[time-nuts] GPSDO Component Selection

[Chris Albertson]

On Sun, Sep 9, 2012 at 6:18 AM, Don Oconnor <eGunk at wowway.com> wrote:
> Hello,
>
>
>
> I would like to design and build a microcontroller based GPSDO. And I have a couple of questions some of you may be able to answer.
>
>
>
>
>
> 1.         I am going to use Trimble SMT GPS timing module but I'm curious. Does a GPS timing receiver produce a more precise 1 PPS output than a standard GPS tracking receiver in a static location? Or does it just produce a 1 PPS signal that is more accurate to GPS time?

Both of the above by several orders of magnitude.  The best timing
mode GPSes today have the PPS within a few tens of nanoseconds of UTC
(one sigma) while the typical navigation mode GPS has the PPS within a
few microseconds.   Also the timing mode GPS can do things like a self
survey to determine location.  The finding a soltion for time within
_knows_ it is not moving removes some uncertainty in the result.   I'd
say it would be almost pointless to use a nav receiver in a GPSDO
especially when decent timming mode receivers sell for under $20.

>
> 2.         I have seen many DOCOXs' listed on eBay produced by Morion Inc St Petersburg Russia. This includes models MV89 and MV180 with manufacture dates in 2008. All of the Ox's' were removed from equipment. I don't know anything about Morion's reputation. The specifications are very good, but do they live up to their data sheet, are they reliable and stable?

Morion has a good reputation but like anything used from eBay expect
unit to unit variation.  Might be worth buying a few.  Actually how
else could you measure your GPSDO?

I think the main thing that will determine performance is the phase
detector that will compare 10MHz to 1Hz.  That is your design and is
the most critical part.

Chris Albertson
Redondo Beach, California