New Most Precise Clock Based On Aluminum Ion

eldavojohn writes "The National Institute for Standards and Technology has unveiled a new clock that will 'neither gain nor lose one second in about 3.7 billion years,' making it an atomic clock twice as precise as the previous pacesetter, which was based on mercury atoms. Experts call it a 'milestone for atomic clocks.' The press release describes the workings: 'The logic clock is based on a single aluminum ion (electrically charged atom) trapped by electric fields and vibrating at ultraviolet light frequencies, which are 100,000 times higher than microwave frequencies used in NIST-F1 and other similar time standards around the world.' This makes the aluminum ion clock a contender to replace the standard cesium fountain clock (within 1 second in about 100 million years) as NIST's standard. For those of you asking 'So what?' the article describes the important applications such a device holds: 'The extreme precision offered by optical clocks is already providing record measurements of possible changes in the fundamental "constants" of nature, a line of inquiry that has important implications for cosmology and tests of the laws of physics, such as Einstein's theories of special and general relativity. Next-generation clocks might lead to new types of gravity sensors for exploring underground natural resources and fundamental studies of the Earth. Other possible applications may include ultra-precise autonomous navigation, such as landing planes by GPS.'"

Re:Plane landings?

[ircmaxell]

You need that kind of precision to do it from several hundred miles away (As is what happens with GPS). The Satellites all have clocks on board that are synchronized and constantly transmit the time. A GPS receiver simply needs to listen to a few of the satellites, and compute the difference between their times to determine the location. The kicker of it is that since the satellites are moving at fast speeds (relative to us), the time of their clocks are different from "our" time. So special relativity is used to counter those relativistic effects. So basically, GPS is only as accurate as the clocks that form its backbone. That's one of the reasons why unaugmented GPS is limited in accuracy to a few meters. Improving the accuracy of the clocks (by orders of magnitude) has the potential to cut down a few meters to potentially tens of centimeters... You'd need that level of accuracy to land a plane... Planes "flare" during landing (slowing the rate of decent to nearly 0 just as the wheels touch down). Plus or minus even one meter in any direction (up, down, forward, back, side to side) could be catastrophic. So current "autoland" autopilots use radar altitude and ground based ILS (radio based navigation) to gain the necessary precision. If GPS accuracy gets good enough to where you don't need those aux systems (or need them as primary at least), complexity of autopilots would drop significantly...

Re:Plane landings?

[Burdell]

The relativistic speed means their clocks run slower than clocks sitting still on Earth, according to special relativity. Another source of GPS time difference is that they are farther away from the center of Earth's gravity than we are, so according to general relativity, their clocks run faster than clocks on Earth. Both factors have to be taken into account.

Re:Plane landings?

[ircmaxell]

Correct, which is why I said in my post that ILS and radar altitude are needed at the present time. Radar altitude systems are prohibitively expensive and big (pretty much only usable on commercial airliners and in military applications). ILS is good for approach, but you can't land off it. It'll get you down to 50 feet AGL (or less in certain areas), but it won't get you though the flare. That's because the glideslope portion of the ILS signal is set a 3 degrees. So it can only tell you your relation to the 3* slope, not distance above the ground (which is what's really needed when you're over the threashold). ILS gets you aligned with the runway, and onto the proper approach path. It's an approach system, not a landing system (and was never designed as such)... For an ILS system without a radar altimiter, the pilot always must handle the actual landing. (hence the classes of autopilot, and existence of a decision height -- The height which you need to either be able to proceed visually, or abort the approach)... That's why autoland based on GPS is such an interesting thing. It would enable cheaper autoland systems which are a lot smaller than present systems (Basically business jets and light aircraft could potentially be equipped) and don't depend on airport infrastructure. So you could theoretically autoland on a small General Aviation airport...

I do have my Private Pilots license with Instrument rating, but I also love physics...

Re:Assuming We're still around

[Anonymous Coward]

Long-distance low-frequency radio syncing is actually quite inaccurate, at least as far as atomic timekeeping goes. There's the simple propagation delay based on your distance from the radio source, plus the possibility of getting waves that were reflected from the ionosphere, possibly more than once, likely varying from sync to sync, and no good way to know how much additional delay that adds even if it were consistent. Not to mention the issues with synchronization itself -- once you get the correct time code you wait for the next "beat" to start counting, but there's some delay between when the beat is received and when counting starts, or there's anticipation of the beat and possible imprecision due to that anticipation.

And there's the issue of non-continous correction. While your watch indicates approximately the correct time after a sync, there's likely some jump in the notion of local time after each sync, and between syncs there's no guarantee that timekeeping is accurate at all -- if you design a clock to sync every 12 hours it can be off by almost 2 seconds/day without displaying the wrong second-accurate time, and that doesn't lead to very consistent timekeeping on intervals shorter than the sync period.

All in all, it's plenty accurate for a wrist watch, but it's not really a high-precision time or frequency source.

Re:Plane landings?

[ircmaxell]

There's a simple answer. The speed of light (in a vacuum) is the absolute speed limit. The red and blue shifts you are talking about are when an object with mass goes faster through a specific medium than light can. But the key point there is that it's through a specific medium. Even if you were inside of a space ship that was going through a dense cloud faster than light could go through it, you'd still be going far less than C. Plus, the light that's measured here is presumably passing through a vacuum inside of the clock... That's without taking relativity into account. Once you add special relativity, the effect that time slows as you approach C causes everything to work itself out. Note that there's no absolute clock. There is no such thing as absolute time. There's only time relative to a reference frame. So while the clock on a space ship going 0.5*C relative to earth will go significantly slower than one on earth, it'll still be just as accurate within that reference frame (So two ships with the exact same velocity (vector) which have these clocks will be as accurate as they would be if sitting on earth).

Even if time did stand still or go backwards, you wouldn't be able to tell since our perception is dependent on time going forwards. Since time must be constant in any non-accelerated reference frame (gravity asside), if a clock was stopped by the slowing of time, so would your heart, and cells, and brain, and electrical impulses, etc...

I would recommend a pair of books for you written by Richard Feynman. Six easy pieces [amazon.com] and Six Not so easy pieces [amazon.com]... They provide a good "foundation" if you've never had a college level calculus based physics series before...

Re:Plane landings?

[plover]

But his assumption that clock error is responsible for all of the current lack of precision is wrong. Clock error is responsible for probably less than a third of the current error. Atmospherics, multipath reflections, and ephemeris errors account for the bulk of the error.

Sure, every improvement is an improvement. But these clocks are not a magic bullet that will magically grant centimeter precision.

Re:Plane landings?

[blueg3]

Red and blue shift are not caused by moving faster than the speed of light in the local medium (though Cherenkov radiation is), but rather by motion of the emitting object relative to the observer. (Not to mention cosmological and gravitational shifts.)

Re:Plane landings?

[mpe]

Improving the accuracy of the clocks (by orders of magnitude) has the potential to cut down a few meters to potentially tens of centimeters... You'd need that level of accuracy to land a plane... Planes "flare" during landing (slowing the rate of decent to nearly 0 just as the wheels touch down). Plus or minus even one meter in any direction (up, down, forward, back, side to side) could be catastrophic. So current "autoland" autopilots use radar altitude and ground based ILS (radio based navigation) to gain the necessary precision.

"Full blind autoland" systems have been around since the 1960's An unexpected problem with the first systems is that they were "too accurate", runways wear out quickly if touchdown always happens in the same place.

If GPS accuracy gets good enough to where you don't need those aux systems (or need them as primary at least), complexity of autopilots would drop significantly...

Most landings are performed by pilots. Even in an autoland situation the pilots go through similar procedures to if they were flying the plane. Otherwise things are likely to end up like TK1951.

Re:Ah, I unplugged the atomic clock...

[Anonymous Coward]

I think you miss the point of accurate clocks. Everything you have said is exactly why we need such accurate clocks. The accuracy of GPS is not due to the accuracy of a single clock. It comes about by comparing clocks on different satellites. It is those relative differences that allow you to measure distance or changes in the fine structure constant, or whatever else you want to measure. Let's say we can get one of these clocks on a GPS satellite. How will we know the exact orbit? By comparing it to a stationary clock. That will tell us its altitude. Then we use the clock on the satellite to compare against other clocks on other satellites. The worse every clock in that chain is, the worse our overall results are. The better every clock in that chain is, the better. Yeah, the whole point of accurate clocks is so that you can compare them so that you can measure all of the things you are worried about.