[Bored chemist]

It jolly well is measuring a frequency; specifically, the frequency of a microwave oscillator. That oscillator is locked to the absorbtion frequency of caesium atoms in a "fountain".
In principle, you could have the events that control the frequency of the oscillator happen as seldom as you like- once a day or whatever. In practice they happen very frequently, but that's just for engineering reasons. They certainly don't happen exactly 9 point something billion times a second.

[Geezer]

It jolly well is measuring a frequency; specifically, the frequency of a microwave oscillator. That oscillator is locked to the absorbtion frequency of caesium atoms in a "fountain".
In principle, you could have the events that control the frequency of the oscillator happen as seldom as you like- once a day or whatever. In practice they happen very frequently, but that's just for engineering reasons. They certainly don't happen exactly 9 point something billion times a second.

I fully agree with your description of how it operates. The atomic activity is used to tune the microwave oscillator to a frequency that corresponds with the resonant frequency of the cesium atom. The microwave resonator might "flywheel" for long intervals between adjustments. There is nothing unusual about that. It's a technique that is used in lots of high stability reference clocks.

Ultimately the clock is comparing the frequency of the microwave oscillator with the resonant frequency of the atoms, although, the process that it uses to do that is rather indirect. It's certainly not like a phase-locked loop or anything that simple. It makes very slight adjustments to the microwave resonator so that it produces events that occur 9 whatever billion times a second.

However, I don't fully agree with you when you say it is "measuring frequency" unless you are referring to the comparison process between the oscillator and the atomic resonator. I would agree that it is producing a frequency based on a comparison. It's a subtle difference that may be of little consequence.

My objection to Farsights comment was that it might imply that the clock, or clocks in general, somehow count events in a given amount of time, which is what it would do if it was "measuring frequency". It's really the other way around. It's producing events with (hopefully)a constant amount of time between each event, then counting off those events.

Hope that makes more sense.

[yor_on]

As with all time measuring we want to split times arrow as closely as possible, I assume? And for one second that would be 1.855094832e+43 'splits' according to Plank time conversion. If we assume that to be the ultimate definition of times arrow we will have the ultimate Plank frequency here it seems to me.

If we now had something oscillating that quick, and we also had something able to measure those oscillations and transfer that to a a dial f.ex we would have a extremely smooth dial movement So how do we fit those oscillations, 1.855094832e+43 times per second to the speed of light,

How far will light 'travel' for one second? Light travels at a speed of 299,792,458 meters roughly per second in a 'perfect' vacuum. So what will we get as a distance then? Splitting it against the oscillations/events I mean? Well as I understand it we will get a Plank length. "1 Planck length per Planck time is the speed of light in a vacuum."

So if we had something that exact I think we could say that we actually was measuring the 'speed of light in a vacuum'. But as it is we don't do that, the events we use for measuring is nowhere that exact as I understands it? That the photons radiated are in themselves traveling at 'C' don't make the clock 'work' at 'C' as I see it.

We would still be unable to measure 'times arrow' in 'real time' even if we had such an exact 'time splitter' though, as we still would measure something happening before we could observe it. So to do it in 'real time' seems quite impossible. But we would be close (Ah, this was a joke:).

In reality we are, all of us, working at 'real' Plank time even though we have no way of measuring it. And that is true as long as we all observe and interact via 'photons'. But to measure is another thing. that isn't about that ethereal 'now' forever disappearing, never to be 'caught' as all processes measuring 'now' will have to take some time, from thoughts to ....

So do this mean that times arrow actually is 'events', as we have a natural limit in the Planck time measured? Not really, it just make a statement about what we see as 'meaningful' for transitions/events observed. Everything faster than that will be unable for us to measure, ever, inside our SpaceTime as I understands it.
==

Well, as I understands it

[Geezer]

Yoron, It's about time you showed up [] for the debate.

I kind of think I understand what your saying.

Just wanted to point out something about the clock and "events".

The actual frequency that the clock oscillates at is not really all that important. If we knew of some substance that had a resonant characteristic that turned out to be more stable than cesium, even though it had a period of multiple seconds, somebody would probably figure out a way to use it to make a more accurate clock. In fact, as BC pointed out, the NIST clock isn't being continuously adjusted. There can be quite long intervals between adjustments.

Once we have a really stable reference, it's possible to synthesize any frequency that electronics will allow, and it will be almost as stable as the "master" clock, and certainly quite accurate over long periods.

So, to measure the speed of light accurately, we "just" need a very accurate clock. It does not necessarily have to operate with a very fast timebase, although it likely will.

Hope this makes sense.

 

[yor_on]

That's a way too Geezer. Still, the clock on my cell phone is sufficient for me. Nowadays I feel it goes to fast too? I liked that Atmos you wrote about

[Geezer]

The Atmos is a good example of what I'm on about. (Actually, it's not really a good example because it does not keep very good time.) Anyway, its pendulum has a full one minute period. It rotates around a torsion spring - 30 secs in one direction, then 30 secs in the other direction.

So it goes tick.............................................tock................................................tick.......

[Bored chemist]

Incidentally, the NIST clock doesn't define time. No single clock does; they use an average of a group of clocks round the world.

[yor_on]

I read that one should send it in for cleaning every twentieth year, The clock I mean, not times arrow. How old is it?

[Geezer]

The Atmos? (Not the NIST I suppose  []) It's been going strong for almost forty years. Never had to wind it up once!

Maybe a cleaning would help, but it's never really kept very good time. Mind you, some of that might just be perception. Because I never have to wind it up I hardly ever adjust it, so it's probably going for many months between adjustments.

BTW - for anyone that's wondering what we are yakking on about - http://www.atmosadam.com/

Probably more a piece of kinetic art that a great timepiece, but possibly a bit more interesting than a quartz digital clock!

[gem]

The ATMOS as manufactured by Jaeger LeCoultre, the makers of world famous Reverso wrist watches, is a time piece that for generations has represented the wonders of science, technology and remarkable Swiss craftsmanship. Possessing one signifies belonging to an exclusive group of world leaders, famous celebrities, business professionals, and, put more simply, people with exquisite taste.

so geezer to which category do you most identify with?

on a serious note on the fractions of a solar day something that soul surfer touched on the leap seconds that are used are not all in one direction [I will exspand on later but will have to go in new theorys]

[Geezer]

so geezer to which category do you most identify with?

Gosh! It's hard to say really. Can I only select one?

(It's one of the rather basic ones from the early 70's. Wedding present from my Mum and Dad.)

[gem]

On behalf of your wife i think it must have been a person with exquisite taste.

[Geezer]

It's not etirely clear Mrs G would agree with that statement  []

[yor_on]

Ah your mum and dad found a excellent present for you both my man. I read that you can expect it to work for hundred of years Up to a thousand if lucky, and cleaning it, nota bene. I was looking for one here in Sweden (curiosity:) But there were none for sale. People that have them seems to want to keep them too.

[syhprum]

One of the characteristics of clocks seems to be the higher the frequency of the basic oscillator the more accurate they are.
The 'Atmos' that Geezer uses has a torsion oscillator working at .033 Hz and is incredibly inaccurate where as most clocks today use 32768 Hz crystals and are pretty accurate while really stable clocks use microwave Cesium transitions.
The only exceptions that I can think of are gravitational ones based on the rotation of planets etc that use periods of about .000078 Hz and quite accurate. 

[LeeE]

The only exceptions that I can think of are gravitational ones based on the rotation of planets etc that use periods of about .000078 Hz and quite accurate.

Heh! - I hadn't thought of that.  Hmm... Jupiter, in its orbit, would make a pretty massive oscillator, and if I've done my maths right, it would have a frequency of 2.672026246839270841e-09 Hz.  Moving even further out, to use Neptune, would result in a frequency of 1.922923089229784694e-10 Hz  []

[Geezer]

Good points! Funny thing about the frequency. I wonder if there is some intrinsic law at work there, or is it more a case of the technology?

Of course, it gets even more complicated, because, the second is really defined by a rather large and very slow pendulum called the Earth  []. The SI second is a very good thing for science and engineering of course, but we always end up having to finagle things to line up with our one year pendulum.

[yor_on]

Syhprum you sure have a way with thinking
That was a very cool idea.

Now we just wait for LeeE to finish it. I'm putting myself first in the queue for buying one, when you finished your work Lee

A stellar watch
A very sweet idea.

(or is it a planetary?)

[Geezer]

I'm pretty sure I could rig up something that regularly adjusted the Atmos to keep it reasonably in sync with the RF time signal from Boulder CO, but it might tend to screw up the aesthetics, don't you think?

Synchronizing it with Neptune might be a wee bit trickier.

[yor_on]

Why stop at Neptune?