This reminds me of a part of Stick and Rudder where a pilot asks with a failed engine “what’s the best glide speed of this airplane?”. And engineer says “It’s simple! All you have to do is solve this equation!” and then displays what looks like an orgy by 142 spiders. Another pilot says “It’s the speed that gives you the lowest rate of descent.”
There is a huge difference between simulation with no risk of severe injury or death, with time to think, consider, and ponder a problem and facing a real life-or-death emergency. Yes, with what we know now, this may be survivable. But when the engine drops off, you’re barely off the ground, the plane is shuddering on the edge of a stall, and you’re rolling over you just don’t have the time to work the problem. You have to figure out what’s wrong and fix it now, and your first guess is the only one that counts. The engineer’s solution will give a very precise answer - but you don’t have enough time to get the answer. The pilot’s solution may not be as precise, but it’s much more useful in a genuine Serious Situationsub[/sub]
And I assure you that dropping the weight of an engine WILL affect the “controllability” of an airplane. Maybe not to the point where all hope is lost (which is what I suspect your captain meant), but you’re gonna notice something has changed drastically. The airplane will lose much of it’s stability and want to roll away from the missing engine - as, in fact, happened in 1979 over Chicago.
A Canadian 747 ran out of gas at, if I recall correctly, 30,000 feet. Glided to a safe landing with no injuries and only minor damage to the foremost landing gear. No engines jettisoned, airframe remained entirely intact.
[sub](By the way - no one had determined a best glide speed for a 747 prior to this embarassing incident, so the pilots had to figure it out in the air. No use of equations to solve the problem - they experimented until they got the lowest rate of descent. Of course, at 30,000 feet they had some time in which to do this)[/sub]
If the engines were jettisoned, meaning no damage to the remaining airframe? In theory the reduced rate might allow it to remain aloft longer, but then your best glide speed and other aerodynamic factors may be significantly altered - or maybe not. You’d be adding a whole 'nother layer of guesswork here.
If the engines rip off or tear loose - gotta wonder what that does structurally to the wings. As a general rule, wanton damage is a bad thing. Frayed and broken parts hanging in the airstream may give you as much, if not more, drag than intact-but-dead engines.
Look, nobody’s claiming that engine separation is a good thing, or that the pilot has any kind of option to “jettison” a bad engine, or that this is even a “good” thing to have happen in an emergency. Under extreme conditions, however, the engines are designed to shear off rather than doing even more damage to the airframe.
My cite? The NTSB, in various statements to the media about this and other crashes. I don’t have time to find convenient webpages for y’all, because I post from work and can’t spend that much time on the 'net not being productive.
There is no “may” to the aircraft being safely flyable after the engine fell off. What the Captain meant, and the NTSB report agrees, was that the aircraft was perfectly flyable as long as the airplane was flown at 159kts or greater. The pilots were following their training, but the training was wrong and they never had the chance to figure that out. A hard won lesson that made the rest of us safer.
I’ll assure you right back. Losing the engine did not create the loss of control. Your own experience as a passenger should bear this out. There is a moment where the aircraft jerks into the dead engine but there is plenty of control authority to correct that with the rudder. I don’t have a weight and balance on a DC-10 but my guess is that the loss of the entire engine assby did not significantly affect (effect?) the center-of-gravity so there should not have been an abrupt or severe nose-up or nose-down moment. Any roll tendency was well within the control authority of the ailerons. From the NTSB report: “Postaccident simulator tests showed that, if the airspeed had been maintained, control could have been retained regardless of the multiple failures of the slat control, or loss of the engine and numbers one and three hydraulic systems.” In fact, the DC-10 initially continued a normal climb-out with no significant change in attitude. It was only when the aircraft was slowed to the recommended airspeed that the aircraft stalled and there was not enough altitude to recover.
Hi all. Thanks for all your replies. You can always count on an aviation question to bring out the fly-boys (and girls). Always plenty of long well-cited answers.
And a debate or two.
Johnny, to answer you, my Q was about this crash in light of what is hypothetically possible. Like Keeve pointed out early in the thread, all the news animations had the plane nose-diving as soon as the VTF snapped; that seemed presumptuous to me.
As near as I can tell, the only other instance of a commercial jet gliding to a landing was also a Canadian (Air Transat) jet that ran out of fuel over the Atlantic and glided to a relatively safe landing in the Azores. More Info
Here is a link to the FAA airworthiness regulation that says, in part:
"(b) For turbine engine installations, the engine mounts and supporting structure must be designed to withstand each of the following:
(1) A limit engine torque load imposed by sudden engine stoppage due to malfunction or structural failure (such as compressor jamming).
(2) A limit engine torque load imposed by the maximum acceleration of the engine."
I don’t know how you get that engines are designed to come off in these kinds of conditions. They are supposed to stay on.
Now, the engine mounts may actually be the weakest link, and therefore would be the first to go - rather than, say, a wing spar buckling - in some other catastrophe. But it isn’t designed specifically to be this way.
kellymccauley, those regs are in reference to an engine failure, not in regard to external forces, such as extreme aerodynamic force, centrifigul loads (as might be caused by uncontrolled flight regimes), or by mechanical forces (striking ground obstructions).
Tranquilis, my cite was directly aimed at situations described earlier by Robot Arm and Sam Stone that addressed engine falure.
And if there’s an airworthiness regulation that addresses structural integrity during uncontrolled flight (how much load factor is that?) or running into obstructions on the ground with your engines, I must have missed it.
I rather doubt that one exists, either, but there are general industry standards, and one basic premise is that if parts must come off, it’s better that they come off in a controlled manner. Engine struts are designed to break-away cleanly rather than tear, even though it takes pretty extreme forces to cause that.
Tranquilis, I think we may be niggling over what the words “designed to…” mean. Once parts start flying off the aircraft, I don’t think of that as a normal design condition. But, given that the parts are comming off, I can see your point.
In my job (military helicopters) I’ve unfortunately been a party to conversations like this:
Contractor: “We’ve designed this part to be fail-safe. If there’s a failure here, then this other part will take up the load. It will be good for at least a 100 hours, plenty of time to catch the failure during inspection. And there will be unusual sounds/vibrations/temperatures that would alert the crew in time to set it down.”
Army: “Okay then, we’d like you to take one of these parts with an induced failure, and have your test pilots fly it for 5 hours or so, to show that it behaves like you say.”
Contractor: “Oh, no, that’s much to dangerous.”
Army: “Tell us again why you think it’s fail-safe?”
Contractor: “Well, it’s designed to take the load this way…”
All I can say about the “detachable engines” is - well, yeah, if you’re skidding across the landscape on your belly scooping up trees, bushes, sod, concrete, houses, and other small obstructions it makes perfect sense to design the engines to “break away” under those circumstances.
But that’s a lot different than dropping off an engine because the engine stopped working.
Airliners can withstand some pretty impressive g forces and structural stress. Parts do not normally come off unless the airplane hits a solid object. In-flight break up due solely to aerodynamic forces is pretty damn rare, and usually connected to spins or steep spirals, not normal take-off climbs. Even encounters with wake turbulence do not normally result in break-ups, particularly not break-ups of airliners. It’s not impossible, but it is extremely unlikely.
Originally posted by Sam Stone:
OK, this is the paragraph I have a problem with. An engine failure by itself is NOT “extreme” or “catastrophic”. It’s serious, no one disputes that, and it’s considered an emergency, but it’s no reason to rip an engine off a wing. Even if the turbine fan was at a complete stop the increased drag is not enough to bring down the plane.
Originally posted by Boxcar:
[sub]>sigh<[/sub]
OK, let’s take this one step at a time.
We all agree that, on a multiengine plane, when an engine stops running (for whatever reason) that the plane turns (yaws) towards the non-functioning engine, correct? [sub](Except for the Cessna Skymaster… oh bother, let’s stick to normal and typical configurations, shall we? Thanks)[/sub]
OK, next item: center-of-gravity does not apply solely to the nose-up nose-down backward-forward aspect of the plane’s flight. Weight and balance can affect all three axis (axii? axises?) of an airplane. Normally, it’s of concern with the aft-to-fore axis (the “pitch” axis) because that’s the one most likely affected by loading of cargo and passengers. However, the roll axis also is affected by weight and balance.
As an example: once upon a time I was flying an extended trip in a Cessna with a companion. This particular model has two gas tanks - one in each wing - and was configured to feed fuel to the engine from both tanks simultaneously. Well, between Fort Wayne, IN and Knox County, IN the right gas gauge fell to zero. Not a good sign. So we set about diagnosing the problem. My buddy maintained it was probably a busted gas gauge (hey, it’s a 30 year old plane, things wear out). I said, maybe that tank really is empty. How to find out? Let go of the yoke (that’s the “steering wheel”). We started a slow roll to the left. Why? Because the right wing was about 84 lbs lighter than the left wing. When we got on the ground we found the right tank was dry. So… yes, an airplane does have the capacity to compensate for left-right weight differences - any plane with the fuel tanks in the wings HAS to have this - but that doesn’t mean these weight differences are insignificant and/or or don’t have an effect in flight. In fact, part of fuel managment is not letting the plane get too far out of balance along the roll axis. Although this model of airplane clearly is capable of flying with one tank dry and the other at full, the out-of-balance aspect would complicate matters if, say, you have a very brisk crosswind on take off or landing, to the point where a combination of the imbalance AND the crosswind might be enough to result in an accident when either condition by itself is not enough to do so.
I am completely sure that, if an entire engine assembly drops off any aircraft it will most definitely have a profound and nasty effect on the weight-and-balance of that aircraft. Why? Because the wing-minus-an-engine is significantly lighter than the wing-with-an-engine, and as these engines are mounted at some distance from the center of gravity there is also a lever effect to account for. It is entirely possible that the ailerons on a DC-10 can compensate for this – but that doesn’t mean the effects aren’t serious. Even if the pilot compensates to the point that a passenger or outside observer thinks normal flight has resumed that still doesn’t mean the problem has gone away or become less serious - it just means you have a competant pilot at the controls and a chance to survive the experience.
There will not be a nose-up or nose-down action of the airplane because that is not the axis affected - it is the roll axis that will experience unexpected movement, not the pitch axis. And that is why the DC-10 in question rolled to one side. The closer you get to stall speed the less effective the controls and the more any deviation from perfect balance will affect the path of the aircraft.
So - by itself losing an engine does not doom the airplane. Even a incipient stall by itself does not doom the airplane. Malfunction of wing parts like slots and flaps by itself does not doom the airplane. A combination of the above, however, can result in a very sudden and very unpleasent end to the flight. Yes, an increase of speed likely would have prevented the immediate cause of the accident by avoiding a stall but that would not garauntee a safe end to the flight. The pilots would still be dealing with an out-of-balance aircraft minus an engine. A lot could still go wrong.
Broomstick: I’m not talking about your everyday engine failure. When a jet engine fails, typically the fan continues to rotate, and air still passes through it. On the other hand, a jet engine that is seized and stopped doesn’t let air flow through it, and essentially acts like a big, flat plate. A GE90 engine is about 13 FEET in diameter. Can you imagine the force on a 13 ft. dinner plate being held into a 400 mph wind?
For comparison, a Boeing 757 has an equivalent flat-plate drag of about 32 sq ft. That means a stopped GE90 engine has more drag than a 757!
BTW, you’re probably thinking back to your pilot training and going, “But hey, a windmiling prop has more drag than a stopped prop!” That’s true, but that’s because a stopped prop only presents a small fraction of the disk size to the relative wind than a windmilling prop represents. The two situations are almost completely reversed.
Now, I can’t give you a cite that says specifically that Engine Pod fuse pins are specifically designed for this reason (such a failure mode may be very rare, and stopping the fan quickly would probably tear the engine off anyway), but I would be willing to put some serious money down that if you stopped a high-bypass engine and prevented the fan from windmilling you’d be in a world of trouble unless the engine departed the airplane.
Airliner Glide Capability: This surprises a lot of people, but large jets are excellent gliders. A Boeing 767 has a glide ratio of about 19:1, meaning it will lose 1 foot of altitude for every 19 feet it travels. That is almost as good as some training gliders, and much better than the average Cessna, which is typically around 12:1 or so. A jet typically cruises about 7-8 miles up, which means it can glide 130-150 miles. The ‘Gimli Glider’, if I recall correctly, was at 41,000 ft, and glided 86 miles to land at a closed airport that was being used for drag racing. It probably won the race, too.
Well, OK, if that’s your opinion, fine. And it’s my opinion that you’d have more trouble without the engine assembly than with in almost all cases. Unless an engineer with cites comes along we may have to leave it at that. I agree sudden engine stoppage is more likely to tear the thing apart than merely stop the fan blades, but I also think that an otherwise intact airliner would be able handle the additional drag.
If I can corner an airline pilot at the local airfield this weekend maybe I’ll ask him his opinion and/or training on this.
Uh… actually, no, I was not thinking that because I am very well aware that jets and props are two very different things. I try to make it clear that I’m a SEL pilot and where my knowledge ends and wild speculation begins.
If you’re talking about a SEL fixed-gear trainer Cessna, it’s actually around a 7:1 ratio. Ultralights, despite being derieved from hanggliders, are often around 5:1. A lot of this has to do with aerodynamic drag. For example, a plane with fixed landing gear (like a small Cessna) has a LOT more drag than one with retractable gear (like a big jet). The end result is that a huge construct weighing hundreds of tons glides better than a dab of aluminum wieghing less than a single ton.
As printed in the Wednesday, 14 Nov issue of the New York Daily News:
So, what does it take to shear the pins? Well, the water landing is fairly obvious. If you’ve got a large pair of water brakes (the engines) digging into the water below of the center of mass on a large object hitting the water at high speed, that object (the plane) will rotate ‘forwards’ and ‘down’, plunging the nose of the aircraft into the water, and causing the plane to flip over onto it’s back, that is, if the planes doesn’t start an end-over-end flip and break up. So, I think it’s pretty clear what the advantage of break-away pylons are, in this case.
What about a belly landing? Provided there are no large obstructions, it makes more sense to keep the engines. But, should the engine, which is one of the components most likely to encounter a serious obstruction, actually encounter serious resistance, it should come away, else there’s the risk of ripping the wing off, teaing open the wing fuel cells, or causing the aircraft to cartwheel. So, shear pins make good sense here, too.
How about catastrophic, uncontained engine failure? According to Investigator Leonard, that might, and usually does, cause engine separation. This kind of failure is quite rare, and even if it happens, the shear pins don’t always let go. <this answers the question I posed in my first post in this thread>
Now, here’s where it gets touchy: What about in-air departure? Commercial aircraft are designed to take forces of about 3G’s accleration without serious damage, in transient conditions. The best information that I have is that it takes about 10G’s worth of acceleration to shear the pins. I would speculate, although I can’t prove it, that the engine pylon was designed to lose the engine to at least partially unload the wing in extraordinary circumstances, and maybe stave-off collapse of the wing for a few more moments. I would further speculate that the violent maneuvers and gyrations of flight 587 in it’s last moments might have produced accelerations up to, or in excess of, 10G.
So, there y’all have it, the results of a couple of days intense digging on the subject. Fire away!
Broomstick: Why are you assuming that I have no education in this area, and am just guessing? For the record, there are not many aircraft that have a glide ratio as low as 7:1. The book I have for the Cessna 172XP lists it at 10:1, and the RG versions are around 12:1. Efficient light aircraft like Mooneys are somewhat higher, but the point I was making was that glide ratio of 19:1 is truly impressive in aircraft, and is generally exceeded only by aircraft that are intended to be gliders.
Parasitic drag from things like fixed landing gear certainly has an effect on glide ratio, but it’s not the predominant one, because best-glide speed is low enough that parasitic drag becomes a smaller factor. Much more important is induced drag, because at best L/D the aircraft is flying at a higher AOA where the induced drag is relatively high. That’s why gliders have long, thin wings. And that’s why big jets glide so well. They have high aspect ratio wings like gliders. They have such wings because they are more efficient at high altitude, and because the jet doesn’t need to be that manoeverable.
And while my opinion that a jet with a high-bypass fan wouldn’t be able to fly if the fan was seized is a ‘guess’, it’s an educated one. As I showed, a seized GE90 engine has as much drag as an entire 757. Putting that much drag outoard on the wing has got to ruin your whole day. But yes, it’s a guess, and I could be wrong.
Tranquilis: The G-loading requirements for big jets is determined by bending the wing up and down, not back and forth. So that’s not really relevant to the discussion, because the loads from a failed engine are not even remotely similar to the loads you are talking about. The load from a failed engine will be transmitted through the pod to the attach points on the wing, and then the surrounding structure. That’s where any failure would occur, and it likely wouldn’t tear the wing off. It would bend whatever it’s attached to, and if it finally broke off it would tear up the immediate structure around the attach fittings. That in turn would rip up fuel lines, control cables, electricals, etc. And it would be unpredictable in what would be damaged. That’s why they’d rather control it with shear pins.
But I’d guess that the biggest problem would be if the engine doesn’t come off at all. Like I said in an earlier message, if an engine digs in to to the ground (or water) ona forced landing, you’d much rather have the engine leave the airplane than have the airplane cartwheel around the engine.
