Air Race Accidents
Safety in air racing depends on the airplane as well as the pilot
![]() |
After a highly modified World War II P-51 Mustang crashed into spectators at the National Championship Air Races at Reno-Stead Airport, Reno, Nev., this past September, there was the predictable criticism of air races being an excessively dangerous sport for participants and spectators alike. The death toll quickly rose to 11, including the pilot with about 70 injured. Enthusiasts and air race organizers defended the activity, noting that only the best-qualified pilots are allowed to participate, and safety is emphasized in everything. The FAA is active in providing oversight and promoting safety for racing events, ranging from prescribing minimum distances between the crowds and aerial activity, to certifying racing check pilots. A records search shows that the NTSB investigated three fatal air racing accidents in the 10 years before the September accident, with about a dozen more if you go back into the '60s. Generally, these previous accidents had in common that the pilots were the fatalities. In the context of extreme speeds and maneuvers, and the informed assumption of risk by the participants, three or four fatal racing accidents in 10 years doesn't seem as grim as a decade's worth of fatal highway accidents, or 10 years' worth of GA fatal accidents involving weather and fuel mismanagement.
Venture M20
On September 13, 2002, an amateur-built Venture M20 airplane crashed after failure of the left and right horizontal stabilizers and elevators. The accident was at the Reno-Stead Airport, during the sport-class race as part of the annual Reno Air Races. The airplane dove into the ground and was destroyed. The pilot was killed.
FAA inspectors from the Reno Flight Standards District Office were at the airport monitoring air race activities. They responded to the accident site. Witnesses told them that as the airplane was rounding pylon Number 1, the horizontal stabilizers and elevators began flexing and then bent down. The FAA inspectors reported that the wreckage field was 450 feet long. The airplane was extensively fragmented. The airspeed indicator was found in the debris. The needle was trapped between the fractured glass and the instrument face at the 300 knots/red line. Official timers for the Reno Air Races reported to FAA inspectors that based on time over the measured course distance, the airplane was doing 330 knots just before the accident. The airplane was designed with a gross weight design-maneuvering speed (Va) of 156 knots indicated airspeed (KIAS). Never exceed speed (Vne) was 300 KIAS, and the maximum structural cruising speed (Vno) was 265 KIAS. At maximum gross weight, 2,000 pounds, the structure was stressed for positive and negative load factors of +5 and -2.5 G's.
Investigators learned during the developmental flight-test work with a prototype airplane that a spring was needed in the elevator control circuit to add feedback resistance to control forces felt by the pilot in order to prevent pilot-induced oscillation. However, the spring recovered from the wreckage wasn't the specified size. The spring tension was measured and found to be 30% of the required value.
The NTSB determined that the probable cause of this accident was the overload failure of the horizontal stabilizers and elevators due to a pilot-induced oscillation at a speed at or above Vne, which exceeded the design stress limits of the structure. A factor was the use of an improper spring in the control system by an unknown person.
P-51A Mustang
On July 27, 2007, an amateur-built P-51A Mustang was destroyed when it struck the tail section of a North American P51-D Mustang as it was landing on runway 36 during the EAA AirVenture Oshkosh Air Show. The airplanes had been part of an air race demonstration involving five aircraft.
The amateur-built aircraft rolled over to the right and impacted the terrain in a wings-level, inverted attitude. The other airplane skidded down the runway, and came to rest about 788 feet from the initial impact point. The pilot in the amateur-built airplane received fatal injuries. The other pilot wasn't injured.
The pilot/builder of the amateur-built P-51 applied for a Special Airworthiness Certificate from the FAA, stating: "This project represents a built-from-scratch exact full-scale replica of the North American P-51A Mustang. It's a 'Plans Built' aircraft built from the original production drawings acquired from the National Archives. The only exception would be the landing gear and a handful of original small components."
The North American P-51D Mustang was manufactured in 1945. The engine was a 1,490 hp Packard-built Rolls Royce V-1650-7. It had a total time of 628 hours since a Special Airworthiness Certificate was issued.
The NTSB determined that the probable cause of this accident was the P-51A pilot's inadequate visual lookout and his failure to maintain clearance from the P-51D.
SNJ Trainers
On September 18, 1994, two World War II trainers, a North American SNJ-5 and a North American SNJ-4, collided about two miles west of the Reno-Stead Airport. Both airplanes were beginning an air race involving six of the trainers. The consensus of witnesses on the ground was that the SNJ-5 overtook the SNJ-4, and struck it with its left wing from below. A participating race pilot reported that the airplanes were to be lined up abreast of each other at the beginning of the race. He said that the SNJ-5 moved out of position before the airplanes reached the staging area. When the airplanes reached the staging area, the SNJ-5 appeared to move back, but the pilot over-corrected and struck the SNJ-4.
The SNJ-5 pitched upward, and its vertical stabilizer and both horizontal stabilizers separated. The rudder remained connected to the airplane by its control cables. When the flight was in a near-vertical climbing attitude, the left wing folded upward, followed by the separation of the left wing's outer panel. The airplane began to cartwheel and spin. It crashed into a house. The pilot was killed.
The pilot of the SNJ-4 closed the throttle and lowered the main landing gears while experiencing severe vibrations. He regained control of the airplane and flew toward Reno-Stead Airport, landing without further incident. The NTSB determined that the probable cause of this accident was that the pilot of the SNJ-5 misjudged the distance between the airplanes.
Cassutt Racer
On September 6, 2008, an amateur-built single-seat Cassutt racing aircraft crashed after an in-flight breakup while maneuvering at the Reno-Stead Airport. The pilot was killed. The airplane was quite small, with a wingspan of 15 feet and length of 16 feet. Its height was four feet, and it had an empty weight of 500 pounds.
An FAA inspector reported that the pilot was participating in an International Formula One Pylon Air Race Class pilot-qualification flight. The flight was intended to be composed of a series of maneuvers that included 360-degree rolls to the left and right, followed by 180-degree rolls to the left and right. These maneuvers were being observed by personnel located on the airport ramp for the purposes of the pilot obtaining a Formula One Racing Pilot License.
The airplane was observed on an east-to-west heading parallel to runway 8 about 1,500 to 1,800 feet AGL. The pilot performed a 360-degree left roll at more than 200 miles per hour. Witnesses reported the airplane seemed to "bobble" and "pitch downwards" prior to abruptly leveling off. They estimated the roll rate was about 400 degrees per second.
The airplane made a climbing 360-degree turn to the left in order to set up for a second pass parallel to runway 8. The airplane exited the 360-degree turn, leveled off, pitched downward in order to gain airspeed, leveled off and proceeded to enter a roll to the right. Two witnesses estimated the airspeed from 220 mph to 240 mph. As the airplane rolled through about 90 to 120 degrees, witnesses stated the right wing separated from the airframe followed by the left wing separating. The airplane crashed at the airport.
Investigation revealed that the airplane was a rebuild of an airplane that had crashed in 2005. The accident airplane was rebuilt using a new Cassutt fuselage and empennage, using Cassutt design plans. However, the manufacturer of the original kit told the NTSB that the trim system on the accident airplane wasn't their design. The accident airplane had a trim system that was adjustable in flight. The original trim system could only be adjusted on the ground.
The accident airplane had a tube-and-fabric fuselage. The fuselage was a tubular steel truss structure that tapered in the cross-section toward the aft end of the airplane. The one-piece wing was of a twin-spar configuration, and manufactured with wood spars, ribs and skins. During a telephone interview, a representative of the company that produces the Cassutt kit told investigators that there's no published maneuvering speed or never-exceed speed for the Cassutt. The representative estimated that if they were to place a maneuvering speed on the airplane, it would be around 200 miles per hour.
The NTSB determined that the probable cause of this accident was the movement of the airplane's modified horizontal-stabilizer trim system during an intentional high-speed aerobatic maneuver that resulted in exceeding the design stress limits of the airplane and an in-flight structural failure. Contributing to the accident was the builder's deviation from the airplane designer's original trim system.

Subscribe to Our Newsletter
Get the latest Plane & Pilot Magazine stories delivered directly to your inbox