PDA

View Full Version : Delta A330-300 Lands Short in Amsterdam


JuniorMan
12th Jan 2023, 18:28
Interesting report coming from EHAM. Link (http://avherald.com/h?article=503a8162&opt=0)

ATC Watcher
12th Jan 2023, 19:38
The first questions that come to my mind are : Why use a short 22 runway when you ahave the main runway 24 with 3000 m and a wind 240 Gusting 40 Kts ? Or was 24 closed ? Or did ATC sent an A330 to Runway 22 due to noise abatment?

Track
12th Jan 2023, 19:51
Not unusual to have 22 as landing and RWY 24 as main TO rwy in these conditions. Very rare that you get 24 even in extreme winds. Approach for RWY 22 lets you fly over downtown Amsterdam so not a noise abatement reason.

( And 240/40 on a rwy 22 is not that bad actually, 15 kts crosswind...)

Avman
12th Jan 2023, 19:55
I don't think 24 is available as a landing runway for heavies. Back in the day I remember watching commuters make an approach on 27 and swing visually to 24 (shorter taxi time to their gates). I personally have never seen a heavy use 24 for landing. In fact, I don't think 24 even has an ILS.

Coastrider26
12th Jan 2023, 20:22
Rwy 24 has no ILS to my knowledge. But I have done an ILS27 with a break off to runway 24 with a 747

avionimc
12th Jan 2023, 20:56
Rwy 24 has no ILS to my knowledge. But I have done an ILS27 with a break off to runway 24 with a 747They all have RNP approaches with LPV minima.

V_2
12th Jan 2023, 21:57
The first questions that come to my mind are : Why use a short 22 runway when you ahave the main runway 24 with 3000 m and a wind 240 Gusting 40 Kts ? Or was 24 closed ? Or did ATC sent an A330 to Runway 22 due to noise abatment?

Also if you use R24 for arrivals, what runway are you going to use for the departures? R24’s 3000m runway is better utilised to get the long haul guys away

Consol
12th Jan 2023, 22:28
Landed on 22 in an A320, no big deal in a 330. Given that it has an ILS the real issue is how come they touched down so early. No glideslope GPWS?

sleeper
13th Jan 2023, 08:32
https://cimg2.ibsrv.net/gimg/pprune.org-vbulletin/960x540/fb_img_1673602141501_58f0a2b3ea22eef9111ed784de3b392a13d63d1 8.jpg
https://cimg3.ibsrv.net/gimg/pprune.org-vbulletin/800x369/fb_img_1673602153186_e686f1d19d8c15dd27a8b17e06bc5fa445bc811 2.jpg

swh
13th Jan 2023, 09:21
Just a ploy by Delta legal to avoid paying landing fees.

MATELO
13th Jan 2023, 10:20
Looks from those pictures just one set of wheels hit the ground.

Equivocal
13th Jan 2023, 10:24
I think you'll find that the two pics each show one of the MLG tracks. Looks like the aircraft was slightly left of the centreline.

Uplinker
13th Jan 2023, 10:54
That looks like an extremely high precision touch-down, almost exactly where the threshold starts. But wrong technique fella !

Or maybe they were on the glide but a gust pushed them or dropped them down onto the threshold ?

vilas
13th Jan 2023, 14:53
What's called brick one landing but ended being brick zero landing.

Iceman49
14th Jan 2023, 19:43
Flying into AMS prior to 2014, when 22 was in use, we used to circle to runway 24 in the 330.

krismiler
14th Jan 2023, 21:59
Given the waterlogged state of the undershoot area, they we’re lucky the gear didn’t dig in any deeper and end up being torn off when it met the tarmac. Obvious heavy rain that day so the weather conditions during the approach come into question, ie wind shear or misjudging the last few seconds with visibility lost in battering rain.

zerograv
14th Jan 2023, 23:40
lucky indeed ...

Given that the main landing gear on the 330 is tilted upwards when the aircraft is airborne, I would hazard a guess that what contacted the soft ground, just before the runway, were the wheels of the rear axle.
When the wheels of the rear axle came in contact with the runway edge, this would have caused a bit of a violent untilting of main gear, slamming the wheels of the front axle on the runway.

Potential to damage the gear interface unit, which can cause the system to think that the aircraft is in the air, making brakes and reverse unavailable, leaving the crew with nothing available to stop the aircraft after the touchdown. Iberia in Quito comes to mind.

safetypee
15th Jan 2023, 09:45
From an on-line AIP, runway 22 appears to have a single PAPI (3deg, 62ft), but no specific installation for long body aircraft.
This appears to to be the situation for all runways at EHAM; is this correct?

Runway 22 appears to have the lowest threshold crossing height of all of the runways - 62ft, agin is this correct?
Assuming this value to be eye height, then what would the wheel height over threshold be for an A330 (+/- wheel bogie tilt)?

Also, that runways 18 C and 22 have no hard-surface prior the the threshold markings;- deduced from online photo maps; is this correct?

https://eaip.lvnl.nl/2023-01-12-AIRAC/html/index-en-GB.html

Search EHAM, index of airports, EHAM AD 2

the_stranger
15th Jan 2023, 10:19
Also, that runways 18 C and 22 have no hard-surface prior the the threshold markings;- deduced from online photo maps; is this correct?

https://eaip.lvnl.nl/2023-01-12-AIRAC/html/index-en-GB.html

Search EHAM, index of airports, EHAM AD 2 The area in front of 22 used to be concrete,back in the olden days. Rumour has it the current grassy area is just a few cm's thick with the old concrete just below it.

Saab Dastard
15th Jan 2023, 12:32
Yes, there was concrete before the threshold as recently as 2008 (removed sometime between 2008 and 2012), as it's visible on google earth historical imagery. It doesn't provide any indication as to the depth of soil, however.

321XLR
15th Jan 2023, 14:33
any general theory yet, on what may have happened ?

Jhieminga
15th Jan 2023, 18:21
is this correct?
The document you're quoting from is the official Dutch AIP, so I would expect that information to be correct.

safetypee
15th Jan 2023, 21:18
Jh, thank-you.

321XLR, a first step is find details to answer "… what would the wheel height over threshold be for an A330" post #18.
Any A330-300 crew with access to the relevant documentation ?

pattern_is_full
15th Jan 2023, 23:15
While waiting for better data:

Here is a "back of the calculator.net" calculation, based on data from pdf page 53 (page 2, section 2-2-0 - diagram) of this Airbus document for the A330-300

https://www.airbus.com/sites/g/files/jlcbta136/files/2021-11/Airbus-Commercial-Aircraft-AC-A330.pdf

Assumptions - based on cockpit windscreen locations, the pilots' eyepoint will be 5.64m above the ground and main gear treads and 3m aft of the nose when flat on the ground. 5.64m is the diameter of the fuselage, and I make the assumption that the gap between the fuselage and the ground, and the pilots' eyepoint below the main cabin ceiling, are about identical.

This forms a triangle:

eyepoint > ground (5.64m)

•------------------------------------• (c)
> main gear tread (aft 29m)) > eyepoint with an angle of 11° (c) at the main gear.

Add to that whatever nose pitch-up is normal on approach - I assume +3°. Which makes the "pitch angle (c again)" between the pilots' viewpoint and the gear at touchdown 14°, and put the pilots' view point 7m above the main gear tread.
-•

-----------------------------------• (c)
Add 1m to that eyepoint-tread height to account for strut extension and eagleclaw-tilted bogie position, to estimate the pilot's eyes will be 8m (26.4 ft) above the trailing main gear wheel tread.

https://www.calculator.net/triangle-calculator.html?vc=&vx=&vy=29&va=90&vz=&vb=14&angleunits=d&x=77&y=20

Naturally...... feel free to critique or massage my assumptions and calculations.

safetypee
16th Jan 2023, 13:50
Thanks pattern; the wheel height over threshold for the A330 could be compared with other aircraft identifying the spread of heights that might be expected in normal operation. (See FAA report; note different references for TCH, ILS, RA, https://hf.tc.faa.gov/people/andrew-cheng/2007-a-study-of-normal-operational-landing-performance/full_text.pdf page 21).

An alternative is to consider the runway distance from the threshold to a common reference point - PAPI location for other runways; this relates the different PAPI TCH altitudes (rwy 22 62ft, rwy 27 70ft).

Approximate distance measurements (G Earth 'measure distance') shows that 22 PAPI is 0.17 'miles' from the threshold markings, grass end. Transposing this distance to rwy 24 PAPI provides an indication of how many touchdown wheel marks could have occurred 'in the grass' - rubber marks on 27 concrete, for the same distance as 22, (photo from G Earth).

The other difference is that the evidence from a low approach on 22 remains for everyone to see (wheel marks in the grass), but may not be so easily determined on 24 - 'that's not my wheel rubber'.
Thus for identical flights and conditions the outcome is determined by what remains to be seen; low TCH on 24 could be a 'non-event'; identical low TCH on 22, … Pprune posts.

Safety thoughts; operators of long body aircraft should choose runways with higher TCH, or use long-body PAPI - none at EDAM?
Also choose the longer runway to avoid risk of ducking under due to a perceived shorter runway.

Out Of Trim
16th Jan 2023, 19:39
They probably had green green indication on the papi. The lights shining through the grass! 🤔😬👀

sb_sfo
17th Jan 2023, 00:13
Nice shout out to Dave Gunson!

the_stranger
17th Jan 2023, 08:17
Safety thoughts; operators of long body aircraft should choose runways with higher TCH, or use long-body PAPI - none at EDAM?
Also choose the longer runway to avoid risk of ducking under due to a perceived shorter runway.Or just keep the basic flying skills up and land the damn plane where it is supposed to.

​​​​​​

procede
17th Jan 2023, 08:41
I think this is likely a case of pilots developing bad habits, such as intentionally landing below the glideslope and with three reds on the PAPI.

nicolai
17th Jan 2023, 10:06
lucky indeed ...

Given that the main landing gear on the 330 is tilted upwards when the aircraft is airborne, I would hazard a guess that what contacted the soft ground, just before the runway, were the wheels of the rear axle.
When the wheels of the rear axle came in contact with the runway edge, this would have caused a bit of a violent untilting of main gear, slamming the wheels of the front axle on the runway.

Potential to damage the gear interface unit, which can cause the system to think that the aircraft is in the air, making brakes and reverse unavailable, leaving the crew with nothing available to stop the aircraft after the touchdown. Iberia in Quito comes to mind.

The Iberia accident at Quito in 2007 (they've had more than one incident/accident there!) didn't cause them to lose all braking. They did lose autobrake (wiring to the wheel tachnomter broke) and they disconnected the anti-skid system. Not sure why they disconnected autoskid, but that combined with the surprise and inevitable delay in applying manual braking probably didn't help them to stop on the runway. https://avherald.com/h?article=422bdd7d has a summary of the accident report.

Losing all brakes might be possible... but is it that likely. Surely you can still apply manual brakes even if the air/ground logic says air?

safetypee
17th Jan 2023, 13:13
O O T, misplaced humour ?
the stranger,
procede,
You appear to have missed the point - that identically flown landings on 27 or 22, with similar 'normal' variation in flight path can have different outcomes because of the relative positions of PAPI and threshold. 22 being more critical in having the lesser distance - lower TCH, and, but not directly relevant, grass immediately before the threshold vice hard surfaces elsewhere.
There is no need to question skills or bad habit; instead question our conclusions for landings on either runway re a touchdown approx 0.17 miles before the PAPI location; why differentiate.

An occurrence such as this is 'statistically' likely from normal operations, the difference is in the outcome 'evidence', a bias in our safety perceptions.
Safety interventions should look at factors other than the human.

Is the lack of long-body PAPI the policy at EHAM, or country wide? If so then consider what advice should be published re rwy 22, which is also an operator choice.

procede, you may wish to calculate the height difference between 2R 2W and 3R 1W at the threshold.

zerograv
17th Jan 2023, 22:42
Not sure why they disconnected autoskid,

I think that disconnecting the antiskid is a action in the event of Loss of braking, which, I think, is a 'memory checklist' on the Airbus.

Have a look at the first pages of the following ...

https://www.pprune.org/accidents-close-calls/649493-korean-air-a330-off-runway-phillipines.html

xetroV
18th Jan 2023, 10:03
For reference, here's a video of a 747-400 landing on that runway:

https://www.youtube.com/watch?v=EdOUZavb6V4

Just posting that here because, judging from a couple of reactions on Twitter and elsewhere, some people seem to think that EHAM runway 22 is not really suitable for anything bigger than a 737. It certainly is.

(Mind you, the FAS of this 747 is comparable to that of a 737-900 under similar conditions, except the latter doesn't have the jumbo's braking performance.)

pattern_is_full
18th Jan 2023, 15:47
Of note in that 747 video, the approach is made at three white/one red on the PAPI, transitioning to four white just above 100 feet ("Minimums").

Am I right in assuming that is SOP for a heavy if there is not a dedicated tall-aircraft PAPI?

the_stranger
18th Jan 2023, 17:28
O O T, misplaced humour ?
the stranger,
procede,
You appear to have missed the point - that identically flown landings on 27 or 22, with similar 'normal' variation in flight path can have different outcomes because of the relative positions of PAPI and threshold. 22 being more critical in having the lesser distance - lower TCH, and, but not directly relevant, grass immediately before the threshold vice hard surfaces elsewhere.

I would be very careful using rubber marks on 24 and 27 for your "case", as it is not unheard for aircraft of a certain local carrier to land as short as possible on both ruwnnways as their parking spots are next to rwy24.

So those marks could be deliberate, skewing your calculations. Further more, if the runway, including papi is up to required specs, it still comes down to basic skills. Especially as this kind of incident has not happened before in the last 10 years on that runway.

MerrillParker
18th Jan 2023, 18:46
Also, in the landing video the aircraft appears to be flown on the ILS glide slope at least to the 100 foot call out. Then slightly above in the flare. In a 747, that should confirm the observation of a visual glide slope of 3 white, one red as the aircraft neared the runway.

safetypee
19th Jan 2023, 07:54
the_stranger, hhmmm

On one hand you caution the interpretation of pictures.

On the other, one operator is sufficiently skilled in deliberately 'landing short" (27), whereas another lacks skill (22) ~ 'blame and train'.

22 has a blacktop surface, no rubber marks for comparison; thus if not seen, short touchdowns are non events. Yet marks in the grass become significant events.
So a 'skilled' landing short on 27 is a non-event, but on 22 it is … …

Better to argue the merits or hazards of the physical situation, 27 vs 22, than to debate behaviour biased by knowledge of outcome - interpretation of pictures.

Uplinker
19th Jan 2023, 08:36
I think that disconnecting the antiskid is a action in the event of Loss of braking, which, I think, is a 'memory checklist' on the Airbus.

Yes, it is :ok:

Edit to add: that switch is how you select alternate brakes.

22/04
19th Jan 2023, 08:47
Can I ask a "threat and error management" question

Why are heavies being obliged to use runway 22 at an airport where five longer runways are available.

Seems to be an unnecessary threat.

Would EHAM have permitted rejection of that runway.as I heard a BA 787 do at New Orleans the other day ( reject 20 in favour of 11).

Uplinker
19th Jan 2023, 09:23
You can always request a longer runway if required 'due performance reasons'.

xetroV
19th Jan 2023, 09:31
Even if runway 22 was the only designated landing runway at that time, I think ATC would likely accommodate a request for a circling to runway 24 if needed. It is not uncommon to see freight carriers using the Sierra apron and KLM Cityhopper aircraft that park on apron Alpha use that procedure.

BraceBrace
19th Jan 2023, 11:25
Am I right in assuming that is SOP for a heavy if there is not a dedicated tall-aircraft PAPI?

I do would like to add these ideas are related to the 747 (maybe A380 as well?). Not "heavies". Even on a 777 you use same procedures as on a 737. There is no reason to go higher.

If memory serves me well I have landed a 777 on that runway (an empty one going into maintenance). It's not a big deal. However those same vague memories also makes me believe, that the general "feeling" the runway gave many pilots was not the most relaxed one because of the lack of shoulders and the "shorter than usual" distance.

the_stranger
19th Jan 2023, 11:48
Can I ask a "threat and error management" question

Why are heavies being obliged to use runway 22 at an airport where five longer runways are available.

Seems to be an unnecessary threat.

Would EHAM have permitted rejection of that runway.as I heard a BA 787 do at New Orleans the other day ( reject 20 in favour of 11).
In this case the wind was (south)westerly at a rather decent speed.
That gives you 4 runways to land on, 22/27/18C/18R (24,18L are not used for approach or landing, with 24 sometimes available for a break-off from 22/27)
The 18C was closed due maintenance and the 18R was deemed to have too much Xwind.
Leaves 22 and 27, but at the time, SPL had an outbound peak, requiring 2 runways to start from.

As 22 is a terrible runway to depart from (logistic wise for aircraft not based on the east part of SPL), 27 was used besides 24.
As 22 is a perfectly fine runway, there was and is no reason to change that. However, if a pilot requires the 27, ATC will give it but with (substantial) delay. The same, or worse for an eventual break-off to 24. Those "extra" landings would screw up the departure flow quite a lot.

Again, it might be the shortest runway and maybe short compared to what most are used to, but by no means a runway which is special, dangerous or to be avoided. Especially an a330, which stops on a dime, should have no more issues than a embraer or b737 (which probably requires more runway).

Doors to Automatic
20th Jan 2023, 09:48
If this short runway is regularly used for landing I am surprised it hasn't been extended to properly accommodate wide-bodies. There is room to extend past the 04 threshold to the RET taxiway that serves 18L. They wouldn't even have to move the 04 threshold, just keep it as displaced similar to the area ahead of 18L. This would give it a similar length to 22L at JFK which has been used for decades for wide-body arrivals without incident (SAS (1984) not withstanding!)

Sailvi767
20th Jan 2023, 13:56
Rumored the aircraft suffered a 20knot loss of Airspeed below 100 feet. Have to wait for the report to get the entire story.

swh
20th Jan 2023, 14:55
Rumored the aircraft suffered a 20knot loss of Airspeed below 100 feet. Have to wait for the report to get the entire story.

That is why aircraft cross the threshold at 50', does not explain how they lost 100' in less than 400' (horizontally, ie 14 degree path) if they were on slope.

BraceBrace
20th Jan 2023, 16:15
Rumored the aircraft suffered a 20knot loss of Airspeed below 100 feet. Have to wait for the report to get the entire story.

What are the rules for the commanded speed with or without autothrust on an Airbus?

On a Boeing, if the autothrottle is used for the landing, there is no need for wind corrections on top of the standard 5kts (all Boeings afaik). If autothrottle is/will be switched off for landing, command speed is increased with a wind correction up to 20kts. Main wind corrections can be bled off over the runway, but gust corrections on the speed should be kept in. With the autothrottle, the system reacts aggressive if speed would drop below command speed, especially if you hit Vref, which is why the standard 5kts is sufficient.

EI_DVM
20th Jan 2023, 17:08
What are the rules for the commanded speed with or without autothrust on an Airbus?

On a Boeing, if the autothrottle is used for the landing, there is no need for wind corrections on top of the standard 5kts (all Boeings afaik). If autothrottle is/will be switched off for landing, command speed is increased with a wind correction up to 20kts. Main wind corrections can be bled off over the runway, but gust corrections on the speed should be kept in. With the autothrottle, the system reacts aggressive if speed would drop below command speed, especially if you hit Vref, which is why the standard 5kts is sufficient.

On airbus using Auto-Thrust the approach speed is the minimum of VREF+5 or VREF+1/3 the headwind component, whichever is greater, (manual thrust is VREF or VREF+1/3 HW, this being the reason landing performance calculations are better with manual thrust in many cases). In this case, given the 28 knot headwind, the default approach speed would be VREF+9 knots. This again can be modified by the pilots up to a maximum of VREF+15, with this additional increments recommended in strong crosswinds by Airbus.

On top of this, G/S Minimum functions as well to provide a buffer if the instantaneous headwind component as computed by the IRs is greater than the tower provided headwind component, the aircraft will add the difference in headwind to the approach speed in anticipation of it being lost as the aircraft descends to the runway. In this case for example, with a VAPP of VREF+9, with the headwind on the ground of 28kts being expected, if at 500' there was a 38knot headwind, the "Fly Speed" would be VREF+19, if at 300' the headwind component was 33kts, the fly speed would be VREF+14, and so on, with the objective being that as the wind dies off the closer you get to the runway, the negative sheer is counteracted by the target speed reducing, and that you cross the threshold at the calculated VAPP (VREF+9 in this case).

All that being said, despite the extra margin being provided by G/S Mini and the VAPP increments in strong headwinds, I've always found the Auto-Thrust, both on the A330 and A320 to let you down at the worst possible moment. It can often get itself out of sync with the gusts and the speed and result in an automatic form of PIO with the thrust and can be quite slow to add power, even when the speed sinks down to or below the VLS (VREF).

Personally I've always found manual thrust the best option on both the A320 and A330 when landing in blustery conditions, it allows better anticipation of speed drops and gains and allows for local knowledge to be applied, as is often known about different airports, such as passing a certain hangar or passing by a certain ditch. It allows you to maintain current thrust setting in areas where you know a gain in airspeed will be short lived and soon die off (whereas auto-thrust would reduce thrust and then whack it back on as the speed bleeds off) or in the case of Schipol, manual thrust allows you to anticipate the drop in speed you often get in the last 40-80 feet as you descend below the tree line.

A320 Auto-Thrust performs reasonably well 90% of the time but the A330 auto-thrust in particular can often let you down for whatever reason, be it the larger engines and inertia, or just a different software gain/adjustment for counteracting airspeed reductions.

While I don't care to overly speculate, it would not surprise me that in the final report the use of auto-thrust, perhaps no additional increment being put on the VAPP due to concerns of the shorter than normal runway, the possibility of the aircraft getting low to begin with due to pilot mental perception of the shorter runway or just blustery conditions may all be listed as contributary factors.

Just worth mentioning as well though from experience, despite the shorter runway length, given the prevailing conditions, headwind, sea level, cool temperatures etc, even with a wet runway, Airbus landing performance figures from rough experience would still give you about a ~60% margin and I'd expect the calculated LDR would be in the region of 1,200m-1,300m, leaving a 700m-800m margin on the landing rollout on the 2,000m runway, and that being with proper technique (crossing threshold at 50', touchdown within touchdown zone etc.), so by no means overly limiting or close to the minimum 15% margin required.

the_stranger
20th Jan 2023, 18:28
If this short runway is regularly used for landing I am surprised it hasn't been extended to properly accommodate wide-bodies. There is room to extend past the 04 threshold to the RET taxiway that serves 18L. They wouldn't even have to move the 04 threshold, just keep it as displaced similar to the area ahead of 18L. This would give it a similar length to 22L at JFK which has been used for decades for wide-body arrivals without incident (SAS (1984) not withstanding!)
Because that would throw money at a problem which isn't a problem.
Most widebodies can land with 50% of the runway remaining and for those who can't or won't, 27 is available (with delay).

22 has been used of a decade at least without any incident except this one. I know a little (with emphasis on little) challenge is scary nowadays, but every pilot worth his salary should be able to land on a runway his performance figures indicate to be adequate.

Capn Bloggs
21st Jan 2023, 02:50
Add 1m to that eyepoint-tread height to account for strut extension and eagleclaw-tilted bogie position, to estimate the pilot's eyes will be 8m (26.4 ft) above the trailing main gear wheel tread.

​​​​​​​I do would like to add these ideas are related to the 747 (maybe A380 as well?). Not "heavies". Even on a 777 you use same procedures as on a 737. There is no reason to go higher.

This document:
https://www.airbus.com/sites/g/files/jlcbta136/files/2021-11/Airbus-Aircraft-Data-for-Visual-Aids-Calibration-v5.0.pdf
Gives a A330-300 eye to wheel height of over 10m.

By my calculations, on that runway, with an assumed PAPI angle of 3° located at just under 287m (giving a MEHt of 49ft), the wheels will cross the fence at 12ft and touchdown, with no flare, at less than 50m in. And that is on 2W2R. That is why they install "big jet" PAPIs with a MEHt of around 65ft.

Anybody thinking a 777 is the same as a 737 is this scenario is, IMO, mistaken.

​​​​​​​with proper technique (crossing threshold at 50', touchdown within touchdown zone etc.
Made more difficult when the PAPI "encourages" your mains to be only 12ft over the piano keys.

The 330 may have great brakes; it's where the PAPI leaves you which is the problem.

pattern_is_full
21st Jan 2023, 04:57
By my calculations, on that runway, with an assumed PAPI angle of 3° located at just under 287m (giving a MEHt of 49ft), the wheels will cross the fence at 12ft and touchdown, with no flare, at less than 50m in. And that is on 2W2R. That is why they install "big jet" PAPIs with a MEHt of around 65ft.

Anybody thinking a 777 is the same as a 737 is this scenario is, IMO, mistaken.

Made more difficult when the PAPI "encourages" your mains to be only 12ft over the piano keys.

The 330 may have great brakes; it's where the PAPI leaves you which is the problem.

Good stuff! Guess my assumptions were too conservative (or not conservative enough).

swh
21st Jan 2023, 05:55
By my calculations, on that runway, with an assumed PAPI angle of 3° located at just under 287m (giving a MEHt of 49ft), the wheels will cross the fence at 12ft and touchdown, with no flare, at less than 50m in. And that is on 2W2R. That is why they install "big jet" PAPIs with a MEHt of around 65ft.


Per the AIP (EHAM AD 2.14 APPROACH AND RUNWAY LIGHTING), the MEHT for RW22 PAPI is 62ft.

https://eaip.lvnl.nl/2023-01-12-AIRAC/html/index-en-GB.html

Capn Bloggs
21st Jan 2023, 06:46
Per the AIP (EHAM AD 2.14 APPROACH AND RUNWAY LIGHTING), the MEHT for RW22 PAPI is 62ft.
True. 62ft equates to a ground point of intercept of 360m; however, the distance between the PAPI and the threshold is 278m, according to Google Earth. Even given the errors in a non-vertical satellite picture, I can't see how 62ft could be close to being correct.

It also looks like the GS ground antenna is around 255m. From those Airbus figures, that would put the GS antenna over the threshold at 44ft, wheel height around 17ft.

Looking at the 747 video, they were 4 whites at the minimum and pretty much on GS. Given the glideslope would well below the PAPI if it was at 62ft, you expect to see at least one red.

BraceBrace
21st Jan 2023, 08:28
Anybody thinking a 777 is the same as a 737 is this scenario is, IMO, mistaken.

First thing they teach you is to fly a normal papi (btw, 3000 777 hours, 200 landings and type rating from the Boeing guys and instructors who've flown for a company with +50 777s in the fleet as a reference). The flare height is almost the same as a 737 as well, as there is very little pitch change required or you will end up at the end of the runway. The 747 might want to fly 3 whites, it is not done and not advised on a 777. Which is also what is stated in the FCTM.

A 2-bar VASI is not to be used (because yes, with a 3° path the main gear will touchdown near the end of the runway), a 3-bar VASI gives 3.25° and this is the "long-body" configuration you keep confusing with the PAPI system. It's a VASI specification, not a PAPI.

The PAPI's are perfectly safe to be used by any 777 "2 reds, 2 whites", whatever the body length, because the touchdown point is not the same as for a glideslope. They will guide an aircraft further along down the runway. This basic parameter is the one where you guys are doing pretty hard "guesswork" it seems (correct me if I'm wrong though). That doesn't mean there are variations to account for, but that is indeed, the pilot's job. The first "P" in PAPI is precision, and it will tell you when you are too low.

safetypee
21st Jan 2023, 09:31
Several inter-related aspects in recent posts.
Runway length per se relates to the landing distance required, which is referenced to "15m (50 ft) above the landing surface", taken to be the nominal 'threshold'. AMC CS 25-1592
Certification accepts (appears to assume) that there will be some variation in TCH re 'use of procedures' and the nebulous 'pilots of average skill'; as indicated by the distribution of wheel touchdown positions.

PAPI is an angular guidance system indicating the required flight path angle.
An approach flown with constant 3Red or 3White will have an angular offset, but still guiding to the same position on the runaway. The TCH for approaches with one light deviation will have different TCH, but not significant wrt to the range of crossing altitudes in normal operations i.e. normal flight path variations (beware combinations of extremes - lower designed TCH, low approach + wind-shear).

An non-standard approach 3W becoming 4W, flown parallel to the approach path (pseudo long body) requires judgement of the longitudinal distance relative to PAPI as to where the change in indications occur (2W - 3W - 4W).
Without real-time computation the height displacement from the PAPI flight path is unknown - a judgement, which if excessive can result in a touchdown further down the runway, invalidating landing distance calculations (risk of overrun). A mis judged deliberate offset vs normal variability.

Increasing the distance between PAPI and the threshold (higher TCH):
If the PAPI was repositioned SW on the existing runway 22 then the TCH would be higher. This would not reduce landing distance available (same threshold); but the hazard of the lip between runway / grass still remains - possibility of wheel brake damage as discussed.

Retaining the current PAPI location but extending the runway by moving the threshold of 22 NE would again increase TCH, and ironically increase the LDA - new threshold, but again the hazard remains.
However, retaining the existing threshold and adding hardstanding to the NE should reduce the hazard - less likely to occur, compare with 27 wheel marks, but this may not reduce the perception of 'safety risk' from an early touchdown 27 vs 22 - already discussed.

Thus if 'it' can happen, 'it' will happen, and if 'it' happens on 22, then the higher safety risk from the hazard.
This is an issue for operational awareness, but not within pilots capability to reduce the risk; that's the 'hard' task for the airport - remove / reduce the hazard.

Bloggs #50 :ok:

swh, thanks for the info; don't forget that with the correct TCH eye ht for the flt deck, the wheels are behind and lower, and lower still when they cross the threshold.

BraceBrace " The PAPI's are perfectly safe to be used by any 777 "2 reds, 2 whites", whatever the body length, " This is debatable; requires the equivalent data for 777 as in #50, and context re PAPI - threshold distance i.e. rwy 22 and grass.

BraceBrace
21st Jan 2023, 09:59
BraceBrace " The PAPI's are perfectly safe to be used by any 777 "2 reds, 2 whites", whatever the body length, " This is debatable; requires the equivalent data for 777 as in #50, and context re PAPI - threshold distance i.e. rwy 22 and grass.

Straight from Boeing (who also state the a 2 bar VASI is unsafe)
"When the airplane is on a normal 3° glide path, the pilot sees two white lights on the left and two red lights on the right. The PAPI may be safely used in relation to threshold crossing height, but may result in landing further down the runway. The PAPI is normally aligned to intersect the runway 1,000 to 1,500 feet beyond the threshold."

If you want more info, the numbers are in the FCTM.

The "blocks" are 1300-1500ft from the threshold and work perfectly as aiming points from a visual approach. If you want to feel safe, you might use the "farther" edge, but you risk long landings, especially on a glider like the 777. I very very vaguely recall an approach where the approach plates indicate wide bodies have to consider "3 whites" on the PAPI's. But I suspect this is because people have been calculating the numbers on the ground and have found variations to the standard setup requiring the note.

If you want to argue, I suggest you contact the Boeing engineers. But I'm pretty sure they've already done the calculations you have been doing here.

Capn Bloggs
21st Jan 2023, 11:08
From Boeing (https://www.boeing.com/assets/pdf/commercial/airports/faqs/icaoadmpart4.pdf):

737-800 Eye height above wheel path: 17.1ft, giving a wheels over the threshold on 22 of 31ft, touchdown at 177m (no flare).

777-300 Eye height above wheel path: 33.5ft, giving a wheels over the threshold on 22 of 15ft, touchdown at 81m in (no flare). On the ILS, you're a little better off, with a wheel crossing height of 27ft and a no-flare touchdown point of 155m.


The PAPI's are perfectly safe to be used by any 777 "2 reds, 2 whites", whatever the body length, because the touchdown point is not the same as for a glideslope. They will guide an aircraft further along down the runway.
Theoretically, but in this case, the PAPI and the ILS are within 20m of each other, so they don't. One would happily fly down the 2W2R PAPI thinking "she'll be right' when in fact, if you go just a tad low for whatever reason, you're going to be very very close to scraping the keys.

Capn Bloggs
21st Jan 2023, 11:15
The PAPI may be safely used in relation to threshold crossing height, but may result in landing further down the runway.
What does that mean?

​​​​​​​The PAPI is normally aligned to intersect the runway 1,000 to 1,500 feet beyond the threshold."
In this case, it's not. Nowhere near it.

BraceBrace
21st Jan 2023, 12:34
Quotes above come from Boeing. Pretty sure it's reliable.

Theoretically, but in this case, the PAPI and the ILS are within 20m of each other, so they don't. One would happily fly down the 2W2R PAPI thinking "she'll be right' when in fact, if you go just a tad low for whatever reason, you're going to be very very close to scraping the keys.

Why care how close they are to eachother? Calibration and vertical positioning change everything. You might as well put a PAPI on a building next to the threshold, it will work fine if calibrated correctly. The PAPI provides a 62ft clearance according to the AIP, +/- 30ft to the wheels according to your document, giving a good 30ft clearance. I don't see the issue with the PAPI.

If you're low, you're low. You're not supposed to think you're fine. If you are "on" the PAPI, you will be fine. From the minima I have never heard anybody say you should "ride" the PAPI's to the ground. It's a visual maneuver with aiming point in the touchdown zone.

Uplinker
21st Jan 2023, 12:48
All that being said, despite the extra margin being provided by G/S Mini and the VAPP increments in strong headwinds, I've always found the Auto-Thrust, both on the A330 and A320 to let you down at the worst possible moment. It can often get itself out of sync with the gusts and the speed and result in an automatic form of PIO......
Personally I've always found manual thrust the best option on both the A320 and A330 when landing in blustery conditions, it allows better anticipation of speed drops and gains and allows for local knowledge to be applied, as is often known about different airports, such as passing a certain hangar or passing by a certain ditch. It allows you to maintain current thrust setting in areas where you know a gain in airspeed will be short lived and soon die off (whereas auto-thrust would reduce thrust and then whack it back on as the speed bleeds off) or in the case of Schipol, manual thrust allows you to anticipate the drop in speed you often get in the last 40-80 feet as you descend below the tree line.


You sound like you know what you're doing and I certainly agree about local conditions caused by hangars etc, but you have misquoted the Airbus G/S mini as reducing thrust with a sudden headwind. I always get slightly nervous when people talk of overriding or preventing G/S mini, which of course will increase thrust if the headwind and therefore the IAS increases, to maintain the ground speed and the energy. i.e., It reacts the opposite way round to the Boeing system.


A320 Auto-Thrust performs reasonably well 90% of the time but the A330 auto-thrust in particular can often let you down for whatever reason, be it the larger engines and inertia, or just a different software gain/adjustment for counteracting airspeed reductions.
.

Yes, I believe the older A330s do not increase the auto-thrust gains below 3,000' ? Hence you need to be ready to intervene - on hot convective days for example. I have only ever had to do that twice though in years of flying it.

safetypee
22nd Jan 2023, 09:21
BraceBrace, I wonder if you are overlooking the aircraft 'flight deck -- main wheel' geometry at TCH. This is of increasing importance in long-body aircraft.

Using the diagram reference at #50:
https://www.airbus.com/sites/g/files/jlcbta136/files/2021-11/Airbus-Aircraft-Data-for-Visual-Aids-Calibration-v5.0.pdf

When the flight deck is over the threshold TCH on rwy 22, the wheels are still over the grass, at the calculated wheel over-surface height.
As the aircraft continues on a descending the flight path, the wheel over-surface height decreases proportionately, thus the wheel height over the threshold will be less than the value calculated when the flight-deck was over threshold; thus there is reduced margin for normal flight deviations.

Also (seeking expert A330 input on thus), if the A330 command to retard thrust depends on rad alt, then this too happens over the grass; thence the thrust could be at idle at the point where the wheels are over the threshold (rwy 22).

A late change in the flight path increasing the descent rate, would reduce the already small wheel-threshold clearance with risk of wheels touching the grass and runway lip contact.
This scenario is specific to rwy 22 due to the reduced PAPI to threshold distance - lower TCH, and no hard standing before the threshold; this combination does not apply to all of the other runways.
In addition, and accepting the visual nature of the operation at this point, the action of flaring the aircraft could further reduced height margin.

Specific expert A330 input required:-
During the flare the aircraft pitches nose up, rotating about the cg. The main wheels behind the cg will be displaced downward relative to the cg (the flight path). In long-body aircraft this contributes further height reduction, which in combination with issues above, could be critical.
The effect could also depend on aircraft type and control system mechanisation. Is the A330 flare control law biased towards pitch rate opposed flight path control ?

See marked up diagram, grass is 'green'.




https://cimg5.ibsrv.net/gimg/pprune.org-vbulletin/1702x617/c05be00f_d088_4d5c_98ca_f5f136200c37_69cfb237c03b40aae4e646d b24710469632c30a8.jpeg

FlightDetent
22nd Jan 2023, 12:14
I have argued long and wide repeatedly that PAPIs are not to be used for picking the aiming point at landing. Nor the so called aiming point markers.

While both feel like a one-size fit-all solution and judging by the typical result work fairly well, there are ample configurations where following that guidance will just take the plane into a wrong place.

​​​​​​What happened here was a geometrical necessity, no further research beyond safetypees post needed.

The question still stands the same, are pilots actually trained where to land, have you been? Last time around the agreed answer was 'there's no reason why we should care, know the best already'.

Amsterdam, Burbank...

If somebody links the LAX THR graphic posted on the Skiathos thread, all is revealed. That is built pilot proof for T7/330. Other places might not be.

Chiefttp
22nd Jan 2023, 12:30
As far as Boeing procedures, on the 767 My companies procedure regarding high wind conditions with the autothrottles On or Off is to add half the steady state headwind component, and ALL of the gust (above the steady state HW component)to VRef. So if the winds are 120 at 14K gusting to 20k and we are landing on Rwy12. we add 7 knots (1/2 the steady wind of 14k) to VRef, and then all of the gust, which in this example is 6 knots above the 14k steady wind. So Our final additive is VRef plus 7 Knots plus 6knots (gusts) for a total of VRef +13k. As we fly the approach we are supposed to reduce our speed so as to land without the gust increment, but maintain the steady state additive until touchdown. So in my example we should land at VRef +7k. Boeing used to say that this additive was unnecessary if the autothrottles were active, and the additives were only required if the autothrottles were inop, but about 5 years ago (at my airline) we changed the procedure to always add the steady state headwind and gust additive.

Capn Bloggs
22nd Jan 2023, 13:43
there are ample configurations where following that guidance will just take the plane into a wrong place.
If they are knowingly misleading, then they shouldn't be used. Why do you think that at the vast majority of long-body runways, the PAPIs are around the 62ft/360m in? Because flying a stabilised approach ie not doing a wiffodil under 200ft to extend the approach into the TDZ with the PAPI at the traditional "short-jet" 50ft/290m is too hairy, with not enough room for error (as, I predict, occurred with this A330). The whole point of a stabilised approach is you fly consistently to the flare point, not go off the PAPI above 200ft to stretch the flight to the TDZ. If you have to do that, such as in this case, then if we're going to apply today's cotton-wool level of expected safety, the aircraft shouldn't be using that runway.

At the very least, there should be written, in the company route info manual, a note of the danger of using 22 with it's low MEHt (not that anybody "knows" that yet because the published MEHt is wrong (IMO)). Keep an eye on the AIP for EHAM 22; I reckon there's a good chance it will be quietly republished down to 48ft as a result of this incident.

At KLAX, all the PAPIs are between 390m/67ft and 450m/77ft. EHAM 22, by my calculation, is 287m/49ft. Additionally, all the LAX runways have significant concrete under-runs.

RatherBeFlying
22nd Jan 2023, 17:35
Location of the radalt antenna will also have a role in determining wheel height over the ground. Location by the main gear will give a result that is less dependent on body angle than location nearer the cockpit.

I do not know whether or how A has programmed radalt antenna location and body angle into their various laws.

BraceBrace
22nd Jan 2023, 21:18
then they shouldn't be used.

EXACTLY. But they are not misleading, you are expecting too much precision from the system.

The PAPI is a (1) VISUAL and (2) APPROACH aid. By the time you are over the fence, you should already be looking elsewhere and get a complete picture of the runway. You should be looking at touchdown zone markings and lights to get a complete picture. If you are on the PAPI you will be fine, the system will not make you crash. It's not a precision approach category, neither a visual flare aid, we’re not doing carrier landings here, it’s not that type of ”precision”. A PAPI is designed to put the aircaft on ie 3° glide 5NM out VISUALLY (using no GS or path indications on your instruments) and bring you close to the fence safely. Nothing more.

Capn Bloggs, everybody understands your ideas about main gear position on long body aircraft. But you make it sound like this simple light system is an automated landing system giving you sufficient protection all the way down to the runway. It’s not.

PS; it's a visual approach tool so legally, if you are cleared for an ILS... you're not even supposed to look at the PAPI, you visually transition from the glideslope indications to the set of runway lights (nothing stops you from using the PAPI, but it's your responsability).