Configuration Warning near V1. Continue or stop?

[Centaurus]

In most take off safety briefs the mention is made of the various events that warrant a rejected take off. A Take off configuration warning is one of those events.

Saw it happen in a simulator where the instructor triggered a config warning at 15 knots below V1 and accelerating fast. After a brief hesitation the pilot elected to continue but remained on the runway past V2 rather in shock than a deliberate ploy. A normal rotation was made and although the leading edge devices had retracted (caused by the inconsiderate instructor pressing a special button) the 737 climbed away without any problem around V2 + 25.

This got me thinking. After all it was a successful outcome and it would have been a trifle boorish to criticise a command decision that went the right way. Close to V1 there is little time to ponder exactly what system failure caused the configuration warning to go off. The alternative is an inevitable heavy weight high speed abort with its attendant well documented risks.

Weighing it up, perhaps the greater danger is the high speed abort as against leaving the aircraft on the runway and accelerating to a higher VR before rotation with either an asymmetric or no leading edge devices situation. Of course it could also be the once in a million false warning like some false fire warnings. It made for an interesting post flight de-briefing.

[411A]

I would be inclined to agree.
Generally speaking, the config warning at or near V1 is 'usually' the speed brake lever having just come up a bit on the type that I fly, however at or near V1, the takeoff should be continued, with possibly an overspeed V2(improved climb) used, if at all possible, runway permitting, of course.

[BEagle]

In the type I flew, the configuration warning was theoretically inhibited once the throttle levers were beyond a certain position...

However, on day we received a configuration warning at around 90 KIAS (well below V1), so rejected the take-off. The fault was later proved to be a faulty 'interruptor unit' - the thing which caused the beeping. This was shared with the cabin altitude warning system and it was incorrectly sensing an excessive cabin altitude.

The rejected take-off was very straightforward, but we had the luxury of about 12 simulator sessions per annum and probably 20 or so high speed abort practices in the box. So it was, to some extent, second nature.

What really annoyed me though was to learn that several of my colleagues had also had the horn go off for the same reaon in this jet - but hadn't bothered to snag it...:mad:

I never accepted the 'just ignore it' advocates who said they'd continue if the configuration warning sounded during the take-off roll; years earlier an airline operating the same type had a nuisance configuration warning when it was tested prior to flight. They decided not to go - which was just as well. The cause of the warning was an unserviceable tailplane incidence sensor; however, the cause of the unserviceability was significant damage to the tailplane and fin caused by a previous encounter with severe turbulence......

[dolly737]

...the 737 climbed away without any problem around V2 + 25...
Hmmm...
sure, but I'm doubting the same outcome with an engine failure ...

[I-2021]

This got me thinking. After all it was a successful outcome and it would have been a trifle boorish to criticise a command decision that went the right way. Close to V1 there is little time to ponder exactly what system failure caused the configuration warning to go off. The alternative is an inevitable heavy weight high speed abort with its attendant well documented risks.
Weighing it up, perhaps the greater danger is the high speed abort as against leaving the aircraft on the runway and accelerating to a higher VR before rotation with either an asymmetric or no leading edge devices situation. Of course it could also be the once in a million false warning like some false fire warnings. It made for an interesting post flight de-briefing.

Hello,

I agree with you, if you consider what Boeing states for aborted takeoff procedures between 80 kias and V1 (engine failure, fire, aircraft unsafe to fly, predictive windshear). But... could the takeoff config warning be a warning of an aircraft that could become unsafe to fly for any reason ?

[Capt Claret]

Would a modernish jet be controllable with the slats on one wing extended, and those on the other retracted?

[I-2021]

I suppose it would:hmm:

[GlueBall]

In the simulator you should abort; in the real aeroplane you should not abort. :eek:

[I-2021]

You mean in general terms or for a specific aircraft ?

[Empty Cruise]

GlueBall,

I take it your remark was rather tounge-in-cheek. If not, please expand on that philosophy - I'm eager to learn. I was just brought up with this wimpy "Train the way you fight - fight the way you train"-attitude that is obviously past its sell-by-date :D

Empty

[NigelOnDraft]

In my company, a Config Warning is a compulsory "Stop" call from either Pilot up to V1. So in the Sim, in a politically correct environment, "Stop" it would (have to) be.

In reality, and IMHO, Config Warnings are there to stop the crew making "plonkers" of themselves i.e. trying to takeoff with some major config error, and as a consequence, once you have got takeoff power set, and no Config Warning, it's real purpose is over.

Therefore a Config Warning occurring near V1 is indicating either some major "change" in configuration (and hardly likely to have been crew selected), or an indication fault feeding into the Config Warning system. It then comes down to the situation on the day, and whether it is a "Go" V1, or a "Stop" V1, and your judgement, about whether to "Stop" or not. I know which I'd tend towards ;)

[alf5071h]

Config is normally a red warning.
If the reason for the warning is identifiable (i.e. a config error) then the take off should be rejected.
If the reason is not identified then crew must not make any assumption as they cannot judge between an undetected fault and a nuisance warning. Thus, with a config warning less than V1, always stop.
Many aircraft have a pre takeoff ‘embarrassment’ config check button e.g. BAe146. This does not prevent an item in the config circuit moving out of alignment after that check or preclude any operating error.
N.B. BAe146 takeoff accident in China. A config warning was given during the pre-takeoff check. The crew failed to identify the fault (a rather obvious failure to select flap), the crew erroneously deduced a flap gauge fault and overrode the flap config system and commenced the take off. The flap config auto reset was re-enable after approx 20 sec, thus a config warning was given as the aircraft passed the end of the runway – it did not get airborne and there were many fatalities.
__________________
Unless specifically authorized everything else is forbidden.

[qwertyuiop]

Surely the answer is;

If you get a Config warning before V1 the aircraft CAN stop, but may not fly. Unless you brief otherwise you STOP.

[InSoMnIaC]

qwertyuiop. Thats the smartest thing I've heard in the whole thread.

[bafanguy]

Admittedly, I've only flown older model airplanes(B737/727, DC9/MD80, CV880) but the takeoff configuration warning was triggered by throttle position or perhaps an EPR setting ( in the neighborhood of 1.4) with the existence a configuration problem.

Either of these occurred rather early in the TO roll...mostly before 80 kts. Stopping the TO under this condition was rather a non-event.

Forgive me, but how does one get anywhere near V1 before a configuration warning is sounded and stopping would be a particularly serious problem.

What have I missed here ?

[Empty Cruise]

...unscheduled config-changes, inadvertent positioning of flap lever, uncommanded slat retraction, inadverdent speedbrake deployment, uncommanded stabiliser movement - to name a few :confused:

I do agree with what many people in this forum have said about near V1-cuts, but when it comes to trying to outsmart the safety systems in a fraction of seconds, I must admit my limit is reached. Yep, we can all be rocket scientists when sitting at home with a cuppa, but there is no time for focus groups or mental artistery at V1 -10. You brief it as you want it done - end of story.

Is it me or is there - in several threads relating to V1 - an attitude creaping in that a) there is room for meddling with V1 and b) serious failures are so rare that it's much better to assume spurious warning & continue, rather than stopping and potentially damaging the aircraft in a 40 kt. overrun?

Are there any examples of rejected takeoffs that have been carried out by the book, from at or below V1 that have resulted in personal injury? If so, when we weigh them against the outcomes of the cases where the take-off should have been rejected (but was continues) - which one would you rather be associated with?

Empty.

[bafanguy]

...unscheduled config-changes, inadvertent positioning of flap lever, uncommanded slat retraction, inadverdent speedbrake deployment, uncommanded stabiliser movement - to name a few : .

EC,

Well, I suppose anything is possible, but I'd be interested to hear from people who've had the things you mention actually happen to them after TO power was set and the IAS neared V1. "...uncommanded slat retraction...inadventent speedbrake deployment...inadvertent positioning of the flap lever..." ? Inadvertent positioning of the flap lever AFTER to power is set and the IAS has progressed to near V1 ?

I guess it's the probability I question. All seems rather unlikely. Admittedly, I haven't heard of every case that ever happened but in 35+ years of operating these airplanes I never had anything of the sort happen or even heard of these cases in the literature I read.

I learn new stuff all the time....give me some documented "new stuff" to support your contention.

[Capt Claret]

Not a jet I know but I've had an unscheduled autofeather of an engine (PW123) just on rotate. There was nothing wrong with the engine as such, just a faulty PFM box.

[mutt]

Bafanguy,

We operate the MD90 with a "Dial A Flap" system, this allows any variable flap setting to be selected within a certain band. we used to have plenty of high speed configuration warnings when operating at the limit of a band (say Flaps 13) due to the external forces moving the flaps sufficiently to activate the micro switch, ie Flaps 12!

Mutt

[GlueBall]

In the simulator you should always abort before V1, do the exercise for training purposes,...to learn the stopping power of your airplane.

Aborting at V1 in a heavy jet can be a violent maneuver, especially when your gross weight is runway limited, not to mention the condition of the brakes and tires and the delay in applying immediate maximum brake pressure if no RTO auto braking is available.

My crew is always briefed as such that when my hand comes off the throttles that it's a "go" decision and not even to think about stopping. In the real airplane my hand may come off the throttles up to 15 knots before V1, depending on gross weight, weather conditions, runway conditions and runway remaining. And that's the way it is. :ooh:

[BEagle]

And then one day your non-standard '15 kts' becomes what? 20...., 30...?

Probably the most potentially dangerous, non-SOP attitude I've read on this thread.

NigelOnDraft and qwertyuiop have the correct answer!

[Empty Cruise]

Bafanguy,
Have had uncommanded flap retraction in a small TP - on a 1300 m runway. Thank God, it occured at V1 -15 kt. and we stopped quite easily - but I sure as hell didn't have time to look around and see what had caused the horn to start blaring. It turned out that flap maintenance had been carried out during the night, and some of the microswitches had been locked out. I took this incident to prove that two things can and do go wrong at the same time :{.

Ever since then, I subscribe to the same school of thought as qwertyuiop :ok:

Empty

[Essential Buzz]

Ten plus years ago, the company I worked for suffered a series of high speed aborts on the 767 fleet due to config warnings arising late in the takeoff roll. The eventual cause was traced to moisture in certain "cards" in the E&E compartment.

The official word from Boeing was: If the config warning hasn't sounded in the low speed regime, DON'T STOP if it sounds later in the takeoff.

[chornedsnorkack]

Surely the answer is;
If you get a Config warning before V1 the aircraft CAN stop, but may not fly. Unless you brief otherwise you STOP.
How do you then define V1?

On a runway, there is a speed above which a plane that aborts will overrun. V1 should not be more than this.

There also is a speed under which the plane cannot fly with a very well defined single-point failure - namely 1 critical engine out, all other systems intact. The cause of being not allowed to fly might be inability to accelerate with the good engine/s to climb above the screen height by the end of runway, or clear obstacles further on flightpath. Or else the inability at slow airspeeds to keep the aircraft on runway against asymmetric thrust plus crosswind.

"Balanced field length" seems to be the length of runway where there are no margins of safety over and above the accelerate-stop distance. An aircraft could take off from a shorter runway, but is not allowed to, because there is a stretch of takeoff where one engine out would cause a certain crash.

Whereas if the craft takes off from a longer runway, there would be a stretch where both aborting and continuing would be safe... there would be leeway for pilot judgment what to prefer, no?

But BFL and V1 are defined by reference to a specific one-point failure - one engine inoperative. Configuration warning is a quite different error condition.

Examples to wrong configuration causing crashes include the Chinese Bae 146, as well as the first 747 to crash in Nairobi. In Nairobi, the slats did not deploy - the aircraft got airborne, but stalled past the end of runway and crashed with many deaths.

Did the pilots at Nairobi have (and miss) any chance to realize what happened and abort?

[john_tullamarine]

"Balanced field length" seems to be the length of runway where there are no margins of safety over and above the accelerate-stop distance ...

I think that you are trying to make BFL into something more complex than it is .. viz., just a convenient procedure which makes the result of the AFM calculation, however performed ...

ASDR = TODR (TODR being whichever TOD case is most limiting)

no more, no less.

Its only real value is that it gives the quickest calculation and/or the simplest derivation of the speed schedule.

[Centaurus]

Essential Buzz. Your reply quoting a Boeing advice on the matter certainly has got my attention. I suppose you wouldn't have that advice on paper would you? It opens up the subject more than I thought. Please PM me with the info or where I could lay my hands on it. Where there is runway length well in excess of the balanced field length content especially allowing for reduced braking on wet or slippery runways then the abort decision may not be critical, although it may frighten the passengers. I sometimes wonder if pilots are fully aware of how much runway is to spare and keep it up their sleeve for a nasty event.

Simulator instructors occasionally see things that horrify them when it comes to rejected take off actions and I have no doubt in my mind that in the real event the chances are delays will occur due indecision and dare I say momentary shock and incompetency and an over-run occurs. Depending on the over-run area characteristics (road, ditch, cliff edge and you name it), someone is going to get hurt. There is no shortage of documentary evidence on this subject going back fifty years in old ICAO Accident Reports.

A high speed rejected take off on minimum runway length is something of a gamble that all actions are faultlessly executed. A decision to continue in the case of a late config warning is also a gamble. Historically the former has resulted in more fatalities than the latter. Or so I believe personally, anyway.

[chornedsnorkack]

Essential Buzz. Your reply quoting a Boeing advice on the matter certainly has got my attention. I suppose you wouldn't have that advice on paper would you? It opens up the subject more than I thought. Please PM me with the info or where I could lay my hands on it. Where there is runway length well in excess of the balanced field length content especially allowing for reduced braking on wet or slippery runways then the abort decision may not be critical, although it may frighten the passengers. I sometimes wonder if pilots are fully aware of how much runway is to spare and keep it up their sleeve for a nasty event.
Logically speaking, using a single V1, or a single other speed decision point, whether V1-15, V1-20, V1-30... means neglecting to take account of how much or how little there is excess runway length to spare. An perhaps not being fully aware of this...

Simulator instructors occasionally see things that horrify them when it comes to rejected take off actions and I have no doubt in my mind that in the real event the chances are delays will occur due indecision and dare I say momentary shock and incompetency and an over-run occurs.
I think official V1 and BFL numbers are supposed to incorporate pilot reaction time somewhere, setting it as 2 seconds. If you think it is going to be more, like 3 or 4 seconds or whatever, you might try and estimate the results - with accompanying changes to V1, BFL, MTOW...
Depending on the over-run area characteristics (road, ditch, cliff edge and you name it), someone is going to get hurt.
Something is going to get hurt, someone may get hurt.
There is no shortage of documentary evidence on this subject going back fifty years in old ICAO Accident Reports.
A high speed rejected take off on minimum runway length is something of a gamble that all actions are faultlessly executed. A decision to continue in the case of a late config warning is also a gamble.
High-speed abort past V1 seems to be a somewhat predictable gamble. There is no chance of the plane coming to stop on the runway to taxi back and try again, but there also is no chance of the plane reaching overrun at a speed greater than whatever speed the plane aborts at. So, the outcome depends on the (known) character of the overrun area and (predictable) speed at reaching the overrun.

Whereas continuing is a much bigger gamble. The plane might take off and climb with no problems if the warning was false alarm. Or the plane might barely pass the screen or obstacles, yet make no contact and suffer no damage, and eventually land unhurt. Or alternatively, the plane might hit something at a great speed, trying desperately to gain speed and climbing capability above V2, or enter stall or spin.

Historically the former has resulted in more fatalities than the latter. Or so I believe personally, anyway.

[alf5071h]

chornedsnorkack I suggest that you revisit your performance course notes, or review the Airbus advice: Getting to grips with aircraft performance (pdf 7.88mb) (www.wingfiles.com/files/performance/gettingtogripswithaircraftperformance.pdf)
Also see the FAA Take off training safety aid. (www.bluecoat.org/)
And Certification rules for large aircraft, (pdf 4.1mb) ( www.easa.eu.int/doc/Agency_Mesures/Certification_Spec/easa_cs25_amendment_1_12122005.pdf) CS25.101 page 10 onwards deals with performance.
If anyone wishes to do their own ‘in service’ flight testing then see: Flight test guide for certification of transport category airplanes, (pdf 3.3mb) (http://www.easa.eu.int/doc/Agency_Mesures/Certification_Spec/easa_cs25_amendment_1_12122005.pdf) page 71 onwards.

[john_tullamarine]

chornedsnorkack,

(a) may I echo alf5071h's observations ...

(b) Logically speaking, using a single V1, or a single other speed decision point, whether V1-15, V1-20, V1-30... means neglecting to take account of how much or how little there is excess runway length to spare

Having selected a V1, the AFM data provides a certification distance for a set of cases - ASDR, TODR (AEO, OEI), TORR (AEO,OEI) etc. It is then a matter of the ops eng (or pilot) repeating the calculation a few times to match the AFM data to the particular runway environment. This can be done conservatively (ie all the cases fit but RTOW is not optimised) or to the optimally limiting case (in which the case the RTOW is maximised).

(c) I think official V1 and BFL numbers are supposed to incorporate pilot reaction time somewhere, setting it as 2 seconds. If you think it is going to be more, like 3 or 4 seconds or whatever, you might try and estimate the results - with accompanying changes to V1, BFL, MTOW

The reaction times vary with the Design Standard version relevant to the particular Type's certification. The specifics are in the individual distance data items (eg ASDR). Particularly with older Standards, the fat is less and the pilot's flexibility (near V1) reduced. If one desires any given time delay during the accel-stop transition, that is fairly straightforward .. but not for estimating ... calculating.

(d) Whereas continuing is a much bigger gamble

Probably we are more interested in risk outcomes than the gamble per se. Regardless, the historicals suggest that your postulate is false in that, for near V1 failures on limiting runways, the probability of an acceptable outcome appears to be weighted towards the GO option. One should also qualify comments by considering whether the case in question is a limiting runway and, if so, which case is limiting. In the conservative case of a grossly non-limiting runway, the arguments can become academic/legal procedural.

[Cool_Hand]

I'm with qwertyuiop,

Surely you should react as briefed, that's the point of V1, if a config sounding horn is an abort then you abort. If it happens at V1 or above then you take off. The whole point of V1 is to remove the processing time for the pilot during a critical phase of flight. You mention all things that are a stop and stop if ANY of them happen.
Just be very clear in your take off brief what your intentions are.

[Intruder]

At my company, a Configuration Warning below V1 in the 744 is one of the specific high-speed Reject criteria. However, in the 747 Classic it is only below 80 KIAS.

[chornedsnorkack]

the Airbus advice: Getting to grips with aircraft performance (pdf 7.88mb) (www.wingfiles.com/files/performance/gettingtogripswithaircraftperformance.pdf)
Also see the FAA Take off training safety aid. (www.bluecoat.org/)


Excellent sources, that explain matters more thoroughly than the certification rules (which omit the underlying reasoning).

The FAA takeoff safety aid seems to complain (in 1993) that data to estimate actual speed where the takeoff could be safely aborted from a non-runway-limiting-weight case are often not available. Is it still the case?

[captjns]

Aircraft at hight gross weights on limiting runways will do better in the air than trying to stop on the runway remaining.

[Gary Lager]

Unless the slats have retracted...;)

[mutt]

Aircraft at hight gross weights on limiting runways will do better in the air than trying to stop on the runway remaining.

Interesting statement, does it apply to ALL aircraft ???

Mutt

[john_tullamarine]

Like all generalisations, the hurt is in the detail. Not such a simple matter, I fear ... consider turbulence, for instance, and whatever get up and go you may have thought you had has just got up and gone ....

The main protection is that we don't have very many critical flight phase failures .. it is always most important to keep in mind very firmly the fact that the book data is based on reasonably good conditions .. turbulence, shear, inversions, speed control etc., for OEI climb are the killers ...

[Essential Buzz]

Centaurus - Sadly I have no documentary evidence to support my reference to Boeing's feedback. At the time discussions were definitely held between the Fleet Manager's Office and the manufacturer, but this was only summarised by way of a Notice to Crew, or similar.

Ironically, I was (unsuccessfully) seeking a copy of that very notice some years later as the same company had a different policy on their 737 fleet! (Anecdotally, I understand the 737 has had similar false config warnings - also due to corruption of circuit boards buried somewhere in the E&E compartment.)

As an aside, I wonder about the possibility/likelihood of a slat retracting during the takeoff roll. Even if it is possible, I would have thought a "slat-generated" warning would be much more likely to be caused by a problem with a micro-switch/proximity sensor instead of something that should be hydraulically locked in place.

In the end, it all gets down to "balance of risk", which is invariably assessed with the wisdom of hindsight!

[chornedsnorkack]

Like all generalisations, the hurt is in the detail. Not such a simple matter, I fear ... consider turbulence, for instance, and whatever get up and go you may have thought you had has just got up and gone ....
The main protection is that we don't have very many critical flight phase failures .. it is always most important to keep in mind very firmly the fact that the book data is based on reasonably good conditions .. turbulence, shear, inversions, speed control etc., for OEI climb are the killers ...

By and large, is it correct to state that
a) the book data are based on a rather specific failure condition - OEI and all other systems intact? And that therefore, the safety margins certified in case of OEI are not necessarily assured to be available in case of any other failure condition, like wrong configuration?

b) that there is no "safe side" of V1 in the case of balanced field/runway limited takeoff weight, because the certification assumption is that rejecting at any speed after V1 is unsafe (overrun) and continuing at any speed below V1 also is unsafe (whether through inability to climb to screen height with OEI, or through veering off the side of runway due to asymmetric thrust of the good engine/s)?

The main protection is that we don't have very many critical flight phase failures ..

But should then one make the assumption that, since there are not very many critical flight phase failures, warnings in critical flight phase should be assumed to be false warnings? That sounds a dangerous assumption.

[Capt Chambo]

[QUOTE=Centaurus]In most take off safety briefs the mention is made of the various events that warrant a rejected take off. A Take off configuration warning is one of those events.

If rejecting the take-off for the TOC warning is part of your companies SOPs, and the Pilot briefed that (s)he would reject the take-off for a TOC warning then they should rejected the take-off.

Rejecting before V1, all the data shows that you will stop safely.

It is also dangerous to try and take comfort from the fact that you got away with it the simulator, the aeroplane may perform somewhat differently. The Lufthansa crash off RW 24 Nairobi in the '70s was attributed to the lack of deployment of leading edge devices, a stark reminder as to just how critical LE devices are to low speed lift.

[RYR-738-JOCKEY]

Back to the topic. I see that different pilots apply different interpretations of a situation which ideally should be treated exactly in the same manner. The T/O config warning in my craft sounds if:

-trailing edge flaps are not in the flaps 1 through 25 T/O range, or
-leading edge devices not configured for T/O, or
-speed brake lever not in the down position, or
-spoilers not down with the speedbrake lever in the down position, or
-parking brake is set, or
-stab trim not set in the T/O range

So FLAPS, STAB TRIM, PARK BRAKE, SPEEDBRAKE not in T/O config. I would say that all warnings should be treated without any assumption that it may be a nuisance warning. And furthermore if you get THIS warning the aircraft IS unsafe to fly, hence the trigger-sources.
So, below V1: Abort. One can dispute weither an abort at V1 is safe or not, but one should not treat abort cases as if they can be independently evaluated from case to case. That is what causes trouble...when pilots have different ways of assessing a particular situation. Stick to your SOP.

[Empty Cruise]

RYR-738-JOCKEY,

Yes, I think we do need a discussion on weather it is safe to abort at V1... My question to start this off is: if you consider it unsafe to abort at V1, why then did you choose a V1 that you don't consider it safe to abort from?

Also, the definition of "safe" is up for debate. Is a 20 kts overrun into the approach lights at the other end unsafe? Is departing with retracted slats safe?

I think it is time these near-V1-reject-debates are put into perspective - if you're not comfy aborting from V1, you've got the wrong speed! If you feel it's safer to continue with a 0,1% chance that the aircraft is no longer in the take-off configuration, priorities might need be re-arranged. Damaging the undercarriage, gear doors and lower fuselage in an overrun - but having everyone walk away - is an option (albeit not on every runway - so the V1 should be set accordingly).

Empty.

[alf5071h]

Empty Cruise don’t try to double think the regulations and SOPs. Over many years of experience, the authorities and industry have debated the safety of RTOs. The accident data shows that the human making the decision is the weakest link.
See all of the text in FAR 25.109 (http://a257.g.akamaitech.net/7/257/2422/12feb20041500/edocket.access.gpo.gov/cfr_2004/janqtr/14cfr25.109.htm), note para (e) … determine the accelerate-stop distance if that means--
(1) Is safe and reliable;
(2) Is used so that consistent results can be expected under normal operating conditions; and
(3) Is such that exceptional skill is not required to control the airplane.
And also note para (h). If the accelerate-stop distance includes a stopway with surface characteristics substantially different from those of the runway, the takeoff data must include operational correction factors for the accelerate-stop distance. The correction factors must account for the particular surface characteristics of the stopway and the variations in these characteristics with seasonal weather conditions (such as temperature, rain, snow, and ice) within the established operational limits.

[john_tullamarine]

chornedsnorkack,


... the safety margins ... are not necessarily assured .. in case of any other failure condition

Depends on the A/L pertaining to the certification .. but that, together with variation in environmental conditions (esp runway surface), is the crux of the worries one should have ...


.. there is no "safe side" of V1 ... the certification assumption is that rejecting at any speed after V1 is unsafe ...

I suggest this misses the underlying point and principal real world risk concern.

The V1 concept (similar to much of the handling and performance aspects of certification) puts one of a set of lines in the sand. If the certification boundary conditions are replicated, then it is valid to presume a reasonable correlation between the certification data and the real world .. if not, then such a presumption will be tenuous to an extent reflecting the divergence.

If the particular runway condition is similar to the certification presumptions AND is distance limited for ASDR, then an accel-stop with a speed excursion likely will be fraught with an unpleasant lack of success .. the consequences are related to the particular environment only and have naught to do with the certification. If, on the other hand, the TODA/TORA is comfortably not limiting (ie not BFL limited), continuing with a delta below V1 (presuming Vmcg is not limiting) likely may be quite successful. In a similar manner, one can argue the complementary situation with limiting TODA/TORA, but non-limiting ASDR.

However, procedurally (ie from an SOP point of view), using the above argument is quite unacceptable as a general principle.


.. should then one make the assumption that ... warnings in critical flight phase should be assumed to be false warnings? That sounds a dangerous assumption.

Couldn't agree more. Two factors at work here ..

(a) procedurally, unless there is reliable evidence to indicate that a warning is false, it should (must ?) be considered valid

(b) historical reality is that many warnings are false and this needs to be considered in the captain's decision making process where there is time to reflect on multiple courses of action.

[Centaurus]

Further to the original post about going near V1and a configuration warning, the recent up-dated edition by Boeing for the FCTM B737-200, April 2005, makes an interesting distinction between low and high speed configuration warning. Quote in part: "Prior to 80 knots, the take off should be rejected for any of the following:
Activation of the master caution system.
System failure(s).
Unusual noise or vibration.
Tire failure.
Abnormally slow acceleration.
Unsafe configuration warning.
Fire or fire warning.
If the airplane is unsafe or unable to fly.

After 80 knots and prior to V1, the take off should be rejected for any of the following:
Fire or fire warning.
Engine failure.
If the airplane .is unsafe or unable to fly" Unquote.

I may be reading more into it than was intended by Boeing, but notice that there is no mention specifically after 80 knots of rejecting for an unsafe take off configuration warning. It is of course left to pilot judgement if he considers the aircraft is unsafe or unable to fly. This would perhaps tie in with a previous Pprune comment on this thread where Boeing suggested that a take off configuration warning nearing V1 was not necessarily a cause to reject the take off. Naturally Boeing cannot cover every conceivable eventuality and one presumes therefore it is odds on that a late configuration warning is a false warning.

[RYR-738-JOCKEY]

Centaurus wrote: "This would perhaps tie in with a previous Pprune comment on this thread where Boeing suggested that a take off configuration warning nearing V1 was not necessarily a cause to reject the take off. Naturally Boeing cannot cover every conceivable eventuality and one presumes therefore it is odds on that a late configuration warning is a false warning."

Frankly speaking, getting closer to V1 should not require more thought processing from the pilots. As an FO my job is to call out any failures. And I expect that my CP acts according to SOP. If close to V1 and the config warning sounds and it is "not neccesarily a cause to reject the take-off", who shall check the STAB/FLAPS/PARK BRAKE/SPEED BRAKE? My point being, I simply cannot see that there is enough time available to assess the situation.

[Empty Cruise]

alf5071h,

Sorry if I did not make myself clear enough - I advocate following yours and JTs advice and going with the numbers, and definitively not doing any second-guessing or double-thinking. In line with that, I take the liberty of questioning those who want to portrait a near-V1-reject as an unsafe manoeuvre which - although demanding - it clearly is not.

My protestations are against those who'd "take the chance, since said chance is so small" that the aircraft is still in the take-off configuration rather than aborting the take-off near V1. If your AFM says that high-speed T/O config warnings should be considered invalid, by all means, that is what we do. If your AFM hasn't got any such statements in it (and your SOP is not specific either), in my view there can be only one call - and that's "stop". You'd end up looking rather silly (and possibly looking slightly dead) having a config warning at V1-10 and then continuing only to find out that the aircraft is unflyable at Vr+10 and then trying to stop from that speed, right?

On the other hand - even if you bungle the reject and run over at 20 kts. - will you be better or worse off than if you continued with the slats retracted, or the stab out of trim? It might just be me, but I find the answer to this question fairly obvious. If you consider only "damage to aircraft", I recognise that the "continue"-crowd have statistics on their side. If you however consider "survivability", I know that the "stop"-mentality puts us in a much safer place. I have not done the maths, but it would be interesting to compare the numbers of fatalities on "mishandled rejects prior to V1" vs. "should not have attempted to leave the ground".

If people feel that a V1-reject is of questionable outcome, that alone is in my view and indication that these people are mentally aligned with the "call-V1-early-crowd" (debated at length elswhere). I suggest you might be better off not derating to the maximum but rather select a lower derate and/or V1 to keep the V1/Vr/V2 relationship intact. So to sum up my view: Fly the numbers, know your AFM, FCTM & SOP - and make your decisions accordingly. Don't make assumptions about the flyability of your aircraft.

Empty

[chornedsnorkack]

On the other hand - even if you bungle the reject and run over at 20 kts.

With a bungled reject, is it actually very probable?

At 20 knots, an aircraft with good braking should be able to stop in less than 20 metres. With poorer circumstances like ice or hydroplaning, this could take longer, but still... an aircraft might come to full stop "five feet" or so before an obstacle, as in the (fictional!) Airport, but it is much more likely either to stop completely hundreds of metres before the end - or far after the end.

So, the consequences of overrun at high speed are not always trivial.

On the other hand, finding the craft unflyable is still worse.

Also, an aircraft at V1 in runway-limited load has spent around 60 % of runway and only 40 % is left for stopping. An aircraft which lands has about 100 %. But continuing a takeoff with flyable craft may be setting up the need to perform an overweight landing...

[Empty Cruise]

Chornedsnokack,

Agreed, 20 kts. might be unrealistic - so let's call it 60 or 80 kts. Even so, the vast majority of airports have RESAs that will damage the aircraft, but not beyond survivability. On those where that is not the case (KIL & FAE spring to mind), you go with the lowest safe V1.

However, at those very same airports - I really don't fancy my chances at FAE if I have to abort at Vr +5 because the aircraft did turn out to be unflyable... and the locals down at the foot of the hill prolly don't fancy a visit by an out-of-control-aircraft (God knows them people have seen some hairy tings happening to aircraft already).

Just my 2 kronurs worth ;)

[AirRabbit]

GlueBall,
I take it your remark was rather tounge-in-cheek. If not, please expand on that philosophy - I'm eager to learn. I was just brought up with this wimpy "Train the way you fight - fight the way you train"-attitude that is obviously past its sell-by-date :D
Empty
Without even getting close to a tongue-in-cheek comment, there isn't anything "wimpy" about that attitude. Those of us who have been "in the business" for more than a few decades (at least those of us who really care - and Centaurus is certainly one!) have been trying unceasingly to instill that very attitude in all pilots; i.e., train the way you fly, and fly the way you train. Unfortunately, it wasn't until the advent of what I would call "modern simulation" (within the last 15 years or so) that really allowed this to be done in earnest. Training in an airplane doesn't really lend itself to proper or compete training and there are specific and documented reasons for this. In today's environment, the idea of having two sets of skills (relying on one in a simulator and the other in the airplane - which tends to get confusing in circumstances when split-second decision making is required) is an outmoded and a "detrimental-to-your-health" idea that should be expunged from every airplane training curriculum.
While I certainly don't mind having serious discussions punctuated every once-in-a-while with a little levity, sometimes "light-hearted" comments can be taken the wrong way. That is why I am glad that you had the temerity to ask the question you did - and it's why I jumped on your comment so quickly. Flying is one of the most rewarding and exciting and entertaining things humans can do (note, I did say "one" of the things...), but it is also tremendously challenging - especially when doing it as a professional - and as expansive as is the "wild blue yonder," it is no place for anything less than professional action all the time. And, speaking for those of us for whom the "art" of flying comes only with great practice and diligence (as opposed to those rare few who are "natural" pilots), let me encourage you that "practicing" anything with less than the attitude you've learned is certainly NOT wimpy. Ignoring that kind of attitude not only takes away valuable time from learning and practicing what is necessary and proper, it also more than likely reinforces incorrect decision making. There are a lot of folks who say, with varying degrees of tongue-in-cheek thought, "I'd rather be lucky than good." If you (not "you" specifically, but anyone), if you fly and you are depending on being "lucky," you're in the wrong business.
Keep that attitude, my friend, and don't let anyone talk you out of it!
______
AirRabbit

[Ignition Override]

Capt Claret: in 1979, following failure of an American Airlines DC-10 engine pylon, with separation of the engine and damage to the leading edge, one slat retracted.

But the crew had no indication of the change. They only thought that the engine had flamed-out, until the plane rolled out of control.

Maintenance technicians at the Tulsa facility used a very wrong procedure to change the engine under the wing.

[chornedsnorkack]

Capt Claret: in 1979, following failure of an American Airlines DC-10 engine pylon, with separation of the engine and damage to the leading edge, one slat retracted.
But the crew had no indication of the change. They only thought that the engine had flamed-out, until the plane rolled out of control.

Had the crew immediately realized that one slat was retracted, could they have done anything to keep roll control?

[Centaurus]

Chorned etc. Lower the nose and increase speed. Or reduce the thrust on the other engine at the same time lowering the nose for more speed. It may be a balancing act, but anything as long as the roll is stopped.