But a hill such as that diagrammed above also allows a different kind of soaring flight . The most obvious difference to the spectator (and even more so to the pilot) is that the glider is flown on the back-side of the hill! Since there isn't any lift there, how does this work?
In the diagram above, notice that the steep drop of the slope on the back of the hill causes the wind to separate from the hill, leaving a zone of still air. This separation is what makes DS possible. The glider is flown along a path like that defined in red. This red shape is not necessarily intended to represent an aerobatic loop (perpendicular to the slope), but rather a loop tilted to one side. However, a vertical loop will work (but is harder to do), as will a loop turned entirely on its side (parallel to the slope face--much easier to do).
The basic principle of DS is that the glider is repeatedly flown across the boundary which separates two bodies of air which have different speeds. So, for example, as the glider is flown from the top of the red circle toward the back of the hill, in addition to its airspeed, it acquires additional ground speed, since it is moving with the wind toward the right.
As the plane dives down and crosses the boundary at point A into the still air, the plane instantly acquires more airspeed because the still air is not moving to the right---in effect, the still air presents "head-wind" to the plane. Then, the plane continues to turn back toward the hill with its additional airspeed. As the plane crosses the boundary again (at point B), it is now moving in a direction opposite to that of the wind over the top of the hill. This results in another sudden, and substantial, increase in airspeed. The plane continues to turn and the process is repeated. With each "lap" through the boundary, the airspeed of the plane increases, directly related to the speed of the wind coming over the ridge.
How fast can a plane go while DS'ing? Where normal soaring is merely about maintaining or gaining elevation, DS is all about speed --- raw, heart-thumping, close-to-the-deck speed. While DS'ing, normal ridge-lift soaring is relegated to "priming" the plane for its initial plunge down the backside of the ridge to kick off the first DS circle. Then, aside from the speed limitations of any particular plane, the maximum airspeed attainable is limited mainly by the difference in speeds of the two bodies of air. The stronger the wind, the faster the plane can go. Speeds in excess of 200mph have been reported via this technique, and it is hoped that a speed record may be set if a sufficiently durable plane can be designed and built to withstand the large forces on the wing and tail surfaces that occur with this type of soaring.
Why does DS cause the destruction of so many planes?? High-speed flutter of either the wing or the tail surfaces, and sudden very strong forces as the plane encounters turbulence at the boundary, are probably the most common causes of spectacular mid-air destruction of DS'ing planes. Of course many planes are also lost when the pilot makes a tiny error in controlling the plane and augers it into the ground or into a tree. The problem here is that most DS takes place only 5-50 feet above the ground, so at high speeds, things happen very fast: a small mistake in controlling the plane can lead to an almost instantaneous crash. That's part of what makes it so much fun, of course.
What kind of plane do you need for DS? Almost any plane except a light-weight floater will do. Generally, the faster and better penetrating a plane is (more efficient, less drag), the better DS it will do. Heavier wing-loading also helps. EPP flying wings (Zagi, Boomerang, etc.) will work well, although they have a lower top speed than stiffer wings. These EPP flying wings are excellent planes to learn DS with, however, since they survive the inevitable crashes a lot better. If you learn DS with a more fragile plane, be prepared to see it destroyed.
