[time-nuts] yet another GPSDO design, or so

EWKehren at aol.com EWKehren at aol.com
Sun Jun 27 18:17:51 UTC 2010

Going from 24 to 100 MHz only gives you smaller steps (resolution) every  
thing else stays the same.
If the he saw 2 to 3 nsec should be more like 8, going to 100 MHz will  
improve it by a factor of 4. In a redesign of the total system I would have two 
 sample sizes maybe stay with 30 or go to 50/60 and in the Rub. mode 200  
maybe 300 sec. Let us not forget what we start out with the GPS signal does 
not  allow us to take advantage of the full resolution.
Do not forget I did this to get smaller D/A steps and am not able to  
rewrite the code, basically fooling the controller that the error should call  
for a 1.7 E-13 correction when in reality the error is 4.3 E-14 and the  
resulting step is also 4.3 E-14  per D/A bit.
In a message dated 6/27/2010 1:43:40 P.M. Eastern Daylight Time,  
stanley_reynolds at yahoo.com writes:

Yes I  see your need for a reduced range with smaller steps. But I was 
looking for  smaller steps to improve the tracking accuracy without a loss of 
the benefit  of averaging. From the QST article:

"Interestingly, it is  desirable to have the frequency of U7 drift slightly 
rather than being  synchronized with the VCXO. A
slight random drift averages out the count  ambiguity that is inherent in 
any pulse-counting device. My  measurements
indicate that the simple phase-measuring circuit I use is  consistently 
accurate to 2 or 3 ns (for a 30-second measurement),  while
without drift, the resolution would be limited to 42 ns. The $5  crystal 
oscillator module drifts adequately"

So the drift should just  cover the area of uncertainty that is one cycle, 
too much drift would reduce  accuracy, not enough and the average is of no 
One extreme no  jitter, average doesn't work as it doesn't distribute the 
samples over the  range of uncertainty. The other case too much jitter and 
the best to expect is  an average weighted to one side or the other (+-1 
count) with the extreme  producing multiple counts of error. Something about this 
makes me nervous  maybe the part about "slight random drift" what is slight 
at 24 Mhz is it also  slight at 100 Mhz ? An average of 30 samples does 
have a limit to what it will  correct. 


----- Original Message ----
From:  "EWKehren at aol.com" <EWKehren at aol.com>
To: time-nuts at febo.com
Sent:  Sun, June 27, 2010 8:58:55 AM
Subject: Re: [time-nuts] yet another GPSDO  design, or so

the faster counter also has the jitter, no  change, as long as it is  not 
tied to the input frequency. The 24 MHz  is not unique, the 100 MHz is same 
technology just four times faster  and thus gives me smaller steps on the 
and  since I use it on  Rub. the full range of the 18 bit covers the full 
tuning range  of  the Rub.

In a message dated 6/27/2010 9:05:12 A.M. Eastern  Daylight Time,  
stanley_reynolds at yahoo.com writes:

I  have  been thinking about a faster counter also but the Shera board was  
depending on  the jitter in the 24 Mhz clock to average out the +-  count. 
faster  clock would reduce the need for this but without  the right amount 
of jitter we  lose the benefit of this  average.



time-nuts  mailing list -- time-nuts at febo.com
To unsubscribe, go to  
and follow the  instructions there.

More information about the time-nuts mailing list