[time-nuts] Pulsars make a GPS for the cosmos

Tom Van Baak tvb at LeapSecond.com
Sat Sep 28 13:30:50 EDT 2013


> Pulsars are not infrinitely stable. They slowly decay, and, worse,
> randomly undergto 'star quakes' which upset their timing. This was proven
> in the 1960s.

John,

You are correct, but it is a simple opportunity rather than a major problem. The solution is to monitor multiple pulsar sources. Not only can you detect the quakes, but you can precisely measure their magnitude as well. Once a quake is detected the new phase and/or frequency offset for that source can be entered into the ensemble equation, and you move forward as if nothing happened.

Exactly the same thing happens at home when you monitor multiple quartz oscillators. You need more than one or two; three or four is better. Confidence grows according to the number of oscillators. N-1 oscillators help you detect any abnormality in 1.

This is not unlike what happens with GPS SV as they come in and out of reception. Easy to detect; easily solved.

And the same solution is used for UTC, where more than 300 clocks participate. Phase or frequency jumps (quakes) are easily identified and excluded from the ensemble mean. The mean is safe and the N-1 clocks can be used to precisely measure the anomaly in 1 clock.

/tvb



More information about the time-nuts mailing list