[time-nuts] Thunderbolt self-survey results...

[Björn Gabrielsson]

Hi Dave,

On Sat, 2008-08-30 at 11:09 +0100, David Ackrill wrote:
> Michael Baker wrote:
> 
> > However, after completing its self-survey it thinks
> > its elevation is 11.2 meters when the actual elevation
> > of its antenna (on my house roof) is 28.4 meters.
> 
> I've not found any GPS receiver, available to the general pubic anyway, 
> that correctly reports its height above sea level.  So, I tend to ignore 
> that reading, personally.

Due to geometric reasons the GPS altitude/height-accuracy is worse by a
factor ca 1.8, than the horizontal (2d) accuracy. 

This gives the basic reason the vertical is worse.

Other factors contributing

  - the earth is flat (!?) ie heights from maps of sea levels etc can be
very accurate.
 
  - user confusion of ellipsoid height (that GPS works in) and geoid
height (mean sea level height). You need to know/check which height is
given by the GPS in each "message" you are looking.
 
  - inaccurate geoid model in the GPS to translate ellipsoid height to
Geoid height.  
 
  - modern high sensitivity receivers will happily track signals that
have bounced a couple times between floor and ceiling, which of cause
gives even worse height accuracy.


> Dave (G0DJA)

--

   Björn