The Naked Scientists

The Naked Scientists Forum

Author Topic: Are atomic clocks directional?  (Read 2583 times)

Offline CliffordK

  • Neilep Level Member
  • ******
  • Posts: 6321
  • Thanked: 3 times
  • Site Moderator
    • View Profile
Are atomic clocks directional?
« on: 16/01/2012 04:48:13 »
Ok, my research has led me to try to learn a little about clocks.  And, thus Cesium, and other Atomic Clocks that are considered the gold standard today.

My question is.  Say you went to Ecuador, and set up 4 atomic clocks.
1 facing North, 1 facing South, 1 facing East, and 1 facing West.
Otherwise everything else is identical.

Of course, in Ecuador, North and South are level, whereas here, North and South are technically at about a 45 degree angle from level.

Anyway, for the four clocks, you compared their times averaged over:
  • 24 hr day
  • 23 hours, 56 minutes, 4.091 second Sidereal day
  • Long-Term Basis
Would they remain synchronized?

I think the fountain clocks point upwards...  maybe they all do, so then orientation would be less important, except that they are all upward facing.

So...  A little background.
Cesium (Caesium) clocks use Cs-133 (stable), rather than Cs-137 (radioactive), and have nothing to do with radioactivity.

Basically, I think they consist of several parts.

Tunable oscillator.
Microwave
Cesium gas.

The microwave then emits a wave (at 9,192,631,770 hz), which carries the correct energy to cause a hyperfine state change to the cesium atoms.  Anyway, it absorbs the energy, and can be detected, which is processed through a feedback loop which then is used to synchronize the microwave.

Anyway, I haven't seen anything indicating that an orientation change would affect the cesium clocks, although, I did see a short note about quartz clocks, although it isn't clear if this is directionality, or acceleration effects.

Crystal oscillators are sensitive to accelerations, even the gravitational field of the earth. A
standard test is to change the orientation with respect to up by 90 degrees. This is called a “tip
over test”. It usually changes the frequency by over 1 part in 108. This means crystal oscillators
are “noisy” in a vibrating environment unless well mechanically isolated.

I'm working on some graphs to demonstrate why I think the atomic clocks may be inherently relativistic, and will try to post them later.

Does anybody have some actual research links?
« Last Edit: 16/01/2012 04:50:01 by CliffordK »


 

Offline Geezer

  • Neilep Level Member
  • ******
  • Posts: 8328
  • "Vive la résistance!"
    • View Profile
Re: Are atomic clocks directional?
« Reply #1 on: 16/01/2012 06:45:28 »
No good links I'm afraid, but a bit more on how the clock works.

The oscillator is tunable, but it is extremely stable and it will "flywheel" for a very long time. There is no phase comparison between the phase of the microwave signal and the state changes in the cesium atoms. Adjustments to the oscillator are very small and they only happen infrequently.

The feedback loop is formed my measuring the photon emissions from the cesium fountain. When the microwave is running at the cesium hyperfine state change frequency, the fountain is at maximum "brightness". As the microwave frequency drifts from that frequency, the fountain get's a bit dimmer (probably not very much!)

As there is no detectable phase relationship between the cesium and the oscillator, it's not obvious how to determine whether the oscillator is running fast or slow. I presume they address this by dithering the oscillator very slightly and observing the effect on the amount of energy radiated by each burst of the fountain, but maybe they have a better way.

However, even although the method is convoluted, the clock is still using the frequency of cesium atomic activity to count time. So, if time determines that atomic activity (which it jolly well should) the clocks will be affected by the usual rules of time, motion and gravity.

 

Offline CliffordK

  • Neilep Level Member
  • ******
  • Posts: 6321
  • Thanked: 3 times
  • Site Moderator
    • View Profile
Re: Are atomic clocks directional?
« Reply #2 on: 16/01/2012 08:45:54 »
As far as I can tell, the whole thing is to measure the frequency of the microwave (known), to determine the optimum frequency to energize the Cesium atoms.

Here are a couple of drawings.  These are optimized for detail in 2 dimensions.  As far as I can tell, the actual clocks use an enclosed microwave generator, usually a Ramsey Microwave Cavity, but it is unclear if the microwaves generated are monodirectional, bidirectional, or omnidirectional.



If there is no relative movement in relation to the fabric of space, then the frequency emitted would be the same in all directions, and all energy levels emitted are the same.



However, if there is movement in space, for example to the right along the X axis, the microwaves emitted perpendicular to the movement are essentially unchanged.  However, those in the direction of the movement are all blue-shifted.  Thus, the frequency is higher, and the energy is higher, that is with respect to the fabric of space.

Those in the opposite direction of the movement are all red shifted.  Thus the frequency is lower, and the energy is lower with respect to the fabric of space.

What I'm having troubles visualizing is that the Cesium atoms will also be in the same frame as the microwave emitter.

So, the view of the Cesium atom should NOT be red/blue shifted, so it wouldn't seem that directionality would be important.

I suppose, this is where I'm stuck.  It would seem the most stable direction to point the microwave would be perpendicular to the direction of movement.  However, I'm not sure if this is actually necessary.

So, it would seem to me that on Earth, and Earth's orbit, it would be best to align the microwaves (assuming directional) either North or South from a tangent to the equator, but not East or West. 

Somebody has to have found a good place along the equator to test this question.
 

Offline Geezer

  • Neilep Level Member
  • ******
  • Posts: 8328
  • "Vive la résistance!"
    • View Profile
Re: Are atomic clocks directional?
« Reply #3 on: 16/01/2012 09:29:47 »
I think you have it a bit downside up. The microwave frequency is not known. It's the cesium transitions that are known to have maximum emission when bombarded with microwaves at that frequency. The oscillator is adjusted now and again to maximize the cesium emissions (wich are also only periodic).
 
 

Offline yor_on

  • Naked Science Forum GOD!
  • *******
  • Posts: 11993
  • Thanked: 4 times
  • (Ah, yes:) *a table is always good to hide under*
    • View Profile
Re: Are atomic clocks directional?
« Reply #4 on: 16/01/2012 14:46:26 »
As it is about clocks, and, as I think of light that way :)

A camera fast enough to watch light move?
 

Offline namaan

  • Sr. Member
  • ****
  • Posts: 195
    • View Profile
Re: Are atomic clocks directional?
« Reply #5 on: 16/01/2012 15:04:33 »
With "A camera" being a technically erroneous term of course. In fact, they didn't "capture film" at all, at least not in the traditional sense. It was a collection of cameras collecting snippets of time and then putting that information together to create a "movie". Just so we don't think camera makers have side-stepped Einstein ;)
 

Offline CliffordK

  • Neilep Level Member
  • ******
  • Posts: 6321
  • Thanked: 3 times
  • Site Moderator
    • View Profile
Re: Are atomic clocks directional?
« Reply #6 on: 17/01/2012 00:58:14 »
The camera is cool.
Of course, it isn't a single image, but the average of multiple events and images.  But, still spectacular nonetheless.

It might be a temporary belief - it may not; but most of the physics at present use this as a fundamental law.   Please bear in mind that SR  correctly predicts time dilation in orbiting satellites, is the basis of modern physics, and had been thoroughly tested - this is what you need to overturn.
I hope you don't mind cross-quoting.

But, this is precisely what I'm trying to do.  And, of course, coming up with testable hypotheses. 

Thus, the question of whether the atomic clocks are directional.  So, start with 4 atomic clocks orientied horizontally, N/S/E/W, and see if I could get the Central University of Quito Ecuador Physics Department to provide facilities for testing.  Just to see if I get the predicted time drift.

Unfortunately, I'm  not a "rocket scientist", so I'm trying to imagine how satellites work.  I presume that each one is put on an orbital plane, probably with respect to the Galaxy/Universe.  If in a polar LEO, then it will appear as if they are precessing, and perhaps they do a little, but for the most part they are orbiting in a fixed plane, and the Earth is spinning under them.  Now, if one induces a slight spin in the satellites, then they will appear to be phase-locked with Earth, just like the moon, and complete one complete revolution with every orbit.

So, to the GPS Satellites. 
My guess is that NASA came up with a good design.  Launched one, and found the clock drift.  Then they decided to fix the clock, and make dozens of clones of the satellite. 

I assume they still have some clock drift issues, but their primary drift seems to be independent of the plane the satellite is orbiting in.  Would they have oriented all the clocks in the same direction by using the same basic chassis (although I do see they have made a few different models so far)?  Presumably a satellite could be twisted at 90°, and still function.  However, it is likely that the solar tracking for the solar panels is all designed for the same orientation, so all the GPS satellites would be orbiting the Earth at about the same speed, same spin to sync with Earth, and thus be in the same orientation. 

Anyway, if the atomic clocks are directional.  Then one would predict speed related time drift.  However, if one oriented the clock to 90° to the satellites plane of motion, then it should be far more stable than if it is oriented in the direction of the satellites plane of motion.  And, of course, such a test would be pretty trivial.  Perhaps they already have the resources available on the International Space Station.

It would seem odd that the same error would be repeated by different countries, and manufacturers, but perhaps we are only hearing a few examples.
 

Offline Geezer

  • Neilep Level Member
  • ******
  • Posts: 8328
  • "Vive la résistance!"
    • View Profile
Re: Are atomic clocks directional?
« Reply #7 on: 17/01/2012 01:38:46 »
I'm probably missing something here (I know that is highly unusual, but it has happened.)

Your very nice diagrams don't seem to describe four clocks. It looks more like one clock with four cesium fountains.

If that is the case, there is no relative motion between the fountains and the microwave oscillator, so why would you expect to see any frequency shift?
 

Offline CliffordK

  • Neilep Level Member
  • ******
  • Posts: 6321
  • Thanked: 3 times
  • Site Moderator
    • View Profile
Re: Are atomic clocks directional?
« Reply #8 on: 17/01/2012 03:37:54 »
Right,
I've kind of simplified the diagram, of 4 possible orientations.

One often sees the clocks as having some kind of chamber that either looks like a cylinder, or a "U" where the microwaves are generated.

And, right, if the emitter, and the observer are in the same reference frame, then logically they should not see a phase shift.  The phase shift is only visible if comparing it to the reference frame of the fabric of space. 

Ok.
Perhaps I'm thinking of it wrong.
Say the microwave oscillator was relatively unaffected by speed and direction.  Then, one would have to figure out what speed does to the Cesium atom.

So, perhaps one could conclude:
The faster one goes, the LESS energy it takes to force the Cesium atom to change hyperfine states.  Thus, one compensates by feeding it a lower energy wavelength.

So, if you are counting off cycles, then you are creating a lower energy microwave at speed, then you would be needing at rest, and thus your clock would in fact be running slow.

Now, why would this make any sense?
The Cesium atom is carrying more Kinetic Energy.  Yes, with respect to the resting frame.  But, the faster one would go, the more kinetic energy one would get, up to the speed of light, by various calculations, the Kinetic Energy then approaches infinity.

So, what one might expect.
If one had a set of atoms that was say 50% low energy, and 50% high energy at rest.  Then, one would accelerate the atoms...  one might find the proportions would change, and one would get 40% low energy, 60% high energy.

So, my new theory of (relativistic) time.

(Relativistic) time is the Energy that it takes Cs-133 to make the transition from low energy to high energy which decreases with increasing velocity.

Time to sit back and think about that one.
 

Offline CliffordK

  • Neilep Level Member
  • ******
  • Posts: 6321
  • Thanked: 3 times
  • Site Moderator
    • View Profile
Re: Are atomic clocks directional?
« Reply #9 on: 17/01/2012 08:30:59 »
The oscillator is tunable, but it is extremely stable and it will "flywheel" for a very long time. There is no phase comparison between the phase of the microwave signal and the state changes in the cesium atoms. Adjustments to the oscillator are very small and they only happen infrequently.
(Relativistic) time is the Energy that it takes Cs-133 to make the transition from low energy to high energy which decreases with increasing velocity.

Time to sit back and think about that one.

Ok, I'm absolutely convinced that the speed of light is not the same in all directions.  But, I can also understand why it is so damned hard to measure the differences.

Now, one question is whether the adjustments made to the oscillators are completely random, or if they in fact have a pattern.  For example slowing them down when earth turns in the direction of the orbit around the sun, or through space.  For example, slowing down at midnight when the directions of spin and solar orbit are additive (faster movement), and speeding up at noon, when the directions subtract (slower movement).

But, in all likelihood, both the oscillator, and the cesium standard suffer from the same relativistic problems.
 

Offline Geezer

  • Neilep Level Member
  • ******
  • Posts: 8328
  • "Vive la résistance!"
    • View Profile
Re: Are atomic clocks directional?
« Reply #10 on: 18/01/2012 01:11:53 »

Ok, I'm absolutely convinced that the speed of light is not the same in all directions.  But, I can also understand why it is so damned hard to measure the differences.

Now, one question is whether the adjustments made to the oscillators are completely random, or if they in fact have a pattern.  For example slowing them down when earth turns in the direction of the orbit around the sun, or through space.  For example, slowing down at midnight when the directions of spin and solar orbit are additive (faster movement), and speeding up at noon, when the directions subtract (slower movement).

But, in all likelihood, both the oscillator, and the cesium standard suffer from the same relativistic problems.

The clock is not measuring the speed of light. It's observing time (I prefer to call it "counting events") at its location in spacetime. The stuff of which the oscillator is constructed, and the cesium atoms, are in the same relative positions within spacetime, so they cannot detect a difference in spacetime when none exists.

As the oscillator and the cesium atoms are both controlled by the same time, the only thing that needs to be done is to adjust the oscillator to keep time with the cesium atom events. That's really all the clock does (in a rather roundabout fashion).

EDIT: I said "adjust the oscillator to keep time with the cesium atom events". What I should really have said is

"adjust the oscillator so that it produces the same number of events as the number of events produced by the cesium atoms"
« Last Edit: 18/01/2012 01:45:49 by Geezer »
 

The Naked Scientists Forum

Re: Are atomic clocks directional?
« Reply #10 on: 18/01/2012 01:11:53 »

 

SMF 2.0.10 | SMF © 2015, Simple Machines
SMFAds for Free Forums