[time-nuts] SE880 GPSDO

[Azelio Boriani]

Here in Italy the DVB-T practice is to use GPSDOs to lock the
transmitters' carriers. This is needed to exploit the Single Frequency
Network (SFN) system. The frequency and time stability are stated in
the ETSI TR 101 190, 8.3.1 and 8.3.2 on page 41, in short they say 1us
is enough for the time and about the frequency it is 1/1000th of the
DVB-T carriers' spacing that is, at worst, 1116Hz so 1.1Hz of
stability.

<http://www.etsi.org/deliver/etsi_tr/101100_101199/101190/01.03.02_60/tr_101190v010302p.pdf>

On Fri, Apr 29, 2016 at 12:35 AM, Ilia Platone <info at iliaplatone.com> wrote:
> Il 28/04/2016 23:22, Attila Kinali ha scritto:
>>
>> On Thu, 28 Apr 2016 21:01:49 +0200
>> Ilia Platone <info at iliaplatone.com> wrote:
>>
>>> Thanks Attila, I know how to build a transmitter and a receiver, and now
>>> is more clear the system you designed. But as I will propose this system
>>> to an astro club, and in this astro club there's the possibility that
>>> not all would have a radio license, I need something "free-to-play", if
>>> it concern.
>>
>> Ok.. that's quite some constraint. This rules out any kind of
>> transmission.
>>
>>> I was wondering if it would be more convenient to lock to a signal from
>>> an AM broadcasting station, if available to a multiple of the OCXO. What
>>> do you think about?
>>
>> AFAIK most radio and TV transmitters are using some stable reference.
>> I don't know though what they use these days. It used to be an Rb.
>>
>> I would guess that using a radio station should in general work.
>> It should be as close as possible, so that you get as little
>> reflection as possible and that any multi-path from the troposphere
>> and ionosphere is minimized. If you still have any AM stations close
>> by, that would work. But these are more and more switched off and
>> replaced by digital broadcast systems.
>
> I know that some AM station still exists, a place where to setup the
> telescopes will have the local repetitors very close.
> I think that airports use AM modulation, but I sincerely don't know if it's
> legal even to listen at those frequencies. Their signal should be strong,
> however.
>
>> The most common radio and TV transmitters these days are DAB and DVB-T.
>> Both use QPSK or QAM signals. This makes locking to those signals
>> quite a bit more difficult. What you can do is, use a DAB/DVB-T
>> tuner chip like the MAX3580 or MAX3541, down convert the signals,
>> then use the FPGA to track the signals and steer the OCXO's EFC DAC.
>> Yes, this is a lot more complicated and you need to build quite a bit
>> of a DVB-T/DAB receiver in the FPGA. Fortunately, this is something
>> people have already implemented in software using GNURadio. Ie you
>> can have a look at what they have done, copy the over the parts that
>> you need. But still, this will be quite some serious effort and will
>> take you months at best.
>
> Will inform on AM stations.
>>
>> I also have no idea what the signal stability of the DVB-T and DAB
>> stations is. Maybe someone else (Magnus?) can comment on that.
>
> I would appreciate his contibution :)
>>
>> As such... I think using an AM station that is close by would be feasible.
>> Using DVB-T/DAB stations would be a lot of effort and I would advise
>> against it in a first step. GPS alone should give you ~1ns when done
>> right.
>> With more expensive equipment (high qual geodetic or timing antennas with
>> L1/L2 receivers) you should be able to go below that (see Michael Wouters'
>> mail).
>> An alternative approach would be to use an Rb reference instead of an
>> OCXO at the telescopes. This way you have a frequency stable reference
>> that you can use like the reference signal I mentioned in the other mail.
>> You would need one that has low phase noise (that rules out the FE-5680's
>> that are so cheap on ebay, ie you would need to go for LPRO, PRS10, FRS
>> or LPFRS). As now you only have a kind of stable reference, but you don't
>> know how far off it is (and probably not how fast its drifting), some
>> precision will need to be spend on determining its exact frequency.
>> But nonetheless it should give you additional precision when doing
>> the post-processing that you can use to increase the timing solution's
>> precision.
>
> The problem is not the absolute stability, but the relative error between
> the various stations. ie the telescopes clocks must not drift too much by
> each other.
>>
>>
>>                         Attila Kinali
>>
>
> --
> Ilia Platone
> via Ferrara 54
> 47841
> Cattolica (RN), Italy
> Cell +39 349 1075999
>
> _______________________________________________
> 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.