Thanks for the paper.
One should first know that there is a lot of papers now on frequency
transfer over fiber. The stability achieved on the best ones so far
greatly below that of the optical clocks that they want to compare.
Then, for those links able to transfer phase/time, most of them is for
point-to-point systems, many relating to relatively short distances.
Only a few relates to larger distances and some form of network style, mesh.
They fill different purposes, and should not be compared between the
groups, as it changes widely what is meaningful.
I did a presentation some EFTFs ago on some experiences of time-transfer
systems. We had some "interesting" failures where the delay jumped 1 ms.
That is what happens when underlying system re-route one side of a
two-way transfer under the feet of you. These are challenges others
don't see, but that comes with commercial telco setups. For some reason
I know far more about behavior and delays in radio links now than I
thought I would need to, again for reasons that would never show up in
dedicated systems. So, the challenges shifts with the field.
On 09/01/2018 08:18 PM, Bob Martin wrote:
> Here is an interesting paper on using fiber:
> I believe it used some of the gear that is(was) used in the GPS ground
> stations around the world.
> Bob Martin
> On 9/1/2018 3:29 AM, Magnus Danielson wrote:
>> It was very telling when I crashed a research group into the reality of
>> phase/time transfer over fiber compared to frequency transfer. Armed
>> with a whiteboard and pens, I derived the forumulas and showed how they
>> worked and not worked. It's a completely different ball-game and their
>> "known tricks" ain't doing nothing good as it comes to time.
>> I had to figure much of this out myself as I did nation-wide system
>> design to achieve the goal. It's a combination of many skills that goes
>> into designing the full system from scratch and make it fit together.
>> It's not hard stuff, it's just many details one needs to get right.
>> Oh the fun.
>> On 08/31/2018 05:15 PM, Bob kb8tq wrote:
>>> That works fine if you are doing things manual to check a local
>>> standard. If you are trying to
>>> disipline a few thousand cell towers 24 hours a day … not so much. It
>>> also works for
>>> checking frequency. What modern systems need is time. That gets you
>>> into a whole
>>> world of resolving and identifying individual edges. The WWVB signal
>>> really was never
>>> set up for this. Loran-C is an example of a signal that was designed
>>> to identify a specific
>>>> On Aug 31, 2018, at 10:30 AM, Martin VE3OAT <***@storm.ca> wrote:
>>>> But the diurnal phase shifts at VLF are predictable and largely
>>>> repeatable. Ignore the phase at night and use only the phase
>>>> records during the day when an all-daylight propagation path
>>>> exists. You might have to "correct" the absolute phase reading by
>>>> some multiple of the RF period, but with a low rate of local
>>>> standard oscillator drift, this is a simple matter of arithmetic.
>>>> Back in the day, I managed Sulzer crystal oscillators at 5 field
>>>> sites from my office and could maintain phase continuity for weeks
>>>> at a time, until we had to diddle the dial on one or several of them
>>>> to correct for crystal aging. Then it was just more arithmetic
>>>> again. Several of the oscillators had such low drift rates that all
>>>> I needed was one daily phase reading from the VLF phase tracking
>>>> receiver (Tracor 599Js) at those sites to know the frequency of the
>>>> Sulzers there.
>>>> ... Martin VE3OAT
>>>> On Thu, 30 Aug 2018 12:27:12 -0400
>>>> Bob kb8tq<***@n1k.org> wrote:
>>>>> WWVB as transmitted ( = right at the input to the antenna) is a
>>>>> wonderfully stable signal. As soon as
>>>>> that signal hits the real world things start to degrade.
>>>>> Propagation between transmit and receive sites
>>>>> is a big deal, even at 60 KHz. On top of that, there is a*lot* of
>>>>> manmade noise at 60 KHz. The receive
>>>>> signal to noise will never be as good as you might like it to be ?.
>>>>> I don't know about WWVB, but for DCF77 it's known that sunrise/sunset
>>>>> causes a phase shift of several 100?s at even moderate distances
>>>>> (like ~500km). Unfortunately I don't have any measurements at hand.
>>>>> Attila Kinali
>>>> time-nuts mailing list -- email@example.com
>>>> To unsubscribe, go to
>>>> and follow the instructions there.
>>> time-nuts mailing list -- firstname.lastname@example.org
>>> To unsubscribe, go to
>>> and follow the instructions there.
>> time-nuts mailing list -- email@example.com
>> To unsubscribe, go to
>> and follow the instructions there.
time-nuts mailing list -- firstname.lastname@example.org
To unsubscribe, go to http://lists.febo.com/mailman/listinfo/time-nuts_lists