N724WDDonohoe Terence E VANS RV-7A · 2006

On December 10, 2011, about 1258 mountain standard time, an experimental, amateur-built Van's RV7-A, N724WD, was substantially damaged when it impacted terrain near Surprise, Arizona, shortly after the pilot reported a problem in cruise flight. The certificated private pilot, who was also the builder and owner of the airplane, was fatally injured. The personal 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 flight.

The pilot based the airplane at Glendale Municipal Airport (GEU), Glendale, Arizona, in a hangar that he shared with two other RV owners. According to information provided by Federal Aviation Administration (FAA) personnel, the airplane departed GEU about 1226. Luke Air Force Base (LUF), Glendale, was situated about 5 miles west of GEU. Review of LUF ground tracking radar data revealed that the entire flight was captured by the radar system. The LUF radar first acquired the airplane at 1227:27, as it was climbing through an altitude of 1,300 feet above mean sea level (msl), and broadcasting a beacon code of 1200 on its transponder. The airplane flew about 25 miles to the northwest of GEU, maneuvered for a short time, and then began tracking back towards GEU. The flight altitude varied irregularly between about 2,700 and 3,300 feet msl.

About 1255:50, when it was about 12 miles northwest of GEU, the airplane began decelerating from its 100-knot ground speed. The airplane slowed to about 60 knots and began a descent. About 1256:25, when the airplane was at an altitude of 2,500 feet, the first 7700 beacon code return was received. The pilot transmitted a "mayday" call to LUF Approach about 25 seconds later, and stated that he was having a problem and was "going down." No further communications from the airplane were received, and the final radar return was received at 1257:21.

A flight instructor and a student who were flying in the same vicinity, and operating on the same radio frequency as the accident airplane, heard the pilot's radio communications. No ELT signals were obtained by any aircraft or facilities. The instructor and student were able to visually locate the wreckage, and guided a first responder helicopter to the site about 45 minutes after the accident.

The wreckage was located about 800 feet north of the final radar return, on flat terrain, at an elevation of 1,330 feet. Examination of the wreckage revealed ground scars and damage consistent with a nose-down impact, with little or no horizontal velocity. All aerodynamic surfaces and flight controls were located in the wreckage. Fuel spillage from the right wing tank was observed, but there was no fire. A handheld Garmin GPSMap 296 unit was recovered from the wreckage, and retained for data download.

According to FAA records, the pilot held a private pilot certificate with a single-engine land rating. His most recent FAA third-class medical certificate was issued in July 2010, and his most recent flight review was completed in January 2011.

Review of the pilot's flight logbook indicated a total flight experience of about 713 hours. The pilot conducted the first flight and all the FAA Stage 1 testing of the airplane. He accrued about 340 hours in the airplane, over a period of 5 years and 2 months. The logbook annotations only contained one explicit reference to aerodynamic stalls; that reference was dated November 24, 2006. At that time, the pilot had about 21 hours in the airplane. The pilot's flight logbook contained multiple references to mechanical issues that were not included in the airplane maintenance records.

Interviews and/or communications with persons familiar with the pilot's flying habits typically described him as a "straight and level flyer" who did not like stalls, and who did not do aerobatics or other more aggressive style flying.

The FAA Civil Aeromedical Institute conducted forensic toxicology examinations on specimens from the pilot, and reported that no carbon monoxide, cyanide, ethanol, or any screened drugs were detected. According to the Maricopa County Arizona Office of the Medical Examiner, the cause of death was "blunt impact injuries."

FAA and pilot records indicated that the airplane was built in 2006 by the pilot. Examination of the pilot's airframe maintenance records did not reveal any significant problems with, or maintenance on, the airframe. A hangar mate of the pilot was not aware of any significant or unusual problems with the airplane.

The airplane was powered by an Eggenfellner conversion of a Subaru EJ-25 series automobile engine. The engine was a 4 cylinder, liquid cooled unit with an electric fuel pump, electronic ignition, and electronic fuel metering. These items were monitored and controlled by the engine control module (ECM), frequently referred to as the engine control unit (ECU). Examination of the pilot's engine maintenance records did not reveal any significant problems with, or maintenance on, the engine. No records containing any age or time in service information for the engine prior to its installation in the airplane were located.

The engine was equipped with Eggenfellner "Generation III" propeller speed reduction unit (PSRU) that reduced the engine output rpm values to values suited to the propeller. Examination of the pilot's engine maintenance records did not reveal any significant problems with the PSRU.

The engine was equipped with a Quinti-Avio propeller hub which provided for electrically-controlled, electrically-driven, in-flight pitch adjustment of the three Warp Drive composite propeller blades. Examination of the pilot's propeller maintenance records did not reveal any significant problems with, or maintenance on, the hub or propeller.

The GEU 1247 automated weather observation included wind from 080 degrees at 5 knots, visibility 20 miles, broken cloud layer at 20,000 feet, temperature 19 degrees C, dew point -7 degrees C, and an altimeter setting of 30.10 inches of mercury.

FAA and pilot records indicated that the airplane was built in 2006 by the pilot. Examination of the pilot's airframe maintenance records did not reveal any significant problems with, or maintenance on, the airframe. A hangar mate of the pilot was not aware of any significant or unusual problems with the airplane.

The airplane was powered by an Eggenfellner conversion of a Subaru EJ-25 series automobile engine. The engine was a 4 cylinder, liquid cooled unit with an electric fuel pump, electronic ignition, and electronic fuel metering. These items were monitored and controlled by the engine control module (ECM), frequently referred to as the engine control unit (ECU). Examination of the pilot's engine maintenance records did not reveal any significant problems with, or maintenance on, the engine. No records containing any age or time in service information for the engine prior to its installation in the airplane were located.

The engine was equipped with Eggenfellner "Generation III" propeller speed reduction unit (PSRU) that reduced the engine output rpm values to values suited to the propeller. Examination of the pilot's engine maintenance records did not reveal any significant problems with the PSRU.

The engine was equipped with a Quinti-Avio propeller hub which provided for electrically-controlled, electrically-driven, in-flight pitch adjustment of the three Warp Drive composite propeller blades. Examination of the pilot's propeller maintenance records did not reveal any significant problems with, or maintenance on, the hub or propeller.

The airplane came to rest upright on level hard desert terrain with a moderate cover of bushy vegetation. The wreckage was very tightly contained, and only a few, non-structural, components (including cockpit canopy and transparency fragments, some cockpit items, and one propeller blade) were separated from the airplane. Ground scarring was consistent with the airplane remaining in essentially the same location that it struck the terrain.

The canopy was impact damaged, and found just forward of the wreckage. The nose was severely crushed in the up and aft direction. There was no rotational scoring damage to the propeller hub or to the two blades that remained attached to the hub. The third blade was fractured in multiple pieces but did not display any significant rotational scoring. All three propeller blades were determined to have been at a similar pitch setting at the time of impact. Visual examination of the engine and PSRU, which included separation of the PSRU case, did not reveal any anomalies that were consistent with pre-impact damage and that would have prevented normal operation. The PSRU gears were undamaged, and the PSRU internal oil was free of metal particles or other debris. Continuity was established from the propeller hub to the engine.

All three fixed landing gear were deformed up and aft. The cockpit/cabin area was severely deformed, and fuselage fractures were observed at the firewall and immediately aft of the cabin aft bulkhead. The cockpit/cabin longitudinal axis faced approximately 248 degrees magnetic, while the aft fuselage axis was oriented about 36 degrees tail-right relative to the forward section.

The leading edges of both wings exhibited crush damage up and aft across their full spans. The wing and nose damage was consistent with an impact angle of approximately 50 degrees nose down. Both ailerons remained securely attached to their respective wings. Both electrically-driven flaps remained attached to the wings; the right flap control link was fractured. Flap and flap actuator positions were consistent with the flaps being in the retracted position at the time of impact. Damage and witness marks indicated contact and relative motion between the right wing and flap inboard edges, and the right fuselage sidewall. However, there were no photographs of that area of the fuselage prior to the removal of the pilot. The right rear spar was fracture-separated from its fuselage attach point. Both the left and right rear spars and fuselage attach points were excised from the airplane and sent to the NTSB Materials Laboratory in Washington DC for examination and analysis. No indications of pre-impact failures were observed.

The horizontal and vertical stabilizers remained attached to the fuselage. The elevators and rudder remained attached to their respective stabilizers. The elevator trim tab remained attached to the left elevator. The reinforced nylon clevis that attached the elevator trim tab to the electric elevator trim servo was fractured, which disconnected the tab from the servo. The elevator servo and clevis segments were sent to the NTSB Materials Laboratory in Washington DC for examination and analysis. The laboratory examination determined that the clevis failure was due to overstress, and no material deficiencies were noted.

Three of the four rivets that secured the left fuselage-mounted rudder stop were sheared, which allowed the rudder stop to pivot from its normal position. That permitted rudder overtravel to the left, which enabled the rudder to contact the left elevator. Damage and witness marks indicated that contact had occurred between the left side of the rudder, and the trailing and inboard edges of the left elevator and elevator trim tab. Manipulation of the control surfaces indicated that some of that contact could not have occurred unless the elevator trim tab was displaced well beyond its normal trailing edge up travel range.

Exclusive of impact damage, control continuity for the flaps, ailerons, rudder, and elevator was established. The investigation was unable to determine the pre-impact condition or continuity of the elevator trim system.

The airplane was equipped with a Trio Avionics brand "EZ Pilot" single-axis autopilot that drove an aileron servo. The "power" toggle switch for the autopilot was found in the down/off position. The airplane was not equipped with either an angle of attack or a stall warning system.

The ELT had separated from its mounting and retaining mechanism, and was found in multiple fragments inside the aft fuselage. No indications of pre-or post-impact fire were observed.

Airplane Flight Profile

Data from the LUF ground tracking radar and the onboard GPS unit was used to reconstruct the profile for the entire flight. The overall ground track was meandering, consistent with the airplane being hand flown, instead of being guided by the autopilot.

After the turnaround at the northwest-most extent of the flight, the airplane proceeded in an essentially straight track to the southeast, back towards GEU. Between about 1255:50 and 1256:15, the groundspeed decreased from about 100 knots to about 60 knots. Between about1256:15 and 1257:15, the airplane began a steady descent from about 2,800 feet down to about 2,000 feet msl. About 1256:25, the transponder code switched to 7700, when the groundspeed was about 62 knots, and the airplane was at an altitude of about 2,700 feet. The 1256:53 "mayday" transmission began when the groundspeed was about 50 knots, at an altitude of about 2,300 feet msl. The ground speed continued to decrease for the remainder of the flight. The ground track began a turn to the south about 1256:55. About 1257:15, the descent rate increased to over 2,000 fpm, and the data ended a few seconds later. Local terrain elevation was approximately 1,300 feet msl.

Airspeed and Stall Information

Since the airplane was experimental amateur-built, flight testing of the completed airplane was required to determine airspeed system calibration (and therefore accuracy) and actual stall speed values. The Vans Construction Manual enumerated the requirements and methods for the airspeed system and stall speed testing.

Comparisons of the airplane's airspeed indicator (ASI) arcs and the Vans published values revealed that while most of the arcs were per the Vans values, the ASI stall speed values were not. The ASI full-flaps stall speed was 48 knots, instead of Vans value of 50 knots. The ASI no-flaps stall speed was 50 knots, instead of Vans value of 56 knots. No records of the airspeed system calibration, or stall speed testing or results were located for the investigation, and therefore the accuracy of the airplane airspeed indication system, the ASI stall speed markings, and actual stall speed values could not be determined by the investigation.

According to the Vans Construction Manual "The ideal is that when a stall is encountered, the nose tends to lower, or can easily be lowered by an easing of stick back pressure or by a forward stick pressure. In most RVs, there is little advance stall warning in the form of pre-stall buffet. The buffet which does occur does so within just a mph or two of the fully developed stall."

The airplane was not equipped with an angle of attack or stall warning system.

Engine Electrical Information

Eggenfellner Aircraft ceased production and support of the Subaru conversions about 2010. The first page of Chapter 6 (Electrical System Installation) of the Eggenfellner Installation Manual (IM) contained the following text in red typeface:

"READ AND UNDERSTAND [Boldfaced type in original] - Your engine requires a constant and stable source of electricity to drive its fuel injection, fuel pumps, and the engine control computer....[T]he Subaru cannot tolerate a loss of electrical power. For this reason, we have designed fully redundant electrical and fuel systems with provisions for automatic fault management. It is imperative that you adhere to this design for your own safety and to assure ongoing maintainability. Eggenfellner Aircraft Inc. cannot be responsible for endorsing or supporting builders who deviate from our design, nor be responsible for any direct or indirect damages which may result from such deviations."

The IM also stated that the "... engine uses a small, rugged, computer to control ignition, timing, and a wide variety of sensors. This computer is known as the Engine Control Module or ECM. It is also commonly referred to as the Engine Control Unit or ECU" and "The ECM must be mounted inside your cabin, away from sources of heat and vibration."

The wiring block- and detail- diagrams for the engine installation depicted three master switches which controlled power to three buses: Engine, Main, and Avionics.

The accident airplane was equipped with three explicitly-labeled master switches, "Master," "Avionics Master," and "Bus Master". The first two were two-position rocker switches located with nine other similar switches in a multi-switch panel typical of general aviation airplanes. The "Bus Master" was a 3-position (ON, OFF, OVERRIDE) toggle switch located a few inches above the other switch array.

Although no wiring diagrams for the airplane were located, comparison of the airplane switch configuration with Eggenfellner and other documentation indicated that the "Bus Master" switch likely controlled electrical power to the engine as follows: The "ON" position powered the Engine Bus via the Main master switch, the OFF position isolated the Engine Bus from electrical power, and the OVERRIDE position powered the Engine Bus independent of the Main master switch setting.

The day after the accident, the Main and Avionics master switches were found to be in their off positions, but the Bus Master was found to be in the OVERRIDE position. The investigation was unable to determine whether any first responders reset any of those switch positions.

The pilot's normal and emergency procedures or procedure checklists for the airplane were not recovered.

The flight departed from GEU, which was equipped with an air traffic control tower (ATCT) that was operating at the time of departure, and for the duration of the flight. However, after takeoff, the airplane departed the local area, and therefore, the pilot was not required to communicate again with the ATCT until he returned to the area. Since the accident occurred prior to the pilot re-contacting the GEU ATCT, no copies of GEU ATCT communications with the airplane were requested or obtained by the investigation.

The bulk of the flight was conducted in the airspace designated as the "Luke Air Force Base Special Air Traffic Rule (SATR) North Area." The SATR was defined by 14CFR 93.176, and 93.177 designated the operating rules. The SATR was depicted on the applicable aeronautical navigation charts. The SATR was not active at the time of the flight, and therefore, the pilot was not required to be, and was not initially, in communication with LUF Approach.

According to ground tracking radar data obtained from LUF, the pilot squawked 1200 on his transponder for almost the entire flight. At 1256:25, the data indicates that the pilot switched his transponder to 7700, the emergency code. At 1256:53, the pilot made his first call to LUF Approach with the transmission "Luke approach, Luke approach, mayday, mayday 724WD right by the substation, I have a (uncertain word) that's out of control." That transmission lasted 15 seconds. The uncertain word sounded variously like "flap" or "craft" or "prop." At 1957:00, the LUF approach controller assigned the airplane a discrete transponder code and requested the pilot's intentions. At 1957:05, the pilot transmitted five phrases regarding the fact that he was "going down," and no further transmissions were received from the airplane. The recorded communications from the airplane were generally clear, and were free from background noise.

The NTSB investigator-in-charge was unable to obtain or listen to the radio communications recordings, or obtain the radar data, until 4 days after the accident, which was after the on-scene examinations had been completed.