[time-nuts] suggestions on getting 24 Mhz ?

Mark Goldberg marklgoldberg at gmail.com
Wed Apr 11 16:21:26 EDT 2018


On Wed, Apr 11, 2018 at 11:15 AM, Bob kb8tq <kb8tq at n1k.org> wrote:

> Hi
>
> > On Apr 11, 2018, at 12:11 PM, Graham / KE9H <ke9h.graham at gmail.com>
> wrote:
> >
> > Phase noise is (usually) more important than absolute frequency accuracy.
>
> It’s probably well worth looking into exactly what you are after in terms
> of phase noise,
> spurs, and stability *before* deciding on a solution. In some cases things
> like DDS spurs
> are a non-issue. In other cases phase noise at certain offsets is a really
> big deal, but at
> other offsets … it’s a non-issue.
>
>
I am using the single output Bodnar GPSDO driving a frequency locked Wenzel
low phase noise oscillator and using that for my SDR ADC clock. I am using
it to take phase noise measurements and it certainly is as good as the
references I have available, and matches pretty well with measurements on a
$90k phase noise meter.

The Wenzel eliminates far out spurs from the GPSDO. Note that different
register values may result in the same frequency output of the GPSDO, but
spurs can be very different. For some frequencies and with the right
register values, I expect the GPSDO alone may be good enough for an SDR
clock, but I have not tried.

This is the best I can measure with this setup. I think the ADC noise
limits the far out phase noise.

Offset       Phase Noise
(Hz)           (dBc/Hz)
10                 -96.9
20               -103.0
50               -106.9
100             -109.7
200             -117.7
500             -128.4
1,000          -136.4
2,000          -140.1
5,000          -143.1
10,000        -144.9
20,000        -145.6
50,000        -145.7
100,000      -146.2
1,000,000   -146.6

Regards,

Mark
W7MLG


More information about the time-nuts mailing list