[time-nuts] Question about SA.33 Rb clock
jimlux at earthlink.net
Tue Oct 17 08:52:51 EDT 2017
On 10/16/17 9:37 PM, Hal Murray wrote:
> kb8tq at n1k.org said:
>> The gotcha is that the spur specâ€™s are not always met. As you might guess,
>> doing testing over really wide DDS ranges is impractical. Some designs use
>> cleanup loops. The gotcha then becomes a spur (say at 0.053 Hz) that is
>> inside the cleanup loop bandwidth â€¦.
> How narrow can I make the loop bandwidth with a traditional analog filter
> using reasonable parts? Do people build GPSDO style digital filters?
Crystal filters can be pretty narrow...
> Is there a reasonable formula for the spurs from a DDS? That sounds like a
> place where guaranteed-by-design would be appropriate.
There's some guidelines out there for DDS spur levels but most folks
just simulate it and see what the spurs are. There's a lot of variation
in the details of implementation of the DDS (phase accumulator length,
cosine table length, whether you're doing CORDIC or not, whether you're
using phase dither or amplitude dither, whether you've got one of the
clever filter schemes to move spurs around).
More information about the time-nuts