NTSB Factual Report

HISTORY OF FLIGHT

On September 14, 2012, about 1219 Pacific daylight time, a Frederick F1, N158CF, was substantially damaged following an in-flight failure of the left elevator while maneuvering in an air race near the Reno Stead Airport (RTS), Reno, Nevada. The airplane was registered to a private individual, and operated by the pilot under the provisions of Title 14 Code of Federal Regulations Part 91 as Race 12. The commercial pilot, the sole occupant of the airplane, was not injured. Visual meteorological conditions prevailed, and no flight plan was filed for the air race flight. The local flight originated from RTS about 1211.

In a written statement to the National Transportation Safety Board (NTSB) investigator-in-charge (IIC), the pilot reported that at 180 knots indicated airspeed, the pace plane released the flight of airplanes to descend and enter the race course. Shortly after, a vibration began at an unknown airspeed, and increased to a heavy flutter event. The pilot stated that he pitched the airplane upwards and reduced airspeed. As the airspeed decreased below 90 knots indicated, the vibration stopped. The pilot subsequently landed uneventfully on runway 14.

PERSONNEL INFORMATION

The pilot, age 57, held a commercial pilot certificate with an airplane single-engine land, airplane multiengine land, and instrument airplane ratings. The pilot was issued a second-class airman medical certificate on March 23, 2012, with no limitations listed.

AIRCRAFT INFORMATION

The two-seat, low-wing, , experimental, amateur built airplane, serial number (S/N) 158, was powered by a Continental IO-550-N engine, serial number 917607, rated at 310 horse power. The airplane was equipped with a Hartzell model HC-J2YF-1BFX adjustable pitch propeller.

The airplane was 21 feet long with a wing span of 24 feet 10 inches. It was equipped with a low cantilever wing, a conventional tail, and fixed conventional landing gear. The airplane was constructed primarily of aluminum alloy, and marketed by Team Rocket LP in the United States. The accident airplane was assembled by the operator between 2008 and 2010. The published never-exceed speed (Vne) for the airplane was 240 KTAS (knots true airspeed).

The airplane was equipped with a Dynon D100 Electronic Flight Information System (EFIS), a Dynon D120 Engine Monitoring System, and a Garmin GPSMAP 396 GPS receiver. The recorded data from all three devices was downloaded, and forwarded to the NTSB Vehicle Recorder Division for review.

The accident airplane was equipped with a video camera mounted behind the pilot looking forward. The video recording was downloaded, and forwarded to the NTSB Vehicle Recorder Division along with the audio recording from the airplane's intercom system for review.

METEOROLOGICAL INFORMATION

A review of recorded data from the Reno-Tahoe International Airport, Reno, Nevada, automated weather observation station, located 12 miles southeast of the accident site, revealed at 1155 conditions were wind calm, visibility 10 statute miles, scattered cloud layer at 14,000 feet, broken cloud layer at 20,000 feet, temperature 27 degrees Celsius, Dew Point -1 degrees Celsius, and an altimeter setting of 30.28 inches of mercury.

WRECKAGE AND IMPACT INFORMATION

Examination of the airplane by the NTSB IIC and an NTSB structures engineer revealed the left elevator upper and lower skins had fractured forward of the trim tab. The trim tab remained attached to the elevator. The rod end clevis at the aft end of the trim tab control rod was fractured. The forward and aft control horn brackets on the lower surface of the trim tab were fractured. The separated control horn and clevis were not recovered. The elevator stops on the control column in the cockpit had no obvious evidence of contact.

The left elevator and attached trim tab were removed from the airplane, and shipped to the NTSB Materials Laboratory in Washington, DC for further examination.

TESTS AND RESEARCH

NTSB Materials Laboratory Examination

Skin

The upper skin portion of the elevator assembly contained a fracture that extended forward from the outboard forward corner location of the trim tab, and a fracture that extended from the inboard forward corner location of the trim tab. The fracture on the outboard forward corner split into two fractures. The upper skin contained another fracture that extended outboard from the outboard area of the trim tab. The lower skin portion of the elevator assembly contained two fractures that extended from the outboard forward corner location of the trim tab. The lower skin of the elevator assembly also contained fractures at the inboard edge. Examination of the skin with the aid of a magnifying glass revealed the skin fracture faces exhibited a luster with no evidence of dark discoloration. The skin fractures were on a slant plane consistent with overstress separation. The thickness of the upper and lower skin each measured approximately 0.023 inch, consistent with the skin thickness specified by the builder of the airplane.

Clevis

The clevis was disassembled from the threaded trim tab rod portion of the elevator assembly. An intact clevis contains a conical base portion and two ear portions that extend beyond the conical base portion. The clevis contains a transition fillet radius between the bottom slot and each ear portion. Binocular microscope examination of the clevis revealed the fracture intersected the transition fillet radius of each ear. The fracture face contained radial fracture features consistent with a fracture origin that emanated from the transition fillet radius of each ear. The fracture face contained a rough texture consistent with overstress separation. Scanning electron microscope examination of the fracture face revealed micro-void coalescence features typical of overstress separation with no evidence of fatigue cracking. According to the airplane builder, the clevis was to be manufactured from a polyamide resin. Fourier transform infrared spectroscopy analysis of the side portion of the clevis produced a spectrum consistent with polyamide resin.

Aft Bracket for the Control Horn

The control horn assembly is made by joining a portion of the forward bracket to the aft bracket. The control horn contains an ear portion in the area where the forward bracket overlaps with the aft bracket portion. The ear portion fractured from the control horn assembly, and was not found. The fractured forward and aft bracket portions remained attached to the lower skin. The aft bracket portion for the control horn was to be attached to the lower skin with four rivets. Detailed examination of the elevator revealed that only three (of four) rivets were installed between the aft bracket and the lower skin. The lower skin contained no evidence of a through hole for the fourth rivet. The surface of the lower skin in area that corresponded to the fractured ear portion of the control horn contained evidence of fretting damage. The fracture face was excised from the aft bracket, and examined using a scanning electron microscope (SEM), which revealed a large portion was obliterated by mechanical damage that resulted from relative movement between the mating fractures. The remaining portion of the fracture contained micro-void coalescence features consistent with overstress separation.

Forward Bracket for the Control Horn

The fracture face for the forward bracket of the control horn was excised. The thickness of the painted forward bracket measured 0.036 inch, consistent with the specified bracket thickness (0.032 inch). SEM examination of the fracture face revealed a large portion was obliterated from mechanical damage that resulted from relative movement between the mating fractures. The remaining portion of the fracture contained micro-void coalescence features consistent with overstress separation.

Threaded Trim Tab Control Rod

The fractured clevis was attached to the threaded trim tab control rod. The threaded rod was analyzed with a hand-held ThermoFisher Scientific model XL3t-980 X-ray fluorescence (XRF) alloy analyzer. The XRF alloy analyzer identified the alloy of the threaded rod as grade 304 alloy steel, consistent with the steel alloy specified by the builder of the airplane.

Airworthiness / Structures

The F-1 Evo Assembly Manual provides guidance to the builder on assembling the trim tab, and attaching the trim tab control horn. The control horn should be manufactured in two pieces, and fastened to the lower inboard surface of the trim tab. The forward control horn bracket should be manufactured from 0.032 inch thick 2024-T3 aluminum and fastened to the trim tab with two rivets. The forward rivet should attach through the skin to the trim tab spar, and the aft rivet should attach through the skin to the root rib. The aft control horn bracket should be manufactured from 0.064 inch thick 2024-T3 aluminum, and fastened to the trim tab with four rivets. All four rivets should attach through the skin to the root rib.

An electric servo is installed in the left elevator to actuate the elevator trim tab. The servo drives an 8-32 threaded rod manufactured of 304 stainless steel. A rod end clevis is installed on the aft end of the rod in order to connect the control rod to the control horn on the trim tab. The rod end should be manufactured from Minlon 22C, a mineral/glass reinforced nylon thermoplastic that has a tensile strength of 17.4 ksi in the dry as molded condition and 10.4 ksi in the 50% relative humidity condition.

Vne is defined as the never exceed speed, or normally the maximum operating speed of the airplane, and generally is that speed, below which, the airplane has been shown to be free from damage, structural failure, or flutter. Vne is normally annotated on the airspeed indicator by a red line. On the accident airplane, Vne is annotated on the EFIS by a red ribbon. The value for Vne is selectable by the operator during setup of the EFIS. The EFIS was set to a Vne of 240 KIAS (knots indicated airspeed) for the accident airplane. The pilot reported that he had previously flown the airplane up to 250 KTAS in preparation for the races, and in accordance with requirements to race the airplane.

The speed generally available to the pilot in the cockpit is that indicated value shown on the airspeed indicator or the EFIS system. Indicated airspeed and true airspeed will only be equal at sea level in standard atmospheric conditions and at low speeds where the compressibility effects will be negligible. As the air density and temperature vary from the standard conditions, the values of indicated and true airspeed will differ. When operating in non-standard conditions where the air density decreases due to changes in altitude or temperature, true airspeed will be much greater than indicated airspeed. The EFIS system can be configured to display the outside air temperature, density altitude, and true airspeed together on the screen.

Using the outside air temperature of 80 degrees F, the maximum indicated airspeed of 236.25 KIAS, and an altitude of 5,775 feet, the true airspeed at the time of the event was calculated to be 266 KTAS.

Recorders

Recovered data from the EFIS unit installed in the airplane recorded a maximum indicated airspeed at 1219:54 of 236.25 KIAS, an indicated altitude of 5,775 feet mean sea level (msl), and an OAT of 80 degrees Fahrenheit as the airplane was descending onto the race course to begin the race. The altimeter setting was 30.28 in Hg as recorded at RNO. Shortly after, the airplane slowed to about 90 KIAS and climbed to 6,581 feet msl by 1220:34 before descending to land.

Recorded video images and audio from inside the accident airplane captured a distinct humming noise about 1219:51; the camera began to shake, and the camera viewing angle moved downward to point at the floor. The video showed shaking and vibration of airplane components, and recorded the humming noise until about 1220:54.

The NTSB Vehicle Recorder Division performed a sound spectrum study on the audio information recorded by the video camera. About 1219:51, there was a distinct change in the spectrogram with multiple new harmonic frequencies becoming evident. The fundamental frequency recorded was 31.8 Hz with multiple harmonics recorded. The most intense frequency was at 63.6 Hz, twice the fundamental. After about 4.5 seconds, the frequency spectrum changed such that the fundamental frequency became 19.1 Hz with the most intense frequency at 57.3 Hz, three times the fundamental.

Using the outside air temperature of 80 degrees F, the maximum indicated airspeed of 236.25 KIAS, and an altitude of 5,775 feet, the true airspeed at the time of the event was 266 KTAS.

NTSB Probable Cause

The pilot’s operation of the airplane above the established never-exceed speed and the incorrect installation of the trim tab control horn, which resulted in the flutter of the left elevator trim tab.