We know from popular movies and books that a pilot can be identified as having "the right stuff" when he "pushes the edges of the envelope." Not wanting to be found lacking the right stuff ourselves, it's useful to review exactly what the envelope is, what its edges refer to, and what that tells us about the aircraft we're flying.
The figure below illustrates the envelope of a typical but hypothetical Normal category general aviation airplane. Such a figure is called a V-n (or sometimes, V-g) diagram, because the envelope is defined in terms of indicated (or calibrated) airspeed (V) and load factor (n).
Any given airplane can have a virtually unlimited number of V-n diagrams because the dimensions of the envelope depend on four variables: the weight of the airplane, its configuration (clean, gear down/flaps up, gear down/flaps down, and so on), symmetry of loading (a rolling pullup, for example, induces different loads than a constant-bank-angle — symmetrically loaded — pullup), and altitude (although this tends to have little effect on most light aircraft).
We know we're dealing with a Normal category airplane because of the limit load factors of +3.8/-1.52 Gs. FAR Part 23 insists that an airplane be capable of withstanding 1.5 times the limit load factor before failing, hence the ultimate load factors are depicted at +5.7/-2.28 Gs. If you exceed the limit load factor, the primary structure of the airplane may (and probably will) be permanently deformed. If you exceed the ultimate load factor, the primary structure may be expected to break. An airplane is theoretically capable of sustaining its ultimate load factor (but not more) exactly once. No one will debate your possession of the right stuff if you choose to go out and test this theory.
An aside here: Some people assume that an airplane can sustain higher loads at lighter weights. This might be true if lift (the load on the wings) were the only factor involved. But there are items whose weights remain the same regardless of the overall weight of the airplane. The amount of stress the engine mounts can stand, for example, has nothing to do with the weight of the airplane and everything to do with the weight of the engine. (Technically, of course, we're talking about mass rather than weight, which is mass times acceleration.)
The curved lines forming the left side of the envelope correspond to the wing's maximum coefficient of lift (CLmax). The coefficient of lift (C L) is a relative measure of an airfoil's lifting capability. To avoid the mathematics, let's just say that it's a characteristic of the wing design and that it increases as angle of attack increases up to the point of the stall. The wing stalls at CLmax. If the angle of attack increases beyond the angle that generates maximum lift, the stall occurs and CL drops off sharply. For the purposes of this discussion, simply remember that CLmax is directly related to the angle of attack at which the wing stalls. Indeed, to the left of the curved lines is the stall region. A level, unaccelerated stall occurs at the airspeed at which the curved CLmax line crosses the horizontal 1-G line (given the four variables stipulated above for the V-n diagram).
The point at which the CLm line crosses the positive load limit factor line is maneuvering speed, Va. Honk the stick back as hard as you like at any speed below V a, and the airplane will stall before the load reaches 3.8 Gs.
Following the load limit factor line out to the right, we encounter a vertical line that corresponds to the maximum structural cruising speed, VNO. Because our V-n diagram assumes gear up/flaps up, the speed range between the 1-G stall speed (CLm, at a load factor of 1 G) and VNO correspond to the green arc on the airspeed indicator.
How is V NO determined? A lot of factors go into it, including gust loading, but it must be no less than the minimum design cruising speed, VC (which, in a Normal or Utility category airplane, is 33 times the square root of the wing loading). VC is also used to determine the design dive speed, VD(For a Normal category airplane, VD is 1.4 times the minimum VC.) For certification purposes, the airplane's manufacturer has demonstrated that the airplane can be dived to VD without coming apart.
But VD serves another purpose: It helps define the never exceed speed, Vne, denoted by the redline on the airspeed indicator. Vne may not be more than 0.9 times VD. The airspeed region between V NO and V ne is the yellow arc on the airspeed indicator of a light, piston-powered aircraft.
The vertical line denoting Vne on the V-n diagram is a limit that offers another opportunity for a pilot to demonstrate the right stuff. Outside the envelope in this direction you are again flirting with structural damage or failure resulting from any of a number of causes.
Flutter is one. Any structure — a control surface, say — will vibrate at a certain natural frequency. One of the purposes of certification testing is to assure that these frequencies will not be encountered in normal flight. Outside the envelope, however, the structure may be forced to vibrate at or near its natural frequency, at which point it will flutter and fail. Catastrophically. The time between the onset of flutter and failure can be extremely short — too short to allow a pilot to recover from it.
Some airplanes encounter aileron reversal at very high speed. At some given speed, dynamic pressure couples with aileron deflection to cause the wing to twist to the extent that aileron effectiveness is lost. At any speed above this, the twisting is great enough that the airplane will roll in the direction opposite the pilot's aileron inputs. Fortunately, this is seldom an issue in light general aviation airplanes.
Another phenomenon, which, like aileron reversal, is caused by the interaction of aerodynamic forces and the elasticity of the structure, is called divergence. This is a violent instability that can produce almost instantaneous structural failure.
Jets that cruise at speeds nearing or surpassing the speed of sound may encounter high-speed buffet or loss of stability or other problems due to air compressibility factors.
But the agent most likely to bring a general aviation pilot to grief while pushing the edges of the envelope is as mundane as an encounter with a vertical gust of wind, particularly in an airplane with high-aspect-ratio wings and relatively low (say, Normal category) load limit factors. What happens is this: The airplane encounters a vertical gust, which very quickly increases the angle of attack and, therefore, the CL. As the lift increases, so does the load factor. It's just the same as honking back on the stick.
Fortunately, airplane designers take gusts into account — substantial ones at that. Broadly speaking, the rules under which most of today's Normal category airplanes were certified required that they be able to withstand a 30-foot-per-second (1,800-fpm) gust at V NO and a 15-fps (900-fpm) gust at V D. Those are substantial gusts, but not unusual ones in convective turbulence. In fact, 45-fps (2,700-fpm) gusts are not unknown.
Obviously, we want to slow down when encountering severe turbulence. But how slow? If we're going too slow, a sharp gust could stall the airplane. If we're going too fast, it could overstress the airframe. For any given weight, there is a minimum-maximum spread of airspeed within which to operate. These speeds are higher at higher aircraft weights, and the spread widens as weight increases. The greatest margin for error, however, is found precisely at VA. Why? Because that's where the airplane can tolerate the highest CL max before either stalling or overstressing. (If you fly an older airplane for which the POH lists no maneuvering speed, it can be estimated by multiplying the clean stall speed at MTOW, V so, by 1.7.)
One more point: For any given weight, configuration, and angle of attack, there is only one stall speed and one maneuvering speed. For that matter, there is also only one speed for maximum range, one speed for maximum endurance, one speed for a maximum distance glide, and one speed for a minimum sink glide. The upshot is that all of these speeds rely on a constant coefficient of lift and all change as a function of the square root of the change in weight. Double the weight, for example, and all the speeds increase by a factor of the square root of 2, or 1.414. If your POH only lists speeds based on maximum takeoff weight, you now have a way of computing them for any reduced weight. Here's the formula: speed at reduced weight = speed at MTOW x (square root of [actual weight/MTOW]). Now that's the right stuff.