[time-nuts] gpsdo 10MHz steering resolution

[Mike Ingle]

Hi everyone,  Thanks for your reply's.  The device is indeed in 50hz land
(germany).  I suspect you are right that it is 50 hz mains pickup.  A
little surprising since it is a "factory," as opposed to homemade unit.
But it is almost 20years old, so the caps may be dry, or it is just a bad
design.  I measured my DDS based signal generator output, and it was spur
free, so my spectrum analyzer and cabling is an unlikely culprit.  I had
been thinking along the lines of working from an OCXO, and fitting the PPS
to the stamped intervals, and assuming that the average interval is exactly
one second. The issue then will be OCXO changes vs number of averages from
the PPS.
I am on vacation till next week,  but I will change out the PSU on the
truetime (temporarily to my bench linear PSU) and move the internal cabling
around to see if the spur amplitudes change.

--mike




On Mon, Nov 25, 2019 at 5:34 PM Bob kb8tq <kb8tq at n1k.org> wrote:

> Hi
>
> I caught the 1KHz and took it to be per division…. oopss … (sorry about
> that)
>
> If the device is indeed in 50 Hz power main land, the 100 Hz spurs might
> make some sense
> as line related. If it’s in 60 Hz power land, they are coming from
> something else.  No idea
> (yet) what part of the world the device is in.
>
> Bob
>
> > On Nov 25, 2019, at 3:55 AM, John Moran, Scawby Design <
> john at scawbydesign.co.uk> wrote:
> >
> > I may be completely wrong here but if the span of that spectrum is 1kHz,
> then the spurs look like 100Hz mains hum   frequency modulation of the
> 10MHz.
> >
> >
> > _______________________________________________
> > time-nuts mailing list -- time-nuts at lists.febo.com
> > To unsubscribe, go to
> http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
> > and follow the instructions there.
>
>
> _______________________________________________
> time-nuts mailing list -- time-nuts at lists.febo.com
> To unsubscribe, go to
> http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
> and follow the instructions there.
>