[ntp:questions] PPS from an external OCXO source. Correcting drift is it possible?
John Ackermann N8UR
jra at febo.com
Sun May 3 11:44:09 UTC 2009
Kat said the following on 05/03/2009 12:05 AM:
> Nero Imhard wrote:
> > A much more reasonable approach would be to hook up the OCXO to
> > computer's hardware clock for better stability.
> Unfortunately not as easy as it seems high frequency clocks are
> notoriously difficult to construct and calibrate and a new custom
> clock or clocks would be needed every time I changed server so I opted
> to do something portable.
I try to avoid shameless plugs, but have you considered the TAPR
Clock-Block (http://www.tapr.org/kits_clock-block.html)? It's a clock
synthesizer specifically designed to allow clocking a PC motherboard
from an external OCXO (or other high quality) reference. It'll take
inputs like 5 or 10 MHz and generate outputs up to a couple of hundred MHz.
I use the Clock-Block to generate 33.333... MHz to drive a Soekris 4501
single board computer which has some unique hardware timing capabilities
such that by adding a high quality PPS you end up with an NTP server
that can keep time to fractions of a microsecond. I have details of
that at http://www.febo.com/pages/soekris.*
TAPR is currently out of stock on the Clock-Block, but we are having
another run made and they should be available Real Soon Now.
* Note -- if you look at the performance plots linked from that page,
things won't look so good right now -- the Cesium clock that I use as
the reference for the offset comparisons died and is now acting like an
OCXO. Haven't had a chance to reconfigure to switch to the alternate Cs
More information about the questions