737NG-Engine Failure during a Double Derate take off

[gearpins]

Greetings to all,
here is the Quote: FCTM
Engine Failure During a Combined ATM and Fixed Derate Appendix A.2.5
….on airplanes with the double derate option, Boeing recommends that the thrust levers not 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 derate limit only.

My understanding:
once airborne, on a single engine and airspeed above V2+15 (to achieve full maneuver margins) one can advance thrust lever all the way to obtain full rated thrust of 27 K with out running out of rudder or exceeding 5dg bank.

And the Question:
Has any Operator come up with a customized procedure ?
More importantly is my understanding correct ? :uhoh:
Appreciate your response
cheers

[john_tullamarine]

is my understanding correct ?

I think not - consider derate as a different engine, ATM as flexible thrust. The OEM is suggesting that you stay within the parameters of the lower thrust engine.

[Denti]

One can advance the thrust levers to the full thrust of the selected derate, but not further without knowing where your relevant VMCa is. It is easy enough knowing that limit (select the derate first, note down the N1, select ATM next), however having it readily available in an engine out scenario is a different thing as it is not shown on the instruments.

We use double derate since it was first approved on the NG in the late 90ies and have not come up with a custom procedure. As all take off performance constraints, including single engine climb out, are met with a double derate, we simply fly the aircraft at the derated setting until it is clean and then use MCT from there. Only thing making it even easier would be automatic switching to EOSID on engine failure, speeds and acceleration altitude are already switching to OEI settings automatically after all.

[Skyjob]

When calculating your takeoff performance it will be based on the fixed derated thrust setting, the single engine stabiliser trim setting will be set for the selected fixed derated thrust setting as well, as is the engine thrust limit indicator on the N1 gauges.

When trying to clear the obstacles on the initial climb out after an engine failure, the crew's priority should be to FLY the aircraft, concentrate on the performance you already have and calculated.

Once the aircraft has been flown to safety, the action of setting Max Continuous Thrust is essentially what you are asking for to do earlier on in the procedure. Your request for additional thrust is thus achieved.

Changing the thrust setting to a higher one prior to this point would require not just flying the engine out procedure, but would require additional FMC selections, thrust lever movements, associated trim changes, all of this leads to reduced monitoring activities by the non-handling pilot, whilst adding configuration changes for the pilot flying to manage on top of the failure trying to fly the aircraft.
The above is not a desirable situation in this critical phase of flight.

Selecting the N1 to Maximum Continuous Thrust (MCT) will set the bugs to the full engine thrust available, it can be done at any time, after liftoff )by the non-flying pilot), but changing the thrust levers to this new setting will cause a configuration change which the pilot flying may not be able to handle whilst flying the aircraft at this point thus causes a threat in the operation.

Although the 737NG has no (known to me) Vmca limitations, its BBJ1 (-700) version does have a Vmcg limitation which requires a minimum takeoff weight prior to selecting a higher fixed derate thrust setting on departure. This limitation is no factor once airborne as sufficient speed is available for full rudder authority, thus MCT is selected as per other models, when in a clean configuration, clear of terrain.

[nitpicker330]

You respect the new Derate limit UNLESS you have severe ground contacting Windshear or any other life threatening situation......THEN YOU FIREWALL THE BASTARD...:ok:

[john_tullamarine]

.. but very carefully and steadily advance the lever .. lest you convert a right way up disaster to an upside down disaster at a similar location ...

One should know one's Vmca limits in terms of weight vs speed schedule .. at higher weights, the thrust limit generally isn't a concern .. however, at minimum weights it may be a whole different ball game.

[Centaurus]

the single engine stabiliser trim setting will be set for the selected fixed derated thrust

I thought the stabiliser trim setting from the load sheet assumed a normal two engine climb out after VR. Are you saying the trim is calculated assuming that an engine has failed at V1? If so, in which document is that published?

[Skyjob]

Stabiliser trim takes account of the available thrust selected by fixed derate, however not by a further reduction for an assumed temperature.
This can be particularly important for underslung engines like the 737, where subject to the selected thrust, the created pitch moment greatly varies.

FCTM: When using derated takeoff thrust, the takeoff thrust setting is considered a takeoff operating limit since minimum control speeds (VMCG and VMCA) and stabiliser trim setting are based on the derated takeoff thrust.

The stabiliser setting is for V2 with thrust available from the operating engine.
This means that slightly higher forces are needed in case of an engine out then normal to rotate but when speed is achieved it is in trim. On normal departures, this is also why a slight trim change is required after reconfiguration of the landing gear.

[gearpins]

john_tullamarine, Denti, Skyjob, nitpicker330, Centaurus -thank you folks for the excellent response. :ok:
feels good to revisit basic performance once in awhile. dusting out the cobwebs :O

I understand the difficulty involved in determining the exact N1 value and attempting to set the same at critical phase of flight all while trying to stay on the right side of the disaster :)

leaving the practical side behind and just as a thought exercise....
I was exploring a simpler alternative in the safety given by "V2 + 15" or the white bug. as this assures full maneuverability... and got thinking that if A/C is above this speed then perhaps one has enough rudder to take on 27K.

here is a quote from AIRBUS getting to grips with performance that actually got me started on this in the first place:
"When a derated takeoff is carried out, TOGA thrust must never be selected until the aircraft is airborne and above the minimum flap retraction speed (“F”speed). The reason for this is that performance calculations are made for minimum control speeds, different from the ones of TOGA"

apples to apples F speed in the BUS is the same as V2 + 15 in the Boeing
and TOGA is the same as full rated 27K

appreciate all inputs. cheers

[BARKINGMAD]

I hear this is now a formally approved way of the beancounters saving even more pennies/cents/groats on engine maintenance at the expense of take-off performance.

How long before the first accident consigns this pracice to where it belongs?

Tin hat on and adjusted...................:ugh:

[soundlover]

The highest Vmca we can have on a 27K is not higher than 110Kts....
Has anyone ever took off with a Vr lower than that?
Basically, you can firewall anytime you want and there is no chance to loose stability.

When we go around, we fly even lower speeds compare to the TO speeds, but then we don't really worry about the Vmca....

Food for thoughts!

[Matey]

Thomson policy...

"If additional climb performance is required on a derated thrust and/or assumed temperature reduced thrust takeoff, the thrust lever of the operating engine may be advanced to full rated thrust, provided the aircraft is airborne, the IAS is V2 or greater, and no directional control difficulties are encountered. This guidance will ensure protection against minimum control speed."

[john_tullamarine]

Methinks you might be in line for a potential fall from grace at some stage ...

(a) with the dynamics of the engine failure situation, any time you are at or near to book Vmca for the speed schedule is a time to be wary of control problems. Any fool can handle the routine V1 failure situation with ease ... a bit more in the way of holding the tiger by the tail if the failure occurs half way through rotate ... especially if the weight is around minimum values. When, do you ask ? .. short range ferry comes to mind ..

(b) light weight takeoff (or missed approach) potentially sets you up for a Vmca encounter - certainly applies to the earlier 737.

(c) whether takeoff or missed approach, the dragon is awaiting to catch you at low weights and high thrust. The concern is thrust setting albeit that there may be slight variations to Vmca with configuration.

(d) the line pilot is at risk of succumbing to complacency as just about all the takeoffs experienced (and trained for in the sim) are at higher weights.

(e) Basically, you can firewall anytime you want and there is no chance to loose stability.

This is one of the most alarming statements I have read in a long while.

Be very wary of the dynamics if you choose to shove the lever up to the stop.

I can recall a turboprop fatal I had some investigation involvement in many years ago .. it is highly likely this is exactly what the crew did .. with quite undesirable consequences ... yaw, roll, cartwheel, burn, die. Very fortunate that the restaurant a few metres further past the ditch the aircraft ended up in didn't get involved as it was lunchtime .. the death list may well have been very much higher than it was.

(f) When we go around, we fly even lower speeds compare to the TO speeds, but then we don't really worry about the Vmca.

You may not .. and for normal weights the statement is reasonable .. however, very light weight and it might be a quite different animal to control ..

(g) and no directional control difficulties are encountered

At minimum weights, the dynamics might just overwhelm your ability to achieve this unless you advance the lever quite gently.

[soundlover]

Hi John,

Thank you for your very competent contribution about the subject.
I believe there is no need to be alarmed after my statement.

I try to be very cautios with my actions as well as with my words.
I see you have a good experience and you made extensive research on the subject, I appreciate it. Just remeber that the title of this Thread thus my replies are solely about the B737-NG.

Applying the same safety rules of a turboprop on a Jet is not entirely correct and might be as well... Alarming! (Allow me the irony)


The theory of which I am aware is the following:
The directional controllability during an engine failure, depends on the available trust vector coming from the remaining engine Vs the aerodinamic authority of the rudder (considering an additional 5° Bank towards the operative engine).

That is why in the FCOM - Performance In-flight table of the B737, the weight is not used as variable factor, but is fixed at the MTOW.

If you have one minute to check that chapter of the FCOM, you will notice that the highest Vmcg (which by the way is higher than the Vmca) is 108 Kts.
This value, is derived from a pressure altitude of -2000 ft and an OAT of -60°C.
All other values are lower than that 108 Kts.
For as low as your take off weight on a B737-Ng can be, your lowest V2 with a TOW of 40T and Flaps 25 is 116 Kts.
Basically, with the two worst and most unlikely extremes, there is still a margin of 8 Kts. Anything in between will bring you to have a considerable higher margin on top of your Vmca. We are talking about margins of over 30 Kts.


If you know something that I don't, please don't be alarmed, don't push any little button, be constructive and I will greatly appreciate your contribution to my knowledge.
I am being trained this way by my actual and former company (which also means there are 4 heads of training and 4 chief pilots involved).
Right now I can say I am confident that on the B737-NG, after an engine failure, anytime we are airborn it is perfectly safe (and I would add... Safer!) to firewall the thrust all the way up.
No risk of "dragons" biting assess.

Agree?


I wish you pleasant and safe lunches! :-)

[Jwscud]

Rapid and/or aggressive thrust inputs when asymmetric strike me as not he greatest idea in the world. At a time when you're asymmetric and low to the ground, do you really want to slam the thrust lever from 22K ATM debate to max? You could be adding >10% N1 in the highest power part of the range.

That could be really rather exciting! Think about how you advance the thrust lever on an SE missed approach. Given that you were Perf A legal when you departed, unless you are about to hit a large mountain (which would have been accounted for in your performance anyway) it would not be a choice I'd make.

[john_tullamarine]

I have no specific background on the NG so my comments are generic but conditioned by a reasonable experience of the older models.

Applying the same safety rules of a turboprop

The dynamics are much the same. Too slow, too rapid a thrust increase and it's all over, Red Rover.

The directional controllability during an engine failure ..

Keep in mind that the flight test exercises will look at dynamic and static Vmca ... with a fixed thrust setting. Shove the lever up and the response depends on the engine status and condition and how rapid the thrust ramp up is .. and what overshoot you may end with ..

That is why in the FCOM - Performance In-flight table of the B737, the weight is not used as variable factor, but is fixed at the MTOW.

Not being familiar with the NG paperwork, I can't comment specifically but, as a general rule, heavy weight failures have far less handling problems than low weight (climb performance is a tad difference, however)

Basically, with the two worst and most unlikely extremes, there is still a margin of 8 Kts.

Not a great deal to play with, I suggest. However, we are talking philosophy rather than specific aircraft numbers. Very low weights are deserving of considerable respect.

Anything in between will bring you to have a considerable higher margin on top of your Vmca. We are talking about margins of over 30 Kts.

Two points - the data is based on having enough performance for the circumstances ie you shouldn't need to ramp up thrust and the worry is the variation in thrust and potential for an overswing at the top.

I am confident that on the B737-NG, after an engine failure, anytime we are airborn it is perfectly safe (and I would add... Safer!) to firewall the thrust all the way up.

So long as you do it tenderly ... or have much higher manipulative skill than the majority of us ...

it would not be a choice I'd make.

sounds like a good strategy to me

[soundlover]

John, I really don't think I am better than anyone in flying ;-)
...especially if I can see hundreds of pilots doing exactly the same!


I also flew turboprops, and it's true, the yaw/roll moment could actually be very critical. But on a Jet, and especially on the B737...

Let's do not forget that the Power output at low speeds, is normally proportionally higher on turboprops compared to Jets. That is due to the fact that at low speeds, the angle of attack of the prop is relatively higher, compared to normal climb and cruising speed profiles. Indeed we all know that the highest power is when the turboprop is stasnding still.



Btw, I have to say that I have the feeling the answer would be the product of a risk assessment on what is more dangerous: Terrain closure Vs Yaw/Roll de-stabilizing moment....

Let's put it this way: Is it "statistically" more difficult for an average pilot to:
1. Add smoothly power on the remaining engine and keep the wings level; or
2. Hold the pitch so that the speed does not decrease/increase too much and fly the climb out path with a minimum separation from the terrain of 35 ft...

I am to small to decide that! :-)
If there is a test pilot or anybody who worked in some sort of study boards, maybe for Boeing (since it is the Boeing B737-NG the plane we are talking about), it could be very interesting to hear the opinion.

[JammedStab]

I thought the stabiliser trim setting from the load sheet assumed a normal two engine climb out after VR. Are you saying the trim is calculated assuming that an engine has failed at V1? If so, in which document is that published?

It appears that the stab trim setting for takeoff is calculated for both the normal and engine failure scenario.

This link is to a Boeing publication detailing the subject. It says on page 32,

• A horizontal stabilizer trim setting is used during all takeoffs. These settings are designed to produce an in-trim condition at:
-The proper all-engines operating climb-out speed of V2+ 15 to 25 kts(V2+ 10 to 20 kts for 3 and 4engine airplanes).
-The proper one-engine inoperative climb-out speed, V2.

• These settings are also designed to produce reasonable column forces for the pilots
• Trim setting is provided as a function of the airplane weight, C.G., flap setting, and engine thrust level.

http://www.scribd.com/doc/133403992/Boeing-Weight-Balance

[Skyjob]

Thanks JammedStab, was looking for that quote...

[BluSdUp]

Dear John,,,
Must say I do not like Your tone on this one, considering You do not know what You are talking about.
Turboprops and B737-800,,, please?

Ps, Do not Firewall, the EEC will give you up to 2000 lb extra !

26k or 27k bump, may be needed when heawy ,,,not light, John, considering we derated to 22k and Assumed down in the 50s because we could from intersection S(hortfield ) with an ET.
Ie, Seneca II hot/hi..
26k or BBC World. Ds
Not Your call John.!

[Matey]

To inject an element of practicality into this...are we really going to need full rated thrust with an engine failure at low weight, leading to controllability issues? Where I would be considering full rated thrust following an engine failure would be somewhere like Sharm el Sheikh to the UK at max weight and 35+ degrees with a strong likelihood of a turbulent departure. Minimum speed considerations would not be an issue then. A lightweight departure on, for example, a ferry flight, would not present the same need for a thrust increase above the calculated figure. Something to be covered at the briefing stage methinks.

[john_tullamarine]

Let's do not forget that the Power output at low speeds, is normally proportionally higher on turboprops compared to Jets.

Bit of a red herring, I suggest. The message we should be getting across to the newchums, in particular, is that thrust increases during the OEI exercise and especially at low weight and speed schedules ... are fraught with potential hazard. For the very great majority of OEI situations, there is no need to resort to increasing thrust ..

the answer would be the product of a risk assessment on what is more dangerous: Terrain closure Vs Yaw/Roll de-stabilizing moment....

Absolutely ! However, perhaps it is better to crash right way up with the aircraft under reasonable control .. rather than upside down and out of control ?

Must say I do not like Your tone on this one, considering You do not know what You are talking about.

My apologies .. not trying to be confrontational. Of course, you are entitled to your opinion.

Turboprops and B737-800,,, please?

While the details differ, the overall resulting effect is similar.

bump, may be needed when heawy ,,,not light

Agreed. However, the thrust of the discussion is to get the message across that there be dragons potentially lurking at very low weights and speed schedules.

We might just have to disagree on philosophy, perhaps ?

To inject an element of practicality into this..

Absolutely the case in practice.


For those who prefer to dismiss the concern out of hand .. I should relate that, with sim exercises intended to highlight the problem, even very experienced pilots (who had not seen the low weight/speed handling problems previously) found themselves caught out until they had the opportunity to play with the dragon.

[Skyjob]

are we really going to need full rated thrust with an engine failure at low weight, leading to controllability issues

No, hence the initial suggestion for fixed derated thrust and additional assumed temperature method to be used. Once a continuous climb clearance has or can be attained, after cutback (1500' AGL default but overridable), select full thrust.

This reduces impact of any failure on departure, such as a bird strike or engine failure, while minimising time to cruise level afterwards and thus saving fuel.

[soundlover]

John, if you take off with double de-rate, you basically cut a good margin from the Actual temperature Vs Assumed Temperature.
If on top of that you add improved climb speed profile, your initial margin from the highest obstacle in the first, second and third segment is reduced even furthef.

If you Take Off from a relatively flat area, the my risk assessment would say that the directional control stability could be of higher importance.
If you take off from a field surrounded by high obstacles / terrain, my personal risk assessment brings me to believe that the worst danger is the terrain.

On the B737-NG JET, with light weights, high flap settings, -60ºC of temperature and -2000ft, your directional control cannot be impaired by the over trust because you are flying with more than 15 Kts overall margin from the Dragon nest.

But if you, John, want to be super-safe, maybe it is just better not to fly at all! Let the dragon fly for you.


Greetings from a newchum!

[PEI_3721]

soundlover, et al, the risk assessment is not quite as is suggested.
First the thread title might be misleading; consider a takeoff with engines rated at a value less than maximum, but where this is the new max thrust. Then a reduced thrust takeoff would be allowed under existing rules.
These ensure that in the event of engine failure the minimum speeds do not jeopardise the flight handling safety margin, or the climb performance detracts from the obstacle clearance safety margin. Thus like-for-like, a ‘double derate’ takeoff provides equivalent safety.
It is only when additive and rare circumstances are considered that many aspects of this debate apply, but even then, the safety of any operation requires other considerations, e.g. consider the wind shear risk before takeoff.

Whilst all of the non-additive situations provide a safety margin, the value of the margin or the rate at which the limit could be approached are not identical; in this instance low weight / speed being more demanding.
Thus whatever reasoning there might be for applying more power in rare circumstances, the risk assessment should consider the aspects above, particularly where the bottom line of doing nothing is no more hazardous than for any take-off with or without the ‘additive’ situations, but alternatively acting hastily, or with surprise, or misjudgement of the situation, then the actions could seriously reduce the safety margins and the the ability to maintain control.
Safety is the art of knowing where the ‘edge’ is and maintaining a sufficient margin for any error – yours or others.

Engine Malfunction Recognition and Response. (http://www.cast-safety.org/engine_malfunction.cfm)

http://www.faa.gov/documentLibrary/media/Advisory_Circular/AC25-13.pdf

http://www.theairlinepilots.com/forumarchive/quickref/reducedandderatedthrust.pdf

[Denti]

Don't understand that obsession with increasing thrust (well, it sounds a bit like ssg). Take off performance calculations take into account the current conditions, obstacle situation and any other special conditions. We do fly into airports with quite a bit of terrain around like SZG or INN and have all the special approvals one can get for those airports, we wouldn't even think about increasing thrust until cleaned up there, there is simply not need to do so. It is much easier to fly the airplane, don't fiddle with any thrust issues and follow the EOSID. And of course the lowest V speeds we use are well into two figure territory on the NG.

[john_tullamarine]

you basically cut a good margin

May I emphasise that, as a pilot and performance engineer, I have absolutely no problem using derate plus ATM plus overspeed schedules when appropriate .. providing that the folks doing the sums are competent. Unfortunately that doesn't always apply from what anecdotes I hear from time to time ..

If you Take Off from a relatively flat area, the my risk assessment would say that the directional control stability could be of higher importance.

Concur

If you take off from a field surrounded by high obstacles / terrain, my personal risk assessment brings me to believe that the worst danger is the terrain.

Caveat - if the sums have been done competently and the monitored performance is appropriate (we do all know what sort of ROC we are looking for on the day, don't we ?) then, in the absence of some obvious performance problem (is the actual wind significantly different from what was anticipated ?), terrain clearance isn't likely to be of any real consequence .. which should increase the control problem risk potential in the HRA.

On the B737-NG

As I made clear, I cannot speak specifically to the NG as I didn't fly it. However, with a rapid thrust increase, 15kts might end up being diddly squat as a buffer regardless of Type and Model.

But if you, John, want to be super-safe

Not looking to be super safe - life is a risk whichever way one looks at it - only as safe as I can reasonably achieve. Knowledge of these various and oddball things is quite material to that aim, I suggest.

Greetings from a newchum!

Welcome aboard the Tech Log sandpit.

well, it sounds a bit like ssg

Come to think of it, we haven't seen our friend around for quite some time, now ...


May I commend PEI_3721's and Denti's repective commentaries, immediately above, to the readership ?

[gearpins]

Thank you folks for sharing your expertise and a lot of valuable insight into the various issues associated with this.
In Summary:
[LIST]
The manufacturer does not recommend you do it :=
unless you establish a procedure (e.g.Matey,Thomson policy.)
Difficult to teach the next gen unless it is documented :ugh:
if push comes to shove it can still be done, albeit very gently and at a safe speed above full rated VMCA (which needs to be predetermined.) :eek:
Not many operators out there who have established a "procedure"
[LIST]

thanks once again to All who contributed
cheers:)