N362TSSeratt Timothy G S-2

On August 23, 2009, about 1305 central daylight time, an experimental, amateur-built Vollenweider/Hurst Van's RV-9A, N362TS, was substantially damaged during a forced landing near Marianna Municipal Airport (MAI), Marianna, Florida. The certificated private pilot/owner was seriously injured. The flight was operated under the provisions of Title 14 Code of Federal Regulations Part 91. Visual meteorological conditions prevailed, and no flight plan was filed for the personal flight.

According to the pilot, he based the airplane at Tri-County Airport (1J0), Bonifay, Florida. The pilot departed 1J0, with no particular destination in mind, and was in a "cruise-climb" passing through an altitude of about 3,500 feet when he heard a "bang." Although the engine continued to run normally, and the pilot could control the engine rpm, he determined that he was no longer obtaining thrust from the propeller, and that he could not control the propeller rpm. When the airplane was approximately 10 miles west of MAI, the pilot reported on the airport's common traffic advisory frequency (CTAF) that he was having engine problems, and that he intended to make a straight in approach for landing on runway 8. He did not switch his transponder code from 1200 to alert air traffic facilities to his problem. When the airplane was approximately 5 miles west of MAI, the pilot realized that he would not be able to reach the airport, and advised an attendant at MAI that he was going to make an off-airport landing. The airplane struck trees, and came to rest in a swamp approximately 3 miles west of MAI.

The MAI attendant notified the operators of an emergency medical services helicopter that was stationed at MAI about the accident, and the operators initiated an airborne search for the accident airplane. They located the airplane about 1 hour after the accident, but due to the terrain conditions, the helicopter was unable to land at the accident site. The helicopter personnel then directed law enforcement personnel on foot to the accident site. When the law enforcement personnel reached the airplane, the pilot was still in the airplane, and he told them that he had no feeling below his waist. A military emergency medical services helicopter from the U.S. Army facility at Ft. Rucker, Alabama was summoned, and the pilot was subsequently evacuated via a basket lowered from the military helicopter.

According to Federal Aviation Administration (FAA) information, the pilot, age 65, held a private pilot certificate with a rating for airplane single engine land, and an experimental aircraft repairman certificate for the accident airplane. His most recent FAA third-class medical certificate was issued in July 2007 with restrictions for corrective lenses for near and distant vision.

According to FAA records, the RV-9A airplane was issued its airworthiness certificate in 2005, when it was owned by the pilot and a co-owner. The airplane was a two-place, all-metal, low-wing design with tricycle-configuration landing gear, and was equipped with an "Eggenfellner 2.5 S" engine, which was a modified Subaru automobile engine. The engine was a four-cylinder liquid-cooled model, with a belt-driven supercharger, and a wastegate that was both manually and automatically controllable. A non-certificated propeller speed reduction unit (PSRU) gearbox was used to drive the propeller at a speed slower than the engine speed. Like the engine, the PSRU was marketed and sold by the Eggenfellner Aviation. The propeller assembly consisted of a Quinti-Avio brand hub, and three electrically-controlled, variable-pitch Warp Drive brand blades.

The 1253 weather observation at MAI included wind from 280 degrees at 9 knots, 10 miles visibility, few clouds at 4,800 feet, scattered clouds at 6,000 feet, temperature 28 degrees C, dew point 16 degrees C, and an altimeter setting of 29.95 inches of mercury.

MAI was equipped with two paved runways, 8/26 and 18/36. Each measured 4,895 feet by 100 feet, and the airport elevation was 109 feet above mean sea level. There was no air traffic control tower at the airport. The radio communications equipment at MAI had no provisions for recording the communications between the station and aircraft on the CTAF frequency.

According to information provided by the FAA inspector who responded to the scene, the accident site was a heavily wooded, swampy area. The inspector stated that the tree damage indicated that the airplane struck the forest canopy approximately 60 feet above the ground, and then traversed only a very short horizontal distance before it descended nearly vertically to the ground. The airplane remained essentially in one piece, and the FAA inspector reported that all debris was located within about 15 feet of the airplane. The airplane sustained crush damage in the aft direction to the left wing and right horizontal stabilizer, and crush damage in the up and aft directions to the cowl, engine, and nose and main landing gear. In addition, the fuselage was fractured across the top, and down both sides, just aft of the cockpit. The engine remained attached to its mount at all attach points, and the hub and PSRU remained attached to the engine. All three composite propeller blades were fracture-separated from the hub. Two blades were separated at the hub, and one was fractured approximately 18 inches from the hub. The propeller and hub assembly was able to be rotated freely by hand, with no discernible resistance from the PSRU or the engine. There was no fire. A handheld Garmin GPSMap 296 global positioning system (GPS) unit was recovered from the wreckage, and retained by the National Transportation Safety Board (NTSB) for data download.

Airplane Glide Performance

A representative of the airplane kit manufacturer (Vans) stated that due to the variability of airplane/engine/propeller combinations, as well as individual variations in airplane and component configurations, the kit manufacturer did not provide kit builders with a template for an RV-9A pilots operating handbook (POH). The kit manufacturer also did not provide kit builders with information regarding glide performance, such as best glide speed, or nominal glide ratio.

Review of exemplar RV-9A POHs found on the internet indicated that while some included glide performance information, the information appeared inconsistent, even within each POH. For example, the "Emergency Procedures" section of one POH listed a "maximum glide" speed of 80 knots, but the glide performance chart in the same POH cited an airspeed of 75 knots.

A non-profit educational organization known as the Comparative Aircraft Flight Efficiency (CAFE) Foundation conducted flight tests, and produced corresponding aircraft performance reports, for a number of amateur-built airplanes. According to the CAFE website, the tests and reports were sponsored and funded by the Experimental Aircraft Association (EAA). The CAFE report for the RV-9A listed a power-off minimum sink rate of 664 feet per minute (fpm) for the conditions tested, which included an airspeed of approximately 82 mph (71 knots). The report cited a best glide ratio of 12 to 1, which was achieved at an airspeed of 95 mph (82 knots).

GPS-Derived Flight Data

The pilot's handheld Garmin GPSMap 296 global positioning system unit was sent to the NTSB recorders laboratory, and data from the accident flight was successfully downloaded. The takeoff roll to the south from 1J0 began about 1251, followed by a climbing gradual turn to the east. About 1258, the GPS-derived ground speed and altitude data both indicated decreases, and subsequent GPS data indicated that the airplane continued to descend and decelerate, which was consistent with a loss of propeller thrust. The peak GPS recorded altitude was 4,049 feet. The GPS data revealed that after the loss of thrust, the airplane remained airborne for about 7 minutes, and traversed a distance of 10.2 miles. The groundspeed decreased steadily at a rate of about 5 mph per minute for the majority of the descent, and the descent rate remained relatively constant at about 620 feet per minute. The initial groundspeed for the stabilized portion of the descent was 109 mph, and the final groundspeed for the stabilized descent was 88 mph. The impact location was 3.15 miles west of the runway 8 threshold at MAI.

At the time of the loss of thrust, the airplane was approximately 12 miles east of the departure airport, and 13 miles west of MAI. Examination of topographic and land-use data within 5 miles to the north, east and south of the gearbox failure location revealed multiple fields with major dimensions of 2,000 feet. Several circular fields approximately 2,700 feet in diameter were also noted. In addition, there was a divided highway that had multiple straight segments over 1 mile long within the 5-mile radius semicircle ahead of the airplane's location at the time of the thrust loss. Review of exemplar RV-9A POH landing performance information indicated that less than 1,000 feet was required for landing.

PSRU Gearbox Information

In a telephone interview with investigators from the NTSB, the owner of Eggenfellner Aviation provided the following background and historical information regarding his company's PSRU gearboxes. The investigation was not able to independently verify some of the information presented below.

The current-production gearbox is the third evolutionary iteration; the three iterations are referred to as "Gen[eration]- I", "Gen-II" and "Gen-III," respectively. The first Gen-I gearbox was delivered in the late 1990s, and approximately 300 to 400 Gen-I units were delivered. The development of the Gen-II gearbox was prompted by the Gen-I gearboxes "making metal," where excessive internal component wear produced metal shavings and/or/particles, and "three to four" gearbox failures, which resulted in the inability of the engine to drive the propeller. The company did not track or maintain a database of the failure modes, gearbox hours, installed propellers, or any other parameters associated with the failed gearboxes. Pending gearbox failures provided "plenty of warning" to the pilots/owners prior to actual failure, since the gearbox oil temperature would be "10 to15 degrees" F higher than normal operating temperature and the gearbox installation manual specified the installation of an oil temperature gauge.

The life expectancy of the Gen-I gearboxes could be as high as 1,200 hours when coupled with "a lightweight propeller," but no guidance that related propeller model to weight categories was available in the company documentation. The company owner stated that the accident propeller was "somewhere in the middle" of the weight range of propellers typically used with the gearbox. Prior to the development of the Gen-II and Gen-III gearboxes, "any propellers" were acceptable for installation; the company did not recommend or restrict which propellers could be utilized with the gearboxes. However, subsequent to the Gen-I gearboxes, the company prohibited the use of metal propellers.

The Gen-II gearbox was essentially the same as the Gen-I gearbox. It had the same housing and gears as the Gen-I, but was equipped with different seals and bearings.

The introduction of the Gen-II gearboxes was a "slow process;" initially the modifications were propagated when the owners had leaks or other problems with the gearboxes, and the gearboxes were retrofitted with the new components. At an unspecified date, the company changed the color of part of the Gen-II cases (from silver/aluminum to blue) to differentiate them from the original gearboxes. According to March 2005 documentation on the company's website, the company "strives to improve our products through continuous testing and refinement," and offered the Gen-II upgrade "that further reduces vibration and extends gearbox performance and reliability." The documentation enumerated several of the "major features of the PSRU GEN-II upgrade."

Initially after the first Gen-II gearboxes were developed, there was no company mandate for Gen-I gearbox owners to upgrade to those gearboxes. When an individual purchased a gearbox from the company, s/he was to sign a "purchase agreement" in the presence of a Notary, in which the purchaser agreed to monitor the company's website and participate in a designated internet-based forum in order to ensure that each purchaser had the most "up to date information" regarding any safety information about the gearboxes.

The company owner stated that it was "very logical" for gearbox purchasers to "want to know" such safety information, and that it was "irresponsible" for any gearbox purchasers not to remain apprised of such information. When certain unspecified safety issues were noted by the company, the company also mailed hardcopy notifications to the gearbox purchasers listed in the company records. For cases where individuals purchased their gearboxes from other individuals instead of directly from the company (e.g., as in the case of a used gearbox), the company relied on the strategy of "requiring" the new purchasers to provide their contact information to the company in order to ensure that they, as the new owners of such gearboxes, receive hardcopy information directly from the company. Corresponding guidance was explicitly stated in the company purchase agreement.

The Gen-III gearboxes were introduced in about 2006, and the cases were red-anodized to differentiate them from the two previous versions. A company Service Bulletin, which was first published in May 2007, noted that the "upgrade" to the Gen-III gearbox was "mandatory" as of January 2008. The service bulletin stated that it was the "official policy of Eggenfellner Aircraft Inc. to upgrade, rather than to service, older revisions of gearboxes. This assures that our customers are flying with the latest and most reliable gearbox available at the time." The service bulletin also stated that "We make no claims as to the expected lifecycle of any gearbox" and "At any time, we reserve the right to declare an upgrade mandatory if, at our sole discretion, we determine that the latest available gearbox is significantly improved and would result in a safer product." The service bulletin also stated that "No gearbox failures have resulted in injury." Finally, the Gen-III service bulletin stated that "It is imperative that the input drive shaft from your engine be perfectly aligned with the gearbox input shaft socket. Even a tiny misalignment will result in undesirable side-loads on the input shaft bearings and ultimately premature seal and bearing wear."

When the Gen-III gearboxes were introduced, the company website was modified to alert owners to the fact that they should not continue to utilize the two previous models. In late November 2007, the company owner's postings on the company's internet user forum included the following: "All [Gen-I and Gen-II] drives are to be grounded at the end of 2008 and replaced with a [Gen-III] unit," and he advised Gen-I and Gen-II owners to "Drain and check the oil on 25 hour intervals to verify that the internals are clean" until they replaced their gearboxes. He also stated that he "cannot, in good faith, allow anyone to use a drive that we now have a superior replacement for." According to the company owner, as an inducement to owners to remove previous gearbox versions from service, the company made the Gen-III gearboxes available "at cost" to Gen-I and Gen-II owners who returned their old gearboxes to the company.

Engine and PSRU Examination

No identification markings or placards were found on the engine, supercharger or PSRU. The PSRU case was constructed of aluminum, and was natural aluminum in color. Examination of the owner's engine purchase and installation documentation, and comparison of the accident hardware with information on the Eggenfellner Aviation website, enabled confirmation that the PSRU was a Gen-I version of the Eggenfellner PSRU.

Several days after the accident, the oil was drained from the PSRU, and the propeller hub and front cover of the PSRU were removed to examine the internal PSRU components. The PSRU contained four primary gears; an engine shaft output gear, two gears machined on a common shaft, and a propeller input gear. The gear geometry was such that the engine output gear drove the aft gear on the two-gear shaft, and the forward gear on that shaft then drove the propeller input gear. The forward gear on the two-gear shaft was found to be completely devoid of all teeth. The PSRU case contained a handful-sized mound of metal flakes that ranged from about 1/16 to 1 inch in length and width.

Accident Pilot-Provided Information

According to the accident pilot, he was originally the co-builder/co-owner of the airplane. He stated that the engine, supercharger, and Gen-I PSRU gearbox were purchased as a unit from Eggenfellner Aircraft, and were installed on the airplane without any additional modifications. Although the co-builders knew that their airplane could have accommodated the installation of certain certificated aircraft engines, they opted for the Eggenfellner engine because it seemed to be "the best option." Documentation provided by the pilot indicated that he had paid the company approximately twenty six thousand dollars for the original engine, gearbox and instrumentation. In June 2008, after the airplane was completed, the accident pilot became the sole owner of the airplane.

The pilot was aware that Eggenfellner Aircraft had revised the gearbox design two times, but he did not know what modifications and/or improvements were incorporated into the Gen-II or Gen-III gearboxes, and he was not aware of the rationale behind, or the reasons for, the two gearbox revisions. According to the pilot, a new Gen-III gearbox cost approximately six thousand dollars. In addition, the pilot understood from various discussions with other builders, owners and pilots who used the company's gearboxes, that a Gen-I gearbox life expectancy was approximately 1,000 hours. The combination of the low amount of time on his Gen-I gearbox and the cost of an upgrade to a Gen-III gearbox caused the pilot to decide not to replace his Gen-I gearbox with the revised version. At the time of the accident, the failed gearbox had accumulated less than 200 hours in service.

The pilot stated that he visually monitored the condition (color and appearance) of the gearbox oil, and that he did not change it until it had accumulated approximately 157 hours in service, since until then the oil did not appear to be "dirty." The pilot recalled that the recommended interval time for changing gearbox oil was 100 hours. He did not submit the used gearbox oil for any particle or other formal condition analysis. At the time of the accident, the new gearbox oil had accumulated only 15 to 20 hours in service.

The accident airplane was equipped with a gearbox oil temperature gauge in the cockpit. Neither the cockpit nor the gearbox contained any other gearbox condition/status instrumentation. The pilot reported that he did not observe any abnormal gearbox oil temperature indications prior the accident. Subsequent to the accident, the pilot learned that this gearbox failure was reportedly the sixth catastrophic failure of an Eggenfellner reduction gearbox.