[time-nuts] carrier phase tracking GPS receiver

Magnus Danielson cfmd at bredband.net
Mon Feb 19 19:51:09 EST 2007


From: Didier Juges <didier at cox.net>
Subject: Re: [time-nuts] carrier phase tracking GPS receiver
Date: Mon, 19 Feb 2007 18:12:57 -0600
Message-ID: <45DA3D09.4080003 at cox.net>

> Dr Bruce Griffiths wrote:
> > Didier
> >
> > Typically with a good local oscillator you can do even better than that, 
> > around 1E-11 in 1 sec is achievable and has been achieved.
> > Even with the on board TCXO typically 3E-11 or so in 1s is achieved.
> >
> > Bruce
> >
> >   
> I don't understand how such good variance can be achieved without some 
> method of electronic frequency control on the TCXO (unless the TCXO does 
> have EFC)?

http://www.oscilloquartz.ch/file/pdf/8607.pdf

See bottom table of Page 2. OK, this is an expensive rock, but looking at
those numbers you can realize that the numbers which Dr Griffiths is giving
can be acheived for much less.

Also, we are talking about their out-of-the-box properties, not in a streered
application. You gotta read some more datasheets to see what it "typical".
It is always good to calibrate onces senses.

The TCXO can have fairly good performance in the short term, but not that
great on longer term since heat waves will have a highly noticeable effect on
them. It is also not really worth it to spend alot on a very good crystal and
oscillator if you cheating out by TCing it. You end up in the OCXO single-oven
range first, and there is plenty of variation of performance right there!
I have 3 cut-open oscillators on my office desk (OK, the ESD pad area) right
now, since the performance difference between them needed a bit of explanation,
and I did get my hunches confirmed.

> Otherwise, we would be back to the case where the receiver would 
> approximate the position of the PPS by skipping pulses, and we would 
> have the typical 40nS jitter (assuming a 25 MHz clock)

Sigh!

> I am more puzzled now :-)

Well, that's how we learn, right? ;)

Cheers,
Magnus



More information about the time-nuts mailing list