[time-nuts] Re: PPS adjustment ... ?

[Bob Camp]

Hi

The GPS puts out a PPS based on it’s internal TCXO. The edges
of that TCXO are not aligned to anything. It’s just a free running 
oscillator. 

To put out a PPS, the module picks the closest edge and uses that.
On some ( like the LEA-6T) modules, the device reports how far
off it thinks the edge is. 

The net result is an error curve that is bounded by the period of the
TCXO’s output. If you graph it, the error looks like a sawtooth. It is
commonly called “sawtooth error”. There is no guarantee that it will
always look that way, 

In some cases the TCXO edge will simply sit a bit early or a bit late
relative to the “correct” location for a while. Normally the sawtooth 
shaped stuff averages out. This “hanging bridge” will not average out
over the short term. Best guess is that’s what you are seeing.

Bob

> On Feb 23, 2023, at 2:36 PM, Dave via time-nuts <time-nuts at lists.febo.com> wrote:
> 
> I have built a GPSDO based around a NEO-6M and it works well.
> 
> However, I have noticed tiny shifts in the PPS signal every few minutes.
> 
> Does the GPS smoothly adjust the PPS signal or does it make small correctional steps from time to time???
> 
> It doesn't affect my usage - I'd just like to understand ...
> 
> Dave
> 
> _______________________________________________
> time-nuts mailing list -- time-nuts at lists.febo.com
> To unsubscribe send an email to time-nuts-leave at lists.febo.com