[time-nuts] Low jitter oscillators

SAIDJACK at aol.com SAIDJACK at aol.com
Mon Aug 16 03:31:57 UTC 2010


Hi Bob,
 
Agreed that at a minimum the instrument will restrict the  bandwidth due to 
it's own bandwidth limitations, and some (like the  Wavecrest units) go up 
to 1GHz or more.
 
But many vendors claim sub 1ps jitter on low cost oscillators, and don't  
specify what bandwidth was used to measure this jitter. That would be like  
giving phase noise in dBc without also specifying the /Hz, and then every 
vendor  using their own internal bandwidth definitions to make their phase 
noise  claims..
 
Sometimes I think vendors assume that E5052A units are being used  
everywhere to measure jitter.
 
bye,
Said
 
 
In a message dated 8/15/2010 17:34:09 Pacific Daylight Time, lists at rtty.us  
writes:

Hi

The bandwidths chosen generally have roots in a telcom spec.  Infinite 
bandwidth generally equates to infinite jitter. You have to restrict  it somehow.

Bob



More information about the time-nuts mailing list