[time-nuts] Accuracy/drift of Garmin GPS 16 HVS 1 PPS output under invalid fix conditions...

[jimlux]

On 3/11/19 6:27 PM, Hal Murray wrote:
> Steve Olney said:
>> The professional guys were impressed with my data, but slightly  disappointed
>> it didn't have accurate time-stamping. I am working to  rectify that for the
>> next glitch.
> 

 >
 > What do they need for accuracy?


It depends on your baseline (assuming you're doing interferometry) - 
uncertainty in sampling instant turns in to uncertainty in angle.
And then it depends on the integration time (which winds up being the 
tau in your ADEV spec, eh?)



 >
 > Some of the radio astronomy people are world class time nuts.  All 
the VLBI
 > stations have a Hydrogen Maser.
 >
 > Don't forget the delay through your radio.  You can avoid that by 
injecting
 > the PPS it a spare channel and using that for the low bits of the 
time stamp.
 > You need something else to number the seconds.  Plan B is use the PPS 
to keep
 > your system clock close to UTC and correct for the delay by measuring 
it once
 > and assume it doesn't change much.
 >


To answer all these sorts of questions (what performance do you need?), 
you want Thompson, Moran, and Swenson, "Interferometry and Synthesis in 
Radio Astrononmy", 3rd edition

which you can download for free (!) from Springer Open

https://link.springer.com/content/pdf/10.1007%2F978-3-319-44431-4.pdf

https://link.springer.com/book/10.1007/978-3-319-44431-4

This is the book all my radio astronomer colleagues point to.



(I'm building a radio interferometer in space using GPS as the timing 
reference to look at the sun - A mission called SunRISE )