New atomic clock claims title of world's most accurate
By Donald Melanson 
posted April 16th 2009 10:26PM
posted April 16th 2009 10:26PM
You may have thought that the previous world's most accurate clock was good at keeping time, but it's apparently nothing compared to this new strontium atomic clock developed by scientists at the University of Colorado, which is supposedly more than twice as accurate and just as atomic. To achieve that impressive feat, the scientists made use of the same so-called "pendulum effect" of atoms as before, but took things one step further by holding the atoms in a laser beam and freezing them to almost -273 degrees Celsius, or the temperature at which all matter stops resonating. In clock terms, that translates to about one second lost every 300 million years. Of course, that's still one second too many for the researchers, and they say they "dream of getting an atomic clock with perfect precision." You just know you never want to be late for a meeting with these guys.
so wont it be perfectly precise until 300 million years after operation.
@bobert
thats if you use 1 whole second as the smallest margin of error. In 150 million years itll be off by half a second, etc etc. until you can say in an hour itll be off my some infinitely small number. What I wonder is the level of physics needed for clocks this accurate.
Secret government time machines!
No... It won't be perfectly precise ever.
They're saying, ever 300,000,000 years, it will need to be adjusted up a second. To remain accurate.
I'm guessing that it's more like it's slowly losing that one second over the period of 300 million years. It's not like when the 300 million year mark hits, all of a sudden "OMG! Our clock is one second behind!"
It losses:
3.333333 * 10^-9 seconds/year
Wow. Same as my energy-saving light bulbs - they're so long-lasting I change them because I get bored of them, not because they broke/don't work any more.
Same thing with the clock. Its always accurate. But I wonder how long will it last till its replaced by something "even better"...
I don't get how clocks are off at all these days. Why cant digital clocks perfectly represent time?
dude, I can't remember to reset my clock every 300 million years. c'mon, scientists! make my life easier!
300 million years!!!
one Second?!?!?
thats too inaccurate!
/sarcasm
Geeks Go Slashdot!
:)
@Chip: It's because we measure the second using information goatherd from natural phenomenon. Thus, we are always going to be at some *level* of accuracy
If they know for sure that it will lose one second every 300 million years, I presume it loses 1/300000000 second every year or 1/(300000000*365) second every day (not counting leap years), and so on...
Can't they just adjust the clock with the appropriate amount whenever they wan't to know the exact time?
If I understand it correctly, if you know exactly how many seconds you are off in 300 million years, you can adjust for this so you have the EXACT time.
Or I must be missing something.
@Dretske
The error isn't so steady and predictable, the error is random and unpredictable and that leaves the average inaccuracy mentioned.
Time is an illusion. Lunch time, doubly so.
Didn't these scientists ever take physics? You're not going to get the perfectly accurate clock because every time you measure your counting device, you're going to mess with its running state. I think it would be like attaining perpetual motion to do; either that or they'd have to build a needlessly complicated work-around for the efficiency loss...
I'll check back in 300 million years to see if it's still accurate. Only what should I use to check it's accuracy?
My question is? Who cares about about a second gained or lost in a human lifetime.
Can't they just set the clock BACK by 3.33x10^10 every second? I'm not sure if that's hard or not, but it would keep the clock in sync.
If this is the most accurate clock in the world then... what do they use to measure it up against when they say it's 1 second behind... behind what?? o.O
Nice. But they could not build a watch with this accuracy.
no watch is an "atomic" watch. the atomic watches you buy are set using the radio signal sent out from NIST (in the states of course). So they can pass the benefits of the accuracy on...
Sort of. But these radio watches don't account for the distance they are from the tower, nor for the time for the signal to register in the clocks microprocessor, which is certainly a solid fraction of a second.
no, but they could use gps to calculate the light speed delay.
Excuse me but it's not only NIST (WWV) that broadcasts atomic clock signals, the germans for instance have DCF77, and I think the asian region has their own too.
Oh and india has a radio station and a satellite service that broadcasts their IST
And then there's something called TAI (from the French name Temps Atomique International) :
I quote wiipedia "TAI as a frequency standard is a weighted average of the time kept by about 300 atomic clocks in over 50 national laboratories worldwide. The clocks are compared using satellites (see BIPM clock comparisons). Many of these are caesium atomic clocks, which are the standard by which the SI second is defined. Due to the averaging it is far more stable than any clock would be alone.
The participating institutions each broadcast, in real time (in the present), a frequency signal with time codes, which is their estimate of TAI."
The radio-controlled watches sold in europe often use the german signal btw, although I saw an ad for one that uses 3 systems so it's usable everywhere.
Also the brits have a time broadcasting station too, mention it to not feel them left out :)
@wWhat:
I wasn't implying that NIST is the only place that sends the radio signal out. I was saying that they are the US organization which does it.
Lots of important scientific discoveries can come out of this development, protein folding is the first to come to mind, as well as a better understanding of fluid dynamics, long distance measurements, chemistry, nano scale engineering, and so much more.
Atomic? Can it blow up stuff?
no, it's just a massive waste of money, like everything else at -273 degrees celcius.
They can never freeze it to absolute zero...
thats why it says almost
i dunno. the nerd in me felt betrayed that the article referenced neither "absolute zero" nor "zero kelvin". im sure it bothers no one else.
Don't worry. It bothered me too.
@austin:
Yes, but the article implys thatthe perfection they want to achieve is dependant on the temperature of the atom...which can never go to absolute zero.
@ a ham sandwich
yeah... they had to explain that -273ºC = "the temperature at which all matter stops resonating"... I hope they actually knew that it = 0K... it's not that clear without the term "absolute 0" as some people will always think they can freeze their coffee to -500ºC.
I got excited about seeing 0K, If only they can create something I can use in my trucks radiator, instead of a stupid clock. 100% efficiency!! My watch works just fine IMO.
"Anybody know what time it is?"
"Time for YOU to get a strontium atomic clock! BAHAHAHAHAHA!"
How many @Beats is this, sorry?
Ha ha, and I thought no one else would remember that.
its over 9000!!!!!!
over 9000?!?!?!
that cant be riiight!
Why the hell are the dudes spending their time and money on this? I understand the lure of achieving a higher goal than the guys previous, but don't we have better things to do and research? Who actually cares how precise their watch will be when they are dead?
Accuracy in timing is important for more than just the clock you use to wake your sorry ass up...lol
More accurate computers, star charts, quantum mechanics, etc...
some things need to be super accurate so that satellites don't collide in space and fall on our head
GPS is possible because of precisions in the realm of atomic clocks... The general and special relativity theories needed to be proven before the calculations for GPS could be precise and those theories were proven by very accurate atomic clocks... there is more to be proven.
So eventually the clocks are going to be so precise that they have to be calibrated for a certain speed you're travelling at due to relativity? Ie. the speed at which the clock, on the rotating earth, orbiting the sun, is zooming towards virgo? I'm pretty sure that's not a constant velocity, hence the accuracy of the clock can never be defined precisely...
Actually 1 second in 300m years is probably already precise enough for relativity to affect it..
That was my thoughts! Surely the existing clocks are accurate enough? Why not just put a post-it note on the existing ones that says 'Please put me forward 1 sec in 150million years. Thx!'
Besides, none of these devices will still be in active service in 200 years, so any measurements required here and now would surely have such minimal drift as to be irrelevant?
And don't they already know that time isn't constant, like the rotation of the earth or the orbit of the moon? Seems to be a case of my cLock is bigger and better than yours.
Actually, cesium frequency standards that can be bought off the shelf (granted, for as much as a new car for a new one) have been precise enough to detect relativity effects even while staying on Earth's surface for years now.
And they're much less precise than this thing.
http://leapsecond.com/great2005/index.htm
@Sirius
hehe, no. Each clock, regardless of its position has its own Eigenzeit. So you can't calibrate them for a different spots since time passes by differently in all of them
Atomic clocks have nothing to do with "clock time". It is for navigation... Ever notice Longitude is in degrees, minutes, seconds, milli-seconds, ect...? It is for positioning and the farther you move from the Earth the more accurate of time you need.
When sending objects into the deep solar system, you need hyper accurate navigation.
what?