[time-nuts] TIC resolution impact on GPSDO's performance

[Magnus Danielson]

From: Dr Bruce Griffiths <bruce.griffiths at xtra.co.nz>
Subject: Re: [time-nuts] TIC resolution impact on GPSDO's performance
Date: Thu, 28 Dec 2006 17:33:19 +1300
Message-ID: <4593490F.5000007 at xtra.co.nz>
> Didier
> 
> Yes, using an integrated high stability oscillator in the GPS receiver 
> provides all the local oscillator signals etc, and is "locked" to the 
> GPS signals eliminates the external TIC that would otherwise be 
> required. Some of the more expensive early GPS receivers used 10.23MHz 
> OCXOs for this purpose.
> 1575.42 = 154 x 10.23
> 1227.6   = 120 x 10.23
> 
> The drawback is that the short term frequency stability is limited to 
> that of the oscillator chosen by the manufacturer, if one happens to 
> have an oscillator (HP10811, HP10544, FTS1200, OSA 8607, etc.)with 
> better short term stability one can do better.
> 
> The OCXO used in the Thunderbolt doesn't appear to have great short term 
> stability specifications.
> It appears to be worse (100X??)  than just about any variety of HP10811 
> for example.
> It appears to be (1000x ??) worse than an FTS1200, FTS1000, or an OSA 8607.
> 
> To use an external oscillator in the way I indicated, one has to modify 
> a standard M12+, M12M timing receiver to allow using an external oscillator.
> I'm not sure that this is possible though it is likely.

It should be possible. I'll flipp my M12M over and check.

Bjorn and I have a little project going where we use the CNC Allstar receivers.
They have the specific feature that their referens XO is at 10 MHz. This will
make it very easy to drop in a suitable oscillator. As it happends I see that a
FTS1200 and an Allstar receiver has showed up in my main rack during some
unexplainable appearance in the lab yeasterday. It is not as much the ability
too hook it up as getting the receiver time to EFC solution into place. I'm
working on that right now, but it will involve a little bit of wizardy.
The Allstar receivers have about 1 cm RMS of carrier noise which should show up
as about 33 ps RMS noise. It should be an interesting proof of concept thing.

Cheers,
Magnus