Upside-down plane crash

[snopes]

Comment: In the book, The Devine Conspiracy, Dallas Willard tells of a
pilot who was testing a plane and attempted a steep incline, but instead
flew the plane straight into the ground because she was unaware that she
had been flying upside down. Willard doesn't site his source for the
story and doing a internet search didn't reveal anything, but I'm
wondering if the story is legit.

[Singing in the Drizzle]

The only thing I can think of is a South American crash of passenger airplane (737 if I remember right). They were not following proper procedures by switching to another artificial horizon when the main one was doing funny stuff during a night flight. In try to keep the airplane level to the instrument that was not working then managed to do a prefect 1g roll and dive. The airplane nearing or exceeding the speed of sound broke apart before hitting the ground.

[Alchemy]

Spatial disorientation followed by impact into terrain is a not uncommon type of deadly pilot error, and a large fraction of these incidents involve inverted or nearly inverted flight at some point. One example:

http://www.aopa.org/asf/epilot_acc/atl02fa043.html

Most spatial disorientation incidents are single-engine fixed-wing aircraft flying in VFR during non-VFR conditions.

A passenger aircraft would be far less likely to suffer this sort of thing because there would be two better-trained IFR pilots in control. But it's not impossible.

[Ezri]

Call me daft but I really don't understand how you can fly upside-down and not notice it....

[Eddylizard]

Originally Posted by Ezri

Call me daft but I really don't understand how you can fly upside-down and not notice it....

I think Alchemy's post above yours covered that, but spatial disorientation is a well documented occurrance. Possibly the final upside down bit happens so soon before the controlled crash into terrain that the pilot has no chance to respond.

[hijack] I have heard a UL that the most common words heard at the end of CVR recordings are "Oh shit" [/hijack]

Disclaimer - Hijack might not be the most appropriate word in this discussion, but it is board lingo.

[WildaBeast]

Originally Posted by Eddylizard

[hijack] I have heard a UL that the most common words heard at the end of CVR recordings are "Oh shit" [/hijack]

I just happen to have a book of CVR recordings sitting in front of me. A brief scan of the book shows that while that particular phrase isn't necessarily the very last words, it does appear to be fairly common.

In my experience, it is extremely easy to get spatially disorientated flying in cloud without reference to instruments. However, I don't know of anyone stupid enough to test fly a purely VFR equiped aircraft (maybe with as little as an altimeter and airspeed indicator) in IMC conditions (basically clouds or low visibility to the layman). If the aircraft was equiped for instrument flight, the test pilot should have been capable of using it as such.
Also, I know when performing a 1g manoeuvre (spelling?) without a visual reference you have no way of knowing which way is up, but the OP may be suggesting that it was in sustained inverted flight before its "climb" into the ground, which most definately would be noticed by the pilot (the inverted flight, not hitting the ground).

In short, definately possible but not definately true IMHO.

[Eddylizard]

Originally Posted by moreton

In my experience, it is extremely easy to get spatially disorientated flying in cloud without reference to instruments. However, I don't know of anyone stupid enough to test fly a purely VFR equiped aircraft (maybe with as little as an altimeter and airspeed indicator) in IMC conditions (basically clouds or low visibility to the layman). If the aircraft was equiped for instrument flight, the test pilot should have been capable of using it as such.
Also, I know when performing a 1g manoeuvre (spelling?) without a visual reference you have no way of knowing which way is up, but the OP may be suggesting that it was in sustained inverted flight before its "climb" into the ground, which most definately would be noticed by the pilot (the inverted flight, not hitting the ground).

In short, definately possible but not definately true IMHO.

Just out of interest, in the inverted position, would the wings not give 'negative lift' that is to say that they are shaped to pull* the plane towards what is normally considered to be their top surface, so that they are now pulling the aircraft downwards?

Following the realisation that his aircraft was inverted would this effect hasten the aircraft's descent and shorten the time in which the pilot could respond? Depending on inital altitude, doubtless the pilot would relise his aircraft was inverted, but how long does he have to take corrective action? Is it possible that in that short timespan the aircraft has accelerated to the point where any corrective action will exceed the design envelope of the airframe such that it breaks up anyway if you try to roll it?

*Not a very technical term. Sorry.

Originally Posted by Eddylizard

Just out of interest, in the inverted position, would the wings not give 'negative lift' that is to say that they are shaped to pull* the plane towards what is normally considered to be their top surface, so that they are now pulling the aircraft downwards?

That is correct, as in straight-and-level unaccelerated flight the plane must still produce lift in the amount of the plane's weight in order to keep it from descending.

[UEL]

Originally Posted by Singing in the Drizzle

The only thing I can think of is a South American crash of passenger airplane (737 if I remember right).... The airplane nearing or exceeding the speed of sound broke apart before hitting the ground.

I watched an excellent documentary on that prior to my flight accident investigation course. Out of the 4 artificial horizons they had, one was gone, one was broken, one was good and the other inaccessible. The one that was doing "wonky" things was actually depicting the reality of what the plane was doing. The one they switched to was the broken one that would stick, and about every few minutes it would slip. Because it was holding steady, and in the 1 g turn it felt like it was reading correctly, the flight crew apparently went with it as their eyes. However, when it slipped and told them that they were in a 95 degree turn to the left, the crew panicked, sending the plane into a mountain.

UEL:
I'm always happy to be proved wrong and learn something new, but I'm pretty sure that there were only 3 Attitude Indicators on the COPA 737-200. Due to a defective pin on some wiring to the Captains (I believe) ADI, the Captains and First Officers ADIs were not synchronised. The F/O's and Standby Horizon would have been displaying the same data. Both main ADI's were also being fed from the same gyro, which is not the normal selection.
By putting their faith in the Captains ADI, and not cross checking the F/O's or Standby, they rolled the aircraft first steeply left, then when they realised, over compensated and rolled it hard right into the inverted position. A steep dive was entered, which they failed to recover from, and the aircraft broke up after passing its maximum design speed.

And yes, in inverted straight and level flight, a small amount of forward pressure is required on the controls.

[RichardM]

Originally Posted by Ezri

Call me daft but I really don't understand how you can fly upside-down and not notice it....

When you are in a roller coaster upside down but going very fast in a loop, the centrifical force pushes you to the bottom of the car. Hence, down feels like up. By the way, this is written from an emperical point of view and not from a personal experience. I refuse to ride roller coasters preferring to take my chances on the rally course.

[Troberg]

Just out of interest, in the inverted position, would the wings not give 'negative lift' that is to say that they are shaped to pull* the plane towards what is normally considered to be their top surface, so that they are now pulling the aircraft downwards?

It depends on the aircraft and attitude. Some aerobatic aircraft has a symmetrical wing profile, which will give equal lift either way. Even with an assymmetrical profile, you can get positive lift by changing the angle of attack. Just lift the nose a little more above the horizon and you'll get lift.

[Eddylizard]

Originally Posted by Troberg

It depends on the aircraft and attitude. Some aerobatic aircraft has a symmetrical wing profile, which will give equal lift either way. Even with an assymmetrical profile, you can get positive lift by changing the angle of attack. Just lift the nose a little more above the horizon and you'll get lift.

I'm going to soung like Mr. Stupid here, but if you have a wing profile that gives equal lift forces upwards and downwards, then don't those two forces cancel each other out - so how does the plane get in the air to begin with? As I understand it, lift depends on a different rate of airflow over one surface of the wing than the other, exploiting the pressure difference. If the wing is symmetrical, then there is no lift.

Sorry if it's a silly question.

[Troberg]

I'm going to soung like Mr. Stupid here, but if you have a wing profile that gives equal lift forces upwards and downwards, then don't those two forces cancel each other out - so how does the plane get in the air to begin with? As I understand it, lift depends on a different rate of airflow over one surface of the wing than the other, exploiting the pressure difference. If the wing is symmetrical, then there is no lift.

Draw a symmetrical wing on a piece of paper. Now tilt it a little. See how the lower side gets shorter and straighter, while the upper gets longer and more curved, as the center point on the leading edge moves down?

Let me do a very quick aerodynamics 101:

There are three guys keeping an aircraft flying: Newton, Coanda and Bernoulli. I argue that Coanda and Bernoully are just riding on the back of Newton. This is their respective contributions:

Newton: The wing has an angle of attack that shoves air downwards. Force and counterforce. Something goes down, something else must go up, in this case the wing.

Coanda: The upper surface of the wing is curved, and a stream of air (or liquid) tends to follow a curved surface. This is easy to test with the back of a spoon under a faucet. The air follows the wing, which means that it moves down. Then Newton takes over, force and counterforce, something goes down and the wing goes up.

Bernoulli: The upper path is longer, which means the air is spread over a larger surface, in other words the pressure is lower. Air is sucked in from above, and once again Newton takes over and the wing moves up.

On a symmetrical wing, Bernoulli has less effect, but is still there, and the other two still do their work. Depending on the wing profile, angle of attack and speed, you vary the influence these three guys have and balance lift vs drag.

In some circumstances, Bernoulli is taken out of the equation completely, such as in some model aircraft with a completely flat wing profile. On the other hand, if you have a really high angle of attack, Coanda is out of the game, as the air will not follow the top of the wing and instead form a turbulent vortex which will induce a lot of drag and probably a stall. You can never really get rid of Newton, though, unless you have an engine strong enough to keep the aircraft flying without a wing, such as some model aircraft do.

If you don't trust me, look at a helicopter. The rotor blades are always symmetrical. Look at these photos of slices of rotor blades I have on my wall (the big one is a Bell 205, the small one an AS-350, and the wing spar is from an old glider):

http://rpglab.net/nobackup/rotor/1.JPG
http://rpglab.net/nobackup/rotor/2.JPG
http://rpglab.net/nobackup/rotor/3.JPG
http://rpglab.net/nobackup/rotor/4.JPG
http://rpglab.net/nobackup/rotor/5.JPG
http://rpglab.net/nobackup/rotor/6.JPG
http://rpglab.net/nobackup/rotor/7.JPG

OK, Little Grasshopper, when you can grasp the concept of laminar flow wings, it will be time for you to leave.