[time-nuts] lightening protection of a GPSDO system / optical isolated distribution amp
kb8tq at n1k.org
Thu Nov 27 10:24:16 EST 2014
If you are happy with 1x10^ -9 frequency accuracy, then there are a number of things you can ignore. If you are after < 1x10^-11 frequency accuracy 99.9% of the time you have to do everything right. Different limits apply to different measures of stability.
To reduce multi path you need a clear sky view. To keep the timing solution in the GPS happy you need to let it see >80 % of the sats it should see. In this case stuff below 20 degrees does not count towards the total.
In both cases the exact numeric impact depends on a bunch of things. It would take a few hundred pages to go into all of it. You ultimately need a site survey to work out the multi path math.
In the extreme case the antenna can not see any sats. The math is simple then. Your stability is just that of your oscillator. That gets us right back to Rb's are better than OCXO's. Both are better than a TCXO.
Best advice is still the same. Get as good a GPSDO as you can afford. Put the antenna up on the roof.
Sent from my iPhone
> On Nov 27, 2014, at 3:24 AM, Dr. David Kirkby (Kirkby Microwave Ltd) <drkirkby at kirkbymicrowave.co.uk> wrote:
>> On 27 Nov 2014 03:06, "Bob Camp" <kb8tq at n1k.org> wrote:
>> You need a good antenna setup to get a good disciplined oscillator. The
> bigger question is if you *need* the sort of “better” it gives you. If you
> are running a simple TCXO based GPSDO, you will have a number of issues to
> deal with compared to an OCXO or Rb based system. If “better” is an
> objective, consider the entire system.
> can you explain more about the effect of antenna performance on a GPSDO
> Now you have told me it is important, I would like to know more! My lab
> has an East facing window.
> The only way I would use an external antenna is if I had optical isolation.
> I live in an area where the lightening risk is considered low, but having
> lost equipment twice, I am probably more concerned than others might be.
> time-nuts mailing list -- time-nuts at febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.
More information about the time-nuts