Bill Crawford’s Flightlab Blog

When Flying magazine arrives each month, I turn to Peter Garrison’s column, “Technicalities,” first. If you like the wonky part of aviation—lift-over-drag ratios, laminar bubbles, and maybe the area rule if you need to go fast—Garrison is dependably interesting and informed. And he’s a wry fellow. I’m such a groupie that I actually rip “Technicalities” out each month and stash it for future reference in a file marked “Garrison.” My theory is that after reading the rest of the issue I can then safely toss the magazine and get on with life. I’ll bet psychologists have a name for this behavior, however, because the entire process, from evisceration upon arrival until a magazine’s belated farewell into the dumpster, still takes around five years. Magazines about flying are just inordinately hard to dispose of, violated or not, don’t you think?

In the February 2007 issue of Flying, Garrison writes about turning performance. This was occasioned by a 2006 accident in a Cirrus involving New York Yankees pitcher Cory Lidle and his CFI, Tyler Stanger. Apparently, they attempted to make a 180-degree turn within the confined airspace boundaries of the East River, but misjudged and instead crashed into a Manhattan apartment building.

So the question Garrison raises is: when you really have to, how do you make the tightest possible turn? Garrison points out that for a level turn (no change of altitude) the radius is a function of airspeed and bank angle. The radius varies directly with the square of the airspeed, and indirectly with angle. You might go to Garrison for the details (or to www.flightlab.net/pdf/8_Maneuvering.pdf ), but what this boils down to is that the tightest level turn under a given load restriction happens at the highest bank angle that the aircraft can sustain structurally, while flying at the lowest speed necessary to maintain that bank angle without stalling. If the aircraft has a normal category load restriction of 3.8 G, that means a 75-degree bank angle. Since stall speed goes up by the square root of the load factor, at that bank angle stall speed would be almost twice what it would be if a building were not in the way.

Since lift is what makes an aircraft turn, I’d add that the aircraft must fly as close as possible to its maximum coefficient of lift. If it’s doing that, it’s flying at its lowest possible speed above stall. If you tufted the wing, you’d probably see flow reversal beginning along the trailing edge. The stall warning would be on—in fact if it weren’t on, you’re not pulling hard enough. But if you pulled too hard, and got into a real buffet, lift would suddenly drop and your turn rate would actually go down. I make sure my students see this when we do accelerated, turning stalls (or pull-ups from spins and such). My Zlin and Air Wolf are disciplined soldiers in the high angle of attack regime, and don’t drop a wing as long as they’re flown in a coordinated manner—meaning no sideslip. In the Cirrus (as I’ve heard but can’t confirm) pulling too hard can cause a sudden roll-off and perhaps an entry into autorotation if you don’t get the stick forward—fast.

In his column, Garrison goes on to point out that a level turn is bit of an over-simplification, and asks, “Does a pull-up from cruising speed into a steep wingover turn allow reversing course within a smaller space, without exceeding G and stall-speed limitations, than a level turn at a lower entry speed would?” Then he adds, “I don’t know the answer, but I would be interested in hearing from readers who do.” I found this later statement disquieting, but strangely thrilling. After all, the man went to Harvard. Don’t they still discuss this stuff in Dunster House, deep into the night?

I suspect Garrison’s spam filter is already choking on the subject of lift vector versus gravity vector. The former, as remembered from primary ground school, points upward in level flight, from its origin at the aircraft’s center of gravity. It remains perpendicular to the wings, so that when you bank the aircraft the lift vector tilts toward the horizon. That tilt creates a horizontal component of centripetal force—or radial G—which causes the aircraft to turn. The gravity vector obeys a higher law and always points toward the center of the planet.

When you bank an aircraft past 90 degrees, the lift vector begins to point below the horizon, and gravity and lift begin to align. As the aircraft rolls toward inverted, gravity increasingly contributes to radial G, and thus to turn rate (a turn which has now become tilted on its side, in the fashion of a loop). There are some illustrations in the Flightlab document noted above, on page 8.7, that make these relationships clearer.

I thought I had better go out and fly some wingovers, and see if I could come up with at least a qualitative answer to Garrison’s question. I took the Zlin (fresh from Bobby Thissell’s annual and sounding smooth) and my parachute (still within Chuck Braga’s packing date and thus making maneuvers past 60-degrees of bank and 30-degrees nose up or down strictly legal) and headed down my favorite reference-point power line. The drill was to figure out how to reverse direction while causing the least amount of lateral shift from the line, and do it in a way that required the least piloting skill. Wind wasn’t a factor, since there didn’t appear to be any. Obviously, a hammerhead turn or a half loop followed by a roll back to upright would normally produce the least, or no, lateral shift, but they both involve a lot of skill and an aerobatically compliant aircraft. If you don’t know what you’re doing it’s easy to run out of flying speed, and thus control.

The wingover is better for dummies. I found that by (1) pitching the nose up to about 60 degrees, and (2) holding some aft stick while rolling the lift vector maybe 30-degrees below the horizon, and (3) pulling on through while bringing the wings back to level as the view out front turned to dirt, I had no trouble remaining tight to the line. Or at least apparently so, because it’s difficult to judge these things from altitude. The G forces started to rise moderately as the wings rolled back to level, then picked up more noticeably during the pull-up from the dive. You can’t turn an aircraft without applying G. In a wingover you apply most of that G in a vertical plane, rather than a horizontal one.

The success of any vertical maneuver, in terms of the final altitude when the excitement is over, is very much a function of the airspeed you bring to the problem at the start. You convert that speed into the altitude gain necessary to complete the maneuver with minimum altitude loss. If you’re going fast, you can end up higher than your entry altitude (perhaps the pilots’ fear of busting the floor of the New York Class B airspace contributed to the accident). But if you’re too slow, you can end up in the East River.

I’ve flown the Hudson numerous times. The best flight was back to Plymouth in a Pitts S1S I’d just bought. It was my first flight in a single-seat Pitts—save for a takeoff and a skittery-Pitts landing at Cape May, New Jersey, after handing over the check. I was happy, excited, and petrified. The worst flight, of course, was the first time down the Hudson after 9/11. But the East River always looked too small, even on the puffed-up Terminal Chart. Why risk it?