[time-nuts] Accurate 1 pps signals

Martyn Smith martyn at ptsyst.com
Tue Aug 18 19:29:29 UTC 2009


Hello,

I am trying to achieve accuracy with two frequency standards disciplined to 
GPS.

I am trying to get their 1 pps outputs close to each other as possible.

The idea being you put each unit in a different part of the world and the 1 
pps outputs are sycronised.

So far I have two units (GPS10RBN) on their own antennas.  The antennas are 
seeing slightly different parts of the sky.  The antennas have different 
length cables but I have equalised the delays by entering a GPS delay into 
one of the GPS receivers inside the GPS10RBN.

I plot the rising edge of the 1 pp signals (derived directly from the 
rubidium) using my SR620 in time interval mode.  The sample time is 1 
second?

My results are that they are within plus/minus 8.5 ns.  This is over a 24 
hour period.  I'm sure they will still this close, I just haven't had enough 
time to make longer measurements.

This is a peak to peak measurment with no sampling.

Does anyone know a way of getting better results?  My target is actually 10 
ns, so I am meeting that, but just wondered what anyone else has achieved.

Regards

Martyn





More information about the Time-nuts_lists.febo.com mailing list