Error in glide slope indication

[Jeroen D]

Yesterday, during a ground school session for my PPL we were discussing Instrument Landings.

Apparently it's possible to get a erroneous glide slope indication. So you need to use the approach plate information to verify that at a given distance/altitude from the runway you are receiving a 'valid' glide slope indication. One of the reasons given for a false reading was the radio waves bouncing off water.

Anybody experienced this in real life? It also made me wonder how do the big commercial planes, like the 744, deal with this? Do they have more advanced equipment that filters out these erroneous indications?

What would happen if you're flying a ILS approach on auto and all of sudden the glide slope moves a couple of hundred feet higher or lower? Plane would be all over the place!

Jeroen

[Mariano]

Jeroen,

Take a look at this video: http://www.youtube.com/watch?v=GelRBhJ4gmI

There was also an incident in which an Air France 747 approaching Rio de Janeiro encountered erroneous glideslope indications and nearly ended up on a hill below the approach path.

I was taught that the best way to protect yourself from erroneous glideslope indications is to perform the distance/altitude check you talked about. The advent of EGPWS (and its Terrain Awareness Alerting feature) also provides protection against this phenomenon.

Mariano

[John H Watson]

What would happen if you're flying a ILS approach on auto and all of sudden the glide slope moves a couple of hundred feet higher or lower? Plane would be all over the place!

Not necessarily. The flight control computers compute an inertial path after the aircraft has established itself on the G/S. Short term signal loss or sudden excessive deviations are recognized as such and the aircraft continues along its inertial path for a short time period, say 30 seconds, before A/P Cautions appear.

Rgds
JHW

[Jeroen Hoppenbrouwers]

I once did ILS "practice" (no formal training, but with a qualified instructor) in a C172 around Den Helder/De Kooy (EHKD) and was specifically warned about yet another ILS feature, the ghost side lobe. Basically a second (and sometimes third) glideslope null that is at 6 degrees angle, 9 degrees angle etc. If you catch that, everything looks OK except of course for a double V/S and associated altitude indication.

The glideslope needle was waving back and forth a lot and totally not stable until we were about 7 nm out. With altimeter and DME you could still maintain a proper situational awareness, but it threw all faith I had in the system out of the window and replaced it with "hey, this is real analog electronics, it is alive" ideas.


Jeroen

[Holger Wende]

Approximately a year ago I had the opportunity to talk to an ATC system maintenance engineer in Munich and he told me that sometime pilots send reports with something like this:

"Hey, today your glideslope was really bumpy".

Then a maintenance engineer analyses the ILS system and calibrates it if necessary.
He confirmed that this happens from time to time.

Regards, Holger

[frumpy]

A ghost lobe led a 737 to crash in 1988:
http://aviation-safety.net/database/record.php?id=19880102-1

[effte]

Additional glide paths above the nominal one is a side effect of how the system builds the signal in space. That's why you not only should do the altitude check once established, but why you must do it.

The LOC also has false nulls outside of the course sector, just outside of 35 degrees off. These are reverse-sensing.

GPs all have structure, some more than others. This is due to the transmitting equipment, but also and in no small part due to the ground conditions in front of the GP mast. The signal in space is built by reflecting the signal from the GP antennae in the ground plane, thus effectively building a mirror image of the GP mast underground - hence the name imaging GP antenna. Some GPs are straight as an arrow, others show significant distortions.

Aircraft receivers are, however, not very sensitive to these fluctuations. They operate using a time constant of three seconds, so fluctuations of even rather long duration are smoothed out. No INS required, luckily.

The facilities are flight checked with receivers using a 0.5 second time constant, meaning structure which will not be detected by normal receivers will stand out clearly. If outside of tolerances at a flight check, the facility is rendered U/S until it can be corrected. If you adjust the facility, a flight check is required.

Nominal GP coverage is ten miles, so at 7 nm it should be reasonably stable. No idea what was going on there.

A "bumpy" GP is typically caused by the clearance signal being reflected off something on the ground up through the GP. The clearance signal is a lobe at a slightly different frequency emitted below the normal GP lobe to ensure full fly up signal if you stray below GP. If this is reflected up, you get fly up when established on the nominal glide - a "bump in the beam". It has to be very significant to be noticeable though, due to the way receiver selectivity works. 45 degree noise abatement berms can do it.

If the GP drops offline, you get a flag and then you go around. The GP signal is only disregarded during the very final segment during CAT II/III approaches, when radio altimetry and possibly INS is substituted, as the signal in space becomes distorted in close.

The signal is monitored, both for GP/LOC position and sensitivity. If either parameter is exceeded, it switches to the backup transmitter. If that doesn't do it, it drops offline.

GP angles and sensitivities vary, especially as ground conditions change. Snow is especially bad. That's why you see significant effort being spent on clearing out the area in front of the GP mast during snowfall. Too much snow will lead to the facility going outside of tolerances and dropping offline.

Don't hesitate to ask away if you have more navaid questions. I'll do my best to answer.

Cheers,
   Fred