# Time Dilation and different clock types

1. Oct 26, 2015

### DAC

Hello PF.
A moving clock is seen by the platform observer to run slow. This applies to all clock constructions. A light clock runs slow because the light path lengthens and the clock takes longer to tick over.

Other types of clocks don't have a light path to lengthen. By what mechanism do they run slow?

If the answer is, there is no mechanism, time just slows and the clocks record that, then why does the light clock not only slow but also shows the mechanism by which it slows i.e. longer path?
Regards.

2. Oct 26, 2015

### Staff: Mentor

By what mechanism is the hypotenuse of a right triangle longer than the side?

3. Oct 26, 2015

### Staff: Mentor

That's not really a "mechanism", that's a skewed observation due to the relativistic effect (it has no effect locally). Any other clock will show similar effects on whatever its particular mechanism of operation is: A pendulum will be observed swing back and forth "slower". A quartz crystal will be observed to vibrate "slower", etc. Point being, there really is no single, direct "mechanism" that you can pinpoint that you can say that speed slows clocks because___________.... unless in that blank is that time is relative.

4. Oct 26, 2015

### DAC

" The way it is ". Bruce Hornsby & the Range?

5. Oct 27, 2015

### SlowThinker

This also answers you original question.
If you come up with a better answer, it will still answer the original question as well.

6. Oct 27, 2015

### facenian

I think the problem is with the concept of time itself and not the clock mechanism.What is a good clock?¿What is an ideal clock? if two clocks show different times how do you know which one show the correct one?Wheeler et al said "time is defined so that motion is simple" and that's a good clue. However I think that time is defined when the constancy of light velocity is postulated and this very definition unites space and time.

7. Oct 27, 2015

### Staff: Mentor

If you are willing to accept that as a mechanism for Euclidean geometry then it should also be acceptable as a mechanism for Minkowski geometry.

8. Oct 27, 2015

### Staff: Mentor

An ideal clock is one who's tick rate is not affected by physical conditions in its environment, such as temperature, pressure, humidity, gravity, etc, as well as having a high precision and repeatability. The way you determine if a clock is better or worse than another is by identifying and quantifying the sources of error.

That description should sound a lot like how you evaluate pretty much any scientific instrument(s).

9. Oct 27, 2015

### Staff: Mentor

It's a bit of an irony to me: Time dilation in some ways seems harder for some people to accept than length contraction because the differences accumulate, are regularly used in real life with GPS, and thus are difficult to ignore...plus, people deal with perspective illusions (of trig) and accept them because they are illusions. But that doesn't mean people accept length contraction, it just means they tend to ignore it.

10. Oct 27, 2015

### facenian

The problem I see is that this entails a circular reasoning because, how do you evaluate the precision or error of the instrument(in this case clock)? I think we can do this with an atomic clock but then again you must suppose arbitrarily that the atomic clock is more precise, in fact the atomic(cesium) clock is really exact by definition!
However all this does not really answer why we take(arbitrarily) the atomic clock to be better than, for instance, the motion of the earth except for Wheeler's observation.

11. Oct 27, 2015

### Staff: Mentor

I don't think that's true - it doesn't really take into consideration what I said at all. Let's take a couple of specific examples:
A pendulum clock is highly impacted by gravity. In a stronger gravitational field it ticks faster as Newton's Laws of gravity and motion predict it must. In orbit, it wouldn't tick at all. This effect is well understood and predictable to a pretty high degree of precision. A quartz oscillator clock is not affected by gravity. It is therefore much more accurate in situations where varying g-acceoerat89hs are present and can be used to measure the inaccuracy of the pendulum clock and match the source of the inaccuracy to the known mechanism that causes it.

I see nothing at all circular about that. Perhaps what you are believing is that the "error" is first detected by comparison to a real clock and then later explained when a more "accurate" real clock is invented, thus the "error" is always referenced to the more accurate clock. That isn't true. The "error" is pre-calculated based on the physical limitations of the clock as an error vs the "ideal" clock, not vs any other real clock.

In short, all real clocks have known mechanisms that cause error. An ideal clock would not. The known mechanisms can't always be easily experimentally verified for the most accurate real clocks and in practice you use known more accurate real clocks to help evaluate less accurate real clocks, but that doesn't make the logic circular.

12. Oct 27, 2015

### Mister T

You build two of them and run them side-by-side to see if they stay in sync.

That's not at all true. Atomic clocks had to be shown to be more precise than rotations of planet Earth before metrologists accepted them as a better standard.

Atomic clocks are more precise than Earth rotations. That's the reason and no other. It's not arbitrary. If you used several atomic clocks to repeatedly measure the time for Earth rotations you would find there's a variation in your measurements. But the variations between the readings on the various atomic clocks would be less than the variations between the atomic clocks and the rotations of Earth.

13. Oct 27, 2015

### facenian

I think you've got a point here and I have to agree with you. You postulated the existence of an ideal time against which you can compare, by calculation, your measurements. I think this way of putting it can break the circularity I was referring to.
Maybe this discussion belongs more to the philosophy of science than to real physics.
By the way could you define easily what is that ideal time?

14. Oct 27, 2015

### Mister T

Maybe. But metrology is not philosophy. Read Post #12.

15. Oct 27, 2015

### facenian

I agree that in a practical sense it is not arbitrary to proclaim that atomic clocks are better that earth rotation.
However this discussion somehow is similar to the old argument about absolute motion against relative motion.
When you compare two clocks(atomic and earth) and find they do not coincide the only rigorous conclusion is just that, you can not tell which one is correct and which one is right, however we all "know" that atomic clocks are better and that's because we unconsciously apply Wheeler's argument(post #6)

16. Oct 27, 2015

### phinds

No, the clock APPEARS to run slow. Locally the clock ticks at exactly the same one second per second as does any other clock.

17. Oct 27, 2015

### Staff: Mentor

Our current standards use the constancy of the speed of light to define distance, not time. We define time, the second, based on a certain number of cycles of the radiation emitted in a certain atomic transition; then we define distance, the meter, such that the speed of light is exactly 299,792,458 meters per second.

Not really. The Earth's rotation looks simplest if we define time based on the Earth's rotation; it looks more complicated if we define time as we currently do, by an atomic standard--just look at how we have to deal with leap seconds and other such corrections.

The reason we say atomic clocks are better is that the clocks themselves are simpler, so it's easier to understand the time they keep as a simple consequence of fundamental physical constants. The Earth is a huge conglomeration of atoms of many different kinds with a complicated structure; obviously the rotation of this thing is going to be a complicated process. Trying to relate that complicated process to something simple like the fundamental constants of the universe is not going to be straightforward at all.

A single energy level transition in a single atom is much, much simpler, and much easier to relate to fundamental dimensionless constants. We want to relate our standards to fundamental dimensionless constants because they are the only physical constants that don't depend on the units we choose; in other words, they don't depend on how we choose to define the second or the meter or any other unit of time or distance (or anything else). For example, the key dimensionless constant that governs atomic energy level transitions is $\alpha$, the fine structure constant. By defining the second using an atomic energy level transition, we are basically using $\alpha$ to define our unit of time.

18. Oct 27, 2015

### Staff: Mentor

Yes. This issue is most certainly not philosophy.

19. Oct 27, 2015

### Staff: Mentor

I already did. It's the first thing you quoted me as saying!
Again: no. You know the atomic clock is better because you can calculate the error of the other clock (in this case, the Earth) and use the atomic clock to measure it. You cannot do the opposite: you cannot calculate the error of the atomic clock and then use the Earth to measure it.
There is nothing unconscious about it: one way, the laws of physics work and the other way, they don't.

20. Oct 27, 2015

### Mister T

It's not a simple comparison of one to the other in that way. You must take repeated measurements with each and compare them to each other.

Either you missed my point entirely or you chose to ignore it. It's an issue of precision. A synonym for precision is reproducibility. The results of measurements of time using atomic clocks are more reproducible than the measurements of time using Earth rotations. It has nothing whatever to do with any argument about motion. If pendulum clocks were more precise that's what the metrologists would use to define the standard.

Last edited: Oct 27, 2015