[time-nuts] GPS antenna with direction orientation?

[J. Forster]

> ---------------------------- Original Message ----------------------------
> Subject: Re: [time-nuts] GPS antenna with direction orientation?
> From:    "Tom Van Baak" <tvb at LeapSecond.com>
> Date:    Sun, April 3, 2011 6:58 pm
> To:      "Discussion of precise time and frequency measurement"
> <time-nuts at febo.com>
> --------------------------------------------------------------------------
>
>> A couple of years ago I picked up a surplus Aeroantenna choke-ring GPS
>> antenna that I think was intended for surveying use.  I finally got it
>> installed today and noticed that it has an arrow on the bottom
>> indicating that the antenna should be oriented with the arrow facing
north.
>>
>> I'm trying to figure out why an omnidirectional antenna should care
>> about which way it is oriented.  The best I can figure is that perhaps
>> it is for repeatability in surveying, so that any minor offset in the
>> phase center would remain consistent when moving the antenna from site
>> to site.
>>
>> Does anyone have a better answer?
>>
>> John
>
> That's my understanding too. Interpret the word "omnidirectional"
> like the words "cable length"; to a first approximation it is obvious
> and very constant, no worries; but if you look too many decimal
> places to the right, well of course, you will see variations.
>
> My Leica/Novatel/Areoantenna575 and Ashtech L1/L2 antennas
> have the same arrow. I pointed mine north only by eyeball. My
> hunch was that at this extreme level issues like cable tempco or
> receiver temperature were more important so I used FSJ1 heliax
> for the feed and kept the lab stable to 1C. I never revisited the
> angle issue.
>
> To be certain, and if your gear is sensitive enough, you can do
> an experiment and plot the phase center variation as a function
> of antenna angle. But given how the GPS constellation changes
> over time I think it would take many weeks before you could get
> results. I'd guess the asymmetry is well below 1 ns. If you have
> a spare ham antenna rotor then vary the angle over 360 degrees
> every N hours and then after you've collected many days of data
> look carefully for correlation between gps phase and antenna
> angle right around tau N hours.
>
> BTW, I gave up I gave up on this sort of problem when I realized
> that plate tectonics (mm to cm per year here in the NW) and local
> earthquakes had a greater effect on my long-term timing. See:
> http://www.leapsecond.com/pages/quake/
>
> If there are any cm-level surveyors on the list I'd like to hear their
> experience in the matter. I could imagine John's question has an
> effect only at the mm level, but not cm or meter level.
>
> /tvb


Correct.  The choke rings have practically perfect cylindrical symmetry
about the (vertical) axis of the antenna, but the feed -- at the center of
the choke rings, and hidden by a radome -- does not.  The reason for
orienting the antenna consistently in azimuth is to improve the
reproducibility of position determinations.

By design and construction, some antennas are more symmetrical than
others.  I was a millimeter-level surveyor.  My own relative-position
determinations were reproducible within less than 1 mm in both horizontal
coordinates and within 1 to 2 mm in the vertical, even when the
determination was repeated months later, with different antennas,
different receivers, different operators, and different data-processors;
but I _never_ bothered with azimuthal alignment of my antennas because, by
design and construction, they were azimuthally symmetrical within less
than 1 mm.  Most geodetic surveyors bought inferior antennas because they
were cheaper.

The symmetry of an antenna is best measured at a test range where the
antenna can be rotated while receiving a signal from a fixed source.  The
antenna must be rotated both in azimuth and in elevation.  Ideally,
far-field wave-fronts are spheres centered on an identifiable point fixed
with respect to the antenna's mount, and this point is the same for each
GPS signal band.

The main reason for departure from spherical symmetry is usually that the
antenna responds to signals arriving from below, after reflection from the
ground, in addition to signals arriving directly from a satellite in the
sky.  Any antenna having finite size will have non-zero response below
horizontal.  This result follows from physical optics, in other words wave
diffraction.  Larger antennas are intrinsically advantaged; but they are
heavier, harder to handle, and more expensive.  See above regarding what
most surveyors buy.

>From a friend,

-John

==============