Slashdot Mirror


NIST Unveils Chip-scale Atomic Clock

grumling writes "The heart of a minuscule atomic clock, believed to be 100 times smaller than any other atomic clock has been demonstrated by scientists at the Commerce Department's National Institute of Standards and Technology (NIST), opening the door to atomically precise timekeeping in portable, battery-powered devices for secure wireless communications, more precise navigation and other applications. "

15 of 172 comments (clear)

  1. OK, so when do I get one in my PC... by Atrax · · Score: 3, Interesting

    ... so my clock doesn't drift by like five minutes a day, necessitating a daily ping to the USNO time servers? anyone?

    --
    Screw you all! I'm off to the pub
  2. Great for GPS by DustMagnet · · Score: 5, Interesting

    With a atomic clock in a GPS you no longer need to solve for time, so you can get the same quality position with one less satellite. There are times where this could make a huge difference.

    --
    'SBEMAIL!' is better than a goat!!
    1. Re:Great for GPS by Detritus · · Score: 3, Interesting

      The Navy has been using atomic clocks for decades. Much of the technology in GPS can be traced back to early Navy programs for satellite assisted navigation. When you launch an ICBM from a submarine, you need a very accurate fix on the position of the submarine. Atomic clocks are also as timing references for secure communications links.

      --
      Mea navis aericumbens anguillis abundat
    2. Re:Great for GPS by Rich0 · · Score: 3, Interesting

      Ok, first a clarification, and then a possible alternative implementation of GPS with fewer limitations if you can assume the receiver knows the time:

      First - if you have only two satellites in the present system, the solution to the distance equation is approximately a hyperbola (it would be exactly one for a flat earth). Three satellites gives you three hyperbolas, which intersect at one point. (Remember, a hyberbola is the set of points where the difference between the distance from the point to two foci is a constant.)

      However, you don't have to use difference in time to implement a GPS-like system. A more direct solution is measuring absolute time to arrival for each satellite signal. This requires an atomic clock in the receiver - which is why they don't generally do it this way. If you measure absolute time to arrival you get a sphere around each satellite you receive signal from. With two satellites the intersection is a circle, but only two points of that circle lie on the surface of the earth. If you have even a remote idea of where you are (within 1000 miles or so), you should be able to figure out which one you're at. since the two points might literally be separated by more than 1000-2000 miles.

      So, the two satellite system might have some use - especially if the software is smart and keeps track of state. For instance, if you are a bomb decending to a target you would use as many satellites as you can - probably 3-4 most of the time. However, if you lost all but 2 as you got close to the ground and some jamming, you can probably bet that you didn't change position much, and so you can pick the closest solution and still have some useful guidance data. So, an atomic clock would be useful for more accurate GPS...

    3. Re:Great for GPS by UnknowingFool · · Score: 2, Interesting
      With a atomic clock in a GPS you no longer need to solve for time, so you can get the same quality position with one less satellite. There are times where this could make a huge difference.

      Say what? GPS satellites have always had atomic clocks but receivers have had to rely on quartz.

      Multiple signals are always needed by GPS for positioning regardless of timing accuracy. It's called triangulation. The more signals you have the better the accuracy due to timing differentials. Most GPS receivers use at least four satellites. Three are for positioning. The fourth is used to correct for timing differences. Some receivers boast of using 6 for better accuracy. Now if you have a better clock in a GPS receiver, it means you can find your position with fewer satellites and the accuracy of your location is better. Most GPS receivers always have an accuracy of 6-12 meters with some good ones in the 3-5 meter range. Using atomic clocks, GPS receivers could reduce that distance to feet.

      --
      Well, there's spam egg sausage and spam, that's not got much spam in it.
  3. Could this be the end of NTP? by feronti · · Score: 3, Interesting

    I mean, if every device has its own atomic clock, the only time you'd have to synchronize them would be when you bring them up, unless you were doing some kind of scientific work that requires ultra-accurate timekeeping. Most other applications (I'm thinking Kerberos, remote logging, etc) would only need to be synchronized to the second (or even less) to be useful.

  4. So now my cell phone will be able to tell time? by Ride-My-Rocket · · Score: 1, Interesting

    My Samsung SPH-A460 cell phone only shows the time when it can get a cellular signal. There's also a host of additional phone book software design flaws that make it less useful than my previous, archaic green-and-black phone. Here's to hoping this makes it easier for developers to easily integrate essential functionality into their products.

  5. Re:Great for GPS... and other things by Anonymous Coward · · Score: 2, Interesting

    It matters not just for navigation although that is the first time I really noticed how accurate time was getting easier and easier.

    Back in the seventies, our boss showed a video about him sailing across the Atlantic on a small sailboat. There was a shot in the cabin showing his digital watch (a new thing then) swinging back and forth. He pointed that out and said, "That's our chronometer." So at that point you could have the equivalent of a ship's chronometer (worth thousands) for less than a hundred bucks! I'm not even sure if a ship's chronometer would have worked on a small boat that rocked back and forth that violently.

    As atomic clocks became generally available, they enabled faster communications because it was no longer necessary to send as much clock information along with the signal.

    Thus far, cheaper accurate time has enabled us to do things that we couldn't before. On the other hand, I'm not prepared to guess what that would be this time.

  6. GPS Devices by KB1GHC · · Score: 2, Interesting

    This device will be excellent for Global Positioning Systems.

    GPS works like this:
    every GPS satellite has an atomic clock, your GPS reciver calculates all the difforences in time and position of the GPS satellites, and based on knowing the distance from each satellite, is able to calculate where you are. Currently, GPS recievers have Quartz clocks that are constantly kept snycrenized by the attomic clocks in the satellites.

    now quartz clock accuracy is nowhere near attomic clock accuracy, so this will make GPS recievers much more accurate. Even though GPS accuracy is already better than 10 feet/3 meters

  7. Another interesting home networking gadget by starbird · · Score: 3, Interesting

    When can I buy a netgear networkable home atomic clock box? Plug it in to your network, and use it to update the times on all your systems, instead of pinging NTP servers.

    Or put it on a pci card, I can just put it in my router box.

  8. Re:Does wifi/cellular marginalize this? by interiot · · Score: 2, Interesting
    Yes, even normal users sometimes need more precise clocks. The most obvious case is GPS mentioned earlier in these posts. Quote:
    • Currently, the GPS system provides time to the general public with uncertainties measured in nanoseconds.

    GPS receivers are required to synchronize very closely to the atomic clocks in the GPS sattelites in order to calculate position... they have to measure how far RF signals (travelling at the speed of light) have gone. GPS receivers (and the article's atomic clock) are more than a million times more precise than "a few ms of lag".

  9. Re:Does wifi/cellular marginalize this? by Detritus · · Score: 2, Interesting
    It would be a better frequency reference, which is important for the accuracy of microwave frequency synthesizers used in transmitters and receivers.

    Less phase noise, which improves overall system performance.

    A more accurate timing reference would make it easier for spread-spectrum systems to acquire and maintain lock on received signals.

    --
    Mea navis aericumbens anguillis abundat
  10. Gravity by Anonymous+Writer · · Score: 4, Interesting

    Since atomic clocks can be used to measure effects of gravity, it would be interesting to see how mass producing atomic clock chips could be used to create maps of gravity, seeing how they can be used to reveal geological information.

  11. Re:This had to come by Guppy06 · · Score: 2, Interesting

    "Give the rich part of the world ten years, and we're all spending our time wearing atomic _and_ digital watches."

    The hard part isn't necessarily keeping an eye on the atom, it's all the math needed to approximate what atomic time should be.

    Over the years since the adoption of the atomic second, all sorts of adjustments and clarifications were made to the definition, that include (among others) accounting for blackbody radiation (it's "supposed" to be at 0 K), special relativity (they're "supposed" to be stationary to the observer), and general relativity (gotta remember gravity and centripetal acceleration from earth's spin at your latitude). All this must be done in a relatively controlled environment so that you know what numbers to put into the equations.

    Throw that all in with the effects that swinging your arm will have while you walk, and putting one on your wrist sounds extraneous since you'll still have to correct it from time to time to get that accuracy you're supposed to have with an atomic timepiece. The correcting will probably have to be done by GPS time transfer, which is a lot more complicated than simply asking your GPS receiver what time it is. The light lag involved is very signifigant when you're counting 9,192,631,770 "ticks" per second.

    My radio-controlled quartz watch, when allowed to update itself every 3 days, is at worst accurate to the nearest second, and I've never seen it get anywhere near that far off. If I concentrate a little I can watch deciseconds tick by, but centiseconds are too much of a blur for a mere mortal like me. Why bother with nanoseconds?

    Put one in a computer? Sure. Network validation schemes like Kerberos would benifit greatly by knowing that all the workstations in the domain are all synchronized to a disgustingly high degree. But quartz crystals continue to be good enough for even high-end GPS receivers, so why would I need better on my wrist?

  12. Re:NTP by AnotherBlackHat · · Score: 2, Interesting

    the drift caused by the el cheapo crystal oscillator.


    Even the "el cheapo" crystal oscillators are guaranteed accurate to better than 1 cycle per 100,000.
    PC clocks drift by more than 1 second a day because of poor software, not poor hardware

    -- less is better.