[time-nuts] Re: TRAK 9000S : propagation delay or antenna lead-in compensation

Bob Camp kb8tq at n1k.org
Tue May 2 02:46:57 UTC 2023


Hi

Assuming you trust the cable delay setting and believe it works, then
you have one set of problems to solve.

If you don’t trust the setting to do what it says ( is 500 ns really 500 
and not 450 …), you have a second set of problems to solve. I assume
you are asking about the first part and not this end of it. 

There are many sources of delay in a GPS setup. The coax cable is only
one of them. Your antenna likely has a delay. The front end of the GPS
module has a delay. Even the PPS output driver is going to have a 
delay. 

The only way to calibrate *all* of that out is to compare to a calibrated
receiver. 

Unfortunately that’s not a full list of all the things that move your PPS 
around. The ionosphere and troposphere induce delay. It is not fully
compensated for by a single band module. Indeed only a very few 
multi-band devices do “active” compensation of these delays. One
answer to that is to subscribe to a data service that will tell you what’s 
going on “right now”. 

Because of uncompensated “stuff” moving the PPS around by 10 to 
20 ns a day, most folks don’t get to crazy about trying to get below 20
ns or so. Even at that level, antenna and module delays can be an issue.

Bob

> On May 1, 2023, at 6:19 AM, comsec22 via time-nuts <time-nuts at lists.febo.com> wrote:
> 
> Hi Folks,
> 
> Perhaps this topic has been already discussed in the past.
> The question is not only related to TRAK's products, but to those GPS
> receivers, capable of being able to compensate for the delay introduced by
> the length of the antenna cable.
> Once the value has been established, either thanks to the cable datasheets,
> or by measuring it with a VNA, how can it be instrumentally verified that
> the value found has actually compensated for the delay?
> I think 1 PPS signal ..
> 
> 73's de IU0CIX
> _______________________________________________
> time-nuts mailing list -- time-nuts at lists.febo.com
> To unsubscribe send an email to time-nuts-leave at lists.febo.com




More information about the Time-nuts_lists.febo.com mailing list