[FOSS-GPS] Good antenna

[beale]

FWIW: I'm using a LEA-6T with a $6 ebay magnetic puck sitting on a 1m x 1m wire-mesh groundplane. This is on a mast about 1 m above my chimney and it has a pretty clear view down to the horizon in most directions.  I have experimented with different satellite mask angles in RTKLIB and 5 degrees gives me the solution with the lowest variability. Using the same antenna in the same place with only a 20 cm groundplane, I did better with a 15 degrees cutoff.

Using a 24-hour observation period and PPP-Static solution (just my single receiver; not using data from any other station) my ground track typically fits into a box 7 cm (E-W) x 3 cm (N-S) x 5 cm (U-D), and the day-to-day variation has been about 10 cm horizontally and 20 cm vertically over a few weeks now.  I have not been taking into account "solid tides" but at this level I probably should.  The scatter in the results does not look random between days; there seems to be a consistent rise in elevation during/after a rain so I'm wondering if I am seeing some local atmospheric effect, or actually a true elevation change due to groundwater (I don't think my 7 m chimney expands 20 cm after a rain).

-John B

>  -------Original Message-------
>  From: Danny Miller <dannym at austin.rr.com>
>  To: Open Source GPS-related discussion and support <foss-gps at lists.osgeo.org>
>  Subject: Re: [FOSS-GPS] Good antenna
>  Sent: 10 Mar '14 22:26
>  
>  Really?  I thought helix antennae were  poor choices because they accept
>  signals at low angles.  In fact  the article makes a point of saying it
>  accepts low-angled signals.
>  
>  From satellites coming in at only 5 deg above the horizon, they're  prone
>  to distortion from the atmosphere and ground obstacles.  Now  the software
>  can just refuse to use those satellites because it  knows the angles from
>  the ephemeris data.
>  
>  But as I understand it, a lot of the multipath reflections-  delayed
>  versions of the original signal bouncing off things- often  comes in at a
>  low angle, so the antenna rejecting the low-angled  reception will help.
>  These signals traveled a longer path and  only distort the pseudorange.
>  Since this is a feature of the  antenna, software can't just elect to
>  reject these signals- and it  if could, it would.