derated take off during Engine failure

[idisaid]

Tm say if an engine failure occurs during take off, any thrust increase beyond the fixed derate limit could result in loss of directional control....

why we can loss control?

[Denti]

With a fixed derate you use legally an engine with that lower thrust limit and the corresponding speeds. Especially Vmcg (minimum control speed ground) and Vmca (minimum control speed air) are based on the thrust setting.

If you increase the thrust beyond the fixed derate limit and your speed is near the approved Vmcg/a you might lose control as the controls cannot cope with that amount of thrust at that low a speed.

[TypeIV]

The greater the thrust, the greater the assymetry in case of an engine failure.

The greater the assymetry, the greater will the control forces required be.

With under-wing mounted enginges this can be a problem. I don't know about aircraft with the engines mounted in the tail though, anyone with experience on the falcon or MD?

[RAT 5]

It is to do with the Vm speeds; so I was instructed years ago. Further, the performance calculations have a little buffer between 100% perfect and the line duffer on the day. Thus they say it is not necessary to increase thrust to achieve the profile.
What I find disconcerting, and another dumb don in knowledge, is that people say it is not allowed to increase thrust. Some say they are nervous that once PF has the correct rudder inputed close to the ground it would be disturbing to cause an adjustment.
Personally, IMHO, there should be a 'note to self' in the training and during the manoeuvre: "choose 1.......a/c is performing as required......leave thrust alone. A/C is not climbing as required.......increase thrust and control a/c."
To not teach people the pro's & con's of use of thrust is negligent, but I see it every day.

[Meikleour]

RAT5 : see Denti's reply. Are you confusing a FLEX take off with a DERATED one?
Very different beasts.

[barit1]

Are you confusing a FLEX take off with a DERATED one? Very different beasts.

Yes! Not to put too fine a point on it, review ops on a contaminated rwy.

[misd-agin]

Since increasing power reduces the controllability margin wouldn't decreasing power increase the controllability margin? Think about it. If you reduced the good engine's power to match the power of the bad engine lateral control issues would be reduced to zero.

[RAT 5]

I know nothing of Flex takeoffs. I understand the effect of thrust/speed on yaw & roll control. It is an effect at slow speed. VMCG is an on ground scenario. VMCA is airborne and again slow speed. If airborne and above V2 I suggest you should not be anywhere near VMCA. If I have the yaw & roll under control, but the ROC is suffering for an unknown reason, I feel capable of slowly increasing thrust to try and solve that problem while retaining control about the other 2 axis. That is what I've been trained to do, and in light twins, non perf A, was a regular airborne training manoeuvre with new pilots on type. Educating and illuminating, too.
I do not advocate increasing thrust as a common technique during an EFATO. (I have flown for one airline whose CP did advocate that as an SOP). What I do advocate is not teaching that it is not allowed.

[FE Hoppy]

because the minimum v1, rotate and v2 speeds are restricted by the minimum control speeds and the minimum control speeds increase as rated thrust increases so increasing thrust beyond the rating used to define minimum control speeds and those speeds limited by them could put you in a position where the yaw effect of engine out is greater than can be countered aerodynamically at your scheduled speeds.

[RAT 5]

This is becoming far too theoretical and in the ideal world for my simple way of thinking. I know the theory and the ratios & %'s for the various speeds. In the sim it works fine and has done for 40 years. In the real world, as they say on the rugby pitch, you play what's in front of you. The coach's best plans may need adjusting on the day. You need to be well aware of what you're doing. Perhaps there-in lies the hesitation. Too many treading into the darkness of the unknown nor understood.
Indeed there have been a/c whose rudder had too little authority, and with an EFATO it was necessary to reduce thrust on the live donk. That was to gain control in yaw and 2ndly roll. Assuming those 2 were acceptably under control I was only considering what was happening in pitch & ROC and seeing if anything needed to be done about that. I might feel rather sheepish trying to explain why I'd descended shallowly into farmer Fred's ploughed earth while having more power to call upon.

[flyhardmo]

This is becoming far too theoretical and in the ideal world for my simple way of thinking. I know the theory and the ratios & %'s for the various speeds. In the sim it works fine and has done for 40 years. In the real world

I agree with rat. We all know the theory behind it but practically nothing is stopping you from increasing thrust once above VMCA or V2. VMCG means Jack all once you pass it and by V1 you are either on or above it.

[Denti]

VMCG means Jack all once you pass it and by V1 you are either on or above it.

Dangerous assumption, especially on a contaminated runway where the lowest possible V1 is used. Now consider having an engine failure at 98kts, just passed the V1 of 97kts but very shy of the full rated Vmcg of 116 kts and increasing the thrust to the maximum possible to get that darn thing airborne. There will be a very swift and violent sideway departure of the runway why still being on the ground. It doesn't help that some airlines do not publish the Vmcg table anymore, but it is understandable if one considers that one table would probably hurt more than help if you use three different thrust settings.

[Goldenrivett]

Hi flyhardmo & RAT 5,
We all know the theory behind it but practically nothing is stopping you from increasing thrust once above VMCA or V2.
Indeed there have been a/c whose rudder had too little authority, and with an EFATO it was necessary to reduce thrust on the live donk.
Do you really understand the theory?

May I suggest you read: ASN Aircraft accident Boeing 707-436 G-APFK Glasgow-Prestwick Airport (PIK) (http://aviation-safety.net/database/record.php?id=19770317-2)
"At the light aircraft weights used for training, standardised takeoff airspeeds were used as follows: V1 - 125 knots, VR - 135 knots and V2 - 145 knots. Based on the prevailing conditions and actual aircraft takeoff weight of 94,580 kg, the true takeoff airspeeds were: V1 - 125 knots, VR - 125 knots and V2 - 142 knots; the VMCG and the VMCA were 125 knots and 119 knots respectively."

But they still crashed.

Please see the AAIB report: https://assets.digital.cabinet-office.gov.uk/media/5422f2fced915d1371000437/6-1978_G-APFK.pdf
Para 2.4 which calculates the VMCA had risen to 160 kts with wings level.

You don't normally get that sort of in depth analysis if you crash the simulator.

[FlightDetent]

Yup. If you're above Vmca that had been calculated for 23k thrust, and decide to put in 27k, you may well find yourself inverted.

Another one here:
ASN Aircraft accident Airbus A330-321 F-WWKH Toulouse-Blagnac Airport (TLS) (http://aviation-safety.net/database/record.php?id=19940630-0)

[john_tullamarine]

The main desirable trait of this thread is the general admonition to be careful and cautious ..

I feel capable of slowly increasing thrust to try and solve that problem (my emphasis) ..

RAT 5's comment should be read carefully by all as it is written in blood. The certification animal doesn't look at thrust increases during a dynamic manoeuvre.

As a f'instance, an aeroplane with which I had some familiarity in a previous life had a reasonably well-behaved static Vmca and, indeed, could get to significant yaw angles. But life could get interesting in the dynamic situation.

I was involved in the on-site investigation of a fatal which (almost certainly - based on eye-witness debriefings) involved one of the pilot's panic advance of one throttle as the aircraft was about to impact the takeoff overrun. This particular aircraft had extremely good engines (very high mod status embodiment) and the thrust response, quite rapidly, saw the aircraft yaw/roll significantly and cartwheel after the wingtip impacted the runway strip end. (There were some related things which made the pilots' job much more difficult than it ought to have been but the thrust/yaw/roll/crash/burn/die exercise was, nonetheless, very obvious and devastatingly effective .. ).


If you elect to increase thrust (I'm not suggesting whether you should or should not) .. do it slowly so that the asymmetric effect can be assessed more easily and the requirement for rudder input adjustment made with some degree of confidence.




As an aside, the B707 report is worth reading to note the importance of bank angle on Vmca ..

[mustafagander]

Denti
One of the conditions to determine V1, at least in the Boeing world, is that it may not be less than Vmcg.

Logic demands this as well - given that after V1 you are committed to fly it will be unpleasant if you are unable to control the aircraft for a while.

[john_tullamarine]

... but note that Vmcg is determined for prescribed wind component conditions .. if you are playing in the sand pit at light weights (ie low speed schedule) and in strong crosswinds ... all bets are off if the "wrong" side engine fails and you find yourself in the weeds.


The go-after-V1 mantra may need to be revised very quickly if you find yourself without adequate control (ie out of control).


.. in strong crosswind conditions at light weights on a local training exercise .. he who takes off without using speeds for higher weights .. is, indeed, a brave training captain.


Real world Vmcg can increase quite significantly with crosswind.

[RAT 5]

RAT 5's comment should be read carefully by all as it is written in blood.

John: Firstly I respect your knowledge and have learnt from many of your previous discussion points. This comment sounds most ominous. I wonder if I have understood it correctly, or if others have understood my comments correctly.
I have not advocated a technique of increasing thrust, ON THE GROUND, after an engine failure. I have not advocated, willy nilly, increasing thrust once airborne. I have certainly not even hinted at stuffing the lever to full thrust. What I've asked in debate & conversation is to consider what you might do, if using reduced thrust, when you find the a/c not accelerating or even sinking, or at any rate not climbing as you consider necessary? Who knows, an engine catastrophe might have prevented the gear retracting fully. What then to all your imperfect calculations?
It is simply a debate about whether pilots should be taught it is NEVER allowed to gently and smoothly increase thrust, once airborne, to assist in solving a performance problem without creating another. Once you reach a thrust setting to satisfy your needs then leave it.
Once again we are back to the level of education and depth of training. What I find disturbing is when asking cadets at FFS during TR courses: "what are VMCG & VMCA?" They might give you a perfect definition. Next question, "what do they mean in practical terms?" Often blank stares. You then move on to "what affects them; how can they change; how can you infringe them?" It is very sad to hear the silence.

[john_tullamarine]

Good sir, my apologies if my meaning was obscure.

I was not expressing any concern with your previous post .. only emphasising my concurrence with your admonition to play with thrust levers .. gently. We appear to be in heated agreement, generally ..

.. is to consider what you might do, if using reduced thrust, when you find the a/c not accelerating or even sinking, or at any rate not climbing as you consider necessary? Who knows, an engine catastrophe might have prevented the gear retracting fully. What then to all your imperfect calculations?

d'accord. Certification is all well and good ... and generally works fine in practice .. but is a postulated animal. As you observe, the real world doesn't always read the rules .. and, sometimes, the Commander is called upon to earn his/her dollar for the day by exercising command prerogative.

Once again we are back to the level of education and depth of training. What I find disturbing is when asking cadets at FFS during TR courses: "what are VMCG & VMCA?" They might give you a perfect definition. Next question, "what do they mean in practical terms?" Often blank stares. You then move on to "what affects them; how can they change; how can you infringe them?" It is very sad to hear the silence.

.. isn't that the oft-observed truth .. ?


It is my view that the main value of Tech Log is the depth of experience and technical expertise collectively put by a significant proportion of posters .. the potential educational value is of the greatest importance, especially for the new chums. Knowing who some of those posters are, in their real worlds .. it amazes me just how great the knowledge base is within PPRuNe ..

[RAT 5]

John: I thank you kind sir.

[flyhardmo]

Denti

I've never known VMCG to be above V1 but I'm happy to be shown otherwise. I think like Rat maybe I was taken out of context in that I'm not is suggesting to firewall the engines but more power is certainly available should you need it after V2. Practical application to get you out of a theoretical situation that doesn't go according to the book is what I'm asking everyone to consider.

[Mad (Flt) Scientist]

@flyhardmo

I believe denti is postulating a case where the declared VMCG for the thrust expected to to be used (a derate thrust) is indeed below V1, albeit only just, but the fully rated thrust VMCG is somewhat higher. To then push the levers to get "full" thrust does indeed expose you to a OEI condition while below VMCG.

In addition to that scenario, there's also all the other variables which are not considered in the definition of VMCG (and hence in limiting V1) but which be very much in evidence on the day. Crosswind is one that comes immediately to mind.

[RAT 5]

KISS: V1 must be >VMCG. There are many reasons to reduce V1 but it can never be
< VMCG.

[Skyjob]

Tm say if an engine failure occurs during take off, any thrust increase beyond the fixed derate limit could result in loss of directional control....

why we can loss control?

There are aircraft types, INCLUDING a 27K 737/BBJ-1, that require a minimum takeoff weight before use of higher thrust settings are allowed.

If fixed derates are used in cases where takeoff weights are light in these aircraft, these fixed derates become limitations. More thrust would not guarantee controllability because of the immense momentum of the available remaining asymmetric thrust and restricted rudder effectiveness.

[tdracer]

We did a flight test last week of a new 'thrust bump'. Naturally we were doing a full rated takeoff (otherwise we'd not be testing the bump), and it was a relatively light aircraft.

Prior to takeoff, the Boeing and FAA pilots discussed the scenario of loosing an engine after V1 - it was specifically mentioned that if we lost an engine, it might be desirable to retard the good throttle a bit to improve controllability.

[Denti]

Denti
One of the conditions to determine V1, at least in the Boeing world, is that it may not be less than Vmcg.

Logic demands this as well - given that after V1 you are committed to fly it will be unpleasant if you are unable to control the aircraft for a while.

It was a post in the context of this thread, which is about derated take offs. Yes, V1 must be above Vmcg. However, if using derates, it is above the Vmcg for the derate used, which might differ quite a bit from the full thrust available. Which in turn means that Vmcg, if the pilot increases thrust above the derated thrust, increases as well.

If the Vmcg for the lowest usable derate is , for example, 20 kts below the Vmcg for full rated thrust, and i believe on the 737NG it is even lower, the pilot will violate the certification specs of the performance data if he increases thrust above the derate. In the air, if done slowly and once the flight path is stable, that is not that bad a suggestion. On the ground just passing the V1 of a low thrust rating slamming the thrust levers forward "because we are commited and safe now that we are above V1" might very well be a recipe for disaster.

If one uses user selectable thrust derates, V1 may be below Vmcg, for a higher thrust rating. That is why Boeing is pretty insistent that pilots do not increase thrust when using derates. Especially on the 737 where there is absolutely no indication what the thrust limit for the used derate is, no idea if it is different on other boeings.

[nick14]

Our thrust display shows the fixed derate, IE the vmcg/vmca protected full debated thrust if you will.

If performing a combined derate and assumed temp take-off, advancing the thrust lever to the fixed derate shown on the N1 reference bugs is permitted (however should not be necessary).

If performing a fixed derate take-off advancing the thrust past the fixed value is not recommended but may be accomplished if terrain contact is imminent. Doing so may result in loss of directional control.

I believe what they are getting at is vmca rather than vmcg but I may be wrong. Worst case vmcg is 106 kts, if you are above that you are fine regardless of conditions on the ground. In flight I'm not sure but maybe there are vmca tables out there?

[john_tullamarine]

You won me until the comment re Vmcg. Very much dependent on crosswind, the actual book figure depending on the cert basis.

If you are at the maximum crosswind and the "wrong" side fails around book Vmcg .. then you WILL find yourself well BELOW the real Vmcg for the conditions .. depending on the configuration, think something in the range 0.5-1.3 kt/kt increase on the book figure. Ergo, tiptoe gyrations in the tulips ...

[nick14]

Ok so 108 kts then? Vmcg is calculated for zero wind and unless the crew is provided with further guidance in the form of crosswind additions then we are unable to do anything other than guess.

It would only be an issue in the continued take-off case where v1 becomes the controlling factor. I am yet to come across a v1 limited by vmcg in the -800.

[john_tullamarine]

That's the one - nil wind (or 7kt for older UK aircraft).


However, you are missing the point of the comment.


For example, if your aircraft had a max crosswind limit of, say, 40kt, and you had something approaching that on the day, you could find your actual Vmcg (as opposed to the book figure) increasing by something in the range of 20-50kt depending on the aircraft Type. Have a failure in that grey area with a low scheduled V1 .. and you might find control to be extremely interesting but, possibly, only for a short time until you found yourself in the grass.


You could well be taking off with a V1 somewhat above book Vmcg .. and still get bitten big time if you believe you are ironclad in a strong wind by virtue of the certification Vmcg animal.


The point is to be aware of the potential problem and, where feasible, avoid min speed schedules ..


I don't know the actual delta for the 738 but, if I may hazard a guess, I would guess something in the 1kt/kt region.


I may have an opportunity to ask the question of a knowledgeable chap next week .. if so, I will come back with something a little more accurate.

[nick14]

Ok I see what you are getting at, missed the point.

It would be good to hear a little more.

[actus reus]

John T,

You are correct; the old BCARs required the 7knots which proved somewhat problematic for the B747 as people involved in the original certification programme have told me.

Mad (F) S is on the money. VMCG is a difficult and demanding test point and not every TP will be called on, or permitted, to fly those profiles.

Nick 14, The difference with VMCG and VMCA, or VMCA(2), is that once you are airborne, you are permitted under FAR 25 to use up to 5 degrees of bank which makes a significant difference to controllability compared to VMCG where the aerodynamic controls only are extremely limited in their effectiveness on the ground.

The other thing which is worth mentioning with regards to certified speeds be they VMU which is a datum, or any other speed 'demonstrated' in flight test, and that is we all know the buffers that are applied for the 'average' pilot etc, but every TP not only knows what they are going to do but also they spend a LOT of time and effort at getting those speeds as low as humanly possible.

The commercial success of the aircraft often will depend on the slightest Balanced Field Length calculation advantage over a competitor's aircraft. What I am saying is that with a VMCG close to a V1 (certified numbers), it is most probably circumspect to leave things alone on the operating engine(s) until airborne.

[Denti]

Our thrust display shows the fixed derate, IE the vmcg/vmca protected full debated thrust if you will.

Interesting, didn't know that that was an option. But then, boeing is probably the only company that has a longer option list than Porsche...

Found something, however only for planes without the double derate option:


If an engine failure occurs during takeoff when using both the reduced thrust (ATM) and fixed derate methods:

- on airplanes with the double derate option, Boeing recommends that the thrust levers not to be advanced since the fixed derate limit is not displayed on engine or flight instruments. However, if operators have developed a procedure that makes the fixed derate immediately available to the crew, thrust may be advanced to the fixed limit only.
- on airplanes without the double derate option, the reference N1 bug shows the fixed derate limit. Thrust may be advanced to the fixed derate limit only.

We do have double derate on all our 737s, therefore never really bothered to read beyond the first bullet point.

[nick14]

We have double derate AND the fixed derate thrust is displayed on the N1 reference bugs so we have an odd option it seems.

In either case Boeing don't recommend increasing thrust so therefore I am unlikely to advocate doing so. As a side point if one is still running the last thing I would want to do it put it under more stress or change the status quo on a possibly damaged engine.

[galaxy flyer]

I did a memorable take-off on a wonderful English day--hard rain and 20 knot crosswind component-in the Challenger 300 some years ago. Vmcg is 111, min speed scheduled take-off, V1 of 116 at Biggin Hill. I wasn't handling, but it took healthy amount of aileron and lots of rudder to keep it straight. After clean up, I mentioned that was a setup for an upwind engine failure which drew a confused look until Vmcg was explained to the pilot. The discussion was interesting for many reasons including how poorly and perfunctorily performance is trained.

GF

[john_tullamarine]

If I may bore you with a few more observations ...

you are permitted under FAR 25 to use up to 5 degrees of bank

One needs to be careful here.

Unless the AFM states something different to 5 deg bank into the operating engine(s), it is a REQUIREMENT to use that bank to achieve book Vmca. Vmca is very dependent on bank angle.

Some useful reads -

(a) earlier in the thread, reference was made to a B707 mishap at Prestwick many years ago. Inadequate bank control saw the actual Vmca ramp up 40 kts or so and bite the crew on the tail big time. It is worth a read of the report (https://assets.digital.cabinet-office.gov.uk/media/5422f2fced915d1371000437/6-1978_G-APFK.pdf) to get a feel for the problem.

(b) the Air North Brazilia thread (http://www.pprune.org/australia-new-zealand-pacific/409686-air-north-brasilia-crash-darwin-merged.html) is worth a sobering read.

(c) The investigation report (http://www.atsb.gov.au/publications/investigation_reports/2010/aair/ao-2010-019.aspx), likewise, is essential reading.

(d) Post 417 (http://www.pprune.org/australia-new-zealand-pacific/409686-air-north-brasilia-crash-darwin-merged-21.html#post7045858) in the Brazilia thread links to a youtube video of a QueenAir Vmca departure .. the speed with which it all happens in the last few knots near the real Vmca on the day is chillingly clear. The video title refers to stall but the event looks to be a Vmca departure.

Boeing recommends that the thrust levers not to be advanced

One of the points which should be emphasised is the very real difficulty for the pilot in juggling thrust response with rudder/aileron inputs. Unless you are about to impact the ground in the next few seconds .. in which case, it's worth a try ... better to leave the thrust where it is.

I mentioned that was a setup for an upwind engine failure which drew a confused look until Vmcg was explained to the pilot. The discussion was interesting for many reasons including how poorly and perfunctorily performance is trained.

Isn't the training implication the truth of the state of the Industry ? By the way, you likely to be out for the next Air Show ?

[actus reus]

JT,
Yes, I was a little lax with my wording wrt the '5 degrees of bank'. You have addressed that; thank you.

The many facets of engine inoperative operations, on or near the runway at take off, come together, in a controlled environment, with '1 engine inoperative ferry flights'.

As far as 'performance' knowledge or training goes, it has been worrisome over the years to hear 'but it works in the simulator...'

So, if you say in reply; 'Yeah, but the simulator does not come with an accurate modelling of, say, rolling friction, ground effect, the aerody package is most probably pretty old...'

Blank looks.

We cannot let an in-depth knowledge of performance and what is and is not reality go out of the body of knowledge of our pilot community but how do we get that addressed?

Mind you, I am not saying it is all doom and gloom; I am just saying...fullstop.

[RAT 5]

The discussion was interesting for many reasons including how poorly and perfunctorily performance is trained.

Yonks ago, when I flew the PA-31 Navajo, Beech 90 & HS-125 there were no sims and type ratings were done on a/c. Thus we stalled them for real and did engine out on departure for real. Note, this was at a safe height with & without flaps. The PA-31 was interesting as you had to descend until the dead donk was feathered. What we also did was fly back to VMCA. This gave me a much better understanding of the reality of what was before just an answer in a theoretical CPL exam. Since then, during numerous Boeing type ratings it has never been introduced into initial training in the sim. No wonder these newbies don't have a clue. It is a definition they might with luck remember, and if they do, it will be spouted with little understanding. Meanwhile they still do access/decel to/from VMo with/without speed brakes - yawn, and other not quite so educational manoeuvres that have been there for decades. Even the 'turns - unto 45 degrees - is a minimum handling manoeuvre. I still advocate the No FD aerial ballet with a smooth combination of turns, climbs, descents, speed changes and angle of bank changes. It teaches feel, use of thrust & trim, control deflection and also where to scan. It takes 20 mins, is great fun and the student learns a more about the a/c than many other mandatory manoeuvres.
And why is UAP recovery something left to airline RST programs which might be glossed over in a trice and not done for 3 years after type rating? Let's get some priority in type rating training.

[john_tullamarine]

We cannot let an in-depth knowledge of performance

d'accord.

Certainly isn't going a nice way as the bean counters exercise an ever-increasing influence over corporate management decision protocols and the counterproductive pressures on training increase continually.

In the dream time of long ago, pilot standards was the catchcry and woe be to any who dragged the chain too much .. nowadays ?

I was fortunate in that my first operator put great emphasis on over-training pilots and accepted the associated dollars as a cost of doing business. A client airline for which I provided ops eng and other services for many years, in similar vein, put yet even more emphasis on pilot initial and recurrent training.

I think, on the balance of probabilities, I would prefer to take my chances, in extremis, with one of the earlier style of folk at the helm on a dark and dirty night ...

Like many, I think it will take considerably more blood on the ground before there is a general management and regulatory level awakening to the underlying problems.




RAT 5, d'accord.

[divinehover]

All far to complicated

For a Flex (Airbus)/Assumed (everybody else) temp takeoff VMCG is based on TOGA (Max) thrust. You can therefore safely use TOGA if need be.

For a Derate takeoff the VMCG is based on the lower engine thrust. Using TOGA at speeds close to V1 might result in loss of directional control.

For most modern airliners, a derate takeoff will only be beneficial on contaminated runways where the lower thrust VMCG is an advantage.

[de facto]

most modern airliners, a derate takeoff will only be beneficial on contaminated runways where the lower thrust VMCG is an advantage.

Want to get back to us with that bit later?
All far to complicated

Based on your earlier quote, I would think,but never too late to learn:E