[time-nuts] GPS position survey

Sarah White kuzetsa at gmail.com
Sun May 5 11:28:55 UTC 2013


On 5/5/2013 6:35 AM, Miguel Barbosa Gonçalves wrote:
> Hi fellow time nuts!
> 
> I've recently bought a Trimble Acutime gold that will be used as a
> reference clock for a NTP server.
> 
> This receiver has the possibility of averaging it's position before
> entering what Trimble calls the overdetermined clock state. The
> default is to average the position with 2000 fixes.
> 
> What do you believe is a good number of fixes for the survey in this case?
> 
> Cheers,
> Miguel

Accurate position might be important. Light travels a distance of
something like 29.9792 centimeters per nanosecond

You might find the results of this study to be helpful:

http://www.syz.com/gps/gpsaveraging.html

I myself did a sampling over a period of 1 million samples
(actually, it was a value of 2^20, not exactly 1 "million")
quickest it could have completed such a survey is 12+ days

... However, I had "masks" set for elevation, signal, etc.
(resulted in occasional periods when there was no "lock")

I'm quite satisfied / happy with the results :)

hope this helps?
--Sarah



More information about the Time-nuts_lists.febo.com mailing list