[time-nuts] Building a GPSDO & trouble using Jupiter-T

[Azelio Boriani]

Timing GPS receivers usually have the sawtooth correction message and I saw
an application of a delay line to correct the PPS before using it. Of
course if you time the PPSes difference with a time-to-digital you don't
need to correct the incoming hardware PPS by a delay line, but to implement
an all-analog PPS PLL then it can be done. For those of you who use a 10KHz
reference signal there is not even the need for a long time constant in the
low pass filter: to average 1000 PPSes you need 1000 seconds but to average
1000 10KHz cycles you need only 0.1 seconds.

On Mon, Jan 30, 2012 at 6:15 AM, David <davidwhess at gmail.com> wrote:

> The sawtooth error in the PPS output and how they were able to correct
> it externally was interesting.  I have seen that kind of problem
> before in DDS and other applications.
>
> I wonder what other GPS receivers provide either PPS outputs without
> sawtooth noise or a correction message.
>
> On Sun, 29 Jan 2012 20:40:52 -0800, Hal Murray
> <hmurray at megapathdsl.net> wrote:
>
> >
> >> What do you mean by "average"?  Do you mean that the GPS and PLL must be
> >> kept on for "20 minutes to hours", or did you mean that the PLL loop
> filter
> >> must have a time constant of 20 minutes to several hours?
> >
> >You have to compare the characteristics of the oscillator with the
> >characteristics of your GPS receiver.
> >
> >If your local oscillator is very stable, then you want to average over
> long
> >times (hours, days).  If your local oscillator is a good thermometer and
> you
> >have a very good GPS receiver, then you want a shorter time constant
> >(minutes) so you can track temperature changes.
> >
> >Do you know about hanging bridges?  If not, please read Timing for VLBI by
> >Tom Clark and Rick Hambly.  It's got some wonderful graphs.  Once you
> >understand those, this discussion will get much more interesting.
> >  http://gpstime.com/files/tow-time2009.pdf
>
> _______________________________________________
> 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.
>