[time-freq] RefLock II behavior

[Luis Cupido]

Hi John,
Hi Michael,

John, I don't know what is the range of Michael's VCXO span,
assuming it to be something like 50Hz or similar, and considering a dac
size of 11bits that makes the quantization step to be 0.024 Hz
so 1 Hz change should have a completely different origin.

Also the loop filter should take care of that dithering by integrating
many seconds, we can make the calculations to see how long we should
integrate for example to get a resolution of 1/10 the dac size...

Or the VCXO range is extremely large for the application (like 1KHz or 
so...)
Or I would suspect of the 1pps quality, some GPS engines have quite bad
pps jitter and this looks like changes when switching satellites or reduced
satellite visibility (calculating with just 3 satellites for a short period 
, for example)
and a non optimum GPS engine software.

Just my impression... not enough info to make a more in deep judgment :(

Luis Cupido.
CT1DMK

----- Original Message ----- 
From: "John Ackermann N8UR" <jra at febo.com>
To: "TAPR time and frequency projects" <time-freq at lists.tapr.org>
Sent: Wednesday, June 21, 2006 1:25 AM
Subject: Re: [time-freq] RefLock II behavior


> This looks similar to, but is orders of magnitude larger than, what I saw 
> when using the Reflock II to control a 10MHz TCXO from GPS.
>
> We concluded what was happening there was that the Reflock D/A was 
> dithering between two discrete steps, causing the noisy state for a while, 
> then the loop would move for a while to a stable state where the voltage 
> wasn't dithering, then it would drift off again.  Not sure if you're 
> seeing a magnified version of the same thing.
>
> John
> ----
>
> Owen, Michael wrote:
>> Dear friends,
>>
>> Thanks to Steve Bible, I got my brain un-stuck and figured out how to
>> set the 0-ohm jumpers (I'll summarize everything in a couple of days
>> when it's all finished and my shame has faded).
>>
>> For right now, I have a question.  RefLock plus my pretty good OCXO are
>> trying to settle at 100 MHz but there are sharp spikes every few
>> minutes.  Here's an image:
>>
>>  The time span is about 40 minutes.  Even after 3+ hours, the spikes are
>> still there (although more rarely).  Does anyone have an idea how to get
>> rid of them?
>>
>> 73, W9IP
>>
>> ------
>> Michael R. Owen, Ph.D.
>> Senior System Engineer
>> Sensis Corporation
>> (315) 
>> 634-3066 ------------------------------------------------------------------------
>>
>> _______________________________________________
>> time-freq mailing list
>> time-freq at lists.tapr.org
>> https://lists.tapr.org/cgi-bin/mailman/listinfo/time-freq
>
>
> _______________________________________________
> time-freq mailing list
> time-freq at lists.tapr.org
> https://lists.tapr.org/cgi-bin/mailman/listinfo/time-freq