Georgia

The twin engine airplane departed Miami-Executive Airport (Kendall Tamiami) at 0628LT on a flight to Kennesaw-Cobb County-McCollum Airport, Georgia. Approaching Atlanta control area, the pilot declared an emergency and requested a clearance to land at the nearest airport. Vectored to Atlanta-DeKalb-Peachtree, the pilot modified his route and completed a left hand circuit. On final approach to runway 21L, the airplane went out of control and crashed at Homeland Drive, in a residential area located in Doraville, about 5 km north northeast of the airport. Both occupants were injured and the airplane was destroyed.

At the conclusion of the instrument flight rules flight, the pilot approached the 2,500-ft-long by 50-ft-wide runway at the destination airport. The pilot was unable to recall any of the events that transpired after the landing flare, but the front seat passenger described that the airplane’s nose pitched up as the airplane descended below the tops of the trees that surrounded the airport. The pilot then initiated a go-around, but the airplane did not climb as expected and drifted left towards the trees. The left wing then dropped before the airplane struck the ground, resulting in a postimpact fire. The fuselage and left wing were substantially damaged during the accident sequence. The pilot reported, and a postaccident examination of the wreckage confirmed, that there were no preimpact mechanical malfunctions of the airplane that would have precluded normal operation.

After landing on the grassy runway 36 at Jackson-Seven Lakes Airfield, the twin engine airplane went out of control, veered off runway to the left and came to rest against a tree. All three occupants were injured and the airplane was damaged beyond repair. It was officially registered N2072 and the registration NC2072 was on the fuselage.

On September 21, 2023, about 1436 eastern daylight time, an Embraer SA EMB-545 MOD airplane, N434FX, operated by Flexjet LLC as Flexjet flight 434, was substantially damaged when it was involved in an accident near St. Simons Island, Georgia. The pilot, copilot, and the six passengers were not injured. The airplane was operated as a Title 14 Code of Federal Regulations (CFR) Part 91K fractional flight. The airplane departed Westchester County Airport (HPN), White Plains, New York, at 1239 destined for St. Simons Island Airport (SSI), St. Simons Island. According to the flight crew, the departure and en route phase of the flight were uneventful, and they performed the checklists in accordance with procedures. The pilot (who was flying from the left seat) stated that he chose to fly the GPS runway 4 approach (versus flying a visual pattern) to allow more time for alignment and setup for landing. He said that he selected the “flaps 3” configuration due to the gusting wind component to be prepared in case it shifted to a crosswind and that the autothrottle was in use. The copilot said they had the runway in sight from 10 miles out, and he made the “500-ft stable, 20-kt headwind” callout when the airplane was descending through 500 ft agl. The NTSB airplane performance study determined that the FDR data and CVR audio for the airplane were consistent with a stable approach up to this point. The flight crew reported that all checklists and callouts were conducted. The FDR data showed that the fly-by-wire system was operating in normal mode, and the airplane’s autopilot was disconnected at 1435:49, when the airplane was about 500 ft. The pilot said that, when the airplane was about 100 to 150 ft agl, it crossed over a tree line, and he felt an “uplift,” then the airplane began to porpoise. The pilot tried to make pitch corrections using the sidestick control, but the airplane did not respond as commanded. The copilot said that the airplane was over the runway displaced threshold when it started pitching up and down. He looked over at the pilot’s hand on the sidestick, and he could see the pilot moving it back and forth in what seemed to him to be ”very aggressive” inputs. The copilot then said something like, “Take it easy there,” and the pilot replied that the airplane was not responding. The copilot estimated that the airplane was about 50 ft above the runway when he asked the pilot something along the lines of, “Would you like me to try?” The copilot said that he pulled back on his sidestick, and it felt like nothing happened. The copilot said that, at this point, it seemed like they were “along for the ride” and that he needed to brace himself for impact. The NTSB airplane performance study determined that, about 1436:19 (13 seconds before touchdown), when the airplane was about 162 ft radio altitude, the airplane’s AOA increased suddenly, likely due to a wind gust. The airplane’s AOA limiter protection system (aerodynamic stall prevention system) engaged, and the airplane’s AOA reduced. The FDR data for the pilot’s sidestick control showed that the pilot provided airplane-nose-up and airplane-nose-down pitch commands (as well as roll commands) and that the AOA limiter protection, which remained engaged until touchdown, limited the airplane’s response to the pilot’s pitch commands. (See the Tests and Research section for more information about the NTSB airplane performance study and systems functions during the accident flight.) In their postaccident statements, neither flight crew member indicated any awareness that the AOA limiter protection had engaged. The CVR transcript showed that, at 1436:28 (4 seconds before touchdown), the flight crew received a terrain awareness and warning system (TAWS) “sink rate” aural alert and reacted in surprise. During the 3 seconds before touchdown, the pilot commanded full airplane-nose-up pitch, and the fly-by-wire control system responded with about 3° airplane-nose-up elevator (full elevator deflection ranges from 25° up to 15° down). During the flare, the airplane was subjected to a horizontal gust which which resulted in a reduction of about 8 kts of indicated airspeed. During the 1.5 seconds before touchdown, the copilot also commanded full airplane-nose-up pitch, and the system responded with 5.5° airplane-nose-down elevator about 0.5 second before touchdown. The airplane impacted the ground at 1436:32. The copilot said that the airplane touched down “hard” before the runway threshold on centerline, and he thought that it bounced and then touched down again. He said the airplane was listing and drifting to the right as it skidded down the runway. The copilot remarked to the pilot that the airplane was not decelerating very much, and the pilot said that he was trying to get it stopped. The copilot then pulled the emergency brake handle, but nothing happened, so he put it back down. The airplane then went off the right side of the runway, hit a concrete base for a sign, then skidded mostly sideways before coming to rest on the grass to the right of the runway. After the airplane came to a stop, the copilot opened the cockpit door to check on the passengers and advised the pilot to shut down the engines so he could open the cabin door and deplane the passengers. The pilot shut down the engines, and all occupants deplaned through the main cabin door.

The student pilot, who was the new owner of the multi-engine airplane, and a private pilot flew commercially to Lubbock, Texas, utilized a ride-hailing service to drive to Portales, New Mexico; they met with the former owner of the airplane to finalize the purchase of the airplane and flew it back to Georgia the same day. The next day, the student pilot commenced flight training with the private pilot who offered to provide flight instruction to the student pilot in the student pilot’s newly acquired multi-engine airplane, even though he did not possess a flight instructor’s rating or a multi-engine airplane rating. Radar data showed that the track of the accident airplane's route consisted of their departure airport, a midway stop, and the third leg of the flight, where it crashed during the approach to their destination airport. Witnesses observed a sharp right turn before the airplane’s spiraling descent and impact with terrain and unoccupied semi-trailers. Surveillance footage from a parking lot security camera captured the airplane in a right spiral turn just before the accident. The airplane was destroyed by impact forces and the postimpact fire. The postaccident examination of the airframe, engines, and propellers revealed no anomalies that would preclude normal engine and airplane performance. Additionally, a review of the maintenance logbook revealed that the airplane was overdue for its annual maintenance inspection; no special flight permit (ferry permit) was obtained from the Federal Aviation Administration (FAA) for its return flight to Georgia. Toxicological testing of the student pilot revealed the presence amphetamine, a prescription Schedule II controlled substance that may result in cognitive deficits that pose a risk to aviation safety; however, its effect, if any on the accident flight could not be determined. It is likely that the private pilot’s failure to maintain aircraft control was exacerbated by his lack of a multi-engine airplane rating, his lack of a flight instructor rating, and his poor decision making.

The captain and first officer were assigned a two-leg overnight on-demand cargo flight. The flight crew were accustomed to flying night cargo flights, had regularly flown together, and were experienced pilots. The first leg of the trip was uneventful and was flown by the captain; however, their trip was delayed 2 hours and 20 minutes at the intermediate stop due to a delay in the freight arriving. The flight subsequently departed with the first officer as the pilot flying. While enroute, about forty minutes from the destination, the flight crew asked the air traffic controller about the NOTAMs for the instrument landing system (ILS) instrument approach procedure at the destination. The controller informed the flight crew of two NOTAMs: the first pertained to the ILS glidepath being unserviceable and the second applied to the localizer being unserviceable. When the controller read the first NOTAM, he stated he did not know what “GP” meant, which was the abbreviation for the glideslope/glidepath on the approach. The controller also informed the flight crew that the localizer NOTAM was not in effect until later in the morning after their expected arrival, which was consistent with the published NOTAM. The flight crew subsequently requested the ILS approach and when the flight was about 15 miles from the final approach fix, the controller cleared the flight for the ILS or localizer approach, to which the captain read back that they were cleared for the ILS approach. As the flight neared the final approach fix, the captain reported that they had the airport in sight; he cancelled the instrument flight rules flight plan, and the flight continued flying towards the runway. The airplane crossed the final approach fix off course, high, and fast. The cockpit voice recorder (CVR) transcript revealed that the captain repeatedly instructed the first officer to correct for the approach path deviations. Furthermore, the majority of the approach was conducted with a flight-idle power setting and no standard altitude callouts were made during the final approach. Instead of performing a go-around and acknowledging the unstable approach conditions, the captain instructed the first officer to use the air brakes on final approach to reduce the altitude and airspeed. Shortly after this comment was made, the captain announced that they were low on the approach and a few seconds later the captain announced that trees were observed in their flight path. The CVR captured sounds consistent with power increasing; however, the audible stall warning tone was also heard. Subsequently, the airplane continued its descent and impacted terrain about .70 nautical mile from the runway. The airplane was destroyed by impact forces and both occupants were killed.

Before the personal instrument flight rules flight began, the pilot obtained a preflight weather briefing that indicated that instrument meteorological conditions, convective activity, and moderate-to-severe icing conditions would be occurring along the route of flight. According to track data, while the pilot was navigating to avoid weather, the pilot was using the autopilot for maneuvering. After the pilot reported to air traffic control that the airplane would be turning direct to its destination, the performance analysis of track data showed that the airplane began a slight left turn with a bank angle of about 10°, which was consistent with the intended route of flight. However, the turn then reversed, and the airplane began banking to the right, reaching about 120° right wing down during the next 70 seconds and showing a slow oscillation in pitch attitude. Satellite imagery showed that the airplane was likely in instrument meteorological conditions when it began a rapid descent, and the airplane’s descent rate was about 7,000 feet per minute. Postaccident examination of the airplane and right engine (the left engine was not recovered) revealed no evidence of any preimpact mechanical malfunctions or failures that would have precluded normal operation. The dynamics of the airplane’s movements after the right turn began indicated that the airplane likely was not being actively controlled when it diverted from the intended flightpath. The circumstances of this accident were thus consistent with the pilot’s lack of timely recognition that the autopilot was disengaged. The available evidence for this accident precluded a determination of where the pilot’s attention was directed while navigating direct to the destination. However, the turbulence would have increased the pilot’s workload, and the restricted visibility conditions would have prevented the pilot’s use of outside cues to detect deviations in the airplane’s attitude. Also, the initial roll rates might not have been sufficient to provide reliable cues to the pilot of the developing bank, and the convective conditions would likely have made it difficult for the pilot to detect and recover from the fully developed unusual attitude.

According the commercial pilot and a flight instructor rated check pilot, they were conducting their first long-duration, aerial observation flight in the multiengine airplane, which was recently acquired by the operator. They departed with full fuel tanks, competed the 5-hour aerial observation portion of the flight, and began to return to the destination airport. About 15 miles from the airport, the left engine fuel warning light illuminated. Within a few seconds, the right engine stopped producing power. They attempted to restart the engine and turned the airplane toward an alternate airport that was closer. The pilots then turned on the electric fuel pump, the right engine began surging, and soon after the left engine stopped producing power. They turned both electric fuel pumps to the low setting, both engines continued to surge, and the pilots continued toward the alternate airport. When they were about 3 miles from the airport, both engines lost total power, and they elected to land on a highway. When they were a few feet above the ground, power returned briefly to the left engine, which resulted in the airplane climbing and beginning to roll. The commercial pilot pulled the yoke aft to avoid a highway sign, which resulted in an aerodynamic stall, and subsequent impact with trees and terrain. The airplane sustained substantial damage to the wings and fuselage. Although both pilots reported the fuel gauges indicated 20 gallons of fuel remaining on each side when the engines stopped producing power, the flight instructor noted that there was no fuel in the airplane at the time of the accident. In addition, according to a Federal Aviation Administration inspector who responded to the accident site, both fuel tanks were breached and there was no evidence of fuel spillage.

The pilot departed on an instrument flight rules cross-country flight with three passengers. While enroute at a cruise altitude about 6,000 ft mean sea level (msl), the pilot discussed routing and weather avoidance with the controller. The controller advised the pilot there was a gap in the line of weather showing light precipitation, and that the pilot could pass through it and then proceed on course. The controller assigned the pilot a heading, which the pilot initially acknowledged, but shortly thereafter, he advised the controller that the airplane was pointed directly at a convective cell. The controller explained that the heading would keep the pilot out of the heavy precipitation and that he would then turn the airplane through an area of light precipitation. The pilot responded, saying that the area seemed to be closing in fast, the controller acknowledged and advised the pilot if he did not want to accept that routing, he could be rerouted. The pilot elected to turn toward a gap that he saw and felt he could fly straight through it. The controller acknowledged and advised the pilot that course would take him through moderate precipitation starting in about one mile extending for about four miles; the pilot acknowledged. Radar information indicated that the airplane entered an area of heavy to very heavy precipitation, likely a rain shower updraft, while in instrument meteorological conditions, then entered a right, descending spiral and broke up in flight. Examination of the wreckage revealed no evidence of a preaccident malfunction or failure that would have prevented normal operation. The airplane was equipped with the capability to display weather radar "mosaic" imagery created from Next Generation Radar (NEXRAD) data and it is likely that the pilot was using this information to navigate around precipitation when the airplane encountered a rain shower updraft with likely severe turbulence. Due to latencies inherent in processes used to detect and deliver the NEXRAD data from the ground site, as well as the frequency of the mosaic-creation process used by the service provider, NEXRAD data can age significantly by the time the mosaic image is created. The pilot elected to navigate the hazardous weather along his route of flight based on the data displayed to him instead of the routing suggested by the controller, which resulted in the penetration of a rain shower updraft, a loss of airplane control, and a subsequent inflight breakup.

While on an instructional flight in icing and instrument meteorological conditions (IMC), the pilots indicated that they were having instrumentation difficulties to air traffic control. They initially reported a problem with the autopilot, then a navigational issue, which they later indicated were resolved, and finally they reported it was a problem with the left side attitude indicator. After air traffic control cleared them to their destination, the airplane entered a descending left turn, which continued into a 360° descending turn. An inflight breakup resulted, with the wreckage being scattered over 7,000 ft of wooded terrain. Examination of the engines revealed there were no anomalies that would have precluded normal operation prior to the accident. Control cable continuity was established from the flight controls in the cockpit to all flight control surfaces through multiple overload failures. The pitot-static system was examined, and no blockages were noted. Since there was rotational scoring noted on the vertical gyro and the directional gyro, it’s likely they were operating at the time of the accident. Furthermore, the left side attitude indicator examination revealed that there were no anomalies with the instrument. Examination of the deice valves for the deicing boots revealed that the left wing deice valve did not operate. Corrosion was visible in all three valves and it could not be determined if the corrosion was a result of postimpact environmental exposure. Furthermore, since the cockpit switch positions were compromised in the accident, it could not be determined if the pilots were operating the deicing system at the time of the accident. However, most of the pilot reports (PIREPs) in the area indicated light icing and the airplane performed a 6,000 ft per minute climb just before the loss of control. Given this information, it is unlikely the icing conditions made the airplane uncontrollable. A review of the pilots’ flight experience revealed that the pilot in the left seat did not hold a type rating for the accident airplane model but was scheduled to attend flight training to obtain such a type rating. The pilot in the right seat, who also held a flight instructor certificate, did hold a type rating for the airplane. Given that the remarks section of the filed flight plan described the flight as a “training flight” and the left-seat pilot’s plan to obtain a type rating for the accident airplane model, it is likely the pilot in the left seat was the flying pilot for the majority of the flight. Although the right-seat pilot's autopsy noted coronary artery disease, the condition was poorly described. The circumstances of the accident are not consistent with sudden physical impairment or incapacitation; therefore, it is unlikely it contributed to the event. Toxicology testing identified diphenhydramine, which can cause significant sedation, in the right-seat pilot’s blood. However, the level present at the time of the accident was too low to quantify. Therefore, it is unlikely effects from diphenhydramine contributed to the accident. Prior to entering the descending right turn, air traffic control noted that the airplane was not following assigned headings and altitudes and the pilots’ reported having autopilot problems. Subsequently, the pilots’ reported they were using the right attitude indicator as they had difficulties with the left-side indicator. Information was insufficient to evaluate whether the reported difficulties were the result of a malfunction of the autopilot or the pilots’ management of the autopilot system. However, the reported difficulties likely increased the pilots’ workload, may have diverted their attention while operating in IMC and icing conditions, resulting in task saturation, and may have increased their susceptibility to spatial disorientation. It is also possible that the onset of spatial disorientation was the beginning of the pilots’ difficulties maintaining the airplane’s flight track and what they perceived to be an instrumentation problem. Regardless, since the left seat pilot was not rated to fly the airplane, the right seat pilot’s workload would have increased by having to diagnose the issue, assess the situation, and maintain positive airplane control. The airplane’s track data are consistent with the known effects of spatial disorientation, leading to an inflight loss of control and subsequent inflight breakup.