Figure: Instrument landing systems, from AFM 51-37, Figure 16-2.
The Instrument Landing System is fantastic, except when someone drives through the antenna beam. Then? Not to fantastic. But it is still the approach of choice just about everywhere in the world.
If you are looking for ILS fundamentals:
But if you've always wondered about "what exactly is 'on course' for an ILS and how much distance you have between you and the trees, well that's here:
The final question, of course, is when can you arm your flight director for the approach? The answer, of course, is it depends on your flight director. Before you begin the approach you need to be on a segment of the approach or above the MSA. While descending on the glide path, you need to have less than half-scale CDI deflection, as is explained below.
What follows are quotes from the relevant regulatory documents, listed below, as well as my comments in blue.
Figure: Instrument landing systems, from FAA-H-8083-15B, Instrument Flying Handbook, Figure 9-33.
[FAA-H-8083-15B, Instrument Flying Handbook, pg. 9-35] Instrument landing system (ILS): An electronic system that provides both horizontal and vertical guidance to a specific runway, used to execute a precision instrument approach procedure. The ILS system provides both course and altitude guidance to a specific runway. The ILS system is used to execute a precision instrument approach procedure or precision approach. The system consists of the following components:
[FAA-H-8083-15B, Instrument Flying Handbook, pg. 7-27]
Figure: Localizer coverage limits, from FAA-H-8083-15B, Instrument Flying Handbook, Figure 9-36.
[FAA-H-8083-15B, Instrument Flying Handbook, pg. G-9] Categories of instrument approach procedures allowed at airports equipped with the following types of instrument landing systems:
All you need to fly a Category I ILS approach, besides the aircraft and ground equipment, is an instrument rating. If you want to go for a higher category, you will have increased aircraft maintenance requirements, a few training hoops to jump through, and you will need some form of authorization. See FAA Order 8900.1, Volume 4, Chapter 2.
The ILS is easy, you can find them at most airports, and you can use airports with ILS approaches as alternates.
The localizer and glide slope beams are subject to interference. There I was, flying a Boeing 747 down to ILS Cat II minimums, just 150' in the air when a truck drives through the localizer beam at the opposite side of the runway. The localizer beam decided it needed to be several hundred feet to the right and the airplane decided it needed to chase the beam with maximum bank angle. But I digress.
Figure: Concept for suppresses-image antenna, from ILS Suppressed Image Antennas, Figure 2.
[ILS Suppressed Image Antennas, page 253.] The most common glide slope antenna reflects an image of its signal off the surface in front of the antenna to create the mirror image of the signal. A more expensive antenna does this electronically and is therefore easier to site. The former can be thrown off by surface snow and the latter by snow in front of the antenna.
I've heard that in either case the beam is deflected upward so you won't be closer to the ground but your descent rate might be more than you bargained for. The airport should have test data to indicate when the glide slope is bent beyond category limits and should NOTAM the glideslope and prohibit descent below localizer minimums. At KBED, for example, 18 to 24 inches of snow in front of the glide slope antenna can bend the beam beyond 3.1 degrees, the limit for Category D aircraft.
[FAA-H-8083-15B, Instrument Flying Handbook, pg. 7-27] The ILS and its components are subject to certain errors, which are listed below. Localizer and glide-slope signals are subject to the same type of bounce from hard objects as space waves.
Figure: Localizer final trapezoid, from Eddie's notes, redrawn from TERPS, Vol. 1, Ch. 9, ¶904, figure 75.
[TERPS, Vol. 1, Ch. 9, ¶903] The final approach dimensions are specified [in the figure]. . . . Calculate the width of the area using the following formulae:
Perpendicular Width from RCL to the Edge of the Primary = 0.10752 (D - 200) + 700
Perpendicular Width from RCL to the Edge of the Transitional Sfc = 0.15152 (D - 200) + 1000
Where D = Distance (ft) from RWT measure along RCL.
RCL is runway centerline and RWT is runway threshold.
In the example shown, when 50,200' (8.26 nm) from the end of the runway, the primary area extends 6,076' (1 nm) either side of centerline. On a localizer only approach, without a glideslope, the required obstacle clearance (ROC) will be 250' below the MDA. But a glide slope changes everything. . .
Figure: Final Segment OEA/OCS, from TERPS, Vol. 3, Ch. 3, figure 3-1.
[TERPS, Vol. 3, Ch. 3, ¶3.0] The area originates 200 feet from LTP or FTP and ends at the PFAF/Glide path intercept point (GPIP). The primary area consists of the “W” and “X” OCS, and the secondary area consists of the “Y” OCS. See figure 3-1.
TERPS appears to draw the OCS right down to the surface but you have to consider the glide slope. What is important to understand at this point is the lateral dimensions are a bit tighter on the ILS than the localizer-only.
Figure: Localizer final trapezoid, from TERPS, Vol. 1, Ch. 2, figure 1-2.
[TERPS, Vol. 1, Ch. 2, ¶203.a.] The obstacle evaluation method for descent of a glidepath is the application of a descending OCS below the glidepath. The vertical distance between the glidepath and the OCS is ROC; i.e., ROC = (glidepath height) - (OCS height). The ROC decreases with distance from the final approach fix as the OCS and glidepath converge on the approach surface baseline (ASBL) height (see figure 1-2). The OCS slope and glidepath angle values are interdependent: OCS Slope = 102 ÷ glidepath angle; or glidepath angle = 102 ÷ OCS slope. This relationship is the standard that determines the ROC value since ROC = (glidepath height) - (OCS height).
Figure: "Run over rise," from Eddie's notes, redrawn from TERPS, Vol. 1, Ch. 2, ¶203.
The OCS formula may look simple but it is made needlessly complex by the Byzantine TERPS practice of "run over rise" slopes, which is opposite of how engineers and pilots think. The slope of a 3° glidepath, for example, is found by dividing the run, say 6,076' for a nautical mile, by the height at that point, which would be 318' — so, 3° slope = 6076 ÷ 318 = 19.11 : 1.
[Table 2-2a. Maximum GPA's, from TERPS, Vol. 3, Ch. 2, ¶2.5.]
|A (80 knots or less)||6.4|
|A (81 - 90 knots)||5.7|
|D & E||3.1|
Figure: Solve h given d and α, from Eddie's notes.
The OCS on a glide path is fairly standard given the distance from the glide path's intercept with the runway and the glide path angle itself. If an obstacle penetrates the OCS, the glide path angle is raised until it doesn't. If the glide path angle then exceeds the maximum for the aircraft category, that category of approach is disallowed.
To find the height of the glide path at any given distance, multiply that distance by the tangent of the glide path angle. For example, a 3° glide path is (6076) tan ( 3° ) = 318' at 1 nm, (5)(6076) tan ( 3° ) = 1592' at 5 nm.
To find the minimum height above obstacle along a glide path, you need to figure the slope of the glide path angle, the slope of the OCS, convert the OCS to an angle, compute the height of each, and subtract. Simple! For example . . .
Given a 3° glide path
Glide path slope is 6076 ÷ 318 = 19.11 : 1.
OCS slope is 102 ÷ ( 3° ) = 34 : 1
OCS angle is arctan ( 1 / 34 ) = 1.68°
Glide path height at any given distance is d tan ( 3 ° )
OCS height at any give distance is d tan ( 1.68° )
ROC at any given distance is the glide path height minus the OCS
For a 3° glide path:
|Distance (nm)||Distance (feet)||Glide path height||OCS height||ROC|
Figure: T-37B Course Indicator, from Technical Order 1T-37B-1, figure 4-8.
I flew my first ILS looking at a course indicator and was told I could consider myself on course when the CDI came "off the wall." On a procedure turn, for example, it was okay to descend to the inbound altitude once the CDI was no longer fully deflected. Years later I adopted a "centered or nothing" attitude, thinking I am better than a nearly fully deflected needle. But it begs the question, do we have the required obstacle clearance when the needle just comes off the wall. If not, what about one dot? FAA-H-8083-15B, Instrument Flying Handbook, pg. 7-27, says one-quarter scale deflection means the airplane is aligned with the runway and full scale deflection shows when the aircraft is 2.5° either side of centerline. (The text says 2.5° and the diagram says 1.4°, we'll use 2.5° to take the worst case scenario.) Will we have the required obstacle clearance at full scale deflection?
Figure: Localizer trapezoid versus 2.5° CDI full scale deflection, from Eddie's notes.
You cannot come up with a one-size fits all rule about where full scale CDI deflection places you because the trapezoid begins 200' before the runway and the angle on the CDI begins at the localizer antenna, on the other side of the runway. We can, however, provide an example. If the runway is 1 nm long (6,076'), and the localizer antenna is 1,000' from the opposite end of that runway, and we check our displacement at 5 nm (30,380'):
This lends credence to the old conventional wisdom that once you are "off the wall," you are close enough to being on course to call it on course. What about the tighter lateral dimensions of the ILS on the obstacle clearance surface? The TERPS example is for 50,200' so using the same methodology:
The needle comes "off the wall" about 300' too early, but that is based on being right on the obstacle clearance surface. Drawing the slope of the secondary area (4:1) you could argue that you are still covered 2,500' either side of centerline as long as your intercept takes place at least 1,200' above the OCS (the 4:1 that gets you the extra 300' of lateral displacement). But when flying the intercept how are you going to know exactly where that OCS is? You won't. But it doesn't matter for those of us who also fly outside the United States. . .
[ICAO Doc 8168, Vol 1, §4, Chapter 5, ¶5.5.5.]
22.214.171.124 The width of the ILS/MLS/GBAS final approach protection area is much narrower than those of non-precision approaches. Descent on the glide path/MLS elevation angle must never be initiated until the aircraft is within the tracking tolerance of the localizer/azimuth.
126.96.36.199 The protection area assumes that the pilot does not normally deviate from the centre line more than half-scale deflection after being established on track. Thereafter the aircraft should adhere to the on-course, on-glide path/elevation angle position since a more than half course sector deflection or a more than half course fly-up deflection combined with other allowable system tolerances could place the aircraft in the vicinity of the edge or bottom of the protected airspace where loss of protection from obstacles can occur.
Conclusion: You might have the required obstacle clearance when descending on an ILS glide path as long as the CDI is just "off the wall," but the ICAO wants you within a half-scale deflection. Using a half-scale deflection also removes all uncertainty from the TERPS secondary area as well.
Portions of this page can be found in the book Flight Lessons 1: Basic Flight, Chapter 29.
Air Force Manual (AFM) 51-37, Instrument Flying, 1 December 1976
FAA-H-8083-15B, Instrument Flying Handbook, U.S. Department of Transportation, Flight Standards Service, 2012
ICAO Doc 8168 - Aircraft Operations - Vol I - Flight Procedures, Procedures for Air Navigation Services, International Civil Aviation Organization, 2006
Lopez, Alfred. Suppressed-Image ILS Glide Slope Antenna, ARL Associates Incorporated, Proceedings of the 51st Annual Meeting of the Institute of Navigation, Colorado Springs, CO, June 1995, pp. 253-259.
Technical Order 1T-37B-1, T-37B Flight Manual, USAF Series, 30 September 1959
United States Standard for Terminal Instrument Procedures (TERPS), Federal Aviation Administration 8260.3B CHG 26, 02/24/2014