[time-nuts] Positional accuracy of the M12+T

[Magnus Danielson]

From: "Poul-Henning Kamp" <phk at phk.freebsd.dk>
Subject: Re: [time-nuts] Positional accuracy of the M12+T
Date: Wed, 03 Jan 2007 19:26:02 +0000
Message-ID: <29724.1167852362 at critter.freebsd.dk>

> In message <D6D8D27B5219D649A005529978F791371CCE0B at SBSERVER.Synergy-GPS.sbs>, "
> Randy Warner" writes:
> 
> >You will see several
> >ns of 1PPS offset due to slight changes in the Position Hold Position,
> >but this is easily removed using the 1PPS offset commands if you are
> >trying to match to another receiver.
> 
> Are you sure these offsets are constant ?  My impression is that they
> depend on the satellite geometry ?

Having looked at the way my position moves around during the day (the surveying
mode in the M12M-T windows software plots it live as it goes along) I am fully
convinced that it does move around due to geometry. A position error cranks out
as a different time error in the pseudorange depending on the bearing. Toss on
the multipath (which is my main error - got to move the antenna).

You need to look on the quality of the survey to determin if you need a longer
survey period. You need to measure for at least a certain period to be able to
be sure that the quality you requested it valid (confidence range is narrow
enought), but you should really maintain the averaging until you have reached
your surveying goal.

Cheers,
Magnus