(December 2011) It will be a while before the NTSB issues its findings about the crash of The Galloping Ghost at the Reno Air Races this year. There were so many witnesses, however, and photographic and video coverage of the disastrous accident was so clear, that it did not take long for theorizing about the cause to converge on a single scenario.
The extensively modified P-51 had been involved in racing on and off, with various pilots and under different names, since 1946, but news reports referring to it as a “vintage” airplane, with their suggestion of overall wear and decay, were misleading. It had been crashed, repaired and modified a number of times and was barely recognizable as a P-51. Ten feet had been clipped from its wings and the iconic belly scoop and radiator had been removed altogether, replaced by a tank of water that boiled off during flight and left behind a telltale trail of steam. It was owned and flown by 74-year-old Jimmy Leeward, an Ocala, Florida, real estate developer and an experienced race and stunt pilot.
The Galloping Ghost was approaching the area of the bleachers, which are 1,000 feet from the southern edge of the egg-shaped 8.5-mile race course, when it suddenly shot upward, rolled to the right and plunged out of a tall, narrow loop to crash just in front of the VIP boxes. The impact and the flying debris killed at least 10 people on the ground and gravely injured dozens. That the accident occurred where it did was a piece of diabolically bad luck. Unlike the stands that entirely surround a NASCAR course or a football stadium, those at Reno occupy a very small fraction of the perimeter of the course. If the accident had occurred anywhere else, only the pilot would have died.
A high quality video of the loss of control (vimeo.com/29519344) shows the airplane rounding a pylon in a steep left bank. It apparently encounters wake turbulence and briefly rolls past vertical before recovering. The pitch-up begins right after this excursion, and the tailwheel is out almost immediately. The tab falls away while the airplane is inverted at the top of the loop.
The initial failure, then, must have been of an actuator or a hinge, not of the tab itself. Photographs provided other clues. In a profile view of the airplane seconds before impact, the cockpit appeared to be empty and the tailwheel was extended. Whatever held the tailwheel up had failed, presumably under a high G load. The apparent absence of the pilot was explained by another photograph, in which the airplane was seen from above and his white helmet was visible far forward under the canopy; he was apparently slumped over the stick, his head below the canopy rails.
The Galloping Ghost was equipped with data recorders that stored various parameters and also transmitted engine and GPS information to the ground. Within a couple of days of the accident it was reported that the telemetry had registered an 11 G pitch-up and that the engine was pulling full power — 105 inches of supercharged manifold pressure — in the final moments of the 440-knot dive.
Race speeds are much higher than the indicated airspeeds for which World War II fighters were originally designed, and the airplanes have to carry a lot of nose-down trim to compensate for excessively negative stabilizer incidence. Even the stock P-51H, the final version of the great fighter, could not trim hands-off for level flight at its maximum power. The effect of the nose-down trim — tab deflected upward — was to push the trailing edges of the elevators downward, as if the pilot were pushing forward on the stick. The tab, making use of the very powerful aerodynamic forces at an indicated airspeed around 390 knots, held the elevators down more forcefully than a pilot could. When the tab failed, the elevators would have jumped to their natural trail position and the airplane would have pitched violently upward. The pilot, slammed down and forward, was probably unconscious for the remaining few seconds of the flight.
The same thing had happened once before at Reno. In 1998 a pilot named Bob Hannah, flying the modified Mustang Voodoo, experienced a similar failure, was briefly unconscious and came to bent over, looking at his feet, inverted, and 4,000 feet above the ground. He managed to regain control of the airplane and landed safely.
As speed increases, pitch excursions become harder to manage. One reason is obvious: Aerodynamic forces upon control surfaces increase, but the strength of the pilot does not. Another reason is subtler, and has to do with the behavior of lift coefficient. Lift coefficient is the ratio between the lifting force produced by a wing and the force of air, at the same speed, against a flat surface. The weight of the airplane being constant, the faster it goes the lower its lift coefficient. At very high speed an airplane might have a lift coefficient of, say, 0.1 — one 10th. Lift coefficient increases at a constant rate with angle of attack. That rate varies from wing to wing, mainly as a function of aspect ratio, but for the sake of discussion let’s say that it increases by 0.1 per degree. If an airplane’s lift coefficient is 0.1 in level flight, then all that is needed to generate 2 Gs is to increase the angle of attack by one degree. At a lower speed, however, say the speed for best rate of climb, it might take six degrees’ change of angle of attack to double the G loading. That same six-degree pitch change, occurring at high speed, would generate six Gs.
Nevertheless, although The Galloping Ghost was going very fast, the reported 11 G pitch-up must have been a transient overshoot. The dynamic pressure of air at race speed is around 480 pounds per square foot. The area of The Galloping Ghost‘s clipped wing was 170 square feet; if the airplane weighed 9,000 pounds its wing loading was 53 pounds per square foot, and so it would have been flying at a lift coefficient — or properly a “normal force” coefficient, because we’re talking about the complete airplane — of 53⁄480, or about 0.11. To pull 11 G, it would have needed a coefficient 11 times greater, or around 1.2. That would be close to the stalling angle of attack of the thin laminar-flow wing. The Galloping Ghost’s stable stick-free attitude without a trim tab is unlikely to have been that nose-up.
One has to wonder why seemingly similar tab failures on The Galloping Ghost and Voodoo had such different outcomes. Perhaps the positions of the pilots as they slumped forward moved the sticks in different ways. But the airplanes were also aerodynamically different. The greatly reduced wing area of The Galloping Ghost, coupled with a basically stock horizontal tail, would have increased the airplane’s longitudinal stability and could have made the untrimmed stick forces even larger. CG location could play a role; so could the fact that Ghost had only one trim tab, and Voodoo two.
The NTSB’s metallurgists will microscopically examine the fracture surfaces on the trim tab itself and on its hinges and actuator to determine whether the failure was due to a momentary overload or to pre-existing cracking or metal fatigue. They should be able to tell whether the failure was preceded by the rapid force reversals of flutter. They will know more than we do today about the airplane’s structure, modifications and service history. With access to both telemetry and memory chips recovered from the wreckage, investigators will probably reconstruct the flight path to see whether any pilot input occurred after the initial failure. Eventually, we will understand a good deal more than we do now about the worst accident in Reno racing history.
In February’s Aftermath column, which concerned an accident that press reports had attributed to people crowding to the front of a small twin turboprop after a crocodile escaped from one passenger’s carry-on bag, I quoted a British accident investigator, Tim Atkinson, as saying that the account seemed to him “extremely unlikely.” I did not make sufficiently clear to readers that what he considered unlikely was not that a large CG shift could precipitate a crash, which is simply a scientific reality, but rather the report that a passenger had smuggled a crocodile aboard the airplane stuffed inside a sports bag. I certainly agree that the story sounds improbable, and I apologize for the ambiguity in the original column.
This article is based on information available at the time of this writing, and is intended to bring the issues raised to our readers’ attention. It is not intended to judge or to reach any definitive conclusions about the ability or capacity of any person, living or dead, or any aircraft or accessory.
Send reader mail to: [email protected] or P.O. Box 8500, Winter Park, FL 32789.