[time-nuts] 1 pps Accuracy in two locations

[jimlux]

VLA and ALMA aren't exactly VLBI - they're all in the same (general) 
place and they distribute LO and timing signals to all the antennas.

Larry D'Addario, recently retired from JPL and now down at Caltech, 
described the setup at ALMA in 2009 - He's talking 1 microsecond, maybe 
100 nanoseconds.

https://library.nrao.edu/public/memos/alma/memo298.pdf

A more recent reference for ALMA (2018) describes a scheme using pairs 
of lasers which are transmitted via fiber heterodyned at the user end to 
regenerate a Local Oscillator - that gives you good frequency accuracy, 
but I don't know about absolute phase. They use fiber "line stretchers" 
to compensate for the fiber delay.  But I didn't see a performance spec. 
You can hunt if you like.

https://www.iram.fr/IRAMFR/ARC/documents/cycle6/ALMA_Cycle6_Technical_Handbook.pdf


  DSN does antenna arraying on receive as well. DSN uses fiber links in 
temperature controlled pipes underground with two way measurements. 
Calhoun 2007 gives a performance of about 1E-15 at 100 seconds, which is 
0.1 ps.  But that's a pretty darn elaborate system (costing the 
equivalent of many houses, I suspect, with Cryogenic Sapphire 
Oscillators for cleanup loops)

M. Calhoun, S. Huang and R. L. Tjoelker, "Stable Photonic Links for 
Frequency and Time Transfer in the Deep-Space Network and Antenna 
Arrays," in Proceedings of the IEEE, vol. 95, no. 10, pp. 1931-1946, 
Oct. 2007.
doi: 10.1109/JPROC.2007.905048



For radio astronomy, one can do a lot of post processing to "back out" 
the time shifts - if there's some bright source you can use that as a 
focusing target. It's not like CygA or CasA are going to be moving 
(other than from Earth rotation and movement)



On 1/15/20 4:12 PM, Dana Whitlow wrote:
> That is true.  Is post-processing not an option here?
> 
> Dana
> 
> 
> On Wed, Jan 15, 2020 at 5:05 PM Attila Kinali <attila at kinali.ch> wrote:
> 
>> On Wed, 15 Jan 2020 16:39:45 -0600
>> Dana Whitlow <k8yumdoober at gmail.com> wrote:
>>
>>> If anybody knows how to do this, I bet it would be the people who do
>>> long-baseline
>>> interferometry at millimeter-wave frequencies, such as at the eVLA in New
>>> Mexico.
>>
>> You probably know that much better than I do, but I thought that
>> VLBI applications usually did all the magic in post-processing.
>> Ie record everything with highly stable local clocks (aka hydrogen masers),
>> then feed everything to a large cluster of computers and let it figure out
>> what the actual time offsets between the stations were.
>>
>>
>>                          Attila Kinali
>> --
>> <JaberWorky>    The bad part of Zurich is where the degenerates
>>                  throw DARK chocolate at you.
>>
>> _______________________________________________
>> time-nuts mailing list -- time-nuts at lists.febo.com
>> To unsubscribe, go to
>> http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
>> and follow the instructions there.
>>
> _______________________________________________
> time-nuts mailing list -- time-nuts at lists.febo.com
> To unsubscribe, go to http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
> and follow the instructions there.
>