[time-nuts] GPS Antennas

[Bob kb8tq]

Hi

Based on the papers I have seen, there is no “magic” shape that can be ignored. 
Part of the issue is that the cover is going to be in the near field of the antenna.
As noted in another post, the impact is more complex than a simple delay. ( consider 
that signals bounce off the cover material …. ).

Also consider that if I’m going to 3D print the device, I have a limited number of 
materials available. Teflon isn’t going to run through a 3D printer. Unless you paint
the gizmo it will not be water tight so humidity will get into the act as well. 

=====

A “good” GPS antenna should have roughly the same phase center regardless of
the angle to the sat it’s receiving. Indeed GPS antennas in general get this right.
There is a modest variation with angle, but it’s not all that great. Units of mm are not 
uncommon. It will be greater on a multi band design. The impact of your cover also 
will be greater on a multi band device. All the various delay /. loss / reflection parameters
are frequency dependent…..

Bob

> On Sep 6, 2019, at 5:57 PM, Hal Murray <hmurray at megapathdsl.net> wrote:
> 
> 
> kb8tq at n1k.org said:
>> The gotcha with putting anything on top of a GPS antenna is that it impacts
>> the phase  shift from the sat’s. How important that is gets quickly into just
>> how picky you happen to be. 
> 
> How picky do I have to be before that is significant?  How geeky would I have 
> to be to measure it?
> 
> Would a hemispherical cover provide equal delays at all elevations?  I think 
> the solidly mounted USGS fault/volcano monitoring setups have a round top.
>  https://www.usgs.gov/media/images/gps-tracks-ground-deformation
> 
> Do any antenna manufacturers advertise uniform phase delay?
> Do any software vendors advertise that their software works better with 
> specific antennas?
> 
> 
> 
> -- 
> These are my opinions.  I hate spam.
> 
> 
> 
> 
> _______________________________________________
> 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.