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

An absolute time clock!

  1. Dec 20, 2010 #1
    Here is my question, starting from a suggested experiment that reminds a known example that Einstein gave about special relativity including a moving train, two mirrors inside the train and a light beam reflected between the mirrors (this suggested thought experiment described here only reminds the historically well known thought experiment –please note critical differences):

    1. Lets question how to measure and compare measurements for an amount of time that it takes a light beam to travel from a bottom mirror to a top mirror in a train, once when the train is not moving (stationary at dock) and once when its moving at constant speed relative to dock (when measured only from within the train). As much as I understand this amount of time has got to be the same in both cases because the speed of light is constant at any direction and relative speed, while regarding the height of the train - there exists only horizontal length contraction, as seen from outside the train (in the same direction that the train is moving), so anyway there is no vertical length contraction from within the train. It seems that the time it takes light to travel from top to bottom mirror or from bottom to top mirror when looked at from inside the train is the same - doesn't matter if train is moving or not.

    I think it's important to repeat that point: we are measuring from within the train, and not what is seen from outside the train. Measuring once when the train is not moving and once when the train is moving. The measurement is made by dividing the height of the train by the speed of light (constant at any speed or direction).

    2. Suppose every time the light beam hits the top or the bottom mirror in the train, there is a mechanism that changes the light color a little bit.

    3. This light apparatus in the train will be a clock for us in here suggested experiment.

    One more thing that should be emphasized here, is that every time the light beam hits the bottom or top mirror, it registers a counter bottom and counter top photo-electric mechanisms with +1 hit.

    4. If we assume as described in bullet 1, that time dilation dose not occur in the duration of the travel that the light beam makes from bottom mirror to top mirror of the train (in other words lets put aside in this experiment the concept of: 'looking at the light beam in the train from outside') - as measured only from within the train once when not moving and once when moving. As well we know time dilation must occur in the apparatus as a whole (assuming time dilation at constant speed was experimentally proved) - Then time dilation must occur somewhere in the parts and components of this clock apparatus that change light beam color, and other parts that make the light bounce back, and in the parts that register the number of light hits – but time dilation does not occur in that clock component built with a vacuum tube where light itself bounces - because the assumption here is that light itself can not undergo time dilation.

    5. If we add one more condition here, and it is that changing the color of light beam and bouncing it back is a faster process than sending a signal over the wires of the apparatus towards the counter (say these wires are long enough), then:

    6. When this experiment is made, results should be as following - as the train is not in motion relative to dock, and after running the clock for a while, the counter will show a certain number in accordance with a certain light color, and when the train is moving relative to the dock, the exact same number that will show on the clock counter must be (according to all above assumptions) coupled with a different light color.

    This is because while the travel of light from the upper mirror to the bottom mirror takes the same amount of time (e.g. X seconds in both cases), in other parts of this clock apparatus, time dilation does make a difference (e.g. changing color of light beam and bouncing the light back takes 2X when train is stationary, and 4X when train is moving, registering light hits takes 10X when train is stationary, and 20X when train is moving). The result is that if we stop the light hit counters of the moving train and the stationary train at the same counter numbers, the corresponding light colors will not be the same.

    7. If such results will appear this might be a serious problem!! That is because, although such results still show in a way the validity of time dilation, they enable experimentally to say if the train is moving or not, without looking out of the window - which brings physics back 105 years to the necessity of hypothesizing some sort of ether.

    8. If none of the above will happen, and when counters will be stopped at the same number, corresponding light colors will be the same, then my question is, won't such results represent a situation, in which light itself undergoes time dilation?

    Diagram 1: 20X/5X=4 but 10X/3X=3.33

    Now, here is a simplification of the 'Bouncing light' experiment:
    While there is no time dilation effect on the wires, there is time dilation effect on each cycle of crystal vibration, as a result of the electric pulse, therefore when the apparatus was adjusted to have light sources turn on together in stationary apparatus, the exact same arrangement of wires, crystal and light sources, will not have the light sources turn on together when in the moving apparatus.

    Diagram 2: 2X+1X+2X=5X but 2X+2X+2X=6X

    Attached Files:

    Last edited: Dec 20, 2010
  2. jcsd
  3. Dec 20, 2010 #2


    User Avatar
    Science Advisor

    Huh? If you're putting aside the concept of looking at the train from the perspective of the outside observer's frame, and only considering what happens in the train's own rest frame, then there is no time dilation "in the apparatus as a whole" either! It's only when you consider things from the perspective of a frame where the train is moving that this clock slows down, and in this other frame there is a change in "the duration of the travel that the light beam makes from bottom mirror to top mirror" (because in this frame the mirror is moving horizontally, so the light has to travel the vertical distance along with some horizontal distance to get from bottom to top and top to bottom).
    In what frame are these statements supposed to be true? In the train's own rest frame, both the time for the light to travel from one mirror to another and the other processes ('changing color of light beam and bouncing the light back') are completely unaffected by whether the train is at rest relative to the ground or whether it's moving relative to the ground. And if it's moving relative to the ground, then in the ground frame all these processes are slowed down by exactly the same amount.
  4. Dec 21, 2010 #3
    If so, I ask if jets flying atomic clocks were looked at from outside in any way while they were flying?

    Please refer me to an experiment that did check this out at relativistic speeds - that light oscillation does not start to differentiate from time dilation of other processes that occur inside matter, as speed of light is approached. I don't think that experiments that are considered to prove time dilation, also prove this point. If they do, please try to give me a detailed answer as possible relating directly to the elements of these experiments.

    Thanks a lot,
    Last edited: Dec 21, 2010
  5. Dec 21, 2010 #4


    User Avatar
    Science Advisor

    No, their total elapsed time was compared with the total elapsed time on Earth clocks at the beginning and end of the trip, and verified that they showed less elapsed time by the amount that relativity would predict. No measurement was done to show that the clock was "running slow" in the Earth frame during any individual segment of the trip, but obviously they were running slow relative to Earth clocks on average over the entire trip. You might compare with the twin paradox. Also, the GPS system requires that atomic clocks on board GPS satellites be continually corrected for the fact that their velocity and distance from the Earth cause them to run slow in an Earth-centered coordinate system, so the fact that the system works correctly is a verification that they do in fact run slow in this coordinate system by just the amount predicted (see here for a little discussion).
    Basically all the physical processes we observe in daily life can be analyzed in terms of electromagnetic laws--even a mechanical process like gears pushing against each other is understood to work because the molecules that make up the objects are interacting via the electromagnetic force. And if you write down the most accurate possible expressions for the equations governing electromagnetic interactions, they have the mathematical property of "Lorentz invariance", which means that if you transform the equations into a different inertial frame according the the "Lorentz transformation" which relates different frames in relativity, the equations are unchanged, implying that any physical system governed by these laws will behave the same way in its own rest frame regardless of how fast it's moving relative to the Earth. For this to be incorrect, the "true" equations of electromagnetism would have to depart somewhat from out best estimate of the equations, in such a way that the true equations were no longer Lorentz-invariant, but the equations have been verified to a high degree of accuracy so such a departure would have to be pretty tiny. There are also phenomena such as radioactive decay which require non-electromagnetic laws to understand them, but our best estimates of the equations governing these forces (the strong and weak nuclear forces) are also Lorentz-invariant.

    For some experiments which more directly test whether the laws governing specific physical systems and devices are Lorentz-invariant, see here.
  6. Dec 21, 2010 #5
    Hello Jesse,

    I don't see how the twin paradox has to do with the fact that in experimenting to establish time dilation, there is no action involved of looking from outside. If you (and me as well) accept the existence of time dilation, then it means that time dilation as a phenomenon regarding in this case exclusively constant speed, was proved in a non-thought experiment without looking from the outside of the moving vehicle into the inside of it.

    You referred in your reply to two experimental proofs of time dilation: GPS and radioactive decay.

    I see how they prove time dilation, but I don't see how in either of these experiments there is a four-fold comparison of light oscillation to matter oscillation at a relativistic speed difference between two identical devices, neither of course in the jets-atomic-clock experiments.

    You say that also matter can be described by electromagnetic-laws, but still the difference between matter and light is well established.

    Please try to help me understand an experiment that actually was done in the past (while I suggest that such an experiment was never done yet) - By explaining such a past experiment detail by detail, and by comparing it to what I suggest. I can not understand these links to these very long lists of past experiments. Please chose one of them that explains your point and compare it in simple words for me.

    Thanks for trying to help me understand!
    Last edited: Dec 21, 2010
  7. Dec 21, 2010 #6


    User Avatar
    Science Advisor

    Hopefully you understand that "outside/inside" is just a metaphor for different frames of reference (coordinate systems)? An observer could be physically inside the train but if he was moving relative to the clock, the clock would be time-dilated in his rest frame (likewise an observer outside the train but moving at the same velocity relative to the ground as the train itself would find that the clock showed no time dilation as measured in her own rest frame). In the case of the GPS clocks, we do measure their moment-by-moment behavior in the context of an Earth-centered coordinate system, and they do slow down relative to the time coordinate in this system by the amount predicted by relativity.
    What do you think is being shown in atomic clock time dilation measurements, time dilation of "light oscillation" or "matter oscillation"? Atomic clocks involve both the oscillation of energy levels of electrons in atoms (matter) and also the photons emitted due to these oscillations (light).
    I don't see how that statement really addresses my argument. Let's break it down into different sub-claims: do you agree that both matter and light behavior are accurately predicted by electromagnetic laws (aside from phenomena involving nuclear forces like radioactivity, or involving gravity) that the most accurate known form of these laws is Lorentz-invariant (as are the most accurate known forms of laws for nuclear forces and gravity), and that it's logically impossible that any system obeying Lorentz-invariant laws could fail to match the predictions of relativity? If you disagree with any of these three statements, which one, and why?
    I don't really understand what type of experiment you're suggesting in the first place, "a four-fold comparison of light oscillation to matter oscillation at a relativistic speed difference between two identical devices" is unclear to me. What precise signals/events relating to each device are being "compared", and how do we investigate "matter oscillation" without using light in any way?
    I don't think it's possible to explain physics in "simple words" (unless these simple words are descriptions of math that you already have some understanding of), for example the notion of "Lorentz invariance" itself is intrinsically mathematical in nature. For example, the experiment here tests the hypothesis there is some slight difference from Lorentz-invariance in the laws of physics, which apparently implies that protons in cosmic rays with sufficiently high energy would be unable to reach the surface of the Earth if the Earth is moving relative to a preferred frame in which the laws of physics are invariant under both rotation and translation (see continuous symmetries). But showing why the implication follows from the premise probably couldn't be explained in any simple nonmathematical way.
  8. Dec 21, 2010 #7


    User Avatar

    Staff: Mentor

    You're implying here that different types of clocks may show different (or not at all...) Relativistic effects, correct? On what basis do you make this claim? Either Relativity is real and affects time or it isn't real and all clocks have their own unique error modes -- and several different types just happen by random coincidence to exactly match the predictions of Relativity (with no alternate explanation for why this could happen).

    Is this what you are suggesting? Do you grasp the magnitude/implications of that claim?
  9. Dec 21, 2010 #8


    Staff: Mentor

    Off the top of my head I can think of laboratory experiments showing time dilation for electromagnetism, the strong force, and gravity. That leaves only the weak force, and I would suspect that there are experiments using that one too, I just never bothered to look considering the existing evidence to be overwhelming.

  10. Dec 22, 2010 #9
    Hello Jesse,

    What I think happens in atomic clocks, is that the process of oscillation that occurs inside matter itself gets dilated, while when light is emitted out of matter, the time it takes to get to the photo-electric detector is not dilated, but always stays the same time at a universal and at an absolute pace (if light path is perpendicular to the path of the clock in the jet). In atomic clocks there is no comparison of light behavior to matter behavior, and matter dilation is seamlessly coupled with light constancy.

    In atomic clocks, you always get only the read of what happens in matter, because you don't have another counter that is positioned in a way that the two different paces can be compared, one counter of light dissemination, and another counter of matter oscillation.

    I think that there might be an inaccuracy in the predictions of the way that light behaves in comparison to matter. That is because light is considered also to undergo time dilation, altogether with the whole moving device, while this contradicts the assumption, or the fact, that light speed is always constant.

    Last edited: Dec 22, 2010
  11. Dec 22, 2010 #10


    User Avatar

    Staff: Mentor

    Muon decay proceeds via the weak interaction, so the observations of muons in storage rings apply here.
  12. Dec 22, 2010 #11


    User Avatar
    Science Advisor
    Gold Member

    Roi, you have proposed that the only reason an absolute ether reference frame has not yet been detected is because no one has yet done the experiment you propose and your logic is that if your experiment shows a difference in results based on the speed of the apparatus, then Special Relativity cannot be true because it says everything looks the same independent of speed, but if it does show a difference then it indicates that Special relativity cannot be true because your experiment would prove that light is experiencing time dilation but Special Relativity says that the speed of light is a constant independent of the speed of the source of the light. Either way, Special Relativity cannot be true and the absolute ether frame becomes the default truth. So the experiment doesn't even have to be performed. Is this a correct summary of your position?
  13. Dec 22, 2010 #12


    Staff: Mentor

    That will do it then. That is all 4 fundamental forces showing time dilation. The OP has no basis for thinking that a clock based on some other process will not show time dilation. All 4 known processes have been experimentally confirmed.
  14. Dec 22, 2010 #13
    Hello ghwellsjr,

    Of course I think this suggested experiment should be performed, if there is to be a proof that an absolute time clock can be built.

    As much as I understand, this is exactly the phenomenon that the machine I propose will use - That the speed of light is a constant, independent of the speed of the source of the light.
    Last edited: Dec 22, 2010
  15. Dec 22, 2010 #14


    Staff: Mentor

    An absolute time clock cannot be built using EM, the strong force, the weak force, or gravity (what else is there) because all of those have been experimentally shown to exhibit time dilation. Your proposed clock uses only EM, so it is clearly going to show time dilation.
  16. Dec 22, 2010 #15
    Hello DaleSpan,

    Please try to explain to me, how does the assumption that light speed is constant, consistent with such experiment results, that say that EM actually does undergo time dilation.

  17. Dec 22, 2010 #16


    Staff: Mentor

    Time dilation is part of the Lorentz transform and Maxwell's equations are invariant under the Lorentz transform. Basically, the speed of light is unchanged but the frequency shows time dilation.
  18. Dec 22, 2010 #17

    As much as I understand, what you say is not to say that the speed of light becomes constant at the very moment it is detected, but before it was detected it had a higher velocity - and that only detection itself translates this speed into frequency change. The frequency is the same, the moment it leaves the light source, independent of the speed of light, right? Or isn't it so? What do you mean by frequency shows time dilation?
  19. Dec 22, 2010 #18


    Staff: Mentor

    Huh? I don't know how you got any of that from either what I said or anything that has been written about relativity.

    You seem to have some idea that the invariance of c is incompatible with time dilation. The Lorentz transform is the mathematical framework of special relativity. It includes time dilation and the invariance of c. Far from being incompatible, they are both implied by the Lorentz transform. Do you understand how time dilation and the invariance of c are obtained from the Lorentz transform?

    I mean that the usual way to test for time dilation is to measure the Doppler effect (frequency shift) and see if it has the classical value without time dilation or the relativistic value with time dilation.
  20. Dec 22, 2010 #19
    If Einsteins posulate borrowed from De Witter is a parameter in the totality of motion described here, how could the reflecting light seem to move along with the moving frame? The reflection should have no effect of dragging the reflecting beam along with the motion of the frame. The model you use here necessarily adds a component of light in the direction of the frame motion resulting in the speed of light being dependent on the speed of the frame, as described by your model - the same flaw is ingrained in virtually every published MichelksonMorely experimental arrangement.
  21. Dec 22, 2010 #20


    Staff: Mentor

    I have never seen any publication showing a reflection dragging a beam. What are you talking about?
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Similar Discussions: An absolute time clock!
  1. Absolute Time (Replies: 32)