Message-ID: <29386.200102141714@gemini>
From: "James Moritz" <j.r.moritz@herts.ac.uk>
Organization: University of Hertfordshire
To: rsgb_lf_group@blacksheep.org
Date: Wed, 14 Feb 2001 17:18:52 +0000
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII; format=flowed
Content-Transfer-Encoding: 8bit
Subject: LF: 3 way QSO/Accurate clocks
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org
Sender: <majordom@post.thorcom.com>

Dear LF Group,

Congrats to G3LDO, VE1ZZ and VE1ZJ on their QSO - if nothing 
else, it proves that the SNR in both directions across the pond can 
be good enough to support 2 way communications in a reasonable 
amount of time.

As regards GPS derrived clocks, etc, it seems to me that errors of 
the order of 1us or so due to the GPS RX are always going to be 
small compared to other uncertainties - for example, the distance 
between the two stations can be determined quite accurately, but 
ionospheric effects are bound to alter the effective length of the 
path by much more than microseconds. Also, the group delay of the 
average CW receiver filter must be at least a few milliseconds, and 
this will be dependent on where the signal is in the passband, as 
well as temperature/ageing effects. So the issue would seem to be 
not so much how to get an accurate clock, but how to calibrate it 
for a QSO over a particular path between stations using a specific 
set of equipment.

Incidentally, the 2000 HP - sorry, Agilent catalogue says their 
HP58503B GPS receiver will give time accurate to 110ns with 
95% probability if I read the specifications correctly. The jitter on 
the 1pps output is <750ps RMS. The 10MHz output is specified in 
parts in 10^11 or 10^12 depending on the measurement period. It 
is in a cute box like a bench digital multimeter. No doubt easy 
mortgage terms are available....

Cheers, Jim Moritz
73 de M0BMU