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

Does acceleration cause time dilation?

  1. May 26, 2008 #1
    Simple question (derived from some unanswered posts from various posters)...
    Does acceleration cause time dilation? Can someone shed light on this one?

    There seems two conflicting claims:

    acceleration cause time dilation
    acceleration does not cause time dilation
  2. jcsd
  3. May 26, 2008 #2
    acceleration causes a change in what the object being accelerated considers to be simultaneous. that is what Hurkyl was referring to. it does not contradict what DaleSpam said.
  4. May 26, 2008 #3
    >> when accelerating away from a clock, it is observed to run slower, or even backwards..

    This sounds pretty much like time dilation to me
  5. May 26, 2008 #4
    nope. look again.
  6. May 26, 2008 #5
    Yep, look again
    Last edited: May 26, 2008
  7. May 26, 2008 #6
    The question "Does acceleration cause time dilation?" is similar in context to "Does HIV kill you?". The answer is "yes, but not directly". Acceleration causes velocity, which causes time dilation. HIV causes AIDS, which causes death. It's not meant to be a perfect analogy, since one can undergo HIV without developing AIDS, but one can never undergo acceleration without developing velocity.

    Just because one experiences time dilation while undergoing acceleration does not mean that acceleration is the direct cause. If that were the case, then one would experience a constant rate of time dilation under constant acceleration. However, we know better: the rate of time dilation only stays constant with constant velocity. Therefore, it's directly related to velocity.

    The real question is: "What is time dilation constant with respect to?"

    The real answer is: "Velocity."
    Last edited: May 26, 2008
  8. May 26, 2008 #7
    how would you change what is considered to be simultaneous without some clocks seeming to move forward and some clocks seeming to move backward?
  9. May 26, 2008 #8


    User Avatar

    So accumulated proper time during acceleration can be found using a simple integral? I thought SR couldn't handle acceleration?
  10. May 26, 2008 #9


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    That's just a misunderstanding that's extremely common because of how SR is taught in introductory texts.

    As I pointed out in the "Einstein simultaneity" thread, Einstein's postulates are ill-defined and can't be the axioms of a mathematical theory. They're just there to help us guess the real axioms of the theory. It turns out we need only one: Space-time can be represented mathematically by Minkowski space.

    The properties of the minkowski metric imply that inertial frames exist, but it would be absolutely preposterous to pretend that those are the only coordinate systems we're allowed to consider, since a coordinate system is just a function that assigns numbers to events. Some authors claim that we're doing GR when we're considering other coordinate systems on Minkowski space. In my opinion that's just an obsolete way of thinking about SR that should have been abandoned decades ago.

    If the space-time we're considering is Minkowski space, then we're doing special relativity. If the manifold is curved, we're doing general relativity.
    Last edited: May 26, 2008
  11. May 26, 2008 #10


    Staff: Mentor

    Acceleration does not cause time dilation. This is known as the http://www.edu-observatory.org/physics-faq/Relativity/SR/experiments.html#Clock_Hypothesis" and has been experimentally verified up to about 10^18 g. Consider also muons created from cosmic rays in the upper atmosphere. They do not accelerate but instead are created at their high relative velocity. They are a textbook example of time dilation without acceleration.

    You can either say that velocity causes time dilation or that time dilation is just what happens when a clock takes a shorter path through spacetime. I prefer the second approach, which is the spacetime geometric explanation.
    Last edited by a moderator: Apr 23, 2017
  12. May 26, 2008 #11


    User Avatar

    So what they mean when saying "SR can't handle acceleration" would for instance be the traveling twins frame during the turnaround?
  13. May 26, 2008 #12

    I tend to agree with Fredrik thoughts in post #9 that suggest these type of statements are a common misconception.

    Special Relativity is perfectly capable of handling twins type paradoxes by plotting paths (even paths that change direction) on Minkowski type diagrams with distance on one axis and time on the other axis. Saying General Relativity is required to explain the twins paradox is a gross exageration. The power of General Relativity is only really required when things get really complicated such as when tidal effects have to be taken into account. (IMHO)
  14. Jun 17, 2008 #13
    It's science, but it's real. Nothing but real science here...
  15. Jun 17, 2008 #14
    Let me guess... you believe that a static axisymmetric body rotating on its axis of symmetry emits gravitational waves? If so, try spinning a disc under water and see how well the disc's edge pushes the water out of the way. Doesn't work quite as well as a spinning stick, does it? PhD? My God, they're giving them out in Cracker Jack boxes these days.

    Hopefully one of us is making sense.

    Either way, if you have an actual point, make it. Otherwise stop trolling.
    Last edited: Jun 17, 2008
  16. Jun 19, 2008 #15
    Time dilation is a symmetric artifact of relative motion.

    Sometimes when people use the term time dilation they actually mean differential aging.

    Differential or asymmetric aging is a consequence of different acceleration histories.
  17. Jun 19, 2008 #16


    User Avatar

    I think this shows pretty conclusively that acceleration by itself isn't the cause of differential aging (time dilation in the OP). Have a look at the spacetime diagram attached. It's a rework of the problem posed in an earlier thread.

    A is an asteroid moving left to right at 0.3c according to M (the midpoint observer). B is a spaceship moving right to left at 0.3c according to M. M sends a light pulse (event 'flash') and A and B set their clocks to zero when they receive the flash (A0 and B0).

    A and B collide at C. Since mass A >> mass B the spacetime path of A is unaffected by the collision.

    Moments before the collision A's clock and B's clock have recorded the same number of ticks. This should be clear from the symmetry and the calculated proper times (A0 to C and B0 to C) are the same. Moments after the collision, B is co-moving with A so from this point on both clocks should accumulate time at the same rate.

    Sometime after the collision (A1_B1), astronauts recover B's clock which is still ticking (it's a Timex). Both clocks will read the same time even though B underwent an (somewhat substantial) acceleration. Which, while it lacks the rigor of a proof, makes it pretty clear acceleration doesn't directly cause time dilation.

    Special thanks to Mentz114 for his 'Spactime plotter' software. Very nice and useful Mentz114.

    Attached Files:

  18. Jun 19, 2008 #17
    There is a very simple proof that acceleration does not cause differential ageing.

    Take 2 turntables. One has a radius that is 10 times bigger than the other. Place observer A on the perimeter of the small turntable and observer B on the perimeter of the large turntable. Spin up both turntables so that they reach the same perimeter velocity relative to inertial observer C who is not on a turntable. A and B experience different proper accelerations but when the turntables have stopped observer A and B note that the proper times recorded on their own clocks are the same despite experiencing different proper accelerations. They also note that the elapsed proper time on their clocks is less than that of observer C and can be accounted for by their equal perimiter velocities.

    Repeat the experiment, but this time spin the turntables so that both A and B experience the same proper centripetal accelerations. This time their perimeter velocities are different and they record different proper elapsed times.

    How is this a proof when I have shown no maths? Easy, an analogue of the experiment was actully carried out in a lab with a centrifuge where the equipment experienced centripetal acceleration millions of times greater than that of the Earth. The time dilation was proportional to the instantaneous linear velocity on the perimeter and independent of the proper acceleration. Case closed.
  19. Jun 19, 2008 #18
    I'd like to point out an example of a gravitational field which produces the exact same metric as the one used for a rotating frame of reference. If one takes a shell which has a finite thickness and a uniform mass distribution and sets it rotating about its axis of symetry while the observer inside remains in a coordinate system which remains non-rotating then the rotating shell will create a a gravitational field incuding frame-dragging effect. Even though the spacetime inside the shell remains flat there will still be a gravitational field. The field manifests itself by causing different clocks at different distances from the center of rotation run at different lengths. Observers who were initially at rest inside will experience a gravitational force directed radially away from the center of rotation.

  20. Jun 20, 2008 #19
    Given two identical oscillators, sitting side by side, keeping time at exactly the same rate, then if neither clock undergoes an acceleration, then there will be no difference between their accumulated times.

    On the other hand, if one of the clocks is accelerated for a certain interval, then brought back to rest beside the unaccelerated clock, then there will be a difference between the accumulated times.

    No acceleration, no differential aging. Isn't this correct, or am I missing something?
  21. Jun 20, 2008 #20


    User Avatar
    Homework Helper

    These posts are confliciting with the thread I created, where the "conclusion" was that the "equivlance" of gravity and acceleration held in the case of time dialation and have the same time dialation effect in GR. 2 objects, one experiencing 1g of gravity, the other 1g of acceleration experience the same time dialation component effect (their clocks would run slower than clocks experiencing 0g assuming the clocks at 0 g are at the same velocity).

    Going back to my analogy:

    Clock #1 on the equator of the earth at sea level, velocity is 465.09 m/s (relative axis of earth), and the clock experiences a gravitational force equivalent to 9.78033 m / s2 of acceleration.

    Clock #2 on the perimeter of a rotating space station in open space, with a radius of 22116.71 meters rotating at 0.0210289 radians / s, with a speed of 465.09 m / s, and experiencing centripetal acceleration of 9.78033 / s2. The space station is traveling at the same velocity as the earth, using rocket engines to duplicate the earth's orbital speed around the sun, as well as the sun's orbital speed around the galaxy.

    Clock #3 at the center of the same space station as clock #2, experiencing zero acceleration.

    Assuming that "equivalency" holds for time dialation in GR with respect to gravity and acceleration, then clock #1 and clock #2 should run at the same rate, but slower than clock #3.
    Last edited: Jun 20, 2008
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook