Time Dilation Formula: Clarifying Confusion

[Grimble]

I know that this is a very basic question but what is the correct formula for time dilation?

In Wikipedia etc. I read ${t^'} = \gamma {t}$ or at least $\Delta{t^'} = \gamma {\Delta{t}}$; yet in this http://en.wikipedia.org/wiki/Twin_p...t_of_differences_in_twins.27_spacetime_paths" 'phase 2' and 'phase 5' imply that the formula is $${t^'} = \frac {t}{\gamma}$$.


Also, if a moving clock is seen to 'go slow' by a stationary observer, then one would expect that less time would be seen to pass in the transformed time, and $${t^'} = \frac {t}{\gamma}$$seems to me to fit that scenario.

I have been looking at this for some time on the internet but, taking heed of the warnings I have been given about believing all I read on there, I have followed the arguments and read the 'derivations' and suchlike, but have a problem:

Whichever way I approach it the formula appears to be the latter viz. $${t^'} = \frac {t}{\gamma}$$ in the same way that $${x^'} = \frac {x}{\gamma}$$ the formula for length contraction.


where:
t is the time on the stationary observer's local clock and
t' is the traveling clock's time, transformed by the Lorentz transformation formulae.

Or are there different formulae applied in different circumstances.

We talk of time dilation - expansion(?) yet also about the moving cock slowing (less time passing)?

 

[dianaj]

In a moving system time seems to go slower while objects seem to get longer. As

$$\gamma = \frac{1}{\sqrt{1 - \frac{v^2}{c^2}}} > 1$$

for v > 0, the correct formula for time dilation must be

$$t' = \frac{t}{\gamma}$$

from which follows

$$t' < t$$

e.g. when t minutes has passed in the rest system only t' has passed in the moving system meaning that times moves slower in the moving system.

The length contraction formula must be

$$x' = x \gamma$$

as the length of an object in the moving system appears to be contracted and not dilated.

I hope this sorted out your confusion. Of course this answer is not a derivation of the equations - this is just my line of thought when I forget when to multiply/divide by $$\gamma$$

 

[diazona]

I often go through the same thinking as dianaj. It definitely helps to remember that $\gamma > 1$.

 

[matheinste]

I know that this is a very basic question but what is the correct formula for time dilation?

In Wikipedia etc. I read ${t^'} = \gamma {t}$ or at least $\Delta{t^'} = \gamma {\Delta{t}}$; yet in this http://en.wikipedia.org/wiki/Twin_p...t_of_differences_in_twins.27_spacetime_paths" 'phase 2' and 'phase 5' imply that the formula is $${t^'} = \frac {t}{\gamma}$$.


Also, if a moving clock is seen to 'go slow' by a stationary observer, then one would expect that less time would be seen to pass in the transformed time, and $${t^'} = \frac {t}{\gamma}$$seems to me to fit that scenario.

I have been looking at this for some time on the internet but, taking heed of the warnings I have been given about believing all I read on there, I have followed the arguments and read the 'derivations' and suchlike, but have a problem:

Whichever way I approach it the formula appears to be the latter viz. $${t^'} = \frac {t}{\gamma}$$ in the same way that $${x^'} = \frac {x}{\gamma}$$ the formula for length contraction.


where:
t is the time on the stationary observer's local clock and
t' is the traveling clock's time, transformed by the Lorentz transformation formulae.

Or are there different formulae applied in different circumstances.

We talk of time dilation - expansion(?) yet also about the moving cock slowing (less time passing)?

To a "stationary" observer, a clock moving relative to him will appear to run at a slower rate than his. If we call the time between ticks on the "stationary" clock as observed by the "stationary" observer one second, then the time between ticks on the "moving" clock as measured by the "stationary" obsever will be greater than one second, and so in this sense the time in the "moving" frame as observed from the "staionary" frame could be described as expanded. That is, the time between ticks appears to be longer. The term dilated is normally used rather than expanded . So in a certain time as measured by the "stationary" observer on his own clock he oberves a smaller number of ticks on the "moving" observer's clock. More observed time between ticks is taken to mean time passing more slowly, and so the time observed in the "moving" frame by the "stationary" observer can be said to be passing more slowly. I expect you already knew all this but were unhappy with the terminology.

As for time passing more slowly or being dilated, there is no absolute time. For an ideal clock, elapsed time IS the time measured by a comoving observer counting the ticks. Time IS the ticks. To any inertial observer the time elapsed on HIS clock is THE elapsed time, or proper time. For this observer, his own physical time rates never alter.

Matheinste.

 

[Grimble]

To a "stationary" observer, a clock moving relative to him will appear to run at a slower rate than his. If we call the time between ticks on the "stationary" clock as observed by the "stationary" observer one second, then the time between ticks on the "moving" clock as measured by the "stationary" obsever will be greater than one second, and so in this sense the time in the "moving" frame as observed from the "staionary" frame could be described as expanded. That is, the time between ticks appears to be longer. The term dilated is normally used rather than expanded . So in a certain time as measured by the "stationary" observer on his own clock he oberves a smaller number of ticks on the "moving" observer's clock. More observed time between ticks is taken to mean time passing more slowly, and so the time observed in the "moving" frame by the "stationary" observer can be said to be passing more slowly. I expect you already knew all this but were unhappy with the terminology.

As for time passing more slowly or being dilated, there is no absolute time. For an ideal clock, elapsed time IS the time measured by a comoving observer counting the ticks. Time IS the ticks. To any inertial observer the time elapsed on HIS clock is THE elapsed time, or proper time. For this observer, his own physical time rates never alter.

Matheinste.

Thank you, one and all, for your inputs.

It is interesting, Dianaj, and Diazona, that you have the two formulae the opposite way round to Wikipedia, whereas I am inclined to think that you each have one right! Confusing isn't it, a slippery thing to keep one's finger on.

You give a nice summary of the problem, Matheinste: there is no Absolute time – agreed; and time is in the eye of the beholder, if I may paraphrase you.

The big difficulty I see is how to describe 'faster' and 'slower' in time when are we counting the 'clicks' but have no agreement on the size of the clicks.
For instance, one clock may be slower than the other yet record the same number of clicks.
Let me refer you to Einstein's 1920 paper: 'Relativity: The Special and General Theory.'
and in particular to chapter XII. - http://www.bartleby.com/173/12.html" where he derives a formula for the time in the stationary system K when ${t^'} = 1$ of
$$t = \frac{1}{\sqrt{1 - \frac{v^2}{c^2}}}$$
in which ${t^'}$ has been replaced by 1 and which we would write:
$$t = \frac{t^'}{\sqrt{1 - \frac{v^2}{c^2}}}$$
He then goes on to state:
'As judged from K, the clock is moving with the velocity v; as judged from this reference-body, the time which elapses between two strokes of the clock is not one second, but
$$\frac{1}{\sqrt{1 - \frac{v^2}{c^2}}}$$

seconds, i.e. a somewhat larger time. As a consequence of its motion the clock goes more slowly than when at rest.'

But, consider just what he is saying here: the time t is the time in system K, the stationary system and is, therefore, 1 second proper time, but the time ${t^'}$ is the time from the system ${K^'}$ transformed into co-ordinate time (as we refer to it).

So where Einstein says that the clock slows one could just as easily say, that the time of system ${K^'}$ has been 'shrunk' or 'contracted' by the transformation such that one second proper time, system K, is now equal to
$$\frac{1}{\sqrt{1 - \frac{v^2}{c^2}}}$$
seconds co-ordinate time, which is greater than one.

So 1 second proper time in system ${K^'}$ (measured in system ${K^'}$ it is in an inertial frame of reference) upon transformation becomes only $$\frac{1}{\frac{1}{\sqrt{1 - \frac{v^2}{c^2}}}}$$ seconds.
So one might say that it has slowed down as the same duration now measures less elapsed time, or, that the units of time have shrunk and that time now passes faster!
It all depends on what one is comparing, number of units or size of units.

My personal preference would be to say that when transformed by the Lorentz equations, the units of time become smaller.

 

[matheinste]

Thank you, one and all, for your inputs.

It all depends on what one is comparing, number of units or size of units.

:

When you compare your clock to a moving clock, the moving clocks units, that is time between ticks, as measured by you, appears to you to be longer than those on your clock. Because number of ticks and length between ticks are inversely proportional, the moving clock will appear to you to have experienced less ticks compared to your clock.

Time dilation refers to the lengthening of the distance between ticks, compared with your clock, of a clock moving relative to you as observed by you. So comparative length between ticks in the frame moving relative to you, appears to you to be longer, or dilated, That is time dilation. But the comparative number of ticks in the frame moving relative to you as observed by you appears to be decreased, that is smaller, not dilated, when compared to your clock. This is true for any observer when observing another clock moving relative to his.

Of course if you interchange the roles of the observers and let the obvserver who was considered to be moving, now consider himself, quite legitimately for the present pupose, to be at rest, then he will consider the other clock to be slower than his. All clocks moving relative to any observer will be observed by him to run slower than his own.

Matheinste.

 

[Grimble]

Time dilation refers to the lengthening of the distance between ticks, compared with your clock, of a clock moving relative to you as observed by you. So comparative length between ticks in the frame moving relative to you, appears to you to be longer, or dilated, That is time dilation. But the comparative number of ticks in the frame moving relative to you as observed by you appears to be decreased, that is smaller, not dilated, when compared to your clock. This is true for any observer when observing another clock moving relative to his.

Matheinste.

Let us take an example of time dilation; if we say that for a clock moving at 0.8c, for each second that passes and is indicated on that clock, how much time, (time dilated), will the stationary observer see pass?

Now if we put the appropriate figures into the formula for time dilation $t = \gamma {t^'}$
where:
t = the proper time for the inertial observer.
v = 0.8c
t' = co-ordinate time in the moving clock's frame
[tex]\gamma = \frac {1}{\sqrt{1 - {\frac{v^2}{c^2}}}} = 1.67[/itex]

then [itex] t = 1.67 {t^'}

So 1.67 seconds on the moving clock is equivalent to 1 second on the observer's clock, i.e. less time is seen to pass by the observer, the clock slows.

But if the moving clock has 'ticked' (one second ticks) 1.67 times for each of the observer's seconds then is it not ticking faster in the observer's frame of reference?

And, in the infamous Twin Paradox, would the traveling twin not have aged 16.7 years in his frame of reference, while only 10 had passed for his sedentary sibling? And would that sibling not, therefore, see his brother aging faster than himself?

 

[dianaj]

So 1.67 seconds on the moving clock is equivalent to 1 second on the observer's clock, i.e. less time is seen to pass by the observer, the clock slows.

You have

$$t' \cdot 1.67 = t$$

and so

$$t' = 1s \Rightarrow t = 1.67 s$$

The time passed describes an event: the 'hand' on the clock moving from on place to another, one second passing. According to the observer this event event takes 1.67 sec's. But according to the moving clock it only takes 1 sec. Therefore time must go slower for the moving clock (after, say, 100 of these events, the moving clock will have aged 100 secs while the observer will have aged 167 secs).

It's a good idea to think of $$t$$ as $$\Delta t$$: as a time interval describing an event. Usually solves my problems, when I get confused.

 

[matheinste]

Let us take an example of time dilation; if we say that for a clock moving at 0.8c, for each second that passes and is indicated on that clock, how much time, (time dilated), will the stationary observer see pass?

In every case you need to make clear who is making the observation and which clock they are observing.

"for each second that passes and is indicated on that clock" as observed by who?

" how much time, (time dilated), will the stationary observer see pass?" on which clock?

Matheinste.

 

[Grimble]

You have

$$t' \cdot 1.67 = t$$

and so

$$t' = 1s \Rightarrow t = 1.67 s$$

Yes but, t' has been transformed and is in co-ordinate units and t, being in the inertial observer's frame of reference is in proper units.

So what we are saying in your first equation is that 1.67 co-ordinate seconds are equal in duration to 1 second of Proper time at the current velocity.


So really the proper time seconds (as seen by an observer traveling with the clock) have shrunk when transformed at 0.8c such that it takes 1.67 of them to have the same duration as 1 second proper time as measured from the observer's inertial frame of reference.

In every case you need to make clear who is making the observation and which clock they are observing.

OK, my friend, I will restate it:
Let us take an example of time dilation; we will take a clock in an inertial reference frame, moving at 0.8c relative to an observer in another inertial reference frame. Then for each second that passes in the moving clock's reference frame, how much time, (which will be time dilated), will the remote observer see pass from his reference frame?

"for each second that passes and is indicated on that clock" as observed by who?

As would be seen by a local observer in the clock's reference frame.

" how much time, (time dilated), will the stationary observer see pass?" on which clock?

A very good question, for the moving clock observed in its own reference frame (i.e. by a local observer) will be displaying 1 sec Proper time. And the remote observer's clock (relative to which the moving clock is traveling at 0.8c) will also be shewing 1 sec (and it will also be displaying Proper time). The dilated time is the moving clock's proper time transformed into co-ordinate time in the remote observer's frame of reference. But his clock will be displaying HIS local (proper) time.

So you seem to have highlighted another conundrum, on which clock could the transformed time be displayed? One might think that it would be the moving clock as observed by the remote observer, but, if it were a clock with hands (as is often supposed) then how could the remote observer read anything but the position of the hands? It is after all the units of time that have changed (from proper time to coordinate time) not the hands of the clock.
Maybe the observer has another clock set to co-ordinate time?

Or is the difference in time exactly that: a difference in the units of time and the clocks would all read the same but be measuring different units of time.
That is to say both observer's read 1 second on the moving clock, but the 1 second the observer moving with the clock reads (proper time), is equal in duration to 1.67 of the second the remote observer reads (co-ordinate time).

 

[matheinste]

OK my friend, I will restate it:
Let us take an example of time dilation; we will take a clock in an inertial reference frame, moving at 0.8c relative to an observer in another inertial reference frame. Then for each second that passes in the moving clock's reference frame, how much time, (which will be time dilated), will the remote observer see pass from his reference frame?


As would be seen by a local observer in the clock's reference frame.

A very good question, for the moving clock observed in its own reference frame (i.e. by a local observer) will be displaying 1 sec Proper time. And the remote observer's clock (relative to which the moving clock is traveling at 0.8c) will also be shewing 1 sec (and it will also be displaying Proper time). The dilated time is the moving clock's proper time transformed into co-ordinate time in the remote observer's frame of reference. But his clock will be displaying HIS local (proper) time.

So you seem to have highlighted another conundrum, on which clock could the transformed time be displayed? One might think that it would be the moving clock as observed by the remote observer, but, if it were a clock with hands (as is often supposed) then how could the remote observer read anything but the position of the hands? It is after all the units of time that have changed (from proper time to coordinate time) not the hands of the clock.
Maybe the observer has another clock set to co-ordinate time?

Or is the difference in time exactly that: a difference in the units of time and the clocks would all read the same but be measuring different units of time.
That is to say both observer's read 1 second on the moving clock, but the 1 second the observer moving with the clock reads (proper time), is equal in duration to 1.67 of the second the remote observer reads (co-ordinate time).

First, remeber that there is no "moving" or "stationary" clock, just two clocks moving relative to each other. As they are at rest in inertial frames they can each, for the sake of simplicity consider themsemselves to be at rest and the other moving.

Next, you must take into account the relativity of simultaneity.

Let the two clocks be colocated at the origin and there be set to both read zero (as your transformation equations imply). They will of course at this point read zero simultaneouly for both observers. However, the time when the clock of the observer who considers himself at rest reads one second IS NOT simultaneous , in his own frame, with the time when he observes the other clock read 1 second. The time of greater than 1 second shown on the clock of the observer at rest IS simultaneous, in his frame, with the time when he observes the moving clock reading 1 second. That is, to the observer at rest the other clock appears to be running slow. Of course the reciprocal case also applies.

Matheinste.

 

[Grimble]

First, remeber that there is no "moving" or "stationary" clock, just two clocks moving relative to each other. As they are at rest in inertial frames they can each, for the sake of simplicity consider themsemselves to be at rest and the other moving.

Yes, exactly, that confirms my understanding. Thank you.

Next, you must take into account the relativity of simultaneity.

Let the two clocks be colocated at the origin and there be set to both read zero (as your transformation equations imply). They will of course at this point read zero simultaneouly for both observers. However, the time when the clock of the observer who considers himself at rest reads one second IS NOT simultaneous , in his own frame, with the time when he observes the other clock read 1 second.

And this too is exactly my understanding, for the other clock's time has been transformed.

The time of greater than 1 second shown on the clock of the observer at rest IS simultaneous, in his frame, with the time when he observes the moving clock reading 1 second. That is, to the observer at rest the other clock appears to be running slow. Of course the reciprocal case also applies.

Matheinste.

This last part is what Einstein described in the following section of his paper: “Relativity: The Special and the General Theory”.
In chapter XII. “The Behaviour of Measuring-Rods and Clocks in Motion” he writes:

Let us now consider a seconds-clock which is permanently situated at the origin (x' = 0) of K'. t' = 0 and t' = 1 are two successive ticks of this clock. The first and fourth equations of the Lorentz transformation give for these two ticks:
t = 0
and

As judged from K, the clock is moving with the velocity v; as judged from this reference-body, the time which elapses between two strokes of the clock is not one second, but

seconds, i.e. a somewhat larger time. As a consequence of its motion the clock goes more slowly than when at rest.

But if t is the time shewn on the observer's clock then it is in the (proper?) seconds of an inertial frame of reference.
Whereas the

is surely the equivalent time in transformed (co-ordinate?) seconds.
So one second on the observer's clock would be simultaneous with

transformed seconds on the other clock.
And the transformed seconds are smaller (contracted) and consequently they pass quicker and so the other clock will be seen to be speeded up, not slowed?

And is not what is shewn on Minkowski diagrams?

And thank you once again for all your help,

Grimble.

 

[matheinste]

But if t is the time shewn on the observer's clock then it is in the (proper?) seconds of an inertial frame of reference.
Whereas the

is surely the equivalent time in transformed (co-ordinate?) seconds.
So one second on the observer's clock would be simultaneous with

transformed seconds on the other clock.
And the transformed seconds are smaller (contracted) and consequently they pass quicker and so the other clock will be seen to be speeded up, not slowed?


Grimble.

Hello Grimble.

The last part should read

-----So one second on the observer's clock would be simultaneous with //www.bartleby.com/173/M5.GIF[/PLAIN] transformed seconds on the other clock.
And the transformed seconds are LONGER. -----

If you interchange observers you get exactly the same result.

We have ${t}^{'} = \gamma t$ and so the transformed seconds are greater in length than the "stationary" observer's seconds. Then for the inverse transform cosidering the frame in which the other observer is "stationary" and ${t}^{'}$ is his time intrerval we have ${t}= \gamma {t}^{'}$

Where

$$\gamma =$$

and is $\geq 1$

Matheinste

 

[Grimble]

Hello again matheinste and apologies for the delay.

The last part should read

-----So one second on the observer's clock would be simultaneous with

transformed seconds on the other clock.
And the transformed seconds are LONGER. -----

I'm sorry, but how can the transformed seconds be longer, when one second on the observer's clock would be simultaneous with

transformed seconds on the other clock?
For

is >1

If you interchange observers you get exactly the same result.

Agreed.

We have ${t}^{'} = \gamma t$

But surely, the formula that Einstein has derived:

when ${t^'} = 1$ gives us ${t} = \gamma {t^'}$

and so the transformed seconds are greater in length than the "stationary" observer's seconds.

No, they must be shorter in length.

Then for the inverse transform cosidering the frame in which the other observer is "stationary" and ${t}^{'}$ is his time intrerval we have ${t}= \gamma {t}^{'}$

Where

$$\gamma =$$

and is $\geq 1$

Matheinste

I'm sorry and mean no disrespect to you, but I think that while changing labels to shew reciprocality is fine if those labels are arbitrary, it is inadvisable to do so where the labels have been given particular meanings.
In this case $t^{'}$ was appropriated by Einstein to denote the transformed co-ordinates. But that is purely my own view.

Grimble.

 

[matheinste]

Hello Grimble.

I really have nothing to add.

Matheinste

 

[PeterDonis]

Grimble:

Try writing down the full coordinates for each event of interest, in each frame. Einstein didn't do that in his book (at least I don't remember him doing so), but it might help, since it will show explicitly how the formulas work out.

We have a clock at the origin of the system K'. Two successive ticks of that clock have coordinates, in K', of:

Tick 1: x' = 0, t' = 0.

Tick 2: x' = 0, t' = 1.

Now transform into the system K. The two events now have coordinates:

Tick 1: x = 0, t = 0 (by definition; this is where the origins of the two systems cross).

Tick 2:
$$x = \gamma \left( x' + v t' \right) = \gamma v$$,
$$t = \gamma \left( t' + v x' \right) = \gamma$$.

So a time interval that "looks like" 1 in system K', "looks like" $\gamma$ in system K. We can interpret this as saying that the clock at rest in K' is "running slow" with respect to K, because viewed from K, the time between two successive ticks of the clock at rest in K' is $\gamma$ instead of 1.

 

[Bob S]

Here is a very specific real-world test. BNL (Brookhaven Nat. Lab.) physicists stored muons with γ=29.4 in a circular ring. The muon's lifetime at rest is about 2.2 microseconds. In the ring, their lifetime was about 65 microsecons in the lab reference frame.
Bob S

 

[Rasalhague]

It all depends on what one is comparing, number of units or size of units.

Be aware that many (most?) texts take the word dilation in the opposite sense to Matheinste’s “Time dilation refers to the lengthening of the distance between ticks, compared with your clock, of a clock moving relative to you as observed by you.” For example, in Spacetime Physics (p. 66, problem 10), Taylor and Wheeler explicitly state that by time dilation they mean an increase in the number of seconds: “This time lapse is more than one meter of light-travel time. Such lengthening is called time dilation. To dilate means to stretch.” Presumably everyone who presents the relation in the form

$$\Delta t' = \gamma \Delta t$$

and calls this “time dilation” is going by Taylor and Wheeler’s interpretation. Of course, this is just a matter of words.

Let us take an example of time dilation; if we say that for a clock moving at 0.8c, for each second that passes and is indicated on that clock, how much time, (time dilated), will the stationary observer see pass?

Now if we put the appropriate figures into the formula for time dilation $t = \gamma {t^'}$
where:
t = the proper time for the inertial observer.
v = 0.8c
t' = co-ordinate time in the moving clock's frame
[tex]\gamma = \frac {1}{\sqrt{1 - {\frac{v^2}{c^2}}}} = 1.67[/itex]

then [itex] t = 1.67 {t^'}

So 1.67 seconds on the moving clock is equivalent to 1 second on the observer's clock, i.e. less time is seen to pass by the observer, the clock slows.

But if the moving clock has 'ticked' (one second ticks) 1.67 times for each of the observer's seconds then is it not ticking faster in the observer's frame of reference?

You identify t as "the proper time of the inertial observer" and t' as "co-ordinate time in the moving clock's frame", but our input--the information we actually have (this one second)--is the time between the clock's ticks in the clock's rest frame.

The time between the clock's ticks in the clock's rest frame is the proper time between these events. That's the co-ordinate time between them in the clock's rest frame. Proper time is co-ordinate time between events in a frame where they happen in the same place.

The value we want to calculate (our output) is the co-ordinate time between ticks with respect to an inertial frame in which the clock is moving at 0.8c, so we need to multiply one second by gamma to find the (longer) amount of time that will have passed between ticks in the frame where the clock is moving, namely 5/3 = 1.667 seconds for every tick of the moving clock. The proper time between two events is always shorter than the co-ordinate time between them in a frame where they don't happen in the same place.

Equivalently, we could refer to the time we're trying to calculate as the proper time between two events, one of which is simultaneous, in the frame where the clock is moving at 0.8c, with one tick of the clock, and the other of which is simultaneous, in the frame where the clock is moving at 0.8c, with the next tick of the clock. There's no paradox because in the clock's rest frame (the frame where it isn't moving)--where these two events happen in different places--even if we arrange for the first event to be simultaneous with a tick of the clock, the other won't be simultaneous with the next tick of the clock but rather will still lie in the future when the clock shows one second.

 

[matheinste]

Be aware that many (most?) texts take the word dilation in the opposite sense to Matheinste’s “Time dilation refers to the lengthening of the distance between ticks, compared with your clock, of a clock moving relative to you as observed by you.” For example, in Spacetime Physics (p. 66, problem 10), Taylor and Wheeler explicitly state that by time dilation they mean an increase in the number of seconds: “This time lapse is more than one meter of might-travel time. Such lengthening is called time dilation. To dilate means to stretch.” Presumably everyone who presents the relation in the form

$$\Delta t' = \gamma \Delta t$$

and calls this “time dilation” is going by Taylor and Wheeler’s interpretation. Of course, this is just a matter of words.

I am sure that ALL authors agree on their use of the term "time dilation". If I appear to use it in the opposite sense then it is wrong of me to do so and my explanation of what I believe them to be saying is flawed. Perhaps one source of some confusion for you may be my use of the words "distance between ticks". This does not refer to the spatial distance traveled by light between ticks.

Matheinste.

 

[Rasalhague]

Perhaps one source of some confusion for you may be my use of the words "distance between ticks". This does not refer to the spatial distance traveled by light between ticks.

I gathered that you meant the temporal distance (amount of time) between ticks.

Incidentally, Taylor and Wheeler refer to "light-travel time" in that quote just to clarify the significance of their use of meters (rather than seconds) as a unit of time.

 

[matheinste]

I gathered that you meant the temporal distance (amount of time) between ticks.

Incidentally, Taylor and Wheeler refer to "light-travel time" in that quote just to clarify the significance of their use of meters (rather than seconds) as a unit of time.

Just for clarification. By time dilation I mean the effect whereby an observer at rest with respect to inertial frame looking at his own clock and observing it give its first tick has to wait before he observes a clock, at rest in a reference frame moving inertially withh respect to him, give its first tick, both clocks having been set to zero when they are colocated in passing and light travel time having been allowed for. The "stationary" observer describes the "moving" clock's time as dilated. If this is in disagreement with the normal use of the term I apologise and will go back to basics.

Matheinste.

 

[Rasalhague]

Just for clarification. By time dilation I mean the effect whereby an observer at rest with respect to inertial frame looking at his own clock and observing it give its first tick has to wait before he observes a clock, at rest in a reference frame moving inertially withh respect to him, give its first tick, both clocks having been set to zero when they are colocated in passing and light travel time having been allowed for. The "stationary" observer describes the "moving" clock's time as dilated. If this is in disagreement with the normal use of the term I apologise and will go back to basics.

If the expression time dilation appeared in isolation, my intuition could go either way, but because it usually appears as a verbal label for the formula

$$\Delta t' = \Delta t * \gamma$$

and because time dilation so often appears alongside "length contraction", typically presented as the inverse of this--which is to say, a formula for deriving a smaller number from a bigger one--I assumed dilation must refer the opposite process, that of deriving a bigger number from a smaller one; otherwise, why not give the same name to the same process? In Spacetime Physics, Taylor and Wheeler clearly take dilation to mean obtaining a bigger number. Tipler and Mosca likewise: "The time interval measured in any other reference frame is always longer than the proper time. This expansion is called time dilation" (Physics for Scientists and Engineers: 5th ed., extended version, p. 1272).

Similarly Lerner: "The interval delta t_0, read by an observer with respect to whom the clock is at rest, is called the proper time. The interval delta t_v is called the dilated time" (Modern Physics for Scientists and Engineers, p. 1053).

http://books.google.co.uk/books?id=Nv5GAyAdijoC&pg=PA1053#v=onepage&q=&f=false

Similarly Schröder: "In the lab system, one measures a dilated time interval for the half life: t' = t gamma" (Special Relativity, p. 42).

http://books.google.co.uk/books?id=sLQ1rSNUjYAC&pg=PA42#v=onepage&q=&f=false

Similarly Petkov: "He projects the event A onto the event A' and finds that the time component [...] is greater than t [...] In S the clock worldline lies along the time axis and has only a time component ('height'); that is why the S-observer measures the proper length of the clock worldline, which we called proper time. In S' the worldline of the clock at rest in S is inclined and thus has both temporal and spatial components. That is why the S' observer measures an apparent or dilated time" (Relativity and the Nature of Spacetime, p. 88).

http://books.google.co.uk/books?id=ZA-yvXu40e0C&pg=PA88#v=onepage&q=&f=false

Fishbane et al. might be interpreting it the other way around: "Time T' is greater than T by a factor of gamma. The observer in frame F' sees longer ticks for the clock; in other words, the moving clock is slower by a factor of gamma. This effect is known as time dilation" (Physics for Scientists and Engineers, 2nd ed., extended, p. 1084). But this looks ambiguous to me. What exactly they're thinking of as being dilated depends on what "this" refers to: the fact of T' being "greater than" T, or the fact of there being "longer ticks" (causing T to be less than T').

Lawden apparenty takes dilation in the opposite sense to Taylor and Wheeler:

$$\Delta t' = \frac{\Delta t}{\gamma}$$

"This equation shows that the clock moving with O' will appear from S to have its rate reduced by a factor gamma. This is the time dilation effect" (An Introduction to Tensor Calculus, Relativity and Cosmology, 3rd ed., p. 13.)

In Simple Nature, Benjamin Crowel doesn't address the issue directly, as far as I can see, but one sentence might suggest that he takes dilation and contraction as synonymous in this context, contra Taylor and Wheeler:

"length contraction occurs in the same proportion as time dilation"
http://www.lightandmatter.com/html_books/0sn/ch07/ch07.html

Online sources are divided on the matter. I'm not sure if there's a tendency either way among the reputable ones. For example, here's one in agreement with Taylor and Wheeler:

"That equation tells me that if the passengers on the train measure so many seconds between two events, then I will measure a larger number of seconds between the same events. That's what it means to say that the train's clock counts dilated time."

http://bado-shanai.net/Map of Physics/moptimedil.htm

And here's one against:

"A clock in a moving frame will be seen to be running slow, or dilated according to the Lorentz transformation."
http://hyperphysics.phy-astr.gsu.edu/hbase/relativ/tdil.html

The Wikipedia article "Time dilation" also conceives of dilation in the opposite way to Taylor and Wheeler, e.g. "Symmetric time dilation occurs with respect to temporal coordinate systems set up in this manner. It is an effect where another clock is being viewed as running slowly by an observer. Observers do not consider their own clock time to be time-dilated, but may find that it is observed to be time-dilated in another coordinate system." And "as observed from the point of view of either of two clocks which are in motion with respect to each other, it will be the other clock that is time dilated."

http://en.wikipedia.org/wiki/Time_dilation

In psychology, subjective time dilation can refer to the perception of more time passing than is shown by a physical clock. A completely different phenomenon, of course, nothing to do with relativity, but perhaps this is what people new to relativity instinctively think of when they first encounter the term time dilation. For example, in this paper, the "dilation" of durations is synonymous with the perception of time passing slower than it would normally (which could be thought of as a greater number of subjective time units passing than physical time units as measured by a clock, i.e. a longer/expanded/dilated subjective time compared to clock time), but because we naturally take our subjective sense of time as the standard, we tend to think of such effects as being like a clock slowing down, rather than our minds speeding up, and for that reason perhaps the word dilation connotes slowing down.

http://www.plosone.org/article/info:doi/10.1371/journal.pone.0001264

(The actual psychological process, as the paper discusses, is more subtle than that, since not all time-dependent perceptions are affected in the same way.)

 

[matheinste]

Hello again Rasallhague,

I think we are agreed that dilation means becoming larger. But what becomes larger is not the number of ticks of a clock but the relative temporal length between ticks. Take two observers, A and B, in relative inertial movement with respect to each other. A will observe B's clock running slow compared to his own. B will observe A's clock running slow compared to his own. If A counts a number of ticks on his own clock then he observes LESS ticks on B's clock. This is because one second as seen by A on his own clock occupies less than one second on B's clock as seen by A. And vice versa. If that intrerpretation is wrong then I have a serious problem with my basic understanding of relativity because all texts that I have read lead me to this interpretation and none of the knowledgeable regulars on the forum have pulled me up on it.

You give many examples of authors and what they say. When interpreted correctly, whether agreeing with me or not, I am sure that all authors of relativity texts are saying the same thing. There is no room for disagreement among them on something so basic which has consequences for any further study of the subject (including the twins). We are of course not talking about psychological or subjective time.

Matheinste.

 

[Grimble]

Here is a very specific real-world test. BNL (Brookhaven Nat. Lab.) physicists stored muons with γ=29.4 in a circular ring. The muon's lifetime at rest is about 2.2 microseconds. In the ring, their lifetime was about 65 microsecons in the lab reference frame.
Bob S

Very interesting, I was unaware of this experiment; but it is good to know that it supports my contention.

For this demonstrates the correctness of t = γt'

i.e. that 2.2 proper microseconds = 'about' 65 co-ordinate microseconds

where the muon's lifetime at rest, 2.2 microseconds (poper time) is equavalent to the muon's lifetime when moving, 'about' 65 microseconds (co-ordinate time), according to the Lorentz transformation.

 

[Rasalhague]

Hello again Rasallhague,

I think we are agreed that dilation means becoming larger. But what becomes larger is not the number of ticks of a clock but the relative temporal length between ticks. Take two observers, A and B, in relative inertial movement with respect to each other. A will observe B's clock running slow compared to his own. B will observe A's clock running slow compared to his own. If A counts a number of ticks on his own clock then he observes LESS ticks on B's clock. This is because one second as seen by A on his own clock occupies less than one second on B's clock as seen by A. And vice versa. If that intrerpretation is wrong then I have a serious problem with my basic understanding of relativity because all texts that I have read lead me to this interpretation and none of the knowledgeable regulars on the forum have pulled me up on it.

You give many examples of authors and what they say. When interpreted correctly, whether agreeing with me or not, I am sure that all authors of relativity texts are saying the same thing. There is no room for disagreement among them on something so basic which has consequences for any further study of the subject (including the twins). We are of course not talking about psychological or subjective time.

Hi there, Matheinste,

It looks to me as though there could well be some disagreement among those quotes, not on the facts of special relativity, of course, just on what exactly they take "dilation" to refer to, or how they conceptualise it. I may be mistaken and the difference only apparent, but it does seem that "dilated time" can mean a bigger number to one author, and a smaller number to another author. I'd be interested to know what our resident experts think on the matter too. Mostly the textbooks, in particular Schröder, Lerner and Petkov who contrast "dilated time" with "proper time", take the dilated thing to be the bigger number. But quotes like Lawden's or the website that said, "A clock in a moving frame will be seen to be running slow, or dilated according to the Lorentz transformation", or the Wikipedia article do appear to be taking it the opposite way, and consider a "dilated time" to be the smaller number shown by a moving clock whose units have been dilated (made bigger) in comparison to a stationary clock. It would be nice if they did agree, and the more authorative sources we've found so far do mostly lean towards the bigger number interpretation, but I wouldn't be surprised if some equally reputable sources thought of it the opposite way round (a bit like the way some authors take their space coordinates as negative and their time coordinate as positive, while others take time as positive and space negative, or like the way some authors use phi and theta the opposite way round to others to label the angles in spherical polar coordinates).

I only mentioned the psychological usage to compare how the similar issue of comparing two rates of time is treated there, and because I thought perhaps it might offer a clue as to how we insinctively think about such things: hence what people are most likely to think time dilation refers to when they first hear of it in the context of relativity.

Then again, when you say, "But what becomes larger is not the number of ticks of a clock but the relative temporal length between ticks", perhaps there isn't such a contradiction; after all, how do we express the idea that the duration (time interval) between the ticks of one clock is longer than the corresponding duration as measured by another clock if not by counting the greater number of ticks on the other clock?

 

[matheinste]

Hello Rasallhague.

I see that there may have been some initial confusion (caused by me) over different uses of the primed and unprimed symbols for the "moving" and "stationary" observers/frames. I have lately in this thread tried to avoid their use. However, the outcome is the same, and, apart from perhaps a different use of this primed/unprimed symbology I am sure that the term dilation is consistently used by authors to mean the loosely described effect of "moving clocks running slow".

Matheinste.

 

[Rasalhague]

I am sure that the term dilation is consistently used by authors to mean the loosely described effect of "moving clocks running slow".

So is it the standard interpretation to think of the time "to be dilated" as being the interval shown on the moving clock, and the result of its dilation--the time "after having been dilated" (the "dilated time" of Lerner, Petkov and Schröder)--as being the length of this interval as recorded by the stationary clock? (This "dilated time" being a bigger number than the proper time between the events that mark the beginning and end of the interval on the moving clock's worldline.)

 

[matheinste]

So is it the standard interpretation to think of the time "to be dilated" as being the interval shown on the moving clock, and the result of its dilation--the time "after having been dilated" (the "dilated time" of Lerner, Petkov and Schröder)--as being the length of this interval as recorded by the stationary clock? (This "dilated time" being a bigger number than the proper time between the events that mark the beginning and end of the interval on the moving clock's worldline.)

That may well be but the wording is still confusing to me but that is not your fault.

Let me try to explain what I am saying in detail. The detailed explanation does not imply any lack of knowledge on your part but is just to avoid any confusion as to my point of view. If we let A be an observer with a clock at rest in an inertial frame of reference and let B be an observer with a clock at rest in a frame moving inertially relative A's frame. Let them pass each other at some point and let them set their clocks to read zero at this point. Now in A's frame, the event of A's clock reading 1 second as viewed by A is simultaneous with the event of B's clock reading less than 1 second as viewed by B. This assumes that the standard definition of simultaneity is used. So according to A, B's time is dilated in the sense that B's seconds are "stretched". Both are of course reading their own proper times on their own clocks and for each of them this is THE time. This may well not be the standard way of explaining time dilation but it is another way of saying the same thing, I hope.

Matheinste.

 

[Rasalhague]

Let me try to explain what I am saying in detail. The detailed explanation does not imply any lack of knowledge on your part but is just to avoid any confusion as to my point of view. [...] This may well not be the standard way of explaining time dilation but it is another way of saying the same thing, I hope.

That's clear. You explain well. It's certainly the same situation that I was struggling to describe. I suppose what I was trying to get at with my contorted wording was that we're given one value, and use the "time dilation formula" to derive a bigger value from it. Something small becomes something big. We could say that the time between ticks is dilated (bigger, stretched) when measured according to a frame where the clock is moving. But maybe others see the word dilation is a slightly different way, for the situation in general (irrespective of whether a big number is being derived from a smaller one or vice versa), and maybe that's what confused me about the use of the same term to refer to the application of the inverse formula, for which I'd have thought "time contraction" would have been a natural name.

 

[matheinste]

That's clear. You explain well. It's certainly the same situation that I was struggling to describe. I suppose what I was trying to get at with my contorted wording was that we're given one value, and use the "time dilation formula" to derive a bigger value from it. Something small becomes something big. We could say that the time between ticks is dilated (bigger, stretched) when measured according to a frame where the clock is moving. But maybe others see the word dilation is a slightly different way, for the situation in general (irrespective of whether a big number is being derived from a smaller one or vice versa), and maybe that's what confused me about the use of the same term to refer to the application of the inverse formula, for which I'd have thought "time contraction" would have been a natural name.

Its still not quite clear to me as to whether we are saying the same thing. I'll try putting it another, very non rigorous way. If you are happy with length contraction think of it as a contraction, or making smaller, of the unit of measurement, the meter. In the same way time dilation can be thought of as making longer the unit of time measurement, the second.

Matheinste

 

[Rasalhague]

Its still not quite clear to me as to whether we are saying the same thing. I'll try putting it another, very non rigorous way. If you are happy with length contraction think of it as a contraction, or making smaller, of the unit of measurement, the meter. In the same way time dilation can be thought of as making longer the unit of time measurement, the second.

Makes sense. Maybe I was wrong to think that these were two different ways of conceptualising it. I think what you say agrees with Tipler & Mosca's definition, doesn't it? "The time interval measured in any other reference frame is always longer than the proper time. This expansion is called time dilation." That just generalises your statement about the second to intervals of any length.

 

[Rasalhague]

Its still not quite clear to me as to whether we are saying the same thing.

Perhaps the difference in my mind could be visualised in the following way.

I was thinking of one image in which we have one object like a rubber tape measure of time that we stretch (dilate) so that one of its seconds covers gamma seconds of any unstretched tape measure we might care to compare it with; rightly or wrongly, this seemed to me more like how you were understanding the word dilation. I was contrasting this in my mind with another image in which we're comparing two objects: a short time (consisting of a small number of seconds) and a longer (dilated) time made up of gamma times the shorter number of seconds.

Of course, this amounts to the same thing mathematically, and such visualisations are just a mental short-hand to account for the word "dilation". They don't clarify the real physical situation in all its aspects as well as a spacetime diagram. But they might affect how an author writes about the subject. In particular, if we think in terms of the first image, it might seem natural to call the shorter period the "dilated time" (as the Hyperphysics site and Wikipedia come close to doing), whereas if we think in terms of the second image, it might seem more natural to call the longer period the "dilated time" (as Lerner, Petkov and Schröder do explicitly). So perhaps there is a significant difference in these interpretations after all...

 

[JesseM]

It looks to me as though there could well be some disagreement among those quotes, not on the facts of special relativity, of course, just on what exactly they take "dilation" to refer to, or how they conceptualise it. I may be mistaken and the difference only apparent, but it does seem that "dilated time" can mean a bigger number to one author, and a smaller number to another author. I'd be interested to know what our resident experts think on the matter too. Mostly the textbooks, in particular Schröder, Lerner and Petkov who contrast "dilated time" with "proper time", take the dilated thing to be the bigger number. But quotes like Lawden's or the website that said, "A clock in a moving frame will be seen to be running slow, or dilated according to the Lorentz transformation",

I don't see that as contradicting the Schroder/Lerner/Petkov usage--"running slow" refers to the rate of ticking rather than the time interval, but obviously a clock that is running at a slowed-down rate (in our frame) will take a greater length of time to tick forward by a given amount.

or the Wikipedia article do appear to be taking it the opposite way, and consider a "dilated time" to be the smaller number shown by a moving clock whose units have been dilated (made bigger) in comparison to a stationary clock.

What specific quote in what Wikipedia article do you refer to? They seem to use "dilated" in the normal way in the second-to-last paragraph of the overview section of the time dilation article, where they write: "Thus, in special relativity, the time dilation effect is reciprocal: as observed from the point of view of either of two clocks which are in motion with respect to each other, it will be the other clock that is time dilated." From either clock's "point of view" (rest frame), when they it is "the other clock that is time dilated", presumably they mean that it takes longer to tick forward by a given amount (its seconds are longer).

 

[Rasalhague]

> Mostly the textbooks, in particular Schröder, Lerner and Petkov who contrast "dilated time" with "proper time", take the dilated thing to be the bigger number. But quotes like Lawden's or the website that said, "A clock in a moving frame will be seen to be running slow, or dilated according to the Lorentz transformation",

I don't see that as contradicting the Schroder/Lerner/Petkov usage--"running slow" refers to the rate of ticking rather than the time interval, but obviously a clock that is running at a slowed-down rate (in our frame) will take a greater length of time to tick forward by a given amount.

Quite, but it didn't seem at all obvious to me that people would chose to call a time-dilated clock one that displays non-dilated time, while dilated time is shown only by a clock that isn't time-dilated!

What specific quote in what Wikipedia article do you refer to? They seem to use "dilated" in the normal way in the second-to-last paragraph of the overview section of the time dilation article, where they write: "Thus, in special relativity, the time dilation effect is reciprocal: as observed from the point of view of either of two clocks which are in motion with respect to each other, it will be the other clock that is time dilated." From either clock's "point of view" (rest frame), when they it is "the other clock that is time dilated", presumably they mean that it takes longer to tick forward by a given amount (its seconds are longer).

This paragraph suggests that a "clock that is time dilated" would be a clock that records a smaller number because, as you say, it takes longer to tick a given amount (the amount being given by the observer's clock). That was also the impression I got from the statement "Observers do not consider their own clock time to be time-dilated, but may find that it is observed to be time-dilated in another coordinate system." It seems natural to think that the time displayed by a "time-dilated clock" would be called "dilated time", no? Well it did to me... But Lerner, Petkov and Schröder all define "dilated time" in the opposite way, as the bigger number recorded by the other clock.

 

[Rasalhague]

it didn't seem at all obvious to me that people would chose to call a time-dilated clock one that displays non-dilated time, while dilated time is shown only by a clock that isn't time-dilated!

Although, I suppose that would fit with the way that we talk of a contracted ruler whose contracted length is shown by the smaller-numbered marks on another, uncontracted ruler.