Airliner acrobatics

June 15, 2005 4:36 AM

Is it possible for a standard jet airliner, say a 737, to succesfully perform an aerial barrel roll or a vertical loop? It's just something I've always wondered.
posted by punkfloyd to grab bag (18 comments total)

Well let's ask Google, shall we? airliner barrel roll and, lookee there, the first half-dozen hits tell you exactly what you want to know.
posted by five fresh fish at 4:47 AM on June 15, 2005

Cecil Adams says no to a loop, and here's video of a 707 doing a roll.
posted by Marky at 4:48 AM on June 15, 2005

Awesome Marky. Thanks!
posted by punkfloyd at 4:57 AM on June 15, 2005

Depends. Almost certainly yes, but....

1) If the plane is light (no pax, low on fuel, etc.), it's likely to survive, but with rather severe structural damage.

2) If the plane is heavy (say, halfway between unloaded and Max Gross Takeoff Weight), then it will probably break. This would be bad.

3) If the plane is very heavy (near or at MGTOW), it won't even get close to looping or rolling.

There was a recorded instance of a 707 doing a full wingover roll. However, this was the result of a storm updraft kicking the wings over 120 degrees, and the pilots decided to keep with the roll, rather than try to reverse. There are arguments to this day if that, as a general class, was the right call. In that specific case, they landed safely.

So, it really all hinges on your definition of "successfully perform". If it means simple "able to do.", sure -- if the planes empty, and the pilots are careful, easy.

If you want to have a useful airliner afterwards, however...

Also: Which 737? The shorter ones (737-100 or 737-600) would be much easier to deal with than the longest one (the 737-900, which is almost the size of a 757.)

On Preview: Note that the 747 and DC-10 are very different planes, with two big differences: they're much largers, and they have more engines. For looping, this is a problem -- a 737 must be able to fly on one engine, so, by design, they are overpowered in normal flight. A DC-10 or 747 are also designed to fly with an engine out, but with three and four engines, that means each engine only has to be 50% or 33% more powerful to carry the load of the missing engine.

So, a 737 has much more reserve power, which is everything in doing loops. (The 757, using the same engines as the 747, is almost dangerously overpowered at time.)

The plane would still be in real trouble afterwards, though -- airlines just aren't built to do such things, and the cost of the structure needed is the weight of the structure -- which mean less fuel (thus, less range) and less passengers (thus, less revenue.)
posted by eriko at 4:59 AM on June 15, 2005

Consider the hijcaking of Fedex flight 705.

To throw the hijacker off his feet and try and stop his attack, the pilot (badly injured) put the aircraft through extreme maneuvers.

The aircraft was rolled over onto it's back, and the pilot considered an outside loop to throw the attacker against the ceiling. He decided against it as he thought the cargo crates might plunge through the top of the aircraft.

Instead he pulled back on the yoke and essentialy did a reverse immelman, reaching a speed of mach .86.

I've heard it said that any aircraft can do a barrel roll as long as it has the speed to get through the maneuver. The pilot who barrel rolled the 707 said that he knew it was safe as it was a 1 g maneuver.

A loop, on the other hand, would be another issue.
posted by tomble at 5:48 AM on June 15, 2005

Off topic, but I was at Farnborough last year for the Air Show and was amazed at the maneouvers the Commercial airlines were doing. I've been to other airshows but they never had commercial aircraft flying displays. But the South African Airlines 747 doing extreme stuff was unbelievable. Also an Airbus creeping along the length of the runway (at about 100-200 feet) with a nose-up attitude of at least 10-15% was amazing. You could hear the pilot goosing the engines to maintain just enough forward velocity.
posted by smcniven at 7:16 AM on June 15, 2005

I believe that modern airliners are designed to withstand the stress of going inverted whilst diving in the event of a decompression event. This is to allow a severe nose-down attitude, in order to lose height rapidly, without subjecting passengers to negative-G. It's also intended to push passengers into their seats, rather than out of them, and to prevent loose objects flying aroung the cabin. I'm not sure if it's ever been put into practice in the event of a real emergency, or if it's even part of simulator training. The airframe would almost certainly be toast, afterwards, though. In general, once an airframe has been stressed beyond its max-recommended, it's decommissioned.

As eriko points out, it takes a lot of energy to roll a plane, because you are, in effect, moving two massive surfaces (the wings) against the atmosphere. In an emergency dive the roll energy would come from trading altitude.

smcniven, it's not the pilot goosing the engines, it's the fly-by-wire system. It's a heck of a trick, and only really made possible by computerisation of the cockpit. At the 1988 Paris Air Show there was a demonstration of this that went horribly wrong.
posted by veedubya at 8:15 AM on June 15, 2005

I routinely asked pilots this question prior to 9/11, and that was back when I was flying nearly every week for work in a variety of aircraft. Every single pilot said that it was possible, and one pilot (when I was flying to Key West in a small jet) said that he heard of guys doing "funny sh*t" with smaller, empty aircraft.

One day I found myself sitting next to a pilot in coach and we got to talking. Turns out he was also a helicopter pilot and instructure, a vet who'd flown hundreds of hours.

After discussing barrel rolls extensively (he said not only were they possible in virtually any commercial airline, he said that they performed rolls at basically all levels of testing) he said he knew a pilot who pulled a barrel roll off during an overnight flight in the late 70s and was fired.

The basis for this anecdote is that a perfectly performed barrel roll on an upward trajectory would be unnoticeable to passengers if it were nighttime and the shades were drawn. If g's are held steady and the roll is executed perfectly, a glass of wine wouldn't spill. Allegedly this pilot pulled it off and no one noticed but his co-pilot or navigator or whatever turned him in. He was fired immediately and allegedly flies commercial in Eastern Europe someplace. Sounds like the stuff of urban legend, but this guy sitting next to me swore to it and he wasn't lying about his background or the nature of his job (he had his briefcase with him and was showing me documents related to flying and his helicopter business.

After 9/11, I decided it wasn't a good idea to ask about evasive manuvers. YMMV.

Finally, I also asked about a loop and every single pilot I ever asked said that it was impossible.
posted by thewiseacre at 8:20 AM on June 15, 2005

it's the fly-by-wire system

Minor threadjack - didn't fly-by-wire used to mean manual control? (as in, the controls were mechanically connected to the flaps etc, operating them by direct cable connection).

I'm kind of interested in whether this term has evolved to mean the exact opposite of what it originally meant, or whether I'm confusing it :)
posted by -harlequin- at 8:56 AM on June 15, 2005

harelquin, fly-by-wire systems (in theory) don't have a direct hydraulic connection between the cockpit controls and the aircraft's control systems. Command input is detected by computer sensors, interpreted by software, and then relayed by wire to the remote hydraulic systems located in the vicinity of the control surfaces. This has two advantages: firstly, the software can judge whether the command input is safe in the prevailing conditions, and modify the response accordingly; secondly, wire is lighter, less bulky, and isn't susceptible to leaks.

In practice, there is, at least in commercial aircraft, a redundant backup hydraulic system that will provide some limited control in the event of a catastrophic computer failure.

In a modern airliner, it would be impossible to move the control surfaces using just wires. Power-assisted hydraulics are required to deliver the necessary force.

thewiseacre, that night time roll makes sense, but it'd be a b*st*rd to pull off. From the ground, I guess that it would look like a lazy roll around an imaginary point some distance above the aircraft's centre-line, the distance could be determined by the dihedral.
posted by veedubya at 9:47 AM on June 15, 2005

veedubya writes "smcniven, it's not the pilot goosing the engines, it's the fly-by-wire system. It's a heck of a trick, and only really made possible by computerisation of the cockpit. At the 1988 Paris Air Show there was a demonstration of this that went horribly wrong."

Gotcha. That makes a lot of sense thinking back about it. Now, back in the late 80's I was at an airshow in Moose Jaw where a CF-18 pilot flew in a large circle with the nose pointed virtually straight up. Same principle or more pilot input?
posted by smcniven at 9:49 AM on June 15, 2005

didn't fly-by-wire used to mean manual control?

Googling briefly, it looks like fly-by-wire originated in NASA test programs (of electronic control) in the 70s. There's probably more in this NASA document about the test program. (thru archive.org)
posted by ROU_Xenophobe at 9:58 AM on June 15, 2005

Further thread-jacking:

That 1988 Paris Air Show incident ... it says that 4 people of the 136 in the plane were killed. Why were there 136 people aboard a plane that was performing air show maneuvers? WTF?
posted by redteam at 11:09 AM on June 15, 2005

I once ran some tests on an instrument I was developing on a DC-8 that was undergoing a re-certification test flight. The test pilot, among other things; pulled +4G, -2G, and rolled the a/c so violently from side to side up to 90 degrees that looking back from the cockpit to the tail (inside) one could see the tail section lagging the rest of the aircraft by 10-12 degrees!! During some up and down oscillations, the wings appeared to be flapping like a sea-gull's. I asked the pilot about this and he said that the aircraft was built to be flexible, which made it much stronger. The wing tips were designed to displace up to 8 ft during turbulence. Modern large airliners may not be quite so flexible, however.

I believe that one of the main reasons why sustained negative G does not go so well is that lubricants and other liquids that the aircraft uses rise in their tanks and are hard to scavenge. That means that the aircraft could only keep going for 10-20 seconds during negative G before serious damage or flame-out would occur. (My information is a bit dated, however, and these problems may have been addressed over the years.)
posted by RMALCOLM at 1:03 PM on June 15, 2005

According to the Wings channel (OK, OK), the test pilot on the 707's maiden flight did a barrel roll. The interviews they taped (including the pilot's son, who saw it) all said it was deliberate, and way cool.
posted by Kirth Gerson at 1:51 PM on June 15, 2005

redteam, there shouldn't have been any passengers aboard. Aside from the fact that they were performing what was, at the time, an experimental manouvre, it was that aircraft's first public flight. Simple common sense implies that there shouldn't have been any passengers. There's also questions about the kind of fuel they were using (I love the idea of airplanes being loaded up with 'drag-fuel'), and the confidence that the pilot's had in the flight control software.

smcniven, I'm not up on the F-18, but what you describe sounds a lot like a demonstration of vectored thrust. That's where the thrust of the engines can be directed in a direction other than the aircraft's flight. It enables the aircraft to be 'danced' in flight.
posted by veedubya at 2:07 PM on June 15, 2005

one other thing: the guy I was referring to also told me that in theory (and I guess in practice) that you wouldn't be doing a true barrel roll anyway: it'd be (from a layman's perspective, which is how he described it to me) technically a corkscrew. In other words, the aircraft wouldn't be rotating 360 degrees in the same geometric plane. And he did reiterate that keeping the G force constant (not only while pulling upward but also in the rotation) would be tricky; I maintain that it would take one hell of a jock to keep the cabin totally blind to what was going on.

and anyway, it's a pretty fun thing to ask the pilots about when you're getting deplaning. Most of them get a big smile on their face.
posted by thewiseacre at 3:25 PM on June 15, 2005

More information on the French airshow crash here.

Interestingly, it was discovered that the black box was switched after the crash!
posted by tomble at 1:37 AM on June 16, 2005