Jump to content

Tch - Threshold Crossing Heights & Air Distances

Don Hudson

Recommended Posts

I have been wrestling with a problem with a number of dimensions, those being what the formal and regulatory documents say, what the manufacturers say, what flight operations departments, what those who design airport approaches say and what pilots say. There is general agreement but when one examines it closely, that agreement appears to fall apart. The discussion is not a new one!

The question concerns the TCH, threshold crossing height, or rather the assumptions behind the "threshold crossing height" made by the various groups mentioned above.

We all know that the "TCH" is supposed to be 50ft or approximately that depending upon ILS Glideslope installation, PAPI installation, terrain leading up to the threshold and so on. But that isn't the question. .

But what part of the airplane crosses "50ft"?, and here is where those who I thought would know, differ!

In numerous documents that one should take seriously, (ICAO, FAA, Boeing/Airbus, etc), the TCH is "50", and, for example, Boeing means the gear. But is it really? From the airplane's (pilot's) point of view, how is that point determined?...the glide slope, or rather the airplane's GS antenna, the airplane's Radio Altimeter and in the end, the pilot's eye-height. All have varying locations depending upon the airframe.

Those who design ILS approaches KNOW that the intersection of the 3deg glide slope and the threshold of the runway is at 50ft above the threshold, period. There is no apparent consideration for which parts are at 50ft.

But the airplane's ILS antenna, the airplane's gear and the pilot's eye-height cannot all cross the threshold at 50ft together, yet some documents, some legal/certification documents, are VERY specific including the FAA Flight Test Guide for Certification of Transport Category Airplanes Aviation Circular, AC25-7C.

I ran across this because of the air distance values from "50ft" to "touchdown". We know that the TCH has a substantial effect upon the air distance from the threshold to the touchdown point. Higher means longer.

We also know that the height at the threshold must protect the airplane from obstacles.

When one uses an aircraft manufacturer's "Performance Advisory Information", (not the Certification Performance Information, with all the built-in factors but not using reverse), the specific requirements are that the threshold is crossed at 50ft, from which a touchdown is made 1000ft later, with a speed loss of 5kts. An IFALPA document states that the certification information means that the landing (air) distance is taken from "50ft" -that must mean before the threshold?

An FAA research document, (A Study of Normal Operational Landing Performance on Subsonic, Civil Narrow-Body Jet Aircraft During Instrument Landing System Approaches DOT/FAA/AR-07/7) wrestles with the same problem.

My own conclusions is that the ILS antenna, or on an approach using PAPI, it is the antenna or the pilot's eye that crosses the threshold at 50ft.

Otherwise, to satisfy many documents' requirements that the gear crosses the threshold at 50ft, the pilot would have to be between 15 and 75ft higher on either vertical guidance system.

For a B737 for example, the antenna is just in front of the pilots under the radome, (at pilot eye height), and the radio altimeter transceivers are mounted under the forward fuselage. The "eye-to-gear" height for those studies that actually recognized the problem was, for the B737, "15ft". So the gear is crossing the threshold at 35ft and the RA would read about 40ft or so, when "on glide slope"!

The air distance from gear height is slightly less than for 50ft. This reduces what I believe to be slightly incorrect air distance assumptions and computed values. Most of the time it doesn't matter because both landing distance factors in certification data and just plain, long runways, absorb a great deal of the "potential runway excursion" threat. But in critical operations or when one is examining such values as a basis for reporting then I think it must be more closely examined.

Anyone have any thoughts on this? Probably as dry as burnt toast I know but I'm curious.

Link to comment
Share on other sites

Guest rattler

Not a pilot but I did find the following information on the 747 that might be of interest and seems to be somewhat on subject.


Landing Geometry:

Two factors make landing the 747-400 a challenge from the perspective of the pilot; the long wheel base of the aircraft and relative height of the cockpit above the runway. To make consistently accurate and safe landings, it is important that the pilot have a firm understanding of the 747-400sgeometry in the landing configuration. The standard ICAO glideslope installation requires the glideslope to intersect the runway surface 1,000 feet from the threshold. In this configuration, a 2.5ºglideslope will have a runway threshold crossing height (TCH) of 66 feet. On the 747-400, however, the ILS receiver sare located on the nose gear doors, 21 feet below the cockpit. As such, if the aircraft is perfectly on glideslope at threshold crossing and flying at the Flight Director commanded pitch angle of 4º nose up, the pilot’s viewpoint will cross the runway threshold at 87 feet. The landing gear of the 747-400are located behind and below both the cockpit and the ILS glideslope receivers however, and will cross the runway threshold at only 44 feet. If the aircraft is flown to the runway in this configuration without a normal flare, themain gear will touch down approximately500 feet from the runway threshold. If a moderate flare is accomplished, rather than simply flying the aircraft onto the runway, the flight path of the main landing gear can be expected to lengthen by between 500 and 1000 feet. It is recommended that the aircraft be flared to touch down on the runway surface between 1,000 and 1,500 feet from the threshold. As such, the pilot should use the1,500 foot markings on the runway as the visual aim point for the approach. Coincidentally, this aim point will provide agood visual reference for flying both a 2.5ºand 3º glide slope, and result in an appropriately placed touchdown us ingnormal flare technique.


At 50 feet radio altitude above therunway surface, the throttles should bemoved to idle. At 30 feet radio altitude, noseup pitch should be increased from theapproach angle to approximately 6º noseup. If accomplished correctly, the aircraftshould settle onto the runway withoutextended floating.Keeping power added during the flare maycause extended floating in ground effect justabove the runway surface, which willsignificantly increase landing distance.Crews are likewise cautioned not to continueto increase nose up pitch during the flare asthis may cause a rapid decay in airspeed,reducing aircraft controllability and reducingthe effectiveness of immediate go aroundthrust should it be needed. In addition, apitch attitude of 11º nose up will causefuselage contact with the runway surfaceupon main gear touchdown.The recommended approach and landingtechnique is to fly a visual aim point 1,500feet down the runway. Reduce thrust to idlebeginning at 50 feet, with the flarecommencing at 30 feet. Fly the aircraft ontothe runway surface and commence therollout procedure.Effective use of this procedure willconsistently result in a runway touchdownbetween 1,000 and 1,500 feet from thethreshold


Link to comment
Share on other sites

Rattler, that quote helped sort out a few things, thank you. The graphic below from the image that Rich supplied seems to verify what the quoted passages are stating.

Rich, thanks for the photograph - great shot!

I haven't found evidence that the RA is calibrated to put the wheels at the 50' TCH (with everybody riding on top of that height). However, your photograph was really useful - here's a very rough attempt at measuring the various dimensions:


The baseline, known dimension is the height of the vertical stabilizer from the ground. That's about 55ft. Because the photo is almost "straight-on", one could reasonably calculate the other heights using the known dimension. I copy-pasted the red line to keep the length, (FastStone Capture won't let the user rotate lines, so the risk was in altering the dimension when putting all of them together - using the cos of the 4° angle to determine the vertical dimension, the difference was ~ 4" ). The dimensions are as shown in the graphic. The brown line is a check for mistakes in the other calculations...wheel base is ~ 85ft.

If the aircraft was spot on the glideslope, then the gear height here is about 38ft.

The dark grey and light grey horizontal lines are theoretical runway heights for pilot-eye-height-at-50ft, & glideslope-antenna-height-at-50ft respectively.

Really interesting!

Of course the entire drawing rests upon the assumption that the airplane was right on the glideslope but there's no reason to assume otherwise until compared with further data.

Thanks guys!

Link to comment
Share on other sites

Yikes! Don, try reading the yellow pages to get some sleep!! :lol:

That's a helluva series of questions. Out of all the terms from TCH to eye-to-wheel-height, the only thing that makes sense to me is that the UHF glide slope beam is aimed so as to pass over the runway threshold (not stopway, not clearway, not displaced threshold) at 50', or slightly greater if close-in obstacles are a factor - one reason for a non-standard (>3 degree g/s). If this is true, then the only answer to your question can be the glideslope antenna placement on the underside of the aircraft. I can't remember: are g/s antennae mounted side by side, or fore/aft of each other?

And again, if that is a logical statement, then is it not up to the aircraft manufacturer to ensure the g/s antennae placements give sufficient clearance of all parts of the aircraft as the aircraft's g/s antennae cross the threshold?

I yearn to learn!

The rest of the questions you ask about don't seem to factor into the discussion from my very, VERY limited understanding of the subject.

Link to comment
Share on other sites

Hi Moon - yeah, sleep , my wife asks the same question :lol: but in truth this puzzle has been developing for some time and it just spilled out! I believe that standard landing distance FOQA events are introducing distance errors because of the assumption that the wheels cross the threshold at 50' or more accurately, "50ft" is where the threshold is, when in fact it is the glideslope antenna that does so and the gear is along for the ride. I think A320 and B737 gear cross the threshold at 35ft because the GS antenna is in the nose and the pilot's eyes are 15ft above the gear. Thirty-five feet to touchdown will produce different numbers than 50ft to touchdown. I was actually surprised at the numbers that came out of Rich's photo. The real issue is, are FOQA "long landing" events really "long"?

Link to comment
Share on other sites

Hi Robert;

Interestingly, from the studies I've read, (DOT/FAA/AR-07/7), autoland distances are well within a normal distribution of manual landing distances as seen below in a graph from the FAA study. In other words, they're a "model" for normal. In fact, some line indoc guys had suggested that watching "how the autopilot did it" on the Airbus was a good teaching tool !

From "A Study of Normal Landing Performance on Subsonic, Civil, Narrow-Body Jet Aircraft During Instrument Landing System Approaches", p. 35, 36


A limited number of landings were conducted with the autopilot and A/THR engaged until touchdown (511 in total). This number of autolands conducted by each of the four aircraft types is too small for meaningful statistical analysis. Therefore, only the results of all aircraft types together are considered. Figure 23 shows the comparison of airborne distance of autolands and manual landings for all aircraft types. As can be expected, the autolands have a lower average airborne distance than manual landings and also show less deviation from the average airborne performance. These findings are logical because autolands are not influenced by any human performance during the airborne maneuver. As a result, a more consistent and shorter airborne distance is realized during an autoland.

On published CRFI data, yeah, it's a "wow...!". The data is also found in FCOM Advisory Performance Inflight information sections, (the data that does not include any extra factors and includes the use of reverse).

(edited PST) Hi Moon - re "deep" ;-), - it's just looking at the nuts-and-bolts of what's meant by "air distance" in terms of actual runway performance in the context of published data and the risk of runway excursions which is all this stuff is about. Depending upon the year examined, Runway Excursions are either #1 or #2, sometimes #3 on various "Most Wanted" lists, alongside CFIT and LOC. For many flight safety departments running FOQA Programs the idea is to increase awareness of "how close did we come?". Refining the exact touchdown point is the first step. That question led to reading the definitions of what "50ft" meant, how glideslopes were designed and I found that (within narrow limits), the industry itself has different answers. What struck me most was, the only part of the airplane that is at the design TCH is the GS antenna. And the GS is almost universally 3°, Placing the GS antenna further below the pilot's eye height on long body aircraft assures wheel clearances just as mrlupin states below. Because there are limited places one can put the GS antenna, there are going to be variations in gear clearances and therefore air distances to touchdown.

Link to comment
Share on other sites

Hi Don,

You would even get variations going from short to stretched models. I was just looking at the B777, the GS antenna are on the nose gear doors on that aircraft. I wonder how much change you would see for gear height over threshold on a 300 vs a 200. It's probably minimal but since the aircraft is longer, the trigonometric calculation would change... Then one would also have to consider any tail strike protection logic that might be in the longer aircraft and might lower flare angle.


Link to comment
Share on other sites

Interesting discussion. A TSB report regarding a 2007 Global Express accident at Fox Harbour, NS highlighted some eye-to-wheel height (EWH) issues, including misunderstandings of EWH amongst pilots and a lack of available information on EWH. http://www.tsb.gc.ca/eng/rapports-reports/aviation/2007/a07a0134/a07a0134.asp

Apparently EWH is defined as "the vertical distance from a pilot's eyes to the lowest portion of the aircraft in the landing attitude".


It was determined that, in general, pilots are not aware of the EWH of the aircraft they operate. Some pilots wrongly believe that the EWH is the distance between the pilot's eyes and the ground while the aircraft is sitting on the ground, as this information is published in the AFM by some manufacturers..........Furthermore, the topic of EWH is rarely addressed in any type of pilot training.

The accident crew wasn't planning to cross the threshold at 50 ft. They were operating onto a short runway and were looking for an early touchdown. They had recently transitioned from a Challenger 604 (no slats) to a Global 5000 (longer aircraft with slats). They positioned the Global visually as they would have the Challenger and struck the ground short of the runway with the main landing gear. This image shows the problem of the differing EWH and approach attitudes:


As with any accident, there were a myriad of factors but a failure to fully appreciate the dimensions and orientation of the aircraft behind them just about cost this crew and their passengers their lives. All of that to say I believe your question is one worth pursuing, Don.

Link to comment
Share on other sites

Hi Éric;

Re, "You would even get variations going from short to stretched models."

Yes, exactly.

All Jepp and CAP charts have "TCH" indications on the ILS approach plates. I learned that PAPI visual indications had a not-unexpected wider tolerance for heights over the threshold because it's a visual aid. The systems are wildly successful given their long histories. Interestingly, I don't see TCH's on RNAV / GNSS approaches.

Link to comment
Share on other sites


Thanks very much! - I was not aware of the accident. The TSB Report includes a statement indicating that pilot awareness of eye-to-wheel heights is not what it should be. In August of 2012 TC issued an Advisory Circular, (AC 302-009) concerning such awareness. The Report is worth reading.

Link to comment
Share on other sites


The Radio Altimeter on the aircraft is calibrated to read 0' with weight on wheels. 50' RA would put the mains somewhere between 47 and 49 feet over the asphalt depending on aircraft type.

Link to comment
Share on other sites

Rich: while they look similar, the Challenger 604 and the Global express look similar they are very different aircraft. The Global is much longer and has leading edge slats. The profile on landing is far different between the two airdraft which is part of what let to the accident. The challenger doesn't "Flare" so much as level off at 0'. the Global has a nose high attitude on approach as would be expected from an aircraft equipped with leading edge devices.

Link to comment
Share on other sites


The Radio Altimeter on the aircraft is calibrated to read 0' with weight on wheels. 50' RA would put the mains somewhere between 47 and 49 feet over the asphalt depending on aircraft type.

This is not my experience. The 767 RA reads -6 with weight on wheels. When skimming the ground (as infrequently as that might happen!) on landing, it reads zero. That would lead me to believe that it reads the actual wheel height in the approach configuration.

The following two links show the position of the GS and RA antennae on the 737 and 767.



These indicate that the G/S receiver antennae are in the nose of the aircraft while the RAs are about halfway back to the wheels (but the actual height indicated on instruments by the RA is obviously adjusted, or it would read 8 ft with weight on wheels).

The airbus A380 is here (search for antennas and probes). The RAs are behind the gear on that aircraft with the GS antenna in the nose gear door area... probably a good idea, given how low to the ground the aircraft would be if they put it in the traditional radome location. (The 320 locates it in the radome, as well.


The 340 has the GS antenna in the nose wheel door area on the -500 and -600, while the -200 has it higher up.

All of this would lead me to believe that there is an effort to have the wheels of all transport aircraft have the wheels at a similar absolute altitude on approach for the same glideslope beam position.

I read recently that the TCH in Canada is reported based on the size of aircraft normally using the airport (can't find the reference right now).

If it was as simple as saying that the receiver antenna would cross at 50 ft, there would be no need to adjust the reported value so, by extension, it must be adjusted based on aircraft size... ergo wheel height. Small aircraft would obviously have a slightly higher TCH at airports that are calibrated for larger aircraft.

Based on a cursory evaluation of the above, at least in Canada, there is a pretty good chance that most transport category aircraft landing at airports at which they are the normal size of aircraft for that airport will cross the threshold, if on the G/S, at very close to the same wheel height, whatever that might be... but I'm guessing it is very close to 50 ft.

Clearly, there would be some variance with respect to the visual PAPI indication and the G/S indication.

Link to comment
Share on other sites

I was just going by the certification process we followed. we had to test the RA to read zero with WOW and calibrate it that way. Other manufacturers may do it differently depending on the antenna locations. On these aircraft the RA Antennae were quite far aft of the main landing gear. in a flare that may actually be the same height as the main wheels.

Link to comment
Share on other sites

Hi inchman, re "This is not my experience. The 767 RA reads -6 with weight on wheels. When skimming the ground (as infrequently as that might happen!) on landing, it reads zero. That would lead me to believe that it reads the actual wheel height in the approach configuration.",

Nor was it my experience, when I was flying and even now in reading data. For a B737, the RA reads about -3 or so. Like definitions of when the threshold is crossed, how the RA is actually calibrated and what we see appear to be different, and so this aspect of the approach too, has "very definite" statements which fundamentally differ. It's for this reason I brought this whole thing up here - beyond just determining actual air distances from flight data, it's confusing!

Great link to the A380 AMM section on description, thank you. I see that the GS antenna for tracking is near/at/on the nosegear doors which is where it seems to be for wide-body aircraft. I think your assessment of why the GS antenna locations are as they are, certainly is logical.

Hi boestar - thanks too, for your information - your statement regarding RA calibration concurs with the B737-400 AMM - that the RA is calibrated to read zero on the ground. Perhaps a fine point regarding the gear being over the asphalt with an RA value of 50', I think I would disagree slightly only because the RA isn't governing aircraft position over the threshold, the GS antenna is, and the RA value will be somewhat less than 50ft depending upon the aircraft and the GS antenna location on the airplane.

For the B737, the GS antenna is right in front of the pilots underneath the radome, making the EWH about 15ft and thus a gear clearance of about 35ft when on glide slope. Looking at it from Rich's and canadairguy's photographs, gear clearance varies with type while a standard-installation 3deg GS ensures sufficient wheel clearance. As we all likely know already, PAPI is categorized for height; the CAP provides P1 through P3 category PAPIs on each approach page's Airport Chart. The Jepps don't seem to indicate PAPI category.

From the CAP:

P1 is for aircraft with EWH up to 10'

P2 " " " 25'

P3 " " " 45'

Ap is Abbreviated PAPI for EWS's of 10'

Link to comment
Share on other sites


There in lies the issue being discussed. in the Global Express accident mentioned above, the GS antenna is under the radome. The RA is at the tail several feet behind the main gear. The pilot was used to flying the challenger which is a much shorter aircraft and does not flare so much as fly onto the runway. The Global being longer and nose high on the final approach led to the accident. The pilot flew the approach as he would have using his eyes to place himself on the runway however the rear end of the aircraft was several feet lower than he was used to. While radio altitude did not come into play in this instance the GS is 3 degrees no matter if you are in an A380 or a C-172. Radio altitude will give you your actual height above the Terra Firma which would equate to you ACTUAL threshold crossing height, The glideslope wil only dictate the height of the antenna over the threshold the other few hundred feet of aluminum will be somewhat lower.

Link to comment
Share on other sites

I came across an interesting TC advisory circular. https://www.tc.gc.ca/eng/civilaviation/opssvs/managementservices-referencecentre-acs-300-302-009-1202.htm Its purpose is to provide guidance for the installation of PAPI units and to enable harmonization of PAPI with ILS. While not aimed at operators, it does help to understand the geometry.

It introduces some new terms (to me):

- Eye-to-(GS) Aerial Height (EAH), which I think would be the blue line minus the yellow line on your illustration, Don.

- Wheel-to-Threshold Height (WTH)

- Minimum Eye Height over Threshold (MEHT). Crossing the threshold on the lower limit of the PAPI , MEHT=EWH+WTH


The fact that the pilot's eyes are not collocated with the GS antenna leads to the following guidance (with diagram). The paragraph numbers don't match the AC paragraph numbers. (I'm not sure how to do that.)

  1. As a general statement, the objective for "harmonization" of PAPI with ILS is to obtain a condition for which the pilot, when the aircraft is following the ILS signal, will observe an "on path" display [two whites and two reds] for the PAPI. That is, the pilot should obtain the same presentation from both the ILS instrumentation and visually from the PAPI.

  2. However, this is not possible to achieve completely because the pilot eye is above the aircraft antenna which senses the ILS signal and the angle "B" of the lower edge of the PAPI approach corridor is less than the angle of the ILS signal.

  3. The geometry is such that the pilot eye may exit from the approach corridor at some point prior to the threshold when the aircraft is following the ILS. Figure 5 illustrates a case for which the pilot eye exits from the bottom of the approach corridor.

  4. Similarly, for a particular PAPI location and pilot eye to aerial height (EAH), it might occur that the pilot eye exits from the top of the approach corridor, but this situation should be avoided for the installation design, since it would present a "fly-down"instruction to the pilot.AC-302-009-5.jpg

  1. In Figure 5 the TCH is the crossing height of the ILS beam which is normally taken to have a value of 50ft (15.24m), although it may vary from 40ft to 60ft. The designer should consult with NAV CANADA as to the exact value. Although Figure 5 indicates the runway intercept of the ILS beam is at the GPI antenna location, this also is not necessarily the case and thus the designer should not use the physical location of the Glide Path Intercept (GPI) antenna as a basis of calculating the TCH or positioning of the PAPI. The position of the PAPI should always be referenced to the threshold.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


  • Create New...