I have just read that this is possible.
The only way I can see it happening is for the little amplifier (which is inside all modern GPS antennae) to fail in a peculiar way which causes it to generate some muck.
Has anyone ever heard of this?
Yes.
Garmin had to replace my antenna and either they or my avionics shop paid for both the new antenna and the work (which is non trivial because the headliner needs to be removed and quite a lot of trim). Apparently this is rather embarrassing to Garmin and they pay for it no questions asked.
Has anyone ever heard of this?
Yes.
And if the amplifier starts to oscillate, it's almost guaranteed to oscillate (and radiate) close to the GPS frequency, due to the filter.
You are much more likely to see this with a WAAS antenna as it has a higher internal gain. This is less common than it used to be, but the only way to troubleshoot is to turn off each GPS, one at a time, to determine which one is breaking into oscillation and leave that unit turned off until the antenna is replaced. If it is your only GPS, revert to VOR.
That is a complete cockup, surely?
Don't they test these things, or is it a failure caused by some design defect which takes time?
I have a combined VHF+GPS antenna (from Comant) but it isn't the WAAS one.
If it is your only GPS, revert to VOR.
That would be highly unsatisfactory, for European IFR. Perhaps one should have a second panel mounted GPS which is non-WAAS?
Every mobile phone is a full function GPS device...
That would be highly unsatisfactory, for European IFR. Perhaps one should have a second panel mounted GPS which is non-WAAS?
Surely that would be a dramatic overreaction to a possible but still highly unlikely risk?
There were a batch of antennas that failed after they had been installed for a period of time. Most of them have been replaced. The only way you can use a portable or installed GPS when this would occur is to turn off the offending unit, otherwise all GPS units don't work as they all use the same L1 frequency.
Yes - a very good point.
A portable GPS will of course work as a "DCT box" (one use of my yoke mounted G496) but it won't do approaches, and certainly won't do autopilot coupled ones 
certainly won't do autopilot coupled ones
That's why we have to do two approaches each year by hand for the IR revalidation 8-)
