Didier Juges wrote:
>  
>
>   
>> -----Original Message-----
>> From: [EMAIL PROTECTED] 
>> [mailto:[EMAIL PROTECTED] On Behalf Of Bruce Griffiths
>> Sent: Sunday, May 04, 2008 7:27 PM
>> To: Discussion of precise time and frequency measurement
>> Subject: Re: [time-nuts] Fast frequency counting question
>>
>> John Miles wrote:
>>     
>>> If you go nuts with sampling 
>>> precision, aren't you just going to get more information than the 
>>> underlying process is capable of generating?  (I'm probably not 
>>> phrasing that very well, but you get the idea, hopefully.)
>>>
>>>   
>>>       
>> No, the device is an oscillator, everything of interest can 
>> be obtained from the times at which zero crossings occur.
>> In effect one is measuring the phase deviations of the 
>> oscillator during startup.
>> With a suitable measurement bandwidth, a measurement system 
>> noise of a few tens of femtosec or less is possible.
>> In most cases the oscillator zero crossing jitter will be 
>> somewhat larger.
>> The real problem lies in verifying the measurement system noise level.
>>
>> All one is attempting is to extend the phase noise 
>> measurement range down to averaging times of a few tens of microsec.
>>
>> Bruce
>>
>>
>>     
>
> I think the potential problem John is pointing out is that if the
> performance down to that level is driven by process variations rather than
> by circuit design (and therefore assumed to be randomly variable from unit
> to unit), how useful is it going to be to measure the heck out of one or a
> few unit if you cannot statistically extrapolate the results to the entire
> population and be able to predict yield?
>
> Didier KO4BB
>
>   
Didier

To come to valid conclusions on that, one needs to develop a model for 
the effect and calibrate it against the process variables etc.
Its always possible that previos models had similar effects bu it hadn't 
been noticed.

Bruce

_______________________________________________
time-nuts mailing list -- [email protected]
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Reply via email to