[time-nuts] Computing GPS Distance Error in Time

[Bob Stewart]

Hi Chris,

The manual says "timing accuracy 10ns".  They further say Position Accuracy without aid: 3.0m (50% CEP), DGPS 2.5m (50%CEP).  I am just learning about all of this, and DGPS is something new to me.  I see PRN-51 in the sky but not being used.  So, I take that to mean that DGPS is turned off.  I'm going to look into the manuals today and turn DGPS on to see what difference that makes.

The problem I'm trying to solve appears in two ways.  One is that the thing marches around the near neighborhood on foxtrotgps.  The other is that I'm seeing some anomalies in the timing when using it to drive the GPSDO I'm coding.  Maybe the DGPS is the answer.  If not, I need to quantify what's happening somehow, thus my question.  So, if it's just a vector sum divided by the appropriate scaling factor, what is negative and what is positive?  Is a southeast vector negative or positive?  If it's split into quadrants, which are positive and which are negative?  And if altitude is a component, how is that factored?  etc.

Bob




>________________________________
> From: Chris Albertson <albertson.chris at gmail.com>
>To: Discussion of precise time and frequency measurement <time-nuts at febo.com> 
>Sent: Monday, November 4, 2013 11:07 AM
>Subject: Re: [time-nuts] Computing GPS Distance Error in Time
> 
>
>On Mon, Nov 4, 2013 at 3:51 AM, Bob Camp <lists at rtty.us> wrote:
>
>> ..... I can guess it's a vector difference between each two successive
>> points converted to ns,
>
>
>That vector difference has to include altitude.   Then as you say, convert
>distance to time via the speed of light.  But I think this is only the
>upper bound of the error, it could be much less because there are multiple
>satellites in view.   That 3 ns per meter rule of thumb is very
>conservative.
>
>But on the other hand,...   On your specific make and model of GPS they
>could have simply dropped precision on the time, thinking that "0.01
>second is good enough" and allowed some approximations.   You can't assume
>the calculation is perfect.  You'd have to measure.  Is there a spec for
>timing error?
>-- 
>
>Chris Albertson
>Redondo Beach, California
>_______________________________________________
>time-nuts mailing list -- time-nuts at febo.com
>To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
>and follow the instructions there.
>
>
>