British atomic clock 'most accurate in world' / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

British atomic clock 'most accurate in world'

1 2 next

Move Thread

LAN_403

Proud Mary

Fri Aug 26 2011, 10:40PM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

British atomic clock 'most accurate in world'

By Nick Collins, Science Correspondent
Daily Telegraph
26 Aug 2011

A British atomic clock is the most accurate measure of time in the world, scientists have announced.

The machine, which is responsible for keeping Britain's clocks on track and also contributes to the international measure of time, is accurate to within two 10 million billionths of a second.

It is one of a handful of similar clocks which determine the exact length of a second by measuring microwaves as they cause reactions in atoms of caesium, a highly volatile element.

But a number of factors including the shape of the microwaves, the influence of nearby magnetic fields and even the clock's position above sea level can cause tiny shifts in its measurements.

By tinkering with the clock, physicists at the National Physical Laboratory (NPL) near London, led by Dr Krzysztof Szymaniec, were able to reduce its margin of error to unprecedented levels.

A new analysis of the clock, conducted by NPL scientists and American colleagues, published in the journal Metrologia, established that it will now drop just a billionth of a second every two months, making it the world's most accurate timekeeper.

Hotwired

Sat Aug 27 2011, 12:44AM

Registered Member #4049 Joined: Sun Aug 07 2011, 03:50AM
Location: UK
Posts: 14

I'm happy if I've got the time to within a few minutes

ScotchTapeLord

Sat Aug 27 2011, 01:32AM

Registered Member #1875 Joined: Sun Dec 21 2008, 06:36PM
Location:
Posts: 635

Useless to me; I'm late to everything unless my watch is fast!

My watch has been 18 minutes ahead of time for about a year now.

PSCG

Sat Aug 27 2011, 07:02AM

Registered Member #3792 Joined: Sun Mar 27 2011, 06:07PM
Location:
Posts: 136

I'll wait for the synchronization my computer's clock until the replacement of the NIST - F1 with the NIST - F2 : 1E - 16 sec uncertainty.

mikeselectricstuff

Sat Aug 27 2011, 08:59AM

Registered Member #311 Joined: Sun Mar 12 2006, 08:28PM
Location:
Posts: 253

Can someone please explain how you verify the accuracy of a 'most accurate' clock (or anything else for that matter) when you don't have a more accurate reference to compare it against.

Avalanche

Sat Aug 27 2011, 12:49PM

Registered Member #103 Joined: Thu Feb 09 2006, 08:16PM
Location: Derby, UK
Posts: 845

wrote ...

A new analysis of the clock, conducted by NPL scientists and American colleagues, published in the journal Metrologia, established that it will now drop just a billionth of a second every two months, making it the world's most accurate timekeeper.

Something isn't quite right here... if the error is predictable (a loss of one billionth of a second every two months) then it becomes easy to compensate for it - making it even more accurate.

I was also wondering the same as Mike in the last post...

Ash Small

Sat Aug 27 2011, 01:41PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

mikeselectricstuff wrote ...

Can someone please explain how you verify the accuracy of a 'most accurate' clock (or anything else for that matter) when you don't have a more accurate reference to compare it against.

I 'assume' the easiest way to check the accuracy would be to use the speed of light as a reference.

Proud Mary

Sat Aug 27 2011, 01:52PM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

Geocentric Coordinate Time is defined within the framework of the General Theory of Relativity.

The relationship between Geocentric Coordinate Time and Barycentric Coordinate Time and Terrestrial Time and Proper Time is defined within the General Relativistic Metric.

Science students who don't make a habit of arriving fashionably late for their classes will know that identical clocks can appear to tick at different rates according to the point of view and relative motion of the observer. Knowledge of this difference is a major issue in astronomy, space flight, and particle physics, hence the struggle for greater and greater accuracy.

Avalanche wrote ...

Something isn't quite right here... if the error is predictable (a loss of one billionth of a second every two months) then it becomes easy to compensate for it - making it even more accurate.

This is the fault of dumbed-down journalism - the article should have refered to a measurement uncertainty of one billionth of a second every two months.

A much better account of the new clock is given in Science Daily:

Atomic clock with the world's best long-term accuracy is revealed after evaluation

And that's just about all I know about it, so don't look to me for further demystification!

PSCG

Sat Aug 27 2011, 03:10PM

Registered Member #3792 Joined: Sun Mar 27 2011, 06:07PM
Location:
Posts: 136

In a frequency standard (cesium, rubidium, hydrogen maser) you can compensate for phase error, frequency error and drift error.

Phase error (how much the "clock" has moved from the ideal time): you can correct this in hardware or software.

Frequency error (how much the "clock" has moved from the ideal frequency): you can correct it by negative looping (as the frequency error increases, loop is trying to decrease the frequency and, ideally, to stay in one value).

Drift error (the speed that the "clock" moves from the ideal value - time or frequency): it is more difficult to compensate for, but with smart computer algorithms you can correct it too.

The only thing that is impossible to compensate is the irregularity of the drift rate: the chaotic
style of this irregularity is that what distinguishes every frequency standard. And this is what the scientists call "Accuracy" of the clock.

We might think that "Oh, two 10 million billionths of a second are too little time". But this is what cesium can do. It may be possible to drop it a magnitude more, but cesium has reached a dead end - it can't do any better. The only way to increase the time accuracy even more is to move the frequencies that these standards produce higher, using optical frequencies instead of microwave.

radiotech

Tue Aug 30 2011, 05:22PM

Registered Member #2463 Joined: Wed Nov 11 2009, 03:49AM
Location:
Posts: 1546

Where in England is the timing point source for the telecom networks and will this improved
standard be used there?

Omaha used to be a point years ago for T1. What is the relevance today for telecom timing?

1 2 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob