Order Tray | Contact Us | Home | SIG Lists

[time-freq] Re: Febo.com's Z3801A Performance

John Ackermann N8UR jra at febo.com
Sat Jun 30 00:10:15 UTC 2007

Hi Keith --

I've been wondering about these spikes myself.  My gut feeling is that
they are probably the result of network or other queueing delays (what
we're looking at is the offset of the Z3801A system as seen by another
box across a 100mbit ethernet).  But why the system with the Z3801A is
consistently the noisiest of the three, while the one using LORAN-C is
the quietest, I'm not sure.  One possibility is that the LORAN-C system
is getting only a PPS signal and doesn't have a serial timecode coming
it; it's possible that somehow the jitter on the serial time marker
could be influencing the results.  But I'm really only guessing...

K&J Aggett said the following on 06/29/2007 07:25 PM:
> Hi John,
> I am curious about the regular, large jumps in offset shown in the
> Current Day graph.  These jumps are much larger than others that are
> present, suggesting that the minimum correction possible is of the order
> of 5 microseconds or so.  Now, assuming that the Z801A is performing
> well, we have a cosseted, high quality xtal oscillator which, apart from
> the relatively rare jump, should just drift gently along, requiring the
> occasional intervention of the controller to keep it on
> frequency/phase.  A further observation is that positive jumps are
> frequently followed by smaller negative jumps.  Over-correction
> somewhere perhaps?
> Is there any indication as to what these jumps mean?  Are they
> representative of actual changes in the Z3801A output or are they
> introduced elsewhere else in the system?  I have been wondering about
> strange spikes on some of my monitoring, too, but as yet lack sufficient
> data to do more than wonder.
> Keith Aggett
> time-freq-request at lists.tapr.org wrote:
>> Send time-freq mailing list submissions to
>>     time-freq at lists.tapr.org
>> To subscribe or unsubscribe via the World Wide Web, visit
>>     https://lists.tapr.org/cgi-bin/mailman/listinfo/time-freq
>> or, via email, send a message with subject or body 'help' to
>>     time-freq-request at lists.tapr.org
>> You can reach the person managing the list at
>>     time-freq-owner at lists.tapr.org
>> When replying, please edit your Subject line so it is more specific
>> than "Re: Contents of time-freq digest..."
>> Today's Topics:
>>    1. Supported operating system questions for (Gregg Levine)
>>    2. Re: Supported operating system questions for (John Ackermann N8UR)
>>    3. Re: Supported operating system questions for (John Ackermann N8UR)
>> ----------------------------------------------------------------------
>> Message: 1
>> Date: Thu, 28 Jun 2007 20:37:02 -0400
>> From: "Gregg Levine" <gregg.drwho8 at gmail.com>
>> Subject: [time-freq] Supported operating system questions for
>> To: time-freq at lists.tapr.org
>> Message-ID:
>>     <18d205ed0706281737q793727f8vab2478f9a0816f3 at mail.gmail.com>
>> Content-Type: text/plain; charset=ISO-8859-1; format=flowed
>> Hello!
>> Our Owner has his TAPR FatPPS connected to a Soekris system running
>> FreeBSD, and is making use of a Clock Block, for a better
>> understanding of what I am getting at and why I am bringing up the
>> merely obvious as opposed to the incredibly obvious, please revisit
>> http://www.febo.com/pages/soekris/ it explains in far greater detail
>> as to why I am doing that.
>> Now the question, will the pair work with the same hardware but with
>> it running Linux?
>> I freely admit that I had known at the TAPR site because they were the
>> first to offer in the US a board that translates a CF card into an IDE
>> form facter gizmo, it was mentioned in a copy of either LinuxJournal,
>> or their late ELJ newsletter that I have here. But that is an
>> inredible job of making three objects work together and do something
>> special. I should also mention that the Soekris people were also
>> brought to my attention in an article in the same magazine on the
>> subject of network managing a sump pump.
> _______________________________________________
> time-freq mailing list
> time-freq at lists.tapr.org
> https://lists.tapr.org/cgi-bin/mailman/listinfo/time-freq

More information about the time-freq mailing list