Saturday, March 19, 2011


Stalls? Yes, stalls. An article in the current issue of Aviation Week and Space Technology describes how loss of control has replaces controlled flight into terrain as the number one cause of accidents, with details on several accidents in which stalls were a factor: the Colgan Air Q400 at Buffalo, NY, in 2009, a West Caribbean Airlines MD-82 in Venezuala, and a Turkish Airlines Boeing 737-800 in Amsterdam.

How could a professional pilot stall an airplane into the ground? There are many contributing factors, but I think one of the must be the way that our current training materials make the meaning of "stall" obscure. To quote the Airplane Flying Handbook [FAA-H-8083-3a], "A stall occurs when the smooth airflow over the airplane’s wing is disrupted, and the lift degenerates rapidly."

A young instructor recited a similar definition to me the other day, and I compared it to hay after the horse has eaten it. The reason is that while this definition describes the aerodynamics effectively, it doesn't tell you how to recover from a stall. How does a pilot smooth the airflow? By polishing it clean? How can you reattach the boundary layer? Is that what duct tape is for?

In the old days (and I'm not that old) we used to teach that a stall meant too high an angle of attack. That definition tells you how to recover! "Stalled," I say, pulling an invisible yoke into my gut. "Unstalled," I say, pushing an invisible stick to the invisible panel. "Stalled." "Unstalled." "Stalled." "Unstalled." Until the point is made.

The FAA also says "the application of more power, if available, is an integral part of the stall recovery." That's wrong; ask any of my glider students! You add power to climb away, not to recover.

So I have taken pen to paper again (metaphorically) and written another cranky letter to the editor of Aviation Week. (AOPA Pilot didn't publish this one, by the way, although I got a nice note from Bruce Landsberg about it). Here's what I wrote, working to fit their 200 word limit:

The increase in loss-of-control accidents is partly explained by the confusing way we teach stalls. Today’s instructors talk about stalling as a boundary layer effect, which does not offer any insight into recovery. We used to teach that a stall was excess angle of attack, which also describes the recovery procedure: Reduce the angle of attack. A pilot who understands stalls this way is unlikely to descend to the ground with the
yoke fully aft.

Pilots are also taught that power is part of stall recovery, although my glider students do not have that option. The purpose of power is to climb away. The net altitude loss from starting recovery before adding power is minimal, and the risk of loss of control is reduced. This recovery sequence is most important in an airplane like the Bombardier Q400, which risks both spin entry (due to the large left-turning tendency from two propellers) and Vmc loss of control (in the event of an engine failure). (Granted, if engines have long spool-up times it is prudent to start that process as soon as possible.)

Advanced training as offered by the major simulator centers offers few opportunities to correct this basic misconception, leading to the tragic results you described.

Labels: ,

Thursday, March 17, 2011

Flying with Finesse

One of my students suggested a book I'd never read, Henk Tennekes The Simple Science of Flight (MIT Press, 1996). My heart sank a little when he lent me his copy. "This is a book I'd like to write," I sighed.

Tennekes has the audacity to apply the basics of Aeronautical Engineering to all flying creatures, "from insects to jumbo jets." The discussion is clear and gentle. He puts all of the material together -- and I mean all of the material. For example, study The Great Gliding Diagram, which shows the polar curves for 10 flying objects: the cabbage white, the Gossamer Albatross, a typical sailplane, a swift (the bird, not the Globe-Temco beauty), a real albatross, a budgerigar, an ultralight, the Fokker Friendship, a pheasant, and the Boeing 747. You can compare the performance requirements directly, and there are many interesting consequences about wing loading, wing shape, power requirements, and the like.

Apparently the French call "L/D," the ratio of lift to drag, the finesse. I will now use this term forever. By the way, we owe the concept of the polar graph that shows vertical speed versus horizontal speed to another French engineer, Gustave Eiffel, of "Tour d'Eiffel" fame.

My current aeronautical thinking is that everything depends on two parameters: energy, which I wrote about in this post, and L/D, which I wrote about here. Those posts are just the beginning, of course, hence my idea to write a book explaining every maneuver and performance calculation in terms of these two.

But it would be a nerdy book, full of equations and tables. Tennekes does not shy away from equations (Stephen Hawking claims that every equation reduces the sales of a book by a factor of two), but his gentle prose and simple ink drawings really evoke the beauty of these simple ideas. Perhaps with Tennekes's inspiration I can write something with a little more finesse, too.

Labels: , , , ,