[time-nuts] question about Thunderbolt geo acuracy

[Jim Lux]

On 5/10/12 6:08 AM, Magnus Danielson wrote:
> On 05/10/2012 02:50 PM, swingbyte wrote:
>> Hi all,
>> Hope this isn't too chat roomy, however, I have need of a survey precise
>> geolocation type gps. I was wondering if the precise timing abilities
>> extend to its precision in position output? I have a thunderbolt and one
>> of those conical white aerials from china and would like to know if this
>> combination will give me accurate height data.
>
> There are many sides to this issue. You will most definitely be best
> served by a choke-ring or similar antenna that suppresses multi-path
> reflections. In addition to that, you would want Lady Heather to do a 24
> hour position averaging. This should give you an OK solution, but really
> not the best achievable.
>
> Accurate height data is complex, since besides the receiver and antenna
> issues, height data has more uncertainty than longitude and latitude
> measures, and also since even if precise WGS84 height is achieved, you
> would need to correct it to your datum, your sea-level etc.
>
> You would also like to have better ionspheric correction than a plain
> GPS solution gives you, but the Thunderbolt does not give you direct
> support for such corrections.
>
> Exactly how much effort you need to do depends on how accurate you need
> it, +/- 10 m, 1 m, 1 dm, 1 cm or 1 mm.
>


If you can get RINEX format files, you can post process them through 
GIPSY at JPL and get higher precision, using post determined ionospheric 
and other corrections.


My friends in the GPS world say that getting to 1 meter absolute 
position is fairly straightforward but once you start getting finer than 
that, all the various factors start ganging up on you: ionosphere, solid 
earth tides, multipath, phase center shifts, etc.etc.

Likewise, getting 1mm + 1 ppm of separation distance sorts of 
uncertainty in a differential measurement is fairly straightforward.