747 - Climb on 2 engines?

[noblues]

Does the 747 takeoff performance cover the 2 Eng scenario?

I know if you loose two engines in flight their is a commital point on finals for a go-arounds .... so presume in the rare occasion of loosing two say at V1 or just after Vr it would not climb at heavey weights??

[john_tullamarine]

For most 3/4 engined beasties, the 2EI scenario addresses cruise only .. ergo, during takeoff, with a second failure, one is going to be very interested in

(a) gross weight .. dumping would be the norm

(b) configuration and speed ... usual to require a descent during which the aircraft is part/wholly cleaned up and a speed more useful than V2 achieved.

Definitely no guarantees in this situation ...

The 747 specialists will, no doubt, talk in respect of that machine ....

[skiesfull]

A few years ago in a 747-400 sim, we flew a profile that required one outboard failing after V1 and the adjacent engine failing at 1000ft above ground. This was at a sea-level airport at +28deg C and a weight of 380,000kgs TOW. As there was no terrain to consider, the aircraft was put into a shallow dive, to accelerate and the flaps retracted just below bug speeds. At the UP bug (approx. 265 kts) the a/c was allowed to climb until MSA achieved and then accelerated to 290 kts - where full manoeuverability to turn into the live engines was possible. A very interesting scenario which required precise handling, but the results were (for me at least) quite surprising. No fuel dumping took place during the exercise - it would have made, perhaps, only a reduction in weight of about 10 - 12,000kgs i.e 3-4kts.
Not something I would like to face in the aircraft, as PAX and hand-baggage often weigh more than the notional weights used for the loadsheet!

[zerozero]

That's an interesting synopsis from 'skiesfull'.

But with respect to the original question: a sudden loss of two engines (on the same side?) I think the concern would be more towards the loss of directional control as the other two engines are developing near max thrust. In fact a thrust reduction *might* be in order.

In the case of two symmetrical engine failures, well, in that case you're not concerned with directional control. I submit that firewall thrust would be acceptable while realizing that any engine will exceed its limitation for a certain length of time. Those engines may never get used again, but at least maybe you can return in one piece.

[G--SPOT]

Done it a few times in the sim. As skiesfull mentions, clean up is a interesting as when flaps are up you lose outboard ailerons and therefore full roll authority. We found it better to get some speed before going F up.

But when you're clean and at a decent climb speed (300 kts works) you will be able to climb (allbeit not very quickly).

Overall it handles quite well, not too dissimilar to a twin on one, but less performance and the go around scenario is a bit more complex.

Edited for spelling

[barit1]

Here's a textbook case (http://www.ntsb.gov/ntsb/brief.asp?ev_id=20208&key=0) in which the trainee cpt couldn't handle a DC-8 with two out during approach. (The NTSB brief has lost track of the fact 13 were killed in a motel)

Simulator fidelity has progressed to the point where you can rerun this scenario day in, day out and not hurt anybody (except a few egos).

[noblues]

I gues you are having a very bad day at heavy weights and at a hot/high airfield to loose 2 on one side ...

The only likely scenario is a multiple bird strike, or a uncontained blade that took out another eng.

Accelerating to clean up is logical ... I've been on type for 1.5yrs and their is no published procedure for this scenario, to the point of 'it wont happen' - never practiced it in the sim. (If if it can - it will is my theroy).

I've been faced with gear stuck down and an engine out (dispatch with 1 or 4 demand pump u/s and loose 1 or 4 eng respectively), the gear overide switch locks the gear lever ...
Heard it said gear stuck down is as good as loosing another eng. in terms of climb performance without the directional issues.

[john_tullamarine]

Simultaneous loss of two on one side during the early takeoff just about guarantees an up close and personal encounter with Mother Earth ...

[Loose rivets]

But with a lot of luck.........

There was the Viscount training incident, where, having failed two on the same side, the crew ended the sortie by exiting--amid huge flames--out of a hole in the fuselage. Then they found themselves hopping from one 40gal drum to another. Yep, came to rest in the fuel farm.

Early 60s at SEN I think.

[Intruder]

The only similar exercise I've done was in the sim when the instructor forgot to clear the previous #1 engine failure at V1 before he set up the #4 failure. I lost both at V1, but was able to get the airplane in the air (max TO weight). The scenario was LAX, standard day, taking off to the west over the water, so terrain was not a factor. The airplane accelerated VERY slowly, but it did fly.

Since an outboard engine failure at V1 in a 747 requires near full rudder to keep the airplane tracking straight, I suspect loss of both engines on the same side at V1 would not be controllable. VMCG is around 121-127 KIAS for the outboard engine failure, but I don't think the numbers are published for the 2-engine failure. Also, I doubt the Classic would have enough power to get airborne except at lighter weights (320T or less?).

For the 2-engine landing scenario, commit point is at landing gear down. While it is possible to get the gear up and climb away again, there is little margin for error.

[Willit Run]

I fly classics. Just got out of the sim. Our sim is ex-KLM and has 'R' powered Pratts on it; thats 54,000 lbs. of thrust each. During a max weight takeoff;833,000 lbs, full rudder trim is required + a bit of aileron to keep it straight.
I have not tried it on two engines/one side, but my firm belief is it would not be controllable. We do a two engine approach and landing scenario at max landing weight 630,000 lbs. and it is quite controllable. VMCG on an 'R' powered plane is 131 knots. V2 is 188 knots and it takes all 11,000 feet to get it off the ground on 3 engines with one quitting at V1. V1 was 157 knots.

A few years ago, our company lost 2 engines on a DC-8-61 coming out of France somewhere, and it was not pretty. However, there were two very experienced flight engineers on board and a rather crafty captain, and they survived.

[West Coast]

I remember reading the narrative of the UAL 747 classic that lost two on the same side, along with a cargo door out of HNL in 1989. Pretty incredible read. Cant seem to find it anymore online.

[barit1]

The 747 was in mid-climb - see the NTSB brief (http://www.ntsb.gov/ntsb/brief.asp?ev_id=20001213X27705&key=1)

[noblues]

An explosive decompression is bad enough -- AND loosing no. 3 and 4 engines at the same time ... the crew did very well to contain the situation ....

[G--SPOT]

It has happened to a BA classic coming out of somewhere in Canada (Montreal or Toronto I think). It hit a flock of geese and had to return. Can't remember the exact story, but was an interesting read, and I'm sure someone can give more detail.

The 400 has a lot more performance than the classic, but out of Canada it would have been at a relatively low weight so would have performed in a similar way to a 400 at 'normal' weights.

[galaxy flyer]

The notoriously underpowered C-5 has had two "one and a half" (one was gone and one was vibrating so badly power had to be reduced) failures after birdstrikes. Both got around OK, but rather hairly. At Dover in the early 80s, wx about 200-1/2, one took multiple geese, the A/C got it around barely a few hundred feet in the air, established a short PAR landed after being told, "too low for safe approach". One of two peacetime DFCs, I personally know.

The other at Westover, same scenario, double semi-failure during flap retraction, managed to land. Bad news was the nearly 200,000 pounds of munitions on-board. Got to close the dumps nearby the field. Good drivers, both, lucky too.

GF

[Joetom]

Some airlines train for two out on one side as you hear the V1 call-out, attention to detail is reqd if you want to tell the story.
.
Talking about Viscount, think I remember reading about one outer Heathrow in the 60s, engine failed during climb and crew shut down good engine on same wing by mistake(think locations of throttles and fuel cocks helped the mistake) crew avoided hi ground over Windsor and were able to tell the story to the CAA, well done to that crew!!!:)

[ABX]

Does anyone train for three engines out in the 747?

EFATO or cruise?

I mean surely it is a possibility, however remote?

What is the response to three out?

I'm guessing from the posts above that it wont climb ... right?

Cheers,

ABX

[Taildragger67]

OK, it wasn't a 747,

... but wasn't the RAAF 707 which went into the sea off East Sale (Victoria, Australia) in 1991 practising asymmetric climb-outs with two out? Result = :sad:

[Albert Driver]

Those with long memories will recall a series of double engine failures on takeoff on BOACs VC10s in the 1970s, caused by debris from one failing engine entering the (very) adjacent one. All survived. Various flight profiles were tried out and later taught in the simulator to deal with such an event. Acceleration and clean-up at the expense of all else is the key. It says much for the Captain of the first one that he worked it all out as he went along.

[411A]

The RAAF B707 had its rudder power switched off at a low airspeed by a rather inexperienced check pilot, to see....what would happen.
He found out.
Ditto with TWA at Atlantic City years ago, although in this case it was an FAA inspector who did the dirty deed.
All died of course, and the FAA bought TWA a brand new 707 as a result.
At SV in the early eighties, with a B747SP at JED bound for JFK an uncontained number two failure at 800agl resulted in bits entering number one, and illuminating the number one fire handle, no bell as I recall.
The Flight Engineer looked up, notices an illuminated fire handle and...yep, pulls it without saying anything to anyone, now on two engines only at about 1200msl.
The Captain, very experienced on type, descended slightly to accelerate, retracted flaps on schedule (minus 10 knots, give or take) started fuel dumping straightaway, and returned....just.
The Flight Engineer was terminated the next day.
I'm surprised he wasn't thrown in jail for such a stupid stunt.
And no, you won't find reference to this incident anywhere, it was hushed up at the time...very tightly.

[flyandbefree]

I know of a case where the crew tried a 3-out scenario in the sim during an approach. Third one was cut just after glideslope interception. With the 4th one firewalled the aircraft made the runway. There was no change in configuration. With 3 out there is no way the aircraft would climb. 3 out is science fiction I think.

[ABX]

Thanks flyandbefree,

I thought 3 out was pretty rare, but someone must thought about it at some stage and run it through the sim.

In the example you mentioned, do you recall which engine they had left?

Also, while I am at it, what is the convention for engine numbers? (In general and on the 747?) Which is no. 1 and 2, 3 and 4? I've never known for sure. Does it change for different a/c ie., 747 and 777 or is it the same convention all over?

Cheers and thanks for the answers.

[TopBunk]

ABX - engine numbering convention:

Looking forwards, the numbering starts with #1 furthest left then work left to right.

[Taildragger67]

411A,

Ditto with TWA at Atlantic City years ago, although in this case it was an FAA inspector who did the dirty deed.
All died of course, and the FAA bought TWA a brand new 707 as a result.

You're better informed than I, but I thought that one was the FAA guy suddenly cutting fuel to two, rather than rudder power?

Or am I confusing that incident with another suddenly thrown in by an inspector? :eek:

Thanks

[ABX]

Thanks Top Bunk.:ok:

[wileydog3]

Did you know the left seater, Ralph? Great fellow and some say it was this event that propelled him to his first star.

[blue_side_up]

The company I'm with keep a few crews qualified to fly the 747 Classic on 3-eng ferry flights. Fortunate enough to have done the training myself, and we do practice in the sim, including lots of handling of a second engine failure (same side usually) after V1. At weights over approx 280,000KG (ISA cond), you're not climbing much (+100'/min if you're lucky), and if the gear were to remain extended for any reason then you're definitely going downhill (now you're having a bad day). At about 220,000KG you get about +200 to +300'/min climb. The company imposes a takeoff weight limit of 250,000KG, and daylight, as a safety margin. Full rudder is req'd at full thrust, with about 3-5 deg bank into the good engines, if 2 eng failed on one side. Definitely directionally controllable though.
It's a hell of a fun excercise in the sim, not sure about in real life!
As a note, during the initial type rating (JAR, not sure about others), we have to do a 2 eng (same side) ILS approach and go-around (no go around after gear down, although it is possible if you have some height and accelerate down the g/s), and a 2 eng landing.

[ABX]

G'day blue_side_up,

What are the reasons that you would normally do a three engine ferry?

I ask this as it makes me wonder if it might be best to t/o using all four and then shut one down after climb.

Although I guess this is impossible if the reason for the ferry is so that a defective engine can be repaired!

Sounds like you've had some interesting training, did you ever discuss three out?

Cheers

[mustafagander]

Taily & 411A,
WRT the RAAF B707 lost in training near East Sale - the major problem was that the rudder hyd power was not "switched off", rather the hydraulics were depressurised. It does have the same NETT effect - no hydraulics to the rudder - but during the hyd pressure decay period enough pressure remains to keep the rudder torque tube locked for quite a length of time. This means no manual rudder and no power rudder.
Just briefly, the B707 - 338 rudder in manual mode used a control tab to "fly" the rudder. This tab was actuated (simplification!!) by a long (30 ft or so) torque tube up the fin. In power mode, this torque tube was locked so that the tab acted in anti balance. When the rudder hydraulics were "switched off" by the control switch, the rudder hydraulic system was ported to return, immediately allowing the torque tube to unlock and provide manual (tab) control albeit at a considerably higher Vmca. When the hydraulics were depressurised with the rudder power switched on, either by accident or design, the torque tube would not unlock until comparatively low pressure leaving a "no mans land" with little or no rudder control possible. It was a regular base training demo.

[RatherBeFlying]

It surprises me to hear of shutting down a good engine on the Viscount.

Autofeather usually takes over and the feather button of the affected engine glows prominently and makes it pretty obvious which levers to pull back.

Failing autofeather, the JPTs (nowadays EGT) and fire lights lined up nicely with the levers and shutoffs.

For a British late '40s design, the cockpit layout was brilliant.

[411A]

I was informed by an oldtimer at TWA, taildragger67, that it was the rudder power switch to OFF.
Now, this really is not very bright.
B707-320B series:
Vmca 120 knots/boost ON, 180 knots/boost OFF.
2 engines out, same side, Vmca 170 knots/boost ON, 235 knots/boost OFF.
3 engine inop approach (IE: thrust to idle), outboard engine at MCT, flaps 14, would keep you on the glidepath, at light weights.
This was done on my type rating ride, many years ago, in the airplane as a demonstration maneuver only.
Standard PanAmerican training.

[Flight Safety]

I recall a 2 engine out TO for an SAA 747SP (possibly could have been a 741 or 742) from SID sometime in the summer of '81. The AC returned safely. Over 25 years ago, I tried a 3 engine out (actually 3 engines at idle) in a 741 sim in the cruise (I forget the altitude), with an outboard engine still operating. With the one engine at 100%, the AC could barely manage a 75 ft/min decent rate as best climb rate. We thought this gave us enough time to actually get somewhere to land.

[blue_side_up]

ABX,

Ya, your guess was correct - we only do the 3-engine ferry if the offending engine cannot be repaired on site. It's meant to be a 'last resort' when normal options are not available.

If you're interested, the u/s engine has to be secured, blanked off, or have the N1 fan blades removed (depending on engine type, damage, etc). If it's an outboard engine that's u/s, the 2 inboard eng's (#2 + #3) are brought up to t/o thrust before brake release, then the operable outboard engine slowly brought up to t/o thrust as speed increases, to be at full thrust by about 100kts. This is because the nose wheel cannot hold the aircraft straight at the (relatively) light weights we use for 3-eng ferry, without some aerodynamic assistance from the rudder. It's very interesting training, and gives us a bit of insight into non-standard ops and the capabilities of the aircraft. Vmca is around 153kts (varies slightly dependant on engines), and we have been told that there's a 'grey area' in this type of takeoff - a time frame between V1 and Vr where you may not get airborne, or be able to stop on the runway! Not a very comforting thought in a big jet (or any aircraft for that matter!).

We have talked about flying on 1-engine. I know one guy who tried it on approach in the sim. But of course you're heading downhill on the ILS at that point anyway. Don't think you'd get much of a climb, maybe if you're very light, ie: 200,000kg or less? Will have to try it next time there's a few spare minutes after a check ride.

[ABX]

Hey blue_side_up,

Yes mate, I am interested and I appreciate your informative post.


... the u/s engine has to be secured, blanked off, or have the N1 fan blades removed (depending on engine type, damage, etc).


I would presume then that this is due to the windmilling engine wreaking further damage to itself in transit? Or is it to stop it overspeeding?

While the three out scenario may never happen it sure is interesting speculation.

411A & Flight Safety obviously have some interesting experience with 3 out. nice posts guys, thanks.

Flight Safety, if it manages 75fpm descent 3 out and firewalled, I guess it is a brick with all out?

blue, if you ever do find a few minutes spare in the sim and try the 3 out, I'd love to read about it.

Cheers all.

[747dieseldude]

2 Engines out is a tricky one.
In LY, apart from the unfortunate accident at AMS, they had a classic hit a flock of bird on takeoff from TLV, hot day, MTOW. Lost two engines on one side, however not at V1, but very soon after rotation.
They managed to climb to a few hundred feet, dump some fuel and return safely.

My personal experience with 2 (or less) engines on a classic involves sim only. We often practice 2 engine failures on takeoff, most common scenario is a flock of birds. Usually the first engine (outboard) fails at V1 and the second engine just after gear up.
In my view, 2nd engine failure before gear up is remotely survivable at high weight/temperature.
But, once you get the gear up, even with flaps 20, you can make it.
With flaps 10, usualy it's not a problem, you climb slowly or maintain altitude to allow flap retraction to flaps 1 (as mentioned before me, you would want to keep flaps 1 to have the outboard ailerons functioning).
With flaps 20 (which is usually the case in heavy t/os) you will sink on 2 engines. The trick is to slowly move the flaps to 10, in small increments, while maintaing altitude or even while losing a few feet, as the speed builds up slowly.
In the sim it works fine.

3 Engines out.....
I did my initial 747 rating with Capt. Jim Smille, may he rest in peace. He is truly one of the greatest.
After the first few days on the training, he let us get to know the edges of the envelope - from the outside.
Among other stuff, we did 3 engine failure, at 5000' level, at 250 tons (not too heavy). Now I can't remember if the one we had left was inboard or outboard, but we did the approach, go-around from 500' (yes, after gear down) and back to landing again.
Same Capt. Smille also gave us a gimly - approach from altitude with no engines, worked fine as well.

My guess is that the -400 will perform even better.

Kids, do not try this at home!

[ruddman]

The trick is to slowly move the flaps to 10, in small increments,


Silly question, but how do you manage that? There a flap position between flap 20 and 10? Or you can actually slow the flap retraction down in smaller increments?

[747dieseldude]

ruddman,

you can do that by slowly moving the lever and holding it in an intermediate position.

[mustafagander]

I am reliably informed that this technique does not work on the B744.

I can vouch for the fact that it does work on the B742!!