[time-nuts] Symmetricom 58532A and 58536A group delay figures

[Hal Murray]

>> Interesting that it's so far off from the 40 ns in the data sheet.

> I think the usually quote worse case.

Both data sheets said "typical".


Worst case makes sense for things like delays through digital gates where you 
have to add up the total delay to see if your design will meet timing.  But 
if you are trying to correct for actual delays, the typicals are probably 
more useful.

Note that typicals can change over time if the manufacturer switches to a 
newer/cheaper/faster process, or you get a batch from a different vendor.  
The numbers in the data sheet may not match current reality.



-- 
These are my opinions.  I hate spam.