmitted faster than light.

    Nimtz experiment works in an analogous way.
  10. Jul 12, 2003 #9
    The Nimtz experiment I have seen works in the following way:

    A microwave pulse of ca. 0.1 ns length is transmitted through air by a distance of 3 meters. Then a specific material which is 3 cm thick is put into the beam. The power of the received pulse is then attenuated by a factor of ca. 400, but the width is the same and there is no signal ribble before or after this pulse. The pulse now arrives ca. 0.1 ns earlier, which means (within the measurement accuracy) that the microwave pulse did not need any time to travel through the material.

    There was an argument that this result can be explained by a frontal wave. That means that only the first little part of the pulse comes through and seems to represent the whole pulse and the rest is extinguished by destructive interference.

    To refute this argument Nimtz has changed his set up. He has transmitted the pulse towards a layer of ca. 0.5 cm thickness where it was reflected towards the receiver. There appeared a pulse of the given length. Then he added another 2.5 cm of this material to the layer. Of course the received pulse amplitude was now increased. But the received pulse was not spread out in time in accordance to the thickness of the reflecting material but it had the same position and the same width. So that one has to conclude that the pulse did not use any time to invade the additional material until it was reflected and also no time for its way back.

    There is a lot of papers of G. Nimtz. One I have at hand is
    A. Enders and G. Nimtz, Photonic-tunneling experiments, Physical Review B, (45,9605) 1993
  11. Jul 13, 2003 #10
    I couldn't read the article Albrecht indicated, but I could read the related article of the same author. You are referring to tunneling time. The phenomenon is releted to an evanescent wave. I don' t know details of tunneling time. But can we say the evanescent wave is traveling? I need to know how the light wave travels in the medium from the microscopic point of view.
  12. Jul 13, 2003 #11
    I don't know if it was the same person but I saw on either PBS or TLC or Discovery a similar setup in which the throught the medium arrives sooner that the one transmitted through air and he was send a Mozart
    piece over the signal. He said that Mozart should satisfy those who claim that no intelligence could be sent faster that light.
    At the time I assumed he was talking about and demonstrating tunneling.
  13. Jul 19, 2003 #12
    Nimtz has in fact transmitted Mozart with a superluminous speed.

    However, whatever the signal transmitted by Nimtz was, an evanescent wave or a tunnelling signal or something different: It was stable enough that it could be used to synchronise clocks in moving systems.

    Normally clocks at different locations and in different moving frames are synchronised in a way that a light like signal (with v=c) is send from one clock to the others. The arrival time of the signal has to be corrected for it's flight time. But because every observer in a different system relates the speed of light c to his own reference system, the result of the synchronisation will look different to the different observers. These differences are symmetric with respect to the systems, so they cannot be used to find that one system is a preferred one. This impossibility fulfils the relativity principle declared by Einstein.

    This, however, changes if a superluminous signal is used. If clocks are now synchronised using on one hand a light like signal (i.e. one moving with c) and using on the other hand a superluminous signal as generated by e.g. Nimtz, both methods will in general have a different result. It can be shown by the Lorentz transformation that there is exactly one inertial system in which both methods have an identical result. So there is a preferred inertial system. And this is in a strong contradiction to the relativity principle of Einstein.

    But it is in no contradiction to the relativity as defined by Lorentz. Lorentz has always stated that there is a preferred inertial system, formerly called the "ether".
  14. Jul 20, 2003 #13
    According to my thought, spacetime is regularized by the light. Every textbook of relativity refer to the speed of light in vacuum. If some material which makes a propagating light an evenescent wave intervenes in the path of light, the situation is different from the case in vacuum. We have to discuss the relativity in consideration of the property of the material. Even if the light signal is transmitted faster through the medium, the transmitting time through the medium must be shortest and the speed at that time should be thought like constant (speed of this case may be different from c), I think. We need to know the behavior of an electromagnetic wave in the medium according to the theory of relativity.
    Last edited: Jul 20, 2003
  15. Jul 20, 2003 #14
    It is right that the textbooks solely refer to the relativity of Einstein. The relativity of Lorentz which produces equivalent mathematical results is almost completely ignored. As well the logical problems which occur with Einstein but not with Lorentz are ignored.

    The point I made regarding clock synchronisation is independent of how one interprets the cause of the higher speed. It causes anyhow a logical deadlock with Einstein.

    You can also consider the following: If the signal of Nimtz passes the material barrier with a speed greater than c and this is observed from an inertial system which moves with high speed into the opposite direction, the moving observer will see that the signal leaves the material before it has entered it. - This is a very strong logical conflict.
  16. Jul 20, 2003 #15
    It's an interesting post but unfortunately not fully accredited as yet. The theory seems to depend on the cascading effect of photons travelling through a medium which would add up to a momentum which is faster than c.
    Last edited: Jul 20, 2003
  17. Jul 20, 2003 #16
    On Superluminal Propagation

    For the record, I, would like to point some articles which explain why SUPERLUMINAL TRANSMISSION does not violate Einstein causality (postulate of SR).

    I understand this is a controversial topic among reputable experts, but I share the opinion of these authors (who I happen to know and respect). Another expert who agrees with this interpretation of the results is, of course, Einstein.

    University of Toronto (currently)
    Prof. Mojahedi - AN ENGINEER

    + Demonstrated superluminal transmission - explicitly mentions that SR is NOT violated.
    + Explains why pulse-reshaping does not constitute faster than c information transmission

    REFERENCE: http://www.grc.nasa.gov/WWW/bpp/pdf/Mojahedi-JPC.pdf
    (see section V - Superluminal Velocities and Einstein Causality)

    (Quotes from)
    University of Toronto
    Prof. Steinberg - PHYSICIST

    + Refers to Ranfagni's microwave experiment. Explains that this experiment still demonstrates pulse reshaping.

    REFERENCE: http://focus.aps.org/story/v5/st23 [Broken]
    Last edited by a moderator: May 1, 2017
  18. Jul 20, 2003 #17
    Quoting Albrecht

    Albrecht said:

    I know that there are controverse discussions about this point. I have personally observed one of the known experiments (of Guenter Nimtz, Cologne) during which by my understanding information was transfered with a speed much greater than c. This experiment was observed by a lot of physicists who are very critical about it. But none of them could explain all aspects of the experiment without admitting a superluminous signal.

    Not only physicists are being very critical about it, so are engineers. I personally, am critical about any interpretation that violates causality in its simplist form (effect preceeding the cause). My previous post refers NOT to a 1993 paper, but two articles sharing a more recent, maturing interpretation of superluminal transmission (which is easily reproducable).

    To clarify my point, I am not saying that any researchers "cheated" or that they somehow erred, I am saying that I disagree with the interpretation that violates causality.

    Try plotting (using MAPLE, Mathematica, Origin or whatever) the following curves (if you can't visulalize pulse reshaping) for a demonstration of how pulse reshaping can cause superluminal effects:

    RED: exp[-(t)^2]
    BLACK: exp[-(t-8)^2]
    BLUE: AMP*exp[-(t-8)^2]

    where the amplitude modulation is AMP = {0.7*[arctan(t-7)]^2}

    **Note that I amplitude modulated the blue pulse (just as some mechanism, an UNKNOWN BOX would), but that the Gaussian component is NOT centered at t~9, but at precisely t=8 (just like the BLACK curve).

    Consider the t-axis to represent time. At some earlier time (much earlier - not shown), a Gaussian wavepacket is generated and it is later detected to pass through some detector. This is the RED curve shown. This detector pinpoints the PEAK of the wavepacket to be temporally centered at t=0. The pulse then traverses some distance, goes through some UNKNOWN BOX that reshapes the pulse. The pulse is then directed along some path and returns to the detector.

    The BLACK curve shows what would be detected IF the pulse wasn't amplitude modulated (i.e. wasn't reshaped). Note that it is centered at t=8 (i.e. its PEAK is located at t=8). The blue pulse shows the pulse that was reshaped by the UNKNOWN BOX. Its peak is NOT at t=8, but at t~9. Did this violate causality?

    The point is that although the blue curve travelled at a speed equal to the speed of light, the PEAK of the curve arrived at its destination FASTER than the speed of light. This is OK!

    Attached Files:

    Last edited: Jul 20, 2003
  19. Jul 20, 2003 #18
    Although I have not read articles sdeliver645 showed and have not conducted calculation sdeliver645 showed, I want to express my opinion.
    I think I understand what sdeliver645 says. Although Albrecht says causality is violated, I believe not so. Yes, what is occuring is pulse reproduction as sdeliver645 says. My problem is how the tunneling time is transformed between inertial frames. As sdeliver645 clearly showed, wave becomes decaying one in the medium in the case of tunneling. So we cannot define phase of decaying wave. We cannot define the speed of decaying wave based on the usual method. If the tunneling occurs at the greater-than-light speed viewed from only one inertial frame, violation of causality occurs. Think that famous diagram which is a sectional view of light cone. But I don't believe causality is violated. Maybe, I speculate that tunneling occurs the same superluminal speed for all inertial frame. If so, I think causality is not violated.
    Last edited: Jul 20, 2003
  20. Jul 20, 2003 #19


    User Avatar
    Staff Emeritus
    Gold Member

    Great illustration sdeliver645!!

    I'm posting the plot you suggested as an attachment. The light blue curve is AMP(t).

    (the attachment won't show until approved by Greg).

    Attached Files:

    • c1.gif
      File size:
      8.1 KB
  21. Jul 20, 2003 #20
    Thank you! The attachment clearly shows your understanding! But I have some question.
    1) according to the explanetion attachment shows, time advancement of light pulse should be within half width of original pulse.
    2)further, attenuated pulse's width should be decreased.
    Albrecht's report says:
    1)width of original pulse was 1ns,
    2)width of attenuated pulse was the same as the original one,
    3)advancement of traveling time was 1ns.
    Is Albrecht's report wrong? Or is the explanation not enough?
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook