Probably zero distance. Common GPS antennas contain only an LNA and don't emit anything that would interfere with another antenna system.
The only possible consideration might be where you're doing advanced science or extremely accurate surveying where you're tracking the actual phase on the incoming carriers, and then you'd need to be careful to place each antenna on its own little ground plane on its own cute little tripod. You have to keep each one out of the others near field to avoid the possibility of distorting the phase of the carriers. It's very likely that is not a concern for you.
Another option is that you can 'split' (in the Cable TV sense of the word) GPS antenna systems are share the one antenna with as many receivers as you wish. The only exception here is that an airplane connot depend on only one antenna, so the system needs to be designed to fail gracefully if one antenna blows off.
If a GPS antenna emitted noise on the GPS frequencies, then it would interfere with itself.
This sort of concern is most-often associated with embedded local oscillators where Unit A can interfere with Unit B. A classic example of this is where one automotive Radar Detector can trigger off another Radar Dectector.
Normal GPS antennas do not contain a local oscillator. They most often contain a nice LNA powered by 5Vdc on the same coax cable.
A 0.5m gap is pretty big from an EMI point of view (for these circumstances) - I was imagining you meant cheek-to-jowl side-by-side.
On the other hand, 0.5m is perhaps a bit tight if you're planning to use differential methods to extract platform orientation (if I'm reading between the lines correctly).
If the application is in any way life-critical, then it might justify further engineering examination as opposed to relying on some guy on the Internet. Have seen that old New Yorker cartoon where two dogs are surfing the 'net, "On the Internet, nobody knows you're a dog."
No i am not planning to use differential methods for orientation, they are 2 patch antennas for 2 seperate systems, room is just a bit tight!
I totally agree with your dog analogy, hee hee, i was checking to see if there was common knowledge out there, you know the type of knowledge 'EVERYONE knows you don't put two patch antenna's within 0.5m of eachother dummy!'
What you have said makes perfect sense to me, i guess i will continue to find some concrete evidence.
If you're tight for space, and you don't require the redundancy of two separate antennas, then don't forget to consider the GPS Splitters. They can be cheaper overall because of the relatively high labour cost to install and maintain the second antenna.
And if you DO require two antennas for redundancy, then perhaps they should be further separated so that when something bad happens to one, it doesn't also take-out the nearby backup.
