[time-nuts] Zero-Crossing Detector Design?

Magnus Danielson magnus at rubidium.dyndns.org
Fri Jul 20 23:54:01 UTC 2012


On 07/21/2012 01:28 AM, Azelio Boriani wrote:
> Yes, using a 'scope and the persistence it seems possible to "visualize"
> the results.

Exactly.

It's very instructive to see the traces separate appart as result of 
deterministic jitter, or just see the soft edges from the random jitter.

I forgot to mention, you typically view the rising transition one cycle 
off. For some scopes one need to feed the trigger separately and what I 
do is use a powersplitter hooked straight onto the trigger input, and 
then use a coax to compensate for the trigger delay + some offset. It's 
the difference in delay between trigger input and sampling input which 
needs to be sufficiently high.

This way I can view the trigger point completely, and I can also observe 
the 1 cycle jitter effect.

Cheers,
Magnus



More information about the time-nuts mailing list