737 runway overruns
Join Date: Apr 2010
Location: IRS NAV ONLY
Posts: 1,230
Likes: 0
Received 0 Likes
on
0 Posts
#22, interesting.
Re “… would have been able to stop the airplane on the runway if …”
There are many accident reports echoing this point; ‘the aircraft operated as designed’, … ‘the crew did not use it correctly’; blame, train. There is little consideration that even a well trained crew might not be able think of adjusting a system operation in critical situations.
After an event the investigator can calculate distances in slow time; during the event the crew would have to fully understand the capability of the aircraft in the actual conditions, not necessarily as reported, and compare achievable landing performance with runway remaining, a change of action if required.
We forget that humans are more often a limiting factor in an man-machine system, and that on occasion a good machine is insufficient to protect the human, be that operator, runway assessor, regulator, or designer.
There are many situations where the industry should not use the human as the safety back stop;- just because pilots are the last link in the chain does not mean that it is the best defence.
Generally additional safety margins are applied in these situations; landing distance margin.
Historically, the US system, manufacturer, regulator, operator, applied landing distance margins based on ‘actual’ data, where only a minimum addition was required. Other countries had a range of methods, some followed the US example, others - Europe, biased towards the AFM and larger factors.
Nowadays the industry is moving towards commonality with OLD/FOLD (although minimum factors still reign). There may be a pattern involving the 737 in this, but it could be difficult to identify amongst the many variables. A starting point could be what is taught by the manufacturer;- dispatchers use a performance manual (AFM data), flight crews the FCOM (actual data).
Re “… would have been able to stop the airplane on the runway if …”
There are many accident reports echoing this point; ‘the aircraft operated as designed’, … ‘the crew did not use it correctly’; blame, train. There is little consideration that even a well trained crew might not be able think of adjusting a system operation in critical situations.
After an event the investigator can calculate distances in slow time; during the event the crew would have to fully understand the capability of the aircraft in the actual conditions, not necessarily as reported, and compare achievable landing performance with runway remaining, a change of action if required.
We forget that humans are more often a limiting factor in an man-machine system, and that on occasion a good machine is insufficient to protect the human, be that operator, runway assessor, regulator, or designer.
There are many situations where the industry should not use the human as the safety back stop;- just because pilots are the last link in the chain does not mean that it is the best defence.
Generally additional safety margins are applied in these situations; landing distance margin.
Historically, the US system, manufacturer, regulator, operator, applied landing distance margins based on ‘actual’ data, where only a minimum addition was required. Other countries had a range of methods, some followed the US example, others - Europe, biased towards the AFM and larger factors.
Nowadays the industry is moving towards commonality with OLD/FOLD (although minimum factors still reign). There may be a pattern involving the 737 in this, but it could be difficult to identify amongst the many variables. A starting point could be what is taught by the manufacturer;- dispatchers use a performance manual (AFM data), flight crews the FCOM (actual data).
I recall a formal study comparing the safety level of aircraft landing performance between those aircraft with thrust reverse and those without.
Much as expected those aircraft without reverse had less overall safety margin - not sure how that was defined.
I also recall a follow-on study considering the failure of reversers to deploy. Here the aircraft with reverse (failed) had the lesser safety margin. And with fast fading memory, the 737 did not fair particularly well (anyone have a reference for this report).
A conclusion was that the 737 thrust reversers contribute more to the landing performance than other aircraft, the 737 being less able to stop than other aircraft without reversers.
Ratio of effectiveness - reverse vs brakes anyone?
Size / capacity of brakes, brake materials, type of anti skid.
From an HF aspect perhaps those crews operating aircraft without reverse are more risk aware than those who can select reverse - but not necessarily safer. But if reverse fails …
Much as expected those aircraft without reverse had less overall safety margin - not sure how that was defined.
I also recall a follow-on study considering the failure of reversers to deploy. Here the aircraft with reverse (failed) had the lesser safety margin. And with fast fading memory, the 737 did not fair particularly well (anyone have a reference for this report).
A conclusion was that the 737 thrust reversers contribute more to the landing performance than other aircraft, the 737 being less able to stop than other aircraft without reversers.
Ratio of effectiveness - reverse vs brakes anyone?
Size / capacity of brakes, brake materials, type of anti skid.
From an HF aspect perhaps those crews operating aircraft without reverse are more risk aware than those who can select reverse - but not necessarily safer. But if reverse fails …
It might be worth remembering that, at least in the FAA world, the landing performance distances give no credit for thrust reversers (EASA is a little different but I don't know details).
FAA landing distances are established by landing on a dry runway with max braking and no T/R. This is done multiple times at different landing weights to establish a weight/distance baseline. Then a factor is applied - a rather large one (2.4 comes to mind but don't hold me to that) - to account for things like low friction runways and less than perfect technique.
Any improvement the T/Rs provide is considered a bonus.
FAA landing distances are established by landing on a dry runway with max braking and no T/R. This is done multiple times at different landing weights to establish a weight/distance baseline. Then a factor is applied - a rather large one (2.4 comes to mind but don't hold me to that) - to account for things like low friction runways and less than perfect technique.
Any improvement the T/Rs provide is considered a bonus.
td, No credit for T/R
That's correct for certificated performance as in the AFM, which already includes a ‘dry’ factor. The AFM data and further factoring (wet) is predominantly used for dispatch; whereas ‘actual’ - unfactored performance is in the FCOM for crew use, and more often includes credit for T/R.
The impression I have is that operators who use ‘actual data’ tend to include T/R as standard and then consider the minimum FAA factor of +15% on top. Like most landing data this is satisfactory if the runway condition is as reported, but if the braking action is less than planned, then (factored) actual data may run out of runway earlier than other factored data. Do 737 operations predominantly use ‘actual’ data?
A 2.4 factor was published in a Canadian report re landing on snow and ice, grooved / ungrooved runways, assessing equivalent safety with dry operations (IIRC they used a 737), i.e. using anything less than 2.4 on ice and snow has a higher risk than normal.
FAA and EASA are harmonised for AFM data, except EASA has additional requirements for contaminated operations. For operating data, EASA has moved rapidly towards Factored Operational Landing Distance lead by manufacturers (is 737 data available yet - all versions ?). FOLD is much more realistic than ‘actual’, but may be less than AFM factored data, although some views argue that there is a good correlation between FOLD and AFM for a wide range of operating conditions - is more tolerant of miss reporting braking action.
P.S. Aviation Investigation Report A10A0032 - Transportation Safety Board of Canada
(727), para 1.9 discusses additional factors (up to 2.4) in very wet conditions, with/without T/R.
TC Reference. History of the Joint Winter Runway Friction Measurement Program. TP 13579
From Joint Winter Runway Friction Measurement Program - Transport Canada
FAA position re improved data https://www.aci-na.org/sites/default...fng_8-4-16.pdf still only advisory, whereas EASA mandates it.
That's correct for certificated performance as in the AFM, which already includes a ‘dry’ factor. The AFM data and further factoring (wet) is predominantly used for dispatch; whereas ‘actual’ - unfactored performance is in the FCOM for crew use, and more often includes credit for T/R.
The impression I have is that operators who use ‘actual data’ tend to include T/R as standard and then consider the minimum FAA factor of +15% on top. Like most landing data this is satisfactory if the runway condition is as reported, but if the braking action is less than planned, then (factored) actual data may run out of runway earlier than other factored data. Do 737 operations predominantly use ‘actual’ data?
A 2.4 factor was published in a Canadian report re landing on snow and ice, grooved / ungrooved runways, assessing equivalent safety with dry operations (IIRC they used a 737), i.e. using anything less than 2.4 on ice and snow has a higher risk than normal.
FAA and EASA are harmonised for AFM data, except EASA has additional requirements for contaminated operations. For operating data, EASA has moved rapidly towards Factored Operational Landing Distance lead by manufacturers (is 737 data available yet - all versions ?). FOLD is much more realistic than ‘actual’, but may be less than AFM factored data, although some views argue that there is a good correlation between FOLD and AFM for a wide range of operating conditions - is more tolerant of miss reporting braking action.
P.S. Aviation Investigation Report A10A0032 - Transportation Safety Board of Canada
(727), para 1.9 discusses additional factors (up to 2.4) in very wet conditions, with/without T/R.
TC Reference. History of the Joint Winter Runway Friction Measurement Program. TP 13579
From Joint Winter Runway Friction Measurement Program - Transport Canada
FAA position re improved data https://www.aci-na.org/sites/default...fng_8-4-16.pdf still only advisory, whereas EASA mandates it.
A picture is worth a thousand words, so the saying goes. I believe that; which is why when telling type rating candidates that reverse is more effective at high speed I use the following demo in the simulator.
For example: 737 Classic. Take off on slippery runway surface ice patches (input into simulator). Flaps One max structural weight with all engines abort at 150 knots on 10,000 ft runway sea level nil wind.
First abort using speed brake and brakes only. No reverse. Result was over-run the end at 30 knots.
Second abort: Full reverse at 150 knots. No brakes and no speed brake. Over-run about 30 knots as well.
Third abort: 150 knots Maximum manual braking. Speed brake and max reverse. Aircraft stops on runway about 300 metres from the end.
Conclusion: It was the full reverse (plus brakes and speed brake) that made the difference and permitted the aircraft to stop with room to spare.
Admitted freely that this was an amateur and unmeasured attempt to prove reverse at high speed is more effective for stopping than at low speed. Candidates were convinced.
For example: 737 Classic. Take off on slippery runway surface ice patches (input into simulator). Flaps One max structural weight with all engines abort at 150 knots on 10,000 ft runway sea level nil wind.
First abort using speed brake and brakes only. No reverse. Result was over-run the end at 30 knots.
Second abort: Full reverse at 150 knots. No brakes and no speed brake. Over-run about 30 knots as well.
Third abort: 150 knots Maximum manual braking. Speed brake and max reverse. Aircraft stops on runway about 300 metres from the end.
Conclusion: It was the full reverse (plus brakes and speed brake) that made the difference and permitted the aircraft to stop with room to spare.
Admitted freely that this was an amateur and unmeasured attempt to prove reverse at high speed is more effective for stopping than at low speed. Candidates were convinced.
Join Date: Oct 2015
Location: 43N
Posts: 264
Likes: 0
Received 0 Likes
on
0 Posts
I'll add my .02¢.
No one behind the cockpit door knows the definition of a good landing, not one of them. To them a smooth landing is a good landing, we know better.
A smooth landing 3000' down the runway is, IMO, is a legal but bad landing. A smooth landing 3001 feet down the runway (or first 1/3 whichever less) according to our manuals constitutes an unstable approach and a go around must be performed.
A firm touchdown at 1000' may not get you any attaboys from anybody behind the door, but it will get a positive comment from me.
If we strive to touchdown at 1000' on each and every landing, when the time comes where we really, really need to we know we can because we do it every day. Do I try to touchdown at 1000' feet on every landing, yup, am I successful doing that on every landing nope, so I keep practicing.
No one behind the cockpit door knows the definition of a good landing, not one of them. To them a smooth landing is a good landing, we know better.
A smooth landing 3000' down the runway is, IMO, is a legal but bad landing. A smooth landing 3001 feet down the runway (or first 1/3 whichever less) according to our manuals constitutes an unstable approach and a go around must be performed.
A firm touchdown at 1000' may not get you any attaboys from anybody behind the door, but it will get a positive comment from me.
If we strive to touchdown at 1000' on each and every landing, when the time comes where we really, really need to we know we can because we do it every day. Do I try to touchdown at 1000' feet on every landing, yup, am I successful doing that on every landing nope, so I keep practicing.
A firm touchdown at 1000' may not get you any attaboys from anybody behind the door, but it will get a positive comment from me.
If we strive to touchdown at 1000' on each and every landing, when the time comes where we really, really need to we know we can because we do it every day. Do I try to touchdown at 1000' feet on every landing, yup, am I successful doing that on every landing nope, so I keep practicing.
If we strive to touchdown at 1000' on each and every landing, when the time comes where we really, really need to we know we can because we do it every day. Do I try to touchdown at 1000' feet on every landing, yup, am I successful doing that on every landing nope, so I keep practicing.
Passengers often complained about his landings, but he countered by saying that at Boeing, pilots from third world countries frequently underwent 737 type ratings; including on the real aircraft as well as simulators. Firm, little flare but firm consistent accurate touch downs on the 1000 ft markers was Boeing policy; especially as flying competency varied widely among foreign crews.
Moderator
.. and, if one doesn't land by shortly after the 1000ft markers, then one is eating into the 1.67 AFM pad ... eat too much and too leisurely and one might get egg on face syndrome if the runway is getting towards being landing distance critical.
As to using actual, rather than AFM, data on the line .. utter madness in this certification engineer's/pilot's view .. just asking for a screw up sooner or later. If it's good enough to use factored data for planning, then it's good enough to use that on the line other than in an emergency situation. Dinosaur ? .. you bet.
As to using actual, rather than AFM, data on the line .. utter madness in this certification engineer's/pilot's view .. just asking for a screw up sooner or later. If it's good enough to use factored data for planning, then it's good enough to use that on the line other than in an emergency situation. Dinosaur ? .. you bet.
Excellent point JT! Been on the same aircraft now for 17 years in the left seat. Now all of a sudden we need to calculate landing perf for every landing. And it is not as we are runway limited anywhere!
-another dinosaur
-another dinosaur
Join Date: Jul 2006
Location: Somewhere over the rainbow
Posts: 946
Likes: 0
Received 0 Likes
on
0 Posts
Pineteam - Your post has reminded me of the BA internal pilot safety report I once saw where the go-around of an A319 at LHR was discussed. They floated on landing and when it became evident that they would just miss the last TDZ marker, they went around. The discussion centred around how sound that decision was, and that this should always be the case if touchdown within the TDZ wasn't certain.
The distance to go on the runway in question (27R) at the point the g/a was initiated was circa 9,800ft, which to put into context is twice the length of runway after the correct touchdown point at Jersey where the A319 regularly operates and around 3x what is required to stop.
I did question the wisdom of such a strict SOP but, as my friend (a 747 skipper) said, safety is paramount with no exceptions.
The distance to go on the runway in question (27R) at the point the g/a was initiated was circa 9,800ft, which to put into context is twice the length of runway after the correct touchdown point at Jersey where the A319 regularly operates and around 3x what is required to stop.
I did question the wisdom of such a strict SOP but, as my friend (a 747 skipper) said, safety is paramount with no exceptions.
Join Date: Jan 2003
Location: UK
Posts: 304
Likes: 0
Received 0 Likes
on
0 Posts
I've flown the 737 for a few years now. The one thing I have noticed is that pilots tend to avoid the advice given in the FCTM as to how to land the aircraft. Boeing is very detailed in its advice; I'll let you look it up and it does work well. No floating and no crashing down, works every time. Yet many pilots seem to like to yank back, chop the thrust, then yank back some more. Thump it on, float it or grease depending on the day. With a slippery or contaminated, shortish runway the floater guarantees trouble.
There are lots of overrun accident reports you could look at. I'd suggest the Southwest Airlines Chicago Midway, Southwest Airlines Burbank and Eastern Airlines La Guardia reports may give food for thought.
There are lots of overrun accident reports you could look at. I'd suggest the Southwest Airlines Chicago Midway, Southwest Airlines Burbank and Eastern Airlines La Guardia reports may give food for thought.
Join Date: Jun 2007
Location: Wanderlust
Posts: 3,406
Likes: 0
Received 0 Likes
on
0 Posts
Pineteam
Landing distance is one issue which will be satisfied by long runway but if you float tail strike comes in picture. One needs to follow percentage technique. Because on one day A320 is fine but on another day same thing in A321 will cause tail strike. So the adage greasers are good for passengers but not pilots
Landing distance is one issue which will be satisfied by long runway but if you float tail strike comes in picture. One needs to follow percentage technique. Because on one day A320 is fine but on another day same thing in A321 will cause tail strike. So the adage greasers are good for passengers but not pilots
but if the guys floats a little bit, I'm never asking him to "put it down" like some captains Love to do
If for some reason let's say the first officer is the culprit of the long float, the captain, rather than risk attempting to salvage a bad situation by talking the PF into "putting it down", should promptly take control and go-around.
Telling the first officer to go-around instead of taking over control of the go-around himself, has the potential for a further critical delay. Especially if there is a language difference between the two crew members which has the potential to further exacerbate an already developing situation.
Bugger whose "leg" it is. There are times when a stuff-up has occurred and that is where the captain needs to recognise his own responsibility for the safe operation of the flight and take control without pondering the why's and wherefore's of whose "leg" it is.
Join Date: Jul 2006
Location: Somewhere over the rainbow
Posts: 946
Likes: 0
Received 0 Likes
on
0 Posts
Pineteam
Landing distance is one issue which will be satisfied by long runway but if you float tail strike comes in picture. One needs to follow percentage technique. Because on one day A320 is fine but on another day same thing in A321 will cause tail strike. So the adage greasers are good for passengers but not pilots
Landing distance is one issue which will be satisfied by long runway but if you float tail strike comes in picture. One needs to follow percentage technique. Because on one day A320 is fine but on another day same thing in A321 will cause tail strike. So the adage greasers are good for passengers but not pilots
I totally agree with you. I'm only flying on the left less than 4 months since training is completed and so far, I never had that situation yet nor a situation that I felt like I had to take control. But I definitely have what you say in mind. I can only fly with senior Fos for the first year. Make life easier for sure. Not meaning bad landing can't happen of course, it does happen to all of us. We fly A319, A320 and mostly A321 and from my little experience so far I only had one "hard landing" of 1.78g on an old A 321. The guy did not flare enough. Still a safe landing tho.
Join Date: Oct 2015
Location: 43N
Posts: 264
Likes: 0
Received 0 Likes
on
0 Posts
I don't know who writes such manual but I doubt they are pilots..I agree if you are concerned about landing distance available. But a light A320 with 3500M+ runway on a perfect day.. Who cares if you land slightly deep when you know you only need half the runway to stop the aircraft safely... not saying you should disregard the TDZ, but if the guys floats a little bit, I'm never asking him to "put it down" like some captains Love to do.. Give the guy a break. It's all about common sense. Especially now with the ipad computation, we know exactly what's our margin.. A go around is not always the safest manoeuver. Emirates knows best with their 777 accident... and I know very well about landing on very short strips. I did bush flying before.
Our stabilized approach criteria as is our abort criteria are independent of runway length. When you brief your companies abort criteria, do you make caveats for runway length? Something like, "Hey I know we aren't suppose to abort above V1 but we are taking off on a long dry runway so..."
Accident reports are littered with the bodies of pilots (and passengers who wished to remain uninvited to their own funerals) who thought SOP doesn't apply to them, or SOP was for less accomplished pilots, or SOP was situation dependent.
Intentional non compliance is a cancer our industry can not afford.
Join Date: Jul 2006
Location: Somewhere over the rainbow
Posts: 946
Likes: 0
Received 0 Likes
on
0 Posts
V1 is not only a limitation by runway lengh... anyway I won't argue. We don't have such restriction in our Sop. So I don't break any rule... once again it's common sense. If you wanna do a go around at 30 feet on a 4000 meters runway just because you pass the 900 meters TDZ, be my guest... LOL I'm gonna land cause I still use my brain and it's telling me it's safe!😜
With all due respect... Better to reinforce the good habit of landing where you are supposed to land, than risking creating a possible bad habit of landing beyond 900m. Just using my brain in a different way than you do.
I don't really see what's to gain in landing long intentionally and I see all kinds of potential risk in it. One day on the last sector of a max FDP day, you might forget that this airport is actually not a 4000m runway but 3000m or even less. And obviously that will also be the day that you are at max landing weight and upon touchdown reversers fail to deploy, spoilers have a day off and/or your brakes are having a ****ty day. Because Murphy.
You have XXX number of people in the back that are your responsibility, why degrade safety margins? Yes, 9 times out of 10 it will probably be fine, I'll just be landing with that tenth one in mind. Just two cents.
I don't really see what's to gain in landing long intentionally and I see all kinds of potential risk in it. One day on the last sector of a max FDP day, you might forget that this airport is actually not a 4000m runway but 3000m or even less. And obviously that will also be the day that you are at max landing weight and upon touchdown reversers fail to deploy, spoilers have a day off and/or your brakes are having a ****ty day. Because Murphy.
You have XXX number of people in the back that are your responsibility, why degrade safety margins? Yes, 9 times out of 10 it will probably be fine, I'll just be landing with that tenth one in mind. Just two cents.
Join Date: Jul 2006
Location: Somewhere over the rainbow
Posts: 946
Likes: 0
Received 0 Likes
on
0 Posts
We always do ipad computation for landing performance so our situation awareness for the landing distance is there... Maybe I was not clear enough: I don't land deep on purpose. It's very rare that I have landed after the TDZ. Of course not, we are professional pilots, or trying to be. I'm just saying that if it happens, most of the time it's safer to continue and land than doing a go around. Most of our incidents in our company happened during a go around. I used to fly in Zambia with extremely short runways, normal runways and long runways... you just don't forget when runway is short. You think about it all the time haha. It's like when you drive at night, you don't forget to turn on the lights don't you? Anyway that's just my point of view. It just amazed me how some pilots can be so conservative and afraid of things going ugly and always overthink like: "If that happens" syndrome. Just fly the plane, relax but be sharp and enjoy. Easy life.
Last edited by pineteam; 26th Sep 2017 at 18:56.