N902ATBeech A36 · 1991

On October 5, 2022, about 1356 eastern daylight time, a Beech A36, N902AT, was destroyed when it was involved in an accident near Arundel, Maine. The private pilot and passenger were fatally injured. The airplane was operated as a Title 14 Code of Federal Regulations Part 91 business flight.

According to a family member and officer of the pilot’s construction company, the pilot and passenger were returning from a business trip to Presque Isle, Maine. Automatic dependent surveillance - broadcast (ADS-B) data revealed that the airplane departed Presque Isle International Airport (PQI) about 1220 for Biddeford Municipal Airport (B19), Biddeford, Maine, on an instrument flight rules (IFR) flight plan.

Weather radar imagery superimposed over the airplane’s flight track indicated that the airplane was flying toward an area of light intensity echoes associated with light rain to drizzle and instrument meteorological conditions as it approached the area of the destination airport. An AIRMET for these conditions and low-level turbulence was valid for the accident time.

Air traffic control communication information revealed that, at 1340:03, the controller instructed the pilot to descend and maintain 3,000 ft. The pilot confirmed to the controller that he had obtained the most recent weather information and requested the RNAV (GPS) RWY 6 instrument approach procedure at B19.

Figure 1 - Profile View of RNAV (GPS) Runway 6 Approach, Biddeford Muni (B19). (NOAA)

At 1340:56, the controller advised the pilot to fly heading, “two five zero… vectors to final” which the pilot acknowledged. The airplane’s ground track approximated a left downwind leg for the runway 6 traffic pattern at B19. The pilot was subsequently vectored through a 90° left turn, and at 1349:29, the controller advised the pilot, “Two miles from DEXXY [the initial approach fix], turn left to 090°, maintain 2,300 (ft) until established on the final approach course, cleared RNAV 6 approach Biddeford.” The airplane’s ground speed was about 125 knots (kts) at that time.

Eleven seconds later, the pilot replied, “Turn… um… repeat that please.” The controller repeated the instructions and the approach clearance, and the pilot replied, “Cleared for the approach, left 090, and, um, I guess that’s it.”

The airplane flew on an approximate 90° ground track through the final approach course, then turned left, and continued about parallel to the final approach course. The airplane passed abeam the initial approach fix at 109 kts ground speed at an altitude of about 2,100 ft, 200 ft below the minimum descent altitude for that segment of the approach.

At 1351:32, the controller issued a radio frequency change and offered options for the pilot to cancel his IFR flight plan in the air or on the ground. The pilot accepted a telephone number to call after landing and advised the controller that he was changing to the B19 common traffic advisory frequency at 1352:01. At that time, the airplane was about 2,100 ft at 110 kts groundspeed.

The airplane passed the final approach fix at 550 ft, which was 750 ft below the minimum descent altitude for that segment of the approach and had slowed to 58 kts ground speed. The airplane continued to descend over the next .75 miles at a ground speed about 60 kts before the data ended in the vicinity of the accident site.

Figure 2 - View of Approach Track of Accident Flight with Altitudes at IAF and FAF (NTSB)

Several witnesses below the airplane’s flight path reported hearing the airplane but stated that they could not see the airplane due to the low clouds, rain, and fog at the time of the accident. One witness about .5 miles from the accident site said that he was drawn to the airplane’s sound and saw it briefly before it disappeared behind trees. He said that he did not hear the airplane contact trees or terrain, but “felt the ground shake.” He said that the sound of the engine was “not the normal rhythm of a piston engine… it would sputter and die out.” Another witness said that he was, “inside my shop when I heard it coming very low. It didn’t sound good. It wasn’t sputtering, but it didn’t sound good. It was steady, but it didn’t sound healthy to me.”

A third witness located below the airplane’s flight path described the sound of the engine as “low, loud, and steady. It sounded [as] usual, just lower and louder.” He said he ran to look out his bay window in the direction of travel but could not see the airplane because “the clouds were too low. Visibility was not good.”

The pilot held a commercial pilot certificate with ratings for airplane single-engine land and instrument airplane. His Federal Aviation Administration (FAA) third-class medical certificate was issued August 2, 2021. Examination of the pilot’s most recent logbook revealed that he had accrued 2,514 total hours of flight experience, of which 977 hours were in the accident airplane make and model.

The pilot completed a flight review on February 22, 2021.

According to FAA and maintenance records, the airplane was manufactured in 1991 and was powered by a Continental IO-550-B, 300-horsepower engine. The airplane’s most recent annual inspection was completed November 1, 2021, at 3,953.2 total aircraft hours.

Airworthiness documentation indicated that dual Garmin GI 275 electronic flight instruments (Attitude/Attitude Direction Indicator) were installed in the airplane on February 4, 2022. (See figure 3.) The FAA Form 337 filed for the installation stated, “Installed Dual Garmin GI275 with internal ADAHARS as Directional Gyro and GI275 with Autopilot interface as Attitude Indicator, and GI275 as Copilot ADI…per STC SA02658SE.” According to a flight instructor who flew with the pilot following installation of the GI275s, the system installed in the accident airplane integrated with the airplane’s S-TEC 55X autopilot.

The pilot logged 33.8 hours of experience in the accident airplane between the installation of the Garmin GI 275 displays and the accident flight, of which 7.2 hours was in actual instrument meteorological conditions. He logged 2 hours of dual flight instruction received on February 28, 2022. In the notes section, he annotated “familiarization” with the new avionics, two RNAV approaches flown, and a “go around” performed. There were no further dual-instruction-received entries annotated.

Figure 3 – Garmin GI 275 Attitude Indicator [AI/ADI] (Garmin.com)

At 1356, the weather reported at Sanford Seacoast Regional Airport (SFM), 11 miles west of B19, included wind from 010° at 7 knots, 2 1/2 miles visibility in light rain and mist, ceiling broken at 700 ft above ground level (agl), overcast at 1,000 ft agl, temperature 13°C, dew point temperature 12°C, and an altimeter setting of 29.97 inches of mercury (inHg).

The next closest weather reporting location was Portland International Jetport (PWM), Portland, Maine, located 13 miles northwest of B19 at an elevation of 76 ft and directly along the airplane’s route of flight. At the time of the accident, the reported conditions included wind from 030° at 8 knots, 10 miles or more visibility, overcast ceiling at 1,400 ft agl, temperature 13°C, dew point temperature 12°C, and an altimeter setting of 29.97 inHg.

The pilot obtained an online weather briefing that contained both text and graphic depictions of AIRMETs for IFR, mountain obscuration, icing, and turbulence in the area of his route of flight and that of the destination airport.

The terminal area forecast (TAF) available prior to the airplane’s departure from KPQI was issued at 0720 and expected marginal visual flight rules conditions to prevail during the period with winds from 030 at 6 knots, visibility 6 miles or more in light rain, with a broken ceiling at 1,500 ft agl. The next scheduled TAF was issued at 1339, or 17 minutes before the accident and forecasted IFR conditions with visibility 2 miles in light rain and mist, with a ceiling overcast at 1,200 ft agl.

According to FAA and maintenance records, the airplane was manufactured in 1991 and was powered by a Continental IO-550-B, 300-horsepower engine. The airplane’s most recent annual inspection was completed November 1, 2021, at 3,953.2 total aircraft hours.

Airworthiness documentation indicated that dual Garmin GI 275 electronic flight instruments (Attitude/Attitude Direction Indicator) were installed in the airplane on February 4, 2022. (See figure 3.) The FAA Form 337 filed for the installation stated, “Installed Dual Garmin GI275 with internal ADAHARS as Directional Gyro and GI275 with Autopilot interface as Attitude Indicator, and GI275 as Copilot ADI…per STC SA02658SE.” According to a flight instructor who flew with the pilot following installation of the GI275s, the system installed in the accident airplane integrated with the airplane’s S-TEC 55X autopilot.

The pilot logged 33.8 hours of experience in the accident airplane between the installation of the Garmin GI 275 displays and the accident flight, of which 7.2 hours was in actual instrument meteorological conditions. He logged 2 hours of dual flight instruction received on February 28, 2022. In the notes section, he annotated “familiarization” with the new avionics, two RNAV approaches flown, and a “go around” performed. There were no further dual-instruction-received entries annotated.

Figure 3 – Garmin GI 275 Attitude Indicator [AI/ADI] (Garmin.com)

Local emergency services personnel located the airplane’s fuselage in hilly, wooded terrain about 1 hour after the accident, and the wreckage was examined at the accident site the following day. The initial impact point was in a tree about 40 ft above the ground. The wreckage path was oriented 050° and was about 150 ft in length; all major components of the airplane were accounted for at the scene. The fuselage came to rest upright with the engine and propeller still attached. The propeller blades displayed similar twisting, bending, and chordwise scratching.

The cockpit, cabin area, and nearly all the empennage, were consumed by a post-crash fire. The avionics were destroyed, and no information was available to determine the modes or settings of the avionics and/or autopilot during the approach. The tail section remained largely intact. The wings displayed impact damage consistent with collision with trees and terrain. Flight control cable continuity was confirmed from the cockpit to the flight control surfaces through several breaks consistent with impact and thermal damage.

The engine was separated from the airframe. The propeller was rotated by hand and continuity was established through the powertrain to the valvetrain and the accessory section. Compression was attained using the thumb method. The magnetos were secure in their mounts and produced spark at all terminal leads when the crankshaft was rotated.

The engine was examined, and a test run was performed at the manufacturer’s facility.

The throttle body was not shipped with the engine from the recovery facility, so an exemplar throttle body was attached to the engine for the engine run. The engine-driven fuel pump remained attached to the engine and exhibited thermal damage. The fuel pump was removed, and bench tested for functionality. The pump operated on the test bench and was reinstalled on the engine before the test run was conducted. The damaged propeller was removed, and a club prop was installed for use in the test cell.

The following items were removed from the engine to facilitate an engine run:

• The alternator was impact damaged, removed, and a plate installed where the alternator attached to the engine.

• The engine baffling was removed.

• The starter exhibited thermal damage and was removed and replaced by an exemplar unit.

• The ignition harness exhibited damage and was replaced with an exemplar unit.

• All the air intakes on the engine were impact damaged, these were removed and replaced with exemplar intakes.

• The oil filter exhibited thermal damage and was replaced with an exemplar filter.

• The exhaust was impact damaged and replaced with straight pipe exhaust.

• The propeller governor was removed and replaced with a test cell plate.

• The tachometer drive was removed, and the opening covered with a plate.

• The vacuum pump was not shipped with the engine and the position on the crankcase was plated over.

• The pushrod and pushrod housing for the No. 5 cylinder exhaust valve were impact damaged and replaced with exemplar units.

The oil cooler leaked at engine start due to impact damage and was replaced with an exemplar unit.

An engine test run was then performed, and the engine accelerated smoothly from idle to a maximum of about 2,550 rpm (maximum-rated output was 2,700 rpm) with the throttle fully open. During the engine test run, the engine idled at about 500 rpm. The engine was then accelerated to 1,800 rpm and ran smoothly at that power setting for about 3 minutes. It was then advanced to about 2,100 rpm, which would be consistent with a power setting used for an instrument approach procedure. The engine ran smoothly at that power setting for about 4 minutes. The engine ran at maximum rpm for about 3 minutes. A magneto check was performed at 2,100 rpm; when switched to only the left magneto, a loss of 124 rpm was observed. When switched to only the right magneto, a loss of 135 rpm was observed.

During the test run, the No. 6 cylinder exhaust gas temperature (EGT) was "low," possibly due to an inoperative test cell EGT probe. The cylinder head temperature was on the "low side" of the normal operating range.

The fuel flow rate was low at full throttle, which was consistent with the lower rpm achieved at that power setting. The altitude-compensating engine-driven fuel pump was disassembled; the aneroid bellows were punctured by impact damage and therefore fully expanded, which was consistent with the low fuel flow.

According to the pilot’s son-in-law and business associate, the pilot went out to “practice instrument approaches” on the day before the accident flight, which was his custom. The pilot did not record the flight in his logbook.

ADS-B and weather data revealed that the airplane departed B19 the day before the accident in visual flight rules conditions and flew a series of left and right turns in the area southwest of B19 before maneuvering toward the initial approach fix for the RNAV (GPS) Runway 6 approach at B19. An ADS-B ground track depiction of the flight revealed only one track at the end of the flight that approximated the instrument approach procedure.

Once the airplane had reversed course and aligned with the inbound course of the approach, the data depicted an unstabilized approach with significant divergences in both airspeed and altitude, which was inconsistent with the use of the autopilot in a rate/descent mode.

Figure 5 – Profile View of Practice and Accident Approaches at B19 (NTSB)

The Instrument Flying Handbook (FAA-H-8083-15B), in a section titled, “Airplane Attitude Instrument Flying using an Electronic Flight Display,” states:

Fixation, or staring at one instrument, is a common error observed in pilots first learning to utilize trend indicators. The pilot may initially fixate on the trend indicator and make adjustments with reference to that alone. Trend indicators are not the only tools to aid the pilot in maintaining the desired power or attitude; they should be used in conjunction with the primary and supporting instruments in order to better manage the flight. With the introduction of airspeed tapes, the pilot can monitor airspeed to within one knot. Fixation can lead to attempting to keep the airspeed to an unnecessarily tight tolerance.

The Pilot Handbook of Aeronautical Knowledge (FAA-H-8083-23B), states that, “Single-Pilot Resource Management (SRM) is about how to gather information, analyze it, and make decisions.” One practical application of SRM is the 5 P model, in which a pilot will regularly evaluate the Plan, Plane, Pilot, Passengers, and Programing. This evaluation should be accomplished before, during, and after a flight has been completed. An article produced by FAA Safety Briefing states:

The programming can refer to both panel-mounted and hand-held equipment. Today’s electronic instrument displays, moving map navigators, and autopilots can reduce pilot workload and increase pilot situational awareness. However, the task of programming or operating both installed and handheld equipment (e.g., tablets) can create a serious distraction from other flight duties. This part of the 5P approach reminds the pilot to mitigate this risk by having a thorough understanding of the equipment long before takeoff, and by planning in advance when and where the programming for approaches, route changes, and airport information gathering should be accomplished, as well as times it should not be attempted.

The Maine State Medical Examiner, Augusta, Maine, performed an autopsy of the pilot and determined the cause of death as "blunt force injuries."

The FAA Forensic Sciences Laboratory performed toxicological testing on the pilot. Test results were negative for the presence of illegal drugs, contraindicated drugs, and alcohol.