Touchdown Point on ILS: how does FOQA do it?

[Capn Bloggs]

Did a search for FOQA but nothing came up, so here goes.

I'm interested in finding out how the FOQA system works out where the aeroplane actually touches down after an ILS (or can it work it out for any landing?). If anybody knows or can point me to the answer, I'd be grateful.

Thanks, Bloggs.

[john_tullamarine]

Take your pick for a simple solution ..

(a) vertical G shows a spike for my landings or, for the aces of the base .. there should be a variation in G trace between air and ground

(b) altitude trace becomes constant

(c) WOW flag

.. or use (a), (b) and (c) as a check each on the other

(d) or, if you are REALLY bored .. do it with a bit of geometry and simple speed-and-distance-back-of-a-fag-packet calculations

At the end of the day .. once the PC is programmed to flag and monitor any event, it is all done automatically in any case

I have no doubt that people with more time to waste than I have .. have found other, more elegant, ways to find the answer .. mainly this will depend on the type of data available from the box .. the more modern types of recorder (DFDR/QAR) have the capability to record many parameters. The main message is that it is pretty much a straight forward exercise ..

Surprised that you couldn't find anything on the net .. try the UK's CAP 739 (link is on the Tech Log URL sticky) for starters

Useful to keep in mind that FOQA (or one of the other buzz terms) is not a magic thing .. it is a disciplined approach to using flight recorder data to seek improvements in procedures and SOP discipline .. not to mention the maintenance trend monitoring use of the data .. the guys and gals doing the work have to figure out just how they are going to go about it .. while keeping themselves nice along the way with respect to a bunch of regulatory, general legal, and Industrial imperatives.

[Capn Bloggs]

J_T,

Thanks, but my question was where, not when. The link is very interesting but doesn't describe how the system works out the touchdown point.

[Karunch]

Our Qar's trigger a deep landing when greater than 12 seconds from passing 50ft on approach till touchdown. As such I don't think it actually determines a touchdown point, merely an excessive float, Cheers

[john_tullamarine]

Capn Bloggs,

I suspect that you may have missed the point .. the folk running the FOQA system have to develop their own techniques to determine the touchdown point .. I gave a few methods and there will be a number of others, depending on the recorder data captured for a given unit. FOQA is not a magic bullet .. it is a generic description for systems put in place to use recorder data for monitoring and related purposes. But, like any procedural gadget an organisation might develop to do something .. someone in the organisation has to do the developing .. or buy a similar system from someone else who previously did the hard yards.

[Capn Bloggs]

Karunch,

Thanks for that. So I guess the FOQA system could, by knowing where the aircraft is WRT the glideslope and the Rad Alt, then be able to predict where the actual touchdown point is by timing the remainder of the flight.

I suppose that, if not on an ILS (and especially if you are above or below 50ft over the fence), the calculated touchdown point may well be spurious.

[TruBlu351]

On a 3* glideslope, mulitply your actual or published TCH (threshold crossing height) by 20 and that's your touchdown point in feet.

ie: 50ft TCH x 20 = TD on the 1000ft mkrs. (+ any extra for greasing it on in ground effect)

[john_tullamarine]

... close, but no banana ... aiming point doesn't equal touchdown point.

[TruBlu351]

Originally Posted by john_tullamarine

... close, but no banana ... aiming point doesn't equal touchdown point.

Yeah, I had to quickly back paddle and slip that caveat in after posting it. Unless you like struts popping out the top of your wing

[019360]

ESTIMATING TOUCHDOWN POINT FROM FOQA DATA
Here are some notes on the way we estimate Touchdown point for our FOQA system for the B777. This is one method. The most interesting is to use a FOQA track overlay on a surveyed and scaled satellite photo of the airport and note where the MLG goes from Tilt to No-Tilt. We only do this for hard landings mostly as they are mostly short landings. I am not sure why someone cannot automate a GPS touchdown point and overlay it onto a known lat/long series for a particular RW off the Jepp database.
• Find a point on the approach where there is surveyed flat terrain (e.g. at 100 ft RA for a CAT II RW).
• Add the correction between Radio Alt and MLG (approx 21.2 ft). This figure is derived from the assumption (airline specified) that when aft MLG touchdown happens at say 5° Pitch the RA should read 0 ft. Your airline technical department should know this.
• Apply correction for RA ht due to terrain at the Cat 11 DH which has been surveyed usually. This gives the height of the NLG Glide Slope aerial at that point.
• Then correct for G/S deviation (1 dot = 0.35 degrees) and for the actual TCH. If TCH is 50 ft then the G/S antenna is 954 ft down the RW etc. Some simple geometry. Therefore you have one side and one angle of a triangle and can calculate height of the gldeslope at that point and hence the Distance To Go.
• Since RA is measured each second and the aircraft is travelling at xxx fps it will travel xxxx ft before touchdown. You should determine the distance from the glide slope receiver on the NLG doors to the aft MLG wheel and that will tell you at your specified FOQA derived Radio Altitude that the aft MLG is XXXX ft from the EOR and has XXX ft to run so touchdown will be XXX ft from the end of the RW.
• To be most accurate the effect of changing ground speed (if any) between that point and touchdown must be calculated. And to be very accurate add in the effect of aircraft pitch, MLG tilt and the MLG strut extension.
All very pedantic and time consuming but worth it for the valuable training lessons that come from FOQA data. This is a very condensed verion of the process but I hope it helps. It is only going to be accurate to within say plus or minus 125 ft but thats not too bad to pick up the major causal factors in hard landings.

[MrBernoulli]

What does FOQA actually stand for chaps?

[john_tullamarine]

Flight Operations (or Operational) Quality Assurance ... one of a number of buzz terms ..

[Screwballs]

Re: aiming point & touchdown point:

I have been wondering about the question of where you actually touchdown for quite some time now. If you fly a B737 and follow a nominal 3 degree GS then, if all things being equal, the 1000ft markers should be your aim point. And as jt stated, your aim point can never be your touchdown point (unless you like being unemployed). So where is your touchdown point? The Boeing FCTM states for a normal approach, when aiming at the 1000ft marker, 50ft TCH, Vref +5, and a flare producing a 'Boeing' landing gives you a flare distance 1000ft to 2000ft beyond the threshold. (Pg. 257 in Adobe, ref 6.9)

And looking at it's US & ICAO runway markings, (Pg. 256, ref 6.8) it paints the aim points 1500' to 1312' into the runway respectively, with touchdown zone markings 500'-2075' US and 492'-2952' min ICAO.

Now what exactly does this mean? Does it mean that you will make a safe landing so long as the wheels touch down within the touchdown zone? If you took a runway that was 8000 feet long and if you used the ICAO touchdown zone (3000ft) and landing at the far end of it, that's 37.5% of your runway eaten up already. Which takes up a fair chunk of the 60% you are allowed use for calculate your landing there in the first place.

Hopefully I haven't rambled too much... However it is an interesting point and any useful answers would be appreciated - always helps when you see those aim points drift under the nose at 130+ knots...!

Cheers

[Capn Bloggs]

Flight Operations Quality Assurance, I think!

019360,
Thanks a lot. It seems that an accurate estimation of the touchdown point would not be possible without a glideslope, as to need it provide the angle. If you didn't have a glideslope, you'd have to assume the aircraft was on the 3° slope WRT the VASI to even come close to producing decent data.

[mangatete]

FOQA, pronounced ("FOLK QUAR") gains data from the quick access recorder on board the aircraft. The event associated with deep touchdown or short touchdown is based on timing from crossing the threshold. This position is acquired from various inputs not limited to the following; DME, Altimeter, Radio Altimeter, G/S on an ILS, Ground Speed. An event will be provided for both an ILS and Non ILS approach and Landing. It is worth noting that the Data is not gained from the Visual Approach guidance systems and these are set to provide a varying difference in TCH ranging from at least 73ft down to in some cases 45ft. This is to suit the "Pilot eye Height" of different aircraft and is normally set for the most critical to give sufficient clearance. On an ILS approach the Glide slope will normally be set to give 50' TCH above the on board glide slope antenna. The aircraft's antenna can be positioned on the aircraft to provide the appropriate aircraft height over the threshold. Now FOQA alerts are normally set for between, less than 1000ft to greater than 2000ft, (300m-600m). Point to note is Runway touch down markings do vary between runways. The "Fixed Distance Marking" large white blocks can be set to start at 1000ft or 1500ft depending on various reasons which is another topic. Prior to these markings are the "Touchdown zone markings" these are the thinner white stripes placed either side of the runway centre line. These "Touchdown zone markings" are located 500ft (150m) from the threshold and then 500ft (150m) after that. Now if you can see two sets of these before the large "Fixed Distance Marking" then these are set at 1500ft (400m) if you can only see one set of touchdown markings (thin white lines), before the Large white "Fixed Distance Markings" then these are set at 1000ft (300m). To help stay in the FOQA limit range use a visual aiming point that is going to have MLG contact runway just past 1000ft point. IE insure you know which is the 1000ft point on the runway, as FOQA does dot make allowances for the different markings. Using a visual aiming point will give a fly up indication on the VASI or PAPI in some cases, however remember the visual approach aid is set to provide guidance for differing "Pilot eye Heights" and these can vary by more than 30ft between aircraft.

Cheers Mang.

[019360]

well said Mang

It might seem a bit geekish to get into the numbers but if you are an instructor or even line captain it is good to understand what is happening and the geometry etc.

[FlightDetent]

The Teledyne AirFASE software evaluates

LONG FLARE:

as time from 30 ft RA to touchdown with limits of 10, 12, and 14 seconds.

LONG TOUCHDOWN (ILS approaches only):

monitored distance from THR (Runway Threshold, position on the ground where ILS glideslope crosses 50ft above ground level). Input parameter is G/S deviation, RA, and distance measured by integrating longitudal acc to speed and then distance. Factory limits are 750 / 900 / 1050 m.

Cheers,
FD
(the un-real)

[darksecrets]

Reality is that FOQA systems will never (accurately) determine where A/C touchdown on the runway, as has been mentioned the FOAQ (depending on how clever) system can derive the predicted point, however this is not the touchdown point. Using a algorithm which calculates speed and time you can determine how much distance has been traveled from the Main gear TD to RWY turn off, and providing you have the information, you may be able to determine which exit/taxi way used therefore calculate the TD point. This only works well where you have few exits or even better a pan on a single runway, for example LHR would be a nightmare, where as BHX would be much easier. If your FOQA program has the addition of an animation program, then the above mentioned becomes a little easier and probably more accurate than using geometry.

[john_tullamarine]

.. come to think of it ... why would we want to identify the precise (as opposed to a reasonably accurate, but nonetheless approximate) touchdown position ? In my approach to data reduction and monitoring (which we have only recently started doing) I am far more concerned with overly short/long touchdowns .. can't say that I can see any need for identifying the touchdown better than, say, 30-50m for practical purposes .. and I reckon I can get that sort of notional accuracy without too much effort on many occasions .. and, comfortably, 100m. All depends on the data one has available with which to play for the particular kit, of course ..

.. always looking to learn, though, if I might ask for comments on the why ? Upon reflection .. training for naval carrier ops might be a case in point ... but for routine airline-style operations ?

[FlightDetent]

Personally, I do not see the need for such a precision in data monitoring. If needed for full depth investigation, one might do a skidmark analyses or whatever you can think of, for heaven's sake. If G/S is within 1 dot at 50 ft, time from 30 to touchdown is good enough as I see it.

The only problem is that short flare limit is 5 / 4 / 3 seconds and I keep busting the 5 sec limit,