NavList:

   

Reply

Örjan Sandström wrote:
> Any hints as to as to why my terrestrial cel-nav intercepts if I use the
> symedian point more or less constantly show me to be 0.5-0.7' east of
> GPS even with over 20 ?

If your fixes were significantly *west* of the true location, I would 
suspect a systematic late time, e.g., from not reading the chronometer 
second hand first. In that case, if you observe a body to the east 
(west), the computed altitude is too high (low), and so a perfect 
altitude observation is lower (higher) than the computed altitude. This 
shifts the fix to the west.

But your error is to the east, so that explanation doesn't work. I don't 
believe someone would consistently read the time early by 2 or 3 
seconds. And I don't think your chromometer checks by radio would have a 
systematic error that large.

You said, "whatever azimuth body has seems to make little difference," 
but it's not clear if you have observed bodies to the east and the west. 
If observations in both directions agree, that would exclude sextant 
error and personal error.

The GPS might be wrong – if the wrong datum is selected. WGS84 is a good 
worldwide datum. But some of the local datums can give large errors if 
you apply them in the wrong part of the world, e.g., if you use the old 
Tokyo datum in the U.S.

Unlikely but possible – deflection of the vertical. The coordinates from 
a map or GPS are with respect to a mathematically perfect surface, an 
ellipsoid. Due to Earth's irregular density and shape, a gravity 
vertical is not exactly perpendicular to the ellipsoid. In most places 
the error is only a few seconds of arc and is not significant in 
practical navigation. However, deflection of the vertical may be 
apparent to a careful observer in excellent conditions. There are areas 
in the U.S. where the deflection of the vertical is as large as the 
anomaly you report.

http://www.ngs.noaa.gov/GEOID/DEFLEC09/

Unfortunately, the online calculator on that page is good only in U.S. 
territory. The NGA has a worldwide calculator, but it gave an error 
message when I tried it today.

http://earth-info.nga.mil/GandG/wgs84/gravitymod/egm96/intpt.html

-- 

   

Reply