ILS: Errors and Operations

Identification and the watchdog

Trust starts with knowing what is being received: an ILS identifies itself by its frequency and by a Morse-code identifier, sometimes supplemented by plain text, checked before the needles are used, exactly as with the VOR (VHF omnidirectional radio range). And as with the VOR, the station watches itself: an ILS installation has an automatic ground monitoring system. What it watches is specific: the monitoring system detects any shift in the LOC (localiser) or GP (glide path) mean course line, and any reduction in signal strength: the two failures that would quietly move the line in space the aircraft is following.

On the aircraft's side, the receiver guards the display with flags: warning flags appear for both the LOC and the GP when the received signal strength falls below a threshold value. Beyond weak reception, the flags appear under three specific circumstances:

absence of the carrier frequency;

absence of the modulation, simultaneously;

the percentage modulation of the navigation signal reduced to 0.

All three say the same thing in different ways: the 90-Hz and 150-Hz comparison that steers the needle has lost its raw material, so the needle can no longer be trusted and the instrument says so.

Reading the needles

The ILS display looks like the VOR's, but the scale underneath it is far finer. Full-scale deflection of the CDI (course deviation indicator) needle corresponds to approximately 2.5 degrees of displacement from the ILS centre line, and full-scale deflection on the GP corresponds to approximately 0.7 degrees from the ILS GP centre line. Compare the VOR's 10-degree full scale: the same needle travel now reports a quarter of the angle laterally, and a sliver of a degree vertically. On the familiar five-dot face this makes one LOC dot half a degree. The same interpretation applies on an HSI (horizontal situation indicator): the deviation bar and the GP pointer read against their dots on these ILS scales whenever an ILS is tuned.

The back beam in practice

The last lesson left the LOC's back beam as a published non-precision approach; here is how it reads in the cockpit. The key fact: the 90-Hz and 150-Hz lobes keep their fixed sides of the extended centre line, in front of the antenna and behind it. What flips is the pilot: flying inbound on the back beam, the aircraft faces the opposite direction, so the lobe that was on the left of a front-course aircraft is now on the right. A plain CDI knows nothing of heading, only modulation depths, so on a back-beam approach the needle shows reverse sensing: it deflects toward the side of the extended centre line the aircraft is on, and the pilot must fly away from the needle. That is also the interpretation rule: needle left means the aircraft is left of the extended centre line, needle right means right.

The course selectors sort this out unevenly. On an OBI (omnibearing indicator), the course selector has no effect at all on an ILS deviation: unlike the VOR, the needle is driven by the modulation comparison alone, so the selector is set to the front-course inbound track purely as a reminder, and the reverse sensing on the back beam remains. The HSI genuinely fixes it: its deviation bar rotates with the compass card, so with the course pointer set to the LOC front course, on a front-beam and on a back-beam approach alike, the bar keeps its correct sense and the pilot flies toward the bar. One setting rule covers both: the HSI course pointer carries the front course, always.

The categories

How low an ILS approach may be flown depends on the installation as much as the aircraft: per ICAO (International Civil Aviation Organization) Annex 10, ILS approaches are divided into facility performance categories. The operation categories, defined by decision height (DH) and RVR (runway visual range):

Category I — DH not lower than 200 ft, with visibility at least 800 m or RVR at least 550 m;
Category II — DH lower than 200 ft but not lower than 100 ft, RVR at least 300 m;
Category IIIA — DH lower than 100 ft, or no DH, with RVR at least 175 m;
Category IIIB — DH lower than 50 ft, or no DH, with RVR below 175 m but at least 50 m;
Category IIIC — no DH and no RVR limitations.

Category III changes what the beams are for: for all Category III operations, guidance information is provided from the coverage limits of the facility to, and along, the surface of the runway: the signal is flown all the way to the ground, and on IIIB and IIIC through the roll-out. That is also why the accuracy requirements are progressively higher for CAT I, CAT II and CAT III: the lower the decision height, the less height and time remain to recognise a guidance error and fly away from it, until at CAT III there is no margin at all and the beams must be good enough to be followed blind to the surface.

Flying tolerances

The pilot's own accuracy has numbers too, from ICAO Doc 8168, and they are the VOR lesson's criterion transplanted onto the finer scale: to be considered established on an ILS track, the LOC must be flown within half-full-scale deflection of the required track, and the aircraft has to be established within half-scale deflection of the LOC before starting descent on the GP. On the glide path itself, the GP is to be flown to a maximum of half-scale fly-up deflection, in order to stay within protected airspace: fly-up deflection means the needle commanding up, the aircraft below the path, and below half scale the obstacle protection surveyed for the procedure no longer holds.

Past those limits the instruction is not to correct harder but to leave: if the pilot deviates by more than half-course deflection on the LOC, or by more than half-dot deflection on the GP, an immediate go-around should be executed, because obstacle clearance may no longer be guaranteed. Notice how much tighter the go-around trigger is stated for the vertical channel: half a dot on the GP, against half the course scale on the LOC. On approach, height is what there is least of.

What bends the beams

The signal itself is not perfectly straight. ILS beam bends are deviations from the nominal LOC and GP respectively, and they can be assessed by flight test: an aircraft flies the beam with precise reference equipment and maps where the radiated line wanders from the geometric one. A major source of trouble is multipath interference, caused by reflections from objects within the ILS coverage area: buildings, terrain, and above all large aircraft and vehicles near the antennas, whose reflections add to the direct signal and move the apparent centre line.

That last source is managed on the ground, with two defined zones:

ILS-critical area — an area of defined dimensions around the LOC and GP antennas, where vehicles, including aircraft, are excluded during all ILS operations;
ILS-sensitive area — extends beyond the critical area; there, the parking or movement of vehicles, including aircraft, is controlled to prevent the possibility of unacceptable interference to the ILS signal during ILS operations.

This is why, in low visibility, holding points move back and taxi clearances slow down: the geography of the aerodrome is protecting the beam.