[time-nuts] Question about precise frequency / phase measurement

EWKehren at aol.com EWKehren at aol.com
Thu Apr 19 21:14:12 UTC 2012


Wolfgang
It would help if you would let us know what equipment other than a scope  
you have. Also what resolution you would want to achieve. One time set up  or 
want to use repeatedly.
Bert Kehren 
 
 
In a message dated 4/19/2012 4:53:40 P.M. Eastern Daylight Time,  
SAIDJACK at aol.com writes:

Depends  on the scope..

if your scope has 100ps A-to-B measurement resolution,  then waiting 5  
minutes in this scenario would give 0.83ns drift,  with 100ps uncertainty 
IF your 
oscillators were synced to ~3ppt which is  very tough to do with a 
free-running  OCXO (It would be unrealistic  to get that stability from the 
two 
sources if they  are free  running).

A more realistic scenario would give say 100ns drift in 5  minutes, then a  
0.1ns resolution on the scope would give a very  accurate reading in just 5 
 
minutes (100ns +/-0.1ns = +/-0.1% error).  The result would be 3.33E-010,  
+/-3.33E-013. Who needs more  resolution than that as the OCXO will likely 
wander  much more than  that in 5 minutes..

bye,
Said


In a message dated  4/19/2012 12:52:07 Pacific Daylight Time,  
azelio.boriani at screen.it  writes:

Yes,  and, as you can see, you have to wait 1  hour.

On Thu, Apr 19, 2012 at  9:49 PM, <SAIDJACK at aol.com>  wrote:

> Hi  Wolfgang,
>
> one of the easiest and  very accurate ways to do this  is simply to 
measure
> the drift of  the two 10MHz signals on an  oscilloscope. Adjust the OCXO 
so
>  that this drift between the two  traces is as slow as you can get it.  
Then
> simply  measure it over  time. Use one signal for  trigger, the other to
> display
>  if  you only have a  one channel scope.
>
> If you get say  10ns drift over 1 hour  (which you can easily measure even
> with the  cheapest scopes),  that is a resolution of 10ns/3600s =  
2.78E012.
>
> Or in  other words 27.7uHz!
>
> This has  been discussed before and  documented in the time nuts archives
> some  time  ago.
>
> bye,
>   Said
>

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




More information about the Time-nuts_lists.febo.com mailing list