Last September an applicant for a private pilot license rented a Piper Tomahawk from an FBO in central Illinois, one of two Tomahawks the company had in its seven-airplane fleet. The applicant and the designated pilot examiner completed the oral portion of the exam and proceeded northwest of the airport to conduct the maneuvers required in the practical test.
The weather was good for the noon flight. A scattered layer of clouds was reported at 4,900 feet agl, the visibility was 10 miles and winds were variable at 5 knots.
During the maneuvers, however, something went horribly wrong. A witness described the airplane diving nose-low toward the ground with a counterclockwise rotation. Both occupants of the airplane were killed.
The wreckage pattern was consistent with a number of Tomahawk spin accidents in that it appears to have been a flat spin. The vertical velocity at impact was apparently so great that the crushed wreckage is almost beyond recognition. Previous certification tests documented a 10,000 foot-per-minute descent rate in a developed spin.
Newspaper accounts after the crash described the flight examiner as a World War II aviator who had worked as a flight instructor for the company that owned the airplane for the past 16 years. He had an unblemished flying record and had amassed some 30,000 hours in his log books. The private pilot applicant was a successful and popular building contractor. He was a licensed glider pilot, intending with this flight to upgrade to powered aircraft.
While experience does not insulate a pilot from misfortune, the only smoking gun in the accident is the Tomahawks troubled past, which is riddled with fatal spins involving well-qualified pilots.
An investigation into Tomahawk accidents by Omer Benn, an associate professor at the University of Illinois Institute of Aviation, uncovered the disturbing possibility that the airplanes control yokes can lock in the full aft position, typical of where a good pilot would have them while practicing stalls or slow flight.
Benn bases his conclusions on several factors. The rams horn yokes are frequently bent forward, parallel to the control yoke torque tube. A nylon bushing that guides the torque tube through the instrument panel shows half-moon cuts on the top and bottom, with the two cuts at opposite ends of the bushing. The bracket that holds the bushing in place shows signs of heavy forward pressure from the pilots trying to lower the nose. Damage marks on the underside of the directional gyro show that a sprocket at the end of the torque tube behind the panel was not free to move forward at the time of impact.
Benns conclusion is that when the pilots pull the yoke full aft, as in raising the nose for a stall, the torque tube and the bushing become misaligned with the bracket on the panel. The yoke locks like a screen door that locks open with a sliding washer.
With recovery from the stall difficult, the pilots push forward harder, succeeding only in locking the yoke more firmly and often bending the bracket. The yoke horns are bent when the plane hits the ground and the pilots body is thrown forward.
The potential for the torque tube to lock appears to have been known for some time. In 1979 Piper issued Service Bulletin 661, instructing mechanics to modify the bushing assembly by removing two support plates.
The service bulletin says the modification comes after reports of roughness or binding of the control shaft in the instrument panel cutout hole that can occur when the yoke is full aft and direct forward or upward-and-forward pressure, rather than linear pressure along the control shaft, is applied.
The net result, however, is that the modification was not much of a solution. Benns analysis showed that the yoke would lock if deflected upward about 15 degrees before the modification, and about 17 degrees afterward. While the 2-degree difference was an improvement, the yoke was free to move around more because of the shortened bracket.
In 1984, Piper issued another service bulletin, No. 800, addressing the same problem but delivering a different fix. SB 800 notes that compliance with the earlier SB is mandatory if the new modifications were also made.
Again, the service bulletin said that when either control wheel is full aft and an upward-and-forward pressure is applied, the potential exists that the pilot may be unable to move the control yoke out of the full aft position. The solution this time was to limit the elevator up-travel, thereby preventing the yoke from going as far aft.
Push and Push Again
A simple operational change on the part of the pilot can remove the threat posed by a locked yoke. Because more pressure serves only to tighten the bind, the solution to a locked yoke would be to release pressure, reposition it to improve its alignment, and then apply pressure again. Of course, coming back on the yoke again and wiggling it around is not an intuitive response to a stall and potential spin entry.
The fact that experienced pilots have had many fatal encounters with Tomahawk spins may be tied to the fact that more skilled pilots performing stalls (or those intent on doing everything by the book, as on a check ride) will use the full aft limit of elevator travel. And once the yoke locks into place, experience teaches nothing other than to push harder.
The Tomahawks long record of fatal stall/spin accidents is unique in that virtually all begin at altitude, often during check rides. Stall/spins in other airplanes tend to be in the traffic pattern or during other operations close to the ground.
Spin accidents began shortly after the first production airplanes were delivered, which is perhaps not surprising given the Tomahawks intended use as a primary trainer. However, it wasnt long before Pipers chief engineering test pilot, William Kelly, recommended to the companys chief legal counsel that Piper buy back all of the Tomahawks it had delivered and destroy them.
He felt this was prudent because of the aircrafts handling characteristics and thought it would be cheaper than facing the product liability lawsuits he was sure would follow. Later, he told investigators of an inadvertent flat spin he encountered in a rented Tomahawk with his son on board. Fortunately he was able to recover.
The question of flat spins has haunted the Tomahawk almost since the beginning. FAR 23 directs that it must be impossible to obtain unrecoverable spins with any use of the flight or engine power control, whether at entry into, or during the spin. The regulation also says, There must be no characteristics during the spin, such as excessive rate of rotation or extreme oscillating motion, that might prevent a successful recovery due to disorientation or incapacitation of the pilot.
However, even the Tomahawk Information Manual states: Delay of more than about one to one and a half turns before moving the control wheel forward may result in the aircraft suddenly entering a very fast, steep spin mode which could disorient a pilot.
FAA data show 22 spin accidents with similar summaries: Aircraft stalled (and spun) while practicing slow flight. Instructor demonstrating spins … failed to recover prior to ground impact. (Student pilot) failed to recover from unintentional spin entered during practice stall. Went into a flat spin. No spin training. Not required. Impacted ground during (dual) stall and slow flight maneuvers. Entered flat spin. Flaps down.
With about 1,100 Tomahawks in the U.S. fleet, 22 spin accidents may not seem like much. But considering that the certification rules stipulate that an unrecoverable spin should be impossible, its clear that other dynamics are at play.
Some critics have blamed the airplanes design and construction, saying the experimental prototypes used to certify the airplane have stiffer wings than the production models and incorporate fairings that were later dropped to reduce costs. The NTSB has repeatedly urged a review of the airplane, but the FAA stands by its approval of the aircrafts handling.
The Checkered Past
In 1978, the Swedish and Australian civil aviation authorities found that the Tomahawk did not comply with the certification requirements outlined in FAR 23 after pilots in both countries suffered fatal stall/spin accidents in the Tomahawk.
In 1979, after a spin accident in Romeo, Mich., the NTSB investigator recommended that the Bureau of Technologys Operational Factors Division review the Tomahawk and other T-tailed airplanes built by Piper.
Despite six fatal Tomahawk accidents in the previous six months, the recommendation was denied by NTSBs Office Of Safety Recommendations, which said, There is no evidence, direct or indirect, which would warrant initiating an aerodynamic/certification review of any of Pipers T-tailed airplanes.
In 1981, the chief of flight tests for the FAA Eastern Region wrote the chief of the Engineering and Manufacturing Branch, describing his concern over three fatal spin accidents within the last 60 days. The letter also mentioned a reported flat spin incident and requested AEA-216 to follow up. Yet spin restrictions were never imposed: Nor is there a record of the reported flat spin incident or any follow-up.
In 1993, an apparent flat spin accident killed the flight instructor and permanently paralyzed the student pilot. The airplane belonged to a military flying club, and an Air Force investigator concluded that the wreckage pattern was classic for a flat spin. The NTSB, however, concluded that the pilot was at fault for failure to maintain airspeed.
In 1994, a Tomahawk crashed due to an inadvertent spin during a biennial flight review. Both the instructor and student were killed. The flight instructor was known to be cautious with stalls and spins in the airplane. The NTSBs finding was unintentional spin. A contributing factor was found to be the airplanes lateral-directional characteristics at or near stall.
In 1995, another, almost identical, fatal spin occurred. It too was dual with a new student pilot on a demonstration flight. Based on witness statements, radar data, and the wreckage pattern, both mishaps were carbon copies of the most recent mishap in Illinois.
In 1996, an FAA inspector reported through FAA safety channels that he had encountered a flat spin while conducting a flight instructor flight check. The inspector, with more than 13,500 hours and credentials as an aerobatics competitor and instructor, recovered only after releasing his seat belt and having the student do the same. They leaned forward over the instrument panel to move the cg forward. He recovered finally at less than 1,000 feet agl. Later examination showed the airplane in compliance with Pipers maintenance and weight and balance specifications.
The FAA stood by its conclusion that certification test data showed the PA 38-112 complied with FAR 23 spin requirements. No investigation of the specific airplane and incident was conducted.
In its investigation of the 1994 accident, the NTSB investigator documented problems with the production version of the airplane, including significant differences in the stall characteristics of the conformity prototypes and the production versions.
One senior Piper engineer testified that production airplanes were significantly different, both structurally and aerodynamically, from the experimental prototype. In addition he noted that the conformity prototypes were hand made by experts at Vero Beach, Fla., while the production airplanes were built by unskilled labor at Lock Haven, Pa.
Investigators found that the wings on the production models were built with four ribs instead of the eleven used in the experimental prototype. The wing spar was lightened by use of pressed sheet metal. Wing skin thickness was decreased in production aircraft. This made the wing quite flexible. In addition wing root fairings along with thrust-line and vertical stabilizer off-set – found necessary with the experimental prototype – were removed to reduce production costs.
The most recent Tomahawk accident has raised many eyebrows over the unwillingness of the NTSB and the FAA to review the airplanes flight characteristics. Some charge that the investigations have become politicized to the point where useful safety data is being compromised.
In any case, Tomahawk pilots who have learned to respect the airplanes quirks may have found new reason to fear the airplane whenever it nibbles on the edge of a stall.
Also With This Article
Click here to view “Control Yoke Brackets in Original Configuration and After SB 661.”
-by John Lowery
John Lowery is an aviation safety consultant and a former Air Force and corporate pilot.