[time-nuts] How do I compare GPS antennas?
kb8tq at n1k.org
Wed Sep 6 08:54:45 EDT 2017
> On Sep 5, 2017, at 10:23 PM, Hal Murray <hmurray at megapathdsl.net> wrote:
> Subject: Re: [time-nuts] GPS Antenna Feed Line Decision
> kb8tq at n1k.org said:
>> There is pretty much no experiment you could run that would show a
>> difference between the two. With a normal GPS, the “front end” of the radio
>> is in the antenna. The filtering and RF amplification there determine a lot
>> of things. The cable is just a chunk of wire in the middle of the system.
> Does that depend on the antenna (and location) being "good" and both coaxes
> being good-enough so that the receiver always has a good signal?
> Alternatively, if the signal is good, you can't tell the difference in a few
> db of attenuation.
It’s a cascaded noise figure problem rather than an attenuation problem. Given
the noise figures involved, once you get to about 10 db of gain, the impact on the
overall noise figure is negligible.
> But suppose the antenna location isn't good. How can I tell if it is
> good-enough? Or how can I compare location A with location B?
If your antenna location is “bad” your sky view is likely cluttered up with terrestrial
“stuff”. The clear sky is “cold” and terrestrial “stuff” is hot relatively speaking. The
higher the noise temperature, the less your noise figure matters.
> The best I have been able to come up requires two identical receivers. You
> can verify that they are identical, or at least close enough, by running them
> from a single antenna with a splitter. I haven't gotten past that.
Mostly what you will measure on antenna A and antenna B is the effect of
local multipath. Yes, if you have two “ideal” antenna locations to compare
that will be a bit less than with normal ones. Not many of us have clear view
30 M tall concrete towers to mount our antennas on.
> Assuming you had a not-good antenna, is there any numerical scale that would
> be useful to describe its goodness?
At some point you simply have to dig into the theory and the math. Measure the
components on a network analyzer / noise figure meter / whatever. Run the numbers
and see what you get.
Somewhat more to the point of Time Nuts. Your best timing solution comes from the
sat’s that are straight overhead. Those are the ones with the strongest signals and
the least impact of all these noise figure issues (and a bunch of other stuff). The close
to the horizon sat’s are what help with the X/Y part of a survey solution …..
> These are my opinions. I hate spam.
> 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.
More information about the time-nuts